Light Acclimation of the Colonial Green AlgaBotryococcus braunii Strain Showa1[OPEN]

Tomas E. van den Berg,a Volha U. Chukhutsina,a,2 Herbert van Amerongen,b Roberta Croce,a,3,4 andBart van Oorta

aBiophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, Vrije UniversiteitAmsterdam and LaserLaB Amsterdam, 1081 HV Amsterdam, The NetherlandsbLaboratory of Biophysics, Wageningen University, 6700 ET Wageningen, The Netherlands

ORCID IDs: 0000-0002-7202-4699 (T.E.v.d.B.); 0000-0002-9783-2895 (H.v.A.); 0000-0003-3469-834X (R.C.); 0000-0002-5470-1545 (B.v.O.).

In contrast to single cellular species, detailed information is lacking on the processes of photosynthetic acclimation for colonialalgae, although these algae are important for biofuel production, ecosystem biodiversity, and wastewater treatment. Toinvestigate differences between single cellular and colonial species, we studied the regulation of photosynthesis andphotoprotection during photoacclimation for the colonial green alga Botryococcus braunii and made a comparison with theproperties of the single cellular species Chlamydomonas reinhardtii. We show that B. braunii shares some high-light (HL)photoacclimation strategies with C. reinhardtii and other frequently studied green algae: decreased chlorophyll content,increased free carotenoid content, and increased nonphotochemical quenching (NPQ). Additionally, B. braunii has unique HLphotoacclimation strategies, related to its colonial form: strong internal shading by an increase of the colony size and theaccumulation of extracellular echinenone (a ketocarotenoid). HL colonies are larger and more spatially heterogenous thanlow-light colonies. Compared with surface cells, cells deeper inside the colony have increased pigmentation and largerphotosystem II antenna size. The core of the largest of the HL colonies does not contain living cells. In contrast with C.reinhardtii, but similar to other biofilm-forming algae, NPQ capacity is substantial in low light. In HL, NPQ amplitudeincreases, but kinetics are unchanged. We discuss possible causes of the different acclimation responses of C. reinhardtii andB. braunii. Knowledge of the specific photoacclimation processes for this colonial green alga further extends the view of thediversity of photoacclimation strategies in photosynthetic organisms.

Evolutionary differences in microalgae photosyn-thesis are directed by the local light environment.Variations in light intensity and quality combined withthe availability of nutrients have shaped a range of

fine-tuned algae, adapted to their environmental niches(Croce and van Amerongen, 2014). Within this range ofdifferent algae types, many different shapes and sizesare found, from unicells no larger than 1 mm to multi-cellular and colonial species of several centimeters(Beardall et al., 2009). While data on unicellular speciesprovide us with an emerging view on the variations ofthe photosynthetic apparatus and its regulation, weknow little about multicellular and colonial microalgaethat have to deal with increased light attenuation(within colonies) and decreased diffusion of nutrients(Beardall et al., 2009).

One of these freshwater colonial microalgae isBotryococcus braunii (Trebouxiophycae, Chlorophyta;Weiss et al., 2010), found in lakes and ponds through-out different climate zones (Metzger and Largeau,2005). This alga has been targeted for biofuel produc-tion since the 1970s because it produces high-qualitylong-chain hydrocarbons, which are excreted in theextracellular matrix (botryococcenes; Metzger et al.,1985). The oil content of B. braunii colonies can bevery high (30%–40% of its dry weight; Metzger andLargeau, 2005) and makes the colonies float close tothe water surface, where they can be subjected to high-light (HL) intensities (Wake and Hillen, 1980). Theirgrowth rates are prohibitively low for commercial uti-lization, despite numerous studies aimed at optimizing

1This work was supported by the Netherlands Organization ofScientific Research (NWO) Earth and Life Sciences (ALW), througha Veni grant to B.v.O. and a Vici grant to R.C., by the NWO-ALWthrough anMMMgrant to R.C., and by a grant from the BioSolar CellProgram, cofinanced by the Dutch Ministry of Economic Affairs,to R.C.

2Current address: Department of Life Sciences, Sir Ernst ChainBuilding, Imperial College London, London SW7 2AZ, UK.

3Author for contact: r.croce@vu.nl.4Senior author.The author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:Roberta Croce (r.croce@vu.nl).

R.C. conceived the original research plans; T.E.v.d.B., B.v.O.,V.U.C., H.v.A. and R.C. designed the experiments; T.E.v.d.B. per-formed the experiments and analyzed the data; B.v.O. and V.U.C.supervised the experiments and data analysis; T.v.d.B wrote the ar-ticle with contributions of all the authors; R.C. and B.v.O. supervisedand completed the writing; all authors approved the final version ofthe article.

[OPEN]Articles can be viewed without a subscription.www.plantphysiol.org/cgi/doi/10.1104/pp.18.01499

1132 Plant Physiology�, March 2019, Vol. 179, pp. 1132–1143, www.plantphysiol.org � 2019 American Society of Plant Biologists. All Rights Reserved. www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from

Copyright © 2019 American Society of Plant Biologists. All rights reserved.

growing conditions (Sakamoto et al., 2012; Yoshimuraet al., 2013; Kim et al., 2014) and despite the occurrenceof natural blooms (Wake and Hillen, 1980). A betterunderstanding of the biology of this alga could identifyfactors that limit growth and lead to selection of targetsfor strain improvement (Barry et al., 2015).Here, we focused on the Berkeley strain (Showa) of B.

braunii (Nonomura, 1988) because it is one of the best-studied strains (Race B; Metzger and Largeau, 2005)with one of the highest growth rates, and its genomewas recently sequenced (Browne et al., 2017). Thegrowth rate depends on light conditions: upon lightintensity rise, the growth rate increases up to a satura-tion level and then a drop is observed (Sakamoto et al.,2012; Yoshimura et al., 2013; Kim et al., 2014). This dropindicates a decrease in photosynthetic efficiency in HL.Colony size increases in HL (Zhang and Kojima, 1998),possibly as a photoprotective strategy (Khatri et al.,2014), and this may be related to the observed de-crease in photosynthetic efficiency.We recently isolated and characterized the light-

harvesting antenna complexes (LHC) of B. braunii(van den Berg et al., 2018). LHCs are responsible formost of the light absorption in green algae and transferthe resulting excitation energy to the reaction centers ofPSI and PSII, where it initiates electron transport cou-pled with the formation of a transmembrane protongradient (light reactions). These electrons and protonsare used by the ferredoxin-NADP+ reductase and theATPase to produce the NADPH and ATP necessary forcarbon fixation (dark reactions). Some LHCs are spe-cifically associated with a single photosystem formingsupercomplexes, such as LHCI in PSI-LHCI (e.g. Chla-mydomonas reinhardtii and plants). The major antenna(LHCII) can be associated with both photosystems(Wientjes et al., 2013a; Drop et al., 2014). General LHCfeatures such as the oligomerization state and chloro-phyll (Chl) a/b ratio are conserved in B. braunii, but thecarotenoid composition and the stability of the trimerare different from that of the LHCs of C. reinhardtii (vanden Berg et al., 2018).Photosynthetic organisms respond to fast changes in

the local light environment by activating photo-protective mechanisms (for review, see Wobbe et al.,2016). A central process of photoprotection is non-photochemical quenching (NPQ) of excitation energy.Regulated NPQ is activated by the acidification of thelumen and reduces the formation of reactive oxygenspecies by the reduction of the lifetime of the Chl ex-cited state. Multiple components are involved in NPQin different organisms and act on different time scales(Quaas et al., 2015; Christa et al., 2017; Kress and Jahns,2017; Farooq et al., 2018). State transitions, in whichLHCs redistribute between PSI and PSII (Goldschmidt-Clermont and Bassi, 2015), take several minutes andalso may be photoprotective in HL in algae (Allorentet al., 2013). Sustained changes in light conditionslead to long-term (up to several generations) photo-acclimation. Photoacclimation induces changes in thephotosynthetic machinery that take place on a time

scale of hours to days. Such changes concern (1) theamount of photosynthetic protein per cell (e.g. modu-lation of antenna size and composition and change ofthe PSI/PSII ratio; Anderson and Andersson, 1988); (2)the amount and content of proteins involved in NPQ(Peers et al., 2009) and carbon-concentrating mecha-nisms (Yamano et al., 2008); and (3) free carotenoidcontent (not bound by protein; Ben-Amotz and Avron,1983). Together, these changes balance light absorptionand electron transport to optimize photosynthesis un-der new sustained light conditions.Detailed knowledge of photoacclimation strategies is

limited to a few single cellular algae (Meneghesso et al.,2016; Polukhina et al., 2016), and these strategies de-pend on the environment from which the alga wasisolated; thus, our view of the diversity in photo-acclimation strategies could be limited. Moreover, re-cent results show that the diversity of photoprotectivemechanisms in green algae is even larger than previ-ously thought (Peers et al., 2009; Tibiletti et al., 2016;Christa et al., 2017). Therefore, detailed studies of algaeranging from unicellular to colonial and multicellular,obtained from multiple environments, are essential tounderstand photoacclimation strategies. In B. braunii,this may help increase growth rates by light control andstrain selection.We measured pigment content, colony size distri-

butions, and NPQ combined with time-resolved fluo-rescence and microscopy during photoacclimation todifferent light intensities (low [LL], medium [ML], andHL). We conclude that B. braunii employs numerousgeneral HL acclimation strategies, such as the reductionof Chl content and the increase of NPQ in combinationwith additional unique strategies. Colony size increaseand accumulation of extracellular echinenone (Echi)content, unique HL acclimation strategies of B. braunii,reduce the light intensity experienced by themajority ofthe cells, leading to lateral differentiation in Chl contentand antenna size in the colony. Such an acclimationstrategy could possibly be more common for nonmotilecolonial algae that live in environments where shade isminimal.

