Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

Contents lists available at ScienceDirect

Journal of Experimental Marine Biology and Ecology

j ourna l homepage: www.e lsev ie r .com/ locate / jembe

Simultaneous determination of coenzyme Q and plastoquinone redoxstates in the coral–Symbiodinium symbiosis during thermallyinduced bleaching

Adrian Lutz a,b,c,d,⁎, Cherie A. Motti b, Marnie L. Freckelton a,b,c, Madeleine J.H. van Oppen b,d,David J. Miller c,d, Walter C. Dunlap b

a AIMS@JCU, James Cook University, Townsville, Queensland 4811, Australiab Australian Institute of Marine Science, PMB No. 3, Townsville MC, Queensland 4810, Australiac School of Pharmacy and Molecular Sciences, James Cook University, Townsville, Queensland 4811, Australiad ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Queensland 4811, Australia

⁎ Corresponding author at: School of Pharmacy and MUniversity, Townsville, Queensland 4811, Australia. Tel.: +

E-mail address: adrian.lutz@my.jcu.edu.au (A. Lutz).

http://dx.doi.org/10.1016/j.jembe.2014.02.0060022-0981/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 1 September 2013Received in revised form 8 February 2014Accepted 10 February 2014Available online 13 March 2014

Keywords:Coenzyme QCoral bleachingLC–MSMass spectrometryOxidative stressPlastoquinone

Coenzyme Q (CoQ; ubiquinone) and plastoquinone (PQ) are metabolic electron carriers that, in their reducedstate, are powerful antioxidants for cellular protection against oxidative damage. Although damage resultingfrom generation of reactive oxygen species (ROS) is strongly implicated in the initiation of symbiotic dysfunctionthat leads to coral bleaching, very little is known about the redox state of these two prenylquinone pools duringthe stress response. Here we describe a quantitative liquid chromatography–mass spectrometry (LC–MS) meth-od that permits simultaneous measurement of the physiological redox state of both CoQ and PQ in whole corals.The application of this method indicates that the CoQ and PQ pools in the coral–Symbiodinium symbiosis aremaintained predominantly in their reduced (antioxidant) forms, and it is the coral CoQ redox state that ismost affected by acute thermal stress.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

There is growing concern over increasing ocean temperatures thatthreaten the health of coral reefs by disrupting the mutualistic partner-ship between reef-building corals (Cnidaria: Scleractinia) and theirdinoflagellate (Symbiodinium sp.) partners causing coral bleaching(Hoegh-Guldberg and Bruno, 2010). Coral bleaching (loss of algal sym-bionts and/or pigments), which can affect entire reef systems over largeareas (“mass-bleaching” events), has been consistently linked to highsolar irradiance in combination with elevated maximum temperaturesattributed to global climate change (Eakin et al., 2009). The consensusis that coral bleaching is a response to acute oxidative stress wherebyexcessive levels of reactive oxygen species (ROS) overwhelm the anti-oxidant defence capacity of the symbiosis (Lesser, 2011; Weis, 2008).Regardless of the primary impact sites of damage, coral bleaching is at-tributed to ROS formation by the electron transport chains (ETC) of coralmitochondria and Symbiodinium chloroplasts (Buxton et al., 2012;Weis,

olecular Sciences, James Cook61 747 81 60 09.

2008). In addition to the indispensable roles of coenzyme Q (CoQ; ubi-quinone) and plastoquinone (PQ) in electron transport for ATPproduction, their reduced forms ubiquinol (CoQH2) and plastoquinol(PQH2) have an important antioxidant function within mitochondrial(Ernster and Forsmark-Andrée, 1993), cellular (Bentinger et al., 2007)and thylakoid (Nowicka and Kruk, 2012) membranes. The CoQ poolredox state has been used as a sensitive plasma biomarker of oxidativestress in human disease (Yamamoto et al., 1998) ageing (Wada et al.,2007), and has been demonstrated to play a role in the hepatic responseof fish exposed to polycyclic aromatic hydrocarbons (Hasbi et al., 2011).To investigate the significance of host and symbiont redox poise duringbleaching, an analytical procedure was developed for simultaneousanalysis of the CoQ and PQ pool redox states in corals. While similarmethods have been reported for the determination of CoQ and PQ inthe chloroplasts of Arabidopsis thaliana (Martinis et al., 2011; Yoshidaet al., 2010), high performance liquid chromatography mass-spectrometry (LC–MS)methods applicable to invertebrate-algal symbi-osis have not previously been available. In addition, as a proof of con-cept, the novel method described here, is used to monitor changes inthe redox states of host CoQ and symbiont PQ in the scleractiniancoral Acropora millepora (Ehrenberg, 1834) during bleaching inducedby experimental thermal stress.

2 A. Lutz et al. / Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

2. Materials and methods

2.1. Reagents

Ubiquinone-9 (CoQ9), ubiquinone-10 (CoQ10), sodium borohydride,EDTA, formic acid, methanol (MeOH), ethanol (EtOH), isopropanol(IPA), hexane and ethyl acetate (EtOAc) were purchased from SigmaAldrich (St. Louis, MO, USA). All solvents used were HPLC grade.Plastoquinone-9 (PQ9) was a kind gift of Professor Ewa Swiezewskafrom the Polish Academy of Sciences, Poland.

