1 Effects of zinc supplementation on growth and colouration of the scleractinian coral 2 Stylophora pistillata

3 Jennifer Tijssen 1* , Tim Wijgerde 2* , Miguel C. Leal 3 and Ronald Osinga 1,4

4 1. Aquaculture and Fisheries Group, Department of Animal Sciences, Wageningen University, 5 Wageningen University and Research centre, Wageningen, The Netherlands

6 2. Coral Publications, Utrecht, The Netherlands

7 3. Department of Fish Ecology and Evolution, EAWAG: Swiss Federal Institute of Aquatic 8 Science and Technology, Center for Ecology, Evolution and Biogeochemistry, Kastanienbaum, 9 Switzerland

10 4. Marine Animal Ecology Group, Department of Animal Sciences, Wageningen University, 11 Wageningen University and Research centre, Wageningen, The Netherlands

12

13 *These authors contributed equally to this work

14

15 Corresponding author: Tim Wijgerde, Livingstonelaan 1120, 3526JS Utrecht, The Netherlands

16 Email: [email protected]

17

18 Keywords: Stylophora pistillata , zinc, toxicity, scleractinian, coral, NOEC

19

20

21

22

23 24

25

26

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 27 Abstract

28 Zinc levels in artificial seawater are often unnaturally elevated, and thus potentially toxic to 29 aquacultured corals. However, our knowledge of how zinc affects corals is still limited. We 30 tested the effects of zinc supplementation (0, 1, 10 and 100 µg L –1) on health, growth, NDVI (a 31 proxy for chlorophyll a) and overall colouration of the stony coral Stylophora pistillata . After 32 two weeks, no signs of necrosis were observed in any of the treatments. However, at 100 µg L – 33 1, we detected a considerable ~62% growth reduction compared to zinc levels of 0 to 10 µg L – 34 1. In addition, NDVI was significantly reduced by ~36% at 100 µg L –1 zinc, indicating loss of 35 chlorophyll a. Zinc did not affect coral colouration in general, although reflection intensity 36 increased markedly at 100 µg L –1, most likely due to a loss of chlorophyll a. In conclusion, the 37 No Observed Effect Concentration (NOEC) after a two–week zinc exposure was 10 µg L –1 for 38 S. pistillata . Our results show that potentially toxic metals such as zinc, found in commercial 39 sea salts and supplements, can have detrimental effects on corals. Therefore, we recommend 40 regular monitoring and restrained supplementation of zinc in coral aquaculture systems.

41

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 42 1. Introduction

43 Coral aquaculture and mariculture are growing activities, due to high market demand for corals 44 as live ornamentals for marine aquaria and biotechnological applications (Leal et al., 2014). In 45 addition, coral culture delivers biomass for restoration activities and, potentially, drug 46 development (Jaap, 2000; Rocha et al., 2011; Young et al., 2012; Leal et al., 2013). Today, 47 coral culture is mostly practiced in Western countries, using aquaculture systems with artificial 48 lighting, water flow and sea salts (Wijgerde et al., 2012; Rocha et al., 2015). Although the 49 effects of light and water flow on coral growth and physiology have been thoroughly 50 investigated (Schutter et al., 2008, 2010, 2011; Wijgerde et al., 2012, 2014; Rocha et al., 51 2013a,b), the role of different sea salts and their variable compositions has been overlooked. 52 Artificial sea salts have chemical compositions resembling that of natural seawater, but they 53 are often unnaturally enriched in several potentially toxic elements such as zinc, with a 54 concentration range of 10–100 µg L –1 at 35 g L –1 salinity (Atkinson and Bingman, 1997; 55 Wijgerde et al., 2014). For comparison, pristine seawater has a zinc content of 0.01–0.36 µg L – 56 1 (Ferrier-Pagès et al., 2005; Houlbrèque et al., 2012), which is up to four orders of magnitude 57 lower compared to freshly prepared artificial seawater. Zinc is an essential cofactor for many 58 enzymes with important biological functions in corals and zooxanthellae, including superoxide 59 dismutase (protection against thermal and light stress) and carbonic anhydrase (calcification, 60 photosynthesis) (Houlbrèque et al., 2012 and references therein). Indeed, elevating the zinc 61 concentration with 0.65 µg L –1 enhances photosynthesis of the scleractinian coral Stylophora 62 pistillata , suggesting that corals are zinc–limited under natural conditions (Ferrier-Pagès et al., 63 2005). However, zinc becomes toxic to scleractinian corals at higher concentrations, imparing 64 sexual reproduction (egg fertilization) in the range of 10–500 µg L –1 (Reichelt–Brushett and 65 Harrison, 1999, 2005). Despite their already high content in many sea salts, metals such as zinc 66 are used as a supplement in the aquarium industry to promote coral colouration, a desirable trait 67 in the aquarium hobby (Balling, 1996; Balling et al., 2008; Leal et al., 2014). 68 Despite anecdotal reports, scientific knowledge of how metals such as zinc affect corals 69 in aquaculture, in terms of health, growth and colouration, is still limited. To address this issue, 70 we tested the short-term effects of zinc supplementation on the scleractinian coral Stylophora 71 pistillata . We subjected S. pistillata colonies to zinc concentrations of 0, 1, 10 and 100 µg L –1, 72 covering the concentration range observed in natural and artificial seawater (Atkinson and 73 Bingman, 1997; Ferrier-Pagès et al., 2005; Houlbrèque et al., 2012; Wijgerde et al., 2014), over