RESULTS

Growth Kinetics

The effect of acclimation to different growth lightintensities on biomass production was assessed bymeasuring the dry weight of the cultures. A culturegrown for 30 d at ML was split into five daughter cul-tures (day 0), which were then each grown under dif-ferent light intensities: three times lower (LL; twocultures), three times higher (HL; two cultures), or un-changed (ML; one culture). The specific growth rate inthe exponential growth phase (days 0–18) was deter-mined by the best fit with an exponential growthfunction (Fig. 1A). Surprisingly, no lag phase was ob-served for LL and HL. The growth rates (Fig. 1B) for the

Plant Physiol. Vol. 179, 2019 1133

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

exponential phase were similar for all light intensities,meaning that the extra light in ML and HL was noteffectively used for growth.

Colony Size

Colony size distributions at day 0 and 20 d after thechange in light intensity (Fig. 2) showed that acclima-tion to LL leads to smaller colonies compared with MLand HL colonies. For LL acclimation, the volume dis-tribution of the diameter of the colony shows twomaxima: one at a similar position as for the ML startingculture and one at 5 to 6 times smaller diameter. Thisindicates the presence of a fraction of the original col-onies together with newly budded smaller colonies. Anindependent longer experiment (34 d of acclimation inthe three different light conditions) showed a singledistribution for LL colonies with the maximum around100 mm, confirming that LL acclimation leads to theformation of smaller colonies (Supplemental Fig. S1).

Pigment Content

Light stress and photoacclimation in algae and plantsaffect pigment content as a result of changes in thecomposition of the photosynthetic complexes in themembrane and of the synthesis of new carotenoids(Ballottari et al., 2007; Bonente et al., 2012; Wientjeset al., 2013b; Meneghesso et al., 2016; Polukhina et al.,2016).

The carotenoid content was monitored at fourtime points (5, 14, 21, and 41 d) during acclimationin all light conditions. All cells contained neoxanthin(Neo), loroxanthin (Lor), violaxanthin (Vio), lutein(Lut), zeaxanthin (Zea), Echi, and b-carotene (b-Car) atall stages, although their relative abundance varied.The carotenoid composition did not change during the

exponential growth phase (0–18 d) in ML (SupplementalTable S1).

In LL, the amounts of Neo+Lor, Vio, and b-Carnormalized to Chl are higher, and Lut, Echi, andZea are lower, than in ML. The most substantialchanges occur in the first 5 d of acclimation (Table 1).The Echi/Chl ratio decreases sharply and remains sta-ble during exponential growth (0–18 d). At the begin-ning of growth limitation, toward the stationary phase,it increases again. This is in good agreement withthe reported increase of Echi/total carotenoids in theintermediate or early stationary phase (Grung et al.,1989).

In HL, the Zea content increases substantially in thefirst 5 d of acclimation, while both the Neo+Lor and Viocontents remain as in ML and decrease only after theexponential growth phase. The Echi content increaseswithin 5 d, remains constant during the exponentialgrowth phase, and increases further in the stationaryphase (day 41).

In summary, two phases can be observed: the accli-mation to the new light condition, leading to changes in

Figure 1. Growth of biomass expressed as the increase of dry weight concentration over a period of 35 d. A, Exponential growthlasted up to day 18, as judged from the best fit with an exponential growth function (straight lines) for all light experiments. Resultsare for two biological replicas for LL (50 mmol photons m22 s21; open symbols) and HL (450 mmol photons m22 s21; closedsymbols) and one for ML (150 mmol photons m22 s21; crosses) intensities. Note the logarithmic vertical axis. B, Growth ratescorresponding to the fits in A. Error bars indicate SE of the fits.

Figure 2. Colony size distributions for the different acclimation culturesafter 20 d. Results are for two biological replicas for LL (open symbols)and HL (closed symbols) and one for ML (crosses). The black curverepresents the colony size of the starting culture (day 0). Error barsrepresent the SD of five technical replicas.

1134 Plant Physiol. Vol. 179, 2019

Photoacclimation in Botryococcus braunii

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

pigment composition within 5 d, and the transitiontoward the stationary phase, leading to further changesin pigment composition when growth rate decreases.The pigment content of the acclimated cultures in

fed-batch is presented in Table 2. The Chl-to-dryweightratio was;2 times higher in LL-acclimated cells than inML-acclimated cells, but it was only slightly lower inHL-acclimated cells than in ML-acclimated cells. Thissuggests an increase in the number of photosyntheticcomplexes in LL and a decrease of photosyntheticcomplexes in HL. The Chl-to-carotenoid ratio (Chl/Car) was in all cases below 2.3, which is lower than inmost photosynthetic complexes of algae and plants(Nicol and Croce, 2018). This is most probably due tothe presence of free carotenoids such as Echi, which islocated in the extracellular matrix (Grung et al., 1994).The relative amount of Echi increases even more in HL,further decreasing the Chl/Car ratio (Table 1). Thechange in Chl a/b ratio upon acclimation to LL or HLis small.

Maximum Quantum Yield of PSII and NPQ

Next, we measured maximum quantum yield of PSII(Fv/Fm) and NPQ values (Table 3) to assess how lightacclimation affects photosynthetic performance. Thesevalues mainly represent surface cells because the mea-suring light (460 nm) does not penetrate deep into thecolonies, especially for ML and HL cells, which containlarge amounts of Echi (see Supplemental Material).Fv/Fm was 0.68 for the LL culture, close to the valuesreported for other algae (Parkhill et al., 2001; Bonenteet al., 2012; Meneghesso et al., 2016; Polukhina et al.,2016), and it decreased with increasing growing lightintensity (LL . ML . HL), which may indicate pho-toinhibition. Both Fv/Fm and the value of NPQ weresimilar at all measurement time points during accli-mation in all cultures, indicating that they reach a newconstant value within 5 d after the transition from ML.The capacity of NPQ was assessed during light ac-

climation using blue (460 nm, 520 mmol photons m22

s21) and red (635 nm, 550 mmol photons m22 s21) ac-tinic light (AL). In both conditions, NPQ was fully

reversible in all cultures, indicating that no photo-inhibition occurred within the 12 min of light exposure(Fig. 3). A total of 550 mmol photons m22 s21 red lightwas sufficient to induce the maximum NPQ in all cul-tures (Supplemental Fig. S2A), although the level washigher in HL (1.8) than in ML (1.5) and LL (1.3). Incontrast, the induction kinetics of NPQ were similar inall cultures at the same AL intensity (Supplemental Fig.S2B). The contributions of the xanthophyll cycle andstate transitions to the NPQ induction curves are pos-sibly small or absent because the fast phase is muchlarger than the slow phase (Fig. 3).The level of NPQ induced by blue light was far

smaller than by red AL (Fig. 4) in ML and HL culturesbut not in LL cultures. The induction kinetics were alsodifferent. This can be explained by the presence ofpigments (e.g. Echi) that absorb blue light (but not red)and do not transfer energy to the photosynthetic ap-paratus, effectively acting as a sunscreen. This is qual-itatively confirmed by the difference between thefluorescence excitation spectra of the acclimated cul-tures (Fig. 4, right). An effect from the changes in scat-tering can be excluded because the absorption spectraof the cultures measured in an integrating sphere(Supplemental Fig. S3) show increased absorption, 550 nm in HL compared with LL, in line with thepigment composition.In summary, the ML- and HL-acclimated cultures

increased their NPQ capacity and displayed a sign oflight stress, namely, reduced Fv/Fm after dark accli-mation. Echi accumulated in HL and had a screeningfunction for high-energy photons.

