Physiological Characteristics of Spirulina Platensis (Cyanobacteria) Cultured at Ultrahigh Cell DENSITIES1

J . Phycol. 32, 1066-1073 (1996)

PHYSIOLOGICAL CHARACTERISTICS OF SPZRULINA PLATENSIS (CYANOBACTERIA) CULTURED A T ULTRAHIGH CELL DENSITIES'

Hu Qiung2 Microalgal Biotechnolog) Laboratory, T h e Jacob Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boker Campus 84990, Israel

Hugo Guterman Department of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Box 653, Beer Sheva, Israel

and

Amos Richmond3 Microalgal Biotechnology Laboratory, T h e Jacob Blaustein Institute for Desert Research, Ben-Gurion University of the Segev, Sede Boker Campus 84990, Israel

Kej i d e x usords: chemical compositiou; chlorophjll Juoresccme; cjambacteria; dark rrspirntiotz; taight biomass loss; photobioreactor; photosjiithrsis; productivitj; Spirulina platensis; rcltrahigh cell d e ~ z ~ i t ~

Received 19 February 1996. Accepted 12 .4ugust 1996. Present address: Marine Biotechnology Institute, Kamaishi Laboratory, Heita, Kamaishi cit!, Iwate 026, Japan. Author for reprint requests. 2

1066

T h e reason for mass cultivation of microalgae outdoors involves solar energy utilization. Trapping solar energy a t as high a n efficiency as theoretically feasible to produce chemical energy in a desirable form a t maximal rates represents a major issue in this biotechnology. I n general, as net photosynthetic production per reactor illuminated surface area increases, the economic basis for this novel biotechnology also increases. Most research on mass cultivation of microalgae r\.as, until recently, focused on open raceways, that is, shallow (ca. 15-20 cm deep) basins covering a n area of 500-5000 m 2 (Dodd 1986), the standard cultivation device presently employed by industry. In contrast to heterotrophic microorganisms, which are cultured in cell densities amounting t o tens of grams dry weight per liter of culture suspension, cell density of photoautotrophs in open raceways is very lo\\., that is, a few hundreds of mg dry weight per liter, representing a heavy burden on the cost of production. Thus, several heterotrophic microorganisms form the basis of most important economic endeavors, and industrial production of photoautotrophs is still in its infancy. In the past few years, however, enclosed photobioreactors have attracted much attention (e.g. Pirt et al. 1983, Lee 1986, Cohen and Arad 1989, Tredici and Materassi 1992, Chaumont 1993, Richmond et al. 1993). An intrinsic advantage of .photobioreactors is t h i capacity t o support cultures of cell densities over twice a n order of magnitude greater than used in open raceways. There are obvious advantages i n Operating cultures at as high a n optimal cell density as possible. First, the higher the cell density, the easier it is to

PHYSIOLOGICAL CHARACTERISTICS O F SPIRULINA PLATENSIS

harvest biomass and maintain monoalgal cultures, a basic requirement for industrial production (Richmond et al. 1990). Also, capital investment in reactor volume, an essential element of production cost, decreases. A newly designed flat plate photobioreactor for outdoor production of microalgae has been described (Hu et al. 1996). It consists of a flat glass container with a narrow (1.3-2.6 cm) light path and in which vigorous stirring by compressed air is provided. Using this reactor made it possible, for the first time, to grow Spirulina platensis outdoors in cell concentrations reaching well over 30 g dry weight (ca. 300 mg chlorophyll) per liter culture. In this work, we report on some elementary physiological parameters that characterize the adaptive response of Spirulina platensis maintained in ultrahigh cell densities (UHCDs, a term suggested for cultures in which the optimal population density is above 10 g dry weight or 150 mg chlorophyl1.L-' cuiture) and resulting in record output rates. We wished to improve the methodology for utilizing high photon flux densities (PFDs) for cultivating microalgae to obtain maximal photosynthetic productivity and efficiency. MATERIALS AND METHODS

