CO2 and vitamin B12 interactions determine bioactive trace metal requirements of a subarctic Pacific diatom.

0.45 MB·English

by  King, Andrew L

TheISMEJournal(2011)5,1388–1396
&2011InternationalSocietyforMicrobialEcology Allrightsreserved1751-7362/11
www.nature.com/ismej
ORIGINAL ARTICLE
CO and vitamin B interactions determine bioactive
2 12
trace metal requirements of a subarctic Pacific
diatom
Andrew L King1, Sergio A San˜udo-Wilhelmy2, Karine Leblanc3, David A Hutchins1
and Feixue Fu1
1DepartmentofBiologicalSciences,UniversityofSouthernCalifornia,LosAngeles,CA,USA;2Departmentof
Biological Sciences and Department of Earth Sciences, University of Southern California,
Los Angeles, CA, USA and 3Centre National de la Recherche Scientifique; Universite´ de la Me´diterrane´e;
Laboratoire d’Oce´anographie Physique et Bioge´ochimique, Marseille, France
Phytoplanktongrowthcanbelimitedbynumerousinorganicnutrientsandorganicgrowthfactors.
Using the subarctic diatom Attheya sp. in culture studies, we examined how the availability of
vitaminB andcarbondioxidepartialpressure(pCO )influencesgrowthrate,primaryproductivity,
12 2
cellulariron(Fe),cobalt(Co),zinc(Zn)andcadmium(Cd)quotas,andthenetuseefficiencies(NUEs)
ofthesebioactivetracemetals(molCfixedpermolcellulartracemetalperday).UnderB -replete
12
conditions,cellsgrownathighpCO hadlowerFe,ZnandCdquotas,andusedthosetracemetals
2
more efficiently in comparison with cells grown at low pCO . At high pCO , B -limited cells had
2 2 12
B50% lower specific growth and carbon fixation rates, and used Fe B15-fold less efficiently,
and Zn and Cd B3-fold less efficiently, in comparison with B -replete cells. The observed higher
12
Fe, Zn and Cd NUE under high pCO /B -replete conditions are consistent with predicted
2 12
downregulation of carbon-concentrating mechanisms. Co quotas of B -replete cells were B5- to
12
14-fold higher in comparison with B -limited cells, suggesting that 480% of cellular Co of B -
12 12
limited cells was likely from B . Our results demonstrate that CO and vitamin B interactively
12 2 12
influencegrowth,carbonfixation,tracemetalrequirementsandtracemetalNUEofthisdiatom.This
suggests the need to consider complex feedback interactions between multiple environmental
factorsforthisbiogeochemicallycriticalgroupofphytoplanktoninthelastglacialmaximumaswell
as the currentandfuture changing ocean.
The ISME Journal(2011)5, 1388–1396;doi:10.1038/ismej.2010.211; publishedonline20 January 2011
Subject Category: microbial ecosystem impacts
Keywords: phytoplankton; iron; vitamin B ; carbon dioxide; ocean acidification; trace metals
12
Introduction archaea (Rodionov et al., 2003). About half of
cultured eukaryotic phytoplankton species, includ-
Marine phytoplankton account for about half
ing many harmful algal bloom dinoflagellates, are
of global primary production (Behrenfeld and
unable to synthesize B and thus require an
Falkowski, 1997), and depend on the availability 12
exogenous supply (Provasoli and Carlucci, 1974;
of dissolved inorganic carbon, major nutrients and
Croft et al., 2005; Tang et al., 2010). The renewed
trace elements. Many eukaryotic phytoplankton
attention to B over the past decade has led to
also require a number of organic growth cofactors 12
numerous field studies that have demonstrated that
including cobalamin—vitamin B (Provasoli and
12 vitaminB isalimitingfactorinavarietyofmarine
Carlucci, 1974)—a cobalt (Co)-containing molecule 12
ecosystems, including temperate and polar coastal
synthesized by many, but not all, bacteria and
waters (San˜udo-Wilhelmy et al., 2006; Gobler et al.,
2007),iron(Fe)-limited/high-nutrientregimesofthe
SouthernOcean(Bertrandetal.,2007)andthehigh-
Correspondence: AL King, Department of Biological Sciences, nitrate, low-chlorophyll (HNLC) Gulf of Alaska
University of Southern California, 3616 Trousdale Pkwy,
(Koch et al., submitted).
LosAngeles,CA90089,USA.
E-mail:[email protected]˜udo-Wilhelmy,Department Another factor potentially affecting phytoplank-
of Biological Sciences and Department of Earth Sciences, tongrowththathasrecentlyreceivedattentionisthe
University of Southern California, 3616 Trousdale Pkwy, availability of dissolved inorganic carbon in the
LosAngeles,CA90089,USA.
contextofoceanacidification(reviewedbyHutchins
E-mail:[email protected]
et al., 2009). Past glacial–interglacial variability in
Received 14 October 2010; revised 10 December 2010; accepted
12December2010;publishedonline20January2011 carbon dioxide partial pressure (pCO ) based on
2
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1389
paleoclimate proxies (Petit et al., 1999) and the them through a column packed with Chelex-100
future predicted anthropogenic rise in pCO (Bio-Rad, Hercules, CA, USA; Price et al., 1988/89).
2
(Solomon et al., 2007) could influence phytoplank- For both B conditions, triplicate acid-washed
12
ton physiology and elemental requirements. For polycarbonate bottles were equilibrated with com-
instance, a higher CO availability could alleviate mercially prepared air:CO mixtures at three differ-
2 2
high metabolic energy and trace metal requirements ent CO concentrations: 200, 370 and 670p.p.m.
