Trade-off between reproduction and lifespan of the rotifer Brachionus plicatilis under different food conditions

Phaeocystis globosa, one of the most typical red tide-forming species, is usually mixed in the food composition of rotifers. To explore how rotifers respond by adjusting life history strategy when feeding on different quality foods, we exposed the rotifer Brachionus plicatilis to cultures with 100% Chlorella, a mixture of 50% P. globosa and 50% Chlorella, or 100% P. globosa. Results showed that rotifers exposed to 100% Chlorella or to mixed diets produced more total offspring and had higher age-specific fecundity than those exposed to 100% P. globosa. Food combination significantly affected the net reproduction rates of rotifers. By contrast, rotifers that fed on 100% P. globosa or on mixed diets had a longer lifespan than those fed on 100% Chlorella. The overall performances (combining reproduction and lifespan together) of rotifers cultured in 100% Chlorella or mixed diets were significantly higher than those cultured in 100% P. globosa. In general, Chlorella favors rotifers reproduction at the cost of shorter lifespan, whereas P. globosa tends to extend the lifespan of rotifers with lower fecundity, indicating that trade-off exists between reproduction and lifespan under different food conditions. The present study also suggests that rotifers may have the potential to control harmful P. globosa.

food for most zooplankton species 15 . However, during the early stage of red tide development, some of the zooplankton species, such as the copepods Temora longicornis and Acartia clause, could have high abundances 16 , indicating that copepods can survive well when fed with P. globosa. Some protozoa, such as ciliates could well graze the solitary cells of P. globosa 17,18 .
To our knowledge, no report exists about the possible trade-off between reproduction and lifespan in the rotifers that fed on P. globosa to date. Considering that rotifer usually co-exists with P. globosa in nature and some zooplanktons survive on P. globosa, we hypothesized that (1) the rotifer Brachionus plicatilis can survive but reproduction may be affected when feeding on P. globosa, (2) trade-off exists between reproduction and lifespan of B. plicatilis under different food conditions, and (3) B. plicatilis may have the potential to control the harmful P. globosa. To test the hypotheses, we used both P. globosa (generally not a good food for zooplankton) and Chlorella sp. (generally a good food for rotifers) as experimental food to culture the widespread rotifer B. plicatilis under three different food combinations with the same dry weight, namely, 100% Chlorella sp. (100%C), a mixture of 50% P. globosa and 50% Chlorella sp. (50%C + 50%P), and 100% P. globosa (100%P), and recorded the life-history parameters. Results supported the hypotheses and indicated that B. plicatilis may have a potential to control the harmful P. globosa.

Results
Survival rate and lifespan. No significant difference was found in the three treatments before 150 h. After 150 h, the survival rate of rotifers in 100% C decreased sharply and was significantly lower than those in the other treatments ( Fig. 1). However, no significant effect was observed between the rotifers fed with P. globosa and mixed food (P = 0.152).
Reproduction. The age-specific fecundity in all treatments increased with time, and then decreased after a plateau was reached (Fig. 3a). The accumulated total offspring of rotifers cultured with 100% P (maximum value: 7.33) significantly decreased by approximately 50% compared with those cultured in 100% C (maximum value: 14.44) and 50% C + 50% P (maximum value: 13.50) (P < 0.001), but no significant difference was observed between the rotifers cultured in 50% C + 50% P and 100% C (P = 0.091) (Fig. 3b).
The net reproduction rate (R 0 ), generation time (T), and intrinsic rate of population increase (r m ) are shown in Table 1. The R 0 of rotifers in 100% P was lower than that of the rotifers in 50% C + 50% P (P = 0.016) or 100% C (P = 0.009), but no significant difference was observed in R 0 of rotifers cultured with 50% C + 50% P and 100% C (P = 0.870). Similarly, the r m of rotifers fed on 100% P was lower compared with that of the rotifers cultured with 50% C + 50% P (P = 0.019) or 100% C (P = 0.005), but no significant difference was observed between the rotifers in 50% C + 50% P and 100% C (P = 0.473). The generation time of the three food treatments had no significant difference (P = 0.622).
Overall performance. As for the overall performance, the rotifers cultured with 100% P was significantly lower than those in other two treatments (P < 0.001), but no significant difference was observed between the rotifers fed on 50% C + 50% P and 100% C (P = 0.082) (Fig. 4).

