Inorganic Carbon Limiitation and Chemical Composition of Two

Vol. 41, No. 1APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1981, p. 60-700099-2240/81/010060-11$02.00/0

Inorganic Carbon Limiitation and Chemical Composition ofTwo Freshwater Green MicroalgaetJOEL C. GOLDMAN* AND STEPHEN J. GRAHAM*

Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543

Two freshwater chlorophytes, Chlorella vulgaris and Scenedesmus obliquus,were grown in inorganic carbon-limited continuous cultures in which HCO3 wasthe sole source of inorganic carbon. The response of the steady-state growth rateto the extemal total inorganic carbon concentration was reasonably well describedby the Monod equation; however, the response to the intemal nutrient concen-tration was only moderately well represented by the Droop equation when theintemal carbon concentration was defined on a cellular basis. The Droop equationwas totally inapplicable when total biomass (dry weight) was used to defineintemal carbon because the ratio of carbon to dry weight did not vary over theentire growth rate spectrum. In batch cultures, maximum growth rates wereachieved at the C02 levels present in atmospheric air and at HC03- concentrationsof 3 mM. No growth was observed at 100% C02. Both nitrogen uptake andchlorophyll synthesis were tightly coupled to carbon assimilation, as indicated bythe constant C/N and C/chlorophyll ratios found at all growth rates. The maininfluence of inorganic carbon limitation appears to be not on the chemicalstructure of the biomass, but rather on cell size; higher steady-state growth rateslead to bigger cells.

Variations in the chemical composition ofphytoplankton are tightly coupled to changes ingrowth rate (18, 42; J. C. Goldman, in P. G.Falkowski, ed., Primary Productivity in the Sea,in press). To a large degree, this growth ratedependence provides a good description of thenutritional state of a cell population in responseto different degrees of nutrient limitation (39;Goldman, in press). For example, significant var-iations in the cell quota (Q) (cellular concentra-tion of a limiting nutrient) for either phosphorusor nitrogen occur when the respective nutrientis limiting in continuous culture and the dilutionrate (=growth rate) is varied (8, 12, 20, 39).Droop (7) has demonstrated that Q is related togrowth rate by a rectangular hyperbolic equa-tion of the following form: ,u = (1 - kQQ-1)(equation 1), where ,tis the specific growth rate,Ai is the specific growth rate for which Q isinfinite, and kQ is the miimum concentration oflimiting nutrient required before growth can pro-ceed. This equation is empirical, and its utilityis related to which limiting nutrient is beingconsidered (17). For nutrients that constitute asmall fraction of total cellular material, such asP043- and vitamin B12, the ratio of kQ to QM (QMis the upper boundary of Q, which is associated

t Contribution 4658 from the Woods Hole OceanographicInstitution.

t Present address: Division of Applied Sciences, HarvardUniversity, Cambridge, MA 02138.

with the true maxim growth rate ['i ]) is verysmall (e.g., <0.1), indicating a large variation inQ for 0 < ,u <, (17). In such cases, according toequation 1, ,u Ai. When nitrogen, which consti-tutes -5 to 10% of the total cellular biomass, isthe limiting nutrient, the variability in Q is morerestricted, and the ratio of kQ to QM is -0.2, sothat jp= 0.8 Ai (17, 20). Under these conditions,the applicability of equation 1 is restricted, andthere is no substitute for determining a and QMexperimentally.To date, virtually no information is available

concerning the degree of cellular carbon varia-tion in phytoplankton when inorganic carbon isthe limiting nutrient. However, Goldman et al.(19) and Pipes (36) did observe that the yieldcoefficient on a weight basis (cellular dry weightper unit of cellular organic carbon) for severalfreshwater green algae was invariant over theentire growth rate spectrum in inorganic carbon-limited continuous cultures. However, cell num-bers were not measured, and thus Q values forcarbon were unavailable.Moreover, Goldman et al. (19) demonstrated

that at a steady state under inorganic carbonlimitation ,u was related to the residual totalinorganic carbon concentration (CT1 ) (sum ofconcentrations of C02, H2CO3, HCO3 , andCo32-) according to the Monod equation: ,u =IPCT, (K. + CTR)` (equation 2), where K. is thehalf-saturation coefficient (the concentration of

60

INORGANIC CARBON LIMITATION IN MICROALGAE

limiting nutrient for which ,t = ,i /2). In addition,K. was found to be a function of culture pH. Formicroalgae it has been difficult to demonstraterelationships between ,u and residual (extemal)limiting nutrients because K. values for the com-mon nutrients studied (e.g., nitrogen, phospho-rus) are typically below levels of detectability,even though the Droop (internal nutrient) andMonod (external nutrient) equations are com-patible at steady state (4, 6, 12). However, Brownand Button (3) were able to measure residualphosphorus levels in the nanomolar range andfound a linear relationship between ,u and resid-ual phosphorus levels for steady-state growth ofthe freshwater chlorophyte Selenastrum capri-cornutum. Moreover, they found that a thresh-old level of 10 nM phosphorus was requiredbefore growth could proceed and concluded thatthe Monod relationship did not describe phos-phorus-limited growth well for this alga. Still,the usefulness of the Monod relationship is thatit provides a reasonable description of the affin-ity of a particular organism for a particular lim-ited nutrient. Like the Droop equation, it ispurely empirical, but because of its simplicityand the usefulness of K., it has popular appeal(12).In this study, we expanded on the previous

study of Goldman et al. (19) and examined theutility of the Droop and Monod equations forinorganic carbon-limited growth of two fresh-water green microalgae, Chlorella vulgaris andScenedesmus obliquus, in continuous cultures.In addition, we studied how the chemical com-positions of these algae varied with growth rateunder inorganic carbon limitation.

