The Effects of Elevated Temperature Stress on the Acquisition and Allocation of Carbon to Lipids in Hawaiian Corals THESIS Prese

Home , Montipora capitata, Porites compressa

The effects of elevated temperature stress on the acquisition and allocation of carbon to lipids in Hawaiian corals

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

Justin H. Baumann, B.S.

Graduate Program in Earth Sciences

The Ohio State University

2013

Master's Examination Committee:

Dr. Andréa Grottoli, Advisor

Dr. Lawrence Krissek

Dr. John Olesik

Justin H. Baumann

2013

Abstract

Understanding the complex processes of coral response to, and recovery from, bleaching events is central to our ability to predict the impacts of current and future climate change on coral reef ecosystems. Lipids are key biomolecules within the coral holobiont, serving as structural components, as well as significant energy reserves. With the frequency and intensity of bleaching events expected to rise in the coming decades, it is important to understand how coral lipids will be effected by, and recover from, bleaching. A bleaching experiment, followed by carbon pulse-chase labeling, was performed to investigate the assimilation and allocation of carbon to coral host and endosymbiont tissues (including lipids) over the course of bleaching and recovery. Here, we show that bleaching results in a decline in the allocation of photosynthetically derived carbon to lipids in the first month of recovery, but that photosynthetic carbon allocation had fully recovered after 11 months in both Porites compressa and Montipora capitata. In contrast, the allocation of heterotrophic carbon to lipids was no different between bleached and control corals in the first month for both species. While this pattern did not change after 11 months of recovery for M. capitata, dramatically higher enrichment values in the lipids built with heterotrophic carbon of bleached compared to non-bleached control Porites compressa

ii corals were observed. This suggests that either the corals are still recovering, or that they have acclimated and are better able to resist additional bleaching. Overall, P. compressa catabolizes newly synthesized non-lipid energy reserves and maintains newly synthesized lipids. In comparison, M. capitata catabolizes heterotrophic carbon from feeding and catabolizes some of its newly synthesized lipids to meet metabolic demand while maintaining its existing lipid, protein, and carbohydrate reserves. Due to maintenance of total energy reserves and only a small (yet significant) amount of lipid catabolism, it appears that M. capitata recovers faster and may be more resilient to bleaching than P. compressa.

iii

This document is dedicated to my family and friends, especially my parents, whose

support has been vital to my successes thus far.

Acknowledgments

I would like to express my gratitude to my advisor, Dr. Andréa Grottoli, for the continuous support of my undergraduate and graduate research, and for all of the opportunities that have been afforded to me over the past 5 years. Her knowledge, patience, and guidance have been paramount in my academic success. I would also like to thank Dr. Lawrence Krissek, whose enthusiasm for ocean science sparked an interest in my and got me started down this path as an undergraduate. His mentoring has been invaluable as well.

My sincerest thanks to the entire committee: Dr. Andréa Grottoli, Dr. Lawrence

Krissek, Dr. John Olesik, and Dr. Christoper Jekeli for their patience and understanding during the preparation of the manuscript and subsequent defense, as well as for comments and questions related to the manuscript.

I would also like to thank Jim Bauer for his tremendous assistance in editing and developing my research proposal, Adam Hughes, who performed fieldwork and preliminary laboratory work associated with this project, and Yohei Matsui for guidance in the laboratory and for teaching me so many techniques and methods.

Finally I would like to thank S. Levas, V. Schoepf, M. Geronimus, C. Treffny, C.

Zerda, C. Gearing, S. Blackhurst, T. Pearse, P. Jokiel, J. Fleming, A. Chrystal, R. Moyer,

C. Paver, Z. Rosenheim, L.-W. Hung, S. Himes, B. Hull, P. Burns, P. Bills, R. Micheli, v

F. Lugo, S. Hughes, the Hawaii Institute of Marine Biology, and the Hawaii Department of Land and Natural Resources for assistance with fieldwork, laboratory work, constructive criticism, and assistance over with this and various other project.

I am also grateful for funding provided to AGG by the National Science

Foundation Program in Biological Oceanography (Grant No. 0542415), and the National

Science Foundation Program in Chemical Oceanography (Grant No. 0610487), and to JB by the Ohio State University College of Arts and Sciences Mayer Travel Fellowship, and the Ohio State University School of Earth Sciences Friends of Orton Hall fellowship.

Vita

June 2007 ...... Lakota West High School, Honors Diploma

June 2011 ...... B.S. Biology and Earth Sciences w/

Research Distinction, The Ohio State

University

2011 to present ...... Graduate Teaching Associate, School of

Earth Sciences, The Ohio State University

Fields of Study

Major Field: Earth Science

Coral Physiology

Marine Ecology

Biogeochemistry

vii

Table of Contents

Abstract ...... ii

Dedication ...... iv

Acknowledgments...... v

Vita ...... vii

List of Tables ...... x

List of Figures ...... xi

Introduction ...... 1

Methods...... 5

Experimental Design ...... 5

Pulse-Chase Labeling ...... 6

Sample Analysis ...... 9

Statistical Analysis…………………………………………………………………….10

Results……………………………………………………………………………………13

Photoautotrophically Acquired Carbon……………………………………………….13

viii

Heterotrophically Acquired Carbon…………………………………………………..14

Proportionate Contribution of Photoautotrophic and Heteroautotrophically Derived C

to Coral Lipids..…………………………………………………………………...…..14

Discussion………………………………………………………………………………..16

Photoautotrophically Acquired Carbon…………………………..…………………...16

Heterotrophically Acquired Carbon…………………………………………………..18

Proportionate Contribution of Photoautotrophic and Heterotrophically Acquired

Carbon………………………………………………………………...………………22

Summary…………………………………………………………………………………23

References………………………………………………………………………………..27

Appendix A: Table……………………………………………………………………….30

Appendix B: Figures……………………………………………………………………..33

Appendix C: Raw Lipid δ13C Data………………………………………………..….….40

Appendix D: Average Lipid δ13C Enrichment per Chase Interval…………...…..……..48

Appendix E: Average Percent Contribution of Heterotrophic and Photoautotrophic C to

Coral Lipids ……………………………………………………………………………..49

Appendix F: Lipid Extraction Method Comparison………………………………,…….50

List of Tables

Table 1: Montipora capitata and Porites compressa Chla concentration ANOVA results…………………………………………………………………………………….30

Table 2: Montipora capitata and Porites compressa DI13C enrichment ANOVA

results……………………………………………………………………………..…...31

Table 3: Montipora capitata and Porites compressa 13C-rotifer enrichment ANOVA

results………………………………………………………………………….………32

Table 4: Average lipid δ13C enrichment values for Montipora capitata and Porites compressa using two different extraction methods (DCM: Methanol or

Chloroform:Methanol)………………………………………………………..…………52

Table 5: Porites compressa method comparison ……………………………………….53

Table 6: Montipora capitata method comparison..………………………………..…....53

List of Figures

Figure 1: Flow diagram of experimental design ………………………………………...33

Figure 2: Average daily temperatures on the reef in the non-bleached control tanks and in the treatment tanks……………………………………………………………………… 34

Figure 3: Photographs of representative coral fragments of bleached and non- bleached control Porites compressa and Montipora capitata following 1 and 11 months of recovery…………………………………………………………………….35

Figure 4: Average chlorophyll a concentration of non-bleached control and bleached

Montipora capitata and Porites compressa corals……………………………………... 36

Figure 5: Photoautotrophic C assimilation in Porites compressa and Montipora capitata lipids……………………………………………………………………………………...37

Figure 6: Heterotrophic C assimilation in Porites compressa and Montipora capitata lipids……………………………………………………………………………………...38

Figure 7: Average percent contribution of carbon from DIC and heterotrophy to holobiont total lipids 24 and 168 hours after pulse-labeling in Porites compressa and Montipora capitata after 1 month and 11 months of recovery………………………………………39

Introduction

Coral reefs are declining globally due to a combination of direct and indirect human impacts, such as greenhouse gas emissions, agricultural runoff, overfishing, and habitat destruction (Hughes et al. 2003; Hoegh-Guldberg et al. 2007; Veron et al. 2009;

Frieler et al. 2012). Mass coral bleaching events, a phenomenon where whole communities of corals loose a significant proportion of their vital endosymbiotic dinoflagellates (commonly called zooxanthellae) and/ or their algal photosynthetic pigments, are largely caused by elevated sea-surface temperatures (Jokiel and Coles

1990; Glynn 1996; Hoegh-Guldberg 1999; D'Croz et al. 2001). Impacts of bleaching include the following: decreased growth in coral tissue and skeleton formation, reduction or cessation of gamete production and fertilization, and increased susceptibility to disease

(Szmant and Gassman 1990; Fitt et al. 1993; Omori et al. 1999; Ward et al. 2000).

Extended and/or extreme warming episodes can lead to mass coral mortality and ecosystem degradation (Wilkinson 2000; Stanley 2003). At the current rate of predicted global warming, mass bleaching events are expected to increase in frequency and intensity in all tropical oceans in the coming decades (Hoegh-Guldberg 1999; Wilkinson

2000; Buddemeier et al. 2004; Wooldridge et al. 2005) resulting in up to 60% coral mortality within the next few decades (Donner 2009; Frieler et al. 2012).

Healthy corals acquire fixed carbon (C) by two means. First, the endosymbiotic algae photosynthetically fix C (photoautotrophy) in excess of their daily metabolic needs and translocate the majority of it to the coral host, thus supplying the host with up to 1

100% of its daily metabolic requirements (Muscatine et al. 1981; Falkowski et al. 1993;

Grottoli et al. 2006). Second, corals can acquire up to 60% of their fixed C by capturing zooplankton (including pico- and nanoplankton) from the water column (heterotrophy)

(Goreau et al. 1971; Grottoli et al. 2006; Rodrigues and Grottoli 2006; Palardy et al.

