Germanium dioxide as agent to control the biofouling diatom Fragilariopsis oceanica for the cultivation of Ulva fenestrata (Chlorophyta

: During the cultivation of Ulva fenestrata in a land-based aquaculture system, the colonisation of the water tanks ’ surfaces and eventually the macroalgal biomass by the biofouling diatom Fragilariopsis oceanica compromises the production process. Since germanium dioxide (GeO 2 ) is an e ﬀ ective growth inhibitor of diatoms, this study aimed to understand how it a ﬀ ects the presence of F. oceanica and the photosynthesis and growth of U. fenestrata as a primary parameter contribution to the biomass production. A toxicological dose-response experiment showed that the diatom ’ s growth was inhibited at the low GeO 2 concentration of 0.014 mg l − 1 . Incontrast,thephotosyntheticperformancesand growth rates of U. fenestrata remained una ﬀ ected under a wide GeO 2 concentration range (0.022 – 2.235 mg l − 1 ) in small-and large-scale experiments in 1-l glass beakers and 100-l Plexiglass water tanks, respectively. In the latter, the diatom density in the tanks was reduced by 40 %. The costs arising from the use of GeO 2 can range between € 2.35 and € 8.35 kg − 1 fresh weight of produced U. fenestrata biomass under growth conditions resulting in growth rates of 20 and 11.5 % d − 1 , respectively. GeO 2 is an e ﬀ ective agent to control biofouling diatoms such as F. oceanica during the land-based biomass production of U. fenestrata .


Introduction
Aquaculture is an important part of the world's food production.In 2019 macroalgal aquaculture accounted for 29 % (34.7 million tons) of the global aquaculture products.While China, Indonesia, South Korea, The Philippines and North Korea together produced 98 % of the macroalgal biomass, Europe only accounted for 0.03 % (Cai et al. 2021).In Norway, macroalgal aquaculture is a relatively young industry sector with a peak production of 336 t of Saccharina latissima and Alaria esculenta (Fiskeridirektoratet 2023).However, other species such as Ulva only play a negligible role despite ambitions to develop aquaculture concepts for them (Roleda et al. 2021;Stévant et al. 2017).
Strains of Ulva fenestrata from North Norway are rich in carbohydrates, dietary fibre, proteins and minerals, while being low in total lipids, fatty acids and iodine (Biancarosa et al. 2017;Rautenberger 2022;Roleda et al. 2021).Important vitamins (e.g.C, B 3 , B 12 ) and bioactive compounds (e.g.ulvan) could be also present (Holdt and Kraan 2011;MacArtain et al. 2007).Due to these commercially interesting properties, there is a wish to develop a year-round mass production of U. fenestrata through land-based aquaculture (Roleda et al. 2021).However, cultivation studies have repeatedly revealed issues with biofouling by the pennate diatom Fragilariopsis oceanica (Cleve) Hasle.This diatom is pumped up into the laboratory together with the deep-seawater from the nearby Saltfjord.Insufficient filtration and sterilisation capacities of the high flow-through seawater treatment system allow them to pass through the 5 µm filters.Eventually, F. oceanica is flushed into the water tanks and grows rapidly into long and dense filaments under the cultivation conditions used for U. fenestrata (Figure 1).At first, they colonise the surfaces of the water tanks before speading to Ulva.Small brown spots appear first on the Ulva's surfaces, which eventually grow into a dense filaments.
The biofouling by F. oceanica not only affects the growth of U. fenestrata and reduces the biomass quality, it also increases the production costs.The shading of the macroalgal thalli by the diatom's dense colonisations on the water tank's surfaces may reduce light-dependent physiological processes such as photosynthesis, which, eventually, diminishes growth.In addition, there is a competition between the diatoms and U. fenestrata for the nutrients dissolved in the seawater.
Therefore, the presence of F. oceanica needs to be effectively controlled.
Regular cleaning of the water tanks' surfaces by hand is a time-consuming and, thus, an expensive work task.Although artificial seawater does not contain any biofouling organisms, its preparation increases production costs of U. fenestrata substantially.A possibly less expensive solution could be the treatment of F. oceanica with germanium dioxide (GeO 2 ).This cytotoxin inhibits the growth of numerous diatom species by interferring with the silica frustule formation (Lewin 1966).Fragilariopsis oceanica was reported to have silicified frustules consisting of different silicification degrees between valves, mantles and bands (Lundholm and Hasle 2010).Although GeO 2 is routinely used to control the contamination of cultures by diatoms, there is still limited information of its effects on green macroalgae such as Ulva.The few studies in which Ulva was treated with GeO 2 showed inconsistent results.While the growth of U. fenestrata from Helgoland, Germany, remained unaffected up to 2.2 mg GeO 2 l −1 , Ulva australis from Japan was insensitive up to 30 mg GeO 2 l −1 (Markham and Hagmeier 1982;Tatewaki and Mizuno 1979).Therefore, this study was conducted to understand the impact of GeO 2 (1) on the growth of biofouling diatom F. oceanica and (2) on the growth and photosynthesis of U. fenestrata as primary physiological processes influencing the macroalga's biomass production.While the growth of F. oceanica is expected to be impaired by GeO 2 due to the competition with silicate, growth and photosynthesis of U. fenestrata remain unaffected because neither germanium nor silicate are essential for the green macroalga.The knowledge gained from this study is useful for the contamination control of land-based aquaculture of Ulva.

