Microb Ecol (2002) 43:341±352 DOI: 10.1007/s00248-002-2002-3 Ó 2002 Springer-Verlag New York Inc.

Dynamic Characteristics of Prochlorococcus and Synechococcus Consumption by Bacterivorous Nano¯agellates U. Christaki,1 C. Courties,2 H. Karayanni,1 A. Giannakourou,1 C. Maravelias,1 K.Ar. Kormas,1; P. Lebaron2 1 2

NCMR, 16604 Aghios Kosmas, Greece UMR CNRS 7628, 66651 Banyuls sur Mer, France

Received: 26 July 2001; Accepted: 7 January 2002; Online publication: 11 March 2002

A

B S T R A C T

We compared the characteristics of ingestion of Prochlorococcus and Synechococcus by the marine heterotrophic nano¯agellate Pseudobodo sp. and by a mixed nano¯agellate culture (around 3 lm in size) obtained from an open sea oligotrophic area. Maximum ingestion rate on Synechococcus (2.7 Syn ¯agellate)1 h)1) was reached at concentrations of 5 ´ 105 Syn mL)1 and decreased between 6 ´ 105 and 1.5 ´ 106 Syn mL)1. In order to validate laboratory data, one set of data on Synechococcus grazing was obtained during a ®eld study in the oligotrophic northeastern Mediterranean Sea. Ingestion rates by heterotrophic nano¯agellates were related to Synechococcus abundance in the water, and the feeding rate showed a clear diel rhythm with consumption being highest during the night, declining during the day hours, and being lowest at dusk. Ingestion rates on Prochlorococcus increased linearly for the whole range of prey density used (i.e., from 1 ´ 103 to 3 ´ 106 Proc mL)1), with maximum ingestion of 6.7 Proc ¯agellate)1 h)1. However, for prey concentrations in the range of 103±l05, which are usually encountered in aquatic systems, ingestion rates were signi®cantly less than on Synechococcus. In our experiments, both Prochlorococcus and Synechococcus proved to be poor food items for support of nano¯agellate growth.

Introduction The photosynthetic prokaryotes Synechococcus and Prochlorococcus often dominate phytoplankton assemblages in terms of both abundance and contribution to primary productivity [7, 19, 59, 61]. Synechococcus occurs abundantly (102±105 cells mL)1) in the euphotic zone of both

 Present address: Woods Hole Oceanographic Institution, Biology Department, Woods Hole, MA 02543, USA. Correspondence to: U. Christaki; E-mail: urania@¯.ncmr.gr

coastal and open ocean waters. Prochlorococcus, despite its narrower geographical distribution, is usually more abundant than Synechococcus (104 to >105 cells mL)1) and seems to be more important in terms of carbon biomass than Synechococcus on a global scale [reviewed in 40]. Ample information is available on the growth dynamics and the distribution of these organisms throughout the oceans [e.g., 3, 23, 31, 39, 57, 62] and it is well known that both populations are stable over a scale of days [23, 56]. The few existing ®eld studies which considered total production and mortality (usually using community

342

approaches Ð such as size fractionation, dilution, speci®c inhibitors) showed a relatively good balance between growth and apparent losses of Prochlorococcus and Synechococcus, which appear to be in the range of one division per day with a corresponding mortality of about 50% of the stock per day [e.g., 2, 27, 28, 33, 46]. Nevertheless, the above-mentioned community approaches do not distinguish between viral lysis and grazing and they may perturb the balance of the microbial food web. For this reason, the exact nature of picoautotroph losses is yet to be investigated and quanti®ed [reviewed in 40, 41]. Grazing, viral lysis, and sedimentation constitute the three known removal processes for photosynthetic picoplankton. Obviously, each of these processes has speci®c impact on de®ning the role of Prochlorococcus and Synechococcus in carbon dynamics, i.e., whether this important biomass is remineralized via sedimentation and/or viral lysis or partly transferred to higher trophic levels via protozoan grazing. Because of their small size (<2 lm) picoplanktonic organisms are not thought to sink as individuals from surface waters. However, it has been shown that signi®cant sedimentation of these organisms can occur when they become associated with larger particles [58 and references therein]. Viral lysis has been shown to have relatively little impact, at least on Synechococcus [6, 44, 60]. Despite the obvious importance of the grazing pathway, there is little quantitative information on grazing by these protozoa on Synechococcus and Prochlorococcus assemblages. The microbial food web in open marine waters, where an important part of the primary production is due to picoautotrophs, is typically dominated by bacterivorous nano¯agellates which are known to be the major consumer group of heterotrophic bacteria [e.g., 49, 50]. It seems plausible that these ¯agellates are also consumers of autotrophic picoplankton [4, 12, 15, 24, 38]. It was demonstrated that ciliates exhibit a marked preference for Synechococcus over Prochlorococcus, leading to the hypothesis that the major consumers of Prochlorococcus are probably nano¯agellates [10]. In the only existing study devoted to the quanti®cation of two nano¯agellates grazing on autotrophic prokaryotes [21], one nano¯agellate was not ingesting or growing on picoautotrophs. Whereas a second one displayed a substantial division rate when fed on Prochlorococcus, no growth was recorded when Synechococcus was used as a prey. Given the severe lack of quantitative data on the consumption of Synechococcus and particularly on Prochlorococcus by nano¯agellates, the objective of our study was

U. Christaki et al.

to quantify their grazing impact and to examine the potential of ¯agellates to control these populations in nature. For this, prey concentrations from 103 to 106 cells mL)1 were used in short term experiments while in a long term experiment, we studied ¯agellate growth on picoautotrophs. In order to validate laboratory data, one set of data on Synechococcus grazing was obtained during a ®eld study in the oligotrophic northeastern Mediterranean Sea.