2.2. Collection and preparation of coral and Symbiodinium samples

A. millepora colonies roughly 50 cm in diameter containing type C2Symbiodinium (ITS1 terminology, see below) were collected fromPelorus Island, Great Barrier Reef, Australia (18°33′ S/146°29′ E). Priorto experimentation, A. millepora colonies were maintained in outdooraquaria at the Australian Institute of Marine Science, Townsville,Australia, with flow-through filtered (5 μm) seawater under 70%shading (maximum 250 μmol photons m−2 s−1) and ambient temper-ature (26–27 °C) replicating the conditions of the collection site.Symbiodinium type C1 was isolated from Acropora tenuis (MagneticIsland; 19°10′ S, 146°51′ E) and Symbiodinium type C2 from A. millepora(Pelorus Island; 18°33′ S, 146°29′ E) as described by Shick et al. (2011).Cultures of Symbiodinium type C1 were grown in f/2 medium preparedin filtered, autoclaved seawater (Guillard and Ryther, 1962) and main-tained under an irradiance of 40–60 μmol photons m−2 s−1 (12:12 hlight:dark; 17,000 K/420 nm fluorescent lights) at 26 °C. TheSymbiodinium genotype was identified based on sequence differences inthe nuclear ribosomal DNA internal transcribed spacer 1 (ITS1) regionusing single-strand conformation polymorphism (SSCP) analysis as de-scribed by van Oppen et al. (2001). Total coral and Symbiodinium DNAwas extracted using a modified protocol (Wilson et al., 2002) and thesymbiont ITS1 region amplified with fluorescently labelled Sym ITS1PCRprimers. The symbiont genotypewas determined using known refer-ence samples for comparison and scored manually using gel images(Fabricius et al., 2004).

2.3. Heat stress experiment

For the hyperthermal stress experiment, A. millepora nubbins wereremoved from five coral colonies (3–4 cm in length; for prenylquinoneextraction: n = 8 per sample point per treatment; for chlorophyllfluorescence: n = 42 per treatment). Coral nubbins were suspendedin aquaria by hanging them from acrylic glass rods with nylon fishingline, andwere allowed to acclimate for twoweeks at controlled ambientconditions (27 °C; 250 μmol photonm−2 s−1; 13:11 h light:dark) priorto experimental treatment. Coral nubbins were distributed in four rep-licate tanks (30 nubbins per tank) in a randomised block design. Alltanks were continuously supplied with fresh, 1 μm filtered seawater ata rate of 1.5 L min−1 from 500 L reservoirs in a temperature-controlled room maintained at 27 ± 0.5 °C. Each reservoir was heatedwith two titanium heaters (3 kW) controlled by a CR1000 datalogger(Campbell Scientific) and temperature sensors in the treatment tanks.All tanks were fitted with a small power head pump to maintain watermovement and an air stone and pump to provide aeration. UV-filtered400 W metal halide lights (BLV, Germany) were mounted aboveeach tank and provided an average underwater light intensity of350 μmol photons m−2 s−1 (13:11 h light:dark). The UV-filterswere used to minimise UV-radiation induced bleaching (Gleasonand Wellington, 1993).

The stress experiment was conducted at three temperatures (ambi-ent [low]: 27 °C, mid: 31 °C and high: 34 °C) with four replicate tanksfor each temperature. Seawater in the two higher temperature treat-ment tanks was heated to 31 ± 0.5 °C and 34 ± 0.5 °C over 3 h whilecontrol tanks were maintained at 27 ± 0.5 °C. One hour after reaching

treatment temperatures, lights were switched on for a 13 hour pho-toperiod. After the photoperiod, pulse amplitude modulated (PAM)chlorophyll fluorescence measurements were taken and seawatertemperatures lowered to ambient (Fig. S1). Coral nubbins were col-lected for prenylquinone analysis: pre-exposure (t=−6 h; in dark-ness, Fig. S1); upon reaching treatment temperatures (t = −1 h; indarkness); 4 h (t = 4 h) and 12 h during light exposure (t = 12 h)and after a recovery period of 8 h in darkness (t = 24 h). Coral nub-bins were immediately snap-frozen at collection and stored in liquidnitrogen for a maximum of 48 h prior to prenylquinone extraction.CoQ and PQ pool redox states were determined by LC–MS quantifica-tion using a Phenomenex Gemini C18 HPLC column. Absolute quan-tities of the prenylquinones were calculated from calibration plotsobtained from standard compounds containing 0.1 μM CoQ9 (inter-nal standard). CoQ and PQ redox states (%PQH2; %CoQH2) wereexpressed as the proportion of reduced to total (oxidised + reduced)prenylquinone.

2.4. LC–MS equipment and conditions

Coral extract separation was performed by reverse-phase high-performance liquid chromatography (RP-HPLC) on an Agilent 1100series HPLC (Agilent, USA) comprising of a solvent degasser, a bina-ry pump, a thermostated column compartment, and a Gilson 215autosampler (USA) equipped with a 20 μL injection loop. For quan-titative measurements the injection loop was overfilled by five-foldexcess of the loop volume. Prenylquinones and quinols were separatedon a Phenomenex Gemini C18 column (150mm× 4.6mm, 3 μmparticlesize) maintained at 28 °C. Mobile phase A: 0.1% formic acid in MeOH;mobile phase B: 0.1% formic acid in IPA. Flow rate: 1mLmin−1. The pro-grammed step gradientwas: 10% to 20%B over 15min, 20% to 40% B over13 min, followed by a clean-up step: isocratic elution at 40% B for 7 min,40% to 10% B over 3 min and column wash for 4 min.