Materials and methods
Ulva fenestrata Postels et Ruprecht was collected from the intertidal in Bodø (67.2759°N, 14.5706°E), Norway, in July 2019 (Hughey et al. 2019).In the laboratory, the collected thalli were cleaned and kept in 600-l water tanks (aeration from the bottom) at 130 ± 5 µmol photon m −2 s −1 (L36W/954 Lumilux de Lux Daylight, Osram GmbH, Munich, Germany) and 9 ± 1 °C for 7 months prior to the experiments.The deep-seawater (250 m water depth) came from the nearby Saltfjord and was fed into the water tanks with a flow rate of 25 l min −1 .The nitrate and phosphate concentrations were 9.75-10.26µM and 0.55-0.91µM, respectively.The ammonium concentration was low (0.01-0.10 µM).
For the toxicological dose-response experiment with F. oceanica, 21 1 l glass beakers (10 % HCl-washed) were filled with 0.8 l of 5 µmfiltered, ESNW-enriched deep-seawater (Berges et al. 2001).NaSiO 3 was omitted because F. oceanica grows in the laboratory without added silicate.Saturated GeO 2 stock solutions were prepared by mixing 1.11735 g of GeO 2 (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) with 0.25 l of sterile-filtered (0.2 µm) seawater, stored at 4 °C in the dark.The mixture was thoroughly homogenised before use (Shea and Chopin 2007).GeO 2 was added to the seawater medium to reach 6 final concentrations between 0.003 and 1.397 mg GeO 2 l −1 , while GeO 2 -free seawater was used as control (0 mg GeO 2 l −1 ).Then F. oceanica was allowed to grow at 137 ± 5 µmol photon m −2 s −1 at 9 ± 1 °C (L36W/954 Lumilux de Lux Daylight; Osram GmbH, Munich, Germany) under aeration from the bottom of the beakers.The seawater was not replenished during the experiment to disturb the growth of the diatoms.The experiment was finished after 22 days when a significant diatom biomass was visible on the beakers' surfaces.This diatom biomass was carefully removed and filtered through previously heated (105 °C), weighed (at room temperature) and washed (0.5 M NH 4 HCO 3 ) 1.0 µm glass microfibre filters (Number 692; VWR International, Leuven, Belgium).Then the filters were washed twice with 20 ml of 0.5 M NH 4 HCO 3 and dried at 105 °C for 24 h.The filters with and without algal biomass were cooled at room temperature in a desiccator.The diatom's dry weights (DW) were measured using an analytical balance with 0.1 mg precision (VWR International, Leuven, Belgium).The volume-based DW of the diatom biomass in the beakers was calculated as follows: where DW +algae and DW −algae are the filters' dry weights with and without diatom biomass, respectively, and V is the volume of the seawater medium in the beakers.For the small-scale cultivation of U. fenestrata with GeO 2 , 12 2 l flasks (10 % HCl-washed) were filled with 1.8 l of 0.45 µm-filtered, ESNW-enriched deep-seawater (no NaSiO 3 added), which was aerated from the bottom and changed daily.GeO 2 was added to the seawater to treat U. fenestrata with three GeO 2 concentrations (0.022-2.235 mg l −1 ), while GeO 2 -free seawater was used as control.Then 12 individuals of U. fenestrata were randomly taken from the stock culture and three algal discs (17 mm diameter) were cut from each individual using a stainless steel cork-borer.The three discs from the same individual were randomly assigned to a flask and exposed to 140 ± 5 µmol photons m −2 s −1 (16 h light:8 h dark, L36W/954 Lumilux de Lux Daylight; Osram GmbH, Munich, Germany) at 9 ± 1 °C for 5 days.