2
ofthecarbon-concentratingmechanismsthatsupply pCO (see below). In-line HEPA filters were used to
2
CO for photosynthesis (Raven, 1991; Morel et al., avoid trace metal and bacterial contamination from
2
1994). the gas tanks or lines.
Despite the global importance of both vitamin B Tracemetalcleansemi-continuousculturingmeth-
12
and pCO to phytoplankton metabolic pathways, ods were used during acclimation periods and
2
virtually nothing is known about the possible steady-state growth. Final sampling was carried
interactions between these two critical factors. We outfollowing4–8weeksofsemi-continuousincuba-
tested the effects of B and CO availability on tion after statistically invariant growth rates were
12 2
specific growth rates, primary productivity and the recorded for at least three consecutive transfers.
requirementsofmajornutrients(C,NandP)andFe, Culturesweregrownsemi-continuouslytomaintain
as well as other trace metals (zinc (Zn), Co and cells in the exponential growth phase to avoid bias
cadmium (Cd)) in steady-state semi-continuous resulting from sampling during different growth
cultures of Attheya sp., a centric diatom isolated phases in different experimental treatments. All
from the Bering Sea (an Fe-limited HNLC region; mediumhandling,culturingandmanipulationused
Leblanc et al., 2005). Attheya sp. belongs to the careful trace metal clean methods under class 100
family Chaetocerotaceae,and was previously classi- conditions.
fied in the genus Chaetoceros (Crawford et al., Seawater medium pCO in the experimental
2
2000),abiogeochemicallyimportantalgalgroupthat bottles was calculated throughout the experiment
dominates blooms in coastal upwelling regions and using measurements of pH and total dissolved
in mesoscale and bottle Fe addition experiments inorganic carbon according to Dickson and Goyet
(Hutchins and Bruland, 1998; de Baar et al., 2005). (1994). To ensure that CO levelsremained constant
2
duringgrowth,thepHineachbottlewasmonitored
daily using a microprocessor pH meter, calibrated
Materials and methods withpH4,7and10buffersolutions.Totaldissolved
inorganic carbon was measured via coulometry
Unialgal axenic Attheya sp. (CCMP207) stock cul- (modelCM140,UIC,Joliet,IL,USA).Thecalculated
tures, isolated from the Bering Sea, were obtained pCO values (using CO2SYS; http://www.cdiac.
2
from the Provasoli-Guillard Culture Collection of ornl.gov/ftp/co2sys/CO2SYS_calc_XLS) from the
Marine Phytoplankton (West Boothbay Harbor, final day of the experiment were 201±13, 367±21
Maine, USA) and grown in microwave-sterilized and 671±31p.p.m. (Table 1); these treatments are
media prepared from 0.2mm-filtered natural sea- referred tousingrounded-offvaluesof200,370and
water(collectedusingtracemetalcleantechniques). 670p.p.m. pCO .
2
Media had Aquil concentrations of nitrate, phos- Primary production was measured in triplicate
phate andsilicicacidwith additions of5mMEDTA, using 24h incubations with 3.7kBqml(cid:2)1 H14CO
3
451nM Fe, 80nM Zn, 50nM Co and no added Cd under the appropriate experimental growth condi-
(Priceetal.,1988/89).Stockcultureswereincubated tions of light and temperature for each treatment,
at 31C with a 12h:12h light–dark cycle and an followed by filtration and scintillation counting.
incident photon flux density of 80mmol photons Aliquots for analysis of particulate organic C and N
m(cid:2)2s(cid:2)1. Although B auxotrophy has not been werefilteredontopre-combusted25mmGF/Ffilters
12
evaluated for Attheya sp., various centric diatoms (Whatman, Maidstone, UK) and analyzed using a
including species from the same family, Chaeto- 4010CHNSElementalCombustionSystem(Costech,
cerotaceae, have been found to require vitamin B Valencia,CA,USA).Samplesforparticulateorganic
12
(Provasoli and Carlucci, 1974). Cultures were ver- phosphorous were processed and analyzed spectro-
ified to be axenic by staining subsamples with photometrically. Details for these methods can be
40,6-diamidino-2-phenolindole, followed by bacter- foundinFuetal.(2005,2007;andreferencestherein).
ial counts using epifluorescence microscopy Particulate samples for trace metal analysis were
directly before the final sampling. filteredontoacid-washed3-mm-pore-sizepolycarbo-
Experimental treatments included trace metal nate filters (Millipore, Billerica, MA, USA), rinsed
clean vitamin B additions of 500 and 10ngl(cid:2)1 with oxalate reagent to remove extracellular trace
12
(370 and 7pM) for the B -replete and B -limited metals (Tovar-Sanchez et al., 2003), and Fe, Zn, Co
12 12
cultures, respectively. The 7pM vitamin B was and Cd were determined with a magnetic sector-
12
used for B -limited cultures because steady-state field high-resolution inductively coupled plasma
12
growthratesatthatconcentrationwereabouthalfof mass spectrometer (ICPMS) (Element 2, Thermo,
maximum growth rate. Vitamin and nutrient stocks Waltham, MA, USA). Procedural filter blanks were
were cleaned of trace metals before use by passing also subjected to the same storage, digestion,
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1390
Table1 Meanandones.d.(inparenthesis)oftheseawatercarbonatebuffersystemmeasurementsfromallsixexperimentaltreatments
(n¼3)
Treatment MeasuredpH MeasuredDIC Calculated
concentration(mM) pCO2(p.p.m.)