Time-Resolved Fluorescence

The small changes in the Chl a/b ratio observedduring light acclimation (Table 2) can be explained inthree ways: (1) the change in antenna size and/or PSI/PSII ratio is small, as is the case in C. reinhardtii whengrown photoautotrophically (Polukhina et al., 2016); (2)the change of the antenna size is compensated for by achange of the PSI/PSII ratio (PSI has a higher Chl a/bthan PSII in all organisms analyzed so far); and (3) the

Table 1. Changes in carotenoid content (relative to 100 Chls) during light acclimation to LL or HL during the culture period

ML is the aggregate result of measurements at different days (Supplemental Table S2). Error margins represent the SD of n (first column) technicalrepetitions in two biological replicas. Neo and Lor as well as b,b- and b,«-Car were not separated in the chromatogram and are therefore quantifiedtogether.

Condition Neo+Lor Vio Lut Zea b-Car Echi

ML day 0–42 (n = 8) 6.5 6 0.4 2.6 6 0.2 21.0 6 1.0 2.3 6 0.3 4.0 6 1.0 34.0 6 5.0HL day 5 (n = 3) 5.6 6 0.4 2.6 6 0.2 23.3 6 0.2 4.0 6 0.4 5.6 6 0.1 48.0 6 2.0HL day 14 (n = 3) 6.9 6 0.7 3.0 6 0.2 25.0 6 2.0 2.9 6 0.8 3.6 6 0.1 48.0 6 4.0HL day 21 (n = 3) 5.0 6 0.2 2.3 6 0.1 22.5 6 0.2 4.5 6 0.4 5.5 6 0.1 49.0 6 0.0HL day 41 (n = 4) 4.2 6 2.7 1.6 6 0.0 24.7 6 0.3 5.3 6 0.4 5.2 6 0.4 62.0 6 1.0LL day 5 (n = 3) 11.4 6 0.1 3.9 6 0.0 13.0 6 2.0 0 4.4 6 0.2 4 0.0 6 1.0LL day 14 (n = 4) 10.8 6 0.4 4.6 6 0.1 10.0 6 0.8 0 5.6 6 0.3 6.1 6 0.9LL day 21 (n = 4) 10.0 6 0.3 4.0 6 0.2 10.4 6 0.5 0 6.1 6 0.6 14.0 6 2.0LL day 41 (n = 4) 8.1 6 0.4 3.7 6 0.1 14.6 6 0.5 0 6.1 6 0.2 25.7 6 0.8

Plant Physiol. Vol. 179, 2019 1135

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

antenna size and/or PSI/PSII ratio varies within acolony for cells at different positions, as was observed incolonies of the cyanobacteria Nostoc sphaeroides (Denget al., 2008). To be able to discriminate between thesepossibilities, wemeasured time-resolved fluorescence bytime-correlated single-photon counting (TCSPC) andfluorescence lifetime imaging microscopy (FLIM).

TCSPC

TCSPC was measured on colonies with open (F0) orclosed (Fm) PSII reaction centers, with preferential ex-citation of LHCs (650 nm, Chl b) or photosystem cores(662 nm, Chl a) and with detection at 685, 700, and 715nm. The PSII reaction center state (fully open [F0] orfully closed [Fm]) was confirmed by the lack of changesin the fluorescence decay curve at different excitationpowers above and below the set values (example inSupplemental Fig. S4) and by comparing Fv/Fm calcu-lated from TCSPC (Table 4) with the values measuredbypulse-amplitude modulated fluorometry (Table 3).

Fluorescence kinetics were fitted globally with a sumof exponential decays while linking lifetimes betweenmultiple decay traces in F0 (see “Materials andMethods” and Supplemental Methods). Examples ofdecay traces are shown in Figure 5, demonstrating theslower decay in Fm compared with F0 and the differ-ences between HL and LL. All decay traces and fit re-siduals are shown in Supplemental Figures S5 and S6,together with the decay-associated spectra (DAS;Supplemental Figs. S7 and S8). The results are shown inSupplemental Table S2.

Themultiwavelength approach enables separation ofthe decay components associated with PSI and PSIIthanks to spectral differences between the two com-plexes (van Oort et al., 2010; Ünlü et al., 2014). Thelifetime of the first component is approximately 0.11 nsin all experiments. The corresponding DAS (DAS1;

Fig. 6) increased toward longer wavelengths in allconditions, and the amplitude was lower for 650 nmthan for 662 nm excitation. Together, these results in-dicate that DAS1 mainly originates from PSI, as wasobserved before in plants and C. reinhardtii (van Oortet al., 2010; Ünlü et al., 2014), with possibly a minorcontribution of PSII at F0. The DAS1 amplitudes aresimilar for HL and LL colonies, showing that the PSI/PSII excitation ratio is the same in both samples (aver-age difference in ratio DAS1/total between HL/LL inthe same condition is 7% 6 4%) and suggesting thatlight acclimation does not affect the excitation ratio ofPSI and PSII in B. braunii.

The PSII antenna size was previously assessed fromthe lifetimes and DAS attributed to PSII in F0 and theirdependence on excitation wavelength (van Oort et al.,2010; Ünlü et al., 2014). However, the likely presence ofphotoinhibition complicates this analysis in B. brauniicultures. Indeed, the average decay times (,t.) of PSII(greater than 0.12 ns and less than 3 ns) detected at 685nm, where PSII emission dominates, are counterintui-tive (Table 4): ,t. is longer for HL than for LL colo-nies, which would correspond to a larger antenna sizein HL colonies, whereas the antenna size is typicallylarger for LL-grown plants or algae (Falkowski andChen, 2003; Wientjes et al., 2013b). At Fm, ,t. is longerfor LL than for HL colonies. Thus, it appears that the HLcolonies contain fluorescent species with less (or none)variable fluorescence and a lifetime longer than that ofPSII at F0 but shorter than PSII in Fm. Possible candidatesfor such species are quenched antennas disconnectedfrom the photosystems and photoinhibited photosystem.The difference between ,t. at F0 and Fm between LLand HL is smaller for 650 nm than for 662 nm, indicatingthat the species is not likely to be an antenna.

In summary, the results suggest that the excitation ratioof PSI and PSII (including species without variable fluo-rescence) is the same for HL- and LL-acclimated coloniesand that photoinhibited PSII is present in HL colonies.

Table 2. The pigment content of fed-batch cultures fully acclimated to different light intensities

Errors represent the SD of two biological replicas with five technical replicas each.

Light Chl (% dry weight) Car (% dry weight) Chl a/b Chl/Car

LL 1.3 6 0.1 0.6 6 0.2 3.0 6 0.1 2.4 6 0.1ML 0.7 6 0.1 0.3 6 0.1 3.1 6 0.1 1.4 6 0.2HL 0.5 6 0.2 0.45 6 0.2 2.8 6 0.1 0.7 6 0.2

Table 3. Fv/Fm and maximum NPQ of cultures in the different conditions

These values mainly represent surface cells (see text). The values of Fv/Fm and maximum NPQ did not change with time, so we report averages ofmeasurements at days 5, 13, 20, 22, and 29. Errors indicate the SD of n technical replicas in two (LL and HL) or one (ML) biological replicas at thesefive time points. Blue/red excitation ratio indicates the ratio of amounts of excitation by the blue/red LEDs, calculated as the ratio of the overlapintegral of each LED spectrum and the fluorescence excitation spectrum (Fig. 4, right).

Culture LL ML HL

Fv/Fm 0.68 6 0.02 (n = 18) 0.65 6 0.02 (n = 12) 0.59 6 0.03 (n = 20)Blue/red excitation ratio 1.16 6 0.01 (n = 2) 0.87 6 0.04 (n = 2) 0.63 6 0.02 (n = 2)Maximum NPQ (red AL, 550 mE m2 s21) 1.3 6 0.1 (n = 10) 1.5 6 0.2 (n = 6) 1.8 6 0.4 (n = 8)Maximum NPQ (blue AL, 520 mE m2 s21) 1.1 6 0.3 (n = 4) 0.9 6 0.3 (n = 2) 0.5 6 0.1 (n = 4)

1136 Plant Physiol. Vol. 179, 2019

Photoacclimation in Botryococcus braunii

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

Confocal and Reflection Microscopy on the Largest of theHL Colonies

Fifteen of the largest (approximately greater than1 mm) HL colonies were cut in half to study heteroge-neity by optical reflection and confocal fluorescencemicroscopy. All colonies showed the same multilayerstructure (Fig. 7). Each colony consisted of a hollowcavity surrounded by an orange and then a green do-main (Fig. 7A). The orange domain shows no Chlautofluorescence (Fig. 7B). Green autofluorescence isseen throughout the colony (Fig. 7C) and originatesfrom the extracellular matrix (Bachofen, 1982). Thelifetime of the green autofluorescence was extremelyshort (less than 5 ps; Supplemental Fig. S9), suggestingthat it could originate from Echi. The orange domaindoes not contain cells, as evidenced by the lack ofSYTOXGreenfluorescence (see “Materials andMethods”).