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 74 a period of two weeks. During this period, we determined coral health, growth rate, NDVI (a 75 proxy for chlorophyll a) and overall colouration. 76 77 78 2. Materials and Methods 79 80 2.1. Coral Husbandry and Fragmentation

81 The Indo–Pacific scleractinian coral Stylophora pistillata (Esper 1797) was used in this study. 82 Captive–bred corals were originally obtained from Burgers' Zoo BV (Arnhem, The 83 Netherlands). The experiment was conducted at Wageningen University (Wageningen, The 84 Netherlands), with permission from Burgers' Zoo BV. At the time of the experiment, the parent 85 colony used had been in culture for approximately seven years at Wageningen UR after being 86 obtained from Burgers’ Zoo. No approval from an ethics committee was required as 87 scleractinian corals are exempted from legislation concerning the use of laboratory animals in 88 the European Union (Directive 2010/63/EU). 89 Apical tips with an approximate length of 1 cm (N=32) were randomly cut from a parent 90 colony and glued onto 5x5 cm PVC tiles (Wageningen UR, Wageningen, The Netherlands) 91 using cyanoacrylate (Gamma BV, Wageningen, The Netherlands). All fragments were weighed 92 before being glued (see Specific Growth Rate). All fragments were allowed to recover for six 93 weeks in a 400 L holding aquarium before the onset of the two–week experiment. The holding 94 aquarium was provided with full spectrum white light, using two 4x54W T5 fixtures containing 95 Aquablue Spezial bulbs (Elke Müller Aquarientechnik, Hamm, Germany). Irradiance was set 96 to 160 µmol m –2 s–1 (photosynthetically active radiation, ~400–700 nm) with a 12h:12h 97 light:dark regime. After an acclimation period of two weeks, this light regime is sufficient to 98 near–saturate photosynthesis in this particular genotype (unpublished results). Water flow was 99 provided by a Turbelle Stream 6085 circulation pump (Tunze Aquarientechnik GmbH, 100 Penzberg, Germany) providing a total flow rate of 8,000 L h –1. 101

102 2.2. Zinc Treatments

103 Eight aquaria with a volume of 25 L each were set up, with two independent aquarium replicates 104 for each zinc treatment. The four zinc concentrations tested were 0, 1, 10 and 100 µg L –1 as 105 Zn 2+ . Each aquarium housed four small Stylophora pistillata fragments ( N=32 total). The light

