Abstract

We studied autotrophic and heterotrophic C metabolism during mixotrophic growth of Karlodinium micrum (Leadbeter et Dodge) Larsen (Dinophyceae) on prey Storeatula major (Cryptophyceae). Our goal was to determine the balance of autotrophy and heterotrophy that supports mixotrophic growth in K. micrum. Assimilation of inorganic 14C and 14C-labeled prey was used to separate the quantity and quality (i.e., lipid, polysaccharide and protein) of C obtained by autotrophy and heterotrophy, respectively. Growth rates (µ) of mixotrophic K. micrum were 0.52–0.75 div.·day−1, equal to or greater than the maximum autotrophic growth rate (0.55 div.·day−1) of K. micrum. Autotrophy represented 27–69% of gross C uptake during mixotrophic growth. Cellular photosynthetic performance (PPcell, pg C cell−1·day−1) was 24–52% lower during mixotrophic growth than during autotrophic growth of K. micrum. Mixotrophic K. micrum assimilated 16% less photosynthate as protein compared to autotrophic K. micrum, while protein was the major net assimilation product (52%) from ingested prey C. Growth efficiency (%GE) of mixotrophic cultures, based on both autotrophic and heterotrophic C sources, averaged 36 ± 2.9%, slightly lower than the 40–50% GE typical of purely autotrophic K. micrum, but higher C gains associated with heterotrophic feeding more than compensated for the decrease in %GE in mixotrophic K. micrum. We conclude that mixotrophic growth of K. micrum is dominated by heterotrophic metabolism, although photosynthesis continues at a lowered rate. This is consistent with a shift toward secondary production in plankton assemblages dominated by mixotrophically growing K. micrum.

Received October 21, 2005; accepted in principle February 22, 2006; accepted for publication April 22, 2006; published online May 3, 2006
 Communicating editor: K.J. Flynn

INTRODUCTION

Planktonic protists exhibit a variety of nutritional strategies, including strict autotrophy and strict heterotrophy, and strategies that combine these modes that are collectively referred to as mixotrophy (Jones, 1994). Mixotrophic organisms have been described in freshwater, estuarine and marine ecosystems where they play an important role in plankton dynamics (Stickney et al., 2000; Tittel et al., 2003). A broad range of mixotrophic behaviors have been described for dinoflagellates (Schnepf and Elbrächter, 1992; Stoecker, 1999), but the physiological bases of mixotrophic growth are speculative for most species (Hansen et al., 2000; Skovgaard et al., 2000). Here, we examined the balance of autotrophy and heterotrophy that supports mixotrophic growth of the dinoflagellate, Karlodinium micrum.

Stoecker (Stoecker, 1998) categorized mixotrophic nutrition using three conceptual models based on the balance of autotrophy and heterotrophy, functional responses of grazing to light and nutrients and possible benefits of mixotrophic nutrition to the organisms. An ‘ideal’ mixotroph (Model I) is equally adept as an autotroph or a heterotroph; phagotrophic ‘algae’ (Model II) are primarily photosynthetic but ingest prey under certain conditions; photosynthetic ‘protozoa’ (Model III) are primarily phagotrophic but photosynthesize in certain conditions using either their own plastids or ‘kleptochloroplasts’. Karlodinium micrum (syn. Gyrodinium galatheanum) best fits Model II (Stoecker, 1998) in a group with other photosynthetic dinoflagellates, including Prorocentrum minimum (Stoecker et al., 1997), Heterocapsa triquetra (Legrand et al., 1998) and Ceratium furca (Bockstahler and Coats, 1993a, 1993b; Smalley and Coats, 2002). A major unresolved issue for mixotrophs categorized as Model II is the physiological role of feeding. Stoecker (Stoecker, 1998) suggested three categories: (i) feeding as a means to obtain C, (ii) feeding as a source of inorganic nutrients (N and P) and (iii) feeding to supply other growth factors. Experimental determinations of these physiological roles of feeding in Model II mixotrophs are lacking.

Karlodinium micrum is a nonthecate, photosynthetic dinoflagellate capable of ingesting a variety of small prey by phagocytosis (Li et al., 1996, 1999). Mixotrophic growth of K. micrum on prey S. major (Cryptophyceae) yields growth rates that are 2- to 3-fold higher than the maximum autotrophic growth rate, although growth and feeding do not occur in the dark (Li et al., 1999). Feeding is stimulated either by inorganic nutrient limitation or by high light (Li et al., 2000a). These observations suggest that phagotrophy supplies nutrients to support autotrophic growth, although feeding in nutrient-saturated cultures indicates that phagotrophy may serve other purposes, such as the acquisition of C or essential micronutrients (Li et al., 2000a). We know that the maximum growth rate of mixotrophic K. micrum exceeds that of autotrophic K. micrum (Li et al., 1999), but the balance of autotrophy and heterotrophy during mixotrophic growth remains unknown.

The approach we used to determine the balance of autotrophy and heterotrophy in mixotrophic K. micrum was based on the following simplified expression of unicellular growth:

 

(1)
\[\mathrm{Growth\ rate}=\frac{\begin{array}{l}\mathrm{[}\mathrm{autotrophy}\mathrm{]}\\\mathrm{(}\mathrm{PP}^{\mathrm{cell}}\ {\times}\ \mathrm{\%}\mathrm{GE}\mathrm{)}+\end{array}\begin{array}{l}\mathrm{[}\mathrm{hetroprophy}\mathrm{]}\\\mathrm{(}\mathrm{PP}^{\mathrm{cell}}\ {\times}\ \mathrm{\%}\mathrm{GE}\mathrm{)}\end{array}}{\mathrm{C\ cell}^{{-}1}}\]

where autotrophy is described as the product of photosynthetic performance (PPcell, pg C·cell−1·day−1) and autotrophic gross growth efficiency (%GE), describing the percentage of fixed C that is incorporated into new biomass. Heterotrophy is described as the product of C ingested per predator [cellular heterotrophic performance (HPcell), pg C·cell−1·day−1] and the percentage of ingested C that is incorporated into new biomass (%GE). Adolf et al. (Adolf et al., 2003) showed that autotrophic growth of K. micrum consisted of relatively low rates of C-specific light absorption leading to low PPcell and low %GE typical of dinoflagellates (Falkowski et al., 1985; Tang, 1996). The basis of low growth rates in heterotrophic dinoflagellates (compared to ciliates) varies among species examined (Hansen, 1992; Strom and Morello, 1998). Framing our experiments within the context of equation (1) allows us to explicitly address issues of C acquisition versus growth efficiency in comparing autotrophic and mixotrophic growth of K. micrum.

Strict interpretation of equation (1) requires that the quality (i.e., % protein) of autotrophic and heterotrophic C assimilated by mixotrophic K. micrum be the same. If the quality of autotrophic and heterotrophic C differs, then the efficiency at which each C source contributes to growth may differ. In autotrophic organisms, for instance, accumulation of alternative photosynthetic end products, that is, lipid and polysaccharides, rather than protein that is essential for cellular growth (Morris, 1981), will result in lower growth rates for equivalent amounts of C fixation. Thus, evaluation of the qualitative and quantitative nature of C assimilated by the cell is important to address the balance of autotrophy and heterotrophy during mixotrophic growth (Putt, 1990).

The recent recognition that mixotrophic nutrition is common among dinoflagellates is important to biological oceanographers working in coastal areas, where dinoflagellates play important ecological roles both as primary producers (Kirk, 1994) and as causative agents of harmful algal blooms (HABs) (Hallegraeff, 1993). Heterotrophic and mixotrophic dinoflagellates can be a significant source of secondary production, and their grazing can impact the biomass of primary producers (Hansen, 1992; Bockstahler and Coats, 1993a, 1993b; Li et al., 1996, 2000b; Stoecker et al., 1997; Smalley and Coats, 2002; Johnson et al., 2003; Jeong et al., 2005). Mixotrophic nutritional strategies observed in autotrophic (plastid-bearing) and heterotrophic (nonplastidic) dinoflagellates raise the important ecological question, What is the trophic balance of carbon (C) flow through these populations? Here we address this question by examining the physiological bases of mixotrophic growth in K. micrum with goals to (i) quantify the effect of feeding on cellular photosynthetic rates, (ii) determine the major assimilation end products of photosynthesis during autotrophic and mixotrophic growth and the end products assimilated from ingested prey, and (iii) compare growth efficiency during autotrophic and mixotrophic growth.