Materials and Methods Culture Conditions The Prochlorococcus MED4, approximately 0.65 lm in diameter and originally isolated from the Mediterranean Sea, and the Synechococcus WH8103, originally isolated from the Sargasso Sea, approximately 1.0 lm in diameter, were provided by F. Partensky and D. Vaulot (Station Biologique de Roscoff, France). Cultures were grown in 500-mL sterile ¯asks in PCR-S11 medium in aged seawater, at 18 ‹ 0.5°C under continuous light (15 lmol quanta m)1 s)1), provided by Cool-White ¯uorescent bulbs wrapped in ``moonlight blue'' ®lters (Lee ®lter, Panavision France). Concentrations of Synechococcus and Prochlorococcus in the exponential cultures used for the experiments were 8 ´ 106 mL)1 and 2 ´ 107 mL)1 respectively. The picoautotroph cultures were not axenic and background heterotrophic bacterial densities during early exponential phase were approximately 1 ´ 106 bacteria mL)1 (i.e., always <5% of picoautotroph biomass). The nano¯agellate Pseudobodo sp. (3 lm) was originally isolated in northwest Mediterranean waters and kindly provided by F. Rassoulsadegan (Station Zoologique, Villefranche-sur-Mer, France). Stock cultures were maintained in bacterized wheatgrain medium [e.g., 47]. To obtain exponentially growing cultures, protozoa inocula were transferred into a bacterized yeast extract medium (15±30 mg L)1, see [9] for details). A mixed culture of naturally occurring ¯agellates was obtained as follows. First, four L of seawater from 80 m depth in the oligotrophic Outer Saronicos gulf (Aegean Sea, Eastern Mediterranean) was screened through 5 lm polycarbonate (Nuclepore) ®lters using a gravity ®ltration device in order to allow small ¯agellates to pass without harm (Baileys Plastic Fabrication Ltd., Canada). Then the <5 lm screened water was further ``concentrated'' by gentle ®ltration on 0.6 lm polycarbonate (Nuclepore) ®lters down to 30 mL. This 30-mL sample was used as an inoculum in 0.45 lm ®ltered water from the same area enriched with yeast extract (as above). The mixed ¯agellate culture was not kept in stock cultures but made 4±5 days prior to requirement. Flagellates were grown at 18 ‹ 0.5°C in the dark in order to eliminate the backround natural pico- and nano autotrophs that were initially present in the inoculum and could interefere with our counts. Prior to the preparation of the inoculum, a 30-mL water sample was ®ltered on 0.6 lm black polycarbonate ®lters stained with DAPI and observed by epi¯uorescence microscopy. The population of the ¯agellates was dominated (75%) by bi¯agellated

Prochlorococcus and Synechococcus Consumption ovoid and spherical ¯agellates of small size (<3 lm). The ¯agellate culture obtained after enrichment (see above) was also almost exclusively dominated by the same forms and sizes (2±3.5 lm). It is worth noting that, when Pseudobodo sp. and ``mixed ¯agellates'' were grown on heterotrophic bacteria (5 ´ 106± 2 ´ 107 bacteria mL)1), exponential growth at rates of 0.05 and 0.09 h)1, respectively, continued for up to 96±120 h. Flagellate cultures used in the experiments were removed from exponential cultures when the ¯agellate concentration was 3.4 and 3.9 ´ 104 cells mL)1 in mixed and Pseudobodo cultures respectively and the heterotrophic bacteria used a food source had dropped to ca. 106 ml)1 (Table 1). For conversion to carbon biomass, we used 20 fg cell)1 for heterotrophic bacteria [32], 250 fg cell)1 for Synechococcus [26], 53 fg cell)1 for Prochlorococcus [3], and 180 fg lm3 for ¯agellates [5]. Growth yield of ¯agellates was assumed to be 40% [17].

Grazing experiments In order to study the dynamics of ¯agellate consumption on picoplanktonic prey, 100 mL ¯agellate cultures were spiked with exponentially growing Prochlorococcus or Synechococcus cultures yielding 6 different ®nal concentrations ranging from 103 to 106 prey mL)1 (Table 1). The heterotrophic bacteria from the nano¯agellate and picoautotroph cultures were present in concentrations of about 106 mL)1, which are of the same order to those encountered in oceanic waters (e.g., [4, 12, 29], and this study ®eld data). Table 1.

343 Control solutions of picoautotrophs were prepared by adding the same concentration of autotrophs to 50 mL of 0.2 lm-®ltered nano¯agellate culture. We determined grazing parameters from short-term experiments (12 h) in order to minimize changes in grazer concentration [10, 21]. Samples for picoautotroph (2 mL) counts by ¯ow cytometry were removed after 0, 3, 6, 9, and 12 h and for heterotrophic bacteria at 0 and 12 h (2 mL). Samples for ¯agellate enumeration by epi¯uorescence microscopy (5 mL) were taken at 0 and 12 h. In the treatments where Synechococcus and Prochlorococcus were added at concentrations 105±106 cells mL)1 Ð where we could expect continued ¯agellate growth Ð samples for ¯agellate counting were also taken at 24 and 36 h. Parameters such as growth, grazing, ingestion, and clearance rates for prey (Synechococcus, Prochlorococcus, and heterotrophic bacteria) and predators were calculated for each time point using the equations devised by Frost [18] and modi®ed by Heinbokel [22]. In order to validate ¯ow cytometry results, we also used an independent method for studying dynamics of food uptake. In experimental bottles containing Synechococcus as prey (which Ð unlike Prochlorococcus Ð can be easily identi®ed inside ¯agellate cells because of its orange ¯uorescence), ingestion rates were determined in 5 mL samples removed after gentle mixing at 10min interval for 60 min. Samples were preserved with 1% glutaraldehyde ®nal concentration, ®ltered on 0.6 lm black Nuclepore ®lters, and stained with DAPI [43]. Synechococcus cells inside individual ¯agellates were counted with an Olympus AX70 epi¯uorescence microscope. Auto¯uorescence of ingested Synechococcus was distinguished under blue light excitation. For

Initial concentrations (cells mL)1) of prey and predators in experimental bottles in grazing and growth experiments

Grazing Experiment (cells mL)1) Synechococcus 0.98 ´ 103 9.9 ´ 103 11.7 ´ 104 4.9 ´ 105 9.9 ´ 105 1.3 ´ 106 Synechococcus 2.1 ´ 103 10.0 ´ 103 8.1 ´ 104 3.4 ´ 105 6.6 ´ 105 1.5 ´ 106 Growth Experiment** (cells mL-1) Synechococcus 8.2 ´ 106

Prochlorococcus 2.3 ´ 103 8.0 ´ 103 11.8 ´ 104 5.5 ´ 105 9.8 ´ 105 3.2 ´ 106 Prochlorococcus 4.9 ´ 103 9.5 ´ 103 7.7 ´ 104 3.8 ´ 105 6.9 ´ 105 2.9 ´ 106

Prochlorococcus 2.2 ´ 107

Het. Bacteria 1.32 ´ 106



denotes experiment duration of 12 h. denotes experiment duration of 72 h.