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 106 source and regime of all aquaria were identical to those of the holding aquarium. To generate 107 water flow, small submersible pumps were used (model 1000, Eheim GmbH Co. KG, Deizisau, 108 Germany). Nearly zinc–free seawater was obtained at Burgers’ Zoo (Arnhem, The 109 Netherlands). The zinc level of this seawater was measured by Inductively Coupled Plasma 110 Mass Spectrometry (ICP–MS) at the Chemical Biological Soil Laboratory (CBLB, Wageningen 111 UR, Wageningen, The Netherlands), resulting in negative values, suggesting values near zero. 112 This water was subsequently enriched in zinc for the various treatments, using zinc sulphate. 113 Two control aquaria received seawater with no added zinc sulphate. Zinc concentrations and 114 other water chemistry parameters were held approximately constant throughout the experiment, 115 by weekly changing a third of the seawater volume in each aquarium with new seawater having 116 the appropriate zinc concentration. To aid the maintenance of stable water chemistry, a 117 relatively low coral biomass was kept in each aquarium (i.e. four small fragments). Water 118 temperature was maintained by placing 50W heating elements (Sicce Srl, Pozzoleone, Italy) in 119 each aquarium. Water parameters were maintained at the following levels: salinity 35 g L –1, 120 temperature 26°C, pH 8.25, ammonium <0.03 mg L –1, nitrate <0.02 mg L –1, phosphate <0.03 121 mg L –1, calcium 419 mg L–1, alkalinity 5.2 mEq L –1. All aquaria were supplied with 0.5 mL of 122 Artemia nauplii suspension (approximately 3,000 nauplii mL –1) twice a week. 123 124 2.3. Specific Growth Rate

125 To determine specific growth rates, we use the buoyant weighing method of Davies (1989). 126 Buoyant mass provides an estimator of skeletal mass increase, since coral tissue has a similar 127 density to that of seawater and therefore barely contributes to buoyant mass (Schutter et al., 128 2008). Measurements were done by suspending each fragment, attached to a perforated PVC 129 plate, on a hook. The hook was attached to an analytical balance (model 300, A&D, Tokyo, 130 Japan) in a defined volume of seawater at constant depth. The seawater in the weighing chamber 131 was maintained at 26 °C and a salinity of 35 g L –1. All buoyant weights were corrected for the 132 mass of each PVC plate and epoxy resin. Specific growth rates were calculated with the 133 following formula:

134 SGR = (lnW t – lnW t–1) / Δt

135 where ln is the natural logarithm, Wt and W t−1 are the final and initial coral net weights 136 expressed in grams (g), respectively, and Δt is the growth interval in days. SGR is expressed in 137 gram coral gram coral −1 day −1, which can be simplified as day −1.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 138

139 2.4. Coral Spectral Reflectance and NDVI

140 Diffusive reflectance spectra were measured after two weeks over a 190–892 nm bandwidth, 141 with a spectral resolution of 0.33 nm, using a USB2000 spectrometer (USB2000–VIS–NIR, 142 grating #3, Ocean Optics, Dunedin, USA) connected to a 400 μm diameter fibre optic cable 143 (QP400–2–VIS/NIR–BX, Ocean Optics, Dunedin, USA). To minimise background reflection, 144 each coral fragment was removed from the aquarium and placed in a white, styrofoam dome– 145 shaped container. The fibre optic sensor was maintained perpendicular to the coral surface, at a 146 fixed distance, defined to match a view field covering a circular area of approximately 3 mm 147 diameter on the surface of each coral fragment. During measurements, all coral fragments and 148 the reference styrofoam (see below) were illuminated from below, using a full spectrum halogen 149 light (Philips, Eindhoven, The Netherlands). The light spectrum reflected from each coral 150 fragment was normalised to the spectrum reflected from a white styrofoam imitation coral. The 151 reflectance spectrum measured in the dark was subtracted from both spectra to account for the 152 dark current noise of the spectrometer. Each coral fragment was measured on four different 153 sides. The four measurements were averaged before being used for subsequent calculations. 154 The Normalised Difference Vegetation Index (NDVI) (Rouse, 1973) was used as a 155 proxy for chlorophyll a content (Rocha et al., 2013a,b; Wijgerde et al., 2014; Leal et al., 2015)

156 and calculated with the formula below, where R 750 and R 675 represent the average diffusive 157 reflectance in the intervals of 749.73–750.39 nm and 674.87–675.55 nm, respectively. 158