B -limited,208p.p.m. 8.36(0.03) 1975(27) 208(15)
12
B -limited,380p.p.m. 8.13(0.02) 2070(20) 380(18)
12
B -limited,680p.p.m. 7.91(0.02) 2208(13) 680(32)
12
B -replete,195p.p.m. 8.38(0.02) 1946(16) 195(10)
12
B -replete,353p.p.m. 8.15(0.02) 2020(8) 353(16)
12
B -replete,661p.p.m. 7.92(0.01) 2197(12) 661(31)
12
Abbreviations:DIC,dissolvedinorganiccarbon;pCO,carbondioxidepartialpressure.
2
SeawatermediumpCO levelswerecalculatedfrommeasurementsofpHandDICfromthefinalsamplingday,butwereconsistentatornearthese
2
valuesthroughoutmostoftheexperimentafterabriefinitialequilibrationperiod(datanotshown).
dilution and analysis processes, and these blank
B -replete
values were subtracted from sample measurements 12
(San˜udo-Wilhelmy et al., 2001). The digestion 0.4 Sp. growth rate B12-limited
protocol consisted of sequential additions of ultra-
pure HCl, HNO3 and HF (Omnitrace Ultra; VWR, 1)
Westchester, PA, USA) and heating to 1001C -d 0.3
(Eggimann and Betzer, 1976). ate (
The measured concentrations of particulate ele- h r
ments (C, N, Fe, Co, Zn and Cd) were normalized wt 0.2
o
to P. Previous studies have demonstrated that gr
c
phytoplankton cellular P quotas are usually not cifi
affected by changing pCO2 (see review by Hutchins spe 0.1
et al., 2009). Cellular elemental ‘quota’ or ‘content’
are reported in this article interchangeably with
‘requirement’, although we acknowledge that cellu- 0.0
lar quotas could possibly be elevated by luxury 200 370 670
uptake and/or storage of trace metals (for instance,
Fe luxury uptake/storage; Sunda and Huntsman, -1h) 3.5 Primary productivity
1995a; Marchetti et al., 2009). Trace metal:C ratios -1hl 3.0
wereusedinconjunction withspecificgrowthrates c
g
to calculate net use efficiencies (NUEs; specific m 2.5
growth rate divided by trace metal:C ratio), repre- C
senting how efficiently trace metals are used by mg 2.0
cells for growth and C fixation (Raven, 1991). This y (
treatment implicitly circumvents concerns about ctivit 1.5
distinguishing trace metal ‘content’ from ‘require- du 1.0
o
ments’, as it quantitatively links the trace metal pr
quota of the cells to growth and photosynthetic ary 0.5
rates. Significant differences in growth parameters, m
nutrient/trace metal quotas and trace metal NUE pri 0.0
200 370 670
were tested using two-way analysis of variance and
ppm pCO
Student’s t-test (SigmaStat, Ashburn, VA, USA). 2
Figure1 GrowthparametersforAttheyagrownat200,370and
670p.p.m. pCO in B -replete (open bars) and B -limited
2 12 12
cultures (filled bars). (a) Specific growth rates (per day) and (b)
Results
14C-basedprimaryproductivity(mgCpermgchlperh).Errorbars
representones.d.(n¼3).
Effect of B and pCO variability on specific growth
12 2
rates and primary production
Specific growth rates (Figure 1a) and 14C-based were positively correlated with pCO (r2¼0.99
2
primary productivity (Figure 1b) of B -limited and 0.97, respectively), and were B30–40% higher
12
cells (grown at 7pM B ) were significantly lower at 670p.p.m. compared with 200p.p.m. pCO
12 2
(B40–60% lower) than B -replete cells (grown at (P40.05) (Figure 1). In B -limited treatments,
12 12
370pM B ) at all three pCO concentrations (200, specificgrowthratesandprimaryproductivitywere
12 2
370 and 670p.p.m. pCO ; Po0.05). Specific growth not significantly different among CO treatments
2 2
rates and primary productivity of B -replete cells (P40.05) (Figure 1).
12
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1391
Effect of B and pCO variability on elemental ratios cells at 370 and 670p.p.m. pCO , respectively
12 2 2
Cellular molar C:P (Figure 2a) and N:P (Figure 2b) (Po0.05; Figure 4a).
ratiosofB -repletecellsweresignificantlyhigherat Co:P ratios (mmol:mol) in B -replete treatments
12 12
670p.p.m. in comparison with the other two lower were negatively correlated with pCO (r2¼0.99)
2
pCO treatments(Po0.05).BothC:PandN:Pofcells (Figure 3b). In B -limited treatments, Co:P ratios
2 12
from 670p.p.m. pCO treatments were also signifi- were an order of magnitude lower ranging from
2
cantlylowerinB -limitedcellsincomparisonwith 0.001 to 0.005mmol:mol and were significantly
12
B -repletecells(Po0.05).Therewerenosignificant lower in comparison with B -replete treatments
12 12
differences in C:P or N:P ratios between the three at all CO treatments (Po0.05) (Figure 3b). Co:P
2
pCO treatments in B -limited cultures (Figure 2; in B -limited treatments were positively correlated
2 12 12
P40.05). with pCO (r2¼0.73), and Co:P of cells grown
2
Fe quotas (mmol Fe:mol P) of B -replete cells at 670p.p.m. pCO were significantly higher than
12 2
were negatively correlated with pCO (r2¼0.92), cells grown at200p.p.m. pCO (Po0.05; Figure 3b).