The HPLC system was coupled to a Bruker Esquire 3000 (BrukerDaltonics, USA) quadrupole ion trap mass spectrometer (LC–MS)equipped with an atmospheric pressure chemical ionisation inter-face (APCI) operating in positive mode. The capillary voltage wasset at 3000 V, the end plate offset at −500 V and the corona chargeat 4 μA. The nebuliser pressure was 30 psi and drying gas flow at5 L min−1. The drying gas temperature was maintained at 300 °Cand the vaporiser at 350 °C. For quantitative analysis the ion trapwas operated in single ion monitoring (SIM) mode alternatelyacquiring (oxidised) parent ions at m/z = 749.6; m/z = 795.5;m/z = 863.7. Each parent ion was isolated using an isolation widthof four mass units and a maximum accumulation time of 200 mswith averaging of four mass spectra to enhance signal-to-noiseratio and signal stability.

2.5. Quantification of coenzyme Q and plastoquinone pool redox states

Absolute quantities of the prenylquinones were calculated from cal-ibration plots obtained from standard compounds containing 0.3 μMCoQ9 (internal standard). The concentrations of the three calibrationpoints for CoQ10, ubiquinol-10 (CoQ10H2), PQ9 and plastoquinol-9(PQ9H2) were 0.03, 0.3 and 3.0 μM. Three replicate points were mea-sured for each concentration using peak area for calibration of the con-centration for each compound.Mid-level calibration standards (0.3 μM)were analysed at the beginning and end of each day. The calibrationplots were repeated if the difference in the relative response factor ascompared to the initial calibration was more than 15% for any analyte.Peak areas were determined manually for all chromatograms afterapplying a Gaussian smoothing filter (1 point). CoQ and PQ poolredox states were expressed as the percent ratio of reduced to total CoQand PQ, respectively, e.g. %CoQH2 = 100 × [CoQ10H2] × ([CoQ10] +[CoQ10H2])−1.

3A. Lutz et al. / Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

2.6. Preparation of standards

Stock solutions of 50 μM CoQ9, CoQ10 and PQ9 were prepared inEtOH. CoQ10H2 and PQ9H2 standards were prepared by reduction withsodium borohydride (Yamamoto and Yamashita, 1997). Methodswere adapted to the lower concentrations of the standard solutions re-quired. In brief: 1 mg CoQ10 or PQ9 was dissolved in 2 mL hexane towhich 25 mg of sodium borohydride (dissolved in 50 μl MeOH) wasadded. The solution was mixed for 3 min and set at room temperaturefor 5min. EDTA (1mL, 100 μM)was added, the reactionmixture shakenand the hexane phase removed after brief centrifugation. The EDTAwash was repeated once and the combined hexane phases driedunder a nitrogen stream. The ensuing prenylquinols were re-dissolvedin 1.5 mL EtOH and exact concentrations of stock solutions determinedon a UV-2550 spectrophotometer (Shimadzu, Japan) using ε275 =14.24 mM−1 cm−1 for CoQ10; ε275 = 14.7 mM−1 cm− 1 for CoQ9;ε290 = 4.01 mM−1 cm−1 for CoQ10H2; ε255 = 17.94 mM−1 cm−1

for PQ9 and ε290 = 3.39 mM−1 cm−1 for PQ9H2 (Kruk et al., 1992;Podda et al., 1999). Complete reduction was checked by LC–MS analysisas described above. Working standard solutions containing eitherCoQ10 and PQ9, or PQ9H2 and CoQ10H2 were prepared prior to eachexperiment. The stability of the stock solutions was monitored usingthe LC–MS method described here and the solutions remained stablefor three months at −80 °C (no oxidation detected).

2.7. Sample extraction

For prenylquinone extraction, the top 1–2 cm of a coral nubbin wasplaced in a 15mL Falcon tubewith 3mL of extraction solvent containingthe internal standard CoQ9 (0.3 μM in 1:1 IPA:EtOAc). Each sample wassonicated for 2 × 10 s on icewith an ultrasonic processor (Cole-Palmer)and vortexed for 4 min after which the solvent was decanted. The vor-tex extraction was repeated twice more. All three extracts were com-bined (9 mL), centrifuged (14,000 g) for 2 min at 0 °C and stored at−80 °C for a maximum of 4 weeks prior to analysis. Stability testsshowed that the quinone redox state of these coral extractsremained stable at −80 °C for up to three months. The extractionefficiency for coral samples was determined by repetitive extractionof the four prenylquinones and the dinoflagellate pigments chloro-phyll a and c2 following the methods described by Dunlap andChalker (1986).