Spirulina platensis strain M2 (obtained from the Culture Collection of the Centro di Studio dei Microrganismi Autotrofi of Florence and provided through the courtesy of Dr. A. Vonshak) was grown in Zarouk nutrient medium (Zarouk 1966). Laboratory experiments were conducted in glass columns (ID = 2.6 cm, 250 mL culture capacity) immersed in a 35" C water bath. Light was provided by a series of halogen lamps (HQI-IS 1500W, ISRAM, Israel). Irradiance was modified by changing the distance of the light source from the culture column. T h e outdoor culture facility consisted of a series of the newly designed flat photobioreactor (Hu et al. 1996), each of which measured 70 cm high, 90 cm long, and 2.6 cm (inner diameter) wide. Each reactor contained 14 L of algal culture agitated by bubbling air through perforated tubes running along the bottom of the reactor and at half its height, at the rate of 2.0-3.5 L of air.L-l of algal suspension .min-', depending on the population density. Culture pH ranged from 9.2 to 9.8 and was maintained by adjusting the ratio of CO, to air using gas flow meters. T h e reactors were set outdoors with a tilt angle of 30" facing south. T h e optimal culture temperature of 35" C was maintained in daylight by evaporative cooling provided by water sprayed on the reactor front glass panel. Analjticai and chemical methods. T h e dry weight of the harvested cell mass was determined using 5-10 mL of culture, which was filtered through predried and preweighed Whatman GF/C glassfiber filters, washed with 0.5 N HCI, and dried overnight in an oven at 105" C. Specific growth rate was calculated by measuring dry weights of samples taken at different time periods, using equation (1): g = (In X, - In X,)/t, - t,, (1) where X, and X I are the mean dry weights at time t, and tl, respectively, and g is the calculated hourly specific growth rate when the culture is at steady state. T h e volumetric output rate (PJ was calculated by measuring dry weight of the biomass, using equation (2): p,

=

(X, - Xl)V/t, - tl,

where V represents total culture volume.

(2)

1067

T h e areal productivity (Pa)was calculated by multiplying P, with the culture volume (V) per square meter of reactor illuminated surface area: P,

=

P;V.

(3)

Night biomass loss (NBL) was calculated as follows: NBL(g.L-')

=

X,

-

X',,

(4) where X, and X,' are the biomass concentrations in dry weights at the end of daylight and just before the following sunrise, respectively. To discern between the percentage of NBL with respect to cell mass produced in daylight and the percentage of loss with respect to overall cell density, the following equations were used: NBL per net daylight output rate (%)

=

100

x, - x:, xe - x m ,

~

(5)

where X, is the cell concentration just before sunrise of the same day. NBL per total biomass concentration (%)

=

100

-x, . - XL

x,

(6) T o replace the growth medium in experiments o n mineral nutrition or in which very high cell densities were used, agitation of the culture was stopped for 2-4 h, permitting the Spirulina filaments to settle. T h e supernatant that was clear and devoid of Spirulina filaments was completely removed and replaced with fresh growth medium. PFD was measured using a Li-Cor meter with a quantum sensor. Variable to maximal chlorophyll fluorescence (FJF,) was measured with a PEA fluorometer (Hansatech, UK); the samples were dark-adapted for 7 min before measurement. T h e volumetric 0, mass transfer coefficient (K,a) was measured using the oxygen production rate (OPR) method (Guterman et al. 1989). Chlorophyll a (Chl a) was extracted with 90% methanol in a 70" C water bath for 2 min and determined from its absorbance at 665 nm according to Vonshak et al. (1982). The c-phycocyanin (cPC) of freeze-dried algal biomass was extracted with 1% CaCI, for 5 h in darkness. T h e crude extracts were analyzed spectrometrically using extinction coefficient El" Icm = 73, at 620 nm absorption maximum (Boussiba and Richmond 1980). Protein was assayed as described by Lowry et al. (1951) after hydrolysis in 0.5 N HC1 for 1 h at 90" C. Carbohydrates were analyzed by the phenol-sulfuric acid method of Kochert (1978). Lipids were determined according to Cohen et al. (1993); they were transmethylated by treating the freeze-dried cells with methanol-acetyl chloride. Heptadecanoic acid was added as an internal standard, and fatty acid methyl esters were identified by co-chromatography with authentic standards (Sigma Co.) and by calculation of the equivalent chain length. Fatty acids were determined by comparing each chromatographed peak area with that of the internal standard. Gas-chromatographic analysis was performed on a Supelcowax 10 fused silica capillary column (30 m x 0.32 mm) at 200" C (FED, injector and flame ionization detector temperature 230" C, split ratio 1:lOO).

RESULTS AND DISCUSSION

Assessment of nutrient su_tfiriPnry Because the commonly used Zarouk nutrient medium (Zarouk 1966) for Spirulina was not originally designed to support UHCD, nutrient depletion may rapidly occur in enclosed photobioreactors. A meaningful evaluation of physiological characteristics affected by UHCD mandates, however, that light be the only limiting factor due to self-shading of cells (Tamiya 1957).

1068

H U QIANG ET .4L. 500

1 0 . 8

v)

m

35

-

30

25

r “\

20

15

4 0.4

1

10

5 >

c 0 d

--_

0 0

5

10

15

20

25

30

35

40

Cell density (g L-’)

Time (days)

FIG. I . Effect of replacing culture supernatant with fresh growth medium on biomass accumulation in outdoor Spirulznn cultures grown with a semicontinuous mode in the flat plate 2.6cm-wide reactor outdoors. Arrou s indicate time of medium replacement. White and black symbols indicate the culture behavior without and with replacement of fresh gro\\,th medium, respectively.