MATERIALS AND METHODSThe continuous culture apparatus (a bank of eight

0.5-liter cultures), the culturing protocols, and theexperimental analyses were virtually identical to thosedescribed previously (17, 20). We used continuouslighting (2,093 J of visible light per m' per min),temperature control (20°C), and mixing with magneticbar stirring in the continuous culture experiments.Aeration with mixtures of 100% C02 and laboratoryair at several bubble rates was used only in some ofthe batch experiments to determine ,A. C02 from a gascylinder and laboratory air were first mixed in thedesired proportions in a two-gas proportioner. Thespecific gas bubbling rate (gas bubbling rate per cul-ture volume) was set by passing the gas mixturethrough a flow meter-regulator before it entered thebottom ofthe culture in these experiments. In all otherexperiments gas bubbling was not employed, andHCO3 was the sole source of inorganic carbon. Thefreshwater chlorophytes C. vulgaris and S. obliquuswere obtained from the laboratory of M. Gibbs, Bran-deis University.

The freshwater medium was similar to the mediumused previously (19) and contained 2.0mM NH4CI, 0.4mM MgCl2, 0.4 mM MgSO4.7H20, 0.2 mM CaCl2.2H20, 0.04 mM H3B04, and trace metals in a twofolddilution offmedium (23). The medium for the contin-uous culture experiments was buffered with 10 mMphosphate buffer consisting of equimolar concentra-tions of K2HP0O and KH2PO4; this resulted in a pH of7.1 to 7.2. The concentration of total inorganic carbonin the medium (CT,) was 10 mg of C per liter and wassupplied by a mixture of NaHCO3 and NaCO3. For thebatch studies the ratio of di-P043- to mono-P043- in25 mM buffer was varied, depending on the amount ofC02 in the gas mixture. Up to 10 mM HC03 wasadded in some of these batch experiments. The me-dium was dispensed into the continuous cultures viaa multichannel peristaltic pump (Harvard 1203). Alltubing was glass, except for small sections of siliconewhich were inserted through the pumps.

Chemical analyses for CT. and CT, were performedwith a Dohrmann DC-54 Ultra-Low Total CarbonAnalyzer modified for inorganic carbon analyses asdescribed by Goldman (13). This instrument has aprecision of ±10 tug of C per liter (or ±2%) and adetection limit of -50 yg of C per liter. Particulatecarbon and nitrogen were measured with a Perkin-Elmer model 240 elemental analyzer. Cells werecounted with a Spencer Bright-line hemacytometer.Dry weights were determined for 100-ml samples re-tained on precombusted glass fiber filters and com-busted at 500 to 550'C for >4 h. Chlorophyll a levelswere measured on acetone-extracted samples by fluo-rometry, using the method of Strickland and Parsons(51). Typically, samples were extracted overnight. Cul-ture and medium pH values were measured with acombination probe mounted on a Corning 110 meter.All measurements were made directly on culture sam-ples at steady state, which was defined as the timewhen culture absorbance at 600 nm, as measured witha Bausch & Lomb Spectronic 88 spectrophotometer,did not vary more than ±10% for at least 2 consecutivedays. The cultures were not axenic for the reasonscited previously (12).

#i was estimated both by the cell washout technique(12) and by the enriched culture batch technique (20).Batch experiments were performed by using eitherbubbled gas or HCO3 as the source of inorganiccarbon. The following three concentrations of HCO3were used: 3, 6, and 10 mM. Gas mixtures included air(0.036% C02) at three bubbling rates (25, 50, and 75h-') and 1, 5, and 100% C02 in air at a constantbubbling rate of 50 h-'. a was determined for eachexperiment by a linear regression analysis of the plotof the natural log of the cell count versus time. In eachexperiment several measurements were made duringexponential growth for cellular carbon (QmA{) and ni-trogen (Qw). The inocula for the batch cultures weretaken from continuous cultures at steady state to giveinitial cell numbers of 0.1 x 10' to 0.3 x 105 cells perml.The kinetic coefficients A and kQ were determined

from regression analyses of the linearized version ofequation 1, as follows: Y - YQ(1-pA-1) (equation 3),where Y is the cellular yield coefficient and YQ is the

61VOL. 41, 1981

62 GOLDMAN AND GRAHAM

maximum cellular yield coefficient (kQ-1) (7). K. and,u were determined from regression analyses of thelinearized version of equation 2, as follows: CT, =

(CTj,_1)ii - K. (equation 4) (19). The ratio of ii towas determined from linear regression analyses byusing equations 3 and 4 or by inserting the ratio of kQto QCM into the limiting expression of equation 1, asfollows: Ai = ,(l - kQ.QcM-') (equation 5) (17). Thevalue of QCM in this case was determined experimen-tally. A total of 46 steady-state measurements weremade for C. vulgaris in the growth rate range 0.17 to2.05 day-', and 33 measurements were made for S.

obliquus in the growth rate range 0.17 to 1.56 day-'.

RESULTS

,u. Estimates of A as determined by the wash-out technique were 1.59 ± 0.028 day-' (mean ±standard deviation) for S. obliquus and 2.11 ±

0.036 day-1 for C. vulgaris. For C. vulgaris thevalues of p as determined by the batch tech-nique, regardless of whether HCO3 (3 to 10mM) or bubbled C02 (0.036 to 1%) was theinorganic carbon source, were comparable to thevalues obtained by the washout technique, av-

eraging 2.02 ± 0.052 day-' (Table 1). With 5%C02 p was diminished considerably (0.97 day-'),and with 100% C02 no growth was observed. Onthe other hand, for S. obliquus, p as determinedby the batch technique (1.56 ± 0.113 day-) wascomparable to A as determined by the washoutmethod when bubbled C02 was the inorganiccarbon source at any C02 level (except 100%C02, at which no growth occurred). Moreover,increasing the concentration of HCO3 beyond3 mM (f = 1.67 day-) led to significant reduc-tions in (down to 1.16 day-' with 10 mMHCO3) (Table 2). There was no effect of bubblerate on for either species when air was theinorganic carbon source. Culture pH values var-ied between 6.8 and 7.7, with the highest valuesoccurring in the 10 mM HCO3 experiments(Tables 1 and 2).