2.8. Optimisation of the LC–MS method

Quantification of prenylquinones and quinols in coral extracts wasoptimised for the ion trap operating in SIM mode alternately acquiringthe parent ions of PQ9, CoQ9 and CoQ10 at m/z = 749.6; 795.5; and863.7, respectively. Given that the reduced quinol forms are oxidisedduringpassage from theAPCI interface to the ion trap, the same quinoneparent ions were used for the detection of CoQ10H2 and PQ9H2 (Hansenet al., 2004). The absence of any co-eluting compounds with a similarmass was confirmed by reducing or oxidising coral extracts withNaBH4 or 1,4-benzoquinone to detect underlying signals (further detailson the optimisation and development of the LC–MS methodology canbe found in the electronic Supplementary).

2.9. Photosystem II photochemical efficiency

Stress levels of the coral symbiosis were assessed during the exper-iment by estimating the photosystem II (PSII) photochemical efficiencyof the algal partner, which is known to be significantly depressed duringcoral bleaching due to temperature induced-photoinhibition of PSII(Warner et al., 1999). PSII photochemical efficiency, expressed as max-imum quantum yield ([FM − FO] / FM = FV / FM) was measured with aDiving-PAM (Walz Gmbh, Germany) for 42 coral nubbins per treatment(n = 42) after 15 min dark adaptation (Schreiber, 2004). A 6 mm fibre

optic probe was placed perpendicular to the surface at least 20 mmfrom the tip of the nubbin and 3 mm from the coral tissue surface(controlled via a rubber spacer) to obtain themeasurements. Minimumfluorescence (FO) was measured using a weak pulsed measuring light(b0.15 μmol photons m−2 s−1; gain = 3) and maximum fluorescence(FM) was measured upon application of a saturating pulse of light(N4000 μmol photons m−2 s−1).

2.10. Surface area calculation and chlorophyll concentrations

Tissue remaining on the coral nubbins was removed by soaking indiluted commercial bleach (0.5% NaClO) overnight. Surface areas ofthe coral nubbins were determined using a wax dipping technique(Veal et al., 2010). Chlorophyll a concentrationsweremeasured fromal-iquots of the prenylquinone extracts on a UV-2550 spectrophotometer(Shimadzu, Japan) and determined using the equations of Jeffrey andHumphrey (1975).

2.11. Data analysis

Maximumquantumyieldswere analysedwith one-way ANOVA andTukey's honestly significant difference test (HSD). The Kolmogorov–Smirnov normality test and Levene's homogeneity of variance testwere used to determine whether the assumptions of the parametricone-way ANOVAs were satisfied. Repeated measures two-way ANOVAwere used to examine the effect of hyperthermal stress on %CoQH2,%PQH2 and chlorophyll a concentrations. Data for %PQH2 were logitand yields were power transformed prior to analysis. Post-hoc analysesfor the repeatedmeasureswere conducted using pairwise comparisons ofthe treatment (temperature) and the interaction (time × temperature)effects (see electronic Supplementary material for post-hoc tables). Mul-tiple pairwise comparisons were assessed using the false discovery ratefollowing Hochberg and Benjamini (1990) to account for the inflation offamilywise (type I) error rates. Statistical analyses were performedusing SPSS (version 17.0). Mass spectra and LC–MS chromatogramswere analysed with Bruker Daltonics Hystar Data Analysis (version 3.2).

3. Results

Tandemmass spectrometry with standards provided unequivocalidentification of both redox forms of CoQ10 (ubiquinone-10 andubiquinol-10) and PQ9 (plastoquinone-9 and plastoquinol-9) incoral extracts. Host CoQ10 could have potentially been contaminatedwith CoQ10 from Symbiodinium as the length of the hydrophobic sidechain of CoQ – the number of isoprenoid units – is species specific(Bentinger et al., 2007); however, analysis of cultured (type C1)and freshly isolated Symbiodinium (type C2) did not reveal any sig-nals for either CoQ9 or CoQ10 (Fig. 1). In fact, concentrations of anyCoQ isoforms in Symbiodinium were so low that using the above de-scribed LC–MS method it was not possible to determine which CoQisoforms might be present despite concentrating the extracts severalfold more than needed for suitable PQ analysis. It was therefore con-cluded that CoQ10 detected in coral extracts can be attributed solelyto the coral host and that CoQ9 is suitable for use as the internalstandard.

3.1. Characterisation of the LC–MS method

The calibration curves of CoQ10, CoQ10H2, PQ9 and PQ9H2 with CoQ9

as internal standard obtained by positive APCI-LC–MS were linear overthree orders of magnitude with a correlation coefficient R2 N 0.98from 0.03 μM to 3 μM. Using a 1:1mixture of EtOAc and IPA, the extrac-tion efficiency was greater than 95% for all prenylquinones and bothchlorophylls after three extractions. Chlorophyll is ideally extractedconcurrently with the prenylquinones as it can potentially be used tonormalise the total PQ content to give information on the size of the

PQ

CoQ9

CoQ10

PQH2

PQ

CoQ10H2CoQ9

CoQ10

0

1

2

PQH2

CoQ10H2

B

A

CCoQ9

PQH2

PQ

D

0 5 10 15 20 25

Oxidised Standards

Reduced Standards

Symbiodinium sp. C1

Acropora millepora

Aquisition time (min)

0

1

2

0

1

2

0

1

2

Inte

nsity

(10

6 A

U)

CoQ9

Fig. 1. LC–MSquantification of the oxidised (CoQ, PQ) and reduced (CoQ10H2, PQH2) formsof coenzyme Q and plastoquinone. (A) Oxidised standards (0.3 μM); (B) reduced stan-dards (0.3 μM); (C) Symbiodinium (type C1) extract; (D) Acropora millepora extract.Ubiquionone-9 (CoQ9; 0.3 μM) was added as an internal standard in each sample. Thechromatograms combine three alternately acquired traces for m/z = 749.6, 795.5 and863.7.

