HIGH PHOTOSYNTHETIC RATES UNDER A COLIMITATION FOR INORGANIC PHOSPHORUS AND CARBON DIOXIDE¹

P_i is often a growth-limiting factor for the phytoplankton in freshwater lakes (Schindler 1977). Also in very acidic mining lakes, phytoplankton growth is P_i limited during most of the summer stratification period (Spijkerman 2008a). P_i limitation lowers maximum photosynthetic rates (P_max) and maximum quantum yield (Φ_II) as shown from laboratory experiments (Harris and Piccinin 1983, Wykoff et al. 1998, Beardall et al. 2005). The low Φ_II of P_i-limited phytoplankton will increase upon P_i repletion. For example, the electron transport rates of the phytoplankton from two very acidic lakes increased after a 24 h incubation with P_i, suggesting P_i-limited photosynthesis (Beulker et al. 2002).

Inorganic carbon, especially CO₂, is another potential limiting factor of growth for the phytoplankton. CO₂ limitation can occur in phytoplankton blooms and seems especially important in very acidic lakes, where due to low pH, virtually no bicarbonate is present and CO₂ is the only inorganic species (Stumm and Morgan 1970). Concentrations of inorganic carbon in the upper water strata of acidic lakes typically reflect equilibrium concentrations of CO₂ with the air (∼14 μM CO₂). These CO₂ concentrations decreased the maximum growth rate, P_max, and Φ_II of C. acidophila compared with high-CO₂ conditions (Tittel et al. 2005, Spijkerman 2008b). Also in other algal species, P_max increased upon CO₂ addition (e.g., Rost et al. 2003). In the green microalga Dunaliella salina, both growth and photosynthesis were higher under CO₂-enriched conditions (Giordano et al. 1994). The results cited above imply that a single nutrient limitation (i.e., either P_i or CO₂) will decrease the Φ_II of algae compared to nutrient-replete conditions.

In contrast to a single nutrient limitation, it is now often acknowledged that the growth of phytoplankton in many lakes is colimited by two or more nutrients (North et al. 2007 and references therein). Of course, a situation can exist in which different species of algae are limited by different nutrients, but possibly a colimitation also can be expected within one species (Saito et al. 2008). Such a colimitation is defined as a situation at which the concentration of more than one nutrient is below the optimal concentration for uptake and growth (multinutrient colimitation, Arrigo 2005). In a former study, the maximum P_i-uptake rate in C. acidophila was higher under high- than under low-CO₂ conditions, and in concert with other results, high-CO₂-acclimated cells were more strongly P_i limited than low-CO₂-acclimated cells (Spijkerman 2007). Supposedly, the low-CO₂-acclimated algae were colimited by P_i and CO₂. Similar to the cultures described in Spijkerman (2007), C. acidophila was grown in P_i-limited semicontinuous cultures under high- and low-CO₂ conditions, and the P_max, Φ_II, and RUBISCO content were measured. A second high-CO₂ treatment was realized by applying glucose to the medium (osmo-mixotrophic conditions). The presence of (co-)limiting conditions was tested in nutrient-enrichment experiments.

Materials and Methods

Photosynthesis. Photosynthetic O₂-evolution rates were obtained in a light dispensation system (Topgallant LLC, Salt Lake City, UT, USA). Cultures were concentrated by centrifugation (1,500g, 5 min) to an OD at 750 nm of ∼0.2 and subsequently dark adapted for 30 min at 20°C. After that, oxygen evolution was measured over a range of 11 actinic light intensities (I, 0–1,500 μmol · m⁻² · s⁻¹) using the light dispensation system. From the concentrated culture suspension, samples for cell enumeration were taken. Rates of oxygen evolution were corrected for cell density (P) and fitted to the equation:

(1)

where P_max is the maximum gross oxygen production, R_d is the respiration rate in the dark, α is the initial slope of the photosynthesis-irradiance (P-I) curve, and β is the slope at high irradiances that describes photoinhibition. The threshold irradiance at which net photosynthesis equals zero (I_c) was calculated by dividing R_d by α. P_max was recalculated on C_p, as cells were largely different in size in the various culture conditions. In addition, P_max was related to the Q_p by the following model:

(2)

The model is based on the Droop model, but instead of growth rate, P_max is the dependent variable. In the model, P_max′ is the estimated maximum P_max, and Q₀ is the minimum cellular P quota for photosynthesis.

