Accepted Manuscript The response of Phaeodactylum tricornutum to quantum dot exposure: acclimation and changes in protein expression Elisabetta Morelli, Elisa Salvadori, Barbara Basso, Danika Tognotti, Patrizia Cioni, Edi Gabellieri PII:

S0141-1136(15)30006-4

DOI:

10.1016/j.marenvres.2015.06.018

Reference:

MERE 4026

To appear in:

Marine Environmental Research

Received Date: 27 January 2015 Revised Date:

19 June 2015

Accepted Date: 30 June 2015

Please cite this article as: Morelli, E., Salvadori, E., Basso, B., Tognotti, D., Cioni, P., Gabellieri, E., The response of Phaeodactylum tricornutum to quantum dot exposure: acclimation and changes in protein expression, Marine Environmental Research (2015), doi: 10.1016/j.marenvres.2015.06.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT

1

The response of Phaeodactylum tricornutum to quantum dot exposure:

2

acclimation and changes in protein expression

3 Elisabetta Morelli*, Elisa Salvadori, Barbara Basso, Danika Tognotti, Patrizia Cioni and

5

Edi Gabellieri

6 7

National Research Council - Institute of Biophysics, Section of Pisa, via Moruzzi, 1, 56124 Pisa, Italy

SC

8 9 10 11 12

RI PT

4

M AN U

*Corresponding author: E. Morelli, Institute of Biophysics - CNR, via Moruzzi, 1 - 56124 Pisa Italy. tel.: +39 0503152757 e-mail: [email protected]

Abstract

14

Nanotechnology has a great potential to improve life and environmental quality,

15

however the fate of nanomaterials in the ecosystems, their bioavailability and potential

16

toxicity on living organisms are still largely unknown, mainly in the marine environment. Genomics and proteomics are powerful tools for understanding molecular mechanisms

TE

17

D

13

triggered by nanoparticle exposure. In this work we investigated the effect of exposure

19

to CdSe/ZnS quantum dots (QDs) in the marine diatom Phaeodactylum tricornutum,

20

using different physiological, biochemical and molecular approaches. The results show

21

that acclimation to QDs reduced the growth inhibition induced by nanoparticles in P.

22

tricornutum cultures. The increase of glutathione observed at the end of the lag phase

23

pointed to cellular stress. Transcriptional expression of selected stress responsive

24

genes showed up-regulation in the QD-exposed algae. A comparison of the proteomes

25

of exposed and unexposed cells highlighted a large number of differentially expressed

26

proteins. To our knowledge, this is the first report on proteome analysis of a marine

27

microalga exposed to nanoparticles.

CC

EP

18

28 29

Keywords }

1

ACCEPTED MANUSCRIPT

30

Ecotoxicology; phytoplankton; growth; Phaeodactylum tricornutum; glutathione;

31

quantum dots; nanoparticles; proteomics.

32 1. Introduction

RI PT

33

The rapid increase of nanotechnology has raised the concern about the potential

34

effects of nanoparticles on the ecosystems and living organisms. With the growing

36

production and use of nanoparticles, an increasing input in the aquatic environment is

37

expected, especially in riverine and coastal areas (Corsi et al, 2014). Despite their

38

importance to assess environmental risk, cellular and molecular mechanisms taking

39

place in model organisms to cope with nanoparticles exposure are, in general, poorly

40

understood (Matranga and Corsi, 2012).

41

Quantum dots (QDs) are semiconductor nanocrystals representing a great promise for

42

nanotechnologies, not only in biomedical applications, but also in a variety of industrial

M AN U

D

43

SC

35

products, including sensors, light emitting materials and solar cells (Winnik and

Maysinger, 2013). Since QDs are becoming more and more attractive in the global

45

market, the need to understand the potential risks for human health and the ecosystem,

CC

EP

TE 44

46

especially for the species at the basis of the food chain, increases. The large variety of

47

types of QDs, the different physicochemical properties, and the complex interactions

48

with environmental factors, make it very difficult to assess their toxicity, especially in the

49

aquatic environment (Hardman, 2006; Leigh et al., 2012). The toxicity of QDs can be

50

affected by their physicochemical properties, thus it is highly recommended to monitor

51

them during exposure experiments. Aggregation and degradation of nanoparticles are

52

common events in natural waters, which can be mimicked in culture media, where

53

nanoparticles are subjected to light irradiation, high saline concentration, presence of

54

biogenic organic matter and living microorganisms (Klaine et al., 2008).

