Coral Reefs

, Volume 31, Issue 2, pp 401–414

Interactive effects of climate change and eutrophication on the dinoflagellate-bearing benthic foraminifer Marginopora vertebralis

Authors

    • Australian Institute of Marine Science
  • N. Vogel
    • Ludwig-Maximilians-University
  • J. Doyle
    • Australian Institute of Marine Science
  • C. Schmidt
    • Department of GeosciencesUniversity of Tübingen
  • C. Humphrey
    • Australian Institute of Marine Science
Report

DOI: 10.1007/s00338-011-0851-2

Cite this article as:
Uthicke, S., Vogel, N., Doyle, J. et al. Coral Reefs (2012) 31: 401. doi:10.1007/s00338-011-0851-2

Abstract

Elevated sea surface temperatures caused by global climate change and increased nutrient concentrations resulting from land runoff both are stressors for calcifying coral reef organisms. Here, we test the hypothesis that increased temperature leads to bleaching in dinoflagellate-bearing foraminifera similar to corals and that increased nutrients through runoff can exaggerate stress on the holobiont. In an experiment manipulating temperatures alone, we have shown that mortality of Marginopora vertebralis increased with temperatures. Most individuals died after 7 days at 34°C, ~5°C above current summer maxima. Survival at 37 days was >98% at 28°C. After 7 days of exposure to 31 or 32°C, photosynthesis of the endosymbionts was compromised, as indicated by several photophysiological parameters (effective quantum yield and apparent photosynthetic rate). In a flow-though experiment manipulating both temperature (three levels, 26, 29 and 31°C) and nitrate concentrations (3 levels, ~0.5, 1.0 and 1.4 μmol l−1 NO3), elevated temperature had a significant negative effect on most parameters measured. At 31°C, most photopigments (measured by UPLC) in the foraminifera were significantly reduced. The only pigment that increased was the photoprotective diatoxanthin. Several other parameters measured (maximum and effective quantum yield, O2 production in light, organic carbon contents) also significantly decreased with temperature. Optode-based respirometry demonstrated that the presence of symbionts at elevated temperatures represents a net carbon loss for the host. Growth rates of M. vertebralis and mortality at the end of the experiment were significantly affected by both temperature increase and nitrate addition. We conclude that these foraminifera bleach in a similar fashion to corals and that global sea surface temperature change and nitrate increases are stressors for these protists. Furthermore, this provides support for the hypothesis that management of local stressors elevates resilience of coral reefs to global stressors.

Keywords

Coral reef ecologyClimate changeLand runoffBenthosSymbiosis

Introduction

Coral reefs are presumed under pressure from a variety of global and local anthropogenic stressors. On the global scale, these are mainly increasing sea temperatures due to climate change, and ocean acidification caused by increasing carbon dioxide concentrations in the atmosphere (Hoegh-Guldberg et al. 2007; Fabricius et al. 2011). Local stressors for coral reefs mainly result from overfishing and land runoff (Pandolfi et al. 2003; Fabricius 2005; Uthicke et al. 2012). Increased seawater temperatures, especially under high light conditions, can lead to photoinhibition of coral symbionts (dinoflagellates of the genus Symbiodinium) and subsequent loss from the symbiosis (Hoegh-Guldberg 1999). The loss of zooxanthellae or reduced concentrations of photopigments lead to distinctly paler corals, which may suffer mortality if the impact is severe or long (“coral bleaching,” Glynn 1993).

Increased inorganic nutrients through land runoff can affect coral reefs in several ways (Fabricius 2005). Calcification rates in coral and densities of symbiotic dinoflagellates can be directly affected through inorganic nutrient addition (e.g., Marubini and Davies 1996; Ferrier-Pages et al. 2001). Coral cover under higher nutrient loads can be reduced when macroalgae gain competitive advantages as algal growth and productivity are stimulated (Schaffelke and Klumpp 1998; Schaffelke et al. 2005). Population outbreaks of the corallivorous asteroid Acanthaster planci may also be promoted by higher nutrient inputs through higher food availability for their feeding larvae (Fabricius et al. 2010).

It has recently been proposed that management of local disturbances may increase resilience of coral reefs to global stressors (Marshall and Johnson 2007). This finds support in a model proposing that reduced inorganic nitrogen levels may increase the bleaching threshold in corals (Wooldridge 2009; Wooldridge and Done 2009). This model is based on the assumption that zooxanthellae in the host are generally nitrogen limited (Muscatine et al. 1989; Dubinsky and Berman-Frank 2001). A release from nitrogen limitation may reduce translocation of organic carbon to the host, and increased symbiont numbers may disturb the balance in the symbiosis (Dubinsky and Berman-Frank 2001).

Large benthic foraminifera (LBF) of tropical coral reefs are calcifying, algal symbiont-bearing organisms that are important to overall calcification rates on reefs and sediment production (Tudhope and Scoffin 1988; Langer et al. 1997). Foraminifera form symbioses with a much wider range of photoautotrophs including bacillariophytes, rhodophytes, chlorophytes and cyanobacteria (Lee 2006). Foraminifera of the family Soritinae are large (up to 2 cm) and host dinoflagellates of the same genus as corals (Lee 2006). Molecular studies have shown that several of the dinoflagellates from soritines are from the same clades as those found in corals; however, several clades nearly exclusively inhabiting foraminifera were also detected (Garcia-Cuetos et al. 2005; Pochon et al. 2007).