Instead, the green outer domain is composed mainlyof living cells containing Chl (Fig. 7B; see also FLIMresults below).

FLIM on the Largest of the HL Colonies

The structural differences from the surface towardthe inside in the largest of the HL colonies observed byconfocal and reflectance microscopy could indicatefunctional heterogeneity. To test this, we used FLIMto compare the time-resolved Chl fluorescence at dif-ferent distances (depths) from the colony surface.Fluorescence decay curves were measured in eachthree-dimensional pixel (voxel) and fitted by a sum ofexponentials (Fig. 8B). FLIMwas measured on 100 cellsclose to the colony surface (outer cells) and on 80 cellsfar from the colony surface (inner cells) in two of thelargest of the HL-acclimated colonies. For a better sig-nal-to-noise ratio, the fluorescence decay traces of allvoxels belonging to each individual cell were summed.Dead cells marked by SYTOX Green fluorescence werelow in number (Supplemental Fig. S10) and were ex-cluded from the analysis. From the resulting traces, wedetermined the following for each cell: (1) the peak in-tensity of the decay trace, which is approximatelyproportional to the number of contributing pigments;and (2) the average lifetime (,t.). The peak intensities(Fig. 8D) were significantly lower for outer cells

Figure 3. NPQ induction curves (averages of days 5, 13, 20, 22, and29). LL (red-pink, triangles; two biological replicas), ML (black, crosses),and HL (blue-green, squares; two biological replicas) colonies areshown. The AL was red (550 mmol photons m22 s21). The shaded arearepresents the SD from nine to 12 technical replicas. The differencesbetween all light conditions are significant as judged by a double-sidedStudent’s t test (P, 0.05). The spread of NPQ in time was the largest forHL, possibly because the error is larger due to the lower Chl content,larger colony size (and size variation), and higher Echi content of HLcolonies.

Figure 4. Screening effect of blue photons in HL-acclimated conditions. Left, One example of NPQ induction curves at day 22 ofLL (red triangles), ML (black crosses), and HL (green squares) cultures measured with red AL (RL; 550 mmol photons m22 s21; redcurves) or blue light (BL; 520 mmol photons m22 s21; blue curves). Right, Excitation spectra of Chl fluorescence (735 nm de-tection) at room temperature of LL (red), ML (black), and HL (green) cultures (spectra are normalized to the peak around 670 nm).Excitation spectra are among others distorted by absorption flattening. Light blue and light red traces represent the emissionspectra of the pulse-amplitude modulated AL light-emitting diodes (LEDs) normalized to their relative intensities.

Table 4. Average fluorescence lifetimes (ns) of LL and HL colonies atF0 or Fm excited at 650 or 662 nm and detected at 685 nm

The components faster than 0.12 ns and slower than 3 ns were ex-cluded from the calculation. The uncertainty is less than 5% as esti-mated from the confidence intervals of the fit of lifetimes andamplitudes (DAS).

lexc LL (F0) HL (F0) LL (Fm) HL (Fm) Fv/Fm LL Fv/Fm HL

650 nm 0.618 0.710 1.795 1.729 0.66 0.59662 nm 0.584 0.683 1.894 1.584 0.69 0.57

Plant Physiol. Vol. 179, 2019 1137

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

(490 6 160) than for inner cells (600 6 230), as testedwith a one-sided ANOVA (P , 0.001). The averagelifetime of outer cells (850 6 160 ps) was significantlyshorter than that of inner cells (1,040 6 160 ps; Fig. 8C;P , 0.001). In contrast, cells in the smaller LL colonieshad a homogenous distribution of the fluorescencelifetime throughout the colony (Supplemental Fig. S11).The longer average lifetimes of the outer cells in theFLIM measurements compared with the TCSPC mea-surements can be explained by a higher level of pho-toinhibition in FLIM, since it focuses exclusively on theoutermost layer of cells. Together, the FLIM resultsshow a spatial variation of the functional compositionof chloroplasts throughout the largest of the HL-acclimated colonies.

DISCUSSION

Photoacclimation strategies in colonial algae arelargely unknown. Here, we used a combination of bi-ochemical and spectroscopic techniques to study the

photoacclimation of B. braunii grown in photoautotro-phic conditions under continuous light in a bubblecolumn photobioreactor. The photoacclimation strate-gies of B. braunii partly overlap with those of singlecellular algae but are also in part unique to this alga,and possiblymore generally for colonial photosyntheticspecies. We will first address the unique strategies ofB. braunii, followed by those that are similar in B. brauniiand other algae, and finish with a discussion on theeffects of HL on Fv/Fm, colony size, and growth rate.Our comparison will focus on C. reinhardtii because it isthe best-studied green alga.

Unique HL Acclimation Strategies of B. braunii: Shadingand Sunscreen

The function of B. braunii colony size changes and therelation with light intensity are unclear (Zhang andKojima, 1998; Khatri et al., 2014; García-Cubero et al.,2018). However, previous studies were performed inthe presence of high CO2 concentration, which does notcorrespond to physiological conditions. Here, we showthat in carbon limitation (ambient CO2, a condition thatactivates carbon-concentrating mechanisms already atlow light; Suzuki et al., 2013), colony size is increased atlight intensities of 150 and 450 mmol photons m22 s21.This suggests that the increase in colony size, whichincreases shading, is a photoprotective strategy, cor-roborating an earlier hypothesis by Khatri et al. (2014).

Additional photoprotection in HL arises from Echi(Table 1), which accumulates in the extracellular matrix(Grung et al., 1994) and blocks approximately 50% ofthe incident light between 350 and 550 nm (compare redand green spectra in Fig. 4, right). Photoprotection byoptical screening by ketocarotenoids (but not Echi)accumulated outside the thylakoid membrane is ob-served in many other green algae under stress condi-tions (Solovchenko, 2013).

The decrease of the Echi/Chl ratio upon transition toLL (34 [ML] to 4 [LL]; Table 1) suggests that the cellsmay be actively degrading Echi, raising the questionwhether B. braunii can catabolize extracellular Echi.Echi/Chl also decreases by increasing the Chl contentper cell (2.63; Table 2) and by dilution by growth(;1.333; Fig. 1), but these two effects led to a decreaseof Echi/Chl to 10 6 4, considerably higher than theexperimental 46 1 (Table 1). Thus, it appears that Echiis indeed catabolized to some extent.

Similar HL Acclimation Strategies in B. braunii and OtherGreen Algae: Decreased Chl Content, Increase of FreeCarotenoids, Increase of NPQ, and AntennaSize Modification

Decreased Chl content and increased carotenoidcontent upon HL acclimation in B. braunii (Table 1) aregeneral strategies observed in algae and plants, al-though the nature of the carotenoid varies betweenspecies (Ballottari et al., 2007; Bonente et al., 2012;

Figure 5. Fluorescence decay traces of HL (blue) and LL (red) coloniesin F0 (dark) and Fm (light) states. Excitation was at 662 nm and detectionat 685 nm.Note the slower decay in Fm relative to F0. HL decays faster inFm but slower in F0 relative to LL.

Figure 6. DAS1 of LL and HL colonies in F0 and Fm, excited at 662 nm(left) and 650 nm (right). The area under each DAS is normalized to thesum of the areas of all DAS within each experiment. The lifetimes areindicated in parentheses.

1138 Plant Physiol. Vol. 179, 2019

Photoacclimation in Botryococcus braunii

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

Meneghesso et al., 2016; Polukhina et al., 2016). Theincrease in Zea/Chl ratio in B. braunii in HL (Table 2)might suggest an involvement of this xanthophyll inNPQ as in plants (Demmig-Adams, 1990) and somealgae (Quaas et al., 2015; Christa et al., 2017). However,the presence of Zea is not necessarily evidence of itsinvolvement in NPQ, since a similar increase of Zeacontent and NPQ capacity in HL also occurs in C.reinhardtii, where Zea does not play a role in NPQ(Bonente et al., 2012; Quaas et al., 2015).The NPQ capacity in LL in B. braunii is considerable

(1.3; Fig. 3) when compared with that of C. reinhardtii(0.3–1; Bonente et al., 2011, 2012; Quaas et al., 2015;

Polukhina et al., 2016), and it is similar to that of fourbiofilm-forming algae (1–1.2; Quaas et al., 2015). In C.reinhardtii, NPQ capacity increases to greater than 3 inHL and requires an inducible actor (LHCSR3; Peerset al., 2009). In B. braunii, the increase in HL is smaller(1.9 times), suggesting that such an actor (e.g. LHCSR orPSBS, both are present in the genome; Browne et al.,2017) may only be weakly inducible in B. braunii.HL acclimation can affect the antenna size of PSII and

PSI but generally not the PSI/LHCI ratio (Ballottariet al., 2007; Bonente et al., 2012; Wientjes et al., 2013a).The PSI excited state lifetimes are similar in LL and HL(Fig. 6), suggesting that in B. braunii, HL acclimation

Figure 7. Structure of the largest of the HL-acclimated colonies cut in half and visualized by optical microscopy. A, Opticalreflectionmicroscopy image of a cross section of a colony grown after 41 d of HL acclimation showing its heterogenous structure.B, 2D average intensity projectionmap of Chl autofluorescence (greater than 635 nm detection) in a z-stack of an HL colony after36 d of HL acclimation showing that the central domain is nonfluorescent/absorbent. C, 2D average intensity projection map ofthe same z-stack region as in B for green autofluorescence (505–545 nm detection) throughout the colony. Bars indicate thefluorescence intensity in arbitrary units.