Het. Bacteria 0.87 ´ 106

Het. bacteria 0.95 ‹ 0.38 ´ 106 Pseudobodo sp. 3.9 ´ 104 Het. Bacteria 1.28 ‹ 0.14 ´ 106 Mixed ¯agellates 3.4 ´ 104 Psedobodo sp. 7.2 ´ 103 Mixed ¯agellates 4.1 ´ 103 Pseudobodo sp. 6.2 ´ 103 Mixed ¯agellates 3.2 ´ 103

344 each sample, the food vacuole contents of 200 ¯agellates were enumerated. Ingestion rates were calculated as the slope of the linear proportion of the uptake curve (30 min) of average prey number per cell versus time.

Growth of Flagellates in Picoprocaryote Cultures In a second series of experiments, inocula from the early stationary phase of ¯agellate cultures were added to exponentially growing cultures of Prochlorococcus MED4 (2 ´ 107 cells mL)1) and Synechococcus WH8103 (8 ´ 106 cells mL)1) yielding a ®nal concentration of 103 ¯agellates mL)1 (Table 1). All experimental bottles were prepared in duplicate. The initial abundance of heterotrophic bacteria from the ¯agellate and autotroph cultures was about 1 ´ 106 cells mL)1. Samples were taken at 0, 12, 24, 36, 48, and 72 h from each of the ¯asks for protozoan counts (5±10 mL).

Field Data One set of data on Synechococcus grazing drawn from a ®eld study was used to validate laboratory experimental data. In order to investigate the importance of periodicity of grazer induced mortality, samples were collected from 5 depths in the 0±100 m water column (5, 20, 50, chlorophyll maxÐbetween 80 and 89 mÐand 100 m) every 6 h over two separate 54-h cycles (90 samples in all) at 2 stations (39°25¢ N, 25°44¢ E and 39°58¢ N, 25°45¢ E, depth 400 and 100 m, irrespectively) in the north Aegean (northeast Mediterranean Sea) in September 1999. The water column was strati®ed; the water temperature was 20±23°C in the surface layer and was around 15°C at 100 m depth. The north Aegean is oligotrophic and the planktonic food web is characterized by low chlorophyll low bacterial production, and overall dominance of bacterial biomass [11]. Samples for nano¯agellate (35 mL) and Synechococcus (10 mL) counts were preserved with 1% glutaraldehyde ®nal concentration and ®ltered on black polycarbonate (Nuclepore) ®lters (0.6 and 0.2 lm respectively), stained with DAPI [43], stored at )20 °C, and counted by epi¯uorescence microscopy. A minimum of 100 heterotrophic nano¯agellates were examined per ®lter and the presence of Synechococcus in food vacuoles was examined using blue light excitation. Ingestion rates based on food vacuole content were calculated applying the digestion rate of nano¯agellate on Synechococcus of 1.1% cell content min)1 determined by Dolan and SÏimek [14] using a mixed ¯agellate culture prepared by size fractionation and enrichment, as in our study. Total Synechococcus consumption for the whole water column (1010 Synechococcus m)2 h)1) was obtained by depth integration (0±100 m) of grazing measured at the 5 different depths for each sampling time.

Flow Cytometry Analysis (FCM) Samples of two 2 mL were preserved with 2% paraformaldehyde [55] and stored in liquid nitrogen before their ¯ow cytometric

U. Christaki et al. analysis. Synechococcus and Prochlorococcus counts were performed with a FACSCalibur ¯ow cytometer (Becton Dickinson, San Jose, CA) equipped with an air-cooled argon laser (488 nm, 15 mW). Synechococcus excited at 488 nm were detected and enumerated according to their right-angle light scattering properties (RALS) and their orange (585/42 nm) and red ¯uorescence (>650 nm) emissions related to phycoerythrin and chlorophyll pigments, respectively. For Prochlorococcus cells, RALS and red ¯uorescence were use for discriminating and counting. In order to calibrate the ¯ow cytometer, ¯uorescent beads (1.002 lm; Polysciences Inc., Warrington, PA) were systematically added to each sample. The precise volume analyzed (between 250 and 500 ll) and subsequent estimations of cell concentrations were calculated by measuring the remaining volume and subtracting it from the initial subsample volume (1 mL). Samples for bacterial counts were stained with SYBR Green I [35], a speci®c dye for nucleic acids (Molecular Probes). Stained bacterial cells, excited at 488 nm, were enumerated using the FACSCalibur cytometrer according to their right-angle light scatter (RALS) and green ¯uorescence (FL1, collected at 530/30 nm) from their stained nucleic acids. Flow cytometer calibration and cell concentrations were performed as for phytoplankton (see above). The number of picoautotrophs was deducted from the bacterial total numbers.

The Analysis The Gabriel's approximate test [54, p. 508] was used to check differences between slopes and intercepts of linear regression models. This procedure is performed to test if two or more regression coef®cients (slopes) or intercepts are statistically different. In the present study, regression coef®cients or intercepts whose 95% con®dence intervals, calculated by the GT-2 method, did not overlap were considered signi®cantly different [54].

Results Both Pseudobodo sp. and ``mixed ¯agellates'' actively ingested Synechococcus and Prochlorococcus (Figs. 1a, 1b, 1c); the grazing parameters were different for the two types of prey (Table 2). The ingestion rate of ¯agellates, from Synechococcus counts in experimental bottles, obtained by ¯ow cytometry, ranged from 0.0002 to 2.7 Syn ¯agellate h)1. Ingestion rates increased linearly for prey concentrations between 102 and 5 ´ 105 Syn mL)1, while above 5 ´ 105 Syn mL)1 the ingestion rate decreased (Fig. 1a). The relationships between ¯agellate ingestion rates (IR) and Synechococcus concentration, up to 5 ´ 105 Syn ml)1, for both Pseudobodo sp. and mixed ¯agellate cultures, were both described by the linear model (Fig. 1b). Although the ranges of ingestion rates by both predator