Similar to the coral–algal symbiosis, some foraminiferal species have been found to bleach (Talge and Hallock 1995; Schmidt et al. 2011). Diatom-bearing species of the genera Amphistegina, Calcarina and Heterostegina have been observed to bleach or mottle during field observations and in aquarium experiments in Florida, the Caribbean, the Pacific and West Australia (Hallock et al. 1992, 2006; Hallock and Talge 1993; Hallock 2000; Hallock and Williams 2006; Schmidt et al. 2011). Both visible and UV light stress (Hallock et al. 1986, 1995; Talge and Hallock 2003; Williams and Hallock 2004) and increased temperature (Schmidt et al. 2011) have been proposed as stressors leading to bleaching in these species. Sorites dominicensis is the only dinoflagellate-bearing species for which bleaching has been described (Richardson 2006), and similar to corals and other foraminifera, temperature, light or UV stress have been proposed as main causative agent.

Physiological and ecological studies have shown that temperature, light and nutrient concentration can influence growth of LBF (e.g., Röttger and Berger 1972; Hallock et al. 1986). Increased temperatures have led to reduced growth rates of some species in situ and in the laboratory (Schmidt et al. 2011; Reymond et al. 2011). Additionally, growth in situ was further reduced on inshore reefs subjected to higher nutrient loads (Uthicke and Altenrath 2010). A study by Reymond et al. (2011) has described a similar decline in growth under enhanced nutrient conditions in both aquarium and in situ experiments. This is likely to be a factor in the findings that the relative abundance of mixotrophic foraminiferal species can be used as an indicator for high nutrient loads on coral reefs (Hallock et al. 2003; Uthicke and Nobes 2008; Uthicke et al. 2010). Negative effects of elevated nutrient concentrations on symbiont-bearing species in conjunction with positive effects of more abundant organic matter on heterotrophic species are cause for this community change (Hallock 1981; Hallock et al. 2003). Many species of symbiont-bearing foraminifera are surprisingly sensitive to high-light conditions (Röttger and Berger 1972; Muller 1978; Hallock et al. 1986), which can drastically reduce growth and photosynthetic yield (Nobes et al. 2008).

In a previous study, we have investigated the effect of temperature and nutrient increase on three diatom-bearing species (Schmidt et al. 2011). That study has illustrated temperature effects on Amphistegina radiata and Heterostegina depressa, and a bleaching response similar to corals at temperatures only slightly elevated above present-day average temperatures in the protist’s environment. However, increase in nitrate concentrations resulted in no measurable change in the bleaching response. In contrast, growth of the dinoflagellate-bearing species Marginopora vertebralis was distinctly reduced when exposed to inshore conditions on the Great Barrier Reef with increased nutrient levels or in laboratory experiments under increased nitrate or phosphate (Reymond et al. 2011). Here, we further explore the interactive effects of elevated temperature and increased nutrients on M. vertebralis. In a first experiment, we tested the hypothesis that M. vertebralis loses its symbionts in a response similar to coral bleaching under elevated temperatures. In an orthogonal two-factor experiment, we tested the hypothesis that temperature and nutrient effects on that species are interactive by measuring a variety of response parameters of the host–symbiont system (e.g., photosynthesis and respiration, composition of photosynthetic and photoprotective pigments, growth and mortality) under flow-through conditions over 37 days.

Materials and methods

Specimen collection and temperature range

Specimens of M. vertebralis Qui and Gaimard 1830 for both experiments were sourced from 3 to 5 m water depth from Shaw Island (S20°31′02.7″ E149°04′48.8″) in the Whitsunday region of the Great Barrier Reef. This species appears quite variable and exists as several different morphotypes on the GBR; all specimens collected from Shaw Island correspond to the morphotype described in detail in Reymond et al. (2011). Approximately 1,000-bp DNA of the 18S region of 5 specimens from Shaw Island was amplified and sequenced using primers S6R, S14rf, S14F3 and 17R (Pawlowski 2000; Holzmann et al. 2001) and sequences deposited on GenBank (Accession Numbers: JN575555-JN575559). These sequences showed a >99% match to specimens used in Reymond et al. (2011) and sequences deposited under that species name on GenBank.

Sea temperature data for the study area were obtained from a temperature logger maintained by the Australian Institute of Marine Science at 6 m water depth at Pine Island (22 km from the collection site, data available at http://data.aims.gov.au). Based on monthly averages for the period from February 2006 to February 2010, the overall average water temperature in the study area was 25.4°C (1 SD = 2.5°C). The average water temperature of the hottest months (January or February) in each year was 28.8°C (1 SD = 0.40°C).

Experimental setup

Static incubator temperature range–finding experiment

To determine the upper temperature limit of M. vertebralis and investigate whether the symbionts of this species bleach, we conducted one experiment in batch culture. Five individual, temperature-controlled incubators (S.E.M. Pty. Ltd., Australia) were set at 28, 30, 31, 32 and 34°C. Four individual 50-ml polypropylene containers (Sarstedt, Australia), each containing 16 specimens of M. vertebralis, were placed into each incubator. Light was supplied by 50:50 actinic 420 nm/10 K trichromatic daylight fluorescent grow tubes (Catalina Compact, 12-h dark/12-h light cycle) and adjusted to deliver between 30 and 35 μmol photons m−2 s−1. This level of light is below the light saturation point for Symbiodinium in M. vertebralis from several GBR locations (Ziegler and Uthicke 2011) and can thus be regarded as non-damaging. Filtered (5 μm) natural seawater was exchanged every second day.

Flow-through experiment

The effects of temperatures (three levels, 26, 29 and 31°C) and three nitrate levels (see below) were studied in 16-l flow-through glass aquaria over a 30-day period. For each temperature by nutrient combination, there were three replicates (distributed at random positions), resulting in 27 aquaria.

Incoming filtered seawater (5 μm) was delivered to four tanks and heated before being pumped into each aquarium. The flow rate was controlled using a liquid flow indicator (RS Components, Ltd., UK) at a range of 450–500 ml min−1. Temperature in the header tanks was controlled using four kW titanium heating bars controlled by a control data logger (CR 1000, Campell Scientific, Australia). To account for variation between temperature replicates, temperatures in all aquaria were measured at least on 5 days per week throughout the experiment (range per aquarium: 1 SD = 0.22–0.82°C).