2008; Rodrigues et al. 2008; Tremblay et al. 2012) and via the uptake of dissolved organic carbon (DOC) (Levas 2012; Levas et al. 2013) and particulate organic carbon

(POC) (Anthony 1999; Anthony and Fabricius 2000). During elevated temperature stress, such as during bleaching events, the symbiosis between coral host and endosymbiont can break down, corals lose significant numbers of their endosymbiotic algae causing decreases in photosynthesis and the incorporation of photoautotrophically acquired C into coral tissues, and a failure to meet metabolic demand through photosynthesis alone

(Warner et al. 1996; Lesser 1997; Grottoli et al. 2006; Rodrigues and Grottoli 2007;

Palardy et al. 2008; Hughes et al. 2010). To counteract decreases in photosynthetic C, corals can meet their daily metabolic energy requirements by doing one or more of the following: 1) catabolizing stored energy reserves (i.e.: lipids, proteins, carbohydrates)

(Porter et al. 1989; Fitt et al. 1993; Grottoli et al. 2004b; Rodrigues and Grottoli 2007;

Grottoli and Rodrigues 2011)(Grottoli et al 2006), 2) increasing their feeding rates (i.e., heterotrophically acquired C) (Grottoli et al. 2006; Palardy et al. 2008; Tremblay et al.

2011; Levas 2012), 3) decreasing metabolic rates (Rodrigues and Grottoli 2007; Levas

2012), and/or 4) decreasing calcification rates (i.e., (Leder et al. 1991; Rodrigues and

Grottoli 2006; Carricart-Ganivet et al. 2012; Levas et al. 2013). Of these four strategies, we further explored how bleaching and recovery affect energy reserves, particularly lipid

2 synthesis.

Lipids are key biomolecules for growth and storage (Patton et al. 1977; Stimson

1987; Grottoli et al. 2004b; Rodrigues and Grottoli 2007; Rodrigues et al. 2008; Christie and Han 2010; Birsoy et al. 2013), and play a significant role in the production of gametes and thus reproduction (Ward 1995). Further, high lipid content promotes resilience to, and recovery from coral bleaching (Anthony et al. 2009). Therefore, physiological strategies that promote lipid synthesis and storage should promote coral resilience to bleaching. In principle, corals should be able to draw on fixed C (i.e., organic matter) for lipid synthesis from both photoautotrophically and heterotrophically acquired C. However, since bleaching significantly reduces photoautotrophically fixed C acquisition and allocation (Hughes et al 2010), it is unclear how bleaching might affect lipid synthesis and utilization, and how that might influence a coral’s capacity to recover.

If the majority of C allocated to lipid synthesis is photoautotrophic in origin, then recovery of lipid content and its associated physiological functions after coral bleaching should depend on recovery of the endosymbionts and their photosynthesis rates, which can take up to 4 months (Rodrigues and Grottoli 2007; Connolly et al. 2012; Levas et al.

2013; Hughes and Grottoli Submitted). However, if the majority of C allocated towards lipids is heterotrophic in origin, then recovery of lipid content in situ would depend on the heterotrophic plasticity of a given species when bleached (Grottoli et al. 2006;

Palardy et al. 2008; Levas et al. 2013). Coral species that are capable of increasing feeding rates when bleached could potentially recover their lipid levels very rapidly or even maintain lipid content (Grottoli et al. 2004a; Rodrigues and Grottoli 2007). To date,

3 no study has determined the proportionate contribution of both photoautotrophic and heterotrophic C to total holobiont lipids (henceforth referred to simply as lipids) in bleached and healthy corals after bleaching or during recovery. Here, we conducted a manipulative experiment to determine the acquisition, allocation, and utilization of C to lipids in bleached and healthy Hawaiian corals Montipora capitata and Porites compressa using 13C-isotope labeling techniques after 1 and 11 months of recovery following bleaching to determine 1) the relative contribution of photoautotrophic vs. heterotrophic C to lipid synthesis in bleached and healthy corals, and 2) how that ratio varies between species, seasonally, and over the course of recovery?

Methods

Experimental Design

The general experimental design and pulse-chase labeling methods on corals collected immediately following bleaching (0 months of recovery) are outlined in detail in Hughes et al. (2010). In this study, we present findings from corals from the same study but collected at 1 and 11 months recovery. The same pulse-chase protocols were followed at 0, 1 and 11 months recovery. Briefly, corals were collected from a fringing reef (2-4 meters depth) surrounding Moku O Lo’e Island at the Hawaii Institute of

Marine Biology in Kaneohe Bay, Hawaii on 11 August, 2006. Five healthy colonies of

Montipora capitata (branching morphology) and Porites compressa (branching morphology) were collected. Five coral branch tip fragments (5 cm tall) were collected from each colony, attached to ceramic tiles, randomly assigned treatments, and placed into outdoor-flow-through seawater tanks and allowed to acclimate for 7 days (Fig. 1). In half of the tanks the average daily temperature was 27.36°C ± .074 (ambient controls). In the other half, temperature was slowly ramped up to 30.45°C (± 0.09) over the course of a week using aquarium heaters (bleaching treatment). These conditions were maintained for 3.5 weeks. Temperature was monitored every 15 minutes in each tank using Hobo

UA-002-08 temperature loggers. Light intensities in the tanks were reduced to that of collection depth (midday: 235.2275 25.641µmol photons/ m2/ sec) by covering the tanks with 2 layers of neutral-density mesh. Throughout the tank experiment, corals were

5 fed freshly caught zooplankton for 1 h at dusk every other night. Coral fragments were rotated within tanks daily, throughout the experiment, in order to avoid positional effects within a tank. In addition, all corals were rotated randomly among tanks of the same treatment every 4 d to prevent tank position effects. After 3.5 weeks the heaters were turned off (6 September, 2006) and the bleached treatment and ambient control fragments were placed on the reef at allowed to recovery for 1 month at 2-3 m depth (Fig. 1) prior to

13C pulse-chase labeling. Fragments immediately following the 3.5-weeks in the tanks

(i.e., 0 months recovery) were not available for this study as they were destructively sampled for bulk tissue isotopic analyses (i.e., skeleton, host tissue, endosymbiont tissue) in Hughes et al (2010).

After 1 and 11 months of recovery on the reef, two fragments of each colony and treatment were incubated in 13C-labelled dissolved inorganic carbon (DI-13C) in seawater, in order to label the photoautotrophically acquired C incorporated into lipids. The fragments were then chased for one week to track how the photoautotrophic C that was incorporated into the lipids during the incubation was utilized over the course of a week.

This technique is called pulse-chase labeling. Two additional fragments from each colony and treatment were incubated overnight with 13C-labeled rotifers after both 1 and

11 months of recovery in order to label the heterotrophic C incorporated into the lipids.

They were also chased for one week to track how the heterotrophic C that was incorporated into the lipids during the incubation was utilized over the course of a week.

DI-13C: pulse-chase labeling photoautotrophically acquired carbon

After 1 month of recovery, DI-13C pulse incubations were conducted on 6 October for Porites compressa and 9 October for Montipora capitata, respectively. Five 40 l glass aquaria were filled with 25 l of seawater at 07:30 h. The aquaria were placed in outdoor flow-through seawater tanks with water circulating around them to maintain ambient temperature during the incubation. Bleached corals were placed in two of the four aquaria, ambient control corals were placed in two aquaria, and the fifth aquarium served as a control (DI-13C added, no coral, Fig. 1). At 08:00 h, 4.5 ml of 0.117 M 98 at. % 13C

NaHCO3 was added to the four coral-containing aquaria. Dissolved inorganic carbon

-1 (DIC) concentrations increased by 50 µmol kg due to the NaHCO3 addition – an

13 increase of roughly 2.5%. Average initial incubation seawater δ CDIC values were 911.90

±27.2‰ and 983.48 ±8.9‰ for M. capitata and P. compresa, respectively. The incubations were performed for 8 h during the day to allow for uptake of DIC during maximum photosynthesis. The incubation was 8 h long in order to take advantage of peak photosynthesizing daylight hours and allowing sufficient time for C assimilation and allocation. Coral fragments were removed from the glass incubation aquaria after 8 h, and returned to unlabeled, natural flow-through seawater. One fragment from each colony and treatment was removed during the first 24 and another at 168 h and immediately frozen at -50°C (Fig. 1). Previous work on these corals shows that the δ13C enrichment values in the bulk host tissue, endosymbiont tissue, and skeleton do not differ significantly over the first 24 h following DI-13C incubations (Hughes et al. 2010), so it was assumed that δ13C lipid enrichment values also do not differ over the course of the first 24 hours. The same procedure was repeated for both species after 11 months of

7 recovery (16 August, 2007 for M. capitata and 18 August, 2007 for P. compressa, Fig. 1).

13 Average initial incubation seawater δ CDIC values were 848.11 ±8.1‰ and 860.23

±10.6‰ for M. capitata and P. compresa, respectively.

13C-rotifer: pulse-chase labeling heterotrophically acquired carbon

The incubations for Porites compressa were conducted on 6 October, and the incubations for Montipora capitata were conducted on 9 October, 2006. As with the DI-

13C labeling, 40 l glass aquaria were filled with 16 l of seawater and placed in outdoor flow-through seawater tanks. Bleached corals were placed into 2 tanks, ambient corals were placed into 2 tanks, 1 tank served as a 13C rotifer control (13C-rotifer added, no coral, Fig. 1), and 2 additional tanks served as seawater controls (no 13C-rotifer, no coral).