K. values. The response of ,u to the external(residual) inorganic carbon concentration (CT,)was described well by equation 4, leading to K.values of 0.20 ± 0.027 mg of C per liter (mean± standard deviation) for C. vulgaris (r > 0.99,P < 0.001) (Fig. 1A) and 0.16 ± 0.052 mg of Cper liter for S. obliquus (r > 0.99, P < 0.001)(Fig. 1B).Cellular carbon variations. The ratio of

cellular carbon to dry weight (Q',) was invariantwith ,u for both species; this ratio was 0.46 ± 0.07(mean ± standard deviation) for C. vulgaris(Fig. 2A) and 0.48 ± 0.08 for S. obliquus (Fig.3A). In contrast, the carbon cell quota (Q,) in-creased with, for both species (Fig. 2B and 3B).The kinetic coefficients kQ and,, as derivedfrom equation 3, were 2.2 ± 0.09 pg of C per celland 2.32 ± 0.087 day-' for C. vu4garis (mean ±standard deviation), respectively (r = 0.93, P <0.001); for S. obliquus those values were 7.3 ±0.04 pg of C per cell and 2.44 ± 0.029 day-',respectively (r = 0.80, P < 0.001) (Table 3).

Estimates of QCM by the batch technique andfrom the experimental data in Fig. 2B and 3Bwere similar for both species; values for QCMwere 13.6 ± 1.98 pg of C per cell, as determinedby the batch technique (Table 1), to .15 pg ofC per cell, as estimated by eye from the data(Fig. 2B) for C. vulgaris and 29.9 ± 5.25 pg of Cper cell (Table 2), to -28 pg of C per cell (Fig.3B) for S. obliquus. The resulting ratios of kQ toQCM were 0.16 ± 0.049 (mean ± standard devia-tion) for C. vulgaris and 0.24 ± 0.045 for S.obliquus (Table 3). The ratio of to Ai, asdetermined from the summarized experimentaldata for and Ai, generally was within the 95%confidence limits of the ratio of to Ai derivedfrom equation 5 with the ratio of kQ to QCMinserted (Table 3); these values were 0.87 ±

0.048, as determined from the experimental val-ues for and ,u, versus 0.86 ± 0.049, as deter-

TABLE 1. Growth and cellular coefficients for C. vulgaris at Ai in batch and continuous culturesGas

Growth bub- Q.,M (pg Q,.,M (pg K. (mg of C Culturemode Carbon source bling Li (day-) of C per Qc. of N per pe(mg pHrate cell) cell) per liter) pH

(h-1)Continuous HC03- (1 mM) 2.11 ± 0.036a ~15b 0.46 ± 0.071c -2.7 0.20 ± 0.027a 7.1-7.2Batch HC03- (3 mM) 1.94 12.2 2.7 6.9

HC03- (6 mM) 2.07 14.2 2.8 7.5HC03- (10 mM) 2.02 14.7 3.1 7.7CO2 (0.036%)d 25 1.97 12.8 2.5 6.8

50 2.06 14.0 2.3 6.875 2.08 11.5 2.6 6.8

CO2 (1%) 50 1.97 16.0 3.3 6.8CO2 (5%) 50 0.97 6.8CO2 (100%) 50 0 6.0

a Estimated by a linear regression analysis of the data in Fig. 1A, using equation 4; mean ± standard deviation.b Estimated by eye from the data in Fig. 2B.c Estimated by determination of the mean of the data in Fig. 2A; mean ± standard deviation.d CO2 was bubbled through the cultures.

APPL. ENVIRON. MICROBIOL.


TABLE 2. Growth and cellular coefficients for S. obliquus at Ai in batch and continuous culturesGas

Growth Carbon source bub- Q,M (pg Q' (pg K. (mg of C Culturemode Cabn ore ln (a- of C per Q~'c of N1per per liter) pH

(h-1)Continuous HC03 (1 mM) 1.59 ± 0.028a 28b 0.48 ± 0.083c -4.7 0.16 ± 0.052a 7.1-7.2Batch HC03- (3 mM) 1.67 38.6 5.4 6.8

HC03- (6 mM) 1.31 22.6 3.4 7.5HCO3- (10 mM) 1.16 20.2 3.0 7.7CO2 (0.036%)d 25 1.50 25.2 4.5 6.8

50 1.53 29.5 4.4 7.075 1.47 27.6 4.8 6.8

CO2 (1%) 50 1.77 33.8 5.3 6.8CO2 (5%) 50 1.58 30.2 4.2 6.8CO2 (100%) 50 0 6.0

a Estimated by a linear regression analysis of the data in Fig. 1B, using equation 4; mean ± standard deviation.bEstimated by eye from the data in Fig. 3B.c Estimated by determination of the mean from the data in Fig. 3A; mean ± standard deviation.d C02 was bubbled through the cultures.

f T\ A B

2.0

ot'.01z...FIIII'14.05

0.0 1 2 3 4 0 1 2 3 4 5 6 7

RES/DUAL TOTAL /AVRGAN/C CARBON IN CULTURECr -mg /iter '

FIG. 1. Relationship between ,u and residual total inorganic carbon in cultures at steady state in inorganiccarbon-limited continuous cultures. Curves were determined by linear regression analyses of equation 4 andplots of equation 2. (A) C. vulgaris. (B) S. obliquus.

mined from equation 5, for C. vulgaris and 0.64± 0.048 versus 0.76 ± 0.048 for S. obliquus (Table3).Cellular nitrogen variations. There ap-

peared to be tight coupling between nitrogenassimilation and carbon assimilation at allsteady-state growth rates. The cellular ratio ofC to N (by weight) was vitually invariant withvarying Iu values for both C. vulgaris (Fig. 4A)and S. obliquus (Fig. 4B), ranging between 5 and6. The maximum nitrogen cellular content(Q,.), as averaged from the batch culture datain Tables 1 and 2, was 2.7 ± 0.40 pg ofN per cell(mean ± standard deviation) for C. vulgaris and4.7 ± 0.59 pg ofN per cell for S. obliquus.