Therefore, w e first attempted to identify possible mineral limitation in rapidly gro\\.ing cultures. T h e relationship of cell density to frequency of replacing culture supernatant with fresh growth medium was studied, Standard Zarouk medium supported growth to a maximal cell density- of ca. 1 2 g . L - ’ in 6 o r 7 days (Fig. 1). When cell density, ho\vever, reached ca. 10 g.L-’, cell chlorophyll and protein content declined sharply followed by an increase in turbidity of the culture supernatant due to cell Iysis and bacterial proliferation (data not shoivn). Once cell density reached ca. 12 g.L-’, growth ceased altogether, suggesting depletion of some nutrient and/or a buildup of an inhibitor. Groij-th was resumed by replacing the entire supernatant Lvith fresh growth medium. In this manner, algal growth and cell density continued to increase (albeit at a reduced growth rate due to increasing light limitation), up to that cell density at which further growth could not occur by increasing the turbulence flow of the algal suspension o r refreshing the culture medium. T h e highest cell density achieved Ivas 34 g dry wt.L-’ (ca. 300 mg Chl n.L-’), an extremelj- high cell density, not yet reported for outdoor cultures of photoautotrophic microorganisms (Fig. 1). In laboratory experiments, the culture was started with a cell concentration of 3 g . L - ’ in Zarouk medium. Growth ceased after ca. 30 h (at ca. 18 g.L-’) at which point 2.5 g NaNO,, or 0.5 g K,HPO, or 4 mL microelement solution (A5 and B6 in the original Zarouk medium) per liter of culture lvas added. Addition of nitrate alone supported g r o u ~ hto a biomass as high as 30 g.L-’. In contrast, addition of phosphate only or of the microelements alone did not result i n any significant effect on gro\\.th. Therefore, nutrient sufficiency \\.as maintained in all our outdoor experiments by doubling the quantity of

FIG. 2. Effect of cell density o n specific growth rate (o),F,/ F, (A), and output rate (0).Cultures w’ere grown in glass columns in the laboratory under nutrient and carbon repletion as well as optimal temperature and pH. Black bar and “OCD” indicate the range of optimal cell density.

sodium nitrate in the Zarouk medium and replacing the entire culture medium with fresh growth medium every 6-4 days, depending on culture density. Raspouse of C‘HCD to escass light. Output rate of biomass was “bell-shaped” (Richmond 1992) with a very wide range (from 7 to 14 g.L-’) of optimal cell density (OCD, defined as that cell concentration in dry weight, which at steady state would result, under given conditions, in maximal output rate of biomass per reactor volume) (Fig. 2). Maximal specific growth rate \\.as obtained when light limitation was minimal, that is, at a relatively low cell density of ca. 2.5 g. L-I (Fig. 2). T h e specific growth rate declined thereafter exponentially with increasing cell density. A cell density lower than 2.5 g.L-’, however, resulted in a sharp decline in the specific growth rate due to photoinhibition (Krause 1988, Vonshak and Guy 1992). We thus focused on UHCD in modifying the extent of photoinhibition that results when photos)-stem I1 (PS 11) is exposed to excessive light (Powles 1984, Krause 1988, Ohad et al. 1994). Vonshak and Guy (1992) were the first to describe the occurrence of photoinhibition in outdoor Spirulina cultures. They observed that reducing irradiance by shading the cultures resulted in a n increase in photosynthetic activity as well as in a n increase in productivity of biomass. In the open raceways, however, the optimal population density was ca. 0.5 g.L-’, whereas cell densities (ca. 10-20 g.L-’) were 20-30 times higher in our flat plate photobioreactor. Would photoinhibition and reduced productivity also occur in U HCD of Spirulina? T h e photochemical efficiency of PS I1 exposed to high PFD of 2000 pmol-m-2. S K I as a function of cell density was studied under laboratory conditions, using Spirulina cultured in glass columns. PS I1 efficiency was estimated by measuring variable to maximal chlorophyll fluorescence (FJF,, Krause 1991). A 6-h exposure to high radiation showed that the higher the cell density, the