Labeled rotifers were added to the coral-containing aquaria when it was dark (20:00 h), at a density of 10-15 rotifers ml-1 of seawater. The rotifers were 13C-labeled by feeding them 13C-labeled Nanocropsis paste for 96 h prior to the incubations. Rotifer δ13C values for M. capitata and P. compressa were 3027.01±119.71‰ ‰ and 10051.03 ±115.272‰

‰ respectively. Rotifer enrichment values varied between incubation because they were prepared on different days with different batches of rotifers. This did not affect our ability to statistically evaluate isotope enrichment patterns within each species. In addition, it did not affect our ability to calculate the mass balances (see details below) as the different rotifer enrichment values were taken into account.

Corals were incubated with 13C-labeled rotifers for 10 h during the night as corals naturally feed at night, and then placed back into unlabeled, flow-through seawater prior

8 to sunrise. A10 h incubation was necessary to allow for assimilation and allocation of heterotrophic C to the lipids during peak feeding times while not exposing the corals to sunlight which would have confounded the results. One fragment from each colony and treatment was removed within the first 24 hours and another after 168 h and immediately frozen at -50°C. The same procedure was followed for both species after 11 months of recovery (August 16, 2007 for M. capitata and August 18, 2007 for P. compressa, Fig. 1).

Rotifer δ13C values during the 11 month recovery incubations for M. capitata and P. compressa were 40601.71±4450.4‰ and 17316.67 ±2217.66‰, respectively.

Sample Analysis

Branch tips were removed from whole coral samples of M. capitata and P. compressa. Approximate surface areas were calculated based on caliper measurements and attributing the most appropriate geometric shape to the tip (i.e., cone, cylinder, or prism). The branch tips were then ground whole (skeleton + animal tissue + endosymbiont) with a mortar and pestle. Total lipids were extracted from ground samples with 2:1 chloroform: methanol, washed in 0.88% KCl, followed by 100% chloroform solution, and another KCl wash (Grottoli et al. 2004a). Total lipid extracts were dried to a constant weight, resuspended in 100% chloroform, and stored in 2:1 chloroform: methanol at -80°C. Tissue biomass was calculated as the sum of the sample ash free dry weight and the dry lipid weight.

In preparation for isotopic analysis, subsamples of total lipid extracts were dried down in tin capsules at 50°C under high-purity N2 gas to a constant weight and analyzed

9 for δ13C as described in Grottoli and Rodrigues (2011). Each sample was combusted in a

Costech elemental analyzer (EA) and the resulting CO2 gas was automatically analyzed for δ13C with a Finnigan Delta IV stable isotope ratio mass spectrometer via a Finnigan

13 13 ConFlow III open-split interface. Lipid δ C values (δ CL) values were reported relative to Vienna Peedee Belemnite Limestone Standard (V-PDB) (δ13C = per mil deviation of the ratio of stable C isotopes 13C:12C relative to V-PDB). Repeated measurements of the

USGS-40 standard (n = 28) had a standard deviation of ±0.03‰, repeated measurements of the USGS-41 standard (n=26) had a standard deviation of ±0.06‰, and repeated measurements of coral lipid samples (n=12) had average differences of 1.86‰ with a standard deviation of ±1.99‰. Since these samples are labeled, and values ranged from -

17 to 223‰, a 1.86‰ difference among duplicates is very acceptable.

Statistical Analysis

Average Chl a values were measured from whole tissue samples of corals at each recovery interval. The effect of treatment (bleached vs. non-bleached control) on Chl a was determined using a one-way ANOVA at each recovery interval. All lipid δ13C values were reported as enrichment values relative to baseline values. Average baseline lipid

δ13C values were measured from the whole lipid extracts of the parent colonies collected from the reef at the beginning of the experiment. Average baseline lipid δ13C values for parent colonies of both P. compressa and M. capitata were -15.5‰ ±0.07. The effect of treatment (bleached vs. non-bleached control), chase interval (first 24 hours vs. 168 hour), and genotype on the lipid δ13C-enrichment of each species, label-type, and

10 recovery interval was determined using a three-way ANOVA. Treatment and chase interval were fixed effects while genotype was a random effect. A posteriori slice tests were used to determine if bleached and control averages significantly differed within a chase interval for each species, label-type, and recovery. Bonferroni corrections were not used due to increased likelihood of false negatives (Quinn and Keough 2002; Moran

2003). We realize that multiple ANOVAs without Bonferroni corrections have inherent limitations, but in this case, they are more informative and have fewer weaknesses than using a Bonferroni correction (Moran 2003) or using multivariate approaches with this dataset. Prior to ANOVA analysis, all data were tested for homogeneity of variance using Shapiro-Wilk’s test. Any data set not meeting this assumption was log transformed with the addition of a constant prior to transformation when necessary. Statistical analyses were generated using SAS software, Version 9.03 of the SAS System for

Windows. Values of p < 0.05 were considered significant. Due to logistical limitations, it was not possible to pulse-chase both species simultaneously, so pulse-chase labeling experiments were conducted on each species on sequential days. As a result, species could not be compared using statistical approaches, but qualitative comparisons between the trends seen in each species were still possible.

To further explore the data, the percent contribution of photosynthetically derived and heterotrophic rotifer-derived C to newly-synthesized lipids of each species and treatment at each chase and recovery interval was calculated. The following formula was used to calculate the percent contribution of photosynthetically derived C to newly synthesized lipids:

(1)

To calculate the percent contribution of heterotrophically derived C to newly synthesized lipids, the subscripts in the numerator terms are switched from DIC to Rot. The δ13C values are those of the seawater DIC and rotifers themselves. Error associated with the average coral atom % values was propagated through the equation. Atom % values were calculated using the following formula,

(2)

Where Rsample values were calculated from the formula,

(3)

( )

13 13 and Rstd is 0.0112240 (deGroot 2004) and δ Csample is the measured lipid δ C value.

Assessment of these results was qualitative only as statistical approaches were not possible.

Results

The average water temperatures in the bleaching and non-bleached control tanks were 30.2 °C (±0.20 SE) and 27.4 °C (±0.08 SE), respectively (Hughes and Grottoli

Submitted) (Fig. 2). Immediately following 3.5 weeks in the tanks (i.e., 0 months of recovery) bleached fragments were visibly white and non-bleached controls remained brown. Chl a values in bleached coral fragments were significantly lower in bleached than in non-bleached controls (Fig 4, Table 1). After 1 month of recovery on the reef, the corals from the bleaching tanks of both species were still visibly bleached (Fig 3) and average chlorophyll a (Chla) values of the bleached corals were significantly lower than those of the control corals (Fig. 4, Table 1). After 11 months of recovery, bleached and non-bleached control fragments appeared to have similar brown coloration (Fig 3) and had the same Chla concentrations in both species (Fig. 4, Table 1).

Photoautotrophically acquired carbon

After 1 month of recovery, lipid δ13C enrichment in P. compressa was significantly lower in bleached that in non-bleached control corals, and declined significantly over the 168 hour chase in control corals (Table 2, Fig. 5A). However after

11 months of recovery, bleached and non-bleached control lipid δ13C enrichment values did not differ, nor did they change over the week-long chase period (Table 2, Fig. 5B). In comparison, after 1 month recovery, M. capitata lipid δ13C enrichment values were 13 significantly lower in bleached than in non-bleached corals only in the first 24hrs of the chase period and no longer differed from controls after that (Table 2, Fig. 5C). After 11 months of recovery lipid δ13C enrichment values did not differ between bleached and non-bleached controls and decreased significantly in over the course of the chase in the non-bleached corals (Table 2, Fig. 5D).

Heterotrophically acquired carbon

After 1 month of recovery, lipid δ13C enrichment did not differ significantly between bleached and non-bleached control P. compressa and M. capitata corals, nor did they change over the course of the 168 hour chase period in P. compressa (Table 2, Fig.

6A, C). Lipid δ13C enrichment values slightly (but significantly) decreased over the course of the chase in M. capitata. However, after 11 months recovery, lipid δ13C enrichment values were higher in bleached P. compressa corals relative to non-bleached control corals in the first 24 hours of the chase period, and no longer differed from controls after 168 hrs (Table 2, Fig. 6B). No significant treatment or chase effects were detected in M. capitata after 11 months of recovery (Table 2, Fig 6D).

Proportionate contribution of photoautotrophic and heterotrophically derived C to coral lipids

Three general patterns were observed in the proportionate contribution of photoautotrophic and heterotrophically derived C to coral lipids (Fig 7). First, 60-85% of lipids were derived from photosynthetically fixed C for both species under all conditions

(Fig. 7). Second, bleaching had no obvious effect on the proportionate contribution of photoautotrophic and heterotrophic C to holobiont lipids at any point in the study for either species (Fig. 7). Third, the proportionate contribution of heterotrophically acquired

C to total lipids was two times higher at 1 month recovery at 30% (October) (Fig 7A, C) than after 11 months of recovery at 15% (August) (Fig 7B, D). All three patterns were similar in both species.

Discussion

Photoautotrophically acquired carbon

Overall, average lipid δ13C enrichment values in the first 24hr of the chase ranged from 5-30‰ in P. compressa (Fig 5A, B) and from 0-10‰ in M. capitata (Fig 5C, D).

This indicates that all but the bleached M. capitata corals at 1 month of recovery rapidly incorporated photosynthetically derived C into their lipids. Thus, bleached M. capitata either do not build new lipids at all when bleached, or they utilize heterotrophic C to do so (see next section for details).