Rͫͩͤ − Rͪͫͩ 159 NDVI = Rͫͩͤ + Rͪͫͩ 160

161 2.5. Data Analysis

162 Normality of data was tested by plotting residuals of each dataset versus predicted values, and 163 by performing a Shapiro–Wilk test. Homogeneity of variances was determined using Levene's 164 test. All data were normally distributed and homoscedastic after a square root transformation 165 (P>0.050). We used a two–way factorial ANOVA to test the interactive effects of zinc 166 concentration and aquarium on specific growth rates and NDVI. A Bonferroni post–hoc test 167 was used to test for differences between zinc levels. A P<0.050 value was considered 168 statistically significant. Statistical analysis was performed with SPSS Statistics 20 (IBM,

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 169 Somers, USA). Graphs were plotted with SigmaPlot 12 (Systat software, San Jose, USA). All 170 data presented are expressed as mean and standard deviation.

171

172 3. Results

173 All corals survived the experiment, with no signs of necrosis. All corals grew during the two 174 weeks, varying from 0.0030 to 0.0095 day -1, i.e. 0.30 to 0.95% day -1 (Figure 1). Zinc had a 175 significant effect on coral growth (Table 1), with a ~62% growth reduction at 100 µg L –1 176 compared to zinc levels of 0 to 10 µg L –1 (Bonferroni, P≤0.001). No growth differences between 177 0, 1 and 10 µg L –1 were found (Bonferroni, P=1.000). No main or interactive effect of aquarium 178 on growth rates was detected (Table 1). 179 After two weeks, coral reflectance was similar between 0, 1 and 10 µg L –1, although it 180 increased markedly at 100 µg L –1 (Figure 2). The shape of the reflection curves remained similar 181 across treatments, with small, but distinct reflectance peaks at 545 and 611 nm. In addition, all 182 corals clearly showed reflection minima at 670 nm, and similar reflection levels at 700 nm and 183 beyond. 184 After two weeks, zinc had a significant effect on NDVI (Table 1), with a ~36% reduction 185 at 100 µg L –1 compared to a zinc level of 0 µg L –1 (Bonferroni, P=0.000, Figure 3). No 186 differences in NDVI were found between 0, 1 and 10 µg L –1 (P=1.000). No main or interactive 187 effect of aquarium on NDVI was detected (Table 1). 188

189 4. Discussion

190 This study revealed that elevated zinc concentrations can have negative effects on scleractinian 191 corals, with a No Observed Effect Concentration (NOEC) of 10 µg L–1 after a two–week zinc 192 exposure for S. pistillata . More specifically, a zinc concentration of 100 µg L –1 considerably 193 impaired coral growth by ~62% as compared to zinc levels of 0 to 10 µg L –1, and reduced 194 chlorophyll a concentrations (as indicated by NDVI) by ~36%. The growth–impairing effect of 195 zinc may be caused by a direct toxic effect on the coral host, and/or an indirect effect on the 196 zooxanthellae. Although zinc is a critical cofactor for the antioxidant superoxide dismutase 197 (Houlbrèque et al., 2012), paradoxically, it induces reactive oxygen species in plant and animal 198 cells at high concentrations, potentially resulting in damage to DNA, organelles and cell 199 membranes (Alia et al., 1995; Prasad et al., 1999; Daniels et al., 2004). Although no necrosis