Nitrate concentrations in the aquaria were manipulated using peristaltic pumps delivering stock solutions (of K+ NO3) at controlled rates (0.8–1.0 ml min−1). We aimed for nitrate levels that represent an ecologically relevant range, with maximum levels being those observed occasionally in inshore areas of the GBR during flood plumes (Devlin and Schaffelke 2009). To increase the baseline nutrient level (no added nitrate treatment), we added nitrate calculated to achieve a maximum enhancement (assuming no uptake by primary producers) of 1.0 and 2.0 μmol l−1 NO3. Dissolved inorganic nutrient concentrations in each aquarium were measured in duplicates five times throughout the experiment following Ryle et al. (1981). The average of these measurements indicated that the enhancement achieved was somewhat below the target, possibly caused by uptake by biofilms. Average nitrate concentrations were 0.53 μmol l−1 (SD = 0.26 μmol l−1, averages given for of 2 duplicates × 9 aquaria × 5 times = 90 samples) for the control and 0.99 μmol l−1 (SD = 0.44 μmol l−1) and 1.42 μmol l−1 (SD = 0.76 μmol l−1) for the +1 and +2 enhanced treatments, respectively.

A six-well plate (Sarstedt) containing 12 specimens of M. vertebralis was placed into each aquarium. Six circles (Ø 3.5 cm) were cut into the lids and a 1-mm plankton mesh inserted between the lid and the plate to allow movement of water. Light quality and quantity were the same as for the static experiment.

Ultra performance liquid chromatography (UPLC) pigment measurements

In the flow-through experiment, photopigments were analyzed from four specimens of M. vertebralis of each replicate aquarium using ultra performance liquid chromatography (UPLC) analysis. Details of this analysis are given in the Electronic Supplemental Material (ESM).

Pulse amplitude modulation fluorometry

Measurements for maximum quantum efficiency (Fv/Fm), effective quantum efficiency of PSII (ΦPSII) and PAR absorptivity of individual M. vertebralis were taken on an Imaging Pulse Amplified Modulated (IPAM) fluorometer (Walz, Germany, Unit IMAG-CM equipped with a Maxi head). Following previous studies on foraminifera (Nobes et al. 2008; Schmidt et al. 2011; Ziegler and Uthicke 2011), we dark adapted samples for 20 min prior to Fv/Fm readings. Effective quantum efficiency (ΦPSII) was read under similar light conditions (30–32 μmol photons m−2 s−1, supplied by the LED array in the Maxi head) as in the experiment. Fv/Fm and ΦPSII were calculated from maximum and minimum fluorescence readings in the dark and light, respectively, following standard procedures (e.g., Ralph and Gademann 2005). Apparent photosynthetic rate (Ps) was calculated by multiplication of PAR absorptivity by effective quantum efficiency and light intensity (Cooper and Ulstrup 2009).

Respirometry

A custom-made respirometer was built to measure respiration and production of individual M. vertebralis. The respirometer used an OXY-4 (Presens, Germany) fiber-optic oxygen meter to track oxygen concentration in enclosed vials via oxygen sensor spots (“optodes”, Ø 0.5 cm, Presens, Germany, details in ESM). During each run, foraminifera were initially incubated in the dark to measure respiration rates, and subsequently under 30–32 μmol photon m−2 s−1. Samples were incubated for a minimum of 25 min in the dark and light to allow for an initial period to stabilize temperatures. Blank chambers were included every 3–4 runs to test for potential changes in O2 concentrations not caused by foraminifera (e.g., microbial respiration). To obtain baseline data on respiration and production of M. vertebralis under the different temperatures, we conducted measurements of 12 specimens for each temperature prior to onset of the experiment. At the end of the experiment, four specimens from each of the 27 experimental aquaria were measured at the respective experimental temperature. Respiration and production rates were normalized to surface area of the foraminifera (derived from calliper measurements of the diameter).

Organic carbon and nitrogen contents

Organic carbon and nitrogen contents of individual M. vertebralis were determined as described elsewhere (Uthicke and Altenrath 2010; Reymond et al. 2011). Individuals were dried and ground with a mortar and pestle; a subsample was weighed and inorganic carbon driven off with 200 μl of 2 M HCl. Upon drying, organic carbon and nitrogen of the samples were determined on a Shimadzu TOC-V analyzer using Miramichi River estuary sediment sample (MESS-1) and Baie des Chalours sediment sample (BCSS-1, Institute for Environmental Chemistry, Canada) as standards.

Growth measurements

Growth of individual M. vertebralis was measured as increase in cross-sectional surface area measured from initial and final digital pictures of all foraminifera in each 6-well plate, taken under standard conditions. Digital pictures were analyzed as described in detail in Uthicke and Altenrath (2010). Surface area growth data were converted to daily growth rates (%) following Kuile ter and Erez (1984).

Statistical analyses

Maximum quantum efficiency, effective quantum efficiency, growth and mortality were arcsine transformed before analysis because they represent proportions. Residual and normality plots for other parameters indicated that assumptions of normality and equal group variance for ANOVAs or linear models were not violated. All pigment data were log (+1) transformed because this distinctly improved agreement with the normality assumption. All data from the static experiment were analyzed with separate one-way ANOVA for the final values and two intermediate points in time. Similarly, the significance of differences in baseline respiration and production rates at three different temperatures was tested using one-way ANOVA. To accommodate slight variations in nutrient concentrations and temperature between individual aquaria in the flow-through experiment, we analyzed all parameters from that experiment with linear models, using average temperatures and nutrient concentrations for each individual aquarium as explanatory variables. All analyses of variance were conducted in NCSS (Hintze 2001), and linear models were conducted in R 2.10 (R Development Core Team 2010).