4 A. Lutz et al. / Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

pool in relation to chlorophyll content (Kruk and Karpinski, 2006; Krukand Szymańska, 2012). Since complex matrix effects from interferingsubstances in the coral extract could potentially lead to under- orover-estimation of the prenylquinone concentrations by LC–MS analy-sis, coral extracts were spiked with a known quantity of CoQ9 andpeak areas compared of spiked and non-spiked extracts to estimatethe extent ofmatrix effects. On average, the recovery of CoQ9was great-er than 96% indicating that matrix effects on MS ionisation using ourmethod were small.

The limit of detection (LOD) and limit of quantification (LOQ) wereestimated by repeatedly injecting the same extract of A. millepora(n = 10). The sensitivity was determined with the inbuilt signal/noise (S/N) function in the Hystar Data Analysis software. LOD andLOQ were then calculated on the basis of a minimal accepted value ofthe S/N ratio of three and ten, respectively, using the lowest obtainedS/N value for each compound. The LODs were 3.36, 1.73, 6.23 and3.78 ng mL−1, and the limits of quantification were 11.2, 5.77, 20.76and 12.59 ng mL−1 for CoQ10H2, CoQ10, PQ9H2 and PQ9, respectively(Table 1). The sensitivity of this method is thus comparable to other

Table 1Features of positive APCI-LC–MSanalysis of prenylquinones in extracts of Acroporamillepora. AbRSD, relative standard deviation; RT, retention time; %CoQH2, CoQ pool redox state, %PQH2, PQ

No. Identification Formula RT (min) L

1 Plastoquinol PQ9H2 C53H82O2 10.4 62 Ubiquinol-10 CoQ10H2 C59H92O4 17 33 Ubiquinone-9 CoQ9 C54H82O4 18 n4 Plastoquinone PQ9 C53H80O2 22.6 35 Ubiquinone-10 PQ9H2 C59H90O4 23.4 1

%PQH2

%CoQH2

MS methods reported for the analysis of CoQ10H2, CoQ10 (Hansenet al., 2004; Ruiz-Jiménez et al., 2007), PQ9H2 and PQ9 (Martinis et al.,2011). The relative standard deviations (RSD) of the individual concen-trations obtained from ten identical injections ranged from 3 to 7% forall compounds which effectively results in a RSD of less than 1% forthe calculation of the redox states for CoQ and PQ pools determined as%CoQH2 and %PQH2, respectively.With recent improvements in columntechnology, a second method was developed using a PhenomenexKinetex C18 HPLC column that reduced the total run time from40 min to 24 min (see electronic Supplementary material).

3.2. Application of the LC–MS method

The LC–MS method was used to determine the CoQ and PQ poolredox states in coral extracts subjected to short-term hyperthermalstress. The thermal stress induced a distinct bleaching response in the34 °C treatment in A. millepora as manifested by a 54.9% decline inchlorophyll a content at the 24 hour time point (ANOVA, time × temp,F8,84 = 2.68, p = 0.011; post-hoc, p b 0.001) (Fig. 2A; Table S2). In ad-dition, a significant decline in PSII photochemical efficiency (measuredas FV/FM; Fig. 2B) was detected in both the 31 °C and 34 °C treatments(ANOVA, F2,123 = 425.4, p b 0.001; for all pairwise comparisons:Tukey HSD, p b 0.001). Although themaximum quantum yield was sig-nificantly lower by 2.2% at 31 °C, the temperature induced stress wasconsiderably greater at 34 °C with a 23.8% lower quantum yield com-pared to 27 °C.

The PQ pool redox state (%PQH2 ± SE) did not differ significantlybetween temperature treatments (Fig. 3A), but differed between timepoints, which is presumed to be a consequence of the normal light–darkcycle (ANOVA, time × temp, F8,84 = 0.47, p = 0.87; time, F4,84 = 7.81,p b 0.001). Mean values of %PQH2 changed (from t = −6 h to 24 h)from 83.86 ± 1.96 to 74.02 ± 3.95 at 27 °C, from 79.32 ± 2.64 to69.90 ± 5.71 at 31 °C and from 82.80 ± 1.89 to 62.31 ± 4.59 at 34 °C.In contrast, the CoQ pool redox state (%CoQH2 ± SE; Fig. 3B) was signifi-cantly lowered at 34 °C (ANOVA, time × temp., F8,84 = 12.32, p b 0.001;post-hoc, p b 0.001) (Table S2), exhibiting a 10% oxidation from 86.26 ±0.83 (t = −6 h) to 76.32 ± 1.87 (t = 12 h). Oxidation of the CoQ poolcontinued after returning coral samples to 27 °C in the dark (72.53 ±1.1 at t=24 h), suggesting that exposure to high temperature had com-promised the metabolic redox state of the coral host. In the low and midtemperature treatments, %CoQH2 remained stable throughout the exper-iment (mean value at 27 °C = 84.81 ± 0.40; at 31 °C = 85.02 ± 0.40),demonstrating that under normal conditions the A. millepora CoQ poolis maintained in a predominantly reduced form as in other animals(Bentinger et al., 2007).