METHODS

Algal cultures, biochemical analyses

Karlodinium micrum (strain GE, 10–15 µm) (Dinophyceae, CCMP 1974) was isolated as a clonal culture from the mesohaline portion of Chesapeake Bay in 1995 by Dr. Aishao Li (Li et al., 1996). A recent study by Bergholtz et al. (Bergholtz et al., 2005) has suggested that K. micrum is synonymous with Gymnodinium veneficum Ballantine (strain Plymouth 101), isolated in 1950 by M. Parke, and that both strains should be renamed Karlodinium veneficum. We will continue to use the name K. micrum here to be consistent with previous work on mixotrophic nutrition in this species. Storeatula major (strain g, 4–9 µm) (Cryptophyceae) was isolated from Chesapeake Bay in 1990 by Dr Alan Lewitus (National Oceanic and Atmospheric Administration). Identification of the cryptophyte was accomplished through the kind assistance of Dr P. Kugrens (Colorado State University). Cultures of both the dinoflagellate and the cryptophyte contained heterotrophic bacteria, although preliminary experiments showed no uptake of fluorescently labeled bacteria by K. micrum (J. E. Adolf, unpublished data). Further, autotrophic growth of bacterized and axenic K. micrum was similar, suggesting no contribution of bacteria to K. micrum nutrition (J. E. Adolf, unpublished data). Cell counts to determine growth rates and frequency of K. micrum cells containing ingested cryptophyte prey were routinely made by filtering 3–5 mL culture aliquots fixed in 1% glutaraldehyde onto 25-mm blackened membrane filters (0.8 µm pore size, Poretics) and by counting at 400× on an epifluorescence microscope (Nikon Eclipse E800) using filter set EF-4 B-2A (exciter filter 450–490 nm, dichromatic beam splitter 500 nm, barrier filter 515 nm). Ingested cryptophytes were counted as orange fluorescent inclusions (OFIs) (Li et al., 1996) that were clearly visible inside the red fluorescent dinoflagellate cells. OFIs were enumerated in the first 100–150 cells encountered on a slide. Chlorophyll a (Chl a) was determined by high-performance liquid chromatography (HPLC) (see below) in all cultures, with the exception of MIXO 1 for which Chl a was measured fluorometrically (Turner Designs model 10-AU, Turner Designs, Sunnyvale, CA, USA) in 90% acetone extracts. Comparison of Chl a determined by HPLC and Turner Designs fluorometer produced the regression, Chl a (HPLC) = 0.418 + 1.001 × Chl a (Turner fluorometer), n = 424, r2 = 0.94. Particulate C and N were measured on samples collected on pre-combusted Whatman GF/C filters with a Leeman Lab 440 HA elemental analyzer.

Semicontinuous batch culture of mixotrophic K. micrum

Commonly used abbreviations are listed in Table I, and experimental designations and descriptions are listed in Table II. Semicontinuous batch cultures (SBCs) were used (MIXO 1 and MIXO 3) to maintain steady-state feeding and growth rates over the course of experiments that lasted ∼1 week. For all SBCs, growth irradiance (Eo) was 10.8 mol photons·m−2·day−1 (250 µmol photons·m−2·s−1, 12 : 12 L : D), temperature was 20°C and salinity was 15. Duplicate SBCs of mixotrophic (MIXO 1 and MIXO 3) K. micrum and prey S. major were maintained in 2.8-L Fernbach flasks in f/2-Si (Guillard, 1975) medium with P at f/20 concentration (883 µM NO3 and 3.6 µM PO4). AUTO 1 (Table I) was maintained as an autotrophically grown SBC control for the MIXO 1 experiment. AUTO 1 was diluted 50% every 2 days, whereas MIXO 1 and prey S. major were diluted 40 and 60% daily, respectively. All dilutions took place immediately preceding ‘lights on’ in the incubation chamber. MIXO 1 was fed daily from the SBC of S. major at the same time that cultures were diluted. Growth rate for each SBC was calculated from cell counts made immediately following dilution and then 24 h later and was averaged over the 6-day period during which SBCs were maintained. The average prey : predator ratio at the time of feeding was 1.4 (±0.20) over the 6-day MIXO 1 experiment and 0.5 (±0.12) over the 4-day MIXO 3 experiment. We computed clearance (F) and grazing rate (I) using the equations of Frost (Frost, 1972) from changes in prey cell densities in MIXO 1 and in an ungrazed prey control grown in MIXO 1 filtrate (GF/F filtrate of MIXO 1 culture), during a 24-h period in which cell samples were taken at 4-h intervals throughout the light period when most grazing occurred (J. E. Adolf, personal observation). HPcell (pg C·cell−1·day−1) was calculated according to

 

(2)
\[\mathrm{HP}^{\mathrm{cell}}=\mathrm{I}^{\mathrm{cell}}{\times}\mathrm{(}\mathrm{prey\ cell\ C\ quota}\mathrm{)}\]

where Icell has units prey (predator × day)−1 and using an average mid-day ‘prey cell C quota’ value of 57 pg C·cell−1 measured for S. major. Photosynthesis–irradiance (PE) curve (see below) measurements and samples for bio-optical and biochemical analyses were taken within 2 h of ‘lights on’.

Table I:

Commonly used abbreviations in the text

Abbreviations Definition 
PPcell Photosynthetic performance (pg C·cell−1·day−1
HPcell Heterotrophic performance (pg C·cell−1·day−1
%GE Growth efficiency [100 × (C production/C intake)] 
OFI Orange fluorescent inclusion 
SBC Semicontinuous batch culture 
Icell Ingestion rate [prey (predator × day)−1
F Clearance rate [µL (predator × h)−1
PE Photosynthesis–irradiance 
Eo Incident irradiance [mol photons (m2 × day)−1
PP Photosynthetic pigment 
NPP Nonphotosynthetic pigment 
Pmcell Maximum cellular photosynthesis rate from PE curve [pg C (cell × h)−1
αcell Cellular light-limited photosynthetic efficiency [pg C (cell × h)−1] [µmol photons (m2 × s)−1]−1 
ΦC(max.) Maximum quantum efficiency of photosynthesis (mol C fixed/mol photons absorbed) 
Abbreviations Definition 
PPcell Photosynthetic performance (pg C·cell−1·day−1
HPcell Heterotrophic performance (pg C·cell−1·day−1
%GE Growth efficiency [100 × (C production/C intake)] 
OFI Orange fluorescent inclusion 
SBC Semicontinuous batch culture 
Icell Ingestion rate [prey (predator × day)−1
F Clearance rate [µL (predator × h)−1
PE Photosynthesis–irradiance 
Eo Incident irradiance [mol photons (m2 × day)−1
PP Photosynthetic pigment 
NPP Nonphotosynthetic pigment 
Pmcell Maximum cellular photosynthesis rate from PE curve [pg C (cell × h)−1
αcell Cellular light-limited photosynthetic efficiency [pg C (cell × h)−1] [µmol photons (m2 × s)−1]−1 
ΦC(max.) Maximum quantum efficiency of photosynthesis (mol C fixed/mol photons absorbed) 

Table II:

Summary of experiments

Experiment Notes 
MIXO 1 Seven-day SBC of mixotrophic Karlodinium micrum fed Storeatula major once each day, f/2-Si (P/20) 
MIXO 2 Mixotrophic K. micrum pulse-fed one time with S. major, f/2-Si (P/20) 
MIXO 3 Four-day SBC of mixotrophic K. micrum fed S. major once each day, f/2-Si (P/20) 
AUTO 1 SBC of autotrophic K. micrum run simultaneously with MIXO 1, f/2-Si (P/20) 
AUTO 2 Autotrophic control culture run simultaneously with MIXO 2, f/2-Si (P/20) 
AUTO Nutrient-replete autotrophic K. micrum cultures, f/2-Si (Adolf et al., 2003
AUTO-N Stationary phase N-deprived autotrophic K. micrum, f/2-Si (N/20) 
AUTO-P Stationary phase P-deprived autotrophic K. micrum, f/2-Si (P/20) 
Experiment Notes 
MIXO 1 Seven-day SBC of mixotrophic Karlodinium micrum fed Storeatula major once each day, f/2-Si (P/20) 
MIXO 2 Mixotrophic K. micrum pulse-fed one time with S. major, f/2-Si (P/20) 
MIXO 3 Four-day SBC of mixotrophic K. micrum fed S. major once each day, f/2-Si (P/20) 
AUTO 1 SBC of autotrophic K. micrum run simultaneously with MIXO 1, f/2-Si (P/20) 
AUTO 2 Autotrophic control culture run simultaneously with MIXO 2, f/2-Si (P/20) 
AUTO Nutrient-replete autotrophic K. micrum cultures, f/2-Si (Adolf et al., 2003
AUTO-N Stationary phase N-deprived autotrophic K. micrum, f/2-Si (N/20) 
AUTO-P Stationary phase P-deprived autotrophic K. micrum, f/2-Si (P/20) 

All experiments were conducted at Eo 250 µmol photons·m−2·s−1 (12 : 12 L : D), salinity 15 and temperature 20°C. f/2-Si is the growth medium with indicated changes. SBC, semicontinuous batch culture. MIXO 1 and MIXO 3 were conducted separately and with different food supply rates. AUTO refers to nutrient-replete autotrophic cultures presented in Adolf et al. (Adolf et al., 2003). Further details of the experiments are given in Methods.