Prochlorococcus and Synechococcus Consumption

345

b Fig. 1. (a) Ingestion rate of Pseudobodo sp. (circles) and mixed ¯agellate culture (triangles) plotted against Synechococcus mean cell concentration for 3-h sampling intervals. Closed and open symbols represent results from ¯ow cytometric counts and microscopic counts of Synechococcus in ¯agellate food vacuoles, respectively (see text for details). (b) Ingestion rate of Pseudobodo sp. (circles) and mixed ¯agellate culture (triangles) plotted against Synechococcus concentration up to 5 ´ 105 cells mL)1. Linear regression models (continuous lines) were ®t to the solid data points; open points were not considered as part of the linear response. Top (thick) line represent linear regression ®tted to solid circles: y = 0.056 + 0.0051x, r2 = 0.916, F = 197.31, p < 0.001. Bottom (thin) line represent linear regression ®tted to solid triangles: y = ) 0.00130 + 0.0035x, r2 = 0.903, F = 207.18, p < 0.001. Dotted lines represent the 95% con®dence intervals of the regressions. (c) Ingestion rate of Pseudobodo sp. (circles) and mixed ¯agellate culture (rectangles) plotted against Prochlorococcus concentration up to 3 ´ 106 cells mL)1. Top (thick) line represent linear regression ®tted to circles: y = ) 0.1324 + 0.0019x, r2 = 0.903x, F = 224.92, p < 0.001. Bottom (thin) line represent linear regression ®tted to rectangles: y = 0.0115 + 0.0012x, r2 = 0.922, F = 395.02, P < 0.001. Dotted lines represent the 95% con®dence intervals of the regressions.

types were of similar magnitude (Table 2), the slopes of the two ¯agellate cultures (up to 5 ´ 105 Syn mL)1, Fig. 1b) were signi®cantly different (p < 0.05). Ingestion rates obtained by ¯ow cytometry were similar to those obtained independently by direct microscopic observation of Synechococcus in ¯agellate food vacuoles

(Figs. 1a, 1b) over short term incubations. The ingestion rate of ¯agellate grazing on Prochlorococcus increased linearly over the whole range of prey density used (from 103 to 106 Proc mL)1, Fig. 1c). Overall, the ¯agellate ingestion rate on Prochlorococcus ranged from 0.001 to 6.7 Proc ¯agellate)1 h)1. As with Synechococcus, although the ranges of ingestion rates by the two predator types were similar, the slopes for the different ¯agellate cultures grazing on Prochlorococcus were signi®cantly different (p < 0.05), being steeper for Pseudobodo sp. than for the mixed ¯agellate culture (Fig. 1c). Interestingly, for prey concentrations up to 5 ´ 105 mL)1, the ingestion rate on Synechococcus was signi®cantly higher than that on Prochlorococcus for both predator cultures tested, i.e., a statistical difference (p < 0.05) was found between the slopes estimated for Synechococcus and Prochlorococcus. Clearance rates by ¯agellates were from 0.4 to 10.9 nL ¯agellate)1 h)1 for Synechococcus and from 0.16 to 3.4 nL ¯agellate)1 h)1 for Prochlorococcus (Figs. 2a, 2b, Table 2). Moreover, clearance was inversely related to Synechococcus concentration, with higher rates at lower Synechococcus concentrations. This relationship was more pronounced for Pseudobodo sp. (see regression lines in Fig. 2a). In contrast, for both ¯agellate cultures, clearance of Prochlorococcus showed no relationship with the concentration of Prochlorococcus (Fig. 2b).

346

U. Christaki et al.

Table 2. Grazing parameters (range and mean in parentheses) from the experiments with Pseudobodo sp. (top), from a culture of mixed ¯agellates feeding on Prochlorococcus and Synechococcus (middle), and from the ®eld measurements based on Synechococcus presence in food vacuoles of ¯agellates at 2 oligotrophic stations in the NE Mediterranean sampled every 6 h over two separate 54 h cycles (bottom) Laboratory experiment Pseudobodo sp. Prey concentration (cell mL)1) Ingestion rate (cell ¯agellate)1 h)1) Grazing rate (h)1) Clearance rate (nl ¯agellate)1 h)1) Mixed ¯agellate culture Prey concentration (cell mL)1) Ingestion rate (cell ¯agellate)1 h)1) Grazing rate (h)1) Clearance rate (nl ¯agellate)1 h)1) Field data Mixed ¯agellates Prey concentration (cell mL)1) Ingestion rate (cell ¯agellate)1 h)1) Grazing rate (h)1) Clearance rate (nl ¯agellate)1 h)1)

Synechococcus

Prochlorococcus

0.15 ´ 103±1.3 ´ 106 0.0005±2.7 (1.0) 0.016±0.43 (0.20) 0.40±10.9 (5.11)

1.1 ´ 103±2.8 ´ 106 0.001±6.7 (1.86) 0.01±0.25 (0.07) 0.16±3.37 (1.6)

1.2 ´ 103±1.4 ´ 106 0.0002±2.4 (0.74) 0.009±0.22 (0.10) 0.66±5.48 (3.17)

1.9 ´ 103±2.7 ´ 106 0.002±6.4 (1.03) 0.02±0.07 (0.04) 0.79±2.69 (1.35)

Synechococcus spp. 0.7 103±5.5 ´ 104 0.008±0.15 (0.055) Ð 0.4±17 (4.9)

Flagellate Growth The number of ¯agellates did not change signi®cantly during the ®rst 12 h in all grazing experimental bottles and remained around 3±4 ´ 104 cells)1 mL)1. Samples for ¯agellate counting taken at 24 and 36 h in the treatments where Synechococcus and Prochlorococcus were added at concentrations of 105±106 cells mL)1 showed a decrease relative to zero time (2±2.5 ´ 104 cell mL)1). Microscopical observation of the mixed culture showed no apparent shift in the population composition which was dominated by the same forms of ¯agellates, i.e., bi¯agellated ovoid and spherical cells of small size (<3 lm). Heterotrophic bacteria from the nano¯agellate exponential cultures and picoautotroph cultures were present in concentrations of about 106 mL)1 which approach natural abundances (Table 1). Such concentrations, although they could not support ¯agellate growth, were exploitable by ¯agellates and thus we calculated their contribution to ¯agellate feeding in our experiments from the difference of bacterial numbers in control and experimental bottles (see Materials and Methods). The number of bacteria ingested by ¯agellates

during 12 h incubation in the experimental bottles was 1.64 ‹ 1.19 bact ¯)1 h)1, and the clearance rate was 1.37 ‹ 0.8 nl ¯)1 h)1 (Table 2) and represented 25 ‹ 10% of the body carbon of ¯agellate d)1. These ®gures are close to bacterial consumption values exerted by small ¯agellates in marine waters [e.g., 4, 11]. Finally, in the experiments where ¯agellates were feeding almost exclusively on picoautotrophs (i.e., ¯agellate inocula were added to picoautotroph cultures and background bacterial carbon was <5% of total prokaryotic carbon in the cultures, see Methods), the ``mixed ¯agellate culture'' showed some growth in the Synechococcus culture during the ®rst 24 h (l = 0.036 h)1, Fig. 3a), while a detectable increase was also observed from 36 to 48 h of incubation in the Prochlorococcus culture (l = 0.016 h)1). The Pseudobodo sp. concentration decreased in both cultures during the ®rst 24 h and then remained at low levels (around 2 ´ 10 cells mL)1) until the end of the experiment (72 h, Fig. 3b). In the cultures containing Synechococcus, microscopic observation revealed that the cyanobacteria were actively ingested by ¯agellates at least during the ®rst 24 h of incubation, after which time ¯agellate cells, for