Figure 8. FLIM of the largest of the HL-acclimated colonies, cut in half. A, Typi-cal FLIM image (153.6 3 153.6 mm) of afew clusters of inner cells in HL-acclimatedcolonies. Grayscale indicates the summedfluorescence intensity in each voxel. Thegreen mask is an example of the voxelsselected within a single cell. B, Typicalsummed fluorescence decay curves ofcells close to the colony surface (black,outer cell) and far from the surface (blue,inner cell; green mask in A). The multi-exponential fits are in red. C, Frequencyhistogram of the average lifetimes of the fitsof the fluorescence decays of individualchloroplasts in inner and outer cells. D,Frequency histogram of peak intensities ofthe fluorescence decay curves of individ-ual chloroplasts. Results in C and D arefrom 100 individual outer cells and 80 in-ner cells, from two different colonies. Thedifference between the average lifetimesand peak intensities of inner/outer cellswas significant as tested with a one-sidedANOVA (P , 0.001).

Plant Physiol. Vol. 179, 2019 1139

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

does not affect the PSI antenna size, since a larger PSIantenna would result in longer average lifetimes (LeQuiniou et al., 2015). It is interesting that the PSI life-time of B. braunii is ;110 ps, which is longer than inArabidopsis (Arabidopsis thaliana; Wientjes et al., 2013b)and C. reinhardtii (Ünlü et al., 2014), suggesting a largerPSI antenna.

Assessing PSII antenna size is complicated in HLcolonies, which are functionally heterogenous, as ob-served by FLIM (Fig. 8). However, although the likelypresence of photoinhibition does not permit the acqui-sition of quantitative data, within large HL colonies,cells at the colony surface have shorter excited statelifetimes than those deeper inside (Fig. 8C), suggestingsmaller PSII antenna size at the surface. Such variationsbetween cells at different depths were observed beforein the large colonies of the cyanobacterium N. sphaer-oides (Deng et al., 2008) and in leaves of Arabidopsis, inwhich the PSII antenna size was found to be smaller onthe sun side (Iermak et al., 2016).

Effects of HL on Fv/Fm, Colony Structure, and Growth Rate

LL colonies are small (80 mm) and homogenous,whereas the ML and HL colonies are larger (;400 mm),and the largest HL colonies are structurally heteroge-nous. In the largest colonies, the cell density decreaseswith increasing distance from the colony surface untilthe central domain, where there are no more cells. Adecrease in cell density was also found in large colonies(100–2,000 mm) from a bloom of B. braunii, with a greensurface layer of cells of approximately 20 mm (Wake,1983), similar to the 20 to 60 mm found for B. brauniibiofilms (Wijihastuti et al., 2016). Additionally, the longfluorescence lifetimes for inner cells may be indicativeof nutrient starvation (no repair of photodamagedPSII). Together, this suggests that deeper inside thecolony (6200 mm; Fig. 2), light and/or nutrients be-come depleted, leading to cell death. Cell death andsubsequent degradation may have caused the cavity inthe center of these colonies. Diffusion of nutrients andlight attenuation can limit the viable region withincolonies (Ploug et al., 1999) and biofilms (Stewart,2003). Moreover, diffusion was recently shown to below in the oil-rich parts of the colony (van Schadewijket al., 2018), making this scenario very plausible.

Although the largest of the HL colonies have de-creased viability deep inside compared with small LLcolonies, this detrimental effect is apparently out-weighed by other advantages (Beardall et al., 2009),including increased mutual shading of cells. Moreover,the viable region is;200mm thick (Fig. 7), meaning thatmost HL colonies (typically 400 mm; Fig. 2) fully sustainliving cells and lack the hollow cavity (this is corrobo-rated by functional magnetic resonance imaging stud-ies; van Schadewijk et al., 2018). Additionally, thepossibility of programmed death of inner cells couldlead to nutrient recycling that could benefit othernutrient-limited inner cells (Durand et al., 2016), and

indications of remnants from cell death were identifiedin the B. braunii extracellular matrix (Weiss et al., 2012).

The HL acclimation of surface cells leads to lowerpigmentation, increased NPQ, and Echi shading, thuslargely diminishing PSII damage. The Fv/Fm of HLsurface cells is similar to that of faster growing speciessuch as C. reinhardtii (Bonente et al., 2012; Polukhinaet al., 2016), suggesting that inner cells are over-protected by the additional shading by layers of cellsand extracellular Echi. Perhaps the increased shadingfor inner cells in the HL colony is a more economicalway of photoprotection: the one-time cost of extraphotosynthetic complexes (to adapt to lower local lightintensity) may be less than the increased cost of repairthat would be required in the absence of shading.Moreover, the additional shading of blue photons willdecrease the photodamage caused by direct absorptionof these blue photons by the manganese cluster(Tyystjärvi, 2008), a source of photodamage uponwhich NPQ has no effect.

Light acclimation changes are considered to optimizegrowth of the algae under the new condition. This isachieved by increasing light harvesting under light-limited conditions and by limiting photodamage andoptimizing carbon assimilation in light-saturated con-ditions (MacIntyre et al., 2002; Falkowski and Chen,2003). Therefore, it was expected that directly afterchanging the light intensity (before acclimation), thegrowth rate would decrease, followed by an increase toa new constant rate. Light acclimation is usually fasteror equal to the doubling time (Falkowski and Chen,2003), and the doubling time of B. braunii is more than9 d (Fig. 1). Indeed, changes in pigment composition,Fv/Fm, andNPQ reach a new constant value within 5 d.During this period, the growth rate was thus expectedto recover to a new constant rate. However, the dryweight measurements (Fig. 1) gave no indication ofsuch an initial dip in growth rate and were well fittedwith a single exponential growth function. This sug-gests that the acclimation effects on the observedgrowth rate are small because acclimation is fast com-pared with the growth rate.

In conclusion, we show that B. braunii has a uniquestrategy for acclimation to HL, strictly correlated withits colonial nature: the accumulation of the extracellularEchi and intracolony shading by photosynthetic pig-ments. The capacity for NPQ is already substantial inlow light, and NPQ kinetics are unchanged upon HLacclimation. This acclimation strategy fits the environ-ment of this flagella-less alga that often floats near thesurface of lakes and ponds, where shade is minimal(Wake and Hillen, 1980).

MATERIALS AND METHODS

Strain and Culture Conditions

Botryococcus braunii Race B strain Berkeley (Showa; Wolf et al., 1985) wascultivated in a modified CHU-13 medium (composition is provided inSupplemental Table S3; Grung et al., 1989). The cultures were grown in batch or

1140 Plant Physiol. Vol. 179, 2019

Photoacclimation in Botryococcus braunii

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

fed-batch in a bubble column photobioreactor (MC1000; Photon SystemsInstruments) in a volume of 75 mL under continuous illumination from a cool-white LED at 26°C 6 1°C. Precultures were inoculated at a biomass concen-tration of 0.5 g L21 dry weight (see below) and grown at least 30 d at 150 mmolphotons m22 s21. From this preculture, five new cultures were inoculated at0.3 g L21 dry weight. Twowere grown at 50 mmol photons m22 s21 (LL), two at450 mmol photons m22 s21 (HL), and one control at 150 mmol photons m22 s21

(ML) for at least 40 d. In fed-batch conditions, 50% (v/v) of the medium wasreplaced every 5 to 6 d, and cultures were diluted once they reached 1.5 g L21.Nitrate concentrations were measured with Quantofix nitrate test sticks(Macherey-Nagel) during cultivation and did not decrease below 125 mg L21.

Dry Weight Measurements

Dry weight concentrations of batch cultures were determined by vacuumfiltration of 3mLof culture onto apreweighed, predried nitrocellulosefilterwitha pore size of 8 mm (N4146-100EA; Merck-Millipore) in a filter funnel (DS0315-0047; Nalgene). This pore size allows bacteria that are not associated with thecolonies to pass but is too small for B. braunii colonies. Filters with colonies werewashed by vacuum filtration with 10 mL of demineralized water in the filterfunnel (DS0315-0047; Nalgene), dried in an oven for 16 h at 80°C, and subse-quently weighed. All dry weight measurements were done in duplicate.