14C tracing of autotrophic and heterotrophic assimilation end products

A generalized diagram of the experimental setup used to measure the assimilation of inorganic 14C and 14C-labeled prey by mixotrophic K. micrum in MIXO 2 and MIXO 3 experiments is shown in Fig. 1. MIXO 2 and MIXO 3 used duplicate cultures grown in medium consisting of f/2-Si with P at f/20 concentration. Eo was 10.8 mol photons·m−2·day−1 (250 µmol photons·m−2·s−1, 12 : 12 L : D), temperature was 20°C and salinity was 15. MIXO 2 was a ‘pulse-fed’ experiment wherein physiological measurements were made over the first L : D cycle following feeding of an autotrophic (AUTO 2) culture. MIXO 3 was maintained as a mixotrophic SBC for 3 days before the experimental procedures outlined in Fig. 1 were started at t = 72 h. For each experiment, a late-exponential phase autotrophic culture of K. micrum was split into four aliquots at the beginning of the dark period (A–D in Fig. 1). One aliquot was not fed (D in Fig. 1; autotrophic control), and of the other three, one was immediately fed with rinsed, 14C-labeled prey (see below) and used to measure heterotrophic C metabolism (A in Fig. 1), while the remaining two aliquots were fed with rinsed, unlabeled prey (see below). The three fed cultures received the same amount of prey. Of the two cultures fed with unlabeled prey, one culture (B in Fig. 1) was spiked with NaH14CO3 (ICN, Irvine, CA, USA) on the following morning and used to measure photosynthetic C uptake and metabolism over the subsequent 24 h encompassing one L : D cycle (see below). The other culture (C in Fig. 1) received no radioactive tracer and was used for cell samples and to determine PE curves ∼2 h after ‘lights on’, corresponding to t = 14 h (Fig. 2B, MIXO 2) and t = 86 h (Fig. 2C, MIXO 3).

Fig. 1.

Schematic representation of the experimental design used to separately measure autotrophic and heterotrophic C metabolism in mixotrophic Karlodinium micrum in the MIXO 2 (pulse-fed) and MIXO 3 (semicontinuous batch culture) experiments. Three parallel mixotrophic cultures and one autotrophic control were run. Mixotrophic cultures received either 14C-labeled prey (A), 14CO2 and unlabeled prey (B) or no radio-label (C). Autotrophic controls (D) received 14CO2. Feeding was designed such that the prey introduced at the start of the dark phase was significantly reduced by ingestion within 12 h.

Fig. 1.

Schematic representation of the experimental design used to separately measure autotrophic and heterotrophic C metabolism in mixotrophic Karlodinium micrum in the MIXO 2 (pulse-fed) and MIXO 3 (semicontinuous batch culture) experiments. Three parallel mixotrophic cultures and one autotrophic control were run. Mixotrophic cultures received either 14C-labeled prey (A), 14CO2 and unlabeled prey (B) or no radio-label (C). Autotrophic controls (D) received 14CO2. Feeding was designed such that the prey introduced at the start of the dark phase was significantly reduced by ingestion within 12 h.

Fig. 2.

Cell dynamics in the three mixotrophy experiments. (A) One representative day from the MIXO 1 experiment that was maintained as a semicontinuous batch culture for 1 week. (B) and (C): The full extent of experiments MIXO 2 and MIXO 3, respectively. OFI, orange fluorescent inclusion, indicative of ingested Storeatula major inside Karlodinium micrum. OFIs are omitted from panel C for clarity. The white diamonds in panel B refer to OFIs as for panel A.

Fig. 2.

Cell dynamics in the three mixotrophy experiments. (A) One representative day from the MIXO 1 experiment that was maintained as a semicontinuous batch culture for 1 week. (B) and (C): The full extent of experiments MIXO 2 and MIXO 3, respectively. OFI, orange fluorescent inclusion, indicative of ingested Storeatula major inside Karlodinium micrum. OFIs are omitted from panel C for clarity. The white diamonds in panel B refer to OFIs as for panel A.

Rinsed 14C-labeled prey and rinsed unlabeled prey were prepared from 900 mL S. major cultures that had been started from a small inoculum (1 : 100 vol.) 6 days prior to use. The 14C-labeled prey culture was spiked with 0.5 mL of NaH14CO3 to give an activity of 8552 DPM mL−1 and grown next to the unlabeled prey to check for adverse effects of prolonged growth in the presence of 14C. The two cultures achieved similar biomass by the end of the growth period. Before feeding, each prey culture was centrifuged for 5–10 min at room temperature (IEC Clinical Centrifuge, 5000×g) in acid-rinsed 50-mL polypropylene tubes to remove nonparticulate 14C. The supernatant was decanted from the pelleted cells, and the tube was refilled with culture medium to resuspend the cells. This procedure was repeated two more times to yield rinsed prey cells. In vivo fluorescence was used to check for equal recovery of cells between centrifuging and to assure that the biomasses in 14C-labeled and unlabeled prey were equal following the rinsing procedure. An aliquot of the unlabeled prey was taken to determine prey cells mL−1, and an aliquot of the labeled prey was used to determine DPM mL−1. These two measurements were used to calculate DPM prey cell−1 that was converted to DPM prey C−1 using the value 68 pg C·prey cell−1 measured in the unlabeled prey culture at the time of feeding.

Several autotrophically grown K. micrum cultures were used as controls in this study. ‘AUTO’ cultures (Table I) refers to batch cultures (Adolf et al., 2003) at the same growth irradiance used for mixotrophy experiments (250 µmol photons·m−2·s−1, 12 : 12 L : D). ‘AUTO 1’ was grown in SBC conditions in the same experiment as ‘MIXO 1’. Duplicate P- or N-deprived cultures (‘AUTO-P’, ‘AUTO-N’) were produced by inoculating a small volume of exponentially growing cells into f/2-Si medium that had either N or P at f/20 concentration (i.e., one-tenth concentration of f/2). Cultures were monitored daily until stationary phase was reached. Measurements of cellular nutrient ratios (mol) in N-deprived (C : N = 8.2) and P-deprived (C : P = 203.5) cultures confirmed intracellular nutrient deficiency compared to nutrient-replete autotrophic cultures (C : N = 5.0, C : P = 37.9).

14C measurements of photosynthesis

PE relationships were determined at 20°C in temperature-regulated ‘photosynthetrons’, essentially as described by Lewis and Smith (Lewis and Smith, 1983). Detailed methods are presented in Adolf et al. (Adolf et al., 2003). To measure PPcell, aliquots (∼300 mL) of parent cultures were placed into 1-L screw-cap culture flasks and spiked with NaH14CO3 to achieve an activity of 1.6 × 105 DPM mL−1 (±10%) at the start of the light period. Spiked aliquots were incubated adjacent to parent cultures in growth chambers. Samples were taken both at the end of the light period and at the end of the dark period to represent gross and net C fixation, respectively. At each time point, duplicate samples were collected on Whatman GF/F filters, followed by a rinse of the filters with culture medium. One of each pair of filters was gently acidified with 0.01 N HCl and placed directly into a vial containing scintillation cocktail (Packard Ultima-Gold AP, Packard Bioscience, Meriden, CT, USA) to measure TOTAL 14C uptake. The other filter was folded (sample on inside), placed into a 1.7-mL microcentrifuge tube with 1 mL dH2O and stored frozen (–20°C) until processing. We analyzed these samples for 14C incorporation into photosynthetic end products using a serial extraction technique similar to that described by Morris et al. (Morris et al., 1974). Specific solvents and extraction methods followed Lancelot (Lancelot, 1984), except that separation of soluble from insoluble material was accomplished by microcentrifugation instead of filtration. Briefly, frozen filters plus water in microcentrifuge tubes were completely lyophilized and then extracted with 1 mL CHCl3 at 4°C for 1 h. Tubes were then centrifuged (Eppendorf microcentrifuge, 5 min, room temperature), and the supernatants were transferred to scintillation vials containing 4.5 mL Packard Ultima-Gold AP scintillation cocktail that accommodated the sample : cocktail ratios required by the method. Tubes were refilled with 1 mL CHCl3, centrifuged and the supernatants transferred to the same scintillation vials. This procedure was repeated following 1-h incubation in hot (85–90°C) anhydrous ethanol (EtOH) and then in hot (85–90°C) 5% trichloracetic acid (TCA). A third rinse with TCA improved recovery of TCA-soluble material. The filter remaining in the tube after the final solvent extraction was placed into a vial containing scintillation cocktail. Summed DPM recovered using this method averaged 97% [coefficient of variation (CV) = 6%] of the DPM in the unfractionated controls. The fraction of 14C in any biochemical class relative to the sum of 14C counts collected in all fractions was calculated to improve the precision of the technique (Cuhel and Lean, 1987). Following Lancelot (Lancelot, 1984), CHCl3-soluble material is referred to as LIPID, hot EtOH-soluble material as low-molecular-weight metabolites (LMW), hot 5% TCA-soluble material as polysaccharides (PSACCH) and hot 5% TCA-insoluble material as protein (PROT). %Δ14C (overnight) was calculated as

 

(3)
\[\mathrm{\%}\mathrm{{\Delta}}^{14}\mathrm{C}=\frac{\mathrm{(}\mathrm{DPM}^{\mathrm{D}}{-}\mathrm{DPM}^{\mathrm{L}}\mathrm{)}}{\mathrm{DPM}^{\mathrm{L}}}\]

where DPML and DPMD Are the activities recovered on the unfractionated controls taken at the end of the light and dark periods, respectively.