The fluorescence of PSII was measured every 30 s over the first 5 min of every (P-I) curve, when the sample was still in the dark, using a pulse-amplitude-modulated (PAM) chl fluorometer (PAM101/103; Walz, Effeltrich, Germany). Saturating light pulses (600 ms, ∼2,500 μmol photons · m⁻² · s⁻¹) were provided to measure maximal (F_m) and minimal (F₀) fluorescence. The maximum quantum yield (Φ_II) was then calculated using eq. 3.

(3)

Part of the sample used for photosynthesis measurements was filtered over a 2.0 μm Nucleopore track-Etch membrane filter (Whatman). This filtrate contained the same density of bacteria, but no algae, and was measured in the light dispensation system to determine the bacterial respiration rate.

Results

Nutrient-enrichment experiments revealed that P_i was the main limiting factor for growth in all cultures, as a P_i addition largely stimulated growth rates (Fig. 1). With increasing dilution rate, enrichment cultures receiving only deionized water (control) had increasing growth rates, probably as a result of the increasing concentration of P_i left in the medium. Therefore, the growth rate in the nutrient enrichments was divided by that in the control for each dilution rate. Growth rates in the cultures enriched with CO₂ or NH₄⁺ were similar to those in the control (i.e., the quotient did not differ from 1), whereas the addition of P_i or both CO₂ and P_i resulted in 4- to 5-fold higher growth rates than those in the control cultures (ANCOVA, df = 3, 53, F = 101, P < 0.001; Fig. 1). A paired sample t-test revealed that relative growth rates in nonaerated cultures were higher in enrichment cultures with both CO₂ and P_i (t = 3.05, df = 4, P < 0.05) than in those receiving only P_i. In CO₂-aerated cultures, this effect was marginally significant (t = 2.39, df = 4, P = 0.076), and in osmo-mixotrophic cultures, there was no significant difference between these two treatments (t = 1.09, P = 0.336).

A former study had shown that the maximum P_i-uptake rate was ∼60-fold higher in the P_i-limited rather than P_i-saturated cultures (Spijkerman 2007). In addition, the C source had a significant effect on the maximum P_i-uptake rate: this rate was higher in the CO₂-aerated than in the nonaerated or osmo-mixotrophic cultures. Here, I show that the C source (high or low CO₂ and the presence or absence of glucose) had a significant effect on the maximum quantum yield (Φ_II) and the maximum photosynthetic rate (P_max) when the effect of dilution rate was accounted for (ANCOVA, df = 2, 42, F ≥ 6.3, P < 0.01 for both; Fig. 2). Remarkably, the highest Φ_II and P_max were obtained in the low-CO₂ cultures (-CO₂ conditions), whereas lower values were measured in high-CO₂ and osmo-mixotrophic cultures, especially at dilution rates <0.6 · d⁻¹ (Fig. 2). The Φ_II and P_max in high-CO₂ and osmo-mixotrophic cultures were similar.

Values of both Φ_II and P_max increased with increasing dilution rate (ANCOVA, df = 5, 39, F > 6.4, P < 0.001 for both; Fig. 2). Φ_II increased from 0.26 to 0.67 with increasing dilution rate, and P_max increased from 1.1 to 9.3 mmol O₂ · g C⁻¹ · h⁻¹. At higher dilution rates, the cellular P quota (Q_p) was also higher than at lower ones (Spijkerman 2007), yielding a positive relationship between Φ_II and P_max with Q_p. All values of P_max were related to Q_p using eq. 2 (Fig. 3), the maximum P_max (P_max′) was estimated as 9.6 mmol O₂ · g C⁻¹ · h⁻¹, and the minimum cellular phosphorus quota, Q₀, as 0.68 mmol P_p:mol C_p (r² = 0.65). The results presented in 2, 3 clearly show that with increasing strength of a P_i limitation (i.e., a lower dilution rate or a lower Q_p), C. acidophila decreased its photosynthetic capacity.