Discussion
The main results of the current study showed that Chlorella sp. favored rotifer reproduction with shorter lifespan, whereas P. globosa tended to extend lifespan with lower reproduction, which demonstrated that trade-off exists between the reproduction and lifespan of the rotifer B. plicatilis under different food conditions. However, the overall performance of the rotifers completely cultured with P. globosa was significantly lower than rotifers fed on Chlorella sp. or mixed foods.
Changes in reproduction and lifespan of rotifer under different foods in our experiments were comparable to the study by Yoshinaga et al. 3 , which suggested that the rotifer B. plicatilis would have a longer lifespan with the simultaneously decreased reproduction rate when they were badly fed or starved. Energy allocation models provide a very useful framework to investigate the changes in rotifer life history 19,20 . As all energy used to new growth and reproductive effort come from the ingested food, rotifers should achieve a two-way trade-off when nutrition is limited 21 , in which the lifespan and reproduction were clearly negatively related 22 . Our result indicated that when the reproduction rate was high, the mean lifespan was short, confirming the trade-off between reproduction and lifespan when rotifers were fed on different types of food, even under sufficient food conditions. As this trade-off generally occurs when rotifers confront stressful environments 23 , P. globosa was not as good food as the Chlorella to B. plicatilis.
As food for zooplankton, the quality of algae depends on the nutrition, toxicity, and morph 24 . In the present study, 100% Chlorella or mixed alga caused the rotifers to produce more offspring. Conversely, 100% P. globosa caused the rotifers to have a lower reproduction rate. Algal size influences the filtration and ingestion rate of zooplankton. Rotifers feeding with large algae generally had a better reproduction than that feeding with small algae 25 . However, in the present study, P. globosa is larger than the Chlorella sp. (7.2 µm vs. 4.8 µm in diameter), indicating other characters of P. globosa rather than the algal size affecting the rotifers. P. globosa lacks unsaturated fatty acid, which is necessary for the growth and reproduction of rotifers [26][27][28] . In addition, Phaeocystis sp. were found to be toxic toward Artemia salina, Epinephelus akaara fish 13 , and Gadus morhua larvae 14 . Thus, the poor quality and toxic property of P. globosa may have contributed to the inhibited reproduction of rotifers 5,29 . When the rotifers were fed on 50% Chlorella + 50% P. globosa, they had longer lifespan compared with those fed on 100% Chlorella and higher accumulated offspring compared with those fed on 100% P. globosa. The overall performance of rotifers under 50% Chlorella + 50% P. globosa was nearly similar to those feeding on 100% Chlorella, indicating that the energy that these rotifers obtained is comparable under the two food conditions. The rotifers under this food combination may partly obtain both the advantages from the two species of algae. For example, some algal extracts and antioxidants in P. globosa could extend the lifespan of rotifers 30,31 . From the above results, it was concluded that rotifers may have excellent potential to control P. globosa, especially under mixed algal community.
In summary, food combination significantly affected the reproduction and lifespan of rotifers. Chlorella favors rotifers reproduction at the cost of shorter lifespan, whereas P. globosa tends to extend the lifespan of rotifers with lower fecundity, indicating that trade-off exists between reproduction and lifespan under different food conditions. This study also suggests that rotifers may have the potential to control the harmful P. globosa.    Experimental design. Three treatments were arranged, and each treatment contained six neonates (<8 h old). The three different food treatments comprised 100% C (3.0 × 10 6 cells mL −1 ), 50% C + 50% P (1.5 × 10 6 cells mL −1 + 2.75 × 10 6 cells mL −1 ), or 100% P (5.5 × 10 6 cells mL −1 ) with a same dry weight (0.22 mg mL −1 ). Each neonate in every treatment was raised in 1 mL of autoclaved seawater. The experiment was conducted in 24-welled culture plates. Every 12 h, the living rotifers and the new-born neonates were counted, and the dead maternal individuals and new-born neonates were removed. Every 24 h, the maternal rotifers were transferred into fresh seawater with the corresponding foods. The experiments were run in triplicates and under similar environmental conditions as described above. The experiments proceeded until all individuals of each cohort died. All 24-welled culture plates were shaken every 12 h to avoid the sedimentation of algae. We calculated the net reproduction rate (R 0 ), generation time (T), intrinsic rate of population increase (r m ), and mean lifespan (L) according to Krebs 32 .
To assess the total energy output of rotifers that were cultured in different food combinations, we adopted an index, overall performance (P RL ), combining reproduction and lifespan together, based on the fitness formula. The demography parameters were calculated using the formula: RL where x is the culture time, l x is age-specific survival, m x is the age-specific fecundity, and N is the total offspring per rotifer.
Statistical analyses. Data were presented as the mean values ± standard error. The effect of food combination on mean lifespan, the accumulated offspring per rotifer, the median lethal time, R 0 , t, R m , and P RL were analyzed by one-way ANOVA. All statistical analyses were performed in Sigmaplot 11.0.