Colony Size Measurements

Colony size measurements (batch cultures) were performed by laser dif-fraction of the colonies in liquid culture analyzed with the Mie scattering model(Allen, 2013) with aMastersizer 2000 (Malvern Instruments) fittedwith aHydroSM sample dispersion unit with a dynamic range between 0.2 and 2,000 mm.Cultures were diluted with water until the diffraction laser obscuration was inthe range of 1% to 2%. Then, the full culture volume was flushed five timesthrough a capillary while continuously measuring the scattering of the 633-nmHeNe laser light (at right angle incidence) by the colonies. Colony size distri-butions were calculated with the Mie scattering model, and the diffraction datawere analyzed with Malvern software (Allen, 2013). Colonies were treated asspherical oil bodies with the refractive index of sunflower oil. The validity of themethod was confirmed with colonies separated by filters of specific pore size.

Pigment Extraction and Analyses

To determine Chl a/b and Chl/Car ratios according to Croce et al. (2002), wedisrupted the fed-batch or batch culture cells with a cell homogenizer andextracted the pigments with 80 (v/v)% acetone (Sigma-Aldrich). For determi-nation of the carotenoid stoichiometry (details in Supplemental Methods andSupplemental Fig. S12), pigments were extracted with a 4:1 (v/v) mixture ofmethanol (Sigma-Aldrich) and petroleum ether (low-boiling-point fraction[30°C–40°C], Sigma-Aldrich 168 UV-VIS detector; Beckman Coulter) using areverse-phase C18 sphereclone column (Phenomenex 5U ODS1, 00G-4143-E0,4.6 mm 3 250 mm; Xu et al. (2015) modified with the following time profile ofthemixing of the solvents (v/v): 0% B (buffer B), 30 s; 0% to 36% B, 2min; 36% B,5min; 36% to 100% B, 5 min; 100% B, 6 min; 100% to 0% B, 30s; and 0% B, 6 min.

NPQ of Chl Fluorescence

Quenching analyseswere performedwith aDual-PAM (Walz) while stirringat 22°C on fed-batch cultures. Cells were dark acclimated for 40 min in a 1-3 1-cm2 quartz optical cell equipped with a 3-mm stirring bar, at a concentration of2 g L21 dryweight. Each measuring sequence consisted of an initial period withmeasuring light (460 nm, 5mmol photons m22 s21) of 5 minwith two saturatingpulses (635 nm, 180 ms, 12,000 mmol photons m22 s21), followed by a period of12 min with the AL (460 nm [550 or 2,090 mmol photons m22 s21] or 635 nm[520 mmol photons m22 s21]; spectra in Fig. 4) switched on with seven satu-rating pulses and a recovery period of 12 min with measuring light with threesaturating pulses. The saturating pulse light intensity was 12,000 mmol photonsm22 s21with 180-ms duration provided by a 635-nm LED. The estimated Fv/Fmwas calculated as Fv

Fm ¼ Fm2 F0Fm and NPQ was calculated as ¼ Fm2 Fm

0

Fm0 . Fm is themaximum fluorescence during a saturating pulse after dark acclimation, andFm9 is the maximum fluorescence during a saturating pulse during or afterexposure to the AL. F0 is the minimal fluorescence after dark acclimation.

Optical Reflection and Confocal Microscopy

The largest colonies from fed-batch cultures were cut in half and imagedwith a 43 air objective of the BI500 Microscope (VWR) equipped with an RGBCMOS camera (10MP; Motic). Illumination was performed from the side or topwith a Schott KL200 cold light source (Galvoptics) at maximum intensity.Autofluorescence microscopy was performed with a Zeiss LSM510 confocalscanning microscope with a Plan-Neofluar objective (103, numerical aperture0.3) with a step size of 2.54 (x), 2.54 (y), and 5 (z) mm (field size 1,3033 1,3033495 mm). Excitation was at 488 nm and detection either above 635 nm (redchannel) or between 505 and 545 nm (blue-green channel).

FLIM

Time-resolved fluorescence of individual chloroplasts within the largestsingle HL colonies from fed-batch culture wasmeasured bymultiphoton FLIM,with a setup described earlier (Broess et al., 2009). In brief, two-photon excita-tion was performed with 860-nm, 150-fs pulses at a repetition rate of 76 MHz(generatedwith a Ti:sapphire laser [Mira900; Coherent]). Excitation pulses werefocused on the sample with a 603 water-immersion objective lens (CFI PlanApochromat, numerical aperture 1.2; Nikon). Imaging was done with a NikonTE300 inverted scanning microscope equipped with an HPM-100-40 detector(Becker & Hickel). Chl autofluorescence was recorded through a 680 6 20 nmband-pass filter (Semrock), and SYTOX Green fluorescence and auto-fluorescence were recorded through a 5256 45 nm band-pass filter (Semrock).Each image consisted of 128 3 128 pixels (153.6 3 153.6 m, 5-m step size in z)with each containing a decay trace of 256 time channels (20 ps per channel).Colonies were acclimated to HL for 55 d (fed-batch). Single colonies were darkacclimated for at least 20 min and moved to a four-well microscope chamber(Ibidi) with fresh medium. Dead cells were stained with 5 mM SYTOX Green(Molecular Probes; Sato et al., 2004). This approach was validated with a pos-itive control of cells killed with 96% ethanol (data not shown). Different laserpowers (40mW, 160mW, 1mW, and 2mWat the sample) were tested to identifya power (160 mW) with sufficient excitation for good signal to noise but lowenough to prevent singlet-singlet or singlet-triplet annihilation or the closing ofPSII reaction centers, ensuring measurements in the F0 state (Supplemental Fig.S13).

Eachmeasurement series started with imaging cells on the surface of a dark-acclimated colony. A z-scan, consisting of images at three horizontal planes(vertically separated by 5 mm), was recorded for several colony regions thatwere in direct contact with the bottom of the sample chamber. Next, cells moreburied in the colony were imaged. To this end, the colony was cut in half with arazor blade and the z-scan was repeated. Care was taken that only cells at thesurface of the cut were measured to avoid distortion by layers of extracellularmatrix or other cells. Dead cells that displayed SYTOX Green fluorescence werequantified and excluded from analysis. Fluorescence decays of individual cellswere integrated by selecting all voxels in the cup-shaped chloroplast (usingGlotaran 1.5; Snellenburg et al., 2012). The resulting decay traces were fittedwith Fluofit software (version 4.6; PicoQuant) with a sum of exponentialdecays, convoluted with the instrument response function (170 ps) that wasmeasured with the same setup using pinacyanol iodide in methanol (van Oortet al., 2008) in an identical sample chamber.

TCSPC

Time-resolved fluorescence measurements were performed using the time-correlated single-photon counting setup described previously (Müller et al.,1992) with modifications (Tian et al., 2017). Excitation at 650 nm was used topreferentially excite Chl b and 662 nm for Chl a (0.5-mm spot size). Time-resolved fluorescence decay traces were measured at three detection wave-lengths (685, 700, and 715 nm), with 4,096 time channels of 2.569-ps width.Colonies from fed-batch culture were placed between two glass plates andmounted in the rotation optical cell (diameter = 10 cm, thickness = 1 mm). Theoptical cell was rotating at 50 rpm and oscillating sideways (78 rpm). Fluores-cence was measured in a front-face arrangement, with PSII in open (F0) orclosed (Fm) state, using the following conditions. (1) For Fm, 50 mM 3-(3,4-dichlorophenyl)-1,1-dimethylurea was added to the growth medium. The ex-citation power was 50 mW, and the repetition rate was 4 MHz. To achieve fullclosure of the PSII reaction centers during the measurement, an additional blueLED (460 nm, intensity ;40 mmol photons m22 s21) was used to preilluminatethe cells just before detection of the signal. (2) For F0, the sample was in com-plete darkness during the experiment and dark adapted 20 min before the

Plant Physiol. Vol. 179, 2019 1141

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

experiment. The excitation power was 7 to 10 mW, and the repetition rate was0.8 MHz. Checks with different powers and repetition rates showed that thesample was in the F0 state (Supplemental Fig. S4A).