}

2

ACCEPTED MANUSCRIPT

Recently, an increasing literature has studied the effect of these nanoparticles in

55

model systems, showing that QDs can induce different degrees of toxicity in organisms

57

with different level of organization, such as bacteria (Yang et al., 2012), unicellular

58

eukaryotes (Mei et al., 2014; Domingos et al., 2011), invertebrates (Ambrosone et al.,

59

2012; Gagnè et al., 2008), fishes (Zhang et al., 2012a) and mammalian cells in culture

60

(Chen et al., 2012; Su et al., 2010). Although several studies have been carried out in

61

freshwater organisms, only a few papers address the effect on marine organisms

62

(Handy et al., 2008; Quigg et al., 2013; Munari et al., 2014). Aquatic microorganisms

63

can be considered suitable model systems to study ecotoxicological hazards of

64

nanomaterials, due to their ubiquity and abundance in the aquatic environment, as well

65

as to their ability to provide an early warning of ecotoxicity (von Moss and Slaveykova,

66

2014).

M AN U

SC

RI PT

56

Conventional biomarkers have been proved to be sensitive indicators to assess the

68

D

67

toxic effects of nanoparticles. The induction of Reactive Oxygen Species (ROS) is a

main target of nanotoxicity research. An enhanced ROS production stimulates a

70

complex network of defence reactions involving enzymes and metabolites with

CC

EP

TE 69

71

antioxidant properties, able to scavenge excess ROS and mitigate oxidative damage in

72

plants. The main defense enzymes include superoxide dismutase (SOD), catalase

73

(CAT), glutathione reductase (GR) and peroxidases. Among the non-enzymatic

74

antioxidants, glutathione represents a major redox component in plant cells and protects

75

plants against oxidative stress and heavy metals. Glutathione also serves as a

76

precursor in phytochelatins (PCs) production, an active detoxification mechanism

77

developed by plants, algae and fungi to avoid heavy metal toxicity (Grill et al., 1985).

78

Common response mechanisms involving heat shock proteins (HSPs) seem also to

79

play a role in nanotoxicity, since their expression was found to be altered in some

80

organisms exposed to a variety of nanoparticles, including QDs (Gomes et al, 2013; }

3

ACCEPTED MANUSCRIPT

Winnick and Maysinger, 2013; Ambrosone et al., 2012). On the other hand,

82

conventional biomarkers are often unable to differentiate nano-specific biological

83

responses from those induced by their similar ionic/bulk forms. Only recently the

84

ecotoxicological research on nanoparticles, including aquatic environments, moved

85

towards toxicogenomics. Genomic and proteomic-based approaches, by looking for

86

specific gene and protein expression alterations, can provide a range of responses

87

indicative of nanoparticle exposure, that can be used as unbiased biomarkers (Zhuang

88

and Gao, 2014). Despite the promising use of these techniques, up to now only a few

89

recent papers report alterations induced by nanoparticles at the molecular level in

90

bivalves (Tedesco et al., 2010; Gomes et al., 2013; Hu et al., 2014) and, to our

91

knowledge, no study concerns microalgae.

M AN U

SC

RI PT

81

Our study was carried out using a model diatom species, Phaeodactylum

92

tricornutum, whose genome has been completely sequenced (Bowler et al. 2008). In

94

D

93

previous papers, we show that CdSe/ZnS QDs inhibit the growth rate of P. tricornutum

and increase both the membrane lipid peroxidation and the activity of important

96

antioxidant enzymes, such as SOD and catalase, in response to increased ROS

CC

EP

TE 95

97

production (Morelli et al., 2013). Other authors report that this alga can adapt to

98

changing environmental conditions, providing early-warning protective mechanisms

99

(Vardi et al., 2006; De Martino et al., 2007). Mechanisms of acclimation to unfavourable

100

conditions, consisting of the modification of the membrane fatty acids profile, have also

101

been described in eukaryotic microorganisms exposed to TiO2 (Rajapakse et al., 2012)

102

or CuO (Mortimer et al., 2011) nanoparticles. Acclimation has been defined as a short-

103

term phenotypic change, which allows survival in suboptimal environmental conditions,

104

lowering the toxic effects (Rajapakse et al., 2012). In the present paper, we investigated the QD toxicity and the capability of P.

