Structure and function of natural sulphide-oxidizingmicrobial mats under dynamic input of light andchemical energy
Judith M Klatt1, Steffi Meyer1, Stefan Häusler1, Jennifer L Macalady2, Dirk de Beer1 andLubos Polerecky1,3
1Microsensor Research Group, Max Planck Institute for Marine Microbiology, Bremen, Germany;2Pennsylvania State University, University Park, State College, PA, USA and 3Department of EarthSciences—Geochemistry, Faculty of Geosciences, Utrecht University, Utrecht, The Netherlands
We studied the interaction between phototrophic and chemolithoautotrophic sulphide-oxidizingmicroorganisms in natural microbial mats forming in sulphidic streams. The structure of these matsvaried between two end-members: one characterized by a layer dominated by large sulphur-oxidizingbacteria (SOB; mostly Beggiatoa-like) on top of a cyanobacterial layer (B/C mats) and the other withan inverted structure (C/B mats). C/B mats formed where the availability of oxygen from the watercolumn was limited (o5 μM). Aerobic chemolithotrophic activity of the SOB depended entirely onoxygen produced locally by cyanobacteria during high light conditions. In contrast, B/C mats formedat locations where oxygen in the water column was comparatively abundant (445 μM) andcontinuously present. Here SOB were independent of the photosynthetic activity of cyanobacteriaand outcompeted the cyanobacteria in the uppermost layer of the mat where energy sources for bothfunctional groups were concentrated. Outcompetition of photosynthetic microbes in the presence oflight was facilitated by the decoupling of aerobic chemolithotrophy and oxygenic phototrophy.Remarkably, the B/C mats conserved much less energy than the C/B mats, although similar amountsof light and chemical energy were available. Thus ecosystems do not necessarily develop towardsoptimal energy usage. Our data suggest that, when two independent sources of energy are available,the structure and activity of microbial communities is primarily determined by the continuous ratherthan the intermittent energy source, even if the time-integrated energy flux of the intermittent energysource is greater.The ISME Journal (2016) 10, 921–933; doi:10.1038/ismej.2015.167; published online 25 September 2015
Introduction
Life is driven by essentially two types of energy:chemical energy, in the form of chemical disequili-bria, and physical energy carried by light of suitablewavelengths. On geological timescales, changes inenergy fluxes and the development of new strategiesof life are closely coupled. One of the most importantchanges in energy availability was the evolution andproliferation of oxygenic photosynthesis. The onset ofthis process introduced the thermodynamically mostfavourable electron-acceptor oxygen to the widerreduced environment. Oxygenic photosynthesis there-fore provided previously unexplored thermodynamicdisequilibria to other microorganisms thereby allow-ing for the evolution and diversification of aerobic
metabolisms (Castresana and Saraste, 1995; Dismukeset al., 2001; Catling et al., 2005).
Insights into the revolutionary milestones asso-ciated with the evolution and proliferation ofoxygenic photosynthesis can be gained by studyingmodern model ecosystems in environments whosegeochemical/energetic characteristics exhibit spatialgradients that are analogous to the temporaltransition from a reduced to an oxidized state ofthe Earth's surface. The thin microbial mats formingat the bottom of light-exposed cold sulphidic springsat Frasassi, Italy, (Galdenzi et al., 2008) are such ananalogue system as they are distributed across agradient from reduced to oxidized conditions in theoverlying water column. These contemporary photo-trophic microbial mats are of particular interest asthey represent analogues to ancient cyanobacterialmats (for example, stromatolites) that are thought tohave been extensive in shallow waters throughoutthe Proterozoic and possibly already in the Archean(Ward et al., 1992; Grotzinger and Knoll, 1999; Allwoodet al., 2009; Seckbach and Oren, 2010; Schopf, 2012).
Correspondence: JM Klatt or L Polerecky, Microsensor ResearchGroup, Max Planck Institute for Marine Microbiology, Celsius-strasse 1, Bremen 28359, Germany.E-mail: [email protected] or [email protected] 19 January 2015; revised 8 August 2015; accepted12 August 2015; published online 25 September 2015
In this study, we used microsensors to quantifyenergy fluxes available for and conserved by thedominant processes in the mats—photosynthesis (P)and aerobic sulphide oxidation (SO)—with the aimto understand how the interaction between theseprocesses under a fluctuating input of chemical andlight energy determines the structure and function ofthe mats. We hypothesized that under the differentconditions the mats develop towards a system thatconserves the available energy optimally. We discussthe important role of energy dynamics on microbialmat structure and possible implications of ourresults in the context of Earth’s oxygenation.
Materials and methods
Study siteThis study was performed in streams and poolswhere the sulphidic waters from the Frasassi Cavesystem in the Frasassi Gorge, Italy, emerge and mixwith waters of the Sentino river (Galdenzi et al.,2008). The study sites included the outflows of twoperennial springs, the Fissure Spring and the MainSpring (43°24ʹ4″N, 12°57ʹ56″E).
Water chemistrySampling for bulk water chemistry analysis wascarried out in May and September 2009 and inSeptember 2012. Samples were taken during nightand around midday within a few cm above themicrobial mat patches and were preserved immedi-ately in gas-tight vials containing a mixture of 100 μlof 20% ZnCl2 and 50 μl of saturated HgCl2 solution.The vials were filled with water samples until therewas no head space and kept at room temperatureuntil quantification in Bremen. Dissolved inorganiccarbon and ammonium were determined using flowinjection analysis (Hall and Aller, 1992). The sum ofnitrate and nitrite was quantified according toBraman and Hendrix (1989) using an NOx analyserequipped with a chemiluminescence detector(Model CLD 66, Eco Physics, Dürnten, Switzerland).pH and concentrations of dissolved O2 and totalsulphide (Stot = [S2− ]+[HS−]+[H2S]) were determinedwith microsensors. Temperature at the mat surfacewas measured with a PT1000 mini-sensor (Umwelt-sensortechnik, Geschwenda, Germany).
MicroscopyThe dominant members of the mat community wereidentified by microscopy. Imaging by bright field,phase contrast and fluorescence microscopy wascarried out using an Axiophot epifluorescencemicroscope (Zeiss, Jena, Germany). Photopigmentswere identified on a single-cell level by hyperspec-tral imaging (Polerecky et al., 2009).
MicrosensorsO2, pH and H2S microsensors with a tip diameter of10–80 μm and response time of o1 s were built,calibrated and used as described previously(Revsbech, 1989; Jeroschewski et al., 1996; de Beeret al., 1997). Calibration of the H2S microsensors wasperformed in acidified spring water (pHo2) towhich NaS2 was added stepwise. The total sulphideconcentrations, Stot, in the calibration solutions weredetermined according to Cline (1969). Calculation ofStot from the local H2S concentrations and pH valuesmeasured with microsensors was carried out accord-ing to Millero (1986), using the pK1 value of7.14–7.17 depending on temperature (Jeroschewskiet al., 1996; Wieland and Kühl, 2000).
In situ microsensor and light measurements over a dielcycleIn situ measurements of O2, pH and H2S in the matswere carried out using previously described micro-sensor setups (Weber et al., 2007, www.microsen-wiki.net). Parallel O2, pH and H2S profiles weremeasured under naturally variable light conditionsby measuring continuously over a complete dielcycle on a cloudless day.