HPLC and absorption reconstruction

Pigment analyses were performed using HPLC according to the methods of Van Heukelem and Thomas (Van Heukelem and Thomas, 2001) at the Analytical Services facility of Horn Point Laboratory (Cambridge, MD, USA). Samples for HPLC were collected on Whatman GF/F filters, folded with the sample on the inside, flash frozen in liquid N2 and stored at –80°C until analysis. Identified pigments were classified as photosynthetic pigments [PPs: Chl a, Chl c2, Chl c3, fucoxanthin, 19′-butanoyloxy fucoxanthin (but. fucoxanthin), 19′-hexanoyloxy fucoxanthin (hex. fucoxanthin), gyroxanthin diester 1 and 2] and nonphotosynthetic pigments [NPPs: diadinoxanthin (DD), diatoxanthin (DT), carotenes, including the Chl a degradation products pheophorbide and chlorophyllide]. Total particulate absorption spectra of the cultures were determined from samples collected on Whatman GF/F filters according to the procedure of Kishino et al. (Kishino et al., 1985). Absorbance spectra (m−1) were reconstructed from HPLC pigment data according to Bidigare et al. (Bidigare et al., 1990) to determine the ratio of absorbance with and without NPP (aPPs : aTOT). The equations of Dubinsky et al. (Dubinsky et al., 1986) and Kirk (Kirk, 1994) were used to weight absorption spectra for the output of different light sources. Further details on these methods are given in Adolf et al. (Adolf et al., 2003).

To determine the contribution of Chl a from uningested prey to total Chl a measured in mixotrophic cultures, prey Chl a was calculated based on microscopic cell counts of uningested prey and a value of 1.1 pg Chl a·prey cell−1. This calculation resulted in determinations of uningested prey Chl a as 1.8, 0.2 and 3.8% of total Chl a in MIXO 1, MIXO 2 and MIXO 3, respectively.

RESULTS

Semicontinuous and pulse-feeding techniques proved effective to grow K. micrum for physiological experiments (Fig. 2A–C). Prey were rapidly ingested and visible as OFI inside predator cells of K. micrum in the feeding conditions we used. OFIs appeared rapidly as prey were grazed and disappeared due to digestion and dilution by cell division (Fig. 2A and B).

PPcell ranged from 29.6 to 61.3 pg C·cell−1·day−1 among autotrophic and mixotrophic K. micrum, with highest values in AUTO (Table III). Grazing cultures had clearance rates (F) between 2.7 and 8.9 µL·predator−1·h−1, with the lowest value corresponding to the lowest ingestion rate, I. Ingestion rates among mixotrophic cultures varied 6-fold and accounted for HPcell of 21.5–96.3 pg C·cell−1·day−1 (Table III). In the experiments presented here, ingestion rates simply reflected the food supply rate. Division rates of mixotrophic K. micrum were equal to or greater than those observed in the autotrophic K. micrum (Table III). Net carbon production (NET, Table III), calculated from observed division rates and cellular C quotas, was similar in AUTO, MIXO 2 and MIXO 3 treatments, despite differences of PPcell. In contrast, the highest value of NET was observed for MIXO 1 where HPcell was also highest. %GE, based on the sum of autotrophic and heterotrophic C gain, was 34–47%, similar among the treatments we used despite their different nutritional status.

Table III:

C-budget summaries for autotrophic (AUTO) and mixotrophic (MIXO 1–3) Karlodinium micrum cultures used in this study

 PP HP SUM µ (div d−1NET %GGE 
AUTO 61.3 (0.73) — — — 61.3 (0.73) 0.55 (0.021) 28.9 (1.21) 47 (2.1) 
MIXO 1 29.6 (4.02) 7.5 (0.40) 1.7 (0.19) 96.3 (11.02) 125.9 (11.73) 0.75 (0.081) 42.3 (4.63) 34 (4.8) 
MIXO 2 37.6 (3.07) 8.9 (1.26) 0.7 (0.02) 49.4 (1.05) 87.0 (3.25) 0.57 (0.150) 30.0 (1.26) 34 (1.9) 
MIXO 3 46.4 (1.38) 2.7 (0.23) 0.3 (0.02) 21.5 (1.09) 67.9 (1.76) 0.52 (0.070) 26.8 (3.66) 39 (5.5) 
 PP HP SUM µ (div d−1NET %GGE 
AUTO 61.3 (0.73) — — — 61.3 (0.73) 0.55 (0.021) 28.9 (1.21) 47 (2.1) 
MIXO 1 29.6 (4.02) 7.5 (0.40) 1.7 (0.19) 96.3 (11.02) 125.9 (11.73) 0.75 (0.081) 42.3 (4.63) 34 (4.8) 
MIXO 2 37.6 (3.07) 8.9 (1.26) 0.7 (0.02) 49.4 (1.05) 87.0 (3.25) 0.57 (0.150) 30.0 (1.26) 34 (1.9) 
MIXO 3 46.4 (1.38) 2.7 (0.23) 0.3 (0.02) 21.5 (1.09) 67.9 (1.76) 0.52 (0.070) 26.8 (3.66) 39 (5.5) 

PP, photosynthetic performance (pg C·cell−1·day−1); F, clearance rate (µL·predator−1·h−1); I, ingestion rate (prey predator−1·day−1); HP, heterotrophic performance (pg C ingested·predator−1·day−1); SUM = PP + HP; NET, net C production (pg C day−1) based on the culture growth rate (µ, div. ·day−1) and a C quota value of 62 pg cell−1; %GE (growth efficiency) = 100 × (NET/SUM). Values in parentheses are SDs.

PPcell and PE curve parameters corresponded to lower photosynthetic capacity of K. micrum in mixotrophic growth compared to nutrient-replete autotrophic growth, a reduction that reflected cellular Chl a content. PPcell was negatively correlated with HPcell in simple linear regression (Fig. 3). Pmcell (Fig. 4A) and αcell (Fig. 4B) were both lower in mixotrophic K. micrum, except for MIXO 2 wherein an autotrophic culture was fed cryptophyte prey and the PE relationship was measured <15 h later. Cellular PE parameters (Fig. 4A and B) in experiments MIXO 1 and MIXO 3 were approximately twice those in AUTO-P, a P-limited batch culture, although Chl a was similar. MIXO 1 and MIXO 3 had higher cellular PE parameters and Chl a than those measured in AUTO-N, a N-limited batch culture. Pmchl and αchl were less variable among treatments and were not significantly related to cellular Chl a content (data not shown). Pmchl (g C·g Chl a−1·h−1) and αchl (g C·g Chl a−1·h−1) (µmol photons·m−2·s−1)−1 averaged 3.5 (SD = 0.33) and 0.014 (SD = 0.0016), respectively, in all cultures excluding nutrient-deprived autotrophic cultures (AUTO-N and AUTO-P), where Pmchl and αchl averaged 2.2 (SD = 0.46) and 0.009 (SD = 0.0014).

Fig. 3.

The relationship between photosynthetic performance (PPcell over the light period, pg C·cell−1·day−1) and heterotrophic performance (HPcell, pg C·cell−1·day−1) in Karlodinium micrum.

Fig. 3.

The relationship between photosynthetic performance (PPcell over the light period, pg C·cell−1·day−1) and heterotrophic performance (HPcell, pg C·cell−1·day−1) in Karlodinium micrum.

Fig. 4.

The relationship between photosynthesis–irradiance (PE) parameters and Chl a cell−1 in autotrophic and mixotrophic Karlodinium micrum. (A) Pmcell is the light-saturated photosynthetic rate normalized to cell number. (B) αcell is the light-limited slope of the PE curve normalized to cell numbers.

Fig. 4.

The relationship between photosynthesis–irradiance (PE) parameters and Chl a cell−1 in autotrophic and mixotrophic Karlodinium micrum. (A) Pmcell is the light-saturated photosynthetic rate normalized to cell number. (B) αcell is the light-limited slope of the PE curve normalized to cell numbers.