More specifically, after 1 month of recovery, average lipid δ13C enrichment values of photoautotrophically labeled corals were significantly lower in bleached than in non-bleached control corals of both species in the first 24hrs of the chase (Fig. 5A, C).

These findings indicate that when photosynthesis rates are depressed during bleaching

(Rodrigues & Grottoli 2007), not only is there less photosynthetically fixed carbon being incorporated into the animal host and endosymbiotic algae (Hughes et al 2010), but there is less being incorporated into coral lipids. After 168 hrs, the decline observed in average lipid δ13C enrichment values of non-bleached fragments of P. compressa (Fig 5A) could be due to one of the following: 1) loss of lipids through mucus secretions (Brown and

Bythell 2005), 2) lipid catabolism (Grottoli et al 2004; Grottoli & Rodrigues 2011), and/or 3) dilution of the lipid δ13C values with newly assimilated non-labeled lipids.

Since existing C stores (as opposed to newly fixed C) are the source for dissolved organic carbon (DOC) secretions in corals (Tanaka et al. 2008) and lipids are a significant source 16 of the C for DOC and mucous secretions (Brown & Bythell 2005), it is unlikely that the rapid decrease in enrichment is primarily due to loss of C through mucus production. In addition, photoautotrophically acquired C has been shown to be preferentially used for short-term respiratory needs in the cnidarian Aiptasia sp. (Bachar et al. 2007) and in P. compressa (Hughes et al. 2010). Thus, it is likely that control P. compressa catabolizes newly synthesized lipids assimilated from photosynthetically derived C to meet daily metabolic need at 1 month recovery. Any dilution of the lipid δ13C enrichment values with newly synthesized unlabeled lipids would further enhance the decrease observed over the 168hrs.

Over the course of the 168 chase, M. capitata corals maintained their lipid δ13C enrichment values irrespective of treatment status (Fig 5, table 2). This is consistent with findings that this species maintains its lipids when naturally bleached (Grottoli et al

2004), has a lower metabolic rate than P. compressa (Coles and Jokiel 1977; Rodrigues and Grottoli 2007), and can dramatically increase feeding rates when bleached (Grottoli et al 2006; Rodrigues & Grottoli 2007).

After 11 months, lipid δ13C enrichment levels of bleached and non-bleached control corals of both species did not differ in the first 24 hours of the chase. This indicates that the rate of lipid synthesis from photoautotrophically acquired C had fully recovered in the bleached corals (Fig 5B, D). The findings are consistent with previous studies showing that both of these coral species fully recover their lipid, protein, and carbohydrate concentrations within 8 months of recovery (Rodrigues & Grottoli 2007;

Rodrigues et al 2008; Grottoli & Rodrigues 2011) and that the photosynthetic pathway is

17 restored within 4 months of recovery (Hughes & Grottoli submitted). In addition, lipid

δ13C enrichment values declined over the 168hr chase in non-bleached M. capitata, but in none of the other cases (Fig 5B, D). This decline is most likely due a combination of lipid catabolism and the dilution of the lipid δ13C values with newly assimilated non- labeled lipids since mucus and DOC secretions are not produced from newly synthesized lipids (Tanaka et al. 2008).

Heterotrophically acquired carbon

Overall, average lipid δ13C enrichment values in the first 24hr of the chase ranged from 49-145‰ in P. compressa (Fig 6A, B) and from 3-42‰ in M. capitata (Fig 6C, D).

This indicates that all corals rapidly incorporated at least some heterotrophically derived

C into their lipids. Thus for all but bleached M. capitata, both heterotrophic and photoautotrophic C is used for lipid synthesis. In the case of bleached M. capitata, thought the rate of lipid synthesis is low, it is almost entirely dependent on heterotrophic

Examination of the results in more detail reveals that after 1 month of recovery, average lipid δ13C enrichment levels did not differ significantly between bleached and control corals of both species indicating that the allocation of heterotrophic C to lipids was the same irrespective of bleaching status (Fig. 6A, C). δ13C enrichment values of bulk animal host and endosymbiont tissues were also not significantly different from one another at 1 month recovery in either species (Hughes and Grottoli Submitted). This is of interest because feeding rates in bleached M. capitata are three times greater than in non-

18 bleached controls (Grottoli et al 2006; Palardy et al 2008). Considering that the enrichment values of holobiont lipids in the first 24 hours in both bleached and non- bleached M. capitata are quite low (i.e., 2-3‰) and that the δ13C values of the endosymbiotic algae and host tissues more than twice as high (i.e., 6-7‰) (Hughes &

Grottoli unpublished), it appears that of the small quantities of heterotrophically acquired

C assimilated by the holobiont, half of it is used for lipid synthesis. This implies that bleached M. capitata must rapidly respire much of the fixed C acquired via heterotrophy to meet metabolic demand and maintain lipid synthesis rates equal to those of non- bleached corals.

Over the course of the 168 hr chase in the 1 month recovery corals, lipid δ13C enrichment values were maintained in P. compressa but declined in M. capitata (Table 3,

Fig 6A, C). Since heterotrophically derived C is maintained in the bulk tissues of healthy and bleached M. capitata (Hughes et al 2010) but declines in the lipids (this study), it holds that carbohydrates and/or proteins must be sufficiently synthesized to offset lipid catabolism and still maintain δ13C enrichment levels in animal host and endosymbiotic algal bulk tissues. Conversely, since heterotrophically derived C is catabolized in the animal host and endosymbiotic algal bulk tissues of P. compressa (Hughes et al 2010), but lipid δ13C enrichment values are maintained (this study), it holds that carbohydrates and/or proteins are the energy reserve component catabolized when the coral is bleached.

Therefore, heterotrophically derived lipids are used in meeting daily metabolic needs in bleached and healthy M. capitata, but are used for storage in bleached and healthy P. compressa.

In P. compressa at 11 months recovery, bulk animal host and endosymbiont δ13C enrichment values were 75‰ more enriched in bleached than in non-bleached control corals (Hughes & Grottoli submitted), while lipids were 100‰ more enriched in bleached than non-bleached controls (Fig 6B). This indicates that for bleached P. compressa, the heterotrophically derived C enrichment observed in the bulk tissue is mostly due to lipids.

It has previously been hypothesized that such an increase in heterotrophic C assimilation during long-term recovery can be either 1) an adaptive response to bleaching, that may enhance production through heterotrophy, thereby increasing resilience to future bleaching events (Hughes and Grottoli Submitted), or 2) a sign that the corals are still recovering and sequestering disproportionate amounts of heterotrophically derived C for ongoing tissue repair. Such a strategy could be adaptive because corals with higher lipid concentrations are more resilient to, and recover faster from bleaching (Anthony et al.

2009). With an adaptive response, corals such as P. compressa would be more resilient to future bleaching events, as they would have more robust energy stores. It is also possible that despite the recovery of other physiological parameters (i.e. proteins, carbohydrates, tissue biomass, endosymbiont density, Chla concentrations (Rodrigues and Grottoli

2007)), the corals are still recovering after 11 months. Previous studies on the Caribbean coral Montastraea annularis (now Orbicella faveolata, (Budd et al. 2012)) have shown that even after the recovery of endosymbionts, lipid and protein energy reserves take over one year to fully recover (Szmant and Gassman 1990; Fitt et al. 1993). It has also been shown that while total lipid concentrations recover (Rodrigues and Grottoli 2007), several lipid class values (triacylglycerides, phospholipids, and wax esters) are low after 8

20 months of recovery (Rodrigues et al. 2008), indicating that lipids may not be fully recovered at 11 months either. If this overcompensation indicates that the coral is not yet fully recovered, it would mean that corals such as P. compressa are more at risk, as bleaching events increase in frequency and intensity (Hoegh-Guldberg 1999; Wilkinson

2000; Buddemeier et al. 2004; Wooldridge et al. 2005; Donner 2009; Frieler et al. 2012), and are predicted to become annual by the second half of this century (Donner et al.

2007; Frieler et al. 2012; van Hooidonk et al. 2013).

In M. capitata at 11 months recovery, bulk animal host and endosymbiont d13C enrichement values were 80‰ higher in bleached than in non-bleached control corals

(Hughes & Grottoli submitted), while lipid δ13C enrichment was statistically no different between bleached and control coral fragments (Fig 6D). This indicates that for M. capitata, the elevated sequestration of heterotrophically derived C observed in the total tissue of bleached corals (Hughes et al 2010) is primarily allocated to other energy reserves such as carbohydrates and/or proteins. This is reasonable as previous studies have shown that the bulk of the lipid in M. capitata is comprised of lipid classes that are used for structural purposes (sterols and phospholipids), and not energy reserves (wax esters and triacylglycerides) (Rodrigues et al. 2008). Further experiments examining the assimilation and catabolism of C over the course of bleaching and recovery in protein and carbohydrate are needed to fully assess how corals are allocating C resources when healthy and bleached, and to determine if corals are still recovering after 11 months or if the high levels of heterotrophic C assimilation into lipids and bulk tissues is a sigh that they have adapted.

Proportionate contribution of photoautotrophically and heterotrophically acquired carbon

Photoautotrophically derived C constituted the majority of C used in lipid synthesis and the proportion of both sources of fixed C to lipids did not differ between bleached and non-bleached control corals at a given recovery month (Fig 7). As such, the data imply that irrespective of feeding rates or bleaching status, there appears to be a physiological setpoint for the ratio of photoautotrophic C to heterotrophic C needed for lipid synthesis at any given time.