PHYSIOLOGICAL CHARACTERISTICS OF SPIRULINA PLATENSIS

smaller the photoinhibitory decline in PS I1 efficiency, and when the population density was higher than 14 g*L-' no detectable decline in FJF, was observed. Clearly, photoinhibition of photosynthesis could be diminished or altogether controlled by manipulating cell density. T h e effect of culture density on modifying response to excess light occurring at the high PFD prevailing outdoors was also studied. Three different population densities established in the flat plate photobioreactors in midsummer (June) were used (Fig. 3). A similar pattern to that obtained under controlled laboratory conditions emerged: the lower the cell density, the higher and faster was the decline in F,/F, taking place from morning to noon in response to the increase in solar energy. Also, the higher the cell density the faster was the recovery following the midday decline. At cell density above 16 g dry weight.L-', F,/F, remained constant at its maximal level during daytime; that is, it was completely unaffected by the maximal rates of irradiation at noon (Fig. 3a). T h e highest hourly areal output rates at all cell densities coincided with the midday peak of solar irradiance. Up to the optimal density, the output rate of biomass during the day was proportional to cell density (Fig. 3b). Spirulina at optimal cell density, however, which yielded the highest areal output rate, was apparently photoinhibited for a few hours at midday as indicated from F,/F, measurements. In contrast, no detectable photoinhibition was evident when cell density was above optimal, which resulted in a decline in the output rate. Photoinhibition may become pronounced upon exposure of plants to bright light at low, chilling, or freezing temperatures, which prevail outdoors in winter and impede growth (Powles 1984, Oquist et al. 1987, Krause 1994). If self-shading as affected by culture density provides protection from excess light, would elevating the cell density arrest low temperature induced photoinhibition? This assumption was tested in the laboratory, by exposing cultures at different densities to a shift down in temperature; cultures with different population densities were all exposed to a PFD of 1000 pmol.m-2.s-1 at 35" C. Steady-state levels of photoinhibition under these conditions were reached after 5 h. Culture temperature was then shifted down to 4"C for the duration of 4 h without altering the PFD. A significant decline in F,/F, took place within 1-2 h, the lower the population density the higher was the reduction in FJF,. Upon a shift-up to 35" C, a rapid increase in FJF, took place within 4 h in all treatments except the one with the lowest cell density of ca. 0.4 g*L-', in which the temperature shift-up affected bleaching and lysis of cells, a clear indication of photooxidation (Cadenas 1989, Foyer et al. 1994). Other experiments conducted outdoors in winter (December) also demonstrated that a high population density-for example, 15 g dry weight. L-' prevented photodamage typically occurring at low temperature.

0.8 0.7 E

1069

a

-

0.6-

LL

\

>

u-

0.5 0.4 -

0.3

'

6

8

10

12

14

16

18

20

22

Time of day (h) FIG. 3. Effect of culture density on a) FJF, and b) output rate of biomass during the day in outdoor cultures of S. platensis in the flat plate photobioreactor during June-July. Cultures were operated on semicontinuous modes. Population density (8.L-I): = 1.8; A = 8.5; 0 = 15.2.

T h e physiological meaning of the F,/F, value in terms of delineating culture potential for growth and productivity requires careful interpretation. If the rate-limiting steps involved in the photosynthetic activity by microalgae include the rate of carbon assimilation as suggested by Sukenik et al. (198'7a, b) and Lee and Low (199 1), the measured potential activity of PS I1 could be irrelevant to the culture growth potential. Thus, in many cases, a decline in capacity of PS I1 as indicated by FJF, may represent a mechanism of adjusting to excess light, bearing no relationship to cell growth and productivity. Accordingly, photodamage to PS I1 would affect microalgal growth only under extreme conditions of excess light, such as in a low population density culture exposed to high PFD as well as in the case of low temperature. OPR as affected b j cell d e m i t j . T h e OPR of a culture was a reliable and accurate parameter for estimating the rate of carbon assimilation and productivity as well as the general state and vitality of the algal suspension (Ben-Yaakov et al. 1985, Guterman et al. 1989). OPR as a function of irradiance and cell density increased with increasing irradiance, the magnitude of the increase depended on cell density (Fig. 4). At 2.0 g.L-', the lowest cell density tested,

1070

H U QIANC ET AL.

/-

F

30 25

v ~

300 20

e

mL

C

0 .c

200

15

0

10

U

E

0 3

-0

2 Q

100

5

5

0

0"

0

0 400

800

1200

1600

Light intensity (pmol m-'

2000

s-')

5

10

15

20

25

30

0 35

Cell density (g L-')

FIG. 4. 0, production rate in cultures of low, high, and optimal cell densities as affected bt irradiance outdoors. Cultures were operated o n a semicontinuous mode. Cell density (g,L-'): 0 = 2.0; a = 9.0: 0 = 16.0.

FIG. 5. Effect of cell density o n night biomass loss in the flat plate reactor outdoors. Culture temperature at night ranged between 12" and 15" C. 0 = calculated on the basis of net daylight output rate; 0 = calculated on the basis of net daylight output rate; 0 = calculated o n the basis of total cell density in dry weight reached at the end of daylight. Bars = * I SD.