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 200 was observed after two weeks, long–term exposure may result in further growth decline, 201 bleaching and/or tissue loss. 202 The NOEC observed for S. pistillata during this study is different from other coral 203 species. Reichelt–Brushett and Harrison (2005) studied the effects of zinc on fertilization 204 success of Acropora tenuis ova, and found that a 5.5 hour zinc exposure resulted in a NOEC 205 below 10 µg L –1, with a 21 percentage point fertilization reduction at 10 µg L –1 compared to 206 controls, and a 90 percentage point reduction at 100 µg L –1. Goniastrea aspera , in contrast, was 207 found to be much less sensitive to zinc in terms of fertilization, with a NOEC of >500 µg L –1 208 (Reichelt–Brushett and Harrison 2005). However, a longer zinc exposure, i.e. days to weeks, 209 may still reveal negative effects such as growth impairment and bleaching. In addition, growth 210 and sexual reproduction are different biological processes which may be differentially affected 211 by zinc. More generally, our results are in agreement with many studies reporting toxic effects 212 of elevated zinc levels on marine invertebrates, including corals, oysters, mussels, sea urchins, 213 brittle stars, crinoids, amphipods, crabs, bryozoans and polychaetes (Wisely and Blick, 1967; 214 Ahsanullah, 1976; Martin et al., 1981; Nacci et al., 1986; Heyward, 1988; Reichelt–Brushett 215 and Harrison 1999, 2005; Sondervan, 2004). 216 The observed increase in reflectance (especially at 670 nm) and decrease in NDVI of 217 corals exposed to the highest zinc concentration suggests a loss of chlorophyll a, zooxanthellae 218 and coral pigments (Halldal, 1968; Jeffrey and Haxo, 1968; Rocha et al., 2013a,b; Wijgerde et 219 al., 2014; Leal et al., 2015). This is congruent to the findings of Jones (1997), who observed 220 zooxanthellae expulsion in Acropora formosa after a copper exposure at 20 µg L –1 and above, 221 i.e. levels comparable to our study. Expulsion of zooxanthellae after heavy metal exposure was 222 also found for the zooxanthellate anemone Anemonia viridis , with zooxanthellae predominantly 223 taking up zinc and copper before being expelled from host tissue (Harland and Nganro, 1990; 224 Harland et al., 1990). Harland and Nganro (1990) proposed that zooxanthellae expulsion could 225 be a mechanism used by anemones to excrete excess metals, thereby regulating metal uptake at 226 elevated concentrations. The same mechanism may be used by corals; even though coral growth 227 is significantly reduced when zooxanthellae are expelled, a temporal loss of photosynthetic 228 symbionts could be a mechanism to cope with metal toxicity stress. 229 Although corals exposed to 100 µg L –1 showed signs of pigment loss, particularly 230 chlorophyll a, no overall change in tissue colouration was observed. All corals in this 231 experiment displayed similar reflection curves, with reflectance peaks at 545 and 611 nm, 232 which could be due to the presence of green and red fluorescent proteins, respectively (Dove et 233 al., 2001; Wilmann et al., 2005; Matz et al., 2006; D’Angelo et al., 2008).

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 234 The toxic effects of zinc on S. pistillata become apparent somewhere between a 235 concentration of 10 and 100 µg L –1, which is specifically the range found in artificial seawater 236 produced from most commercial sea salts (Atkinson and Bingman, 1997; Wijgerde et al., 2014). 237 Thus, it is possible that scleractinian corals produced in aquaculture are currently negatively 238 affected by elevated zinc concentrations, although the common practice of limited water 239 changes may result in relatively low zinc values due to rapid uptake by large numbers of 240 organisms usually present in culture systems (Balling et al., 2008). As sensitivity to the toxic 241 effects of zinc is species-specific (Schutter et al., 2011; Wijgerde et al., 2014), and possibly 242 genotype-specific, NOEC values for other coral species and genotypes will have to be 243 determined in future experiments. In addition, long–term studies on the effects of zinc exposure 244 are also warranted. 245 In conclusion, we found no beneficial effects of zinc supplementation on S. pistillata , 246 with a NOEC of 10 µg L –1 after a two–week exposure, in terms of coral health, growth, 247 reflectance and NDVI. As trace metals such as zinc can have detrimental effects on corals, we 248 recommend their regular monitoring in coral aquaculture systems. Finally, using zinc 249 supplementation without prior professional seawater analysis is not recommended. 250 251 Acknowledgements

252 This work was funded by Wageningen UR. We would like to thank the staff of Carus 253 experimental facility at Wageningen UR for technical support, and prof. dr. A.J. Murk for 254 providing constructive comments on a draft version of this manuscript.

255 256 References

257 Ahsanullah, M., 1976. Acute toxicity of cadmium and zinc to seven invertebrate species from 258 Western Port, Victoria. Australian Journal of Marine and Freshwater Research 27, 187–196.