Cellular chlorophyll variations. Cellularchlorophyll content, like carbon and nitrogencontents, increased with increasing ,u, rangingfrom -0.04 to -0.25 pg of chlorophyll per cellfor C. vulgarus (Fig. 5B) and -0.1 to 0.25 pg ofchlorophyll per cell for S. obliquus (Fig. 6B) for0 < ,u < p. The ratio of carbon to chlorophyll

decreased slightly from - 75 at zero IL to 50 at,ii for C. vulgaris (Fig. 5A), but generally wasinvariant at - 100 at all values of,u for S. obliquus(Fig. 6A), although there was substantial scatterin the data at I < 0.3 day-'.

DISCUSSIONInorganic carbon-limited growth kinet-

ics. Any interpretation of inorganic carbon ki-netic data is premised on the knowledge that theactual substrate for assimilation is known or thatthe rate reactions within the C02-HCO3--CO32-chemical system are all fast enough so that thetotal flux of inorganic carbon into biomass viaphotosynthesis is the rate-limiting step (19). Ofthe several rate reactions in the C02-HCO3-CO32- system, only the following reactions arerelatively slow (26): H2CO3 0C02 + H20 (equa-tion 5) at pH < 8; HC03- - C02 + OH- (equa-tion 6) at pH > 10; and both of these reactionsat pH 8 to 10. Thus, regardless of whether analga is an obligate C02 user or can assimilate

VOL. 41, 1981 63


4-)

4-)

s 0.6

0.44-)

t, 0.24zo

4-)4-)

v0 0.2 0.4 0.6 0.8 4.0 4.2 4.4 1.6 1.8 2.0

SPEC/F/C GROwTH RATrE,-doy-'FIG. 2. Relationship between and carbon cell quotas for C. vulgaris in inorganic carbon-limited contin-

uous culture (A) Q',~ dry weight basis. The curve was based on the mean values of Q'c. (B) Qc, cellular basis.The curve was determined by linear regression analyses of equation 3 and plots of equation 1. The asteriskindicates the QC from the averaged batch culture data in Table 1.

0.6

fa0--4

* . -00

FIG30

20

0 0.2 0.4 0.6 0.6 4.0 4.2 4.4 4.6 4.8

spEC/F/c GRowrH RATE.p.-doy4

FIG. 3. Relationship between 1L and carbon cell

quotas for S. obliquus grown in inorganic carbon-limited continuous cultures. (A) Q'¢, dry weight basis.The curve was based on the mean values of Qc. (B)Q,, cellular basis. The curve was determined byregression analyses of equation 3 and plots of equa-tion 1. The asterisk indicates the QcM from the aver-

aged batch culture data in Table 2.

HC03- directly, the uptake ofany carbon speciesis indistinguishable from the uptake of the totalinorganic carbon pool when the reactions de-scribed above are not rate limiting. For example,Goldman et al. (19) demonstrated that for a

range of growth rates, algal biomasses, and totalinorganic carbon concentrations similar to thoseused in the current study and for most natural

water situations in which the amount of totalinorganic carbon present is in excess relative tothe demand of phytoplankton, the reactions ofequations 5 and 6 generally are not rate limiting;hence, under these conditions it is valid to use

CT, as the substrate in equation 2.Moreover, when the pH is varied but the

chemical reactions in the C02-HC03--CO32- sys-

tem remain nonlimiting, it is impossible to de-termine the form of inorganic carbon used inphotosynthesis by comparing K. values that arebased on the relative C02 and HCO3 concentra-tions. The two K. values under these conditionsare always related by the equilibrium constantsdefining the chemical system (16). Lehman (27)carried these arguments further by showing thateven when C02 was the source of inorganiccarbon for photosynthesis, facilitated transportof HCO3 across cell membranes supplementedC02 transport to maintain high total fluxes ofinorganic carbon to the sites of photosynthesis.As in the previous study of Goldman et al.

(19), the relationship between u and CT, wasdescribed reasonably well by the Monod equa-tion, particularly for C. vulgaris (Fig. 1A). ForS. obliquus in contrast, although K. and u werewell described by the linear version of equation2, the shape of the curve in Fig. 1B is probablybetter described by a first-order-zero-order typeof relationship than by a rectangular hyperbola(equation 2). Without a more detailed investi-gation of this possible effect, which is beyond

** *0*

., 10

.B450

40 *~~

eoF *~~~~~~~~~~~-Og

* .. __*

"

0. **f . | .


IA


the aims of the present study, it is fruitless to > 4speculate further about the true shape of the S " 4

curves in Fig. 1. .8 oeooThe values of K. (based on CTI) for C. vulgaris o H

(0.20 ± 0.027 mg of C per liter) and S. obliquus m a(0.16 ± 0.052 mg of C per liter) were of the same o o o omagnitude as the K. values determined for two *other chlorophytes, S. capricornutum (0.40 mg Aof C per liter) and Scenedesmus quadricauda o Lo co(0.22mg ofC per liter), grown atpH 7.1 to 7.2 3 oool(19). These K. values are considerably lower u o V ;Nthan the values found for cultured and naturalpopulations of estuarine and marine phyto- eplankton measured during short-term 140 incu- l zbation studies (5, 28). However, the K, values t 0O C;determined in these latter experiments are not -'Acomparable to those of steady-state continuous t- tculture experiments. In the former case, thevalues were based on total inorganic carbonuptake over 1- to 2-h incubations, whereas inthe latter studies the values were determined as LO.ea function of the steady-state growth rate. Pho- Atosynthetic rates, particularly when measured * Xover short intervals, are not necessarily coupled . X Ito growth rates, mainly because short-term pho-tosynthesis is to a large degree dependent on thephysiological state of the cells at the time of X IH H * o osampling (i.e.,K.foruptakeisafunctionofu) 4P c X aO(32).When K. values for carbon uptake from dif- , P