Cellular photopigment quotas in mixotrophic and nutrient-deprived K. micrum differed from expected values established for nutrient-replete autotrophic K. micrum cultures (Fig. 5, Table IV). In autotrophic K. micrum, all photopigments, except DD + DT, but. fucoxanthin and hex. fucoxanthin, showed a significant negative relationship with growth irradiance (Fig. 5). Cellular photopigment quotas from mixotrophy and nutrient starvation experiments were compared to cellular nutrient quotas measured in autotrophic cultures grown at the same growth irradiance (10.8 mol photons·m−2·day−1, 12 : 12 L : D). All cellular photopigment quotas in mixotrophic cultures were significantly lower (P < 0.05) than those measured in autotrophic K. micrum. Cellular photopigment quotas, except hex. fucoxanthin, but. fucoxanthin and DD + DT, were significantly lower in nutrient-deprived autotrophic K. micrum than in nutrient-replete autotrophic K. micrum (Table IV).

Fig. 5.

Changes in photopigment cell−1 as a function of growth irradiance in nutrient-replete autotrophic Karlodinium micrum (black circles, regression line), mixotrophic K. micrum (white circles, MIXO 1 and MIXO 3) and nutrient-deprived autotrophic K. micrum (white squares, AUTO-N, AUTO-P).

Fig. 5.

Changes in photopigment cell−1 as a function of growth irradiance in nutrient-replete autotrophic Karlodinium micrum (black circles, regression line), mixotrophic K. micrum (white circles, MIXO 1 and MIXO 3) and nutrient-deprived autotrophic K. micrum (white squares, AUTO-N, AUTO-P).

Table IV:

Comparisons of photopigment cell1 between nutrient-replete autotrophic Karlodinium micrum and mixotrophic (MIXO) or nutrient-deprived autotrophic (AUTO-N and AUTO-P) K. micrum at the same growth irradiance (10.8 mol photons ·m2·day1)

 MIXO (n = 4)
 
 Nutrient-deprived AUTO (n = 4)
 
 
 % change P % change P 
Chl a cell−1 −28.3 0.004 −42.1 0.072 
Chl c2 cell−1 −30.0 0.005 −46.5 0.053 
Chl c3 cell−1 −45.8 0.000 −55.4 0.006 
Fucoxanthin cell−1 −36.4 0.005 −67.7 0.001 
Hexanoyloxy fucoxanthin cell−1 −33.6 0.003 −28.6 0.287 
Butanoyloxy fucoxanthin cell−1 −29.7 0.006 −30.8 0.237 
DD + DT cell−1 −36.4 0.025 −19.4 0.384 
Gyroxanthin diester cell−1 −53.7 0.007 −78.8 0.000 
 MIXO (n = 4)
 
 Nutrient-deprived AUTO (n = 4)
 
 
 % change P % change P 
Chl a cell−1 −28.3 0.004 −42.1 0.072 
Chl c2 cell−1 −30.0 0.005 −46.5 0.053 
Chl c3 cell−1 −45.8 0.000 −55.4 0.006 
Fucoxanthin cell−1 −36.4 0.005 −67.7 0.001 
Hexanoyloxy fucoxanthin cell−1 −33.6 0.003 −28.6 0.287 
Butanoyloxy fucoxanthin cell−1 −29.7 0.006 −30.8 0.237 
DD + DT cell−1 −36.4 0.025 −19.4 0.384 
Gyroxanthin diester cell−1 −53.7 0.007 −78.8 0.000 

‘% change’ refers to the treatment relative to nutrient-replete autotrophic cultures [100 × (treatment – AUTO)/AUTO]. DD, diadinoxanthin; DT, diatoxanthin. Statistical P is derived from a t-test where P < 0.05 was considered significant.

Representative pie diagrams from MIXO 1 show an accumulation of NPP in mixotrophic K. micrum on % weight (Fig. 6A and B) and % absorption (Fig. 6C and D) bases. In all treatments, Chl a was the dominant pigment by weight and absorption. Likewise, fucoxanthin was the dominant accessory pigment. The increased proportion of NPP in mixotrophic K. micrum was the result of accumulated pheophorbide, alloxanthin and β-carotene (Fig. 6B and D). The effect of accumulated NPP on photosynthesis would be mediated through a decrease in the ratio of absorption by photosynthetically active pigments (PPs) to TOTAL pigments. Figure 7 shows a statistically significant, positive relationship between PPs : TOTAL absorption and ΦC(max.) for K. micrum cultures spanning a range of nutritional conditions, suggesting that changes in composition of the pigment system of K. micrum accompany changes in the efficiency at which K. micrum uses absorbed light to fix C [ΦC(max.)].

Fig. 6.

The relative proportion of photopigments in autotrophic (AUTO 1) and mixotrophic (MIXO 1) Karlodinium micrum. Panels A and B show the relative pigment concentration by weight for AUTO 1 and MIXO 1, respectively. Panels C and D show the relative pigment concentration by absorption for AUTO 1 and MIXO 1, respectively. Cultures were sampled at the beginning of the light period immediately before re-feeding (MIXO 1), and dilution of these semicontinuous batch cultures occurred. Values from duplicate cultures for each treatment were pooled to make these graphs.

Fig. 6.

The relative proportion of photopigments in autotrophic (AUTO 1) and mixotrophic (MIXO 1) Karlodinium micrum. Panels A and B show the relative pigment concentration by weight for AUTO 1 and MIXO 1, respectively. Panels C and D show the relative pigment concentration by absorption for AUTO 1 and MIXO 1, respectively. Cultures were sampled at the beginning of the light period immediately before re-feeding (MIXO 1), and dilution of these semicontinuous batch cultures occurred. Values from duplicate cultures for each treatment were pooled to make these graphs.

Fig. 7.

The relationship between ΦC(max.) (the maximum quantum efficiency of photosynthetic C fixation with unit mol C·mol photon−1) and PPs : TOTAL (the absorption-based ratio of photosynthetic to total pigments) in autotrophic and mixotrophic Karlodinium micrum.

Fig. 7.

The relationship between ΦC(max.) (the maximum quantum efficiency of photosynthetic C fixation with unit mol C·mol photon−1) and PPs : TOTAL (the absorption-based ratio of photosynthetic to total pigments) in autotrophic and mixotrophic Karlodinium micrum.

Figure 8 shows the time course of C assimilation into macromolecular fractions in an autotrophic culture (AUTO 2) and MIXO 2, which had been started by feeding an aliquot of AUTO 2 at the beginning of the previous dark period (0 h, not shown on graph). Changes of photosynthetic C assimilation were observed within 24 h of introducing food to autotrophically grown K. micrum (Fig. 8). At the end of the light period, the most pronounced changes were reduced assimilation of polysaccharide C in mixotrophic cultures (24 h, Fig. 8C). By the end of the dark period, AUTO 2 and MIXO 2 had similar polysaccharide C levels (36 h, Fig. 8C), but MIXO 2 contained lower levels of protein C than AUTO 2 (36 h, Fig. 8D). Tracing of C assimilation from ingested prey in the same experiment (MIXO 2) is shown in Fig. 9. Comparison of 14C-labeled organic material in the culture between t = 0 h (when 14C was initially added as evenly labeled prey cells) and t = 12 h (when all prey cells had been ingested) indicated assimilation of ∼67% of ingested C, most of which was assimilated into protein (Fig. 9). Combined results of 14C tracing experiments for MIXO 2 and MIXO 3 showed that allocation of photosynthetically assimilated C to protein was significantly lower in mixotrophic K. micrum compared to autotrophic K. micrum (Table V). Reduced allocation of photosynthetic 14C to protein in mixotrophic K. micrum was compensated by increased allocation to lipid and polysaccharide. Allocation of 14C assimilated by heterotrophy, at a time point 36 h post-feeding, was predominantly protein and reflected the distribution of 14C in the prey it was fed (Table V).

Fig. 8.

Time course of photosynthetic (A) lipid, (B) low-molecular-weight (LMW) metabolites, (C) polysaccharide and (D) protein following introduction of prey to autotrophically grown Karlodinium micrum. AUTO 2 was maintained under autotrophic conditions for this experiment, while MIXO 2 was started by feeding an aliquot of AUTO 2 with Storeatula major prey.

Fig. 8.

Time course of photosynthetic (A) lipid, (B) low-molecular-weight (LMW) metabolites, (C) polysaccharide and (D) protein following introduction of prey to autotrophically grown Karlodinium micrum. AUTO 2 was maintained under autotrophic conditions for this experiment, while MIXO 2 was started by feeding an aliquot of AUTO 2 with Storeatula major prey.

Fig. 9.

Accumulation of metabolic end products in Karlodinium micrum after introduction of 14C-labeled prey. HP, heterotrophic performance; LMW, low-molecular-weight metabolites.

Fig. 9.

Accumulation of metabolic end products in Karlodinium micrum after introduction of 14C-labeled prey. HP, heterotrophic performance; LMW, low-molecular-weight metabolites.