Interestingly, the proportion of heterotrophic C used for lipid synthesis was not static over time – it was double in October (i.e., 1 month recovery) what it was in August

(i.e., 11 months recovery) – regardless of bleaching status (Fig 7). This variation is most likely related to spawning. M. capitata spawns normally one year after a bleaching event presumably because it acquired more C by increasing feeding rates (Cox 2007). In P. compressa, bleaching leads to significant decreases in gamete production, but not cessation (Sudek et al. 2012). Therefore, it is likely that P. compressa also spawns after a bleaching event, just not as effectively as it would if healthy. While spawning does not have an effect on lipid concentrations in P.compressa (Stimson 1987) or M. capitata

(Padilla-Gamino and Gates 2012), it seems to effect the percent contribution of photoautotrophic and heterotrophic C to lipids. Increased demand on lipid reserves due to spawning in June, July, and August appears to increase the demand for heterotrophically derived C in the lipid synthesis. Therefore the ratio of photoautotrophically and

22 heterotrophically derived C utilized for lipid synthesis is relatively constant at any given point in time (i.e., within each recovery interval), but varies seasonally most likely in response to spawning. Further research exploring the effects of bleaching and spawning on coral energy reserves and lipid classes is needed to verify if spawning is the cause of the change in the proportionate contribution of heterotrophic C to coral lipids. This will allow us to further understand the dynamics of coral lipid synthesis and determine the relative importance of photoautotrophic and heterotrophic C to sexual reproduction in these corals.

Summary

Overall, bleached Porites compressa maintain newly synthesized lipids that are built with heterotrophically and photoautotrophically derived C (present study) but catabolize pre-existing lipid stores (Rodrigues & Grottoli 2007; Grottoli et al 2004) primarily in the form of triacylglycerides and phospholipids (Rodrigues et al 2008).

Since these corals do catabolize energy stores when bleached (Hughes et al 2010), we hypothesize that this species must be catabolizing newly synthesized protein and carbohydrates along with existing energy stores (Rodrigues & Grottoli 2007) to fuel metabolic demand when bleached.

After 11 months of recovery, bleached P. compressa synthesize significantly more lipids derived from heterotrophic C than non-bleached controls, as indicated by significantly higher enrichment values in bleached corals compared to controls (Fig. 6), and fully recover their total energy reserves (Rodrigues & Grottoli 2007), yet have lower

23 triacylglycerides, wax esters, and phospholipids after 8 months of recovery (Rodrigues et al 2008). Since these corals are clearly building tissue and/or energy reserves at a much higher rate than non-bleached controls (Hughes et al 2010), it would seem that these energy reserves are in the form of proteins and/or carbohydrates. In addition, due to the lack of recovery of the previously mentioned lipid classes, it would also appear that the enhanced lipid synthesis observed in bleached corals after 11 months of recovery is in fact an indication that the corals have not fully recovered. If P. compressa is in fact not fully recovered 11 months after a bleaching event, it appears to be at serious risk in the future as bleaching events become more frequent.

Bleached M. capitata also maintain newly synthesized lipids built with photoautotrophically acquired C (present study), maintain their existing lipid stores

(Grottoli .et al 2004), and retain newly synthesized bulk tissues derived from heterotrophic C (Hughes et al 2010) while at the same time catabolizing newly synthesized lipids built with heterotrophically acquired C (this study) in the form of phospholipids (Rodrigues et al 2008) despite a 10-fold increase in feeding rates (Grottoli et al 2006; Palardy et al 2008). Together these data indicate once again that bleached M. capitata respire the vast majority of their heterotrophically derived C to meet metabolic demand. By deduction it would also appear that the phospholipid bilayer of cell membranes are being damaged due to coral bleaching and that heterotrophically derived

C is needed for their repair.

After 11 months of recovery, for M. capitata, bleached coral lipid synthesis (this study), lipid metabolism (this study), lipid, protein, and carbohydrate concentrations

(Rodrigues & Grottoli 2007), and lipid class composition (Rodrigues et al 2008) have fully recovered. Since these corals are clearly building tissue and/or energy reserves built with heterotrophic C at a much higher rate than non-bleached corals (Hughes et al 2010), it would appear that proteins and carbohydrates are the molecules being synthesized.

Tissue building coupled with recovery of energy reserves, tissue biomass, and endosymbiont density (Rodrigues and Grottoli 2007) indicates that M. capitata is recovered at 11 months post-bleaching. Thus, M. capitata appears to be able to manage metabolic need and energy stores well by increasing heterotrophic C intake, likely allowing it to be resilient to future bleaching events. These results again highlight the importance of heterotrophic plasticity in the resilience of corals to bleaching (Grottoli et al. 2006; Ferrier-Pages et al. 2010).

Overall, M. capitata maintains most of its energy reserves and uses C from feeding and from newly synthesized lipids made with heterotrophically derived C to meet metabolic need at 1 month of recovery. M. capitata seems to recover faster than P. compressa, and looks to be more resilient to bleaching events. Additionally, lipids built with photoautotrophic C make up the bulk of the total lipids, but lipids built with heterotrophic C are present as well.

This is the first study to show that heterotrophic C is required for lipid synthesis and can account for 5-35% of C in lipids. This highlights the importance of heterotrophic

C in lipid synthesis, particularly post-spawning, as the percentage of lipids built with heterotrophic C increases after spawning season. Understanding the relative importance of lipids in corals following bleaching, and how lipid synthesis and metabolism are

25 affected by bleaching and recovery helps us to better understand how corals cope with stress events. This study adds to the growing depth of knowledge about coral bleaching physiology, which will allow scientists and policy makers to make informed conservation management decisions in the future, as threats to reefs such as mass bleaching continue to become more prevalent.

References

Anthony KRN (1999) Coral suspension feeding on fine particulate matter. Journal of Experimental Marine Biology and Ecology 232:85-106 Anthony KRN, Fabricius KE (2000) Shifting roles of heterotrophy and autotrophy in coral energetics under varying turbidity. Journal of Experimental Marine Biology and Ecology 252:221-253 Anthony KRN, Hoogenboom MO, Maynard JA, Grottoli AG, Middlebrook R (2009) Energetics approach to predicting mortality risk from environmental stress: a case study of coral bleaching. Funct Ecol 23:539-550 Bachar A, Achituv Y, Pastemak Z, Dubinsky Z (2007) Autotrophy versus heterotrophy: The origin of carbon determines its fate in a symbiotic sea anemone. Journal of Experimental Marine Biology and Ecology 349:295-298 Birsoy K, Festuccia WT, Laplante M (2013) A comparative perspective on lipid storage in animals. Journal of Cell Science 126:1541-1552 Brown BE, Bythell JC (2005) Perspectives on mucus secretion in reef corals. Marine Ecology Progress Series 296:291-309 Budd AF, Fukami H, Smith ND, Knowlton N (2012) Taxonomic classification of the reef coral family Mussidae (Cnidaria: Anthozoa: Scleractinia). Zool J Linn Soc 166:465-529 Buddemeier RW, Kleypas JA, Aronson RB (2004) Coral reefs and global climate change: potential contributions of climate change to stresses on coral reef ecosystems. Pew Center on Global Climate Change, Arlington, VA 56 Carricart-Ganivet JP, Cabanillas-Teran N, Cruz-Ortega I, Blanchon P (2012) Sensitivity of Calcification to Thermal Stress Varies among Genera of Massive Reef-Building Corals. PloS one 7 Christie W, Han X (2010) Lipid Analysis: Isolation, Separation, Identification and Lipidomic Analysis. Oily Press Coles SL, Jokiel PL (1977) Effects of temperature on photosynthesis and respiration in hermatypic corals. Mar Biol 43:209-216 Connolly SR, Lopez-Yglesias MA, Anthony KRN (2012) Food availability promotes rapid recovery from thermal stress in a scleractinian coral. Coral Reefs 31:951-960 D'Croz L, Mate JL, Oke JE (2001) Responses to elevated seawater temperature and UV radiation in the coral Porites lobata from upwelling and non-upwelling environments on the Pacific coast of Panama. Bulletin of Marine Science 69:203-214 deGroot PA (2004) Handbook of stable isotope analytical techniques. Elsevier, Amsterdam [u.a.] Donner SD (2009) Coping with commitment: projected thermal stress on coral reefs under different future scenarios. PloS one 4 Donner SD, Knutson TR, Oppenheimer M (2007) Model-based assessment of the role of human-induced climate change in the 2005 Caribbean coral bleaching event. Proceedings of the National Academy of Sciences of the United States of America 104:5483-5488 Falkowski PG, Dubinsky Z, Muscatine L, McCloskey L (1993) Population-control in symbiotic corals. Bioscience 43:606-611 Ferrier-Pages C, Rottier C, Beraud E, Levy O (2010) Experimental assessment of the feeding effort of three scleractinian coral species during a thermal stress: Effect on the rates of photosynthesis. Journal of Experimental Marine Biology and Ecology 390:118-124 Fitt WK, Spero HJ, Halas J, White MW, Porter JW (1993) Recovery of the coral Montastrea annularis in the Florida Keys after the 1987 Caribbean "bleaching event". Coral Reefs 12:57-64 Frieler K, Meinshausen M, Golly A, Mengel M, Lebek K, Donner SD, Hoegh-Guldberg O (2012) Limiting global warming to 2 °C is unlikely to save most coral reefs. Nature Climate change Nature Climate Change 3:165-170