The measurement time at a single wavelength was limited to a maximum of20 min to avoid changes in the colonies due to prolonged measurement in therotating optical cell. Each first experiment was repeated at the end to make surethat the fluorescence decay did not change during the measurement(Supplemental Fig. S4B). The decay traces were globally analyzed with TRFAsoftware (Digris et al., 1999). The methodology of global analysis was describedpreviously (van Stokkum et al., 2004). In short, a number of parallel, nonin-teracting kinetic components were used as a kinetic model, so the total data set

was fitted with function f (t, l): fðt; lÞ ¼ ∑N1;2.DASiðlÞexp

�2 t

ti

�⊗irf ðt;lÞ,

where DASi is the wavelength-dependent amplitude factor associated with adecay component i having a decay lifetime ti, and instrument response function(t, l) is measured by scattering light. Four to six components were enough to geta good fit of the LL and HL data upon 650- or 662-nm excitation (for residualsand fitted traces, see Supplemental Figs. S5 and S6).

Supplemental Data

The following supplemental materials are available.

Supplemental Figure S1. Colony size distributions.

Supplemental Figure S2. NPQ induction curves.

Supplemental Figure S3. Absorption spectra (integrating sphere).

Supplemental Figure S4. TCSPC: powerseries and control experiments.

Supplemental Figure S5. TSCPC: decay curves F0.

Supplemental Figure S6. TSCPC: decay curves Fm.

Supplemental Figure S7. TSCPC: DAS F0.

Supplemental Figure S8. TSCPC: DAS Fm.

Supplemental Figure S9. Ultrafast green autofluorescence.

Supplemental Figure S10. FLIM: HL colony.

Supplemental Figure S11. FLIM: LL colony.

Supplemental Figure S12. HPLC chromatogram.

Supplemental Figure S13. FLIM: powerseries.

Supplemental Table S1. Carotenoid composition in the ML controlculture.

Supplemental Table S2. TCSPC: global fit results.

Supplemental Table S3. Medium composition.

Supplemental Methods. Pigment extraction.

ACKNOWLEDGMENTS

Dr. Joao Gouveia is acknowledged for the kind gift of the B. braunii Showastrain and assistance with the colony size measurements. Dr. Arjen Bader isacknowledged for providing training and technical assistance for the micros-copy measurements.

Received December 12, 2018; accepted January 8, 2019; published January 16,2019.

LITERATURE CITED

Allen T (2013) Particle Size Measurement. Springer, New YorkAllorent G, Tokutsu R, Roach T, Peers G, Cardol P, Girard-Bascou J,

Seigneurin-Berny D, Petroutsos D, Kuntz M, Breyton C, et al (2013) Adual strategy to cope with high light in Chlamydomonas reinhardtii. PlantCell 25: 545–557

Anderson JM, Andersson B (1988) The dynamic photosynthetic membraneand regulation of solar energy conversion. Trends Biochem Sci 13:351–355

Bachofen R (1982) The production of hydrocarbons by Botryococcus braunii.Experientia 38: 47–49

Ballottari M, Dall’Osto L, Morosinotto T, Bassi R (2007) Contrasting be-havior of higher plant photosystem I and II antenna systems during ac-climation. J Biol Chem 282: 8947–8958

Barry AN, Starkenburg SR, Sayre RT (2015) Strategies for optimizing algalbiology for enhanced biomass production. Front Energy Res 3:

Beardall J, Allen D, Bragg J, Finkel ZV, Flynn KJ, Quigg A, Rees TAV,Richardson A, Raven JA (2009) Allometry and stoichiometry of unicel-lular, colonial and multicellular phytoplankton. New Phytol 181: 295–309

Ben-Amotz A, Avron M (1983) On the factors which determine massivebeta-carotene accumulation in the halotolerant alga Dunaliella bardawil.Plant Physiol 72: 593–597

Bonente G, Ballottari M, Truong TB, Morosinotto T, Ahn TK, Fleming GR,Niyogi KK, Bassi R (2011) Analysis of LhcSR3, a protein essential forfeedback de-excitation in the green alga Chlamydomonas reinhardtii.PLoS Biol 9: e1000577

Bonente G, Pippa S, Castellano S, Bassi R, Ballottari M (2012) Acclimationof Chlamydomonas reinhardtii to different growth irradiances. J BiolChem 287: 5833–5847

Broess K, Borst JW, van Amerongen H (2009) Applying two-photon exci-tation fluorescence lifetime imaging microscopy to study photosynthesisin plant leaves. Photosynth Res 100: 89–96

Browne DR, Jenkins J, Schmutz J, Shu S, Barry K, Grimwood J, ChiniquyJ, Sharma A, Niehaus TD, Weiss TL, et al (2017) Draft nuclear genomesequence of the liquid hydrocarbon-accumulating green microalga Bo-tryococcus braunii Race B (Showa). Genome Announc 5: 16–17

Christa G, Cruz S, Jahns P, de Vries J, Cartaxana P, Esteves AC, Serôdio J,Gould SB (2017) Photoprotection in a monophyletic branch of chlor-ophyte algae is independent of energy-dependent quenching (qE). NewPhytol 214: 1132–1144

Croce R, van Amerongen H (2014) Natural strategies for photosyntheticlight harvesting. Nat Chem Biol 10: 492–501

Croce R, Canino G, Ros F, Bassi R (2002) Chromophore organization in thehigher-plant photosystem II antenna protein CP26. Biochemistry 41:7334–7343

Demmig-Adams B (1990) Carotenoids and photoprotection in plants: A rolefor the xanthophyll zeaxanthin. Biochim Biophys Acta Bioenerg 1020:1–24

Deng Z, Hu Q, Lu F, Liu G, Hu Z (2008) Colony development and physi-ological characterization of the edible blue-green alga, Nostoc sphaer-oides (Nostocaceae, Cyanophyta). Prog Nat Sci 18: 1475–1483

Digris AV, Skakoun VV, Novikov EG, van Hoek A, Claiborne A, VisserAJWG (1999) Thermal stability of a flavoprotein assessed from associa-tive analysis of polarized time-resolved fluorescence spectroscopy. EurBiophys J 28: 526–531

Drop B, Yadav KNS, Boekema EJ, Croce R (2014) Consequences of statetransitions on the structural and functional organization of photosystem Iin the green alga Chlamydomonas reinhardtii. Plant J 78: 181–191

Durand PM, Sym S, Michod RE (2016) Programmed cell death and com-plexity in microbial systems. Curr Biol 26: R587–R593

Falkowski PG, Chen YB (2003) Photoacclimation of light harvesting systemsin eukaryotic algae. Adv Photosynth Respir 13: 423–447

Farooq S, Chmeliov J, Wientjes E, Koehorst R, Bader A, Valkunas L,Trinkunas G, van Amerongen H (2018) Dynamic feedback of the pho-tosystem II reaction centre on photoprotection in plants. Nat Plants 4:225–231

García-Cubero R, Cabanelas ITD, Sijtsma L, Kleinegris DMM, Barbosa MJ(2018) Production of exopolysaccharide by Botryococcus braunii CCALA778 under laboratory simulated Mediterranean climate conditions. AlgalRes 29: 330–336

Goldschmidt-Clermont M, Bassi R (2015) Sharing light between two pho-tosystems: Mechanism of state transitions. Curr Opin Plant Biol 25: 71–78

Grung M, Metzger P, Liaaen-Jensen S (1989) Primary and secondary ca-rotenoids in two races of the green alga Botryococcus braunii. BiochemSyst Ecol 17: 263–269

Grung M, Metzger P, Berkaloff C, Liaaen-Jensen S (1994) Studies on theformation and localization of primary and secondary carotenoids in thegreen alga Botryococcus braunii, including the regreening process. CompBiochem Physiol B Biochem 107: 265–272

Iermak I, Vink J, Bader AN, Wientjes E, van Amerongen H (2016) Visu-alizing heterogeneity of photosynthetic properties of plant leaves with

1142 Plant Physiol. Vol. 179, 2019

Photoacclimation in Botryococcus braunii

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.

two-photon fluorescence lifetime imaging microscopy. Biochim BiophysActa Bioenerg 1857: 1473–1478

Khatri W, Hendrix R, Niehaus T, Chappell J, Curtis WR (2014) Hydro-carbon production in high density Botryococcus braunii Race B continu-ous culture. Biotechnol Bioeng 111: 493–503

Kim HS, Weiss TL, Thapa HR, Devarenne TP, Han A (2014) A microfluidicphotobioreactor array demonstrating high-throughput screening for mi-croalgal oil production. Lab Chip 14: 1415–1425

Kress E, Jahns P (2017) The dynamics of energy dissipation and xanthophyllconversion in Arabidopsis indicate an indirect photoprotective role ofzeaxanthin in slowly inducible and relaxing components of non-photochemical quenching of excitation energy. Front Plant Sci 8: 2094

Le Quiniou C, Tian L, Drop B, Wientjes E, van Stokkum IHM, van Oort B,Croce R (2015) PSI-LHCI of Chlamydomonas reinhardtii: Increasing theabsorption cross section without losing efficiency. Biochim Biophys ActaBioenerg 1847: 458–467