2 2
andFe:Pratiosat370p.p.m.(24±11)and670p.p.m. CoNUE in both B -replete and B -limited treat-
12 12
(14±1) were B40% and B70% lower, respectively, ments were comparable at 670p.p.m. pCO , but
2
in comparison with Fe:P at 200p.p.m. pCO CoNUE of B -limited treatments were B5-fold and
2 12
(41±13; Po0.05; Figure 3a). For B -limited cells, B15-fold greater in comparison with B -replete
12 12
Fe:PwaspositivelycorrelatedwithpCO (r2¼0.93), treatments at 370 and 200p.p.m. pCO (Po0.05),
2 2
andFe:Pofcellsgrownat200p.p.m.wasB50%and respectively (Figure 4b).
B60%lowerthanthatofcellsat370and670p.p.m. Zn:P and Cd:P ratios (mmol:mol) of B -replete
12
pCO , respectively (Po0.05). FeNUE (mol C fixed and B -limited treatments displayed a similar
2 12
per mol Fe per day) of B -replete cells were negativeandpositivecorrelationwithpCO ,respec-
12 2
comparable with FeNUE of B -limited cells at tively, as observed for Fe:P (Figures 3c and d).
12
200p.p.m. pCO , but FeNUE of B -replete cells ZnNUE and CdNUE were approximately threefold
2 12
were B4-fold andB15-foldgreaterthanB -limited higher in 670p.p.m. pCO /B -replete cells when
12 2 12
compared with 670p.p.m. pCO /B -limited cells
2 12
(Po0.05), but were not significantly different
300 between B treatments at 200 and 370p.p.m. pCO
C:P 12 2
B -replete (P40.05) (Figures 4c and d).
12
250
B -limited
12
ol) 200 Discussion
m
ol: 150 Growth rates, primary productivity and major
m
P ( elemental ratios
C: 100 Phytoplankton B12 limitation has been previously
observed in natural phytoplankton assemblages in
50 the US East Coast waters for the 45mm size class
as well as for specific taxa, such as diatoms
and dinoflagellates (San˜udo-Wilhelmy et al., 2006;
0
200 370 670 Gobler et al., 2007). B limitation or B /Fe colimi-
12 12
tation has also been documented in phytoplankton
50 communities in HNLC areas of the Southern
N:P
Ocean and the Gulf of Alaska (Panzeca et al., 2006;
Bertrand et al., 2007; Koch et al., submitted).
40
Although Attheya sp. (CCMP207) has not been
mol) 30 pourervrioesuuslltys t(eFsitgeudrefo1ra)decpleeanrdlyendceemoonnsvtriatatemginrowB1t2h,
ol: rate limitation at B concentrations similar to those
m 12
P ( 20 measured in the coastal and open ocean (0.2–7pM;
N: Panzeca et al., 2009) and in the range of previously
reported B half-saturation constants reported for
12
10 other diatoms (B0.1–0.8pM; reviewed by Droop,
2007). Specific growth rates and primary productiv-
0 ity were significantly lower under B12 limitation at
200 370 670 all pCO concentrations (Figure 1). Similar reduc-
2
ppm pCO tions in growth rate and primary production by B
2 12
limitation have previously been reported for other
Figure2 ParticulateorganicCandNnormalizedtoPforAttheya
diatoms in culture (Carlucci and Silbernagel, 1969;
grownat200,370and670p.p.m.pCO inB -replete(openbars)
2 12 Swift and Taylor, 1974). In our semi-continuous
andB -limitedcultures(filledbars).(a)C:P(mol:mol)and(b)N:P
12
(mol:mol).Errorbarsrepresentones.d.(n¼3). culture design, regular dilutions were used to
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1392
60 B12-replete 0.10
Fe:P B12-limited Co:P
50
0.08
mol) 40 mol)
ol: ol: 0.06
m 30 m
m m
P ( 20 P ( 0.04
e: o:
F C
10 0.02
0 0.00
200 370 670 200 370 670
5 0.05
Zn:P Cd:P
4 0.04
ol) ol)
m m
ol: 3 ol: 0.03
m m
m m
P ( 2 P ( 0.02
n: d:
Z C
1 0.01
0 0.00
200 370 670 200 370 670
ppm pCO2 ppm pCO2
Figure3 Fe,Co,ZnandCdquotasforAttheyagrownat200,370and670p.p.m.pCO inB -replete(openbars)andB -limitedcultures
2 12 12
(filledbars).(a)Fe:P(mmol:mol),(b)Co:P(mmol:mol),(c)Zn:P(mmol:mol)and(d)Cd:P(mmol:mol).Errorbarsrepresentones.d.(n¼3).
8000 2e+7
-1-1Fe d) 6000 FeNUE BB1122--rliempilteetde -1 -1Cod) 2e+7 CoNUE
ol ol
m m
C 4000 C 1e+7
ol ol
m m
E ( 2000 E ( 5e+6
U U
N N
e o
F 0 C 0
200 370 670 200 370 670
1e+5 2e+7
-1-1Zn d) 8e+4 ZnNUE -1-1Cd d) 2e+7 CdNUE
ol 6e+4 ol
m m
C C 1e+7
ol 4e+4 ol
m m
UE ( 2e+4 UE ( 5e+6
N N
n d
Z 0 C 0
200 370 670 200 370 670
ppm pCO2 ppm pCO2
Figure4 TracemetalNUEs(molCfixedpermolmetalperday)forAttheyagrownat200,370and670p.p.m.pCO inB -replete(open
2 12
bars)andB -limitedcultures(filledbars).(a)FeNUE,(b)CoNUE,(c)ZnNUEand(d)CdNUE.Errorbarsrepresentones.d.(n¼3).