Table V:

Summary statistics for proportions (%Cfrac.) of C incorporation from photosynthesis and heterotrophy into end products in autotrophic and mixotrophic Karlodinium micrum

 % Cfrac. (photosynthesis)
 
  % Cfrac. (heterotrophy)
 
  
 AUTO (n = 4) MIXO (n = 4) P MIXO (n = 4) PREY (n = 6) P 
LIPID 8 (0.5) 12 (3.4) n.s. 11 (2.4) 10 (1.5) n.s. 
LMW 20 (1.2) 18 (4.4) n.s. 13 (4.7) 11 (1.0) n.s. 
PSACCH 14 (1.0) 22 (7.9) n.s. 25 (0.1) 25 (1.4) n.s. 
PROT 57 (1.4) 48 (3.5) 0.003 52 (4.4) 54 (2.3) n.s. 
 % Cfrac. (photosynthesis)
 
  % Cfrac. (heterotrophy)
 
  
 AUTO (n = 4) MIXO (n = 4) P MIXO (n = 4) PREY (n = 6) P 
LIPID 8 (0.5) 12 (3.4) n.s. 11 (2.4) 10 (1.5) n.s. 
LMW 20 (1.2) 18 (4.4) n.s. 13 (4.7) 11 (1.0) n.s. 
PSACCH 14 (1.0) 22 (7.9) n.s. 25 (0.1) 25 (1.4) n.s. 
PROT 57 (1.4) 48 (3.5) 0.003 52 (4.4) 54 (2.3) n.s. 

The column ‘PREY’ indicates the 14C composition of evenly labeled prey at the time of feeding. Values in parentheses are SDs of the mean. LMW, low-molecular-weight metabolites; PSACCH, polysaccharide; PROT, protein; n.s., not significant.

DISCUSSION

Previous studies of physiological adaptations accompanying mixotrophic nutrition in K. micrum focused on conditions that increased the propensity of autotrophic cultures to graze (Li et al., 2000a) or on short-term responses of photosynthesis to grazing (Li et al., 1999) but did not address the balance of autotrophy and heterotrophy during mixotrophic growth. Here, we showed that autotrophy contributed 27–69% to gross C uptake of K. micrum during mixotrophic growth with S. major as prey. 14C assimilation patterns showed a decreased allocation of photosynthetic 14C to protein in grazing compared to nongrazing K. micrum and a higher allocation of prey-derived 14C to protein in mixotrophic cells. Growth efficiencies during mixotrophic growth of K. micrum were typical of dinoflagellates, changed little from autotrophic growth efficiencies and implicated elevated C acquisition from prey as the mechanism responsible for elevated growth rates in mixotrophic versus autotrophic K. micrum. These results contrast with the inability of K. micrum to grow as a strict heterotroph on the same prey (Li et al., 1999) and underscore the importance of both heterotrophic and autotrophic C metabolism during mixotrophic growth of K. micrum.

Photosynthetic performance during mixotrophic growth of K. micrum

Photosynthetic regulation accompanying mixotrophic growth plays a critical role in the balance of autotrophy and heterotrophy in mixotrophic organisms. Whether photosynthesis is stimulated, unchanged or depressed during grazing is a fundamental question with clear ecological implications. Decreases in total PPcell and autotrophic % protein during mixotrophic growth reflected a decreased role of autotrophic nutrition during mixotrophic growth of K. micrum. Changes of PPcell that accompanied grazing in K. micrum were similar to observations of the effects of grazing on PPcell for Fragilidium subglobosum (Hansen et al., 2000; Skovgaard et al., 2000). PPcell of food-saturated F. subglobosum was decreased 48–69% compared to autotrophic cells (Skovgaard et al., 2000), with the strongest effect of grazing on PPcell associated with increased food concentrations (Hansen et al., 2000). Our experiments with K. micrum were not food saturated, thus the best comparison of our data and those of Skovgaard et al. (Skovgaard et al., 2000) is for MIXO 1 in which PPcell was decreased 52% compared to AUTO, and the grazing rate was approximately half of the maximum values reported by Li et al. (Li et al., 2000a). These observations support a conclusion that K. micrum and F. subglobosum regulate PPcell similarly during grazing, with each species showing a decrease of PPcell associated with increased heterotrophy.

Differences of Pmcell between autotrophic and mixotrophic K. micrum are consistent with the conclusion that lower PPcell reflects a significant reduction in cellular investment in photosynthesis. Decreases in Pmcell observed in mixotrophic K. micrum suggest decreased cellular quotas or activities of Calvin cycle enzymes (Falkowski, 1992; Geider and MacIntyre, 2002). Parallel changes in Pmcell and Chl a cell−1 in autotrophic (Adolf et al., 2003) and mixotrophic (Fig. 4A) cultures of K. micrum suggest that acclimation of the photosynthetic apparatus primarily involves changes in photosynthetic unit (PSU) ‘numbers’ (cf. Prézelin, 1987). Reductions of PPcell (day-long incubations) were accompanied by reductions of Pmcell (30 min PE curve), except in MIXO 2 that showed PE curve parameters similar to autotrophic cultures, but PPcell similar to other mixotrophic treatments. This discrepancy between Pmcell and PPcell in MIXO 2 most likely reflects a lag between prey ingestion and effects on cellular photosynthesis as, in the MIXO 2 experiment, PE curve measurements were made at a time when 14C uptake rates of mixotrophic and autotrophic cultures had not yet diverged (Fig. 8). Thus, our PE data suggest that lower PPcell in mixotrophic K. micrum results from a reduction of Pmcell rather than from a reduction of cellular Chl a.

Mixotrophy and nutrient starvation significantly lowered cellular photopigment quotas from expected values established for nutrient-replete autotrophic K. micrum (Table IV). Quotas of Chl a and Chl c3 were lower in mixotrophic cells despite potential contributions from ingested cryptophytes contained in food vacuoles. This pattern was consistent with reductions in Pmcell suggested by PE curve data and further supports the conclusion that heterotrophic metabolism dominates when K. micrum uses a mixotrophic nutritional mode.

We measured the effects of pigment composition on ΦC(max.) in autotrophic and mixotrophic K. micrum to examine potential effects of prey-derived and/or endogenously produced NPP on photosynthesis (Bidigare et al., 1989). Pheophorbide constituted a significant fraction of NPP in these cultures, and its specific and spectral absorption properties, which are similar to those of Chl a (Jeffrey et al., 1997), underlie its significant contribution to cellular light absorption measured as a decline of PPs : TOTAL absorption. Herbivorous protists produce pheophorbides as chlorophyll digestion products (Strom, 1993), supporting a conclusion that the presence of this pigment in mixotrophic K. micrum resulted from digested prey. We observed ∼77% decrease of ΦC(max.) accompanied by ∼11% decline of PPs : TOTAL pigment absorption (Figs 6 and 7), suggesting that absorption by NPP did not explain the decline of ΦC(max.) we observed in K. micrum for a range of nutritional conditions. Mixotrophic chlorophytes show altered metabolic demands associated with heterotrophic C metabolism that may impact ΦC(max.). Photoheterotrophic assimilation of acetate in Pyrobotrus stellata was associated with a shift in light harvesting and energy transfer toward PS I, a strategy favoring cyclic ATP production needed in acetate metabolism and disfavoring NADPH production and C fixation (Brandt and Wiessner, 1984; Boichenko et al., 1992; Wolf et al., 1996). A shift in the fate of absorbed light from C fixation to ATP production to support heterotrophic metabolism may be an important physiological mechanism in mixotrophic dinoflagellates deserving further study.

14C allocation patterns from photosynthesis and prey

Differences of 14C allocation patterns between photosynthetically and prey-derived C suggest differences in the use of C from these two sources by K. micrum for growth. Reduced allocation of photosynthetic 14C to protein, as we observed for mixotrophic K. micrum, has been shown for light- and nutrient-limited phytoplankton and is generally accompanied by a decrease of growth rate (Morris, 1981; Harding et al., 1985; DiTullio and Laws, 1986; Harding and Jones, 1988; de Madariaga, 1992). In our study, N- and P-deprived, stationary phase, autotrophic K. micrum cultures allocated 31 ± 0.01% of fixed 14C to protein, lower than in nutrient-replete exponential phase cultures (Table V). 14C-labeling patterns for the ‘kleptoplastidic’ ciliate, Laboea strobila (Putt, 1990), are qualitatively similar to the pattern we found for K. micrum, showing reduced protein synthesis by plastids retained in the ciliate compared to the same plastids in free-living prey cells. At the same time, protein assimilation from prey ingested by L. strobila was high (Putt, 1990). The increased allocation of prey-derived 14C to protein suggests that prey-derived C contributes more to cellular growth than photosynthetically derived C in mixotrophic K. micrum. This qualitative difference in C assimilation patterns of autotrophy and heterotrophy in mixotrophic K. micrum reinforces the quantitative differences and further underscores the importance of heterotrophic metabolism to mixotrophic growth.