In parallel to microsensor measurements, down-welling irradiance of the photosynthetically avail-able radiation was recorded using a calibrated lightlogger (Odyssey Dataflow Systems, Christchurch,New Zealand) or a calibrated scalar irradiancemicroprobe (Lassen et al., 1992) placed next tothe microsensors. The irradiance microprobe wasadditionally used to quantify reflectance of the mats,as previously described (Al-Najjar et al., 2010).
Microsensor measurements under controlled lightconditionsMeasurements under controlled light conditionswere performed both in situ and ex situ. In situ, wefirst measured steady-state profiles of O2, H2S andpH in the dark and at an incident irradiance of 650μmol photonsm− 2 s− 1 generated by a halogen lamp(KL1500, Schott, Müllheim, Germany) correspond-ing to the natural illumination around midday.Subsequently, volumetric rates of gross oxygenic Pwere determined using the O2 microsensor-basedlight–dark shift method of Revsbech and Jørgensen(1983). Finally, the same light–dark shift approachwas applied using H2S and pH microsensors insteadof a O2 microsensor, which allowed quantification ofvolumetric rates of gross anoxygenic P in terms ofconsumed Stot (Klatt et al., 2015). For this measure-ment, the light was switched on and off over a fewminute intervals while the signals were recorded in0.3-s intervals. All of these measurements werecarried out in the same spot of the mat.
To confirm that the cyanobacterial population inthe mats was able to perform simultaneous oxygenicand anoxygenic P, similar measurements were
Energy conversion in illuminated sulphidic springsJM Klatt et al
performed ex situ. The mat sample was placed in atemperature-controlled (15 °C) flow chamber, andO2, H2S and pH were measured in parallel in thesame spot of the mat under variable illuminationfrom the halogen lamp. To assess the potential role ofobligate anoxygenic phototrophs that can use light inthe near infrared part of the spectrum and H2S as theelectron donor for anoxygenic P, similar measure-ments were additionally performed using nearinfrared light-emitting diodes (maximal emission atλmax = 740 and 810 nm; H2A1 series, Roithner-Laser-technik, Vienna, Austria). The measurements withand without the additional near infrared light werecarried out before and after the addition of 5 μM ofDCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea;Aldrich, Seelze, Germany), a specific inhibitor ofoxygenic P.
Calculation of fluxes and daily energy budgetsDiffusive fluxes of O2 and Stot were calculated fromthe measured concentration gradients multiplied bythe corresponding diffusion coefficients and factorscorrecting for the azimuthal angle at which theprofiles were measured (Berg et al., 1998; Polereckyet al., 2007). Diffusion coefficients, D, were correctedfor temperature, salinity and, in case of Stot, for thecomposition of the total sulphide pool as previouslydescribed (Sherwood et al., 1991; Wieland and Kühl,2000). Specifically, diffusion coefficients were1.78 ×10− 5 cm2 s−1 for O2 and 1.35× 10− 5 cm2 s− 1
for Stot, varying by p3% depending on the tempera-ture in each particular measurement.
Fluxes of incident light energy were estimatedfrom the measured downwelling irradiance, assum-ing that the average energy content of photons ofphotosynthetically available radiation is 217.5 kJ(mol photons)− 1 (Al-Najjar et al., 2010). The fractionof the incident light energy absorbed by thecyanobacterial population in the mats was estimatedfrom the measured reflectance of the mats (seeSupplementary Information 1).
Fluxes of energy conserved by oxygenic andanoxygenic P as well as by aerobic SO wereestimated by multiplying the estimated CO2 fluxeswith the molar energy required for CO2 fixation bythe respective process. For oxygenic and anoxygenicP, the CO2 flux was estimated from the measuredfluxes of O2 produced and Stot consumed inthe photosynthetically active zone assuming thestoichiometry of O2/CO2 =1 and Stot/CO2= 2, respec-tively. For CO2 fixation coupled to aerobic SO, weemployed the approach of Klatt and Polerecky(2015), which allows the prediction of the completestoichiometry of autotrophic SO based on themeasured Stot/O2 flux ratios. Prerequisite for thesecalculations is the relationship between the S0/SO4
2−
production ratio and the energy conservation effi-ciency, that is, the ratio between the energy demandfor CO2 reduction and the energy available from O2
reduction. To derive this relationship, we made two
main assumptions: (i) When SO is in a steady state,sulphide is oxidized completely to SO4
2− (Jørgensenet al., 2010); (ii) When the Stot/O2 consumption ratioin the SO layer varies strongly over a diel cycle, thehighest measured Stot/O2 consumption ratio occurswhen sulphide is oxidized incompletely to S0,whereas the lowest measured Stot/O2 consumptionratio corresponds to complete sulphide oxidation toSO4
2− . These assumptions enabled us to determinethe energy conservation efficiency as a function ofthe S0/SO4
2− production ratio, based on which weestimated the fluxes of CO2, energy gained andenergy conserved by SO from the measured fluxes ofStot and O2 at different times during the diel cycle.Details of the CO2 and energy flux calculations aregiven in Supplementary Information 2.
Daily budgets were calculated by integrating theinstantaneous fluxes over the 24-h period for allrelevant reactants and energy-harvesting processesinvolved.
Results
Mat structure vs chemical composition of the springwaterThe light-exposed microbial mats from the Frasassisulphidic springs are dominated by two functionalgroups: cyanobacteria, identified by the presence ofthe characteristic cyanobacterial pigments chloro-phyll a and phycocyanin in the cells revealed byhyperspectral microscopy (Supplementary Figure S1),and large Beggiatoa-like sulphur-oxidizing bacteria(SOB; filamentous, containing sulphur inclusions,motile through gliding) (Figures 1a and b). Inferringfrom microscopy, hyperspectral imaging and reflec-tance measurements of the mats (SupplementaryFigure S2), the abundance of phototrophs other thancyanobacteria was not significant in the studied matlayers.
A close inspection revealed that the structure ofthe mats fell essentially between two end-members:mats with a distinct cyanobacterial layer on top of adistinct SOB layer (Figure 1c), and mats with SOB ontop of cyanobacteria (Figure 1d). These two mattypes are hereafter referred to as C/B and B/C mats,respectively.
In the spring outflow streams, the distribution ofthese two mat types was patchy and varied on acentimetre to decimetre scale. Water chemistry andlight measurements revealed that the physico-chemical parameters directly above the mats alsochanged at this scale. This was mainly due to mixingof the spring water with the freshwater from theSentino river under highly spatially variable flowconditions, which lead, for instance, to the formationof stagnant anoxic pools next to an aerated watercolumn. The formation of a certain mat type did notcorrelate with the daily light dose, with temperature,pH, nitrate, ammonium, dissolved inorganic carbon,H2S or Stot concentrations in the water column
Energy conversion in illuminated sulphidic springsJM Klatt et al
923
The ISME Journal
during day and night nor with the maximum H2Sand Stot concentrations measured inside the mat(Table 1 and Supplementary Table S1). However, thelocations harbouring the two end-member mat typesclearly differed with respect to the dissolved O2
concentration in the water column above the mats,with the C/B and B/C mats exclusively found inareas where the O2 concentration during the nightwas o5 μM and 445 μM, respectively (Table 1).Changes in the water column O2 concentrationduring the day, which occurred owing to the highrates of oxygenic P in the mats when the overlyingwater column was stagnant, had no effect on theformation of a certain mat type (compare data formats C/B-1 and C/B-2 in Supplementary Figure S3and Supplementary Table S1).