Glynn P (1996) Coral reef bleaching: facts, hypotheses and implications. Global Change Biology 2:495- 509 Goreau TF, Yonge CM, Goreau NI (1971) Reef coral- autotrophs or heterotrophs? Biol Bull 141:247-& Grottoli AG, Rodrigues LJ (2011) Bleached Porites compressa and Montipora capitata corals catabolize delta C-13-enriched lipids. Coral Reefs 30:687-692 Grottoli AG, Rodrigues LJ, Juarez C (2004a) Lipids and stable carbon isotopes in two species of Hawaiian corals, Porites compressa and Montipora verrucosa, following a bleaching event. Mar Biol 145:621-631 Grottoli AG, Rodrigues LJ, Juarez C (2004b) Effect of bleaching on the coral host total lipid content, lipid class composition, and skeletal stable isotopic composition. In: The effects of combined sea temperature, light, and carbon dioxide on coral bleaching, settlement and growth In: Hendee JC (ed) NOAA Research Special Report Silver Spring, MD, pp661-664 Grottoli AG, Rodrigues LJ, Palardy JE (2006) Heterotrophic plasticity and resilience in bleached corals. Nature 440:1186-1189 Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the world's coral reefs. Marine Freshwater Research 50:839-866 Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, Gomez E, Harvell CD, Sale PF, Edwards AJ, Caldeira K, Knowlton N, Eakin CM, Iglesias-Prieto R, Muthiga N, Bradbury RH, Dubi A, Hatziolos ME (2007) Coral Reefs under Rapid Climate Change and Ocean Acidification. Science 318:1737-1742 Hughes AD, Grottoli AG (Submitted) Heterotrophic compensation: A possible mechanism for resilience of coral reefs to global warming, or a sign of prolonged stress? PLoS ONE Hughes AD, Grottoli AG, Pease TK, Matsui Y (2010) Acquisition and assimilation of carbon in non- bleached and bleached corals. Marine Ecology-Progress Series 420:91-101 Hughes TP, Baird AH, Bellwood DR, Card M, Connolly SR, Folke C, Grosberg R, Hoegh-Guldberg O, Jackson JBC, Kleypas J, Lough JM, Marshall P, Nystrom M, Palumbi SR, Pandolfi JM, Rosen B, Roughgarden J (2003) Climate change, human impacts, and the resilience of coral reefs. Science 301:929-933 Jokiel PL, Coles SL (1990) Response of Hawaiian and other Indo-Pacific reef corals to elevated temperature. Coral Reefs 8:155-162 Leder JJ, Szmant AM, Swart PK (1991) The effect of prolonged bleaching on skeletal banding and stable isotopic composition in Montastrea annularis- preliminary observations. Coral Reefs 10:19-27 Lesser MP (1997) Oxidative stress causes coral bleaching during exposure to elevated temperatures. Coral Reefs 16:187-192 Levas SJ (2012) Biogeochemistry and physiology of bleached and recoverying Hawaiian and Caribbean corals Levas SJ, Grottoli AG, Hughes A, Osburn CL, Matsui Y (2013) Physiological and Biogeochemical Traits of Bleaching and Recovery in the Mounding Species of Coral Porites lobata: Implications for Resilience in Mounding Corals. PLoS ONE 8:e63267 Moran MD (2003) Arguments for rejecting the sequential Bonferri in ecological studies. Oikos 100:403- 405 Muscatine L, McCloskey LR, Marian RE (1981) Estimating the daily contribution of carbon from zooxanthellae to coral animal respiration. Limnology and Oceanography 26:601-611 Omori M, Fukami H, Kobinata H, Hatta M (1999) Significant drop of fertilization of Acropora corals in 1999: an after-effect of heavy coral bleaching? Limnology and Oceanography 46:704-706 Padilla-Gamino JL, Gates RD (2012) Spawning dynamics in the Hawaiian reef-building coral Montipora capitata. Marine Ecology Progress Series 449:145-U176 Palardy JE, Rodrigues LJ, Grottoli AG (2008) The importance of zooplankton to the daily metabolic carbon requirements of healthy and bleached corals at two depths. Journal of Experimental Marine Biology and Ecology 367:180-188 Patton JS, Abraham S, Benson AA (1977) Lipogenesis in the intact coral Pocillopora capitata and its isolated zooxanthellae: Evidence for a light-driven carbon cycle between symbiont and host. Mar Biol 44:235-247 28

Porter JW, Fitt WK, Spero HJ, Rogers CS, White MW (1989) Bleaching in Reef Corals: Physiological and Stable Isotopic Responses. Proceedings of the National Academy of Sciences of the United States of America 86:9342-9346 Quinn GP, Keough MJ (2002) Experimental design and data analysis for biologists. Cambridge University Press, Cambridge, UK; New York Rodrigues LJ, Grottoli AG (2006) Calcification rate and the stable carbon, oxygen, and nitrogen isotopes in the skeleton, host tissue, and zooxanthellae of bleached and recovering Hawaiian corals. Geochimica Et Cosmochimica Acta 70:2781-2789 Rodrigues LJ, Grottoli AG (2007) Energy reserves and metabolism as indicators of coral recovery from bleaching. Limnology and Oceanography 52:1874-1882 Rodrigues LJ, Grottoli AG, Pease TK (2008) Lipid class composition of bleached and recovering Porites compressa (Dana, 1846) and Montipora capitata (Dana, 1846) corals from Hawaii. Journal of Experimental Marine Biology and Ecology 358:136-143 Stanley GD, Jr. (2003) The evolution of modern corals and their early history. Earth-Science Reviews 60:195-225 Stimson JS (1987) Location, quantity and rate of change in quantity of lipids in tissue of Hawaiian hermatypic corlas. Bulletin of Marine Science 41:889-904 Sudek M, Aeby GS, Davy SK (2012) Localized bleaching in Hawaii causes tissue loss and a reduction in the number of gametes in Porites compressa. Coral Reefs 31:351-355 Szmant AM, Gassman NJ (1990) The effects of prolonged bleaching on the tissue biomass and reporduction of the reef coral Montastrea annularis. Coral Reefs 8:217-224 Tanaka Y, Miyajima T, Koike I, Hayashibara T, Ogawa H (2008) Production of dissolved and particulate organic matter by the reef-building corals Porites cylindrica and Acropora pulchra. Bulletin of Marine Science 82:237-245 Tremblay P, Peirano A, Ferrier-Pages C (2011) Heterotrophy in the Mediterranean symbiotic coral Cladocora caespitosa: comparison with two other scleractinian species. Marine Ecology-Progress Series 422:165-177 Tremblay P, Naumann MS, Sikorski S, Grover R, Ferrier-Pagès C (2012) Experimental assessment of organic carbon fluxes in the scleractinian coral Stylophora pistillata during a thermal and photo stress event. MARINE ECOLOGY PROGRESS SERIES 453:63-77 van Hooidonk R, Maynard JA, Planes S (2013) Temporary refugia for coral reefs in a warming world. Nature Climate Change 3:508-511 Veron JEN, Hoeg-Guldberg O, Lenton TM, Lough JM, Obura DO, Pearce-Kelly P, Sheppard CRC, Spalding M, Stafford-Smith MG, Rogers AD (2009) The coral reef crisis: The critical importance of <350 ppm CO2. Marine Pollution Bulletin 58:1428-1436 Ward S (1995) Two patterns of energy allocation for growth, reproduction and lipid storage in the scleractinian coral Pocillopora damicornis. Coral Reefs 14:87-90 Ward S, Harrison PJ, Hoegh-Guldberg O (2000) Coral bleaching reduces reproduction of scleractinian corals and increases susceptibility to future stress. Proceedings of the 9th International Coral Reef Symposium 2:1123-1128 Warner ME, Fitt WK, Schmidt GW (1996) The effects of elevated temperature on the photosynthetic efficiency of zooxanthellae in hospite from four different species of reef coral: a novel approach. Plant, Cell & Environment 19:291-299 Wilkinson C (2000) Status of Coral Reefs of the World: 2000. Australian Institute for Marine Science, Townsville Wooldridge S, Done T, Berkelmans R, Jones R, Marshall P (2005) Precursors for resilience in coral communities in a warming climate: a belief network approach. Marine Ecology Progress Series 295:157-169

Appendix A: Tables

Table 1: Montipora capitata and Porites compressa Chl a concentrations. Analysis of variance (ANOVA) of the Chl a concentrations for bleached and non-bleached control corals at each recovery interval.

Montipora capitata Porites compressa Source DF SS F P DF SS F P Treatment 1 10252965 188.54 <0.001 1 26174978 34.47 <0.001 Residual 19 1033234 41 31133390 Total 20 11286199 42 57308368 Recovery T=0

Montipora capitata Porites compressa Source DF SS F P DF SS F P Treatment 1 9481689 86.62 <0.001 1 2683013 15.59 0.001 Residual 18 1970404 18 3098363 Total 19 11452093 19 5781376 Recovery T=1

Montipora capitata Porites compressa Source DF SS F P DF SS F P Treatment 1 276678 1.24 0.28 1 3 0.001 0.996 Residual 18 4009815 17 1794724 Total 19 4286493 18 1794727 Recovery T=11

Table 2: Montipora capitata and Porites compressa DI13C enrichment. Three-way analysis of variance (ANOVA) of the effect of treatment (bleached vs. non-bleached

30 control corals), chase (first 24 hrs vs. 168 hrs), and coral genotype on δ13C enrichment above baseline in DI13C pulse-labeled P. compressa after 1 (n= 19) and 11 (n=17) months of recovery and M. capitata after 1 (n=19) and 11 (n=17) months of recovery. Significant effects are in bold (p<0.05).