The relative contribution of the dark respiration (R_d) to P_max decreased with increasing dilution rate when the effect of the C source was accounted for (ANCOVA, df = 5, 40; F = 3.2; P < 0.05; Table 1). This finding means that the stronger P_i-limited algae (grown at low dilution rates) had a higher relative R_d (up to 41% of P_max) when compared with 17.5% in the algae growing at higher dilution rates. This was solely a result of a decrease in the P_max, as R_d did not change over dilution rate (results not shown). There was no effect of the C source on the value of R_d:P_max (ANCOVA, df = 2, 43, F = 2.3, P = 0.11): the dark respiration in osmo-mixotrophic cultures was the same as in low- and high-CO₂ cultures. The effect of bacterial contamination on the measured respiration rates was tested by measuring the O₂ evolution in the filtrate of the sample used for photosynthesis measurements (through a membrane with 2.0 μm pore size) and enumerating the bacteria. The percentage of bacterial R_d to total measured R_d was on average (±SD, n = 5) 21 ± 16, 42 ± 14, and 33 ± 5, for low-CO₂, high-CO₂, and osmo-mixotrophic cultures, respectively. Both observations on R_d agree with the fact that the bacterial densities were the same in osmo-mixotrophic and high- or low-CO₂ cultures (results not shown) and ranged from 7.3 × 10⁷ and 3.6 × 10⁸ bacteria · L⁻¹, mainly varying over the dilution rate (i.e., bacterial densities were higher in cultures with a low dilution rate). The bacterial respiration rates were not correlated with the bacterial density (results not shown; Pearson, n = 15, r = 0.06, P = 0.83). Therefore, bacterial contamination did not have a significant impact on the R_d, which was further supported by the observation that bacterial densities increased with decreasing dilution rate, whereas R_d remained constant.

The compensation light intensity (I_c, which is R_d:α) remained the same over dilution rate when the effect of the C source was accounted for (ANCOVA, df = 5, 40, F = 0.17, P = 0.97; Table 1). The results thereby show that independent of the P_i limitation, algae required the same light intensity for photosynthesis to compensate for R_d and were likely not light limited in their growth. In addition, I_c was the same in osmo-mixtrophic as in high- or low-CO₂ cultures (ANCOVA, df = 2, 43, F = 0.37, P = 0.69). Independent from the C source applied, a more stringent P_i limitation resulted in higher values of R_d:P_max and no differences in I_c.

Differences in photosynthetic rates may relate to differences in RUBISCO content, motivating the measurement of this parameter. The RUBISCO content varied by a factor of three in C. acidophila, depending on the growth rate and C source (Fig. 4). Under low-CO₂ conditions, C. acidophila had the highest cellular content of RUBISCO, which was on average 1.6 times higher than under high-CO₂ and osmo-mixotrophic conditions (ANCOVA, df = 2, 89, F = 22.4, P < 0.001). The RUBISCO content under high-CO₂ and osmo-mixotrophic conditions was similar (Bonferroni, P = 0.95). There was no effect of dilution rate on the RUBISCO content when the effect of C source was accounted for (ANCOVA, df = 5, 86, F = 0.51, P = 0.77; Fig. 4). Therefore, the RUBISCO content was constant over dilution rate (and hence constant over the strength of P_i limitation) in all cultures and positively correlated to the P_max (compare Fig. 2b with Fig. 4; Pearson correlation, n = 46, r = 0.309, P < 0.05).

Discussion

The photosynthetic characteristics of C. acidophila depended on its nutrient limitation. In general, a more stringent P_i limitation resulted in a lower Φ_II and P_max, and higher values of R_d:P_max. Colimiting conditions of P_i and CO₂ were most pronounced in low-CO₂-acclimated cultures that had the highest Φ_II, P_max, and RUBISCO content. A colimitation for P_i and CO₂ enhanced the mismatch between C-fixation capacity and growth rate.

In C. acidophila, Φ_II and P_max positively correlated with each other, and lower values were measured under more stringent P_i-limited conditions. The lower Φ_II with a more stringent nutrient limitation coincided with observations on other algal species where single-nutrient-limiting conditions resulted in a decreased Φ_II (LaRoche et al. 1993, Graziano et al. 1996, Geider et al. 1998). Comparing Φ_II in P_i-replete and P_i-limited conditions in C. acidophila, the highest Φ_II from the P_i-limited low-CO₂ cultures (0.66) was higher than that of P_i-replete cultures (i.e., 0.62; Spijkerman 2008b). In contrast, the highest Φ_II in P_i-limited high-CO₂ cultures (0.67) was the same as that measured in P_i-replete cultures (i.e., 0.67; Spijkerman 2008b). In C. acidophila, carbon-specific P_max under maximum P_i limitation amounted to ∼20% of the P_max at the least P_i-limited conditions. This decrease is rather large compared with Chlamydomonas reinhardtii, for example, where P_max decreased 2-fold over a decrease in dilution rate from 2.0 to 0.25 · d⁻¹ (Harris and Piccinin 1983). Also in other algal species, photosynthetic rates measured by O₂ evolution often decrease under nutrient-limiting conditions (Wykoff et al. 1998, Beardall et al. 2005, Young and Beardall 2005), although in some cases the effect was minor (Geider et al. 1998) or even contrasting: P_max was higher in P_i-limited than in P_i-replete Chlorella vulgaris (Kozlowska-Szerenos et al. 2004). Kozlowska-Szerenos and colleagues explain this as similar to a sun-type (i.e., high light) acclimation response to insufficient phosphate feeding, but as they describe only a minor decrease in the chl content of the algae with increasing P_i limitation, their study rather shows a complex response in the photosynthesis of cells exposed to a transition from P_i-replete to P_i-deplete conditions.