Results

Static temperature range experiment

An experiment with M. vertebralis specimens in enclosed containers was conducted to describe the response of this species to temperature alone over a larger temperature range than the flow-through experiment. In the 34°C treatment, the first mortality was observed after 48 h of exposure. All individuals in that treatment were dead after 9 days. In the remaining treatments, no further mortality was observed until day 19 (at 32°C). At the experiment termination (day 37), mortality rates remained low in the 28°C (1.5%, SD = 2.9%), 30°C (4.4%, 1 SD = 7.7%) and 31°C (14.3%, 1 SD = 2.3%) treatments, while 61.6% (1 SD = 20.4%) of all M. vertebralis in the 32°C treatment had died.

Photophysiological parameters showed distinct differences in foraminifera at different temperature treatments over time (Fig. 1). Maximum quantum efficiency of PSII (Fv/Fm), effective quantum efficiency (ΦPSII) and apparent photosynthetic rate (Ps) were reduced in the 34°C treatment from the first measuring point after 24 h (=day 2) and all subsequent measuring points. No photophysiological measurements in this temperature group could be obtained after day 7, as no living specimens remained. Also, the 32°C treatment, and to a lesser extent the 31°C treatment, had distinctly reduced photophysiological parameters when compared to 28°C. In general, effects were more distinct in the effective quantum efficiency (e.g., 32°C, day 37: 15% reduction) and Ps (39%) when compared with the maximum quantum efficiency (8%, Fig. 1). Analysis of variance conducted for the first, last and two intermediate points in time confirmed the patterns observed, with most comparisons being highly significant (Table 1). Post hoc tests (Tukey–Kramer) revealed that maximum and effective quantum efficiency after 24 h of exposure were significantly higher in the 28°C treatment compared to all other temperatures. Averages of the photophysiological parameters at 32°C treatments were significantly below those from lower temperatures for most comparisons from day 7 onward (Table 1, post hoc tests). Some of the photophysiological parameters were also reduced at 31°C, but less consistently than at 32°C.
https://static-content.springer.com/image/art%3A10.1007%2Fs00338-011-0851-2/MediaObjects/338_2011_851_Fig1_HTML.gif
Fig. 1

Photophysiological parameters of Marginopora vertebralis exposed to 5 different temperatures during a static experiment. Note that days on the x-axes are not equidistant. Error bars represent 95% confidence intervals

Table 1

ANOVA and post hoc (Tukey–Kramer) results for photophysiological parameters of Marginopora vertebralis in the static temperature experiment

 

Day 2

Day 7

Day 16

Day 37

DF

MS

F

P

DF

MS

F

P

DF

MS

F

P

DF

MS

F

P

Maximum yield

 Temperature

4

0.13645

56.9

0.0000

4

0.04101

18.5

0.0000

3

0.02615

5.92

0.0102

3

0.01182

4.63

0.0225

 Container (T)

15

0.00240

1.61

0.0728

13

0.00222

1.91

0.0303

12

0.00442

3.94

0.0000

12

0.00255

1.88

0.0410

 Residual

212

0.00149

211

0.00116

226

0.00112

156

0.00136

 Total

231

228

241

171

Effective yield

 Temperature

4

0.13374

28

0.0000

4

0.04867

19.7

0.0000

3

0.06249

54.3

0.0000

3

0.01433

2.37

0.1215

 Container (T)

15

0.00477

2.45

0.0025

13

0.00247

1.27

0.2348

12

0.00115

0.47

0.9301

12

0.00604

1.67

0.0774

 Residual

212

0.00195

211

0.00195

226

0.00244

156

0.00361

 Total

231

228

 

241

 

171

Ps

 Temperature

4

0.00046

1.19

0.3538

4

0.00101

3.14

0.0517

3

0.00070

5.51

0.0130

3

0.00352

12.81

0.0005

 Container (T)

15

0.00039

2.12

0.0104

13

0.00032

2.23

0.0094

12

0.00013

0.79

0.6567

12

0.00027

2.88

0.0013

 Residual

212

0.00018

211

0.00014

226

0.00016

156

0.00010

 Total

231

228

0.03861

 

0.03992

0.02872

Post hoc tests

 Maximum yield

28 > 30, 31, 32 > 34

28, 30, 31, 32 > 34

28 > 32

28, 30, 31 > 32

 Effective yield

28 > 32 > 34

28 > 31 > 32 > 34; 30 > 32 > 34

28, 30 > 31 > 32

 Ps

30, 31 > 34

28 > 34

28, 30 > 32

28, 30, 31 > 32

ANOVAs for maximum quantum efficiency (Fv/Fm), effective quantum efficiency (ΦPSII) and apparent photosynthetic rate (Ps) are presented for days 2, 7, 16 and 37 of the experiment. Respective post hoc tests (α = 0.05) are given at the lower section of the table

Flow-through experiment

With the exception of beta carotene, all pigments, including specific pigment ratios, showed significant effects due to temperature. In contrast, there was no effect due to nitrate addition on the pigment concentrations or specific pigment ratios (Table 2). Diatoxanthin (Dtx) increased with increasing temperature, but all other pigments (except beta carotene) had highly reduced (17–57%) concentrations at the higher temperatures (Fig. 2). Chlorophyll a significantly decreased by 54% at 31°C. The ratio of Dtx over the sum of Dtx and diadinoxanthin (Dtx/(Dtx + Ddx) is an indicator for the state of the xanthophyll cycle, which is often correlated to the ΔpH-dependent component (qE) of non-photochemical quenching (NPQ) (Müller et al. 2001; Goss et al. 2006). This ratio increased with temperature, with values close to zero measured at 26°C and >0.10 at 31°C. Similarly, the ratio of photoprotective pigments (Ddx, Dtx, dinoxanthin and beta carotene) over light-harvesting pigments (chlorophyll a, chlorophyll c2 and peridinin) more than doubled at temperatures higher than 29°C (Fig. 2).
Table 2