Recovery 1 month 11 months Source SS DF F p SS DF F p P. compressa

Model 2743.17 7 13.05 0.0002 419.77 7 2.06 0.1544

Treatment 1228.35 1 40.89 0.0001 54.43 1 1.87 0.2046

Chase 893.51 1 29.74 0.0002 105.53 1 3.63 0.0893

Geno 83.16 4 0.69 0.6127 261.31 4 2.25 0.1442

Treat*Chase 328.32 1 10.93 0.0070 7.66 1 0.26 0.6202

M. capitata

Model 2.04 7 3.37 0.0355 251.68 7 2.16 0.1400

Treatment 0.82 1 9.53 0.0103 0.14 1 0.01 0.9286

Chase 0.03 1 0.32 0.5829 86.67 1 5.20 0.0485

Geno 0.92 4 2.66 0.0893 98.81 4 1.48 0.2857

Treat*Chase 0.05 1 0.58 0.4612 4.82 1 0.29 0.6037

Table 3: Montipora capitata and Porites compressa 13C-rotifer enrichment. Three-way analysis of variance (ANOVA) of the effect of treatment (bleached vs. non-bleached controls), chase (first 24 hours vs. 168 hours), and coral genotype on δ13C enrichment above baseline in 13C-rotifer labeled P. compressa after 1 (n= 17) and 11 (n=17) months of recovery, and M. capitata after 1 (n=16) and 11 (n=15) months of recovery.

Significant effects are in bold (p<0.05).

Recovery 1 month d 11 months Source SS DF F p SS DF F p P. compressa N 17 17

Model 4978.76 7 0.87 0.5612 42983.52 7 1.92 0.1776

Treatment 1429.77 1 1.75 0.2179 15946.94 1 5.00 0.0522

Chase 426.24 1 0.52 0.4879 8844.40 1 2.77 0.1303

Geno 2060.85 4 0.63 0.6520 7767.09 4 0.61 0.6668

Treat*Chase 378.63 1 0.46 0.5126 4000.00 1 1.25 0.2919

M. capitata N 16 15

Model 38.78 7 5.59 0.0136 2660.68 7 1.47 0.3131

Treatment 2.91 1 2.94 0.1249 386.86 1 1.49 0.2414

Chase 16.99 1 17.1 0.0033 486.23 1 1.88 0.2312

Geno 15.17 4 3.82 0.0504 1341.98 4 1.29 0.3584

Treat*Chase 0.00 1 0.00 0.9604 252.34 1 .097 0.3567 32

Appendix B: Figures

4 (1 mo rec) 2 DIC- Pulse (chase 24, 168 hrs) 2 Rotifer-Pulse (chase 24, 168 hrs) 8 bleached

4 (11 mo rec) 2 DIC- Pulse (chase 24, 168 hrs) 2 Rotifer-Pulse (chase 24, 168 hrs)

X 16 fragments

P. compressa 2 DIC- Pulse (chase 24, 168 hrs) parent colony 4 (1 mo rec) 2 Rotifer-Pulse (chase 24, 168 hrs) 8 non-bleached X 5 colonies/species controls 4 (11 mo rec) 2 DIC- Pulse (chase 24, 168 hrs) X 2 species 2 Rotifer-Pulse (chase 24, 168 hrs)

Figure 1: Flow diagram of experimental design. This method was used for both Porites compressa and Montipora capitata corals. Mo rec = months of recovery on reef, DIC-

13 13 pulse = C-labeled dissolved inorganic carbon pulse-chase, Rot-pulse = C-labeled rotifer pulse-chase, hrs = hours. Figure adapted from Hughes and Grottoli, submitted.

Reef Tank Control 30 Tank Treatment

o 28

26 * *

Temperature ( Temperature 24

01-Jul-07

01-Oct-06 01-Apr-07

01-Jan-07 01-Jun-07

01-Mar-07

01-Feb-07

01-Nov-06

01-Dec-06

01-Aug-06 01-Aug-07

01-Sep-06 01-Sep-07

01-May-07

Date

Figure 2: Average daily temperatures on the reef (grey), in the non-bleached control tanks (black), and in the treatment tanks (white). Error bars are the same size or smaller than the symbols. * indicate 1 and 11 months recovery time points, respectively. Figure from Hughes and Grottoli, submitted.

1 Month Recovery 11 Months Recovery

Control Bleached Control Bleached

compressa Porites

Control Bleached Control Bleached capitata

Montipora

Figure 3: Photographs of representative coral fragments of bleached and non- bleached control Porites compressa and Montipora capitata following 1 and 11 months of recovery. Photographs by A.G. Grottoli and A. Hughes.

Porites compressa Montipora capitata 2500 A Control B Bleached 2000

1500

± 1 SE ± 1 (µg/gdw) a 1000 *

500 * * *

Chlorophyll Chlorophyll 0 0 1 11 0 1 11 Recovery (months)

Figure 4: Average chlorophyll a concentration (± 1 standard error (SE)) of non-bleached control (black) and bleached (gray) A) Montipora capitata and B) Porites compressa corals. Within each species, statistically significant differences between non-bleached control and bleached corals at each recovery interval are indicated with an *. Figure adapted from Hughes and Grottoli (submitted).

DI13C Label 1 Month Recovery 11 Months Recovery A * Control B

30 (5) Bleached DB) DB)

20 (5)

* (5) (5) C (‰, VP (‰, C 10

Porites compressa Porites (2) 13 (5) †

(± 1 S.E.) 1 (± (4)

0 (5) 30 C D 25

nrichment of δ of nrichment 20 15 10 * (5) (4)

Montipora capitata Montipora (5) (5) relative to baselines to relative 5 (5) (5) † (2) 0 (5)

-5 Average Lipid e Lipid Average 24 168 24 168 Chase (hours)

Figure 5: Photoautotrophic C assimilation in compressa and Montipora capitata lipids.

Average (± 1 S.E.) lipid δ13C enrichment of holobiont lipids 24 and 168 hours after pulse-

labeling with 13C-labeled dissolved inorganic carbon in seawater. Porites compressa

after A) 1 month and B) 11 months of recovery and Montipora capitata after C) 1 month

and D) 11 months of recovery. Bleached = open circles, non-bleached control = filled

circles, VPDB = Vienna Peedee Belemnite Limestone Standard. Numbers in parenthesis

are sample size (n) for each average. * indicate significant differences between control

and bleached treatments within a chase interval. † indicate significant differences

between chase intervals within a treatment.

D13C Rotifer Label 1 Month Recovery 11 Months Recovery

A Control B Bleached

150 (5)

C (‰, (‰, C 100 13

(5) *

(± 1 S.E.) 1 (±

(4) (5) Porites compressa Porites 50 (4) (4) (4) (3) 60 C D 50 nrichment of δ of nrichment 40 (4) 30 (5) (2) 20 (4)

5 (5)

Montipora capitata Montipora DB) relative to baselines to relative DB)

(5) Average Lipid e Lipid Average VP (3) † 0 (3) †

24 168 24 168 Chase (hours)

Figure 6: Heterotrophic C assimilation in Porites compressa and Montipora capitata

lipids. Average (± 1 S.E.) lipid δ13C enrichment of holobiont lipids 24 and 168 hours

after pulse-labeling with 13C-labeled rotifers. Porites compressa after A) 1 month and B)

11 months of recovery and Montipora capitata after C) 1 month and D) 11 months of

recovery. Bleached = open circles, non-bleached control = filled circles, VPDB =

Vienna Peedee Belemnite Limestone Standard. Numbers in parenthesis are sample size

(n) for each average. * indicate significant differences between control and bleached

treatments within a chase interval. † indicate significant differences between 24 and 168

hour chase intervals within a treatment. 38

1 Month Recovery 11 Month Recovery 140 %lipid from DIC A B %lipid from Rot 120 100 80 60 40

Porites compressa 20 0 140 C D 120 100 80 60 40

Montipora capitataMontipora Average Percent of Newly Synthesized Total Lipid Total Synthesized Percent ofAverage Newly 1 S.E.) +/- (%, 20 0 24 Hours 168 hours 24 Hours 168 hours Control Bleached Control Bleached Control Bleached Control Bleached

Figure 7: Average (± 1 S.E.) percent contribution of carbon from DIC and heterotrophy to holobiont total lipids 24 and 168 hours after pulse-labeling in Porites compressa after

A) 1 month and B) 11 months of recovery and in Montipora capitata after C) 1 month and D) 11 months of recovery.

Appendix C: Raw Lipid δ13C Data

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Porites compressa P1 0 -15.44125498 0 Porites compressa P2 0 -15.1294687 0 Porites compressa P3 0 -15.47223375 0 Porites compressa P4 0 -17.65673635 0 Porites compressa P5 0 -13.83435645 0 Porites compressa PCBR02042 4 124.502352 140.009162 Porites compressa PCBR02043 4 69.0638628 84.57067285 Porites compressa PCBR02045 4 11.6957649 27.20257495 Porites compressa PCBR02121 12 4.914226634 20.42103668 Porites compressa PCBR02122 12 0.406805949 15.91361599 Porites compressa PCBR02125 12 -15.57713972 -0.07032967 Porites compressa PCBR02241 24 4.495493437 20.00230348 Porites compressa PCBR02242 24 61.31441809 76.82122813 Porites compressa PCBR02245 24 2.379826008 17.88663605 Porites compressa PCBR02 1 average 4, 12, 24 4.704860036 0.2960891 20.21167008 0.296089083 Porites compressa PCBR02 2 average 4, 12, 24 62.07452533 62.051265 77.58133538 62.05126475 Porites compressa PCBR02 3 average 4, 12, 24 69.0638628 0 84.57067285 0 Porites compressa PCBR02 5 average 4, 12, 24 -0.500516269 13.862724 15.00629378 13.86272377 Porites compressa PCBR02921 168 53.33900285 68.8458129 Porites compressa PCBR02922 168 -5.191474534 10.31533551 Porites compressa PCBR02923 168 36.63414984 52.14095989 Porites compressa PCBR02925 168 52.89705966 68.4038697 Porites compressa PCBR14042 4 45.97864383 61.48545387 Porites compressa PCBR14045 4 223.3380701 238.8448802 Porites compressa PCBR14121 12 52.70961242 68.21642246 Porites compressa PCBR14122 12 60.04815824 75.55496828 Porites compressa PCBR14123 12 214.6772256 230.1840356

Full Code: PC= Porites compressa, MC= Montipora capitata, B= bleached, N= non- bleached control, R= rotifer label, D= DIC label, 02, 11= months recovery, 4, 12, 24,

168= chase interval (hours), 1-5= genotype.