105 106

tricornutum cells to adapt to the presence of QDs in the culture medium, using different }

4

ACCEPTED MANUSCRIPT

physiological, biochemical and molecular approaches. In particular, at first we followed

108

the effect of QDs on the growth curve and on the intracellular concentration of

109

glutathione and phytochelatins, used as stress signals. Furthermore, we characterized

110

the QD particle size at different times, by using a sequential fractionation technique,

111

trying to relate the toxicity of the particle to its size. By using a semi-quantitative RT-

112

PCR approach, we tested the expression of selected genes, presumably involved in

113

defence mechanisms. Finally, we compared the proteomic pattern of acclimated QD-

114

exposed algae with that of unexposed algae, in order to highlight differences of protein

115

expression due to QD-exposure.

M AN U

SC

RI PT

107

116

2. Materials and Methods

117 118

2.1 Quantum dots

Lumidot orange quantum dots (QDs) emitting at 590 nm were purchased by

TE

120

D

119

Sigma-Aldrich. These QDs are constituted by a CdSe core with a ZnS shell, stabilized

122

by hexadecylamine (HDA) as a ligand coating surface, and shipped in 5 mg ml-1 toluene

CC

EP

121

123

dispersion. QDs were transferred into aqueous media by encapsulating with the

124

amphiphilic polymer poly(styrene-co-maleic anhydride) terminated with cumene (PSMA)

125

and ethanolamine (EA), by following the procedure reported by Lees et al. (2009), with

126

some modifications. The detailed preparation together with the chemical and optical

127

characteristics of water-soluble QDs have been described elsewhere (Morelli et al.,

128

2012). Briefly, a 100 µl aliquot of QDs (5 nmoles) was vacuum-dried, in order to remove

129

toluene, and re-suspended in 1 ml of chloroform containing 500 nmoles of PSMA. After

130

tumbling overnight at room temperature, 2 ml of water containing 2 µl of EA were added

131

to the QDs/PSMA chloroform solution and tumbled for further 4-5 hours. During this time

132

the QDs were transferred from the organic phase to water. Finally, the sample was }

5

ACCEPTED MANUSCRIPT

centrifuged at 10000×g for 10 minutes and the colored aqueous phase was used

134

throughout. The same encapsulation procedure was performed with HDA, without QDs,

135

in order to obtain a procedural blank (named HDA/PSMA), useful to evaluate the

136

possible effect of the organic constituents of QDs on algal toxicity. An amount of HDA

137

(0.5 mg) equal to the total amount of QDs (coated by HDA) was used, thus the

138

concentration of HDA was surely higher than that occurring in the QD stock suspension.

139

A separate pilot test showed no growth reduction in cultures exposed to HDA/PSMA,

140

indicating that the organic coating of QDs encapsulated with PSMA was not toxic to

141

algae (Fig 1 of Supplementary Information). Moreover, we exposed algae to 1.2 µM

142

CdCl2, a concentration of Cd equivalent to that derived from a complete degradation of

143

2.5 nM QDs, in order to assess the effect of a possible dissolution of Cd2+ from

144

nanoparticles. No significant alteration of the growth was observed (Fig. 1 of

145

Supplementary Information), in agreement with results reported elsewhere (Morelli and

SC

M AN U

D

146

RI PT

133

Scarano, 2001). QD concentration was measured spectrophotometrically using the molar absorption

148

coefficient, ε569=1.6×105 M-1cm-1, provided by the manufacturer. Total Cd was measured

CC

EP

TE 147

149

by Atomic Absorption Spectrometry (Perkin Elmer, Ueberlingen, Germany), after

150

acidification with HNO3 (0.3% v/v). QD and total Cd concentrations in the stock aqueous

151

suspension were 1.3 µM and 620 µM, respectively. As previously stated by fluorescence

152

correlation spectroscopy, the size of the nanoparticle was 17 ± 3 nm (Morelli et al.,

153

2013). Stock suspensions of water-soluble QDs were stored in the dark at + 4° C for a

154

maximum of 3 months and used for exposure experiments with algae. Water was

155

purified by a Milli-Q system from Millipore (Vimodrone, Italy).