Photosynthetic activity of the cyanobacterialcommunityLaboratory microsensor measurements revealedlight-induced production of O2 and consumption ofH2S in the cyanobacterial layer of C/B (Figure 2) andB/C mats (data not shown). Although O2 productionwas due to oxygenic P, removal of H2S could becaused by three processes: chemical or biologicallymediated oxidation with the photosyntheticallyproduced O2, a shift in the equilibrium of sulphidespeciation (H2S, HS− and S2− ) induced by a pHincrease associated with the uptake of dissolvedinorganic carbon by oxygenic P, or by anoxygenic Pthat uses H2S as the electron donor. First, H2Soxidation with O2 could be excluded because theH2S dynamics changed abruptly upon light–darktransitions and were independent of O2 concentra-tions. Additionally, the depth-integrated rate of grossoxygenic P matched the net rate derived from steady-state diffusive profiles (see below), showing thatthere is negligible oxygen reduction activity (such asowing to the reaction with H2S) in the photosynthe-tically active layer. Second, assuming that totalsulphide concentrations, Stot, stayed constant duringthe light–dark transitions and pH varied as measured(by about 0.005 pH units; Figure 2), the calculated
variation in H2S would be about 1–2 orders ofmagnitude lower than measured (data not shown).Therefore, we exclude also the second possibilityand conclude that the measured light-inducedvariation in H2S was a direct result of anoxygenic Pin the cyanobacterial layer of the mat.
This conclusion was confirmed by ex situ mea-surements in a C/B mat treated with DCMU, aspecific inhibitor of oxygenic P but not of anoxygenicP. Specifically, H2S concentrations in the DCMU-treated mat decreased and increased upon theaddition and removal of light, respectively, whilethe effect on pH was marginal (data not shown) andO2 was below detection limit (Figure 3a). For theincident irradiance of 100 μmol photonsm− 2 s− 1, thevolumetric rates of anoxygenic P derived from light–dark shift microsensor measurements rangedbetween 1 and 4mmol Stot m−3 s− 1 (Figure 3b), andtheir depth-integrated value (1.53 μmol Stot m− 2 s− 1)closely matched the flux of Stot removed bythe cyanobacterial anoxygenic P derived from thesteady-state microprofiles (1.55 μmol Stot m− 2 s− 1).This implies that anoxygenic P was the onlysignificant sink of sulphide in the cyano-bacterial layer.
The observed light-induced variability of Stot couldalso be due to the activity of obligate anoxygenicphototrophs. The activity of such phototrophs is,however, expected to be affected by the local oxygenconcentration. For example, filamentous anoxygenicphototrophs (also known as green non-sulphurbacteria) can switch from sulphide-driven anoxy-genic P to photoorganoheterotrophy or aerobicrespiration when oxygen and cyanobacterialexcudates are available in the light (Van der Meeret al., 2005; Polerecky et al., 2007). Additionally,anaerobic anoxygenic phototrophs, such as greenand purple sulphur bacteria, are expected to bepoisoned by high oxygen concentrations (VanGemerden and Mas, 1995). In our measurements,however, the net and gross rates of anoxygenic Pbefore the addition of DCMU to the mat sample, thatis, in the presence of oxygenic P, were almost equalto those measured after inhibition of oxygenic P by
Figure 1 Microscopic images of the dominant cyanobacterial (a) and SOB (b) morphotypes in sulphide-oxidizing microbial mats from thelight-exposed sulphidic streams in Frasassi, Italy, and photographs of end-member mat structures (c: C/B mat; d: B/C mat) (scale bars: a–b:10 μM, c: 20 cm. d: 5 cm).
Energy conversion in illuminated sulphidic springsJM Klatt et al
924
The ISME Journal
DCMU (data not shown), suggesting that obligateanoxygenic phototrophs were not significantly con-tributing to the light-driven sulphide consumption inthe cyanobacterial layer. This was confirmed by theobservation that exposure to near infrared light,which specifically targets bacteriochlorophylls ofobligate anoxygenic phototrophs, did not inducesulphide consumption (Supplementary Figure S4).
Together these data show that the cyanobacterialcommunity in the studied mats is able to performoxygenic and sulphide-driven anoxygenic P simul-taneously and that cyanobacterial anoxygenic P isfully responsible for the measured light-induced H2Svariation in the cyanobacterial layer of the mats.
Activity of C/B matsContinuous in situmicro-profiling in C/B mats undernaturally variable illumination gave consistentresults with those obtained in the laboratory. In thedark, sulphide consumption rates were minute orT
able
1Physico-ch
emical
param
etersin
theov
erlyingwater
columnan
din
thestudiedmats
Mat
type
Site
H2S(S
tota)(μ
M)b
O2(μ
M)b
DIC
(mM)c,d
pH
bNH
4+
(μM)c,d
NO
x
(μM)c,d
PARe
(μmol
photon
sm
−2s−
1)
T(°C)
Wf
Cf
SOBf
C/B
-1MainSpring
73(582
)34
5(275
5)51
0(333
9)0
4.2(±0.18
)8.0
77(±3.3)
34.4
(±1.9)
133
14C/B
-2MainSpring
23(151
)36
(287
)36
(353
)5
3.9(±0.6)
8.0
56(±4.7)
60(±1.8)
9314
C/B
-3Fissu
reSpring
401(628
)41
5(642
)43
5(681
)5
8.9
6.9
150
o1
7814
.5B/C
-1MainSpring
89(278
)12
5(390
)95
(175
)45
4.4(±
.15)
7.5
206(±
10.2)
3.6(±0.07
)12
413
B/C
-2MainSpring
50(92)
81(149
)73
(88)
654.7(±0.11
)7.3
−−
163
13B/C
-3Fissu
reSpring
16(248
)61
3480
−8.3
o1
50−
15
Abb
reviations:
DIC,dissolved
inorganic
carbon
;PAR,photosyn
thetically
availableradiation
;SOB,su
lphur-ox
idizingba
cteria.
aTotal
sulfideco
ncentration,Stot,was
calculatedfrom
themeasu
redH
2Sco
ncentrationan
dpH.
bValues
arederived
from
microsensormeasu
remen
tsperform
edduringnight.Theva
lues
obtained
duringtheday
aresh
ownin
Supplemen
tary
Tab
leS2.
cValues
weredetermined
from
water
samplestake
nduringnight.Theva
lues
determined
from
samplestake
nduringtheday
aresh
ownin
Supplemen
tary
Tab
leS2.
dValues
inparen
theses
referto
s.d.(n
=3–
6).
eDow
nwellingirradiance
ofthePAR,give
nas
theav
eragephoton
fluxduring15
hof
ligh
t.f Con
centrationsin
thewater
column(W
)insidethecy
anob
acterial
laye
r(C)an
dthelaye
rdom
inated
bylargeSOB,as
derived
from
microsensormeasu
remen
ts.
I
Figure 2 Light-induced dynamics of O2, H2S, Stot and pH insidethe cyanobacterial layer of a freshly collected C/B mat. Totalsulphide concentrations, Stot, were calculated from the measuredH2S concentrations and pH. The incident irradiance was 30 μmolphotonsm−2 s− 1. Distance between sensor tips was about 5mm.