4. Discussion

In this study, we present the first direct measurements of the redoxstates of CoQ and PQ in coral–Symbiodinium symbiosis based on quanti-tative LC–MS analysis. Themethodwas found to be not only highly sen-sitive, but also ion-selective with respect to CoQ and PQ isoforms(Fig. 1), and could easily be adapted to measure the CoQ and PQ redox

breviations: LOD, limit of detection; LOQ, limit of quantification;m/z, mass-to-charge ratio;pool redox state. See text above for further details on the LC–MS methodology.

OD (ng mL−1) LOQ (ng mL−1) RSD (+)APCI m/z

.23 20.76 2.9 (M + H)+ 749.6

.36 11.20 4.7 (M + H)+ 863.7/a n/a n/a (M + H)+ 795.5.78 12.59 5.8 (M + H)+ 749.6.73 5.77 5.4 (M + H)+ 863.7

0.740.60

temp. p < 0.001time x temp. p < 0.05

*

0

5

10

15

Chl

orop

hyll

a (µ

g cm

−2 )

27°C 31°C 34°C

Time (hours)-10 302015 2510-5 50

a

c

b

0.8

FV/F

M

0.7

0.6

0.5

0.4

0.327 °C 31 °C 34 °C

A

B

Treatment

Fig. 2. Coral bleaching in response to ambient (27 °C), mid (31 °C) and high (34 °C)temperature treatments. (A) Chlorophyll a content relative to surface area of Acroporamillepora. Grey shaded areas indicate dark periods. * Denotes a significant differenceof high to ambient and mid temperature treatments (p b 0.001, Table S2). Error bars repre-sent± standard deviation (n=8specimens per sample point per treatment). (B) Boxplot ofphotosystem II photochemical efficiency (t=13.5 h) expressed asmaximumquantumyield(FV/FM). n= 40 coral nubbins per treatment. Black circles denote samples outside the 90thand 10th percentile. Different letters denote significant differences (p b 0.001) betweentreatments.

time p < 0.001temp. p < 0.001time x temp. p < 0.001

**

time p < 0.001time x temp. p = ns.

a

b

aA

B

%P

QH

2%

CoQ

H2

70

80

90

100

60

30

40

50

27°C 31°C 34°C

-10 302015 25

27°C 31°C 34°C

10-5 50

Time (hours)

-10 302015 2510-5 50

70

80

90

100

60

50

Fig. 3. (A) Plastoquinone (%PQH2) and (B) coenzyme Q (%CoQH2) pool redox states ofAcropora millepora in response to low (27 °C), mid (31 °C) and high (34 °C) temperaturetreatments. First sampling point (t = −6 h) at 27 °C, then seawater was heated to 31 °Cand 34 °C, respectively, over 3 h. Control samples were maintained at 27 °C. One hourafter reaching treatment temperatures, actinic light was applied (t = 0) for the 13 hourphotoperiod. Subsequently, temperatures were returned to ambient (27 °C) providing arecovery period of 8 h in darkness (t=24). Grey shaded areas indicate dark periods. Let-ters denote significant differences (p b 0.01) between sampling points. * Denotes signifi-cant differences of high to low and mid temperature treatment (p b 0.001, Table S2).Error bars represent ± standard deviation (n = 8 specimens per time point pertreatment).

5A. Lutz et al. / Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

states in other cnidarians, as well as other systems (algae, higher plants,other animals), providing an alternative to existing methods. Previousanalyses by Kokott (2004) indicated that only the ubiquinone CoQ10

and plastoquinone PQ9 isoforms are present in acroporid corals andtheir dinoflagellate symbionts and our analyses have confirmed this.

Data obtained during thermal bleaching ofA.millepora demonstratesthe utility of this method to examine the physiological redox responseof coral–Symbiodinium symbiosis under environmental stress. The re-sults clearly indicate that in this symbiosis CoQ and PQ pools are main-tained predominantly in their reduced (antioxidant) forms, and thehost CoQ redox state is most affected by thermal stress (Fig. 3B). Themain conclusions are as follows: First, in both light and dark periods,the PQ pool is maintained at a highly reduced level, which is consistentwith the hypothesis that at night the symbiont PQ pool undergoes darkreduction via chlororespiration (Hill and Ralph, 2005; Jones and Hoegh-Guldberg, 2001). Nonetheless, the PQ pool redox state exhibited somevariation that is likely caused by photosynthetic reduction of thePQ pool during illumination and oxidation in darkness (Kruk andKarpinski, 2006). These natural daily rhythms will be very importantto document and to understand in order to interpret the results derivedfrommeasuring the PQ pool redox state. Second, Symbiodinium is likelyto contain a large pool of the antioxidant PQH2 that is not photoactive,but which is reported to accumulate in the plastoglobuli after acclima-tion to high levels of irradiance (Szymańska and Kruk, 2010). Third,

the redox state of the PQ pool did not shift significantly duringthermally-induced bleaching, even when the photochemical efficiencyof PSII was severely compromised at the conclusion of the light expo-sure period (t = 12 h; Fig. 2B). Lastly, perhaps the most significantnovel finding reported here is that the host CoQ pool undergoesoxidation under acute thermal stress. In eukaryotes, mitochondrialmembranes contain the highest cellular CoQ/CoQH2 concentrations(Bentinger et al., 2007; Morré andMorré, 2011). Accordingly, oxidationof the coral CoQ pool is not only consistent with the oxidative theory ofcoral bleaching (Downs et al., 2002; Lesser, 1997), but might provide acrucial connection to mitochondrial dysfunction as proposed in currentbleaching models (Lesser, 2011; Weis, 2008).