Prochlorococcus and Synechococcus Consumption

347

Field Data In 90 samples from the 0±100 m water column at two northeast Mediterranean stations, cells <3 lm dominated the heterotrophic nano¯agellate assemblage and abundance varied from 0.1 to 2.6 ´ 103 mL)1. Synechococcus sp. concentration was from 0.07 to 5.5 ´ 104 mL)1 and heterotrophic bacteria from 0.3 to 0.8 106 mL)1. Concentrations of Synechococcus cells in division were the highest at dusk, during both diel studies (Fig. 4a). However, total heterotrophic nano¯agellates and total Synechoccocus numbers did not show a clear diel pattern (Figs. 4b, 4d). Heterotrophic nano¯agellate ingestion rates were related to Synechococcus abundance in the water (Pearson adjusted r2 = 0.40, p < 0.0001). Nano¯agellate feeding rate showed a clear diel rhythm with the highest consumption at night, and lowest rates toward the end of daylight (Fig. 4c). Clearly, the nano¯agellate grazing pattern was the inverse of the variation in percentage of dividing Synechococcus (Fig. 4a, 4c). Summing up the depth integrated values (0±100 m) of Synechococcus consumption for all sampling times (every 6 h over 54 h, Fig. 4c), we found that nano¯agellates daily consumed 15% (station 1) and 10% (station 2) of the Synechococcus standing stock.

Discussion

Fig. 2. (a) Clearance rate by Pseudobodo sp. (circles) and mixed ¯agellate culture (rectangles) plotted against Synechococcus concentration. Top (thick) line represent linear regression ®tted to circles: y = 6.9802 ) 0.0046x, r2 = 0.56, F = 29.217, p < 0.001. Bottom (thin) line represent linear regression ®tted to squares: y = 3.7387 - 0.0016x, r2 = 0.313, F = 15.239, P < 0.001. (b) Clearance rate by Pseudobodo sp. (circles) and mixed ¯agellate culture (squares) plotted against Prochlorococcus concentration.

unknown reasons, showed signs of deterioration, losing typical morphology and ¯agella.

Our results clearly showed that for prey concentrations in the range of 103±105, which covered the whole range of abundance encountered in aquatic systems, ingestion and clearance rates on Synechococcus were signi®cantly higher than those on Prochlorococcus. In a ``typical'' marine environment, which might contain 103 ¯agellates mL)1 and 104 Synechococcus mL)1 (e.g., see Table 3), according to our experimental data, ¯agellates consume approximately 16% of the Synechococcus standing stock per day. In a recent study based on vacuole content analysis in a longitudinal transect of the Mediterranean Sea, we found that ¯agellates consumed 13% of the Synechococcus standing stock d)1 (from 0.4 to 45 % at 4 different depths of the euphotic zone [12]). Similar results were obtained in the present study during diel cycles at the two north Aegean stations (10 and 15% of the Synechoccocus standing stock d)1) and by Dolan and SÏimek [15] during a diel study in surface coastal waters of a northwest Mediterranean station (14% of the Synechococcus standing stock d)1).

348

U. Christaki et al.

Fig. 4. Field data: (a) percent of Synechococcus cells in division (doublets), mean values for 5 sampling depths; (b), (c), (d) depth integrated values 0±100 m of: (b) Synechococcus abundance, (c) Synechococcus ingested by heterotrophic ¯agellates h)1, and (d) heterotrophic nano¯agellate (HNAN) abundance. Dark bars denote night-time hours.

Fig. 3. Changes in cell concentrations of (a) mixed ¯agellates and (b) Pseudobodo sp. feeding on cultures of Synechococcus (open symbols) and Prochlorococcus (closed symbols).

One important dynamic feature of picoautotrophs in marine environments is that they show higher division rates at dusk [23, 56, 57, this study]. Despite strong diel rhythms, the stability of prokaryotic populations over a scale of days indicates that, in nature, losses do not take place uniformly over a diel cycle [23, 28, 57]. Dolan and SÏimek [15] and our own ®eld data showed a clear diel variation of Synechococcus consumption by nano¯agellates with enhanced feeding rates after cell division [see also 34, 57]. The total Synechoccocus number varied little and did not follow any clear pattern, which implies that nano¯agellates might preferentially remove smaller newly produced Synechococcus cells. Thus, small protists seem to optimize selection of preferred prey size by removing the ``smallest'' Synechococcus and the ``largest'' heterotrophic bacteria [e.g., 13, 20, 50]. For bacterial populations, the ecological implication of selective removal of bacteria is community composition control [e.g., 25, 42, 53]. Although the ecological implication for Synechococcus is not known, one could suppose that by removing smaller newly produced Synechococcus cells, nano¯agellates optimize their role as Synechococcus consumers.