the magnitude of the response to increasing irradiance was also the lowest. A linear relationship existed between irradiance and OPR up to 1500 pmol. m--2--1 only, above which an increase in irradiance exerted little effect on OPR, indicating the system was light-saturated. Likewise, at a cell density of 16.0 g . L-', significantly above the optimal, the magnitude of response to increasing irradiance was similar to that found for the verj- low cell density, except that no saturation in light \vas evident. Indeed, the productivity of the overly high 16 g. L-' culture was similar to that of the too lo\v 2.0 g,L-' culture, reflecting the result of inhibited photosynthetic activity and increased maintenance energy (Pirt 1982) due to extreme light limitation. At the optimal cell density of 9 g.L-', OPR responded most significantly to increasing irradiance, reflecting exposure of cells to the optimal light regime (Richmond 1988) associated with the highest productivity and photosynthetic efficiency (Fig. 4). Efect ofC'HCDs o n .YBL. NBL, \vhich may strongly affect net photoautotrophic productivity, is a combined term to account for several processes leading to a reduction in biomass at night (e.g. dark respiration, cell death). Dark (nighttime) respiration is the most important component of NBL. Torzillo et al. (199 1) reported that the rate of dark respiration depended on the temperature and irradiance to which the algae were exposed during the preceding day. I t also depended on culture temperature prevailing during the night (Grobbelaar and Soeder 1985, Torzillo et al. 1991). Reports concerning the effects of cell density on dark respiration and NBL are, however, conflicting. According to Guterman et al. (1989) and Sukenik et al. (1991), in open systems maintained at a relatively low cell density (less than 2 g ' L - l , dark respiration was enhanced by increasing cell density. In contrast, Vonshak et al. (1982) and Radmer et al. (1 987) have reported that

a reverse correlation existed between cell density and dark respiration. O u r results indicate that whereas NBL as percentage of the daylight output rate was high a t low culture densities, it decreased to minimum as cell density increased from 0.8 to 6.5 g.L-' (Fig. 5). When cell density rose further from ca. 7 to 20 g.L-', the rate of NBL remained fairly stable a t its lowest level, consisting of ca. 5% of daylight output rate. Further increasing cell density to ultrahigh densities of 25-30 g.L-' resulted, however, in elevation of NBL as a fraction of the daylight biomass output. In contrast, on the basis of the total biomass concentration in the culture, NBL decreased exponentially as cell density increased. Indeed, when cell density was 8 g.L-' or above, NBL as a fraction of the total algal mass in the culture became negligible. These findings a r e not consistent with reports of NBL as accounting for 1050% of net daytime production (Grobbelaar and Soeder 1985, Guterman et al. 1989, Sukenik et al. 1991, Torzillo et al. 1991). We thus wish to stress that the pattern of low NBL (Fig. 5) was repeatedly obtained in our UHCD cultures for 2 years of experiments in both winter and summer. Measurements of dark respiration by the algae as indicated by 0, uptake essentially confirmed the pattern for NBL based o n measurements of dry weight alone (Fig. 6). O n the basis of culture volume low cell density cultures exhibited relatively high dark respiration rates, reflecting the high oxygen concentration per cell, whereas dark respiration declined as cell density was increased. At a wide range of cell densities (e.g. from 5 to 20 g'L-'), dark respiration was minimal and was fairly constant. A further increase in cell density, however, resulted in increased dark respiration d u e probably t o the heavy cell load. An exponential decrease in dark respiration, however, was evident on the basis of cell mass in response to a n increase in cell density, indicating

PHYSIOLOGICAL CHARACTERISTICS O F SPIRULINA PLATENSIS

50

r

1071

50

7 C

40

2

50

m

30

3

40

(o

a,

20

9

30

s

10 I0 0 05

_ 10

15 j

20

25

30

35 0

2 3 --=,

20

0"

10 0

5

10

Cell density (g L")

FIG.6. 0,uptake per culture volume (0)and per culture mass (0)as affected by cell density. Measurements were carried out in situ. Culture conditions were the same as for Figure 5.

that the higher the cell density the lower the dark respiration rate of the individual cells. Measurements of the volumetric mass transfer coefficient (K,a) of oxygen in outdoor cultures may explain this. T h e K,a of 0, declined linearly as cell density increased (Fig. ?), resulting in reduced O2 pressure per cell, which probably blocked the full respiratory capacity. T h e decrease in dark respiration rate associated with increasing cell density augments net productivity and represents an additional advantage for UHCD cultures. The effect of cell density o n major cell components and their production rates. Microalgae undergo significant physiological and chemical changes in response to variations in light, temperature, and nutrient availability (Tedesco and Duerr 1989, Thompson et al. 1992, Brown et al. 1993, Cohen et al. 1995). T h e effect of cell density on cell chemical composition has not been given sufficient attention; the few available reports relate to cultures with relatively low cell densities (Cohen et al. 1993, Chrismadha and Borowitzka 1994, Grobbelaar 1995). Cell composition as affected by UHCDs as used in this work has not been reported. Cell content of Chl a , c-PC, proteins, carbohydrates, and total fatty acids (TFAs) varied greatly over a wide range of cell densities in outdoor cultures. Highest cell contents in protein and TFAs were evident in cultures maintained at relatively low cell densities and thus low light limitation, with a relatively fast growth rate. With increasing cell density, both components decreased gradually, reaching their minimum at the highest cell density. In contrast, the lowest concentration of cell carbohydrates was found at low cell densities, accumulation of carbohydrates taking place in direct proportion to the increase in cell density and the concomitant decline in growth rate (Figs. 2, 8a). Chlorophyll a increased with increasing cell density, reaching maximum at ca. 10 g.L-', that is in the range of the optimal density yielding the highest photosynthetic efficien-

15

20

25

30

35

Cell density (g L.')