Figure 3 Ex situ microsensor measurements in a C/B mat afterthe addition of DCMU. (a) Steady-state depth profiles of O2 andH2S and Stot measured in the light (open symbols) and in the dark(filled symbols). (b) Depth profile of volumetric rates of anoxy-genic photosynthesis, as derived from the light–dark shift methodadapted for Stot. For both panels, total sulphide concentrations,Stot, were calculated from the measured H2S concentrations andpH, and the incident irradiance during the light measurementswas 100 μmol photonsm−2 s−1.
Energy conversion in illuminated sulphidic springsJM Klatt et al
925
The ISME Journal
below detection limit, as implied by essentially flatStot profiles across the mats during the night, and O2
was not detected (Supplementary Figure S3). Atsunrise, while still at low light, the rate of sulphideremoval owing to anoxygenic P in the cyanobacteriallayer sharply increased, resulting in a gradualdecrease in H2S concentrations (Figure 4a). Aroundmidday, when the incident irradiance exceededabout 300 μmol photonsm−2 s− 1, sulphide concen-trations within the cyanobacterial layer decreasedand the photosynthetic production of oxygen sharplyincreased (Figure 4a). These dynamics occurredessentially in reverse order at sunset.
In situ measurements under controlled lightconditions revealed additional insights into theactivity of C/B mats. At high incident irradiance(650 μmol photonsm− 2 s− 1), the cyanobacterialcommunity performed both oxygenic and anoxy-genic P simultaneously, as shown by the light–darkshift measurements (Figure 5b). Aerobic respirationin the photosynthetically active zone was negligible,as the net diffusive flux of oxygen from the zone wasvery similar to the gross rates of oxygenic P depth-integrated over the zone (2.43 and 2.47 μmolm−2 s−1,respectively). Although oxygenic P was detectable atdepths 0–0.6mm, anoxygenic P was detectable onlyat depths 0–0.4mm, that is, in zones of thecyanobacterial layer where both H2S and light wereabundant. The overlap between the zone of anoxy-genic photosynthetic activity and the decrease in Stot
from the overlying water indicated that the sulphide
used by the cyanobacteria originated exclusivelyfrom the overlying water. This was supported by theclose match between the depth-integrated rates ofgross anoxygenic P (1.56 μmol Stot m− 2 s− 1) and thedownward Stot flux from the water column (1.55μmol Stot m− 2 s− 1) (Figures 5a and b). When oxygenicP was, however, not active (that is, in the morningand later in the afternoon), the flux of Stot consumedby anoxygenic P had significant contributions fromboth the downward and upward components of thediffusive Stot flux.
Sulphide consumption in the SOB layer wasdetectable only after the onset of oxygenic P by thecyanobacteria (Figures 4b and 5a). This observationis consistent with the absence of an external supplyof an electron acceptor, that is, O2, from the watercolumn and additionally indicates that anaerobicsulphide oxidation with NO3
− as the electron accep-tor was not important. Once O2 became available,sulphide consumed by the SOB originated exclu-sively from below, while the sulphide entering themat from above was consumed by the cyanobacteria(see above).
Analysis of the profiles in Figure 5a revealed thatthe photosynthetically produced O2 (downward flux0.85 μmol O2m−2 s− 1) and the sulphide suppliedfrom below (upward flux 0.48 μmol Stot m− 2 s− 1)were consumed at a Stot/O2 flux ratio of about 0.6.However, a detailed analysis of the complete sets ofprofiles measured in the C/B mat over a diel cycle(Supplementary Figure S3) revealed that the Stot/O2
l l
l
l
l l
l
ll
ll
ll
ll
Figure 4 Activity of the cyanobacterial (a and c) and SOB (b and d) populations in the C/B (a and b) and B/C (c and d) mats exposed tonatural light fluctuations over a 24-h period. The fluxes were derived from in situ microsensor profiles of O2 and H2S concentration(Supplementary Figures S5 and S6) and pH. The corresponding average H2S concentration in the cyanobacterial layer, and the Stot/O2 fluxratio in the SOB layer, are also shown. Open and shaded areas correspond to light and dark, respectively.
Energy conversion in illuminated sulphidic springsJM Klatt et al
926
The ISME Journal
consumption ratio changed substantially over thecourse of the day, reaching about 2 after O2 becameavailable, gradually decreasing to about 0.7 beforeincreasing back to about 2.6 just before O2 depletiontowards the night (Figure 4b). Similar patterns in thediurnal variations of the Stot/O2 consumption ratiobetween 0.7 and 2.6 were also observed in a replicateC/B mat (Supplementary Figures S5A and B).
The observed diurnal variation in the Stot/O2
consumption ratio in the SO zone can be explainedby assuming that the SOB activity varied betweenincomplete oxidation of sulphide to S0 and completeoxidation of sulphide to SO4
2− when the Stot/O2 ratioreached the maximum and minimum value, respec-tively. This assumption made it possible to estimate
the energy conservation efficiency of the SOBcommunity. Specifically, using the maximum valueof Stot/O2 = 2.6 (oxidation to S0) and additionallytaking into account the measured local concentra-tions of the substrates and products involved, wefound the efficiency values of 16.6% and 16.5% forthe two replicate C/B mats. Conversely, the mini-mum value of Stot/O2 = 0.7 (complete oxidation toSO4
2− ) measured in the respective mats gave theefficiency values of 17.2% and 17.1% (calculationdetails given in Supplementary Information 2.2).These calculations suggest that the energy conserva-tion efficiency of the SOB community in the studiedmats was around 16.9% and did not significantlydepend on the S0/SO4
2− production ratio.The Stot/O2 consumption ratio in both replicate
C/B mats was significantly negatively correlatedwith the downward O2 flux from the cyanobacteriallayer (R2 = 0.91, Po0.0001), as well as with the localconcentration of O2 (R2 = 0.91, Po0.0001) and H2S(R2 = 0.66, P=0.008), but not correlated with theupward Stot flux (R2 = 0.27, P=0.15; SupplementaryFigure S6). In one of the C/B mats, the availability ofsulphide from below was low and thus the minimumStot/O2 consumption ratio of 0.7, which indicatescomplete oxidation of sulphide to sulphate (see above),was reached early during the day (SupplementaryFigure S5B). When this ratio was reached, weobserved a pronounced downward shift of thezone where sulphide was aerobically consumed(Supplementary Figure S5C). This downward shiftdid not affect the correlation between the Stot/O2
consumption ratio and the O2 flux or the local O2 andH2S concentrations.
Taken together, our data suggest that in the layerunderneath the photic zone aerobic sulphide oxida-tion by SOB was the only significant sink of oxygenin the layer while aerobic heterotrophic activity wasnegligible. Therefore, we conclude that the SOBpopulation in the C/B mats most likely varied itsactivity between incomplete and complete aerobicsulphide oxidation to S0 and SO4
2− , respectively(that is, at a variable S0/SO4
2− production ratio),whereby this variability was regulated by the avail-ability of O2 produced locally by the overlyingcyanobacteria and by the local H2S concentrationbut not by the flux of sulphide from below.