ferent studies are compared, another important , 0consideration is the actual substrate measuredto calculate K.. For example, Markl (29) dem- 0 0oonstrated that there was a gradient in the C02 X Nmconcentration between the bulk fluid and the i Ncell surface; thus when C. vulgaris was main-tained in inorganic carbon-limited turbiostats, X m. m Nthe K. value was l1 yg of C perliter, based on T_ o EC02 levels at the cell surface, which were much C>. o 00 B

S +141+1H+1lower than the C02 levels in the bulk liquid (29). e- X .By comparison, this K, value is more than 2 0orders of magnitude lower than the K, valuesreported in this study, which were based on total .dinorganic carbon concentrations in the bulk 0A.fluid. Thus, it appears that when inorganic car- 0 o 0DCXbon is supplied pimarily in the gaseous form, -Z lathe true affinity for inorganic carbon at the cell 0surface is so high that the main mass transport 2 co3bottleneck occurs at gas-liquid interfaces. When 2,E SHC03- is the major source of inorganic carbon, -S.= .9the chemical conversion rates of HC03- to C0 20for obligate C02 users and the efficiency of eHC03- transport across cell membranes for spe-cies capable of facilitated HC03- transport (27).are the major potential rate bottlenecks. How- -l ,ever, it has been demonstrated repeatedly that 2L Xthe enzyme carbonic anhydrase, which catalyzes ao e u

the reactions of equations 5 and 6, is produced c- ci

VOL. 41, 1981 65


I-)

1JJ

8

4

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6SPEC/F/C GROWTrH RATE, --doy 1


1.8 2.0

FIG. 4. Relationship between,u and cellular carbon-nitrogen ratios in inorganic carbon-limited continuouscultures. (A) C. vulgaris. (B) S. obliquus. Lines were drawn by eye and are meant to demonstrate trends only.

-1i

~4J

:1:;;i

0~II

153O I ,I I I I I I I . r -

A

.100

50~~~~~~~~~~~

0~~~~~~~~~~~~~~~~~~~~.

50F *-i-.* -....... . :t--

.B

0.2

0.10i . *

o~~~~~~~~~00 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

SPEC/F/C GROWrH RA rE, - doy f

FIG. 5. Relationship between ji and cellular chlorophyll levels in inorganic carbon-limited continuouscultures of C. vulgaris. (A) Ratio ofcarbon to chlorophyll. (B) Cellular chlorophyll content. Lines were drawnby eye and are meant to demonstrate trends only.

-4 150

k A

8 50

0.2-

0.4

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 4.6 4.8SPEC/I/C GROWrH RATf, p-doyv

FIG. 6. Relationship between ,u and cellular chlorophyll levels in inorganic carbon-limited continuouscultures ofS. obliquus. (A) Ratio ofcarbon to chlorophyll. (B) Cellular chlorophyll content. Lines were drawnby eye and are meant to demonstrate trends only.

when cells are grown in low CO2 environments fore, it is virtually impossible to distinguish be-(1, 11, 21, 22, 25, 33); this provides additional, tween uptake of a particular form of inorganicalbeit indirect, evidence that microalgae have carbon and the response to the entire carbonvery high affinities for inorganic carbon. There- pool without rapid kinetic experiments, such as

A v.0 * *

-

I I , .

.0 0~~~~~-0--- 0.B . 0. * a_.% _ ._

4

u-


those used by Lehman (27) and Sikes et al. (44).Sources of inorganic carbon and ,i. The

ability of C. vulgaris or S. obliquus to grow atmaxiimum rates in batch culture at inorganiccarbon concentrations as low as 0.036% CO2(Tables 1 and 2) appears to be a common char-acteristic of many freshwater and marine algae(1, 38, 45, 52) and is another indication of theremarkable affinity which these organisms havefor inorganic carbon. There also is general agree-ment that maximum photosynthetic rates ofspe-cies such as Chlorella spp. and Scenedesmusspp. can be sustained with similar and evenlower C02 concentrations (2, 10, 47, 48).However, the percentage of C02 in the air

supplied to a culture is a relatively meaninglessterm in trying to ascertain the amount of C02required for maximum photosynthesis if no ac-counting is made of the concentration of the C02in solution which is really available to the algae(31). As demonstrated by Markl (29), this con-centration is a function of the sparging rate, thedegree of turbulence, and the combined effect ofthese processes on the C02 tension at the cellsurface, where the demand for inorganic carbonoccurs; for example, with optimum turbulencemaximum phytosynthetic rates were attainedwhen the percent C02 at the cell surface was0.0005% (29). The lowest sparging rate (gas bub-bling rate = 25 h-1) used in the current studyclearly was high enough to prevent any masstransport limitations.The apparent toxic effect of 100% C02 on both

species has been observed previously (46), al-though no satisfactory explanation exists for thisphenomenon. The decrease in ft at 5% C02 in airobserved for C. vulgaris (Table 1) is not sub-stantiated by similar data in the literature, as 5%C02 has been used commonly to prevent carbonlimitation in Chlorella and other algal cultures(25, 49). Possibly a lack of conditioning at thisC02 level led to the apparent reduction in,ii (49).Likewise, the decrease in ,i for S. obliquus withincreasing HCO3 concentrations greater than 3mM is difficult to explain. Osterlind (34) founda decrease in it with increasing HCO3 concen-trations and concomitant increasing pH values,which he attributed to C032- toxicity. In ourcultures the pH rose only slightly from 6.8 at 3mM HC03- to 7.7 at 10 mM HCO3 (Table 2),so that C032- levels were always minimal. Pratt(37) observed that the sodium salts of HCO32-and C032- had deleterious effects on algalgrowth; such an effect in our study cannot beruled out.