259 Alia, Prasad, K.V.S.K., Pardha Saradhi, P., 1995. Effect of zinc on free radicals and proline in 260 Brassica and Cajanus . Phytochemistry 39, 45–47.

261 Atkinson, M.J., Bingman, C., 1997. Elemental composition of commercial seasalts. Journal of 262 Aquariculture and Aquatic Sciences 8, 39–43.

263 Balling, H.–W., 1996. Spurenelemente im Riffaquarium, Funktion, Festlegung, Ergänzung, 264 Part 3. DATZ 2, 108–111.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 265 Balling, H.–W., Janse, M., Sondervan, P.J., 2008. Trace elements, functions, sinks and 266 replenishment in reef aquaria. In: Leewis, R.J., Janse, M. (Eds.) Advances in Coral Husbandry 267 in Public Aquariums – Public Aquarium Husbandry Series 2, 143–156, Burgers’ Zoo, Arnhem, 268 The Netherlands. 444 p.

269 D’Angelo, C., Denzel, A., Vogt, A., Matz, M.V., Oswald, F., et al., 2008. Blue light regulation 270 of host pigment in reef-building corals. Marine Ecology Progress Series 364, 97–106.

271 Daniels, W.M.U., Hendricks, J., Salie, R., van Rensburg, S.J., 2004. A mechanism for zinc 272 toxicity in neuroblastoma cells. Metabolic Brain Disease 19, 79–88.

273 Davies, P.S., 1989. Short–term growth measurements of corals using an accurate buoyant 274 weighing technique. Marine Biology 101, 389–395.

275 Dove, S.G., Hoegh–Guldberg, O, Ranganathan, S., 2001. Major colour patterns of reef-building 276 corals are due to a family of GFP–like proteins. Coral Reefs 19, 197–204.

277 Ferrier-Pagès, C., Houlbrèque, F., Wyse, E., Richard, C., Allemand, D., Boisson, F., 2005. 278 Bioaccumulation of zinc in the scleractinian coral Stylophora pistillata . Coral Reefs 24, 636– 279 645.

280 Halldal, P., 1968. Photosynthetic capacities and photosynthetic action spectra of endozoic algae 281 of the massive coral Favia . Biological Bulletin 134, 411–424.

282 Harland, A.D., Bryan, G.W., Brown, B.E., 1990. Zinc and cadmium absorption in the symbiotic 283 anemone Anemonia viridis and the non-symbiotic anemone Actinia equina . Journal of the 284 Marine Biological Association of the UK 70, 789-802.

285 Harland, A.D., Nganro, N.R., 1990. Copper uptake by the sea anemone Anemonia viridis and 286 the role of zooxanthellae in metal regulation. Marine Biology 104, 297–301.

287 Heyward, A.J., 1988. Inhibitory effects of copper and zinc sulphates on fertilization in corals. 288 Proceedings of the 6th International Coral Reef Symposium 2, 299–303.

289 Houlbrèque, F., Rodolfo–Metalpa, R., Jeffree, R., Oberhänsli, F., Teyssié, J.–L., Boisson, F.,

290 Al–Trabeen, K., Ferrier–Pagès, C., 2012. Effects of increased pCO 2 on zinc uptake and 291 calcification in the tropical coral Stylophora pistillata . Coral Reefs 31, 101–109.

292 Jaap, W.C., 2000. Coral reef restoration. Ecological Engineering 15, 345–364.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 293 Jeffrey, S.W., Haxo, F.T., 1968. Photosynthetic pigments of symbiotic dinoflagellates 294 (zooxanthellae) from corals and clams. Biological Bulletin 135, 149–165.

295 Jones, R.J., 1997. Zooxanthellae loss as a bioassay for assessing stress in corals. Marine 296 Ecology Progress Series 149, 163–171.

297 Leal, M.C., Calado, R., Sheridan, C., Alimonti, A., Osinga, R., 2013. Coral aquaculture to 298 support drug discovery. Trends in Biotechnology 31, 555–561.

299 Leal, M.C., Ferrier–Pagès, C., Petersen, D., Osinga, R., 2014. Coral aquaculture: applying 300 scientific knowledge to ex situ production. Reviews in Aquaculture 6, 1–18.