For the large-scale cultivation of U. fenestrata with GeO 2 , nine Plexiglass water tanks were filled with 63 l of deep-seawater and enriched with ESNW medium (no NaSiO 3 added).The seawater was aerated from the bottom of the tanks and not replenished during the experiment.GeO 2 was added to the seawater to treat U. fenestrata with two GeO 2 concentrations (0.223 and 2.235 mg l −1 ), while GeO 2 -free seawater was used as control.The two GeO 2 concentrations were used because they are expected to cause the largest effects on the physiology of U. fenestrata.Afterwards, individuals of U. fenestrata were randomly taken from the stock culture and cut back to a length of 5 cm above the discoid holdfast (or "attachment disc").Then 3-5 individuals were placed in a water tank to start the experiment with an initial FW of 4-5 g.The macroalgae were cultivated at 140 ± 5 µmol photons m −2 s −1 (16 h light:8 h dark, L18W/954 Lumilux de Lux Daylight; Osram GmbH, Munich, Germany) and 12 ± 1 °C.
The changes in FW of all macroalgal samples in a flask or a water tank during the experiments were determined gravimetrically.The daily relative growth rates (RGR) were analysed by comparing the initial and final FWs and calculated according to (Lüning 1990).
The photosynthetic performances of U. fenestrata were measured on randomly taken samples using PAM-fluorometry (Diving-PAM, Walz GmbH, Effeltrich, Germany) at 10 ± 1 °C at the end of the experiments (Rautenberger et al. 2015).After 5 min of dark adaptation, U. fenestrata was exposed to incrementally increasing actinic irradiances (E AL ) after short saturation pulses (>9000 µmol photons m −2 s −1 , 0.8 s) every 30 s to determine the effective PSII-quantum yields [Fv′/Fm′ = (Fm′ − F t )/Fm′].Electron transport rates (ETRs) were calculated by multiplying Fv′/Fm′ by E AL , the proportion of E AL absorbed by an Ulva disc ('thallus absorbance'), and the fraction of absorbed light that was most probably received by PSII (Baker 2008;Figueroa et al. 2009;Lüning and Dring 1985): The maximum ETRs (ETR max ), the light saturation points of photosynthesis (E k ), and the initial slopes of light curves (α ETR ) as photosynthetic parameters were estimated from ETR-E curves, which were fitted after Eilers and Peeters (1988).
The samples for the pigment analyses were taken at the end of the two experiments.All macroalgae in each flask or water tank were frozen together and stored at −80 °C until analysis.Chlorophyll a (Chl a) and b (Chl b) were extracted from the frozen samples (26-36 mg FW) in 5 ml of 100 % N-N-dimethylformamide (DMF) in the dark at 4 °C for 72 h.The pigments were analysed spectrophotometrically (UviLine 9400, SCHOTT Instruments GmbH, Germany) from spectra (350-750 nm) with 1.0 nm resolution at 647 and 664 nm (20 °C) with 100 % DMF as reference.Readings at 750 nm were used as correction factors for scattered light.The Chl a, b and total Chl (Chl a + b) contents were calculated after Porra et al. (1989) and normalised on macroalgal FW.
All experiments were designed such that each sample was independent.Means and standard deviations (SD) were calculated from three replicates per GeO 2 concentration.The Shapiro-Wilk test and the Levene test were used to test the normal distribution of residuals and the homoscedasticity of the primary data, respectively.When these assumptions were met, one-way ANOVAs were performed to identify statistical differences of means between the treatments with Tukey's honestly significant difference (HSD) as post hoc test.In the case of heteroscedasticity, Welch-ANOVAs were performed with Games-Howel post-hoc tests.The statistical analyses were performed with a 5 % significance level (P = 0.05) using JMP 14.0 (SAS Institute Inc., Cary, NC, USA) and R version 4.0 (The R Foundation for Statistical Computing, http:// www.R-project.org).