Effect of growth rate on carbon cellquota. The invariance in Q', with changing ,iobserved, representing -45 to 50% carbon in the

biomass (Fig. 2A and 3A), was identical to pre-vious results (19, 36) and conclusively demon-strated the inapplicability of equation 1 for de-scribing the relationship between,u and internalcarbon levels when inorganic carbon is limitingand Q is defined on a dry weight or total biomassbasis. Droop (9) pointed out that his originalformulation of equation 1 was based on theconsideration that total biomass was the properunit for calculating Q and that only when cellvolume was invariant with changing AL was itacceptable to replace biomass with cell numberin this term. Yet, the general convention in mostphytoplankton studies, both experimental (42)and theoretical (6), has been to use the concen-tration of internal nutrient per cell number as ameasure of Q. The choice of biomass units ac-tually need not be well defined because equation1 is empirical and has no fundamental theoreti-cal basis.On a cellular basis there was significant vari-

ation in Q, for both species. The degree of vari-ation in Qc, as indicated by the ratio of kQ to QCM(Table 3), is somewhat more pronounced for C.vulgaris (ratio of kQ to QCM, 0.16) than for S.obliquus (ratio of kQ to QCM, 0.26), althoughthere is overlap in the two values at the 95%confidence limits (Table 3). C. vulgaris, whichis considerably smaller than S. obliquus, mustbe capable of larger relative increases in cell sizewith increasing IL than S. obliquus, assumingthat cell size in the chlorophytes tested is di-rectly proportional to Q,. This latter inferenceseems reasonable for chlorophytes, which do notcontain vacuoles (30, 43, 50), even though cellsize was not measured in this study. Thus, cellsize may be as important a parameter in dictat-ing the potential range of Q for a particularlimiting nutrient as it is in influencing the abso-lute value of kQ (43).

Thus, the utility of equation 1 for describinginorganic carbon limitation in algae is restricted,and a and QCM must be determined experimen-tally, even when Q is defined as cellular carbon.

Cellular chemical ratios. The tight cou-pling between carbon assimilation on the onehand and nitrogen uptake (Fig. 4) and chloro-phyll synthesis (Fig. 5 and 6) on the other is bestrepresented by the lack of variance in the C-Nand C-chlorophyll ratios with changing ,A. Thevalue of the ratio of C to N for both species (5 to6) represents the lower limit possible with mi-crobes and indicates a cell population in a well-balanced nutritional state (i.e., .50% protein intotal biomass) (Goldman, in press). Similarly,ratios of C to chlorophyll of 50 to 100 are indic-ative ofwell-nourished cells (Goldman, in press).Holm-Hansen (24) recently demonstrated that

67VOL. 41, 1981


acetone extraction (as used in this study) leadsto slightly lower chlorophyll recoveries thanwhen methanol is used, particularly for analyseson chlorophytes. Hence, in retrospect our chlo-rophyll values may represent a systematic un-

derestimate; however, this would not change thetrends of the data in Fig. 5 and 6 nor the generalconclusion regarding the nutritional states ofthetwo chlorophytes.The effect of inorganic carbon limitation on

cellular chemical composition is quite differentthan the effect of nitrogen or phosphorus limi-tation. Under nitrogen limitation, the nitrogencell quota increases with increasing 1A, but gen-erally the carbon and phosphorus cellular con-

tents either remain constant (35, 40, 41) or in-crease in a threshold fashion only close to Iu (20).For phosphorus limitation both carbon and ni-trogen cellular contents typically are independ-ent of A (14, 35, 40). In contrast, the cellularchlorophyll content seems to increase with in-creasing,u regardless ofwhich nutrient is limiting(42).

It appears that the major effect of inorganiccarbon limitation on cell physiology is not so

much an effect on the chemical structure of thecells, but rather the influence of this limitationon cell size; decreases in macromolecular com-

ponents and corresponding decreases in cell sizeare related to decreasing ,u, which in turn repre-sents an increasing degree of inorganic carbonlimitation.Algal productivity. An important conse-

quence of the very low K. values established forinorganic carbon-limited growth is that thesteady-state level of algal carbon virtually isequal to CTO at all growth rates until just beforea' because CT. >> CT, (Fig. 1). Then algal produc-tivity increases linearly with increasing ,u and ismaximum just before ,u; this is followed by a

rapid decrease in productivity to a value of zeroat a (Fig. 7). Under these conditions peak pro-

ductivity is concomitant with high u. However,

I4 "

,S

a

0v

I I . I ,

A

.#1 s

/O I

-f. It .1 I

0 0.5 1.0 1.5 2.0 0 0.5 1.0

SPECIFIC GROWTr RArE, ,'-daK

1.5 2.0

FIG. 7. Relationship between A and algal produc-tivity in inorganic carbon-limited continuous cul-tures. (A) C. vulgaris, (B) S. obliquus. Lines were

drawn by eye and are meant to demonstrate trendsonly.