FIG. 7. T h e volumetric mass transfer coefficient (K,a) of 0, as a function of cell density in situ. Culture conditions were the same as for Figure 5 .

cy and the highest net productivity of biomass per unit of culture volume. When cell density was above optimal, Chl a declined consistently, as was also found for c-PC. Highest c-PC was obtained at a relatively low cell density (ca. 2.5 g.L-'), decresing gradually with increasing cell density above this value. A de-

0.7 0

5

10

15

20

25

30

35

Cell density (g L-')

FIG. 8. Effect of cell density on a) protein, carbohydrate, and total fatty acid content and b) cell chlorophyll and c-PH of S. platensis grown outdoors in the flat plate reactor. Experiments were carried out in the reactors (2.6 cm light path) outdoors on a semicontinuous mode during May and July. T h e temperature was maintained at 35" C during the daylight period. Culture temperature at night followed the gradual decline in ambient temperatures, reaching 12"-15" C before sunrise. 0 = Protein, A = carbohydrate, A = TFAs 0 = Chl a , 0 = c-PC. Bars = _+ 1 SD.

HU QIANG ET AL.

1072 5

-n

-25

4

N

-20

-. 2 s.

2 C

-10

oLL--

0

__

-~

5

10

15

20

(P

Whereas cell contents of each major cell component differed in response to modification in cell density, maximal production rates of c-PC, T F A and protein per unit of culture volume were all obtained in a similar cell density range, that is, 5-10 g.L-', the optimal for maximal productivity of cell mass under the experimental conditions (Fig. 9). Maintaining cultures a t O C D therefore not only sustains maximal biomass output rate but, as reported for other microalgae, may also result in highest production rates of special products.

- -----a0

25

30

35

Cell density (g L ' )

FIG. 9. Effect of cell density on output rate of protein (a), c-PC 0,and T F A (A)of S platpuss Culture conditions were the same as for Figure 8.

cline in c-PC was also evident \%.hencell density was below 2 g.L-' (Fig. 8b). We interpret the changes in cell composition with the increase in cell density and light limitation to represent a photoadaptive strategy. It is well known cell densit), coupled with a high that a relatively 10%~ irradiance results in relatively low cell contents of Chl n and c-PC. Increasing cell densities a r e associated with decreasing light availability for the average cell, and a buildup of photosynthetic pigments to augment the light-harvesting capability under these circumstances is well documented (e.g. Post et al. 1984, Sukenik et al. 1990). Continued photosynthetic pigment buildup, however, would lead to extreme attenuation of incident light and a decrease in the average optical cross-section of the photosynthetic pigments (Dubinsky et al. 1986). Thus, we suggest that continued reduction in light per cell followed with the concomitant acceleration of light limitation (e.g. above 10 g.L-') requires a reverse strategy regarding the photosynthetic pigments, that is, decreasing cell contents in both Chl n and c-PC. Thus, the continued decline in Chl c1 and c-PC after optimal culture density Tvas reached seems to reflect light adaptation to very severe light limitation (imposed by the UHCDs clearly evident in the very small growth rate exhibited by such cultures (Figs. 2, 8b). Accordingly, the augmented photosynthetic machinery built in response to a moderate decrease in light availability becomes superfluous ivhen light limitation becomes more extreme, resulting in suppression of synthesis of photosynthetic pigments. T h e very limited amount of energy that the individual cells in a UHCD can harvest is insufficient to support but very minimal growth. Excess energy and carbon are turned to carbohydrate biosynthesis, forming storage energy in support of the growing requirements for maintenance energy (Pirt 1982) needed for survival when growth is greatly retarded due to the extreme light stress.