Dose-response relationship in
Fragilariopsis oceanica After 22 days of seawater cultivation in 1 l glass beakers, brown biofilms of Fragilariopsis oceanica became clearly visible on the bottom and walls of the beakers of both the control (0 mg l −1 ) and the lowest GeO 2 concentration (0.003 mg l −1 ) with 27.13 mg DW m −2 and 24.23 mg DW m −2 , respectively.In contrast, all GeO 2 concentrations ≥0.014 mg l −1 had lower diatom biomass (6.56 mg DW m −2 ) than the control and 0.003 mg GeO 2 l −1 (P < 0.0001, 1-way ANOVA; Figure 2).

Small-and large-scale experiments with
Ulva fenestrata treated with GeO 2 In the small-scale experiment, the RGR and all three photosynthetic parameters of Ulva fenestrata were statistically similar between the control (0 mg GeO 2 l −1 ) and the 3 GeO 2 concentrations tested between 0.022 and 2.235 mg l −1 (Figure 3).In addition, the contents of Chl a and Chl b in U. fenestrata remained unaffected by the presence of different GeO 2 concentrations in the seawater (Table 1).However, there was slight decrease in the Chl a/b ratio of U. fenestrata from 1.61 ± 0.08 in the control by 10-15 % in the presence of GeO 2 (P = 0.0208, 1-way ANOVA; Table 1).
In the large-scale experiment, the RGR, all three photosynthetic parameters, and the chlorophyll contents, including the Chl a/b ratio of U. fenestrata, were statistically similar between the control and the two tested GeO 2 concentrations after 14 days of cultivation (Figure 4 and Table 1).However, the addition of GeO 2 to the seawater decreased the density of F. oceanica on the wall surfaces of the Plexiglass water tanks by 36-43 % at 0.223-2.235mg GeO 2 l −1 compared to the control (P = 0.0077, 1-way ANOVA; Figure 5).

Discussion
In hatcheries and cultivation systems (e.g.land-based aquaculture) of macroalgae, biofouling through unwanted contamination by diatoms can cause serious damage.Therefore, GeO 2 is often used to control biofouling diatoms.This study demonstrated that GeO 2 controls the growth of Fragilariopsis oceanica, while both the growth and the photosynthesis of Ulva fenestrata remain unaffected.