We compared the ingestion rates of the present study, plotted against Synechococcus concentration, with those from other studies (Fig. 5, Table 3). With the exception of the study by Dolan and SÏimek [15], available information suggests that nano¯agellate ingestion rates increase with increasing picoplankton concentrations [4, 8, 11, 29, 26, 30, 37, this study]. Most of the ®eld data considered in this study testify to such a relationship and fall well between the 95% con®dence intervals of the present experimental data (Fig. 5, Table 3). Comparison with ®eld data, such as that shown in Fig. 5 for Synechococcus, could not be undertaken for Prochlorococcus since in the few ®eld studies dealing with its mortality [28, 33, 34, 46], the number of predators has not been considered. According to our experimental data, in a ``typical'' oceanic environment, which might contain 103 ¯agellates mL)1 and 5 ´ 104 Prochlorococcus mL)1, ¯agellates would consume <5% of the Prochlorococcus stock d)1. This ®gure is close to the estimated consumption of Prochlorococcus by nano¯agellates, based on standing stocks and measured grazing rates of other picoplankton groups in the Mediterranean Sea (mean 6% of the standing stock d)1 [12]). Thus, although Prochlorococcus consumption data, obtained by ¯ow cytometry counts, could not be validated with microscopic counts and/or ®eld data, we believe that the good ®t of ¯ow cytometry results with food vacuole content observations and with ®eld data for Synechococcus (Figs. 1a, 5) indirectly supports the results on Prochlorococcus. It should also be mentioned that phagotrophic protozoa (¯agellates, ciliates) and Synechococcus are abundant in

Prochlorococcus and Synechococcus Consumption Table 3.

Source [4]

Details of data presented in Fig. 4

Location

Coastal, Vineyard Sound, surface water (5 combined depths) [29] Coastal, northern Baltic, surface water (1 depth) [15] Coastal NW Mediterranean station, diel study, surface water (1 depth) [12] Longitudinal Mediterranean transect, euphotic zone (4 depths) This study Diel studies, NE Mediterranean sea (5 depths) This study Laboratory experiment (Pseudobodo sp., mixed ¯agellate culture feeding on Synechococcus WHS 103) a

349

Flagellates (103 cells mL)1)

Synechococcus Ingestion rate (104 cells (Syn ¯ag)1 )1 mL ) h)1) n

Metabolic inhibitors

1±2

0.6±4.4

Size fractionation

0.7±3.4

Food vacuole content analysis

Technique

z

0.07±0.22

4

0.7±50

0.02±2.6

11

1±2

1.7±3.8

0.04±0.35

28

Food vacuole content analysis

0.05±3.0

0.02±3.0

0.004±0.12

36

Food vacuole content analysis Flow cytometry, food vacuole content analysis

0.1±2.6

0.07±5.5

0.008±0.15

90

30±40

0.02±50

0.0005±2.68

64

n = number of observations plotted in Fig. 4. Ingestion rates on Synechococcus from [15] were extrapolated from their Fig. 2.

surface waters, while Prochlorococcus extends deep in the water column and often presents a sharp maximum concentration of the order of 105 cells mL)1 near the bottom of the euphotic zone where abundance of protozoa declines signi®cantly [12, 36, 40, 41]. Although grazing parameters of both predator types used in this study had very similar ranges (Table 2), the difference between slopes of ingestion rates obtained at all

cases (Figs. 1b, 1c) indicated that ¯agellates exhibited different grazing models and behavior [e.g., 1, 21]. Flagellates used in our experiments were grown on bacterial prey and were supposedly bacterivores; however, our results showed that these ¯agellates were active predators of both Synechococcus and Prochlorococcus. The zoo¯agellates of the genus Pseudobodo are ubiquitous free-living forms abundant in marine environments [e.g., 16, 45] and

Fig. 5. Flagellate ingestion rates plotted against Synechococcus concentration; data from different ®eld studies superimposed on data from the present study (see also Table 3). The regression line and con®dence limits (95%) correspond to the results of the present experimental study (squares) for concentrations up to 5 ´ 105 Syn mL)1. Note that the x-axis was cut off at 1 ´ 105 for better visualization of the results.

350

although considered as bacterivores they have been also found to actively feed on picoautotroph prey [38]. It should be also stressed that the inoculum for our mixed ¯agellate culture was obtained from an open-sea oligotrophic area (80 m) where both Synechococcus and Prochlorococcus are common. Moreover, in both coastal and oceanic environments the nano¯agellate populations are usually dominated by small bacterivorous ¯agellates around 3 lm [e.g., 4, 11, 12, 15, 48] similar to those used in our experimental study. In our experiments for concentrations from 105 to 106 Syn mL)1, the Synechococcus carbon ingested by ¯agellates would theoretically result in 1±2 generations per day (assuming 40% growth yield for ¯agellates). Equally, for Prochlorococcus concentrations >106, we would theoretically expect one or more generations of ¯agellates per day. Contrary to these expectations, ¯agellates did not grow in our experiments. Several authors have reported moderate growth and/or survival of ¯agellates and ciliates on Synechococcus [4, 10, 51, 52], while in a recent study Boenigk et al. [1] showed that two freshwater ¯agellates excreted a Synechococcus species a few minutes after ingestion and that they were not able to grow when fed on this species. Interestingly, the only comparable data about the food value of Prochlorococcus for other bacterivoresÐnamely the scuticociliate Uronema sp. [10]Ðsuggested that Prochlorococcus inhibited or interfered with cell division, while it permitted a moderate growth for the omnivorous ciliate Strombidium sulcatum. Interestingly enough, in the only existing study dealing with a ¯agellate (Picophagus ¯agellatus) capable to grow on Prochlorococcus [21], the calculated ingestion rate (on average 1 Prochlorococcus ¯agellate)1 h)1) determined for a single prey concentration (7 ´ 105 Prochlorococcus mL)1) was almost identical to ours (1.19 Prochlorococcus ¯agellate)1 h)1 according to the equation determined in our study (see legend of Fig. 1). In conclusion, although the scarce information from ®eld studies indicates that the picoautotrophs constitute only secondary food for ¯agellates [4, 12,15, 28], the existence of at least one Prochlorococcus feeder is now documented [21]. In helping to improve the understanding of photosynthetic prokaryote's loss processes in situ, future studies should focus on ¯agellate community composition in waters dominated by picoautotrophs.

U. Christaki et al.

Acknowledgments This study was supported by the Greek±French collaboration programme PLATON and the the EU project KEYCOP (MAS-3-CT97-0148). We are grateful to Dr. F. Rassoulzadegan for providing the Pseudobodo sp. culture, to Drs. D. Vaulot and F. Partensky for providing the Synechococcus and Prochlorococcus cultures, and to Mme F. Legall for detailed explanations on culture conditions of picoautotrophs. We thank Dr. J.R. Dolan for comments on the manuscript.