12
prevent the cultures from entering stationary phase, Furthermore, consistent with the influence of B
12
thus prohibiting any evaluation of biomass yield concentrations on growth and carbon assimilation
limitation.Ourexperimentsthereforeexaminedrate rates, C:P and N:P ratios of B -limited cells were
12
limitationonly,andfutureworkusingbatchculture significantly lower in comparison with B -replete
12
experiments is needed for direct comparisons with replicates at 670p.p.m. pCO (Figure 2). Potential
2
B limitation of biomass yield (for example, Droop, changesinphytoplanktonproductivity,C:PandN:P
12
1957). at elevated pCO could have large implications for
2
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1393
global nutrient cycles and the ocean’s biological Fe –complexes, such as the cytochrome and ferre-
pump (Riebesell et al., 2007; Hutchins et al., 2009), doxin. Although future studies will be needed to
and these empirical results suggest that vitamin B test the validity of this hypothesis, the ‘photosyn-
12
availabilitycouldalsohaveanimportantinteractive thetic architecture’ of diatoms has been shown to
role in modulating such stoichiometric shifts in the depend on Fe availability (Strzepek and Harrison,
future high-CO ocean. 2004).Thedominantelectrontransfermechanismin
2
photosystem I could also vary with ambient B
12
supply.
Fe requirements and NUEs There are likely other indirect biochemical rela-
CellsgrownunderB -replete/high-pCO conditions tionships between B /CO availability and Fe
12 2 12 2
had lower Fe quotas than B -replete/low-pCO requirements,asB ,CO andFeareeachintimately
12 2 12 2
treatments (Figure 3a). The negative correlation involved in multiple fundamental biochemical
between Fe requirements and pCO could be a pathways, such as photosynthesis (see above),
2
reflection of the lower external availability of Fe at nutrient acquisition (for example, involvement of
high pCO —reduced uptake rates of Fe bound to Fe in nitrate assimilation; Milligan and Harrison,
2
modelcarboxylicacidandhydroxamate-typeorgan- 2000) and amino-acid synthesis (for example,
icligandswererecentlyreportedindiatom cultures B -dependent methionine synthase, metH; Croft
12
at high pCO /low pH (Shi et al., 2010). In our et al., 2005). Because B is required for key
2 12
experiments, Fe was buffered by EDTA (via car- enzymes, such as metH and methylmalonyl CoA
boxylic acid moieties), and the lower observed Fe:P mutase (Rodionovetal.,2003; Croft etal., 2005), Fe
inB -repleteculturesat670p.p.m.pCO (Figure3a) requirements may be affected downstream of where
12 2
might be a result of lower Fe availability. However, these enzymes are located in metabolic pathways.
Fe:PofB -limited/670p.p.m.pCO -growncellswas The role of B in methionine synthesis could affect
12 2 12
the highest of all CO treatments (Figure 3a), cell-wide processes because of the need of methio-
2
suggesting that another mechanism, in this case nineforbiosynthesisofproteinsandvariouscritical
perhaps the influenceof CO on cellular Fe require- transmethylation reactions via S-adenosyl methi-
2
ments, might have a larger effect than the reduction none (Fontecave et al., 2004).
in Fe bioavailability.
A reduced Fe requirement for photosynthesis
and respiration-generated energy to power carbonic Co requirements and NUEs
anhydrases (CAs) or other carbon-concentrating Under B -replete conditions, Co:P ratios were
12
mechanism-related cellular machinery could be B5- to 14-fold higher in comparison with B -
12
reason for lower Fe:P (Figure 3a) and higher FeNUE limitedcells(Figure3b),suggestingthatCo-contain-
(Figure 4a) of B -replete/670 p.p.m. pCO cells. ing vitamin B could comprise a significant
12 2 12
Theoretical calculations predict that cells saturated fraction of the total cellular Co quota, and/or that
with CO require less ATP and thus approximately B limitationcouldresultinlowerCorequirements.
2 12
one-third less Fe in the photosynthetic and respira- In B -limited experiments, we calculated the
12
tory electron transport chains (Raven, 1991). It is percent of cellular Co that likely originates from
notable that in B -replete experiments, there is a assimilated B , assuming that the B added to the
12 12 12
counterintuitive relationship between Fe:P and media was completely bioavailable. If B was fully
12
physiological measurements—growth rate and pri- utilized,CofromB (oneCoatomperB molecule)
12 12
mary productivity increase with pCO , whereas accounted for B78 to 4100% of measured mean
2
Fe requirements simultaneously decrease. Further cellularCo(Table2).WhenB waslimiting,CoNUE
12
studiesassessingtherelativeimportanceofreduced at 200 and 370p.p.m. pCO were relatively high
2
insituFebioavailability(Shietal.,2010)andlower because of the substantially lower Co quotas
cellular Fe requirements (this study) is warranted. (Figures3band4b).Weareunabletomakeasimilar
The synergistic relationship between vitamin B
12
availability andintracellular Fequotasisnottotally
unexpected. Both Fe (Strzepek and Harrison, 2004) Table 2 Mean cellular Co and s.d. (in parentheses), percent of
and phylloquinone (vitamin K ; a macromolecule mean cellular Co from Co–B and range (in parentheses), and
1 12
that requires B -dependent S-adenosyl methinone mean dissolved Co utilization (calculated by difference) for
12
for its synthesis; Lohmann et al., 2006) serve as B12-limitedcells,assumingB12iscompletelybioavailableandutilized
electroncarriersinphotosystemI.Ourempiricallab
pCO (p.p.m.) CellularCo %From %From
results (Figure 3a) suggest that Fe requirements of 2 ((cid:3)10(cid:2)12M) Co-B12 dissolvedCo
Attheyaarelower(andFeNUEhigher)whenB was
12
repleteat370and670p.p.m.pCO2,perhapsbecause 200 3.8(1.5) 4100 —
photosynthetic electron transfer is more dependent 370 8.8(0.8) 84(73–95) 16
on phylloquinone. In contrast, under B -limited 670 9.4(4.4) 78(54–148) 22
12
conditions, the Fe requirement of the organism
Abbreviations:Co,cobalt;pCO,carbondioxidepartialpressure.
was higher (and FeNUE lower), suggesting that 2
[Co]containedinB12addedtomediawas7.4(cid:3)10(cid:2)12M;dissolved[Co]
electron transfer could be more dependent on addedtomediawas5(cid:3)10(cid:2)8M.