Results of linear models for photopigments in Marginopora vertebralis at the end of the flow-through experiment. An interaction term between the two main factors (temperature and nitrate) was not significant, and inclusion did not improve the model (ANOVA test for model comparison); thus, the interaction was removed from all models

 

Estimate

SE

t

P

R2

Chlorophyll a

 Intercept

0.2611

0.0337

7.74

<0.0001

0.59

 Temperature

−0.0074

0.0012

−6.17

<0.0001

 Nitrate

0.0029

0.0055

0.53

0.6030

Chlorophyll c2

 Intercept

0.0641

0.0108

5.91

<0.0001

0.44

 Temperature

−0.0018

0.0004

−4.65

0.0001

 Nitrate

0.0004

0.0018

0.23

0.8170

Peridinin

 Intercept

0.1912

0.0331

5.78

<0.0001

0.44

 Temperature

−0.0055

0.0012

−4.69

0.0001

 Nitrate

0.0038

0.0054

0.70

0.4930

Diadinoxanthin (Ddx)

 Intercept

0.0268

0.0047

5.66

<0.0001

0.38

 Temperature

−0.0007

0.0002

−4.22

0.0003

 Nitrate

0.0005

0.0008

0.61

0.5483

Dinoxanthin

 Intercept

0.0042

0.0011

3.81

0.0009

0.10

 Temperature

−0.0001

0.0000

−2.17

0.0404

 Nitrate

0.0002

0.0002

0.83

0.4156

Diatoxanthin (Dtx)

 Intercept

−0.0028

0.0004

−6.63

<0.0001

0.68

 Temperature

0.0001

0.0000

7.22

<0.0001

 Nitrate

0.0000

0.0001

0.40

0.6970

Beta Carotin

 Intercept

0.0015

0.0012

1.22

0.2340

0.00

 Temperature

0.0000

0.0000

−0.61

0.5470

 Nitrate

0.0001

0.0002

0.55

0.5890

Dtx/(Dtx + Ddx)

 Intercept

−0.1189

0.0534

−2.23

0.0357

0.39

 Temperature

0.0076

0.0019

4.02

0.0005

 Nitrate

0.0052

0.0087

0.60

0.5526

PP/LH

 Intercept

−0.1451

0.0410

−3.54

0.0017

0.48

 Temperature

0.0068

0.0015

4.69

0.0001

 Nitrate

0.0057

0.0067

0.84

0.4069

PP/LH: ratio of photoprotective over light-harvesting pigments. All pigment data where log (+1) transformed for analysis. R2, the amount of variance explained by the overall model

https://static-content.springer.com/image/art%3A10.1007%2Fs00338-011-0851-2/MediaObjects/338_2011_851_Fig2_HTML.gif
Fig. 2

Photopigments and pigment ratios of symbiotic dinoflagellates in Marginopora vertebralis at the end of the flow-through experiments. Data were pooled across nitrate treatments because that factor was not significant for any of the comparisons (Table 2)

Similar to the static experiment, maximum quantum efficiency (Fv/Fm), effective quantum efficiency (ΦPSII) and apparent photosynthetic rate (Ps) were significantly affected (linear models, Table 3) by treatment temperature, but not by nitrate addition. All three parameters declined with increasing temperature. Average reduction (comparing 26–31°C) was most distinct in apparent photosynthetic rate (~24%) and smallest in maximum quantum efficiency (~4%, Fig. 3).
Table 3

Linear models for photophysiological parameters (maximum quantum yield [Fv/Fm], effective quantum yield [ΦPSII], apparent photosynthetic rate [Ps]), organic carbon and nitrogen content, respiration, production, growth and mortality, in Marginopora vertebralis

 

Estimate

SE

t

P

R2

Fv/Fm

 Intercept

0.9678

0.0492

19.7

0.0000

0.44

 Temperature

−0.0080

0.0017

−4.6

0.0001

 Nitrate

−0.0001

0.0080

−0.01

0.9901

ΦPSII

 Intercept

0.8118

0.0723

11.3

0.0000

0.47

 Temperature

−0.0127

0.0026

−5.0

0.0000

 Nitrate

0.0066

0.0118

0.6

0.5800

Ps

 Intercept

24.9606

2.2927

10.9

0.0000

0.63

 Temperature

−0.5509

0.0810

−6.8

0.0000

 Nitrate

0.2877

0.3740

0.77

0.4490

Organic carbon

 Intercept

2.854

0.45619

6.256

0.0000

0.40

 Temperature

−0.0675

0.01613

−4.189

0.0003

 Nitrate

−0.0288

0.07441

−0.387

0.7025

Nitrogen

 Intercept

0.418195

0.111805

3.74

0.00101

0.12

 Temperature

−0.00761

0.003952

−1.925

0.06612

 Nitrate

−0.01769

0.018238

−0.97

0.34166

C/N ratio

 Intercept

0.08032

0.01119

7.178

0.0000

0.31

 Temperature

−0.0014

0.0004

−3.599

0.0014

 Nitrate

0.00266

0.00183

1.455

0.1585

Respiration

 Intercept

0.3416

0.2269

1.5

0.1453

0.09

 Temperature

−0.0151

0.0080

−1.9

0.0726

 Nitrate

−0.0234

0.0370

−0.6

0.5331

Production

 Intercept

0.8867

0.1839

4.8

0.0001

0.43

 Temperature

−0.0267

0.0065

−4.1

0.0004

 Nitrate

−0.0359

0.0300

−1.2

0.2433

Growth

 Intercept

1.2047

0.4414

2.73

0.012

0.40

 Temperature

−0.0370

0.0151

−2.45

0.0223

 Nitrate

−0.9393

0.4190

−2.24

0.0349

 Temp. × Nitrate

0.0288

0.0141

2.04

0.053

Mortality

 Intercept

−0.4874

0.2926

−1.7

0.1087

0.30

 Temperature

0.0178

0.0103

1.7

0.0988

 Nitrate

0.1302

0.0477

2.7

0.0118

In case the interaction term between the two main factors was not significant and inclusion did not improve the model (ANOVA test for model comparison), the interaction was removed. Fv/Fm, ΦPSII, mortality and C/N ratios were arcsine transformed for analysis. R2: the amount of variance explained by the overall model

https://static-content.springer.com/image/art%3A10.1007%2Fs00338-011-0851-2/MediaObjects/338_2011_851_Fig3_HTML.gif
Fig. 3