References Boenigk J, Matz C, JuÈrgens K, Ardnt H (2001) The in¯uence of preculture conditions and food quality on the ingestion and digestion of three species of heterotrophic nano¯agellates. Microb Ecol 42:168±176 2. Campbell L, Carpenter EJ (1986) Estimating the grazing pressure of heterotrophic nanoplankton on Synechococcus spp. using the sea water dilution and selective inhibitor techniques. Mar Ecol Prog Ser 33:121±129 3. Campbell L, Nolla HA, Vaulot D (1994) The importance of Prochlorococcus to community structure in the central Paci®c Ocean. Limnol Oceanogr 39:954±96 4. Caron DA, Lim EL, Miceli G, Waterbury JB, Valois FW (1991) Grazing and utilization of chroococcoid cyanobacteria and heterotrophic bacteria by protozoa in laboratory cultures and a coastal plankton community. Mar Ecol Prog Ser 76:205±217 5. Caron DA, Dam HG, Kremer P, Lessard EJ, Madin LP, Malone TC, Napp JM, Peele ER, Roman MR, Youngbluth MJ (1995) The contribution of microorganisms to particulate carbon and nitrogen in surface waters of the Sargasso Sea near Bermuda. Deep-Sea Res 42:943±972 6. Carza DR, Suttle CA (1998) The effect of cyanophages on the mortality of Synechococcus spp. and selection of UV resistant viral communities. Microb Ecol 36:281±292 7. Chisholm SW, Olson RJ, Zettler ER, Waterbury JB, Goericke R, Welschmeyer N (1988) A novel free-living prochlorophyte occurs at high cell concentrations in the oceanic euphotic zone. Nature 334:340±343 8. Choi JW (1994) The dynamic nature of protistan ingestion response to prey abundance. J Euk Microbiol 41:137±146 9. Christaki U, Dolan JR, Pelegri S, Rassoulzadegan F (1998) Consumption of picoplankton-size particles by marine ciliates: effects of physiological state of the ciliate and particle quality. Limnol Oceanogr 43:458±464 10. Christaki U, Jacquet S, Dolan JR, Vaulot D, Rassoulzadegan F (1999) Differential grazing and growth on Prochlorococcus and Synechococcus by two contrasting ciliates. Limnol Oceanogr 44:52±61 1.

Prochlorococcus and Synechococcus Consumption 11. Christaki U, Van Wambeke F, Dolan JR (1999) Nano¯agellates (mixotrophs, heterotrophs and autotrophs) in the oligotrophic eastern Mediterranean: Standing stocks, bacterivory and relationships with bacterial production. Mar Ecol Prog Ser 181:297±307 12. Christaki U, Giannakourou A, Van Wambeke F, GreÂgori G (2001). Nano¯agellate predation on auto- and heterotrophic picoplankton in the oligotrophic Mediterranean Sea. J P1 Res 23:1297±1310 13. Chrzanowski TH, SÏimek K (1990) Prey-size selection by freshwater ¯agellated protozoa. Limnol Oceanogr 35:1429± 1436 14. Dolan JR, SÏimek K (1998) Ingestion and digestion of an autotrophic picoplankter, Synechococcus, by a heterotrophic nano¯agellate, Bodo saltans. Limnol Oceanogr 43:1740± 1746 15. Dolan JR, SÏimek K (1999) Diel periodicity in Synechococcus populations and grazing by heterotrophic nano¯agellates: Analysis of food vacuole contents. Limnol Oceanogr 44: 1565±1570 16. Fenchel T (1982) Ecology of heterotrophic micro¯agellates I. Some important forms and their functional morphology. Mar Ecol Progr Ser 8:211±223 17. Fenchel T (1986) The ecology of heterotrophic ¯agellates. Adv Microb Ecol 9:57±95 18. Frost BW (1972) Effects of size and concentration of food particles on the feeding behavior of the marine planktonic copepod Calanus Paci®cus. Limnol Oceanogr 17:805±815 19. Glover HE, Campbell L, Prezelin BB (1986) Contribution of Synechococcus spp. to size-fractionated primary productivity in three water masses in the Northwest Atlantic Ocean. Mar Biol 91:193±203 20. Gonzalez JM, Iriberri J, Egea L, Barcina I (1990) Differential rates of digestion of bacteria by freshwater and marine phagotrophic protozoa. Appl Environ Microbiol 56:1851± 1857 21. Guillou L, Jacquet S, Chretiennot-Dinet MJ, Vaulot D (2001) Grazing impact of two heterotrophic ¯agellates on Prochlorococcus and Synechococcus. Aquat MicrobEcol 26: 201±207 22. Heinbokel JF (1978) Studies on the functional role of Tintinnids in the Southern California Bight. I. Grazing and growth rates in laboratory cultures. Mar Biol 47:177±189 23. Jacquet S, Lennon JF, Marie D, Vaulot D (1998) Picoplankton population dynamics in coastal waters of the northwestern Mediterranean Sea. Limnol Oceanogr 43:1916±1931 24. Johnson PW, Xu HS, Sieburth JMcN (1982) The utilization of chroococcoid cyanobacteria by marine protozooplanktres but not by calanoid copepods. Ann Inst Oceanogr, Paris, 58:297±308 25. JuÈrgens K, Pernthaler J, Schalla S, Amman R (1999) Morphological and compositional changes in a planktonic bacterial community in response to enhanced protozoan grazing. Appl Environ Microbiol 65:1241±1250