Mixotrophic growth

Gross growth efficiency (%GE in Table III) of mixotrophic K. micrum was typical of dinoflagellates (Straile, 1997), varying little as a function of nutritional status, leading us to conclude that increased growth rate associated with mixotrophic nutrition in K. micrum was attributable to increased C gain by the cells. A similar conclusion can be drawn for F. subglobosum where a 3.8-fold increase of growth rate was accompanied by no change of GE (calculated as our %GE from Table IV in Hansen et al., 2000). Although mixotrophic F. subglobosum respired less of its assimilated C than did autotrophic F. subglobosum (Hansen et al., 2000), the magnitude of this difference did not explain the reported growth rate differences between autotrophic and mixotrophic cells in that study.

These findings lead to the conclusion that potential C gain from prey is more important than growth efficiency in explaining growth rate changes arising from mixotrophic nutrition. The relatively large size of the prey compared to predator and constraints on maximum cellular photosynthesis (Adolf et al., 2003) underlie the differences in potential C gain from photosynthesis and phagotrophy in K. micrum. At maximum grazing rates (Li et al., 1999), C gain from phagotrophy would be 260–340 pg C·K. micrum−1·day−1. PE curve measurements on nutrient-replete autotrophic K. micrum indicated potential fixation of <100 pg C·K. micrum−1·day−1 over a 12-h period in saturating light conditions. Thus, in terms of potential gross C gain, K. micrum has a 2- to 3-fold higher capacity associated with heterotrophy than with autotrophy. In a natural population, of course, the realized balance would depend on the relative availability of prey and light, which would be exaggerated by the decrease in illumination as both predator and prey biomass developed, driving cells in high biomass situations toward enhanced heterotrophy.

Our results, stressing the importance of heterotrophic C metabolism during mixotrophic growth of K. micrum, appear counterintuitive, considering previous observations that K. micrum is an obligate autotroph (Li et al., 1999). One resolution may be that photosynthesis during mixotrophic growth provides essential micronutrients that K. micrum cannot obtain from prey, consistent with the lack of dark heterotrophic growth in K. micrum. In contrast to K. micrum, F. subglobosum (Skovgaard, 1996) and Gyrodinium resplendens (Skovgaard, 2000) can grow as strict heterotrophs, perhaps reflecting their ability to acquire essential nutrients from prey, Ceratium spp. The physiological basis of this difference in strict heterotrophic capabilities remains to be determined.

Evolutionary and ecological implications

The study of mixotrophs that combine phagotrophy and photosynthesis is relevant to protistan evolution, as the diversity of protistan autotrophs arose through uptake and retention of plastids (Raven, 1997; Okamoto and Inouye, 2005). Thus, the study of extant mixotrophic dinoflagellates can be regarded as a study of the ecological context within which these evolutionary events took place. A significant insight comes from the observation that reduced photosynthetic capacity accompanies grazing in K. micrum. In terms of the C flux through pelagic ecosystems where mixotrophic organisms predominate (Stickney et al., 2000), our study suggests a shift toward heterotrophy during grazing due to both a reduction in the photosynthetic capacity of the mixotrophs and the removal of autotrophic prey. Important research that remains to be done includes expanding our knowledge of the simple balance between autotrophy and heterotrophy that operates in the variety of mixotrophic protists that are observed in pelagic ecosystems.

ACKNOWLEDGEMENTS

This work represents part of the PhD work of J.E.A. and was partially supported by NSF grants DEB-9412113 (L.W.H.) and NSF OCE-9819670 (D.K.S.). J.E.A. received partial support from a Horn Point Laboratory Graduate Assistantship, a Maryland Sea Grant Graduate Fellowship and the Horn Point Laboratory Student Small Grants Program. The authors gratefully acknowledge the assistance of Michael E. Mallonee, Daniel E. Gustafson, Christy Jordan and W. David Miller.