Activity of B/C matsDespite the fact that B/Cmats were exposed to similarlyhigh incident light intensities as the C/B mats (Table 1),sulphide removal in these mats owing to anoxygenic Pwas not very pronounced (Figure 4c). Also oxygenic Pwas very low or even below detection limit in the B/Cmats (Figure 4c; Supplementary Figure S7). This wasconfirmed by ex situ measurements under controlledlight conditions. Specifically, the depth-integrated ratesof gross anoxygenic and oxygenic P were about oneorder of magnitude lower than in C/B mats at similarirradiances and remained low even when the applied
l
l
Figure 5 Microsensor measurements in the C/B and B/C matsunder artificially controlled light conditions. Measurements in theC/B (a and b) and B/C mats (c and d) were performed in situ andex situ, respectively. Left panels show steady-state depth profilesof O2, Stot and pH in the light (open symbols) and in the dark(closed symbols). Right panels show the corresponding depthprofiles of volumetric rates of oxygenic and anoxygenic photo-synthesis, as derived from light–dark shift microsensor measure-ments of O2 and Stot, respectively. For all panels, total sulphideconcentrations, Stot, were calculated from the measured H2Sconcentrations (not shown) and pH. The incident irradianceduring the light measurements in the C/B and B/C mats was 650and 710 μmol photonsm−2 s−1, respectively. For orientation,approximate structures of the mats are also shown.
Energy conversion in illuminated sulphidic springsJM Klatt et al
927
The ISME Journal
irradiance exceeded the maximal values measuredin situ (Figure 5d; Supplementary Figure S7). Addi-tionally, the rate of O2 consumption in the cyano-bacterial layer (such as by aerobic respiration),calculated as the difference between the gross andnet rates of O2 produced in the layer (Figures 5c and d),was not significant in the B/C mats. Also, the depth-integrated rates of gross anoxygenic P (0.20 μmolStot m− 2 s−1) matched closely the net Stot flux (0.19μmol Stot m− 2 s− 1) consumed in the cyanobacteriallayer, indicating that anoxygenic P was the onlysignificant sink of sulphide in this layer. The sourceof this photosynthetically removed sulphide wasexclusively from below.
Sulphur and oxygen cycling in B/C mats wasdominated by the light-independent aerobic chemo-lithotrophic SO in the top SOB layer (Figures 4dand 5c). In the dark, sulphide for aerobic SO originatedfrom both the water column and the underlyingsediment. During high light conditions, the downwardsulphide flux from the water column was the mainsource, while the upward sulphide flux was mostlyconsumed in the cyanobacterial layer (see above;Figure 5c). Additionally, during the maximum inci-dent irradiance, consumption of O2 produced locallyby the underlying cyanobacterial layer only contrib-uted o15% to the overall O2 consumption in the SOBlayer. Thus SO in the top layer relied mostly on O2
supplied externally from the overlying water column.Over the course of the day, the Stot/O2 consumption
ratio remained constant at about 0.67 (Figure 4d).Thus aerobic SO by the SOB community in the B/Cmats was in steady state and therefore most likelyproceeded completely to sulphate (Nelson et al.,1986; Jørgensen et al., 2010). Using this stoichiometry,we estimated the energy conservation efficiency of theSOB community in the two replicate B/C mats to be16.4% and 17.1% (calculation details given inSupplementary Information 2.2), which is verysimilar to the value estimated for the SOB communityin the C/B mats (see above).
Light absorption by the cyanobacteriaC/B and B/C mats back-reflected 4.5% and 81% ofthe incident irradiance, respectively. This suggeststhat the estimated fractions of the incident fluxof light energy absorbed by the cyanobacterialpopulations were 95.5% and 19%, respectively(see Supplementary Information 1). However, thesevalues likely define the upper limits of the lightenergy utilized by cyanobacteria in the mats, as lightabsorption could occur also owing to abiotic compo-nents in the mats (Al-Najjar et al., 2012).
Carbon and energy budgetsAssuming that photosynthesis by cyanobacteria andaerobic sulphide oxidation by SOB were the onlyprocesses responsible for oxygen and sulphidecycling, we used the measured fluxes of light, oxygen
and sulphide together with the estimated energyconservation efficiency of the SOB populations toestimate daily carbon and energy budgets in the twostudied mat types (Table 2). The C/B mats conservedabout 2.42% of the incident light energy directly by P(split into 1.83% by oxygenic and 0.59% by anoxy-genic P). This conservation efficiency was increased toabout 2.54% when additionally considering that thephotosynthetically produced oxygen is used for SOB-driven carbon fixation coupled to aerobic chemosyn-thetic SO. When expressed in terms of fixed carbon,the exploitation of the thermodynamic disequilibriumbetween O2 and sulphide, which was internallygenerated by oxygenic P, thus increased the totalprimary productivity in the C/B mats by 15% (from67.5 to 79.5mmol Cm−2 d−1).
In contrast, the B/C mats conserved only about0.12% of the incident light energy by P (mainly byanoxygenic P), although the daily flux of light energyavailable for both mat types was similar. Instead,energy conservation in the B/C mats was dominatedby chemosynthesis (about 69% of the total energyconserved), although the daily flux of light energyavailable to the system was about 67-fold larger thanthe flux of chemical energy (in the form of oxygen andsulphide diffusing from the water column, utilized byaerobic SO). Thus, because of the small contributionof P, the overall energy conversion efficiency in the B/C mats was only 0.37%, that is, sevenfold lower thanin the C/B mats. This grossly inefficient utilization ofthe available light energy in the B/C mats is alsoreflected in the estimated overall primary productiv-ity, which was about threefold lower than in the C/Bmats (23.4 vs 79.5mmol Cm−2 d−1; Table 2). Analysisof the replicate measurements led to similar conclu-sions although with slightly different numerical values(Supplementary Table S2).
Discussion
The dominant functional groups in the Frasassispring mats, photosynthetic cyanobacteria andaerobic chemolithoautotrophic SOB are directlycoupled as both oxygen and sulphide are involvedin their energy-generation pathways. The twofunctional groups, however, depend on differentenergy sources, that is, on light and chemical energy,respectively. Our results show that activity in themats is driven by both energy sources. However,depending on the direction and temporal dynamicsof the energy supply, the mats stratified in essen-tially two distinct structures, C/B mats and B/C mats,characterized by substantially different activitypatterns and, most strikingly, utilization efficienciesof the externally available energy.
C/B matsIn C/B mats, cyanobacteria inhabiting the photiclayer switched between anoxygenic and oxygenic
Energy conversion in illuminated sulphidic springsJM Klatt et al
928
The ISME Journal
P over a diurnal cycle, with microzones performingoxygenic and anoxygenic P simultaneously (Figures 4and 5). As C/B mats formed in areas characterized bya very low O2 concentration (o5 μM) in the overlyingwater (Table 1), oxygenic P was the exclusive providerof electron acceptor for aerobic SO.
Because of the fluctuating availability of light, thesupply of oxygen to the SOB population residingunder the cyanobacterial population was also fluc-tuating. A possible adaptation of SOB to live undersuch conditions is to rapidly adjust their SOstoichiometry, that is, to vary the S0/SO4
2− productratio, depending on the availability of O2 and Stot,whereby the range of Stot/O2 consumption rationsthat can be covered by the SOB through the variationbetween the complete and incomplete oxidation ofsulphide to sulphate and zero-valent sulphur,respectively, is determined by the correspondingenergy-conservation efficiency (Klatt and Polerecky,2015). Our data suggest that this strategy wasadopted by the SOB in the C/B mats. Specifically,we estimated that the dominant SOB performedaerobic SO with an energy conservation efficiency of~ 16.9%, which allowed them to adjust their Stot/O2
consumption ratio to the range of Stot/O2 flux ratiosimposed by the environment (between ~0.7 and~2.6) and thus harvest the available fluxes ofsulphide and oxygen optimally.