351 26. Kana T, Glibert PM (1987) Effect of irradiances up to 2000 lE m)2s)1 on marine Synechococcus WH 7803±I. Growth, pigmentation and cell composition. Deep Sea Res 34:479±516 27. Kudoh S, Kanda J, Takahashi M (1990) Speci®c growth rates and grazing mortality of chroococcoid cyanobacteria Synechococcus spp. in pelagic surface waters in the sea. J Exp Mar Biol Ecol 142:201±212 28. Kuipers BR, Witte HJ (2000) Prochlorophytes as secondary prey for heterotrophic nano¯agellates in the deep chlorophyll maximum layer of the (sub)tropical North Atlantic. Mar Ecol Prog Ser 204:53±63 29. Kuosa H (1991) Picoplanktonic algae in the northern Baltic sea: seasonal dynamics and ¯agellate grazing. Mar Ecol Prog Ser 73:269±271 30. Landry MR, Hass LW, Fagerness VL (1984) Dynamics of microbial plankton communities: Experiments in Kaheone Bay, Hawaii. Mar Ecol Prog Ser 16:127±133 31. Landry MR, Kirshtein J, Costantinou J (1996) Abundances and distributions of picoplankton populations in the central equatorial Paci®c from 12° N to 12° S, 140° W. Deep Sea Res 43:871±890 32. Lee S, Furhman JA (1987) Relatioships between biovolume and biomass of naturally derived marine bacterioplankton. Appl Environ Microbiol 53:1298±1303 33. Liu H, Campbell L, Landry MR (1995) Growth and mortality rates of Prochlorococcus and Synechococcus measured with a selective inhibitor technique. Mar Ecol Prog Ser 116:227±287 34. Liu H, Nolla HA, Campbell L (1997) Prochlorococcus growth rate and contribution to primary production in the equatorial and subtropical North Paci®c Ocean. Mar Ecol Prog Ser 12:39±47 35. Marie D, Partensky F, Jacquet S, Vaulot D (1997) Enumeration and cell cycle analysis of natural populations of marine picoplankton by ¯ow cytometry using the nucleic acid stain SYBR Green I. Appl Env Microbiol 63:186±193 36. Moore LR, Goericke R, Chisholm SW (1995) Comparative physiology of Synechococcus and Prochlorococcus: in¯uence of light and temperature on growth, pigments, ¯uorescence and absorptive properties. Mar Ecol Prog Ser 116:259±275 37. Nagata T (1988) The micro¯agellate±picoplankton food linkage in the water column of Lake Biwa. Limnol Oceanogr 33:504±517 38. Parslow JS, Doucette GJ, Taylor FJR, Harrison PJ (1986) Feeding by the zoo¯agellate Pseudobodo sp. on the picoplanktonic prasinomonad Micromonas pusilla. Mar Ecol Prog Ser 29:237±246 39. Partensky F, Blanchot J, Lantoine F, Neveux J, Marie D (1996) Vertical structure of picophytoplankton at different trophic sites of tropical northeastern Atlantic Ocean. Deep Sea Res 43:1191±1213. 40. Partensky F, Blanchot J, Vaulot D (1999a) Differential distribution and ecology of Prochlorococcus and Synechococcus in oceanic waters: a review. Bull Inst Oceano, Monaco, n° special 19:457±475

352 41. Partensky F, Hess WR, Vaulot D (1999b) Prochlorococcus, a marine photosynthetic procaryote of global signi®cance. Microb Mol Biol Rev 63:106±127 42. Pernthaler J, Posch T, SÏimek K, Vrba J, Amann R, Psenner R (1997) Contrasting bacterial strategies to coexist with a ¯agellate predator in an experimental microbial assemblage. Appl Environ Microbiol 63:596±601 43. Porter KG, Feig YS (1980) The use of DAPI for identifying and counting aquatic micro¯ora. Limnol Oceanogr 25:943±948 44. Proctor LM, Fuhrman JA (1990) Viral mortality of marine bacteria and cyanobacteria. Nature 343:60±62 45. Rassoulzadegan F, Laval-Peuto M, Sheldon RW (1988) Partinioning of the food ration of marine ciliates between picoand nanoplankton. Hydrobiologia 159:75±88 46. Reckermann M, Veldhuis MJW (1997) Trophic interactions between picophytoplankton and micro- and nanozooplankton in the western Arabian Sea during the NE Monsoon 1993. Aquat Microb Ecol 12:263±273 47. Rivier A, Brownlee DC, Sheldon RW, Rassoulzadegan F (1985) Growth of microzooplankton: a comparative study of bacterivorous zoo¯agellates and ciliates. Mar Microb Food Webs 1:51±60 48. Sanders RW, Berninger UG, Lim EL, Kemp PF, Caron DA (2000) Heterotrophic and mixotrophic nanoplankton predation on picoplankton in the Sargasso Sea and Georges Bank. Mar Ecol Prog Ser 192:103±118 49. Sherr EB, Sherr BF, Paffenhofer GA (1986) Phagotrophic protozoa as food for metazoans: a ``missing'' trophic link in marine pelagic food webs. Mar Microb Food Webs 2:61±80 50. Sherr BF, Sherr EB, McDaniel J (1992) Effect of protistan grazing on the frequency of dividing cells in bacterioplankton assemblages. Appl Environ Microbiol 58:2381±2385 51. SÏimek K, Bodkova J, Macek M, Nedoma J, Psenner R (1995) Ciliate grazing on picoplankton in a eutrophic reservoir during the summer phytoplankton maximum: A study at the species and community level. Limnol Oceanogr 40:1077± 1090

U. Christaki et al. 52. SÏimek K, Macek M, Pernthaler J, Straskrabova V, Psenner R (1996) Can freshwater ciliates survive on a diet of picoplankton? J Pl Res 18:597±613 53. SÏimek K, Kojecka P, Nedoma J, Hartman P, Vrba J, Dolan JR (1999) Shifts in bacterial community composition associated with different microzooplankton size fractions in a eutrophic reservoir. Limnol Oceanogr 44:1634±1644 54. Sokal RR, Rohlf FJ (1995). Biometry 3rd ed. WH Freeman and Company, New York 55. Trousselier M, Courties C, Vaquer A (1993) Recent applications of ¯ow cytometry in aquatic microbial ecology. Biol Cell 78:111±121 56. Vaulot D, Marie D, Olson RJ, Chisholm SW (1995) Growth of Prochlorococcus, a photosynthetic procaryote, in the Equatorial Paci®c Ocean. Science 268:1480±1482 57. Vaulot D, Marie D (1999) Diel variability of photosynthetic picoplankton in the Equatotial Paci®c. J Geophys Res 104:3297±3310 58. Waite AM, Sa® KA, Hall JA, Nodder SD (2000) Mass sedimentation of picoplankton embedded in organic aggregates. Limnol Oceanogr 45:87±97 59. Waterbury JB, Watson SW, Valois FW, Franks DG (1986) Biological and ecological characterization of the marine unicellular cyanobacterium Synechococcus. Can Bull Fish Aquat Sci 214:71±120 60. Waterbury JB, Valois FW (1993) Resistance to co-occuring phages enables marine Synechococcus communities to coexist with cyanophages abundant in seawater. Appl Environ Microbiol 59:3393±3399 1 61. Weisse T (1993) Dynamics of autotrophic picoplankton in marine and freshwater ecosystems. In: JG Jones (ed) Advances in Microbial Ecology, vol 13. Plenum Press, New York, 327±370 62. Zubkov MV, Sleigh MA, Burkill PH, Leakey RJG (2000) Picoplankton community structure on the Atlantic Meridional Transect: a comparison between seasons. Progr Oceanogr 45:369±386