Acknowledgements

This projectwas funded by the Australian Institute ofMarine Scienceand the AIMS@JCU program of James Cook University. The researchwasperformed under GBRMPA permit number G08/25734.1. [SS]

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jembe.2014.02.006.

6 A. Lutz et al. / Journal of Experimental Marine Biology and Ecology 455 (2014) 1–6

References

Bentinger, M., Brismar, K., Dallner, G., 2007. The antioxidant role of coenzyme Q. Mito-chondrion 7 (S), S41–S50.

Buxton, L., Takahashi, S., Hill, R., Ralph, P.J., 2012. Variability in the primary site of photo-synthetic damage in Symbiodinium sp. (Dinophyceae) exposed to thermal stress.J. Phycol. 48, 117–126.

Downs, C.A., Fauth, J.E., Halas, J.C., Dustan, P., Bemiss, J., Woodley, C.M., 2002. Oxidativestress and seasonal coral bleaching. Free Radic. Biol. Med. 33 (4), 533–543.

Dunlap, W.C., Chalker, B.E., 1986. Identification and quantitation of near-UV absorbingcompounds (S-320) in a hermatypic scleractinian. Coral Reefs 5 (3), 155–159.

Eakin, C.M., Lough, J.M., Heron, S.F., 2009. Climate variability and change: monitoring dataand evidence for increased coral bleaching stress. In: van Oppen, M.J.H., Lough, J.M.(Eds.), Coral Bleaching: Patterns, Processes, Causes and Consequences. Springer,Berlin Heidelberg, pp. 41–67.

Ernster, L., Forsmark-Andrée, P., 1993. Ubiquinol: an endogenous antioxidant in aerobicorganisms. Clin. Investig. 71 (8), S60–S65.

Fabricius, K.E., Mieog, J.C., Colin, P.L., Idip, D., van Oppen, M.J.H., 2004. Identity and diver-sity of coral endosymbionts (zooxanthellae) from three Palauan reefs with contrast-ing bleaching, temperature and shading histories. Mol. Ecol. 13 (8), 2445–2458.

Gleason, D.F., Wellington, G.M., 1993. Ultraviolet radiation and coral bleaching. Nature365.

Guillard, R.R.L., Ryther, J.H., 1962. Studies of marine planktonic diatoms. I. Cyclotella nanaHustedt and Detonula confervacea Cleve. Can. J. Microbiol. 8 (2), 229–239.

Hansen, G., Christensen, P., Tüchsen, E., Lund, T., 2004. Sensitive and selective analysis ofcoenzyme Q10 in human serum by negative APCI LC–MS. Analyst 129, 45–50.

Hasbi, G., de Nys, R., Burns, K., Whalan, S., Dunlap, W.C., 2011. Hepatic coenzyme Q redoxbalance of fishes as a potential bioindicator of environmental contamination by poly-cyclic aromatic hydrocarbons. Biol. Lett. 7 (1), 123–126.

Hill, R., Ralph, P.J., 2005. Diel and seasonal changes in fluorescence rise kinetics of threescleractinian corals. Funct. Plant Biol. 32 (6), 549–559.

Hochberg, Y., Benjamini, Y., 1990. More powerful procedures for multiple significancetesting. Stat. Med. 9 (7), 811–818.

Hoegh-Guldberg, O., Bruno, J.F., 2010. The impact of climate change on theworld's marineecosystems. Science 328 (5985), 1523–1528.

Jeffrey, S.W., Humphrey, G.F., 1975. New spectrophotometric equations for determiningchlorophylls a, b, c1 and c2 in higher plants, algae and natural phytoplankton.Biochem. Physiol. Pflanz. 167 (19), 1–194.

Jones, R.J., Hoegh-Guldberg, O., 2001. Diurnal changes in the photochemical efficiency of thesymbiotic dinoflagellates (Dinophyceae) of corals: photoprotection, photoinactivationand the relationship to coral bleaching. Plant Cell Environ. 24, 89–99.

Kokott, J., 2004. Measurement of Coenzyme Q and Plastoquinone Redox Balance as Indi-cators of Oxidative Stress in Coral Bleaching. University of Bremen, Bremen 90.

Kruk, J., Karpinski, S., 2006. An HPLC-based method of estimation of the total redox state ofplastoquinone in chloroplasts, the size of the photochemically active plastoquinone-pool and its redox state in thylakoids of Arabidopsis. BBA-Bioenerg. 1757 (12),1669–1675.

Kruk, J., Szymańska, R., 2012. Singlet oxygen and non-photochemical quenching contrib-ute to oxidation of the plastoquinone-pool under high light stress in Arabidopsis. BBA-Bioenerg. 1817 (5), 705–710.