301 Leal, M.C., Jesus, B., Ezequiel, J., Calado, R., Rocha, R.J.M., Cartaxana, P., Serodio, J., 2015. 302 Concurrent imaging of chlorophyll fluorescence, Chlorophyll a content and green fluorescent 303 proteins-like proteins of symbiotic cnidarians. Marine Ecology 36, 572–584.

304 Martin, M., Osborn, K.E., Billig, P., Clickstein, N., 1981. Toxicities of ten metals to 305 Crassostrea gigas and Mytilus edulis embryos and Cancer magister larvae. Marine Pollution 306 Bulletin 12, 305–308.

307 Matz, M.V., Marshall, N.J., Vorobyev, M., 2006. Symposium-in-Print: Green fluorescent 308 protein and homologs – Are corals colorful? Photochemistry and Photobiology 82, 345–350.

309 Nacci, D.E., Jackim, E., Walsh, R., 1986. Comparative evaluation of three rapid marine toxicity 310 tests: Sea urchin early embryo growth test, sea urchin sperm toxicity test and microtox. 311 Environmental Toxicology and Chemistry 5, 521–525.

312 Prasad, K.V.S.K., Paradha Saradhi, P., Sharmila, P., 1999. Concerted action of antioxidant 313 enzymes and curtailed growth under zinc toxicity in Brassica juncea . Environmental and 314 Experimental Botany 42, 1–10.

315 Reichelt–Brushett, A.J., Harrison, P.L., 1999. The Effect of Copper, Zinc and Cadmium on 316 Fertilization Success of Gametes from Scleractinian Reef Corals. Marine Pollution Bulletin 38, 317 182–187.

318 Reichelt–Brushett, A.J., Harrison, P.L., 2005. The effect of selected trace metals on the 319 fertilization success of several scleractinian coral species. Coral Reefs 24, 524–534.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 320 Rocha, J., Peixe, L., Gomes, N.C.M., Calado, R., 2011. Cnidarians as a source of new marine 321 bioactive compounds – an overview of the last decade and future steps for bioprospecting. 322 Marine Drugs 9, 1860–1886.

323 Rocha, R.J.M., Bontas, B., Cartaxana, P., Leal, M.C., Ferreira, J.M., Rosa, R., Serodio, J., 324 Calado, R., 2015. Development of a standardized modular system for experimental coral 325 culture. Journal of the World Aquaculture Society 46, 235–251.

326 Rocha, R.J.M., Calado, R., Cartaxana, P., Furtado, J., Serôdio, J., 2013a. Photobiology and 327 growth of leather coral Sarcophyton cf. glaucum fragments under low light in a recirculated 328 system. Aquaculture 414–415, 235–242.

329 Rocha, R.J.M., Pimentel, T., Serôdio, J., Rosa, R., Calado, R., 2013b. Comparative performance 330 of light emitting plasma (LEP) and light emitting diode (LED) in ex situ aquaculture of 331 scleractinian corals. Aquaculture 402–403, 38–45.

332 Rouse, J., Haas, R., Schell, J., Deering, D., 1973. Monitoring vegetation systems in the Great 333 Plains with ERTS. 3 rd ERTS–1 Symposium, Washington, DC: NASA SP–351, Greenbelt MD, 334 pp 309–317.

335 Schutter, M., Crocker, J., Paijmans, A., Janse, M., Osinga, R., Verreth, A.J., Wijffels, R.H., 336 2010. The effect of different flow regimes on the growth and metabolic rates of the scleractinian 337 coral Galaxea fascicularis . Coral Reefs 29, 737–748.

338 Schutter, M., Kranenbarg, S., Wijffels, R.H., Verreth, J., Osinga, R., 2011. Modification of light 339 utilization for skeletal growth by water flow in the scleractinian coral Galaxea fascicularis . 340 Marine Biology 158, 769–777.