Photophysiological parameters of symbionts in Marginopora vertebralis measured by PAM fluorometry at the end of the flow-through experiment (top row), and respiration, oxygen production and 24-h net production of measured with optode-based respiration chambers (bottom row). Data were pooled across nitrate treatments because that factor was not significant for any of the comparisons (Table 3)

Respiration rates measured at the start of the experiment were highly significantly different between temperature treatments (1-way ANOVA, F2,33 = 8.51, P = 0.0010). Average rates at 26°C (0.105 μgO2 h−1 mm−2, SD = 0.047 μgO2 h−1 mm−2) and 29°C (0.125 μgO2 h−1 mm−2, SD = 0.029 μgO2 h−1 mm−2) were significantly (post hoc Tukey–Kramer test, P < 0.05 for both comparisons) lower than those at 31°C (0.164, SD = 0.026). Photosynthetic production of foraminifera at the onset of the experiment was also significantly different between temperatures (1-way ANOVA, F2,33 = 4.08, P = 0.026). Production rates at 26°C (0.153 μgO2 h−1 mm−2, SD = 0.071 μgO2 h−1 mm−2) were significantly (post hoc Tukey–Kramer test, P < 0.05) higher than those at 31°C (0.075 μgO2 h−1 mm−2, SD = 0.082 μgO2 h−1 mm−2). Oxygen production rates at 29°C assumed intermediate values (0.114 μgO2 h−1 mm−2, SD = 0.040 μgO2 h−1 mm−2).

Respiration and photosynthetic production rates measured at the end of the experiment were on a similar level to those measured initially and showed similar trends with temperature (Fig. 3). Respiration showed no effect due to nitrate addition (Table 3), and an increase in respiration at higher temperatures was only marginally significant (Table 3, Fig. 3). In contrast, average hourly net production rates declined significantly (linear models, Table 3) from 0.134 μgO2 mm−2 h−1 at 26°C to values near 0 at 31°C. Assuming that respiration rates measured were representative for the 12-h dark phase and production rates for the 12-h light phase, we estimated daily net production rates for M. vertebralis at the three treatments. Average rates were similar and positive at 26 and 29°C (Fig. 3). In contrast, daily net production at 31°C was negative.

Organic carbon content of M. vertebralis at the end of the experiment was significantly reduced at higher temperatures (Table 3, Fig. 4); values at 31°C were approximately 31% below those at 29°C. A decrease in N content in foraminifera at higher temperatures (Fig. 4) was only marginally significant (linear models Table 3). Molar C/N ratios significantly declined with higher temperatures (linear models, Table 3) (Fig. 4), mainly reflecting disproportional declines in organic carbon. Nitrate addition had no significant effect on organic carbon and nitrogen concentrations or the C/N ratios.
https://static-content.springer.com/image/art%3A10.1007%2Fs00338-011-0851-2/MediaObjects/338_2011_851_Fig4_HTML.gif
Fig. 4

Organic carbon and nitrogen contents in Marginopora vertebralis at the end of the flow-through experiment. Data were pooled across nitrate treatments because that factor was not significant for any of the comparisons (Table 3)

In contrast to other parameters discussed above, growth and mortality of M. vertebralis were significantly affected by both temperature and nitrate addition (linear models, Table 3). Growth, measured as increase in surface area, remained on a similar level between 26°C (0.057% days−1) and 29°C (0.060% days−1) and then distinctly declined. Growth rates at 31°C were close to zero (Fig. 5). Similar to temperature, increasing nitrate concentrations decreased growth in M. vertebralis, with the most distinct decrease occurring between 0 and 1 μmol nitrate addition (Fig. 5). In the case of growth, inclusion of a marginally significant interaction term improved the model (Table 3). This interaction was inspected graphically with an interaction plot (not shown). As suggested by the individual plots, the effects of temperature and nitrate addition were additive. However, at high temperatures and high nitrate concentrations, the effect was slightly less than the additions of both effects would predict. This was likely to be due to additive effects of the highest temperatures and nitrate concentrations reducing growth to zero.
https://static-content.springer.com/image/art%3A10.1007%2Fs00338-011-0851-2/MediaObjects/338_2011_851_Fig5_HTML.gif
Fig. 5

Growth and mortality of Marginopora vertebralis in the flow-through experiment. Data were pooled across nitrate levels (left hand site) and temperatures (right hand site) to illustrate the significant effects of temperatures and nitrogen

An increase in mortality with increasing temperatures (~11% at 26°C and 21% at 31°C) was only marginally significant (linear models, Table 3). However, mortality significantly increased with higher nitrate concentrations (Fig. 5); mortality was around 8% at 0 μmol l−1 nitrate addition and 23% at 2 μmol l−1 addition.