Interestingly, when the Stot/O2 consumption ratioreached the minimum value of 0.7, that is, whensulphide oxidation proceeded entirely to sulphate, afurther increase in the rate of oxygen supply byoxygenic P led to a downward migration of the SOB.We suggest that this is because the SOB could notadjust the Stot/O2 consumption ratio below theminimum value determined by their energyconservation efficiency (Klatt and Polerecky, 2015),leaving the adjustment of their position as the onlyoption to maintain optimal utilization of theavailable substrates in the dynamically changinggradients of sulphide and O2. There are two plausiblehypotheses concerning the exact trigger formigration: (i) the biomass-dependent maximum rateof O2 consumption was reached, or (ii) the rate of O2
consumption by the SOB became limited by thesupply of H2S. In both cases, any further increase inoxygenic P would lead to an increase in the local O2
concentration triggering downward migration owingto a phobic response to O2 (Møller et al., 1985).
Intriguingly, the Stot/O2 consumption ratio wasonly determined by the O2 flux and the local H2Sconcentration but not by the Stot flux (SupplementaryFigure S6). This is consistent with the expectationthat SOB should adjust their S0/SO4
2− productionratio so as to maximize utilization of the available O2.This is because the carbon yield per oxygen isexpected to be almost constant irrespective of theS0/SO4
2− production ratio, that is, the O2 reductionrate directly translates into growth, as suggested bythe fact that the energy-conversion efficienciesestimated for the incomplete and complete SO inT
able
2Daily
budgets
ofco
mpou
ndsan
den
ergy
utilize
dby
photosyn
thesis
(P,an
oxyg
enic
andox
ygen
ic)an
dby
aerobicsu
lphideox
idation(SO)in
thetw
oen
d-m
embe
rmat
types
C/B
mat
B/C
mat
P(anox
yP;ox
yP)
SO
Total
(P;SO)
P(anox
yP;ox
yP)
SO
Total
(P;SO)
O2fluxa
(mmol
O2m
−2d
−1)
43.93
−34
.11
9.82
0.0
−56
.30
−56
.30
Stotfluxa
(mmol
Stotm
−2d
−1)
−47
.12
−32
.28
−79
.34
−8.63
−37
.67
−46
.30
CO
2fluxa
(mmol
CO
2m
−2d
−1)
−67
.49(34.9%
;65
.1%)
−12
.04
−79
.53(84.9%
;15
.1%)
−4.32
(100
%;0%
)−19
.06
−23
.38(18.5%
;81
.5%
)Energy
fluxav
ailableb
(kJm
−2d
−1)
1572
.53
13.22c
1585
.75(99.2%
;0.8%
)14
55.74
21.68
1477
.34(98.5%
;1.5%
)Energy
fluxco
nserved
(kJm
−2d
−1)
38.10(24.4%
;75
.6%
)2.24
40.34(94.5%
;5.5%
)1.70
(100
%;0%
)3.71
5.41
(31.43
%;68
.6%)
Con
servationefficien
cyd
2.42
%(0.59%
;1.83
%)
16.9%
2.54
%(2.40%
;0.14
%)
0.12
%(0.12%
;0)
17.1%
0.37
%(0.12%
;0.25
%)
aPositivean
dnegativeva
lues
areusedwhen
theco
mpou
ndis
producedan
dco
nsu
med
,resp
ective
ly.
bEnergy
gained
byaerobicSO
was
assu
med
torepresentalso
theen
ergy
availableforthis
process.
cRep
resents
energy
generated
internally
within
theecosystem
asaco
nsequ
ence
ofox
ygen
icP.
dCalcu
latedas
energy
conserved
divided
byen
ergy
available.
Thetotalen
ergy
conservationefficien
cyof
thematsis
highligh
tedin
bold.
Energy conversion in illuminated sulphidic springsJM Klatt et al
929
The ISME Journal
the SOB inhabiting the C/B mats were almost equal(Klatt and Polerecky, 2015).
Together, the flexible O2-dependent adjustment ofthe S0/SO4
2− production ratio seems to allow the SOBto efficiently exploit the chemical energy in thesystem and to additionally build up storage ofintracellular S0 that might serve as an electronacceptor for anaerobic respiration during night whenthe O2 is not available (Schwedt et al., 2011). Thedominant SOB therefore appear to be highly adaptedto co-exist with oxygenic phototrophs.
Overall, under the oxygen-limited conditions inC/B mats, both photosynthetic and aerobic chemo-synthetic activity are regulated by light energysupplied to the system. Despite the dependence ofboth anoxygenic P and chemolithotrophy on H2S asan electron donor, its light-dependent depletion wasnot disadvantageous for the SOB because thechemolithotrophic and phototrophic layers did notcompete for the same sulphide pool. Specifically,chemolithotrophic activity was exclusively suppliedby the sulphide flux from below while the photo-synthetically oxidized sulphide originated from theoverlying water column. Aerobic SO by SOB, or forthat matter any aerobic activity, was directly coupledto both oxygenic and anoxygenic P, the latterrequired to locally deplete the reductant (H2S) so asto enable the former. Therefore, both anoxygenic andoxygenic photosynthetic activities were beneficialfor the SOB, as they together allowed for the life-enabling production of oxygen required for the SOBto thrive under the particular oxygen-limitingconditions where C/B mats developed.
B/C matsIn B/C mats, SOB formed a layer on top of thecyanobacteria, where they could access a continuoussupply of energy (H2S and O2) from the water columnthroughout the entire diel cycle (Figure 4d). Incontrast, the energy source (light) for the cyanobac-teria in the B/C mats was discontinuous. Wehypothesize that the continuity of the energy supplywas the main advantage that the SOB had overcyanobacteria in the B/C mats. Specifically, in theabsence of light during the night the cyanobacteriawere not triggered to assert themselves in theuppermost position of the mat, that is, closest tothe energy sources for both functional groups duringthe day. This has been taken advantage of by theSOB, whose chemolithotrophic activity in thisposition could continue uninterrupted. During theday the position of the cyanobacteria remained in thelower part of the mat, where the light availability wassignificantly reduced (at least fivefold) owing tointense back-scattering in the overlying layer of SOB(Supplementary Information 1). Therefore, the cya-nobacteria were not able to harvest the lightoptimally. It remains unclear how exactly the verydistinct layering was maintained during the day andwhy the cyanobacteria were not able to invade the
SOB layer. An important factor was probably fastermotility of the SOB that made them more successfulcompetitors for the 'prime spot' on top of the mats.
Importance of energy flux direction and dynamics format structure and functionWhen oxygen provided externally from the watercolumn was limiting, light availability regulated theactivity and spatial organization of the dominantfunctional groups, leading to the formation of C/Bmats. In these mats, primary productivity wasexclusively driven by light. Namely, light energywas directly utilized by anoxygenic and oxygenic Pand indirectly facilitated productivity of the SOBthat exploited the chemical energy in the form of thethermodynamic disequilibrium driven by photo-synthetically produced O2. Consequently, processesstratified predictably according to the direction andmagnitude of the energy source, and the dominantfunctional groups, photosynthetic cyanobacteria andaerobic SOB, beneficially interacted. This led to anefficient utilization of the available external energy.