this situation is true only when HCO3- is thesource of the limiting nutrient and is supplied tothe culture as part ofthe influent liquid medium.When bubbled C02 which is supplied independ-ent of the medium is the source of inorganiccarbon, a decrease in algal biomass occurs withincreasing A, and peak productivity occurs when1 is considerably less than a (36). In attempts tooptimize productivity in algal mass cultures,consideration must be given to these bioengi-neering constraints (15).Conclusions. In this study the relationship

between growth rate and inorganic carbon limi-tation was reasonably well described by theMonod equation. The Droop equation was in-applicable when total biomass was used in placeof cell number in defining Q and of restrictedusefulness when ,u was related to cellular carbon.Both of these equations are empirical and mustbe used with caution in descriptions of algalgrowth response to nutrient limitation. Ratherinterestingly, the applicability of the Monodequation increases (i.e., measurable K8 values),whereas the utility of the Droop equation de-creases (i.e., large ratios of kQ to QM) when thelimiting nutrient comprises a larger fraction ofcellular biomass. Phosphorus, which constitutesa minute fraction of cellular biomass, and car-bon, which makes up -50% of the total cellularbiomass, represent the extreme examples of thisconcept.The affinity that algae have for inorganic car-

bon is high enough to prevent distinguishingbetween C02 uptake and HCO3 uptake on thebasis of chemical equilibrium considerations.The major factors controlling inorganic carbonuptake are physical mass transpsort bottlenecksat gas-liquid interfaces and the photosyntheticprocess itself. The mass flux of C02 or HCO3 orboth across cell membranes does not appear tobe a rate-limiting step.

ACKNOWLEDGMENTSThe technical assistance of H. Stanley and J. P. Clarner is

gratefully acknowledged.This work was supported in part by contract DE-AC02-

78ET20604 from the Department of Energy.

LMRATURE CITED1. Berry, J., J. Boynton, A. Kaplan, and M. Badger.

1976. Growth and photosynthesis of Chiamydomonasreinhardtii as a function ofC02 concentration. CarnegieInst. Washington Yearb. 75:423-432.

2. Briggs, G. E., and C. P. Whittingham. 1952. Factorsaffecting the rate of photosynthesis of Chlorella at lowconcentrations of carbon dioxide and in high illumina-tion. New Phytol. 51:236-249.

3. Brown, E. J., and D. K. Button. 1979. Phosphate-lim-ited growth kinetics of Selenastrum capricornutum(Chlorophyceae). J. Phycol. 15:305-311.

4. Burmaster, D. E. 1979. The continuous culture of phy-


INORGANIC CARBON LIMITATION IN MICROALGAE 69

toplankton: mathematical equivalence among threesteady-state models. Am. Nat. 113:123-134.

5. Caperon, J., and D. F. Smith. 1978. Photosyntheticrates of marine algae as a function of inorganic carbonconcentration. Limnol. Oceanogr. 23:704-708.

6. DiToro, D. M. 1980. Applicability of cellular equilibriumand Monod theory to phytoplankton growth kinetics.Ecol. Mod. 8:201-218.

7. Droop, AL R. 1968. Vitamin B12 and marine ecology. IV.The kinetics of uptake, growth and inhibition in Mon-ochrysis lutheri. J. Mar. Biol. Assoc. U. K. 48:689-733.

8. Droop, M. 1973. Some thoughts on nutrient limitation inalgae. J. Phycol. 9:264-272.

9. Droop, M. R. 1979. On the definition of X and Q in thecell quota model. J. Exp. Mar. Biol. Ecol. 39:203.

10. Emerson, R., and L. Green. 1938. Effect of hydrogen-ion concentration on Chlorella photosynthesis. PlantPhysiol. 13:157-168.

11. Findenegg, G. R. 1976. Correlations between accessibilityof carbonic anhydrase for external substrate and regu-lation of phytosynthetic use of CO2 and HC03- byScendesmus obliquus. Z. Pflanzenphysiol. 79:428-437.

12. Goldman, J. C. 1977. Steady state growth of phytoplank-ton in continuous culture: comparison of internal andexternal nutrient equations. J. Phycol. 13:251-258.

13. Goldman, J. C. 1979. Bioengineering aspects of inorganiccarbon supply to mass algal cultures, p. 25-32. In Pro-ceedings of the Third Annual Biomass Energy SystemsConference. Report SERI/TP-33-285. Solar Energy Re-search Institute, Golden, Colo.

14. Goldman, J. C. 1979. Temperature effects on steady-state growth, phosphorus uptake, and the chemicalcomposition of a marine phytoplankter. Microb. Ecol.5:153-166.

15. Goldman, J. C. 1979. Outdoor algal mass cultures. II.Photosynthetic yield limitations. Water Res. 13:119-136.

16. Goldman, J. C., D. Jenkins, and W. J. Oswald. 1974.Discussion-the kinetics of inorganic carbon-limited al-gal growth. J. Water Pollut. Control Fed. 46:2785-2787.

17. Goldman, J. C., and J. J. McCarthy. 1978. Steady stategrowth and ammonium uptake of a fast-growing marinediatom. Limnol. Oceanogr. 23:695-703.

18. Goldman, J. C., J. J. McCarthy, and D. G. Peavey.1979. Growth rate influence on the chemical composi-tion of phytoplankton in oceanic waters. Nature (Lon-don) 279:210-215.

19. Goldman, J. C., W. J. Oswald, and D. Jenkins. 1974.The kinetics of inorganic carbon limited algal growth.J. Water Pollut. Control Fed. 46:554-574.

20. Goldman, J. C., and D. G. Peavey. 1979. Steady-stategrowth chemical composition of the marine chlorophyteDunaliella tertiolecta in nitrogen-limited continuousculture. Appl. Environ. Microbiol. 38:894-901.

21. Graham, D., C. A. Atkins, M. I Reed, B. D. Patterson,and R. M. Smillie. 1971. Carbonic anhydrase, photo-synthesis, and light-induced pH changes, p. 267-282. InC. B. Osmond and R. 0. Slater (ed.), Photosynthesisand photorespiration. Wiley-Interscience, New York.

22. Graham, D., and M. L. Reed. 1971. Carbonic anhydraseand the regulation of photosynthesis. Nature (London)New Biology 231:81-83.

23. Guillard, R. R. L., and J. HI Ryther. 1962. Studies onmarine plankton diatoms. L. Cyclotella nana Hustedtand Detonula confervacea (Cleve) Gran. Can. J. Micro-biol. 8:229-239.