341 Schutter, M., van Velthoven, B., Janse, M., Osinga, R., Janssen, M., Wijffels, R., Verreth, J., 342 2008. The effect of irradiance on long–term skeletal growth and net photosynthesis in Galaxea 343 fascicularis under four light conditions. Journal of Experimental Marine Biology and Ecology 344 367, 75–80.

345 Sondervan, P., 2004. The acute effects of zinc toxicity that possibly occurred during a routine 346 change of seawater at the Artis Aquarium. In: Bentivegna, F., Ounais, N., van den Sande, P., 347 (Eds.). Proceedings of the 19 th meeting of European Union of Aquarium Curators (E.U.A.C.), 348 Rotterdam, October 20 th –25 th , 2001. Bulletin de l’Institut Océanographique, Monaco 77(1445), 349 95–103.

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 350 Wijgerde, T., Henkemans, P., Osinga, R., 2012. Effects of irradiance and light spectrum on 351 growth of the scleractinian coral Galaxea fascicularis – applicability of LEP and LED lighting 352 to coral aquaculture. Aquaculture 344–349, 188–193.

353 Wijgerde, T., van Melis, A., Silva, C.I.F., Leal, M.C., Vogels, L., Mutter, C., Osinga, R., 2014. 354 Red Light Represses the Photophysiology of the Scleractinian Coral Stylophora pistillata . PLoS 355 ONE 9(3), e92781.

356 Wilmann, P.G., Petersen, J., Pettikiriarachchi, A., Buckle, A.M., Smith, S.C., Olsen, S., 357 Perugini, M.A., Devenish, R.J., Prescott, M., Rossjohn, J., 2005. The 2.1 Å Crystal structure of 358 the far-red fluorescent protein HcRed: Inherent conformational flexibility of the chromophore. 359 Journal of Molecular Biology 349, 223–237.

360 Wisely, B., Blick, R.A.P., 1967. Mortality of marine invertebrate larvae in mercury, copper, 361 and zinc solutions. Australian Journal of Marine and Freshwater Research 18, 63–72.

362 Young, C.N., Schopmeyer, S.A., Lirman, D., 2012. A review of reef restoration and coral 363 propagation using the threatened genus Acropora in the Caribbean and Western Atlantic. 364 Bulletin of Marine Science 88, 1075–1098.

365

366 Figure 1: Specific growth rates of Stylophora pistillata nubbins exposed to four different zinc 367 concentrations, ranging from 0 to 100 µg L –1, over two weeks. Values are means + s.d. ( N=4). 368 *At 100 µg L –1, corals grew significantly slower compared to the other zinc treatments 369 (P<0.001), irrespective of aquarium series. 370

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 371

372 Figure 2: Reflectance of Stylophora pistillata nubbins exposed to four different zinc 373 concentrations, ranging from 0 to 100 µg L –1, after two weeks. For clarity, mean values of 374 aquarium replicates have been pooled (N=4 measurements per coral, N=32 measurements total 375 per curve) and error bars have been omitted. 376 377

378 379 Figure 3: Normalised Difference Vegetation Index (NDVI) of Stylophora pistillata nubbins 380 exposed to four different zinc concentrations, ranging from 0 to 100 µg L–1, after two weeks. 381 Values are means + s.d. ( N=4 measurements per coral, N=16 per bar). *At 100 µg L –1, NDVI 382 was significantly lower compared to the other zinc treatments ( P<0.001), irrespective of 383 aquarium series. 384 385

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017 386 Table 1. Two–way factorial ANOVA, demonstrating main and interactive effects of zinc 387 concentration and aquarium on specific growth rates and Normalised Difference Vegetation 388 Index (NDVI) of Stylophora pistillata (N=4 per group).

Source of variation Variable F df error P Specific growth rate Zinc concentration 13.30 3 24 0.000 1 * Aquarium 0.698 1 24 0.412 Zinc concentration * Aquarium 2.559 3 24 0.081 NDVI Zinc concentration 18.58 3 24 0.000 3 * Aquarium 0.088 1 24 0.770 Zinc concentration * Aquarium 2.359 3 24 0.097 389 *Indicates significant effect ( P<0.050)

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2858v1 | CC BY 4.0 Open Access | rec: 9 Mar 2017, publ: 9 Mar 2017