Discussion

Two long-term experiments with the dinoflagellate-bearing foraminiferal species M. vertebralis were conducted to investigate its tolerance to warming sea surface temperatures and to test whether this species bleaches in a manner similar to corals. Potential temperature–nutrient interactions were investigated to test the hypothesis that reduction in land runoff can “buy time” for symbiont-bearing species until global climate change can be reversed by management action. Growth rates of M. vertebralis in control treatments of our experiment were in a similar range as those observed on inshore reefs in the study area in situ (Reymond et al. 2011). We have demonstrated that temperatures only 1–2°C above current summer maxima (~29°C, see “Materials and Methods”) exert stress on the symbiosis. Although symbiont cells were not counted, we conclude from the strong decline in chlorophyll and reduced organic carbon and nitrogen content that symbionts were expelled or digested in a manner similar to coral bleaching. Furthermore, nitrate addition significantly increased mortality rate and decreased growth. Thus, both temperature and nitrate increase acted as additive stressors, and it can be deduced that reduction of stress exerted by one stressor reduces overall stress on the system.

Temperature effects: bleaching

Nearly every parameter measured both in the static and in the flow-through experiments showed a temperature-related response. In both experiments, temperatures higher than 30°C negatively impacted the health of the specimens, although in both experiments, a high proportion of individuals survived temperatures around 31°C for over 5 weeks. In contrast, at 34°C, all individuals died after 7 days. The response was clearly akin to bleaching in corals. Individuals gradually became paler, starting from the distal part of their shell. It is currently unresolved if symbionts were digested or expelled by the host. It is also possible that loss of pigmentation from individual symbionts can contribute to these patterns. In a manner similar to coral bleaching (Jones et al. 1998; Warner et al. 1999), the photosynthetic efficiency of algae remaining in the foraminifera was reduced. In both experiments, this effect was more pronounced for the effective quantum yield and for apparent photosynthetic rate when compared to the maximum quantum yield. Reductions in the latter parameter represent chronic photoinhibition (Brown et al. 1999). However, reductions in the other parameters allow a better appraisal of the physiological and ecological consequences for the symbiosis under actual experimental conditions.

In corals, temperature stress can lead to increased proportions of photoprotective xanthophylls (Venn et al. 2006). The photoinhibition model of bleaching states that heat stress interrupts the flow of energy to the dark reactions causing a buildup of reactive oxygen species from the light reactions with the dark reactions being initially affected (Jones et al. 1998; Hoegh-Guldberg 1999). An increase in the ∆pH with excess protons being pumped to the lumen activates diadinoxanthin (Ddx) de-epoxidase to form the protective carotenoid diatoxanthin (Dtx) (Jakob et al. 2001). It may also be postulated that, similar to zeaxanthin in other algae (Tardy and Havaux 1997), Dtx assists in the stabilization of the thylakoid membrane during heat stress by increasing the rigidity and thermostability of the thylakoid membrane. These mechanisms have not been investigated previously in foraminifera. However, the increase in Dtx/(Ddx + Dtx) and the ratio of photoprotective over light-harvesting pigment observed here indicate the presence of oxidative stress resulting from temperature exposure. Although Dtx/(Ddx + Dtx) ratios are in a similar range as found in corals, it should be noted that these are distinctly below those found in some free-living microalgae (Warner and Berry-Lowe 2006). It is thus likely that additional mechanisms to protect from oxidative stress exist in Symbiodinium. Reduced pigment concentrations may be caused either by reduced pigment concentrations per dinoflagellate cell or by reduced (through expulsion or digestion) numbers of symbionts. The reduced chlorophyll content at higher temperatures was mirrored by reduced organic carbon content. Although this could originate from either foraminiferal or symbiont biomass loss, we suggest that at least a fraction of the organic carbon reduction (31%) resulted from reduced symbiont biomass. However, given that chlorophyll reduction (54%) was higher than organic carbon reduction, it appears likely that a reduction in symbiont-specific chlorophyll content also occurred.

Most previous studies describing bleaching in foraminifera have focused on diatom-bearing species. Several studies have reported partial to extensive bleaching in Amphistegina species from Florida (e.g., Hallock et al. 1992; Talge et al. 1997) and elsewhere including West Australia (Hallock and Talge 1993; Hallock 2000). Although temperature and high UV were noted as potential stressors contributing to bleaching, it was argued that the main reason for bleaching in Amphistegina species is photoinhibitory stress (Hallock et al. 2006). However, Talge and Hallock (2003) did observe some symbiont loss in Amphistegina gibbosa also bleached under elevated temperature and optimal PAR intensities. Similarly, three species from the GBR adapted to low light (<15 μmol photons m−2 s−1) bleached under temperatures only slightly elevated above current summer maxima without additional light stress (Schmidt et al. 2011).

Bleaching has also been described from one Symbiodinium-bearing foraminiferal species in Florida (Richardson 2006, 2009). As for corals (Brown 1997), the main causes for this bleaching were presumed to be high temperatures coupled with high irradiance, although other factors such as low salinity may also cause bleaching and disrupt the symbiosis.

Optode-based respirometry in the flow-through experiment corroborated the reduced apparent photosynthetic rate under higher temperatures measured by PAM fluorometry. These measurements, based on O2 development, suggested that the consequences for the symbiosis were more drastic (relative effect size nearly 100% reduction at 31°C) than those suggested by PAM-based measurements (relative effect size 24% for apparent photosynthetic rate). Average production rates under the light conditions used in the aquaria were positive and of a similar level for both 26°C and 29°C, and daily net production rates were also positive. In contrast, daily net production at 31°C was distinctly negative, and thus, the host derives no net carbon gain from the remaining symbionts. Interestingly, baseline data on respiration and production rates at different temperatures obtained at the beginning of the experiment indicated that the effects of temperature on respiration of the holobiont and production of the symbionts are near instantaneous.