CONCLUSIONS

Photobioreactors tilted toward the sun for maximal absorbance of solar radiation, with a narrow light path coupled with very intensive stirring rates a r e capable of supporting ultra high cell densities and output rates of cell mass. Elevating the population density arrested high PFD, and low temperature induced photoinhibition. Light damage to PS I1 in ultrahigh density cultures would therefore affect outdoor cultures of microalgae only under extreme conditions of excess lightfor example, a culture of low population density exposed to high PFD as well as to low temperature. T h e increase in cell density was associated with a marked decline in the mass transfer coefficient (K,a), accounting perhaps for the low rates of dark respiration on a per-cell basis, thus giving an additional advantage to UHCD cultures. Significant changes in cell composition were observed with the increase in population density and the concomitant light limitation, representing a photoadaptive strategy. Manipulating cell density represents an optimal mode by which to maximize productivity of valuable chemicals. We thank Ben Freihoff for the technical assistance provided in construction and maintenance of the photobioreactors. T h e assistance of Dr. Hu Zhengyu in analyzing fatty acid composition and of Mrs. Li Zhen in cultivation and chemical analysis of the algae is gratefully acknowledged. Ben-Yaakov, S., Guterman, H . , Vonshak, A . & Richmond, A. 1985. An automatic method for on-line estimation of the photosynthetic rate in open algal ponds. Biotechnol. Bioeng. 28:1136-45. Boussiba, S. & Richmond, A. 1980. C-phycocyanin as a storage protein in the blue-green alga Spirulina platensis. Arch. Mirrobiol. 125:143-7. Brown, M. R., Dunstan, G. A.,Jeffrey, S. W. & LeRoi, J. M. 1993. T h e influence of irradiance on the biochemical comuosition of the prymnesiophyte Isochrjsis sp. (Clone T-ISO).J. Phycol. 29:60 1- 12. Cadenas, E. 1989. Biochemistry of oxygen toxicity. Annu. Rezi. Biochtm. 58:79-110. Chaumont, D. 1993. Biotechnology of algal biomass production: a review of system for outdoor mass culture. J . Appl. Phycol. 5:593-604. Chrismadha, T. & Borowitzka, M. A. 1994. Effect of cell density and irradiance on growth, proximate composition and eicosapentaenoic acid production of Phaeodactjlurn trzcornutum grown in a tubular photobioreactor. J . Appl. Phycol. 6:6774.

PHYSIOLOGICAL CHARACTERISTICS O F SPIRULINA PLATENSIS Cohen, E. & (Malis) Arad, S. 1989. A closed system for outdoor cultivation of Porphyridium. Biomass 18:59-67. Cohen, Z., Norman, H. A. & Heimer, Y . M. 1995. Microalgae as a source of w3 fatty acids. In Simopoulos, A. P. [Ed.] Plants in Human Nutrition. Basel, Karger, 77: 1-31, Cohen, Z., Reungjitchachawali, M., Siangdung, W. & Tanticharoen, M. 1993. Production and partial purification of y-linolenic acid and some pigments from Spirulina platensis. J . Appl. Phjcoi. 5:109-15. Dodd, J. C. 1986. Elements of pond design and construction. In Richmond, A. [Ed.] Handbook ofMicroalga1Mass Culture. CRC Press, Boca Raton, Florida, pp. 265-83. Dubinsky, Z., Falkowski, P. G. & Wyman, K. 1986. Light harvesting and utilization by phytoplankton. Plant Cell Physiol. 27:1335-49. Foyer, C. H., Lelandais, M. & Kunert, K. J. 1994. Photooxidative stress in plants. Phyiol. Plant. 92:696-717. Grobbelaar, J. U . 1995. Influence of areal density on &carotene production by Dunaliella salina. J . Appl. Phycol. 7:69-73. Grobbelaar, J. U. & Soeder, C. J. 1985. Respiration losses in planktonic green algae cultivated in raceway p0nds.J. Plankton lies. 7:497-506. Guterman, H., Ben-Yaakov, S. & Vonshak, A. 1989. Automatic on-line growth estimation method for outdoor algal biomass production. Biotechnol. Bioeng. 34: 143-52. Hu, Q., Guterman, H. & Richmond, A. 1996. A flat inclined modular photobioreactor (FIMP) for outdoor mass cultivation of photoautotrophs. Biotechnol. Bioeng. 5 1:51-60. Kochert, C. 1978. Carbohydrate determination by the phenolsulphuric acid method. In Hellebust, J. A. & Craigie, J. S. [Eds.] Handbook of Phycological Methods: Ph~siotogicaland Biochemical Methods. Cambridge University Press, London, pp. 95-7. Krause, G. H. 1988. Photoinhibition of photosynthesis. An evaluation of damaging and protecting mechanisms. Physiol Plant 74:566-74. 1991. Chlorophyll fluorescence and photosynthesis: the basics. Annu. Rev. Plant Physiol. Plant Mol. Bid. 42:3 13-49. 1994. Photoinhibition induced by low temperatures. In Baker, N . R. & Bowyer, J. R. [Eds.] Photoinhibition oJPhotosjnthesis. Bios Scientific, Oxford, pp. 33 1-48. Lee, Y. K. 1986. Enclosed bioreactor for the mass cultivation of photosynthetic microorganism: the future trend. Trends Biotechnol. 4: 186-9. Lee, Y. K. & Low, C. S. 1991. Effect of photobioreactor inclination on the biomass production of an outdoor algal culture. Biotechnol. Bioeng. 38:995-1000. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. 1951. Protein measurement with folin reagent. J . Bid. Chem. 193:265-75. Ohad, I., Keren, N., Zer, H., Gong, H., Mor, T . S., Gal, A., Tal, S. & Domovich, Y. 1994. Light-induced degradation of the photosystem I1 reaction center D 1 protein in vivo: an integrative approach. In Baker, N. R. & Bowyer, J. R. [Eds.] Photoinhibition of Photosqnthtsis. Bios Scientific, Oxford, pp. 161-73. Oquist, G., Greer, D. H. & Ogren, E. 1987. Light stress at low temperature. I n Kyle, D. J., Osmond, C. B. & Arntzen, C. J. [Eds.] Photoinhibition. Elsevier Science, B. V., Barking, England, pp. 67-82. Pirt, S.J. 1982. Maintenance energy: a general model for energylimited and energy sufficient growth. Arch. Mzcrobiol. 123: 300-2. Pirt, S. J., Lee, Y. K., Walach, M. R., Pirt, M. W., Balyuzi, H. H. M. & Bazin, M. J. 1983. A tubular bioreactor for photosyn-

