Growth, stoichiometry and cell size; temperature and nutrient responses in haptophytes

Temperature and nutrients are key factors affecting the growth, cell size, and physiology of marine phytoplankton. In the ocean, temperature and nutrient availability often co-vary because temperature drives vertical stratification, which further controls nutrient upwelling. This makes it difficult to disentangle the effects of temperature and nutrients on phytoplankton purely from observational studies. In this study, we carried out a factorial experiment crossing two temperatures (13°and 19°C) with two growth regimes (P-limited, semi-continuous batch cultures [“−P”] and nutrient replete batch cultures in turbidostat mode [“+P”]) for three species of common marine haptophytes (Emiliania huxleyi, Chrysochromulina rotalis and Prymnesium polylepis) to address the effects of temperature and nutrient limitation on elemental content and stoichiometry (C:N:P), total RNA, cell size, and growth rate. We found that the main gradient in elemental content and RNA largely was related to nutrient regime and the resulting differences in growth rate and degree of P-limitation, and observed reduced cell volume-specific content of P and RNA (but also N and C in most cases) and higher N:P and C:P in the slow growing −P cultures compared to the fast growing +P cultures. P-limited cells also tended to be larger than nutrient replete cells. Contrary to other recent studies, we found lower N:P and C:P ratios at high temperature. Overall, elemental content and RNA increased with temperature, especially in the nutrient replete cultures. Notably, however, temperature had a weaker–and in some cases a negative–effect on elemental content and RNA under P-limitation. This interaction indicates that the effect of temperature on cellular composition may differ between nutrient replete and nutrient limited conditions, where cellular uptake and storage of excess nutrients may overshadow changes in resource allocation among the non-storage fractions of biomass (e.g. P-rich ribosomes and N-rich proteins). Cell size decreased at high temperature, which is in accordance with general observations.

152 dilution growth period due to the periodic dilution with fresh medium. The high P treatment 153 (hereafter referred to as "+P") received standard IMR 1/2 with 12.5 μM phosphate resulting in a 154 dissolved N:P ratio of 10:1. To make sure that we had no nutrient limitation in the +P cultures, 155 these were run as batch cultures in turbidostat mode, and diluted down to the same, low cell 156 density every 2-3 days; 50 000 cells mL -1 for E. huxleyi and C. rotalis and 100 000 cells mL -1 for 157 P. polylepis. This ensured that the cells were P-saturated and always growing exponentially.

158
All experiments were run in 40 mL nunclon filtercap flasks (Thermo Scientific). The

172
For the +P cultures, specific growth rate was calculated as log(N t+Δt /N t )/Δt, where N t was 173 the cell density after dilution, N t+Δt the cell density at the end of the growth interval (before the 174 next dilution), and Δt the time interval between measurements (in days). For the -P cultures, 175 growth rate was calculated as log(N t+Δt /(N t /DF))/Δt, where N t was the cell density before 176 dilution, DF the dilution factor (see above), and N t+Δt the cell density at the end of the growth 177 interval.

178
Since the cultures (and growth rates) remained stable over time, we here report mean Manuscript to be reviewed 213 the unknown samples against the fluorescence of the known RNA standards. We converted the 214 weight of total RNA to moles of ribonucleotide monophosphates (which each contains one P-215 atom) using the average molecular weight of the four different ribonucleotide monophosphates 216 (339.5 g mol -1 ).

217
Alkaline phosphatase activity (APA) may serve as an independent biomarker for P-   282 The corresponding analyses for cell quota of C, N, P, and RNA ( Fig. S2 and . 2 A-C, table 2), we found that cell volume-specific 289 concentrations of P and RNA were lower under P-limitation for all species at both temperatures.
290 The relative differences in P concentration between +P and -P was a factor 5.2 at 13°C and 5 at 291 19°C for E. huxleyi; 1.7 at 13°C and 6.3 at 19° for C. rotalis; and 1.3 at 13°C and 1.8 at 19°C for 292 P. polylepis. For C and N, the effect of P-regime was less consistent, but generally, the 293 concentrations increased or were unaffected by P-regime ( fig. 2, table 2). Due to the strong 294 correlation between cell-volume specific concentrations of C and N (r = 0.96; all data pooled), 295 C:N was not significantly affected by P-regime nor temperature (data not shown). For 296 comparison, the correlations between C and P, and N and P were 0.92 and 0.85, respectively.

297
Cell quotas of C, N, P, and RNA were generally also higher in cultures growing at +P.
298 However, for cell quotas the effect of P-regime was weaker and in more cases non-significant 299 compared to the cell-volume specific concentrations C, N, P and RNA (Fig. S3 & Table S3).
300 This can likely be explained by the overall trend of reduced cell size at +P (Fig. S4, table S4).

301
302 Specific effects of temperature 303 We found significant temperature effects on cell volume-specific concentrations of C, N, P, and 304 RNA, but the effect differed between species, P-regime, and element ( fig. 2; p- Table 2: Estimates and p-values from linear models on the form log 10 (y) = β 0 + β 1 ×P-regime + 468 β 2 ×temperature + β 3 ×P-regime : temperature, where y is either C, N, P, RNA, C:P, or N:P. P-469 regime and temperature are both represented as factor variables with -P and 13°C as reference 470 levels, respectively. Hence, β 1 is the estimated difference (on log 10 scale) in y between +P and -P 471 cultures growing at 13°C, and β 2 the difference between 19°C and 13°C for cultures grown at -P.