156 157

2.2 Cultures

}

6

ACCEPTED MANUSCRIPT

The unicellular marine diatom Phaeodactylum tricornutum (Bohlin) was obtained

158

from the Culture Collection of Algae and Protozoa, Dunstaffnage Marine Laboratory,

160

UK. Stock cultures were grown in axenic conditions at 21 ± 1° C and fluorescent

161

daylight (100 µmol photons x m-2 x s-1) in a 16:8 light-dark cycle. Culture medium was

162

natural seawater enriched with f/2 medium (Guillard, 1975) modified to obtain a f/10

163

medium for trace metal concentration. Seawater was collected in an uncontaminated

164

area, 3 miles offshore from the Island of Capraia (Tyrrhenian Sea, Italy), filtered through

165

0.45 µm membrane filters (Millipore) and stored in the dark at + 4° C. Exponential

166

growth was maintained by inoculating cells weekly into a fresh sterilized medium.

M AN U

SC

RI PT

159

In this study we performed a number of exposure experiments, which followed the

167

same basic procedure. They were carried out by inoculating algae from a stock culture

169

to a fresh medium at an initial cell density of 5 x 104 cells ml-1. Batch cultures were

170

manually stirred once a day. Control cultures (no QDs added) were always used. Three replicates of the exposure experiments were carried out. Cell density was measured by

TE

171

D

168

using a haemocytometer under a microscope or, alternatively, by recording the

173

absorbance of chlorophyll at 680 nm (JASCO V-550 UV/Vis Spectrophotometer, Lecco,

CC

EP

172

174

Italy). In a first set of exposure experiments, designed to evaluate the effect of QDs on the

175

176

growth, aliquots of a P. tricornutum culture were spiked with suitable amounts of

177

CdSe/ZnS QDs, in order to obtain a final range of QD concentrations from 0.5 to 2.5

178

nM. The growth was monitored daily by counting the number of cells. Growth is

179

characterized by a non-linear function, which can be described by the following equation

180

according to the Gompertz model (Seber and Wild, 1989):

181

y(t) = y0 + a⋅ exp(-exp(-k(x-xc))

182

where y(t) and y0 represent the cell density at time t and t = 0, respectively; a represents

183

the amplitude of the growth curve; k is a coefficient used to calculate the maximum }

(equation 1)

7

ACCEPTED MANUSCRIPT

growth rate (µmax) by the equation: µmax = a⋅⋅k/e; xc represents the x-value at the point of

185

inflection of the curve, with y=a/e, at which point the maximum growth rate occurs. This

186

value is the time needed to reach the 1/e (36.8%) of a, the amplitude of the growth

187

curve and is related to the lag-time. All the parameters were fitted by non-linear

188

regression using OriginPro 8.5.1 software (Northampton, MA, USA).

RI PT

184

Successive exposure experiments were conducted by spiking cultures with QDs at

190

a final concentration of 2.5 nM, corresponding to an equivalent Cd concentration of 1.2

191

µM. In the first set of incubation experiments, 10 ml aliquots of cultures or culture

192

medium (without algae), maintained under the same experimental conditions, were

193

sampled at time intervals and used for the experiment of QD size-fractionation

194

described below. In the second set of incubation experiments, designed to follow the

195

temporal pattern of glutathione and PCs, suitable aliquots of 500 ml cultures were

196

sampled at different time intervals and processed for the analytical determination.

M AN U

D

Finally, a 4-day exposure experiment at 2.5 nM QDs was carried out in order to extract

TE

197

SC

189

proteins and RNA for proteomic and semiquantitative Real Time-PCR assays,

199

respectively.

CC

EP

198

200 201

2.3 Size fractionation of QDs Aliquots of 10 ml of the culture medium at 2.5 nM QDs, in the absence or in the

202 203

presence of algae, were subjected to sequential filtration by the following procedure,

204

according to Zhang et al. (2012b) with some modifications. The first and the second

205

filtrations were carried out with a 1.2 µm and a 0.2 µm membrane (Millipore),

206

respectively. After filtration, membrane filters were rinsed two times with natural

207

seawater to remove any residual of the exposure medium, re-suspended in 2 ml 5%

208

HNO3 and subjected to Cd measurements. According to this operational procedure, we

209

obtained a first fraction of Cd bound in aggregates larger than 1.2 µm (Cd>1.2 µm) and }

8

ACCEPTED MANUSCRIPT

a second fraction of Cd in aggregates ranging from 1.2 to 0.2 µm (0.2 µm