The R_d:P_max of C. acidophila was similar (ranging between 10% and 50%) to that of C. reinhardtii (ranging between 28% and 43%; Harris and Piccinin 1983). In contrast to the study of Harris and Piccinin, who detected no change in the R_d:P_max over dilution rate, the R_d:P_max of C. acidophila was the highest in the cultures with the lowest dilution rates (i.e., more stringent P_i-limited cells), and values decreased with increasing dilution rates. This was solely a result of changes in the P_max, as R_d did not change over dilution rate. Possibly, in C. acidophila, R_d is rather high as to produce additional ATP necessary for energizing proton efflux and maintaining a neutral internal pH (Gerloff-Elias et al. 2005, 2006). The rather high P_max compensates for this high R_d (Gerloff-Elias et al. 2005), and as the external pH remains low, only changes in P_max can be realized in this algal species.

In contrast to an expected enhanced R_d under osmo-mixotrophic P_i-limited conditions, no such effect was observed here. There is a tendency of R_d:P_max, to be higher in the high-CO₂ and osmo-mixotrophic than in the low-CO₂ cultures, which coincides with the reported increased R_d values for plants grown under high CO₂ (Leakey et al. 2009). Effects of intracellular high-CO₂ concentrations might therefore cover a possible effect of enhanced glucose concentrations on photosynthesis under P_i-limited conditions.

In many physiological characteristics (Φ_II, P_max, R_d:P_max and RUBISCO content), osmo-mixotrophic and high-CO₂-acclimated cells of C. acidophila showed a similar response. Only in the nutrient-enrichment experiment was a different response measured as the osmo-mixotrophic algae realized the same growth rate when enriched with only P_i or with both P_i and CO₂ (Fig. 1). These algae were therefore not colimited in their growth for P_i and CO₂. Both the low- and high-CO₂-acclimated algae grew faster when cultures were enriched with both P_i and CO₂ than when enriched with P_i only, suggesting a colimitation in growth. It is, however, possible that the growth response to both nutrients did not occur simultaneously, but chronologically. From the data presented here, this finding cannot be distinguished as the biomass yield was measured at only one point in time (after 4 d). Therefore, daily monitoring of enrichment experiments, similar to those performed by North and colleagues (North et al. 2007), will be necessary to elucidate the growth response.

P _max related to Q_p in a nonlinear manner to which a saturation model was fitted. From this curve, the minimum cell quota (Q₀) for photosynthesis was estimated as 0.68 mmol P_p:mol C_p. This value is in the lower range of the Q₀ estimated via a normal Droop relation where growth is the dependent parameter (i.e., 0.59–1.14 mmol P_p:mol C_p; Spijkerman 2007) and provides the possibility to estimate Q₀ via photosynthesis measurements. This fact allows the estimation of Q₀ in transiently growing P_i-limited cultures that are not in a steady state, but where Q_p and photosynthesis can be determined.

In C. acidophila, high-CO₂ levels enabled enhanced Φ_II and P_max under P_i-saturating conditions (Spijkerman 2008b), and also in other algal species, P_max increased upon CO₂ addition (e.g., Rost et al. 2003). For example, in Dunaliella salina, photosynthesis was higher under high-CO₂ than low-CO₂ conditions (Giordano et al. 1994). Under P_i and CO₂ colimiting conditions, there was a contrasting effect (i.e., the Φ_II and P_max were lower at high-CO₂ than at low-CO₂ conditions; Fig. 2). The enhanced Φ_II and P_max in the P_i and CO₂ colimited algae reflect an enhanced “uptake capacity for CO₂” following the physiological response to CO₂-limiting conditions. In contrast, the high-CO₂-acclimated and osmo-mixotrophic algae were predominantly P_i limited, which resulted in a lower P_max. As a consequence, algae colimited by P_i and CO₂ will have the physiological properties resulting from a P_i limitation but also from a CO₂ limitation. Present studies support this hypothesis as the maximum uptake rates of P_i and CO₂ were both increased under colimiting conditions (E. Spijkerman, unpublished data).