B/C mats, on the other hand, were exposed to twoenergy sources, both of which had the same direc-tion, that is, they were externally supplied from thewater column, and were therefore most abundant atthe upper surface of the mat. One of these sources,namely, chemical energy in the form of oxygen andsulphide, was continuous, while the other, light, wasdiurnally fluctuating. As the functional groupscompeted for the space closest to their energysources, specific adaptation mechanisms (for exam-ple, motility) and phenotypic features (for example,light-scattering S0 globules) gained in importance.Intriguingly, the organisms specialized in utilizingchemical energy, SOB, outcompeted all photosyn-thetic microbes from the position closest to the light,even though the availability of light energy per daywas orders of magnitude higher than that of chemicalenergy. This means that the continuously availablechemical energy was used preferentially, while theadditional potential for gaining oxygen via internalrecycling of the other available external energysource (light) was neglected or even suppressed. Asa consequence, the competition for a favourableposition in the mat led to a comparatively inefficientuse of the bulk external energy.
It is also likely that nitrogen retention, storage andremoval in the two mat types strongly differ owing tothe differences in the O2 concentrations in the watercolumn overlying the C/B and B/C mats and thedifferences in the activity of the SOB and cyanobac-teria in the mats. To understand possible effects ofthis on the mat stratification, more detailed measure-ments of bioavailable nitrogen in the mats would berequired, which could, however, not be done duringthis study owing to technical limitations. Never-theless, the lack of correlation between the mat typeand the water column concentrations of NO3
− andNH4
+ (Table 1) suggests that nitrogen cycling is an
Energy conversion in illuminated sulphidic springsJM Klatt et al
930
The ISME Journal
unlikely factor that determines the formation of acertain mat type.
ImplicationsThe scenario in C/B mats resembles what is thoughtto have been the arena for the evolutionary transitionfrom anoxygenic to oxygenic P in the ancientphototrophic microbial mats in isolated shallowwater environments. The environment where thesecritical steps in evolution occurred was chemicallyreduced (for example, Tice and Lowe, 2004; Sessionset al., 2009; Lyons et al., 2014). Anoxygenic P usingthe available reductants, for example, H2S, isexpected to be an important process in suchenvironments. However, H2S can become locallydepleted, which provides a selective advantage tocyanobacteria that are able to switch from anoxy-genic to oxygenic P (Cohen et al., 1986; Jørgensenet al., 1986; Klatt et al., 2015) and that are thereforenever limited by electron donors. In C/B mats, thisversatility leads to the introduction of aerobichotspots in the otherwise reduced environment.
The C/B mats also demonstrate the revolutionaryconsequences of creating such aerobic hotspots.Oxygen as the most favourable electron acceptoroffers a myriad of thermodynamic strategies for othermicroorganisms, and oxygenic P has therefore set thebasis for the co-evolution of aerobic organisms.Initially, as the proliferation of oxygenic P andaerobic metabolisms took place in otherwise anoxicand reduced environments, these two processeswere spatially and temporarily closely coupled. Thisis illustrated in C/B mats, where aerobic SO wascompletely dependent on the internal conversion oflight energy into chemical energy driven by thephotosynthetic production of oxygen.
Despite the substantial consumption of oxygen byaerobic chemolithotrophy in the layer underneath thephotic zone, C/B mats were net sources of oxygen tothe water column. It is assumed that, in ancient, morereduced oceans, excess reductants have initiallyscavenged the photosynthetically produced oxygen,and it was only upon depletion of these sinks thatoxygen could persistently accumulate in the atmo-sphere and upper water column of the oceans duringthe Great Oxidation Event. However, the GreatOxidation Event was possibly predated by transientaccumulations of oxygen in the atmosphere (the‘whiffs of oxygen’) (for example, Anbar et al., 2007;Lyons et al., 2014), and it is not unlikely that, despitereductant availability, oxygen had also persistentlyaccumulated in the shallow water column abovemicrobial mats, similarly to the scenario in theFrasassi sulphidic springs. Independent of the exacttiming, water column oxygenation freed aerobicorganisms from consortia and microenvironmentswhere O2 was provided exclusively through the closeproximity to oxygenic phototrophs (that is, oxygenoases; for example, Buick, 2008). A paradoxicalconsequence of this is strikingly demonstrated in the
Frasassi B/C mats. Although these mats were exposedto light and contained significant populations ofoxygenic phototrophs, they became a net sink foroxygen once it accumulated in the water column. Thisis because they were mainly driven by an aerobic,light-independent process (SO). This demonstratesthat an aerobic metabolism, which had, by necessity,evolved dependent on light, could even outcompetethe oxygenic phototrophs that formerly served as theexclusive provider of oxygen for its energymetabolism.
The oxygenation of the Earth’s atmosphere repre-sents an event where oxygen has evolved from afluctuating internal source of chemical energy into acontinuous external source. We suggest that aerobicchemosynthesis might have become competitiveagainst photosynthesis, thus tempering photo-synthesis-driven primary productivity and providinga negative feedback on the proliferation of oxygenicphototrophic organisms. As demonstrated by theFrasassi mats, the effect of this negative feedbackcould have been as radical as a turn from a hugelyproductive ecosystem that acts as a net O2 source (C/Bmat) into a considerably less productive ecosystemthat acts as a net O2 sink (B/C mat). Further research isrequired to identify possible biosignatures of thiscompetitive interaction between phototrophs andaerobic chemolitotrophs that could be searched forin geological records to explore whether the negative-feedback hypothesis could be generalized beyond thehighly localized scale of this study.
Conflict of Interest
The authors declare no conflict of interest.
AcknowledgementsWe thank Daniel S Jones (University of Minnesota) forinvaluable help in the field, Alessandro Montanari andPaula Metallo for providing laboratory facilities andenjoyable atmosphere at the Osservatorio Geologico diColdigioco and Tim Ferdelman for fruitful discussions. Wethank the technicians of the microsensor group formicrosensor construction. This work was financiallysupported by the Max Planck Society, by a 2011 ‘ForWomen in Science Award’ to JMK and by NASA Astro-biology Institute (PSARC, NNA04CC06A) and Hanse-Wissenschaftskolleg Fellowship funds to JLM.
ReferencesAl-Najjar MAA, de Beer D, Jørgensen BB, Kühl M,
Polerecky L. (2010). Conversion and conservation oflight energy in a photosynthetic microbial mat ecosys-tem. ISME J 4: 440–449.
Al-Najjar MAA, de Beer D, Kühl M, Polerecky L. (2012).Light utilization efficiency in photosyntheticmicrobial mats. Environ Microbiol 14: 982–992.
Energy conversion in illuminated sulphidic springsJM Klatt et al
931
The ISME Journal
Allwood AC, Grotzinger JP, Knoll AH, Burch IW,Anderson MS, Coleman ML et al. (2009). Controls ondevelopment and diversity of Early Archean stromato-lites. Proc Natl Acad Sci USA 106: 9548–9555.
Anbar AD, Duan Y, Lyons TW, Arnold GL, Kendall B,Creaser RA et al. (2007). A whiff of oxygen before thegreat oxidation event? Science 317: 1903–1906.
Berg P, Risgaard-Petersen N, Rysgaard S. (1998). Inter-pretation of measured concentration profiles in sedi-ment pore water. Limnol Oceanogr 43: 1500–1510.