Kruk, J., Strzałka, K., Leblanc, R.M., 1992. Monolayer study of plastoquinones, α-tocopherol quinone, their hydroquinone forms and their interaction withmonogalactosyldiacylglycerol. Charge-transfer complexes in a mixed monolayer.BBA-Biomembr. 1112 (1), 19–26.

Lesser, M.P., 1997. Oxidative stress causes coral bleaching during exposure to elevatedtemperatures. Coral Reefs 16 (3), 187–192.

Lesser, M.P., 2011. Coral bleaching: causes and mechanisms. In: Dubinsky, Z., Stambler, N.(Eds.), Coral Reefs: An Ecosystem in Transition. Springer, Netherlands, pp. 405–420.

Martinis, J., Kessler, F., Glauser, G., 2011. A novel method for prenylquinone profiling inplant tissues by ultra-high pressure liquid chromatography–mass spectrometry.Plant Methods 7 (1), 23.

Morré, D.J., Morré, D.M., 2011. Non-mitochondrial coenzyme Q. Biofactors 37 (5), 355–360.Nowicka, B., Kruk, J., 2012. Plastoquinol is more active than α-tocopherol in singlet oxy-

gen scavenging during high light stress of Chlamydomonas reinhardtii. BBA-Bioenerg.1817 (3), 389–394.

Podda, M., Weber, C., Traber, M.G., Milbradt, R., Packer, L., 1999. Sensitive high-performance liquid chromatography techniques for simultaneous determination oftocopherols, tocotrienols, ubiquinols, and ubiquinones in biological samples. In:Packer, L. (Ed.), Methods in Enzymology. Academic Press, San Diego, pp. 330–341.

Ruiz-Jiménez, J., Priego-Capote, F., Mata-Granados, J.M., Quesada, J.M., Luque de Castro,M.D., 2007. Determination of the ubiquinol-10 and ubiquinone-10 (Coenzyme Q10)in human serum by liquid chromatography tandem mass spectrometry to evaluatethe oxidative stress. J. Chromatogr. A 1175 (2), 242–248.

Schreiber, U., 2004. Pulse-amplitude-modulation (PAM) fluorometry and saturation pulsemethod: an overview. In: Papageorgiou, G.C., Govindjee (Eds.), Chlorophyll Fluores-cence: A Signature of Photosynthesis. Springer, Netherlands, pp. 279–319.

Shick, J.M., Iglic, K., Wells, M.L., Trick, C.G., Doyle, J., Dunlap, W.C., 2011. Responses to ironlimitation in two colonies of Stylophora pistillata exposed to high temperature: impli-cations for coral bleaching. Limnol. Oceanogr. 56 (3), 813–828.

Szymańska, R., Kruk, J., 2010. Plastoquinol is the main prenyllipid synthesized during ac-climation to high light conditions in Arabidopsis and is converted to plastochromanolby tocopherol cyclase. Plant Cell Physiol. 51 (4), 537–545.

van Oppen, M.J.H., Palstra, F.P., Piquet, A.M.-T., Miller, D.J., 2001. Patterns of coral-dinoflagellate associations in Acropora: significance of local availability and physiolo-gy of Symbiodinium strains and host-symbiont selectivity. Proc. R. Soc. B 268 (1478),1759–1767.

Veal, C.J., Carmi, M., Fine, M., Hoegh-Guldberg, O., 2010. Increasing the accuracy of surfacearea estimation using single wax dipping of coral fragments. Coral Reefs 29 (4),893–897.

Wada, H., Goto, H., Hagiwara, S.-I., Yamamoto, Y., 2007. Redox status of coenzyme Q10 isassociated with chronological age. J. Am. Geriatr. Soc. 55 (7), 1141–1142.

Warner, M.E., Fitt, W.K., Schmidt, G.W., 1999. Damage to photosystem II in symbioticdinoflagellates: a determinant of coral bleaching. Proc. Natl. Acad. Sci. U. S. A. 96(14), 8007–8012.

Weis, V.M., 2008. Cellular mechanisms of cnidarian bleaching: stress causes the collapseof symbiosis. J. Exp. Biol. 211 (19), 3059–3066.

Wilson, K., Li, Y., Whan, V., Lehnert, S., Byrne, K., Moore, S., Pongsomboon, S.,Tassanakajon, A., Rosenberg, G., Ballment, E., 2002. Genetic mapping of the blacktiger shrimp Penaeus monodonwith amplified fragment length polymorphism. Aqua-culture 204 (3), 297–309.

Yamamoto, Y., Yamashita, S., 1997. Plasma ratio of ubiquinol and ubiquinone as a markerof oxidative stress. Mol. Aspects Med. 18, 79–84.

Yamamoto, Y., Yamashita, S., Fujisawa, A., Kokura, S., Yoshikawa, T., 1998. Oxidative stressin patients with hepatitis, cirrhosis, and hepatoma evaluated by plasma antioxidants.Biochem. Biophys. Res. Commun. 247 (1), 166–170.

Yoshida, K., Shibata, M., Terashima, I., Noguchi, K., 2010. Simultaneous determination ofin vivo plastoquinone and ubiquinone redox states by HPLC-based analysis. PlantCell Physiol. 51 (5), 836–841.