In addition, Harris and Piccinin (1983) showed complex interactions between P_i limitation, growth rate, and C metabolism, and their figure 2 shows a higher calculated growth rate from C-uptake data than the actual dilution rate in P_i-limited C. reinhardtii continuous cultures, especially at the lower dilution rates. In their study, a maximal 2-fold overcapacity for CO₂ fixation compared with the dilution rate was observed. For C. acidophila, growth rates can be calculated from net photosynthetic rates, as at in situ light intensities, P_max is realized. Calculated growth rates ranged between 0.15 and 0.89 · d⁻¹, whereas dilution rates ranged between 0.1 and 0.8 · d⁻¹, suggesting an overproduction of organic substances under most conditions tested. This overproduction is reflected in enhanced concentrations of dissolved organic carbon (DOC) in the culture medium (Spijkerman 2007). Relative to the dilution rate, growth rates calculated from P_max were 1.0 to 1.6 times higher in high-CO₂ and osmo-mixotrophic algae, being in the range of that determined for C. reinhardtii under P_i-limited conditions (Harris and Piccinin 1983). In low-CO₂-acclimated cells, the overcapacity of photosynthesis to growth rate in C. acidophila was higher, ranging between 1.4 and 4.8, suggesting an increasing mismatch between photosynthesis and growth at P_i and CO₂ colimiting conditions than under a single P_i limitation. Interestingly, this enhancement did not result in higher DOC concentrations in low- than in high-CO₂ conditions. At present, it is not clear how to explain this discrepancy other than that the low-CO₂-acclimated algae released more CO₂ or that the photosynthesis measurements overestimated in situ rates. The synthesis of a (larger) mucous sheath under low-CO₂ conditions seems unlikely as the cellular carbon content was comparable and cell size smaller than under high-CO₂ conditions (Spijkerman 2007).

The cellular RUBISCO content was on average 1.6-fold higher in the low-CO₂ cells than in the high-CO₂ cells (Fig. 3). This extent confirms previous findings where under P_i-replete conditions C. acidophila also regulated its cellular RUBISCO content in response to CO₂ concentrations (Spijkerman 2008b): at low-CO₂ conditions, C. acidophila had ∼1.5 times the amount of RUBISCO protein than under high-CO₂ conditions. A positive correlation between cellular RUBISCO content and P_max is often described in algae (Raven 1991) and is confirmed here in C. acidophila under P_i-limited conditions. Also, in Dunaliella tertiolecta, changes in P_max caused by nutrient-limiting conditions could be accounted for by changes in the cellular RUBISCO content (Geider et al. 1998). In contrast, under P_i-replete conditions, cellular RUBISCO content and P_max were negatively correlated in C. acidophila (Spijkerman 2008b). Algae having a carbon concentrating mechanism (CCM) tend to increase the RUBISCO content under increased CO₂ conditions (Raven 1991). Although C. acidophila has a CCM (Spijkerman 2005), the RUBISCO content was higher under low- than under high-CO₂ conditions both under P_i-replete and P_i-deplete conditions. Its RUBISCO content therefore correlates with CO₂ condition and not with P_max.

In conclusion, this study has demonstrated that under P_i-limited conditions both Φ_II and P_max decreased. By performing nutrient-enrichment experiments, a colimitation for P_i and CO₂ was detected in nonaerated algae that had an increased Φ_II, P_max, and RUBISCO content compared to CO₂-aerated (marginally colimited) and osmo-mixotrophic cultures (not colimited). The effect of a P_i limitation on photosynthesis is important to understand primary productivity measurements in lakes, as P_i is often considered a growth-limiting nutrient. In theory, light intensities during an extended period of the day and over several meters of depth would support P_max. However, in-lake measurements show that P_max often exceeds ¹⁴C-fixation rates (e.g., Carignan et al. 2000), which is likely the result of P_i-limited conditions. With the recent acknowledgment that phytoplankton are often colimited by more than one nutrient, studies on the effect of a colimitation on photosynthesis and primary productivity are required—not only for understanding processes, but also for ecosystem modeling, as net primary productivity explains a significant part of food-web properties in ecosystem modeling (Vermaat et al. 2009). The results of this study show that phytoplankton colimited by P_i and CO₂ will have misleadingly high photosynthetic rates, which could result in the false conclusion that the plankton are not nutrient limited.