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1394
ComassbalancecalculationfortheB -repletecells,as Inaddition,Attheyabelongstoafunctionalgroup
12
ourB -repletemediahad370pMCo–B and50nMCo, of chain-forming centric diatoms that have been
12 12
which were both in excess of the measured cellular observed to proliferate in Fe addition bottle experi-
Co concentrations (62–133pM; Table 2). In addition to mentsandmesoscaleFefertilizationexperimentsin
Co in vitamin B , Co has also been identified as a the HNLC regimes of the subarctic North Pacific,
12
metabolic substitute in diatoms for Zn in CA (Morel Southern Ocean and coastal California upwelling
et al., 1994; Sunda and Huntsman,1995b). (Martin et al., 1989; Hutchins and Bruland, 1998;
deBaaretal.,2005),andarebelievedtoformresting
spores that contribute to C export in the coastal
Zn and Cd requirements and use efficiencies
subarctic and northeast Pacific (Grimm et al., 1996;
When Attheya was grown under B -replete/high
12 Saino et al., 1998). Based on our experimental
pCO , Zn:P and Cd:P ratios were low and ZnNUE
2 results, without an adequate supply of B , and
and CdNUE were high (Figures 3c, d, 4c and d). 12
despite a high availability of Fe, Chaetoceros-like
Lower Zn:P is consistent with higher CO avail-
2 diatoms would have a similar Fe requirement but
ability and a decreased Zn requirement for
use Fe much less efficiently to achieve only about
Zn-containing CA used for catalyzing the conver-
half the growth and carbon fixation rates.
sion of HCO(cid:2) to CO (Morel et al., 1994; Lane and
3 2 Thepresentstudyhighlightsthelinkagesbetween
Morel, 2000a). Previous laboratory culture experi-
organicgrowthfactorlimitation,tracemetalrequire-
ments with the diatoms Thalassiosira weissflogii
ments and NUE, and phytoplankton productivity
and Thalassiosira pseudonana have also found
under varying availabilities of pCO . For example,
higher Zn quotas of cells grown at lower pCO in 2
2 our results under conditions of the last glacial
comparison with cells grown at present-day pCO
2 maximum suggest that increases in eolian Fe
(Lane and Morel, 2000a; Sunda and Huntsman,
deposition alone to surface waters of HNLC areas
2005).Lower Cd:PratiosathighpCO inB -replete
2 12 during that period may not have been sufficient to
treatmentscouldbetheresultofthedownregulation
reduce atmospheric CO via the biological pump,
of a Cd-specific CA (Lane and Morel, 2000b; Lane 2
contrarytoMartin’s‘ironhypothesis’(Martin,1990).
et al., 2005; Shi et al., 2010).
Our results showed that high primary productivity
In addition to CO , vitamin B availability might
2 12 was only attained when B was available. The
also affect Zn-based CAs. Methionine, among 12
relevance of Fe availability to the biological draw-
two other amino acids, was found to be a critical
downofatmosphericCO duringglacialperiodshas
component of a hydrophobic cluster on the 2
also recently been questioned by Boyd et al. (2010).
C-terminal a-helix that predicated formation, and
Further work toward elucidating these types of
possibly function, of the b-type Zn-containing CA
interactions is needed to understand the bio-
(PtCA1) of the diatom Phaeodactylum tricornutum
geochemical consequences of natural and anthro-
(Kitao and Matsuda, 2009).
pogenic Fe fertilization in HNLC regimes, and the
potential impacts of changing pCO on global C
2
Ecological implications cycling in the present-day and future ocean.
Extrapolation from culture-based studies to natural
ecosystemsrequirescaution,butthese resultscould
have implications for altered (co)limitation patterns
Acknowledgements
of phytoplankton in a changing ocean. At present-
day pCO2 (370p.p.m.) and estimated pCO2 before This study was supported by the US National Science
year 2100 (670p.p.m.), B limitation resulted in FoundationgrantsOCE0962209toSAS,OCE0850730to
12
higher Fe quotas and lower NUE of Fe, Zn and Cd. FXF,andOCE 0722337and0825319to DAH.
Although diatoms in field experiments in the
subarctic Pacific and Southern Oceans have been
showntobelimitedbyFeand/orB (Bertrandetal.,
12 References
2007; Koch et al., submitted), our experimental
results imply that the consequences of low B12 BehrenfeldMJ,FalkowskiPG.(1997).Photosyntheticrates
availability at current and future pCO2 are a higher derived from satellite-based chlorophyll concentra-
Ferequirement,lowergrowth/productivityandthus tion.LimnolOceanogr42:1–20.
lowerFeNUE.B limitationofdiatomslikeAttheya Betrand EM, Saito MA, Rose JM, Riesselman CR, Lohan
12
could drive the subarctic Pacific and Southern MC, Noble AE et al. (2007). Vitamin B12 and iron
Oceans(thatarealreadyrelativelylowinFe)further colimitationofphytoplanktongrowthintheRossSea.