-

1073

thetic production of biomass from carbon dioxide: design and perf0rmance.J. Chem. Tech. Biotechnol. 33B:35-58. Post, A. F., Dubinsky, Z., Wyman, K. & Falkowski, P. G. 1984. Kinetics of light-intensity adaptation in a marine planktonic diatom. Mar. Biol. (Berl.) 83:231-8. Powles, S. B. 1984. Photoinhibition of photosynthesis induced by visible light. Annu. Rev. Plant Physiol. 35: 15-44. Radmer, R., Behrens, P. & Arnett, K. 1987. Analysis of the productivity of a continuous algal culture system. BiotechnoL Bioeng. 24:488-92. Richmond, A. 1988. A prerequisite for industrial microalgaculture: efficient utilization of solar irradiance. In Stadler, T., Mollion, J., Verdus, M. C., Karamanos, Y . , Morvan, H. & Christiaen, D. [Eds.] Algal Biotechnoloa. Elsevier Applied Science, Essex, pp. 237-44. - 1992. Open systems for the mass production of photoautotrophic microalgae outdoors: physiological princip1es.J. Appl. Phycol. 4:281-6. Richmond, A., Boussiba, S., Vonshak, A. & Kopel, R. 1993. A new tubular reactor for mass production of microalgae outdoors. J . Appl. Phycol. 5:327-32. Richmond, A., Lichtenberg, E., Stahl, B. & Vonshak, A. 1990. Quantitative assessment of the major limitations on productivity of Spirulina platensis in open raceways. J. Appl. Phycol. 2~195-206. Sukenik, A., Bennett, J. & Falkowski, P. 1987a. Light-saturated photosynthesis limitation by electron transport or carbon fixation? Biochim. Biophys. Acta 891:205-15. Sukenik, A., Bennett, J., Mortain-Bertrand, A. & Falkowski, P. G. 1990. Adaptation of the photosynthetic apparatus to irradiance in Dunaliella tertiolecta. A kinetic study. Plant Physiol. 92~891-8. Sukenik, A., Falkowski, P. G. & Bennett, J. 1987b. Potential enhancement of photosynthetic energy conversion in algal mass culture. Biotechnol. Bioeng. 30:970-7. Sukenik, A., Levy, R. S., Levy, Y., Falkowski, P. G . & Dubinsky, 2. 1991. Optimal algal biomass production in an outdoor pond: a simulation model. J . Appl. Phycol. 3: 19 1-20 1. Tamiya, H. 1957. Mass culture of algae. Annu. Rev. Plant Physiol. 8:309-34. Tedesco, M. A. & Duerr, E. 0. 1989. Light, temperature and nitrogen starvation effects on the total lipid and fatty acid content and composition of Spirulina platensis UTEX 1928. J. Appl. Phycol. 1:201-9. Thompson, P. A., Guo, M. X . & Harrison, P. J. 1992. Effects of variation in temperature. I. On the biochemical composition of eight species of marine phytoplankton. J . Phqcol. 28:481-8. Torzillo, G., Sacchi, A., Materassi, R. & Richmond, A. 1991. Effect of temperature on yield and night biomass loss in Spirulina platensis grown outdoors in tubular photobioreactors J. Appl. Phycol. 3:103-9. Tredici, M. R. & Materassi, R. 1992. From open ponds to vertical alveolar panels: the Italian experience in the development of reactors for the mass cultivation of photoautotrophic microorganisms. J. Appl. Phycol. 4:221-31. Vonshak, A., Abeliovich, A., Boussiba, S., Arad, S. & Richmond, A. 1982. Production of Spirulina biomass: Effects of environmental factors and population density. Biomass 2: 175-85. Vonshak, A. & Guy, R. 1992. Photoadaptation, photoinhibition and productivity in the blue-green alga, Spirulina platensis grown outdoors. Plant Cell Environ. 15:613-6. Zarouk, C. 1966. Contribution i l’ttude d’une cyanophycie. Influence de divers facteurs physiques et chimiques sur la croissance et al photosynthise de Spirulina maxima. Ph.D. thesis, Paris, 138 pp.