MacIntyre HL, Kana TM, Anning T, Geider RJ (2002) Photoacclimation ofphotosynthesis irradiance response curves and photosynthetic pigmentsin microalgae and cyanobacteria. J Phycol 38: 17–38

Meneghesso A, Simionato D, Gerotto C, La Rocca N, Finazzi G,Morosinotto T (2016) Photoacclimation of photosynthesis in the Eustig-matophycean Nannochloropsis gaditana. Photosynth Res 129: 291–305

Metzger P, Largeau C (2005) Botryococcus braunii: A rich source for hy-drocarbons and related ether lipids. Appl Microbiol Biotechnol 66:486–496

Metzger P, Berkaloff C, Casadevall E, Coute A (1985) Alkadiene andbotryococcene-producing races of wild strains of Botryococcus braunii.Phytochemistry 24: 2305–2312

Müller MG, Griebenow K, Holzwarth AR (1992) Primary processes inisolated bacterial reaction centers from Rhodobacter sphaeroides studiedby picosecond fluorescence kinetics. Chem Phys Lett 199: 465–469

Nicol L, Croce R (2018) Light harvesting in higher plants and green algae. InR Croce, R van Grondelle, H van Amerongen, I van Stokkum, eds, LightHarvesting in Photosynthesis. CRC Press, Boca Raton, FL, pp 73–90

Nonomura A (1988) Botryococcus braunii var. Showa (Chlorophyceae) fromBerkeley, California, United States of America. Jpn J Phycol 36: 285–291

Parkhill J, Maillet G, Cullen JJ (2001) Fluorescence based maximal quantumyield for ps2 as a diagnostic of nutrient stress. J Phycol 37: 517–529

Peers G, Truong TB, Ostendorf E, Busch A, Elrad D, Grossman AR,Hippler M, Niyogi KK (2009) An ancient light-harvesting protein iscritical for the regulation of algal photosynthesis. Nature 462: 518–521

Ploug H, Stolte W, Epping EHG, Jørgensen BB (1999) Diffusive boundarylayers, photosynthesis, and respiration of the colony-forming planktonalga, Phaeocystis sp. Limnol Oceanogr 44: 1949–1958

Polukhina I, Fristedt R, Dinc E, Cardol P, Croce R (2016) Carbon supplyand photoacclimation cross talk in the green alga Chlamydomonas rein-hardtii. Plant Physiol 172: 1494–1505

Quaas T, Berteotti S, Ballottari M, Flieger K, Bassi R, Wilhelm C, Goss R(2015) Non-photochemical quenching and xanthophyll cycle activities insix green algal species suggest mechanistic differences in the process ofexcess energy dissipation. J Plant Physiol 172: 92–103

Sakamoto K, Baba M, Suzuki I, Watanabe MM, Shiraiwa Y (2012) Opti-mization of light for growth, photosynthesis, and hydrocarbon produc-tion by the colonial microalga Botryococcus braunii BOT-22. BioresourTechnol 110: 474–479

Sato M, Murata Y, Mizusawa M, Iwahashi H, Oka S (2004) A simple andrapid dual-fluorescence viability assay for microalgae. Microbiol CultCollect 20: 53–59

Snellenburg JJ, Laptenok SP, Seger R, Mullen KM, van Stokkum IHM(2012) Glotaran: A Java-based graphical user interface for the R packageTIMP. J Stat Softw 49: 1–22

Solovchenko AE (2013) Physiology and adaptive significance of secondarycarotenogenesis in green microalgae. Russ J Plant Physiol 60: 1–13

Stewart PS (2003) Diffusion in biofilms. J Bacteriol 185: 1485–1491Suzuki R, Ito N, Uno Y, Nishii I, Kagiwada S, Okada S, Noguchi T (2013)

Transformation of lipid bodies related to hydrocarbon accumulation in agreen alga, Botryococcus braunii (Race B). PLoS ONE 8: e81626

Tian L, Xu P, Chukhutsina VU, Holzwarth AR, Croce R (2017) Zeaxanthin-dependent nonphotochemical quenching does not occur in photosystem I

in the higher plant Arabidopsis thaliana. Proc Natl Acad Sci USA 114:4828–4832

Tibiletti T, Auroy P, Peltier G, Caffarri S (2016) Chlamydomonas reinhardtiiPsbS protein is functional and accumulates rapidly and transiently underhigh light. Plant Physiol 171: 2717–2730

Tyystjärvi E (2008) Photoinhibition of photosystem II and photodamage ofthe oxygen evolving manganese cluster. Coord Chem Rev 252: 361–376

Ünlü C, Drop B, Croce R, van Amerongen H (2014) State transitions inChlamydomonas reinhardtii strongly modulate the functional size ofphotosystem II but not of photosystem I. Proc Natl Acad Sci USA 111:3460–3465

van den Berg TE, van Oort B, Croce R (2018) Light-harvesting complexes ofBotryococcus braunii. Photosynth Res 135: 191–201

van Oort B, Amunts A, Borst JW, van Hoek A, Nelson N, van AmerongenH, Croce R (2008) Picosecond fluorescence of intact and dissolved PSI-LHCI crystals. Biophys J 95: 5851–5861

van Oort B, Alberts M, de Bianchi S, Dall’Osto L, Bassi R, Trinkunas G,Croce R, van Amerongen H (2010) Effect of antenna-depletion in pho-tosystem II on excitation energy transfer in Arabidopsis thaliana. BiophysJ 98: 922–931

van Schadewijk RV, Berg TEVD, Gupta KBSS, Ronen I, de Groot HJM,Alia A (2018) Non-invasive magnetic resonance imaging of oils in Bo-tryococcus braunii green algae: Chemical shift selective and diffusion-weighted imaging. PLoS ONE 13: e0203217

van Stokkum IHM, Larsen DS, van Grondelle R (2004) Global and targetanalysis of time-resolved spectra. Biochim Biophys Acta Bioenerg 1657:82–104

Wake LV (1983) Characteristics of resting state colonies of the alga Bo-tryococcus braunii obtained from a bloom of the organism. Aust J Bot 31:605–614

Wake L, Hillen L (1980) Study of a “bloom” of the oil‐rich alga Botryococcusbraunii in the Darwin River Reservoir. Biotechnol Bioeng XXII: 1637–1656

Weiss TL, Johnston JS, Fujisawa K, Sumimoto K, Okada S, Chappell J,Devarenne TP (2010) Phylogenetic placement, genome size, and GCcontent of the liquid-hydrocarbon-producing green microalga Bo-tryococcus braunii strain Berkeley (Showa) (Chlorophyta). J Phycol 46:534–540

Weiss TL, Roth R, Goodson C, Vitha S, Black I, Azadi P, Rusch J,Holzenburg A, Devarenne TP, Goodenough U (2012) Colony organi-zation in the green alga Botryococcus braunii (Race B) is specified by acomplex extracellular matrix. Eukaryot Cell 11: 1424–1440

Wientjes E, van Amerongen H, Croce R (2013a) LHCII is an antenna of bothphotosystems after long-term acclimation. Biochim Biophys Acta Bio-energ 1827: 420–426

Wientjes E, van Amerongen H, Croce R (2013b) Quantum yield of chargeseparation in photosystem II: Functional effect of changes in the antennasize upon light acclimation. J Phys Chem B 117: 11200–11208

Wijihastuti RS, Moheimani NR, Bahri PA, Cosgrove JJ, Watanabe MM(2016) Growth and photosynthetic activity of Botryococcus braunii bio-films. J Appl Phycol 29: 1123–1134

Wobbe L, Bassi R, Kruse O (2016) Multi-level light capture control in plantsand green algae. Trends Plant Sci 21: 55–68

Wolf FR, Nonomura AM, Bassham JA (1985) Growth and branched hy-drocarbon production in a strain of Botryococcus braunii (Chlorophyta).J Phycol 21: 388–396

Xu P, Tian L, Kloz M, Croce R (2015) Molecular insights into zeaxanthin-dependent quenching in higher plants. Sci Rep 5: 13679

Yamano T, Miura K, Fukuzawa H (2008) Expression analysis of genes as-sociated with the induction of the carbon-concentrating mechanism inChlamydomonas reinhardtii. Plant Physiol 147: 340–354

Yoshimura T, Okada S, Honda M (2013) Culture of the hydrocarbon pro-ducing microalga Botryococcus braunii strain Showa: Optimal CO2, sa-linity, temperature, and irradiance conditions. Bioresour Technol 133:232–239

Zhang K, Kojima E (1998) Effect of light intensity on colony size of micro-alga Botryococcus braunii in bubble column photobioreactors. J FermentBioeng 86: 573–576

Plant Physiol. Vol. 179, 2019 1143

van den Berg et al.

www.plantphysiol.orgon August 2, 2020 - Published by Downloaded from Copyright © 2019 American Society of Plant Biologists. All rights reserved.