LimnolOceanogr52:1079–1093.
toward B /Fe colimiting HNLC conditions. In
12 Boyd PW, Mackie DS, Hunter KA. (2010). Aerosol iron
contrast, if B availability increases in the future
12 deposition to the surface ocean-modes of iron
ocean, a lower Fe requirement (and higher Fe/Zn/
supply and biological responses. Mar Chem 120:
CdNUE) by diatoms might result in community
128–143.
shifts toward this group, a decoupling of their Carlucci AF, Silbernagel SB. (1969). Effect of vitamin
growth from Fe availability and strengthening of concentrations ongrowthanddevelopmentof vitamin-
the biological C pump in these regimes. requiring algae.JPhycol5:64–67.
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1395
CrawfordRM,HinzF,KoschinskiP.(2000).Thecombina- Leblanc K, Hare CE, Boyd PW, Bruland KW, Sohst B,
tion of Chaetoceros gaussii (Bacillariophyta) with Pickmere S et al. (2005). Fe and Zn effects on the Si
Attheya. Phycologia39:238–244. cycle and diatom community structure in two con-
Croft MT, Lawrence AD, Raux-Deery E, Warren MJ, trasting high and low-silicate HNLC areas. Deep-Sea
Smith AG. (2005). Algae acquire vitamin B through a ResI52:1842–1864.
12
symbioticrelationshipwithbacteria.Nature438:90–93. LohmannA,Scho¨ttlerMA,Bre´he´linC,KesslerF,BockR,
de Baar HJW, Boyd PW, Coale KH, Landry MR, Tsuda A, Cahoon EB et al. (2006). Deficiency in phylloquinone
Assmy P et al. (2005). Synthesis of iron fertilization (vitamin K ) methylation affects prenyl quinone
1
experiments:fromtheironageintheageofenlighten- distribution, photosystem I abundance, and antho-
ment.J GeophysRes—Oceans 110:C09S16. cyanin accumulation in the Arabidopsis AtmenG
Dickson AG, Goyet C. (1994). Handbook of Methods for mutant. JBiol Chem 281:40461–40472.
the Analysis of the Various Parameters of the Carbon Marchetti A, Parker MS, Moccia LP, Lin EO, Arrieta AL,
Dioxide System in Sea Water (Ver. 2). Department of RibaletFetal.(2009).Ferritinisusedforironstorage
Energy,ORNL/CDIAC-74: Oak Ridge, Tenn. in bloom-forming marine pinnate diatoms. Nature
Droop MR. (1957). Vitamin B in marine ecology. Nature 457:467–470.
12
180:1041–1042. Martin JH. (1990). Glacial-interglacial CO change: the
2
DroopMR. (2007). Vitamins, phytoplankton andbacteria: iron hypothesis.Paleoceanography 5: 1–13.
symbiosisorscavenging?JPlanktonRes29:107–113. Martin JH, Gordon RM, Fitzwater S, Broenkow WW.
Eggimann DW, Betzer PR. (1976). Decomposition and (1989).Vertex:phytoplanktonironstudiesintheGulf
analysis of refractory oceanic suspended materials. of Alaska.Deep SeaResPart A36:649–680.
Anal Chem48:886–890. Milligan AJ, Harrison PJ. (2000). Effects of non-steady-
Fontecave M, Atta M, Mulliez E. (2004). S-adenosyl- stateironlimitationonnitrogenassimilationenzymes
methionine: nothing goes to waste. Trends Biochem in the marine diatom Thalassiosira weissflogii
Sci29:243–249. (Bacillariophyceae). J ofPhycol 36:78–86.
FuFX,WarnerME,ZhangY,FengY,HutchinsDA.(2007). Morel FMM, Reinfelder JR, Roberts SB, Chamberlain CP,
Effects of increased temperature and CO on photo- LeeJG,YeeD.(1994).Zincandcarbonco-limitationof
2
synthesis, growth and elemental ratios of marine marine-phytoplankton. Nature 369: 740–742.
Synechococcus and Prochlorococcus (Cyanobacteria). Panzeca C, Beck AJ, Tovar-Sanchez A, Segovia-Zavala J,
J Phycol43: 485–496. Taylor GT, Gobler CJ et al. (2009). Distributions of
Fu FX, Zhang Y, Leblanc K, San˜udo-Wilhelmy SA, dissolved vitamin B and Co in coastal and open-
12
Hutchins DA. (2005). The biological and biogeo- ocean environments. Estuar Coast Shelf Sci 85:
chemicalconsequencesofphosphatescavengingonto 223–230.
phytoplankton cell surfaces. Limnol Oceanogr 50: Panzeca C, Tovar-Sanchez A, Agusti S, Reche I, Duarte
1459–1472. CM, Taylor GT. (2006). B vitamin as regulators
Gobler CJ, Norman C, Panzeca C, Taylor GT, San˜udo- of phytoplankton dynamics. EOS Trans AGU 87:
Wilhelmy SA. (2007). Effect of B-vitamins (B , B ) 593–596.