Effects of GeO 2 on the growth of Fragilariopsis oceanica
GeO 2 is widely used as an effective growth inhibitor for diatoms and other silicifying microalgae (e.g.chrysophytes).It competitively inhibits the uptake of Si through Si-transporters (SITs), which eventually interrupts cell wall formation in the algae (Thamatrakoln and Hildebrand 2008).Table : Contents of photosynthetic and accessory pigments of Ulva fenestrata after  days ("small-scale cultivation") and  days ("large-scale cultivation") of cultivation at - GeO  concentrations.
Small-scale cultivation of Ulva fenestrata in glass beakers (after  days) Large-scale cultivation of Ulva fenestrata in Plexiglass water tanks (after  days) Toxicological studies have shown species-and strain-specific responses of diatoms to GeO 2 .While concentrations of up to 1 mg GeO 2 l −1 inhibited the growth of highly silicified diatom species (e.g.Amphiphora paludosa, Cylindrotheca fusiformis), diatoms with a low degree of silicified cell walls (e.g.Phaeodactylum tricornutum) were insensitive even to 10 GeO 2 mg l −1 (Lewin 1966;Markham and Hagmeier 1982;Tatewaki and Mizuno 1979).By using these results as benchmark for the present study, F. oceanica seems to be highly sensitive to GeO 2 because its growth was inhibited at a low concentration of 0.014 mg GeO 2 l −1 .Assuming that Si uptake in F. oceanica is mediated by SITs at low Si concentrations in seawater (<30 µM) as shown for Thalassiosira pseudonana (Thamatrakoln and Hildebrand 2008), a growth inhibition by GeO 2 could have been expected because the seawater Si concentration used was approx.2.1 µM (Busch et al. 2014).However, if Si was added to the seawater according to formulation of the ESNW medium with a final concentration of 106 µM, a different result would have been observed because diffusive Si uptake predominates at higher Si concentrations (Thamatrakoln and Hildebrand 2008).Interestingly, the use of GeO 2 in the glass beakers and Plexiglass water tanks showed different effects on F. oceanica.While low GeO 2 concentration inhibited the growth of F. oceanica, the colonisation of the Plexiglass walls by F. oceanica could not be fully prevented even by the use of 2.235 mg GeO 2 l −1 .This could be possibly ascribed to the different physical surface properties between glass and Plexiglass.Insoluble extracellular polymeric substances (EPS) secreted by diatoms allow them to adsorb better on hydrophobic surfaces such as Plexiglass than on the hydrophilic surfaces of the glass beakers (Finlay et al. 2013;Holland et al. 2004;Krishnan et al. 2006; reviewed by Thompson and Coates 2017).In addition, EPS-rich biofilms from Roseobacter and Sulfitobacter, which are associated with Ulva, could have enhanced the attachment of F. oceanica on the Plexiglass walls (Bruckner et al. 2011;Buhmann et al. 2016;Spoerner et al. 2012).Thus, the bacterial biofilms could help F. oceanica to overcome the negative effects of GeO 2 .Other factors such as the different treatments of the glass beakers (acid-washed) and water tanks (hand washing and sodium hypochlorite) prior to the experiments could also have contributed to the different outcomes of the two experiments.Nevertheless, since the large-scale experiment showed a considerable reduction in diatom density on the Plexiglass surfaces by 0.223 mg GeO 2 l −1 (f.c.), the costs for the biomass production of U. fenestrata are expected to be lower than without employing any GeO 2 at all.
4.2 Effects of GeO 2 on the growth and photophysiology of Ulva fenestrata Since there are toxic effects of GeO 2 on F. oceanica, it is crucial to understand its implications for U. fenestrata.This study demonstrated that GeO 2 concentrations of up to 2.235 mg l −1 had no adverse effect on growth and photosynthesis of U. fenestrata.The capture and transfer of light energy through photosynthetic and accessory pigments, as well as the photosynthetic activity, remained unaffected by GeO 2 at different cultivation scales (2-l flasks and 70-l water tanks).Consequently, the photosynthetically produced carbon could be allocated to macroalgal growth.Because photosynthesis also interacts with nitrogen metabolism, the constant growth rates imply that nitrogen assimilation remained unaffected by GeO 2 even though this was not tested (Huppe and Turpin 1994).
Constant growth and photosynthetic performance suggest that U. fenestrata has neither a GeO 2 -sensitive uptake mechanism nor it is limited by Si, indicating that Si is not essential for the green macroalga.This is in line with a previous study showing that the growth of U. fenestrata (studied as U. lactuca) is unaffected up to 2.2 mg GeO 2 l −1 (Markham and Hagmeier 1982).However, this macroalga's growth was reported to be inhibited at higher concentrations of GeO 2 (4.4-8.95mg GeO 2 l −1 ).In contrast, the growth of Ulva australis (studied as Ulva pertusa) and Monostroma angicava was not inhibited up to 30 mg GeO 2 l −1 (Markham and Hagmeier 1982;Tatewaki and Mizuno 1979).GeO 2 seems to be a safe anti-fouling agent for the biomass production of U. fenestrata and other green macroalgae.Nevertheless, its effects on the development of the life-cycle stages have yet to be studied because 0.5-1.0mg GeO 2 l −1 are often used to prevent diatoms from growing during these stages (Lotze et al. 2000;Marshall et al. 2006;Wang et al. 2012).
When GeO 2 is used during the production of U. fenestrata for human and animal consumption, knowledge about the bioaccumulation of GeO 2 is crucial to understand its safety as a component of human food and animal feed.So far, neither the European Food Safety Authority (EFSA) nor its North American counterparts have any regulations on the food safety of germanium (Ge).To the author's knowledge, there are no studies available on the bioaccumulation of GeO 2 in U. fenestrata and other macroalgae.Nevertheless, U. fenestrata may accumulate Ge due to the treatment with GeO 2 because this has been detected across taxonomic borders: bacteria, cyanobacteria, microalgae and higher plants (Cheong et al. 2009;Chmielowski and Kłapcińska 1986;Choi et al. 2013;Djur 2020;Yanagimoto et al. 1983).In the green microalga Chloroidium ellipsoideum (studied as Chlorella ellipsoidea), the bioaccumulation of Ge was lowest under optimal pH conditions (6.0-7.4) for growth while it increased 8-fold at a high pH (>9), which inhibited growth (Yanagimoto et al. 1983).Since U. fenestrata was cultivated under optimal conditions giving high growth rates, the Ge-content is assumed to be low.Because inorganic Ge-containing compounds, including GeO 2 , have low toxicity with acute LD 50 values of 500-5000 mg kg −1 (oral application), the assumed low Ge-contents in the tissue of U. fenestrata could be considered as food-safe (Gerber and Léonard 1997).Since single oral doses (2000 mg Ge kg −1 body weight) of Ge-fortified lettuce did not result in physiological and pathological changes in mice, the bioaccumulated Ge was considered to be safe as food and feed (Kim et al. 2009).However, whether GeO 2 -treated biomass of U. fenestrata is actually a concern for food and feed safety still needs to be determined after we have a better understanding of the bioaccumulation of Ge in Ulva and other macroalgae.