Braman RS, Hendrix SA. (1989). Nanogram nitrite andnitrate determination in environmental and biologicalmaterials by vanadium(III) reduction with chemilumi-nescence detection. Anal Chem 61: 2715–2718.
Buick R. (2008). When did oxygenic photosynthesis evolve?Philos Trans R Soc Lond B Biol Sci 363: 2731–2743.
Castresana J, Saraste M. (1995). Evolution of energeticmetabolism: the respiration-early hypothesis. TrendsBiochem Sci 20: 443–448.
Catling DC, Glein CR, Zahnle KJ, McKay CP. (2005). WhyO2 is required by complex life on habitable planets andthe concept of planetary 'oxygenation time'. Astrobiol-ogy 5: 415–438.
Cline JD. (1969). Oxygenation of hydrogen sulfide inseawater at constant salinity, temperature and pH.Environ Sci Technol 3: 838–843.
Cohen Y, Jørgensen BB, Revsbech NP, Poplawski R. (1986).Adaptation to hydrogen sulfide of oxygenic andanoxygenic photosynthesis among cyanobacteria. ApplEnviron Microbiol 51: 398–407.
de Beer D, Schramm A, Santegoeds CM, Kühl M. (1997).A nitrite microsensor for profiling environmentalbiofilms. Appl Environ Microbiol 63: 973–977.
Dismukes GC, Klimov V V, Baranov S V, Kozlov YN,DasGupta J, Tyryshkin A. (2001). The origin of atmo-spheric oxygen on Earth: the innovation of oxygenicphotosynthesis. Proc Natl Acad Sci USA 98:2170–2175.
Galdenzi S, Cocchioni M, Morichetti L, Amici V, Scuri S.(2008). Sulfidic ground-water chemistry in the FrasassiCaves, Italy. J Cave Karst Stud 70: 94–107.
Grotzinger JP, Knoll AH. (1999). Stromatolites in Precam-brian carbonates: evolutionary mileposts or environ-mental dipsticks? Annu Rev Earth Planet Sci 27:313–358.
Hall POJ, Aller RC. (1992). Rapid, small-volume, flowinjection analysis for ∑(CO2) and NH4+ in marine andfreshwaters. Limnol Oceanogr 37: 1113–1119.
Jeroschewski P, Steuckart C, Kühl M. (1996). An ampero-metric microsensor for the determination of H2S inaquatic environments. Anal Chem 68: 4351–4357.
Jørgensen BB, Cohen Y, Revsbech NP. (1986). Transitionfrom anoxygenic to oxygenic photosynthesis in aMicrocoleus chthonoplastes cyanobacterial mat. ApplEnviron Microbiol 51: 408–417.
Klatt JM, Al-Najjar MAA, Yilmaz P, Lavik G, de Beer D,Polerecky L. (2015). Anoxygenic photosynthesis con-trols oxygenic photosynthesis in a cyanobacteriumfrom a sulfidic spring. Appl Environ Microbiol AEM81: 2025–2031.
Klatt JM, Polerecky L. (2015). Assessment of the stoichio-metry and efficiency of CO2 fixation coupled toreduced sulfur oxidation. Front Microbiol 6: 484.
Lassen C, Ploug H, Jørgensen BB. (1992). A fibre-opticscalar irradiance microsensor: application for spectrallight measurements in sediments. FEMS Microbiol Lett86: 247–254.
Lyons TW, Reinhard CT, Planavsky NJ. (2014). The rise ofoxygen in Earth’s early ocean and atmosphere. Nature506: 307–315.
Millero F. (1986). The thermodynamics and kinetics of thehydrogen sulfide system in natural waters. Mar Chem18: 121–147.
Møller MM, Nielsen LP, Jørgensen BB. (1985). Oxygenresponses and mat formation by Beggiatoa spp. ApplEnviron Microbiol 50: 373–382.
Nelson DC, Jørgensen BB, Revsbech NP. (1986). Growthpattern and yield of a chemoautotrophic Beggiatoa sp.in oxygen-sulfide microgradients. Appl Environ Micro-biol 52: 225–233.
Polerecky L, Bachar A, Schoon R, Grinstein M, Jørgensen BB,de Beer D et al. (2007). Contribution of Chloroflexusrespiration to oxygen cycling in a hypersaline microbialmat from Lake Chiprana, Spain. Environ Microbiol 9:2007–2024.
Polerecky L, Bissett A, Al-Najjar MAA, Färber P, Osmers H,Suci PA et al. (2009). Modular spectral imagingsystem for discrimination of pigments in cells andmicrobial communities. Appl Environ Microbiol 75:758–771.
Revsbech NP, Jørgensen BB. (1983). Photosynthesisof benthic microflora measured with high spatialresolution by the oxygen microprofile method:capabilities and limitations of the method. LimnolOceanogr 28: 749–759.
Revsbech NP. (1989). An oxygen microsensor with a guardcathode. Limnol Oceanogr 34: 474–478.
Schopf J. (2012). The fossil record of Cyanobacteria.In: Whitton BA (ed). Ecology of Cyanobacteria II.Springer: Dordrecht, Netherlands, pp 15–36.
Schwedt A, Kreutzmann A-C, Polerecky L, Schulz-Vogt HN.(2011). Sulfur respiration in a marine chemoli-thoautotrophic Beggiatoa strain. Front Microbiol 2:276.
Seckbach J, Oren A (eds). (2010). Microbial Mats: Modernand Ancient Microorganisms in Stratified Systems.Springer: Dordrecht, Netherlands.
Sherwood JE, Stagnitti F, Kokkinn MJ, Williams WD.(1991). Dissolved oxygen concentrations inhypersaline waters. Limnol Oceanogr 36: 235–250.
Tice MM, Lowe DR. (2004). Photosynthetic microbialmats in the 3,416-Myr-old ocean. Nature 431:549–552.
Van der Meer MTJ, Schouten S, Bateson MM, Nübel U,Wieland A, Kühl M et al. (2005). Diel variations incarbon metabolism by green nonsulfur-like bacteria inalkaline siliceous hot spring microbial mats fromYellowstone National Park. Appl Environ Microbiol71: 3978–3986.
Van Gemerden H, Mas J. (1995). Ecology of phototrophicsulfur bacteria. In: Blankenship RE, Madigan MT,Bauer CE (eds). Anoxygenic Photosynthetic Bacteria,Advances in Photosynthesis and Respiration, Vol. 2.Springer Netherlands: Dordrecht, The Netherlands, pp49–85.
Energy conversion in illuminated sulphidic springsJM Klatt et al
932
The ISME Journal
intermittently oxygenic/anoxygenic, thermal, eukaryo-tic, and terrestrial. In: Schopf JW, Klein C (eds).The Proterozoic Biosphere: A Multidisciplinary Study.Cambridge University Press: NY, USA, pp 309–324.
Weber M, Färber P, Meyer V, Lott C, Eickert G, Fabricius KEet al. (2007). In situ applications of a new diver-operated
motorized microsensor profiler. Environ Sci Technol 41:6210–6215.
Wieland A, Kühl M. (2000). Short-term temperature effectson oxygen and sulfide cycling in a hypersalinecyanobacterial mat (Solar Lake, Egypt). Mar Ecol ProgSer 196: 87–102.
Supplementary Information accompanies this paper on The ISME Journal website (http://www.nature.com/ismej)
Energy conversion in illuminated sulphidic springsJM Klatt et al