Jones et al. (1998) suggested that damage to PS II is only a secondary effect of heat stress in corals and that the primary damage occurs during carboxylation in the carbon cycle. Similar to corals, it can be assumed that a breakdown in photosynthesis, be it in PS II or in the dark reaction, leads to unwanted byproducts such as reactive oxygen species that could damage host tissues (Lesser 1997). It is likely that the mechanism in foraminifera is similar. However, production measurements also suggested that keeping symbionts not capable of photosynthesis under high temperatures is likely to be a burden for the host from an energetic point of view.

Nutrient effects

In the flow-through experiment, only growth and mortality had distinct negative effects due to nutrient addition. Given the importance of these parameters for the individual and population dynamics, we concluded that nitrate addition had a clear negative impact on M. vertebralis.

Several studies have investigated effects of nutrients on foraminiferal growth, but methods applied differed widely. Both the diatom-bearing species Amphistegina lobifera and the Symbiodinium-bearing species Amphisorus hemprichii showed no growth response to nitrate or phosphate enhancement in static experiments (Lee et al. 1991). In separate experiments, growth in those species (Kuile ter et al. 1987) and two other diatom-bearing species (Röttger et al. 1980) increased somewhat under inorganic nutrient addition when compared to starved individuals. Neither of the three diatom-bearing species investigated by Schmidt et al. (2011) in the same experimental setup as the present study showed changes in growth, chlorophyll content or photosynthetic efficiency that could be related to nitrate addition. Contrary to aquarium studies, growth in two diatom-bearing species was distinctly reduced in inshore areas, and statistical analyses suggested that nutrients were the main driver for reduced growth (Uthicke and Altenrath 2010).

In contrast, M. vertebralis showed a decline in growth rates in a static experiment both under increased N or P concentrations in a previous study (Reymond et al. 2011). The same study showed reduced growth of that species in inshore areas with elevated nutrients, supporting the findings from the present study. However, given the small number of species investigated to date, it is too early to speculate whether this is a symbiont-type or species-specific response.

We have shown in previous field studies that C/N ratios in foraminifera reflect dissolved nutrient availability, with lower ratios indicating higher N availability (Uthicke and Altenrath 2010; Reymond et al. 2011). C/N ratios in M. vertebralis found at the end of the flow-through experiment were relatively low, thus indicative of high nitrogen availability. This matches previous observations because even the controlled nitrate concentrations in the flow-through experiment were slightly above average conditions found on inshore reefs at the locations of the field studies (Cooper et al. 2007; Uthicke and Altenrath 2010). However, further increase in nitrate in the flow-through experiment did not further increase nitrogen content in the foraminifera (or, as a consequence, change C/N ratios). Thus, it is possible that the relationship between external nitrogen availability and internal nitrogen concentrations saturates at some point and leads to no additional nitrogen increase within the foraminifera.

In corals, increased nitrate concentration can lead to larger zooxanthellae volume and chlorophyll content per cell (Schloeder and D’Croz 2004). The latter study found reduced zooxanthellae numbers when nitrate and temperature stress interacted. Although we did not directly count zooxanthellae, the fact that neither the photopigments nor organic carbon and nitrogen changed due to nitrate addition suggests that no distinct changes in zooxanthellae numbers occurred due to that factor in our experiment. However, given that C/N ratios in the field varied with dissolved inorganic nitrogen (DIN) availability (Uthicke and Altenrath 2010), it is possible that symbiont population densities are regulated by nutrient availability at lower water column DIN concentrations.

Additive effects of temperature and nutrients

The bleaching mortality threshold for corals from Daydream Island in the Whitsunday Group (only 36 km from the sample location of M. vertebralis) was found to be dependent on exposure time. At 29.1°C ~40 days exposure was required, whereas at 30.1°C less than one day of exposure was required for bleaching to commence (Berkelmans 2009). Thus, it appears that the threshold of M. vertebralis is slightly higher than that of this particular coral species tested. However, the temperature threshold is still only 1–2°C above current summer maxima in that area. Seawater temperatures in the tropics have increased by 0.2–0.4°C in the last four decades alone (Kleypas et al. 2008), and the predicted increase for GBR waters for 2,100 is a further 1–3°C (summarized in Lough 2007). Thus, in the near future, M. vertebralis is likely to be above its bleaching threshold with increased frequency. It was hypothesized that improved management of local stressors, such as nitrogen runoff, can “buy time” and improve resilience for corals on the GBR to global change-related threats such as increases in SST (Wooldridge 2009; Wooldridge and Done 2009). This hypothesis was mainly derived from data modeling and summarizing published data on individual stressors (temperature, nitrate increase). To our knowledge, this hypothesis has not been tested for corals by manipulation of both stressors in parallel. Given that both stressors had an additive effect for growth and mortality in M. vertebralis, we argue that this hypothesis has merit for the foraminifera–symbiont system studied here. An assumption of the model is that under increased nutrient levels, symbionts are released from nutrient limitation (Dubinsky and Berman-Frank 2001), increase in numbers and translocate less organic matter to the host. Higher symbiont concentrations may also be detrimental for host growth because photosynthesis and calcification in foraminifera compete for inorganic carbon (Kuile ter et al. 1989). However, given that we found no evidence for increased symbiont numbers under higher nutrients, the mechanisms of the additional stress due to nitrate addition remain unresolved. The nature of the foraminiferal–algal symbiont system and ease of maintenance in experiments warrant its use as a possible model for symbioses between symbiotic algae and calcifying animals.

Acknowledgments

This research was supported by the Australian Government’s Marine and Tropical Sciences Research Facility, implemented in North Queensland by the Reef and Rainforest Research Centre Ltd. We acknowledge the Reef Plan Marine Monitoring Program, developed by the Great Barrier Reef Marine Park Authority and funded by the Department of Environment, Water, Heritage and the Arts, for providing data and information.

Supplementary material

338_2011_851_MOESM1_ESM.doc (112 kb)
Supplementary material 1 (DOC 112 kb)

Copyright information

© Springer-Verlag 2011