Economic considerations of the use of GeO 2
For land-based aquaculture of U. fenestrata, it was important to understand if the use of a given GeO 2 concentration is economically reasonable.Since GeO 2 can be currently purchased for an average price of €18.69 g −1 from two main laboratory suppliers in Norway, the preparation of 250 ml of saturated GeO 2 stock solution costs €20.89.Thus, the use of 5 ml of the GeO 2 stock solution for a final concentration of 0.223 mg GeO 2 l −1 seawater resulting in a 40 %-growth inhibition of F. oceanica in a 100-l Plexiglass water tank would cost €0.42.Assuming the average RGR of 11.5 % d −1 (this study; over 14 days) with 10 g of initial FW, the cost of adding GeO 2 would be €8.35kg −1 FW of U. fenestrata produced.However, these costs for GeO 2 decline by 70 % to €2.53 kg −1 FW of Ulva with a higher RGR of 20 % d −1 , which is not unusual for Ulva under optimal growth conditions.Consequently, the 250 ml of GeO 2 stock solution costing €20.89 can be used for a production of 2.5 and 8.25 kg FW of Ulva under growth conditions resulting in 11.5 and 20 % d −1 , respectively.However, the costs arising from the use of GeO 2 in other laboratories depend on the species-specific toxicological effects of GeO 2 on predominant diatoms in the cultivation systems used for Ulva or other macroalgal species.While controlling highly silicified diatoms involves lower costs due to the use of lower GeO 2 concentrations, less sensitive species may require higher expenditure for GeO 2 .In addition, operational costs are determined by the effects of GeO 2 on the growth and development of the macroalgal species being cultivated.In addition to Ulva, GeO 2 concentrations need to be found for possibly more sensitive red and brown macroalgae that do not cause adverse effects on the macroalgae but keep biofouling diatom densities low.

Conclusions
GeO 2 is an effective agent for controlling the presence of biofouling diatoms in aquaculture systems and hatcheries.
It can be used to control the contamination of cultures of U. fenestrata by F. oceanica without compromising the macroalga's biomass production.However, studies are needed on the food and feed safety due to the bioaccumulation of Ge when Ulva is produced for human and animal consumption.The concentrations of GeO 2 employed to control biofouling diatoms need to be adjusted to the specific diatom species occurring at an aquaculture facility.
In either case, the use of GeO 2 seems to keep the high quality of the produced Ulva biomass, while reducing the workload and production costs.

Figure 1 :
Figure 1: Photographs of the water tanks used for the cultivation of Ulva fenestrata in the laboratory while being (A) moderately and (B) strongly colonised by Fragilariopsis oceanica.

Figure 2 :
Figure 2: Total diatom biomass at different GeO 2 concentrations after 22 days of cultivation at 137 µmol photons m −2 s −1 and 9 °C.Data are means of three replicates per treatment (n = 3) and error bars represent standard deviations.Lowercase letters above columns indicate statistically significant differences between the treatments (P < 0.001, 1-way ANOVA, Tukey-Kramer HSD post-hoc test).

Figure 3 :
photons)  a Fresh weight (FW)-based contents of chlorophyll a (Chl a), chlorophyll b (Chl b), total chlorophylls (Chl a + b) and the Chl a-to-b ratio (Chl a/b).Data are means ± standard deviations of three replicates per treatment (n = ).Different lowercase letters behind the data in each column indicate statistically significant differences of the pigments between the GeO  concentrations (P < ., -way ANOVA, Tukey-Kramer HSD post-hoc test).