REFERENCES

Adolf
,
J. E.
, Stoecker, D. K. and Harding, L. W. Jr. (
2003
) Autotrophic growth and photoacclimation in Karlodinium micrum (Dinophyceae) and Storeatula major (Cryptophyceae).
J. Phycol
 .,
39
,
1101
–1108.
Bergholtz
,
T.
, Daugberg, N., Moestrup, Ø. et al. (
2005
) On the identity of Karlodinium veneficum and description of Karlodinium arminger Sp. Nov. (Dinophyceae), based on light and electron microscopy, nuclear-encoded LSU rDNA, and pigment composition.
J. Phycol
 .,
42
,
170
–193.
Bidigare
,
R. R.
, Ondrusek, M. E., Morrow, J. H. et al. (
1990
) In vivo absorption properties of algal pigments.
Ocean Optics X. Proc. SPIE
 ,
1302
,
290
–302.
Bidigare
,
R. R.
, Schofield, O. and Prézelin, B. B. (
1989
) The influence of zeaxanthin on quantum yield of photosynthesis of Synechococcus clone WH7803 (DC2).
Mar. Ecol. Prog. Ser
 .,
56
,
177
–188.
Bockstahler
,
K. R.
and Coats, D. W. (
1993
) Spatial and temporal aspects of mixotrophy in Chesapeake Bay dinoflagellates.
J. Eukaryot. Microbiol
 .,
40
,
49
–60.
Bockstahler
,
K. R.
and Coats, D. W. (
1993
) Grazing of the mixotrophic dinoflagellate Gymnodinium sanguineum on ciliate populations of Chesapeake Bay.
Mar. Biol
 .,
116
,
477
–487.
Boichenko
,
V. A.
, Wiessner, W., Klimov, V. V. et al. (
1992
) Hydrogen photoevolution indicates an increase in the antenna size of photosystem I in Chlamybotrys stellata during transition from autotrophic to photoheterotrophic nutrition.
Plant Physiol
 .,
100
,
518
–524.
Brandt
,
P.
and Wiessner, W. (
1984
) Induction of the de novo formation of the photosystem I-related light-harvesting chlorophyll protein complex LHCPa by photoheterotrophic nutrition.
Plant Physiol
 .,
75
,
253
–254.
Cuhel
,
R. L.
and Lean, D. R. S. (
1987
) Protein synthesis by lake plankton measured using in situ carbon dioxide and sulfate assimilation.
Can. J. Fish. Aquat. Sci
 .,
44
,
2102
–2117.
DiTullio
,
G. R.
and Laws, E. A. (
1986
) Diel periodicity of nitrogen and carbon assimilation in five species of marine phytoplankton: accuracy of methodology for predicting N-assimilation rates and N/C composition ratios.
Mar. Ecol. Prog. Ser
 .,
32
,
123
–132.
Dubinsky
,
Z.
, Falkowski, P. G. and Wyman, K. (
1986
) Light harvesting and utilization by phytoplankton.
Plant Cell Physiol
 .,
27
,
1335
–1349.
Falkowski
,
P. G.
(
1992
) Molecular ecology of phytoplankton photosynthesis. In Falkowski, P. G. and Woodhead, A. D. (eds),
Primary Productivity and Biogeochemical Cycles in the Sea
  (Environmental Science Research), Plenum Press, New York, Vol.
43
, pp.
47
–67.
Falkowski
,
P. G.
, Dubinsky, Z. and Wyman, K. (
1985
) Growth-irradiance relationships in phytoplankton.
Limnol. Oceanogr
 .,
30
,
311
–321.
Frost
,
B. W.
(
1972
) Effects of size and concentration of food particles on the feeding behavior of the marine planktonic copepod Calanus pacificus.
Limnol. Oceanogr
 .,
17
,
805
–815.
Geider
,
R. J.
and MacIntyre, H. L. (
2002
) Physiology and Biochemistry of photosynthesis and algal carbon acquisition. In Williams, P. J. LeB., Thomas, D. N. and Reynolds, C. S. (eds),
Phytoplankton Productivity. Carbon Assimilation in Marine and Freshwater Ecosystems
 . Blackwell Science, Oxford, UK, pp.
16
–43.
Guillard
,
R. R. L.
(
1975
) Culture of phytoplankton for feeding marine invertebrates. In Smith, W. L. and Chanley, M. H. (eds),
Culture of Marine Invertebrate Animals
 . Plenum Press, New York, pp.
26
–60.
Hallegraeff
,
G. M.
(
1993
) A review of harmful algal blooms and their apparent global increase.
Phycologia
 ,
32
,
79
–99.
Hansen
,
P. J.
(
1992
) Prey size selection, feeding rates and growth dynamics of heterotrophic dinoflagellates with special emphasis on
Gyrodinium spirale. Mar. Biol
 .,
114
,
327
–334.
Hansen
,
P. J.
, Skovgaard, A., Glud, R. N. et al. (
2000
) Physiology of the mixotrophic dinoflagellate Fragilidium subglobosum. II. Effects of time scale and prey concentration on photosynthetic performance.
Mar. Ecol. Prog. Ser
 .,
201
,
137
–346.
Harding
,
L. W.
Jr. and Jones, T. W. (
1988
) Carbon metabolism in estuarine phytoplankton: regulation of photosynthetic fixation patterns.
Biol. Ocean
 .,
5
,
183
–214.
Harding
,
L. W.
Jr., Meeson, B. W. and Fisher, T. R. Jr. (
1985
) Patterns of photosynthetic carbon metabolism in light-limited phytoplankton.
Mar. Biol
 .,
89
,
121
–133.
Jeffrey
,
S. W.
, Mantour, R. F. C. and Bjornland, T. (
1997
) Data for the identification of 47 key phytoplankton pigments. In Jeffrey, S. W., Mantour, R. F. C. and Wright, S. W. (eds),
Phytoplankton Pigments in Oceanography
 . UNESCO Publishing, France, pp.
449
–559.
Jeong
,
H. J.
, Yoo, Y. D., Kim, J. S. et al. (
2005
) Mixotrophy in the phototrophic harmful alga Cochlodinium polykrikoides (Dinophycean): prey species, the effects of prey concentration, and grazing impact.
J. Eukaryot. Microbiol
 .,
51
,
563
–569.
Johnson
,
M. D.
, Rome, M. and Stoecker, D. K. (
2003
) Microzooplankton grazing on Prorocentrum minimum and Karlodinium micrum in Chesapeake Bay.
Limnol. Oceanogr
 .,
48
,
238
–248.
Jones
,
R. I.
(
1994
) Mixotrophy in planktonic protists as a spectrum of nutritional strategies.
Mar. Microb. Food. Webs
 ,
8
,
87
–96.
Kirk
,
J. T. O.
(
1994
)
Light and Photosynthesis in Aquatic Ecosystems
 , 2nd edn. Cambridge University Press, Cambridge, UK.
Kishino
,
M.
, Takahashi, M., Okami, N. et al. (
1985
) Estimation of the spectral absorption coefficients of phytoplankton in the sea.
Bull. Mar. Sci
 .,
37
,
634
–642.
Lancelot
,
C.
(
1984
) Metabolic changes in Phaeocystis pouceti (Hariot) Lagerheim during the spring bloom in Belgian coastal waters.
Estuar. Coast. Shelf Sci
 .,
18
,
593
–600.
Legrand
,
C.
, Granelli, E. and Carlsson, P. (
1998
) Induced phagotrophy in the photosynthetic dinoflagellate Heterocapsa triquetra.
Aquat. Microb. Ecol
 .,
15
,
65
–75.
Lewis
,
M. R.
and Smith, J. C. (
1983
) A small volume, short-incubation-time method for measurement of photosynthesis as a function of incident irradiance.
Mar. Ecol. Prog. Ser
 .,
13
,
99
–102.
Li
,
A.
, Stoecker, D. K. and Adolf, J. E. (
1999
) Feeding, pigmentation, photosynthesis and growth of the mixotrophic dinoflagellate
Gyrodinium galatheanum. Aquat. Microb. Ecol
 .,
19
,
163
–176.
Li
,
A.
, Stoecker, D. K., Coats, D. W. et al. (
1996
) Ingestion of fluorescently labeled and phycoerythrin-containing prey by mixotrophic dinoflagellates.
Aquat. Microb. Ecol
 .,
10
,
139
–147.
Li
,
A.
, Stoecker, D. K. and Coats, D. W. (
2000
) Mixotrophy in Gyrodinium galatheanum (Dinophyceae): responses to light intensity and inorganic nutrients.
J. Phycol
 .,
36
,
33
–45.
Li
,
A.
, Stoecker, D. K. and Coats, D. W. (
2000
) Spatial and temporal aspects of Gyrodinium galatheanum in Chesapeake Bay: distribution and mixotrophy.
J. Plank. Res
 .,
22
,
2105
–2124.
de Madariaga
,
I.
(
1992
) Interspecific differences in the photosynthetic carbon metabolism of marine phytoplankton.
Mar. Biol
 .,
114
,
509
–515.
Morris
,
I.
(
1981
) Photosynthetic products, physiological state, and phytoplankton growth. In Platt, T. (ed.),
Physiological Bases of Phytoplankton Growth. Can. Bull. Fish. Aquat. Sci
 ., 210,
83
–102.
Morris
,
I.
, Glover, H. E. and Yentsch, C. S. (
1974
) Products of photosynthesis by marine phytoplankton: the effect of environmental factors on the relative rates of protein synthesis.
Mar. Biol
 .,
27
,
1
–9.
Okamoto
,
N.
and Inouye, I. (
2005
) A secondary symbiosis in progress?
Science
 ,
310
,
287
.
Prézelin
,
B. B.
(
1987
) Photosynthetic physiology of dinoflagellates. In Taylor, F. J. R. (ed.),
The Biology of Dinoflagellates I. Dinoflagellata
 . Blackwell Scientific Publishing, Oxford, UK, pp.
172
–223.
Putt
,
M.
(
1990
) Metabolism of photosynthate in the chloroplast-retaining ciliate Laboea strobila.
Mar. Ecol. Prog. Ser
 .,
60
,
271
–282.
Raven
,
J. A.
(
1997
) Phagotrophy in phototrophs.
Limnol. Oceanogr
 .,
42
,
198
–205.
Schnepf
,
E.
and Elbrächter, M. (
1992
) Nutritional strategies in dinoflagellates. A review with emphasis on cell biological aspects.
Eur. J. Protistol
 .,
28
,
3
–24.
Skovgaard
,
A.
(
1996
) Mixotrophy in Fragilidium subglobosum (Dinophyceae): growth and grazing responses a function of light intensity.
Mar. Ecol. Prog. Ser
 .,
143
,
247
–253.
Skovgaard
,
A.
(
2000
) A phagotrophically derivable growth factor in the plastidic dinoflagellate Gyrodinium resplendens.
J. Phycol
 .,
36
,
1069
–1078.
Skovgaard
,
A.
, Hansen, P. J. and Stoecker, D. K. (
2000
) Physiology of the mixotrophic dinoflagellate Fragilidium subglobosum. I. Effects of phagotrophy and irradiance on photosynthesis and carbon content.
Mar. Ecol. Prog. Ser
 .,
201
,
129
–136.
Smalley
,
G. W.
and Coats, D. W. (
2002
) Ecology of the red-tide dinoflagellate Ceratium furca: distribution, mixotrophy, and grazing impact on ciliate populations of Chesapeake Bay.
J. Eukaryot. Microbiol
 .,
49
,
63
–73.
Stickney
,
H. L.
, Hood, R. R. and Stoecker, D. K. (
2000
) The impact of mixotrophy on planktonic marine ecosystems.
Ecol. Modell
 .,
125
,
203
–230.
Stoecker
,
D. K.
(
1998
) Conceptual models of mixotrophy in planktonic protists.
Euro. J. Protistol
 .,
34
,
281
–290.
Stoecker
,
D. K.
(
1999
) Mixotrophy among dinoflagellates.
J. Eukaryot. Microbiol
 .,
46
,
397
–401.
Stoecker
,
D. K.
, Li, A., Coats, D. W. et al. (
1997
) Mixotrophy in the dinoflagellate Prorocentrum minimum.
Mar. Ecol. Prog. Ser
 .,
152
,
1
–12.
Straile
,
D.
(
1997
) Gross growth efficiencies of protozoan and metazoan zooplankton and their dependence on food concentration, predator-prey weight ratio, and taxonomic group.
Limnol. Oceanogr
 .,
42
,
1375
–1385.
Strom
,
S. L.
(
1993
) Production of pheopigments by marine protozoa: results of laboratory experiments analyzed by HPLC.
Deep-Sea Res
 .,
40
,
57
–80.
Strom
,
S. L.
and Morello, T. A. (
1998
) Comparative growth rates and yields of ciliates and heterotrophic dinoflagellates.
J. Plank. Res
 .,
20
,
571
–584.
Tang
,
E. P. Y.
(
1996
) Why do dinoflagellates have lower growth rates?
J. Phycol
 .,
32
,
80
–84.
Tittel
,
J.
, Bissinger, V., Zippel, B. et al. (
2003
) Mixotrophs combine resource use to outcompete specialists: implications for aquatic food webs.
Proc. Natl. Acad. Sci
 .,
100
,
12776
–12781.
Van Heukelem
,
L.
and Thomas, C. S. (
2001
) Computer-assisted high performance liquid chromatography method development with application to the isolation and analysis of phytoplankton pigments.
J. Chromatogr
 .,
910
,
31
–49.
Wolf
,
A. H.
, Brüden, A., Giersberg, M. et al. (
1996
) Influence of photoheterotrophy on the expression of chlorophyll a/b-binding proteins in the green alga Pyrobotrys stellata.
Photosyn. Res
 .,
49
,
49
–56.

Author notes

1Horn Point Laboratory, University of Maryland Center for Environmental Science, Cambridge, MD 21613, USA and 2Maryland Sea Grant, University of Maryland, College Park, MD 20740, USA