24. Holm-Hansen, 0. 1978. Chlorophyll a determination:improvements in methodology. Oikos 30:438-447.

25. Ingle, R. K., and B. Colman. 1975. Carbonic anhydraselevels in blue-green algae. Can. J. Bot. 53:2385-2387.

26. Kern, D. M. 1960. The hydration of carbon dioxide. J.Chem. Educ. 37:14-23.

27. Lehman, J. T. 1978. Enhanced transport of inorganic

carbon into algal cells and its implication for the biolog-ical fixation of carbon. J. Phycol. 14:33-42.

28. Loftus, M. E., A. R. Place, and H. H. Seliger. 1979.Inorganic carbon requirements of natural populationsand laboratory cultures of some Chesapeake Bay phy-toplankton. Estuaries 2:236-248.

29. Mark], H. 1977. CO2 transport and phytosynthetic pro-ductivity of a continuous culture of algae. Biotechnol.Bioeng. 19:1851-1862.

30. Mullin, M. M., P. R. Sloan, and R. W. Eppley. 1965.Relationship between carbon content, cell volume, andarea in phytoplankton. Limnol. Oceanogr. 11:307-311.

31. Myers, J. 1944. The growth of Chlorella pyrenoidosaunder various culture conditions. Plant Physiol. 19:579-589.

32. Myers, J. 1970. Genetic and adaptive physiological char-acteristics observed in the chlorellas, p. 447-454. In I.Setlik (ed.), Prediction and measurement of photosyn-thetic productivity. Proceedings of the IBP/PP Tech-nical Meeting. Centre for Agriculture Publications andDocumentation, Wageningen, The Netherlands.

33. Nelson, E. B., A. Cenedella, and N. E. Tolbert. 1969.Carbonic anhydrase levels in Chlamydomonas. Phyto-chemistry 8:2305-2306.

34. Osterlind, S. 1949. Growth conditions of the alga Scen-desmus quadricanda with special reference to inor-ganic carbon sources. Symb. Bot. Ups. X:1-141.

35. Panikov, N., and J. J. Pirt. 1978. The effects of coop-erativity and growth yield variation on the kinetics ofnitrogen or phosphate limited growth of Chlorella in achemostat culture. J. Gen. Microbiol. 108:295-303.

36. Pipes, W. O. 1962. Carbon dioxide-limited growth ofChlorella in continuous culture. Appl. Microbiol. 10:281-288.

37. Pratt, R. 1943. Studies on Chlorelha vulgaris. VIII. Influ-ence on photosynthesis ofprolonged exposure to sodiumbicarbonate and potassium bicarbonate. Am. J. Bot. 30:626-629.

38. Pruder, G. D., and E. T. Bolton. 1979. The role of CO2enrichment of aerating gas in the growth of an estuarinediatom. Aquaculture 17:1-15.

39. Rhee, G.-Y. 1973. A continuous culture study of phos-phate uptake, growth rate and polyphosphate in Sce-nedesmus sp. J. Phycol. 9:495-506.

40. Rhee, G.-Y. 1974. Phosphate uptake under nitrate limi-tation by Scenedesmus sp. and its ecological implica-tions. J. Phycol. 10:470-475.

41. Rhee, G.-Y. 1978. Effects of N:P atomic ratios and nitratelimitation on algal growth, cell composition, and nitrateuptake. Limnol. Oceanogr. 23:10-24.

42. Rhee, G.-Y. 1980. Continuous culture in phytoplanktonecology, p. 151-203. In M. R. Droop and H. W. Jannasch(ed.), Advances in aquatic Microbiology, vol. 2. Aca-demic Press, Inc., New York.

43. Shuter, B. J. 1978. Size dependence of phosphorus andnitrogen subsistence quotas in unicellular microorga-nisms. iimnol. Oceanogr. 23:1248-1255.

44. Sikes, C. S., R. D. Roer, and K. M. Wilbur. 1980.Photosynthesis and coccolith formation: inorganic car-bon sources and net inorganic reaction of deposition.Limnol. Oceanogr. 25:248-261.

45. Small, L. F., P. L. Donaghay, and R. M. Pytkowicz.1974. Effects of enhanced CO2 levels on growth char-acteristics of two marine phytoplankton species, p. 183-204. In N. R. Anderson and A. Malahoff (ed.), The fateof fossil fuel CO2 in the ocean. Plenum Press, New York.

46. Sorokin, C. 1962. Inhibition of cell division by carbondioxide. Nature (London) 194:496-497.

47. Steeman-Nielsen, E. 1953. Carbon dioxide concentra-tion, respiration during photosynthesis and maximumquantum yield of photosynthesis. Physiol. Plant. 6:316-332.

48. Steeman-Nielsen, E. 1955. Carbon dioxide as carbon

VOL. 41, 1981


source and narcotic in photosynthesis and growth ofChlorella pyrenoidosa. Physiol. Plant. 8:317-335.

49. Steeman-Nielson, E., and M. Willemoes. 1966. Theinfluence ofC02 concentration and pH on two Chlorellaspecies grown in continuous light. Physiol. Plant. 19:279-293.

50. Strathmann, R. R. 1967. Estimating the organic carboncontent of phytoplankton from cell volume or plasma


volume. Limnol. Oceanogr. 12:411418.51. Strickland, J. D. HI, and T. R. Parsons. 1972. A prac-

tical handbook of seawater analyses, 2nd ed. FisheriesResearch Board of Canada Bulletin 167. Fisheries Re-search Board of Canada, Ottawa.

52. Swift, E., and W. R. Taylor. 1966. The effect of pH onthe division rate of the coccolithophorid Cricosphaeraelongata. J. Phycol. 2:121-125.

Documents

Inorganic Carbon Limiitation and Chemical Composition of Two