1 12
and inorganic nutrients on algal bloom dynamics Petit JR, Jouzel J, Raynaud D, Barkov NI, Barnola JM,
in a coastal ecosystem. Aquat Microb Ecol 49: BasileIetal.(1999).Climateandatmospherichistory
181–194. of the past 420000 years from the Vostok ice core,
Grimm KA, Lange CB, Gill AS. (1996). Biological forcing Antarctica.Nature 399:429–436.
of hemipelagic sedimentary laminae: evidence from PriceNM,HarrisonGI,HeringJG,HudsonRJM,NirelPMV,
ODP Site 893, Santa Barbara Basin, California. J Sed PalenikBetal.(1988/89).Preparationandchemistryof
Res66:613–624. theartificialalgalculturemediumAquil.BiolOceanogr
Hutchins DA, Bruland KW. (1998). Iron-limited diatom 6:443–461.
growth and Si:N uptake ratios in a coastal upwelling Provasoli L, Carlucci AF. (1974). Vitamins and growth
regime.Nature 393: 561–564. regulators. In: Stewart WDP, Abbott MR (eds). Algal
Hutchins DA, Mulholland MR, Fu FX. (2009). Nutrient Physiology and Biochemistry. Blackwell Science:
cycles and marine microbes in a CO -enriched ocean. Malden,Mass, pp741–787.
2
Oceanography22:128–145. Raven JA. (1991). Physiology of inorganic C acquisition
KitaoY,MatsudaY.(2009).Formationofmacromolecular andimplicationsforresourceuseefficiencybymarine
complexes of carbonic anhydrases in the chloroplast phytoplankton: relation to increased CO and tem-
2
of a marine diatom by the action of the C-terminal perature.Plant CellEnviron 14:779–794.
helix.Biochem J 419: 681–688. RiebesellU,SchulzKG,BellerbyRG,BotrosM,FritscheP,
Koch F, Marcoval A, Panzeca C, Bruland KW, San˜udo- Meyerho¨fer M et al. (2007). Enhanced biological
Wilhelmy SA, Gobler CJ. (submitted). The effect of carbon consumption in a high CO ocean. Nature
2
vitamin B , nitrogen and iron on phytoplankton 450:545–548.
12
growthandcommunitystructureintheGulfofAlaska. Rodionov DA, Vitreschak AG, Mironov AA, Gelfand MS.
Lane TW, Morel FMM. (2000a). Regulation of carbonic (2003). Comparative genomics of the vitamin B
12
anhydrase expression by zinc, cobalt, and carbon metabolismandregulationinprokaryotes.JBiolChem
dioxide in the marine diatom Thalassiosira weiss- 278:41148–41159.
flogii. PlantPhysiol123: 345–352. SainoT,ShangS,MinoY,SuzukiK,NomuraH,SaitohS
Lane TW, Morel FMM. (2000b). A biological function for et al. (1998). Short term variability of particle fluxes
cadmiuminmarinediatoms.ProcNatlAcadSciUSA and its relation to variability in sea surface tempera-
97:4627–4631. ture and chlorophyll a field detected by ocean color
Lane TW, Saito MA, George GN, Pickering IJ, Prince RC, and temperature scanner (OCTS) off Sanriku, north-
Morel FMM. (2005). A cadmium enzyme from a western north Pacific in the spring of 1997.
marine diatom.Nature435:42. JOceanogr 54:583–592.
TheISMEJournal
DiatomCO ,B andtracemetalinteractions
2 12
ALKingetal
1396
San˜udo-Wilhelmy SA, Gobler CJ, Okbamichael M, Taylor and geochemical implications. Limnol Oceanogr 40:
GT.(2006).Regulationofphytoplankton dynamics by 1404–1417.
vitaminB . GeophysResLett33:L04604. Sunda WG, Huntsman SA. (2005). Effect of CO
12 2
San˜udo-Wilhelmy SA, Kustka AB, Gobler CJ, Hutchins supply and demand on zinc uptake and growth
DA, Yang M, Lwiza K et al. (2001). Phosphorus limitation in a coastal diatom. Limnol Oceanogr 50:
limitation of nitrogen fixation by Trichodesmium in 1181–1192.
the centralAtlanticOcean. Nature411:66–69. Swift DG, Taylor WR. (1974). Growth of vitamin
Shi DL,XuY,HopkinsonBM, Morel FMM.(2010). Effect B -limitedculturesThalassiosirapseudonana,Mono-
12
of ocean acidification on iron availability to marine chrysis lutheri, and Isochrysis galbana. J Phycol 10:
phytoplankton.Science 327:676–679. 385–391.
Solomon S, Qin D, Manning M, Chen Z, Marquis M, Tang YZ, Koch F, Gobler CJ. (2010). Most harmful algal
Averyt KB et al. (2007). Climate Change 2007: The bloomspeciesarevitaminB andB auxotrophs.Proc
1 12
Physical Science Basis. Contribution of Working NatlAcad SciUSA 107:20756–20761.
Group I to the Fourth Assessment Report of the Tovar-SanchezA,San˜udo-WilhelmySA,Garcia-VargasM,
Intergovernmental Panel on Climate Change. Weaver RS, Popels LC, Hutchins DA. (2003).
Cambridge University Press: Cambridge, United A trace metal clean reagent to remove surface-
Kingdom andNewYork, NY. bound iron from marine phytoplankton. Mar Chem
StrzepekRF,HarrisonPJ.(2004).Photosyntheticarchitec- 82: 91–99.
ture differs in coastal and oceanic diatoms. Nature
431:689–692.
This work is licensed under the Creative
Sunda WG, Huntsman SA. (1995a). Iron uptake and
CommonsAttribution-NonCommercial-No
growth limitation in oceanic and coastal phytoplank-
ton.MarChem 50:189–206. Derivative Works 3.0 Unported License. To view a
Sunda WG, Huntsman SA. (1995b). Cobalt and zinc copy of this license, visit http://creativecommons.
interreplacementinmarinephytoplankton:Biological org/licenses/by-nc-nd/3.0/
TheISMEJournal