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Porites compressa PCBR14241 24 61.48701338 76.99382343 Porites compressa PCBR14242 24 -10.80942994 4.69738011 Porites compressa PCBR14243 24 194.8815761 210.3883861 Porites compressa PCBR14 1 average 4, 12, 24 57.0983129 6.2065597 72.60512294 6.206559743 Porites compressa PCBR14 2 average 4, 12, 24 31.73912404 37.513629 47.24593409 37.51362943 Porites compressa PCBR14 3 average 4, 12, 24 204.7794008 13.997638 220.2862109 13.99763801 Porites compressa PCBR14 5 average 4, 12, 24 223.3380701 0 238.8448802 0 Porites compressa PCBR14922 168 25.0702753 40.57708535 Porites compressa PCBR14923 168 69.5376665 85.04447654 Porites compressa PCBR14925 168 57.55486634 73.06167639 Porites compressa PCNR02121 12 33.50655181 49.01336186 Porites compressa PCNR02123 12 40.48465496 55.991465 Porites compressa PCNR02124 12 60.35611577 75.86292582 Porites compressa PCNR02125 12 56.3676136 71.87442365 Porites compressa PCNR02921 168 37.95498008 53.46179013 Porites compressa PCNR02922 168 47.77771762 63.28452766 Porites compressa PCNR02923 168 100.0230028 115.5298129 Porites compressa PCNR02924 168 57.59145631 73.09826636 Porites compressa PCNR02925 168 81.0064469 96.51325695 Porites compressa PCNR14121 12 67.00826336 82.51507341 Porites compressa PCNR14122 12 26.32889176 41.83570181 Porites compressa PCNR14123 12 -0.822726613 14.68408343 Porites compressa PCNR14124 12 -7.681713984 7.825096062 Porites compressa PCNR14125 12 49.71112064 65.21793068 Porites compressa PCNR14921 168 46.3492478 61.85605785 Porites compressa PCNR14922 168 12.82341344 28.33022349 Porites compressa PCNR14923 168 -13.98885208 1.517957962 Porites compressa PCNR14924 168 14.2067367 29.71354674

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Porites compressa PCNR14925 168 -2.271113868 13.23569618 Porites compressa PCBD02041 4 -14.26042349 1.24638656 Porites compressa PCBD02042 4 0.120113137 15.62692318 Porites compressa PCBD02 2 average 4,12,24 -5.006504155 7.2501317 10.50030589 7.250131704 Porites compressa PCBD02242 24 -10.13312145 5.373688599 Porites compressa PCBD02243 24 -8.867722252 6.639087794 Porites compressa PCBD02245 24 -13.15307415 2.353735895 Porites compressa PCBD02921 168 -16.8719733 -1.36516326 Porites compressa PCBD02922 168 -14.59026656 0.916543489 Porites compressa PCBD02923 168 -14.58226759 0.924542461 Porites compressa PCBD02924 168 -16.47802393 -0.97121389 Porites compressa PCBD02925 168 -16.74298988 -1.23617983 Porites compressa PCBD14041 4 -5.745737361 9.761072685 Porites compressa PCBD14042 4 -4.365301875 11.14150817 Porites compressa PCBD14043 4 -2.58974174 12.91706831 Porites compressa PCBD14121 12 -12.27793625 3.228873795 Porites compressa PCBD14123 12 4.871267845 20.37807789 Porites compressa PCBD14124 12 -4.081882856 11.42492719 Porites compressa PCBD14125 12 -3.301782594 12.20502745 Porites compressa PCBD14241 24 1.470539156 16.9773492 Porites compressa PCBD14242 24 -9.203828284 6.302981763 Porites compressa PCBD14243 24 -1.360353905 14.14645614 Porites compressa PCBD14 1 average 24 -5.517711485 6.8770736 9.989098561 6.877073568 Porites compressa PCBD14 2 average 24 -6.784565079 3.4213548 8.722244967 2.419263204 Porites compressa PCBD14 3 average 24 0.3070574 4.0002327 15.81386745 4.000232662 Porites compressa PCBD14 4 average 24 -4.081882856 0 11.42492719 0 Porites compressa PCBD14 5 average 24 -3.301782594 0 12.20502745 0

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Porites compressa PCBD14922 168 -6.327171431 9.179638615 Porites compressa PCBD14923 168 -5.122016539 10.38479351 Porites compressa PCND02121 12 16.4928553 31.99966535 Porites compressa PCND02122 12 9.987925462 25.49473551 Porites compressa PCND02123 12 9.994926361 25.50173641 Porites compressa PCND02124 12 14.4464985 29.95330854 Porites compressa PCND02125 12 20.950587 36.45739705 Porites compressa PCND02921 168 -15.91509628 -0.40828624 Porites compressa PCND02922 168 -14.97221747 0.534592579 Porites compressa PCND02923 168 0.781321729 16.28813177 Porites compressa PCND02924 168 1.711615208 17.21842525 Porites compressa PCND02925 168 -11.47254399 4.034266051 Porites compressa PCND14121 12 -8.522574718 6.984235329 Porites compressa PCND14122 12 -5.52187351 9.984936536 Porites compressa PCND14123 12 18.72879342 34.23560346 Porites compressa PCND14124 12 5.876790144 21.38360019 Porites compressa PCND14125 12 -2.154931647 13.3518784 Porites compressa PCND14921 168 -6.553200481 8.953609565 Porites compressa PCND14922 168 -5.637082738 9.869727308 Porites compressa PCND14923 168 -2.835243474 12.67156657 Porites compressa PCND14924 168 -4.489935301 11.01687475 Porites compressa PCND14925 168 -7.793359873 7.713450174

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Montipora capitata M1 0 -13.86435985 0 Montipora capitata M2 0 -15.07550569 0 Montipora capitata M3 0 -17.38891993 0 Montipora capitata M4 0 -15.61113853 0 Montipora capitata MCBR02041 4 -13.11006058 2.374920419 Montipora capitata MCBR02042 4 -9.881041924 5.603939077 Montipora capitata MCBR02043 4 -16.05098836 -0.56600735 Montipora capitata MCBR02122 12 -14.26291613 1.222064873 Montipora capitata MCBR02124 12 -12.2874545 3.197526501 Montipora capitata MCBR02243 24 -13.51018681 1.974794191 Montipora capitata MCBR02244 24 -15.1156933 0.369287703 Montipora capitata MCBR02245 24 -12.31681034 3.168170665 Montipora capitata MCBR02 1 average 24 -13.11006058 0 2.374920419 0 Montipora capitata MCBR02 2 average 24 -12.07197903 3.098453 3.413001975 3.098452964 Montipora capitata MCBR02 3 average 24 -14.78058758 1.796618 0.704393418 1.796618003 Montipora capitata MCBR02 4 average 24 -13.7015739 1.9998668 1.783407102 1.999866833 Montipora capitata MCBR02 5 average 24 -12.31681034 0 3.168170665 0 Montipora capitata MCBR02921 168 -14.48321968 1.001761316 Montipora capitata MCBR02923 168 -16.33545774 -0.85047674 Montipora capitata MCBR02925 168 -15.8713981 -0.3864171 Montipora capitata MCBR14043 4 40.90013852 56.38511952 Montipora capitata MCBR14044 4 48.47753235 63.96251335 Montipora capitata MCBR14045 4 52.63439416 68.11937516 Montipora capitata MCBR14122 12 -7.347471059 8.137509941 Montipora capitata MCBR14123 12 6.902911531 22.38789253 Montipora capitata MCBR14124 12 21.2022159 36.6871969 Montipora capitata MCBR14242 24 0.821597402 16.3065784 Montipora capitata MCBR14244 24 2.397517164 17.88249816 Montipora capitata MCBR14245 24 72.03678188 87.52176288

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Montipora capitata MCBR14 2 average 24 -3.262936829 5.7764037 12.22204417 5.776403705 Montipora capitata MCBR14 3 average 24 23.90152502 24.03967 39.38650603 24.03966974 Montipora capitata MCBR14 4 average 24 24.02575514 23.169403 39.51073614 23.16940289 Montipora capitata MCBR14 5 average 24 62.33558802 13.71956 77.82056902 13.71955993 Montipora capitata MCBR14922 168 14.20518214 29.69016314 Montipora capitata MCBR14925 168 6.821875236 22.30685624 Montipora capitata MCNR02121 12 -10.51303985 4.971941149 Montipora capitata MCNR02122 12 -11.510773 3.974207997 Montipora capitata MCNR02123 12 -14.47596591 1.009015087 Montipora capitata MCNR02124 12 -14.0248605 1.460120502 Montipora capitata MCNR02125 12 -11.18119599 4.303785013 Montipora capitata MCNR02921 168 -13.17405142 2.310929576 Montipora capitata MCNR02923 168 -14.9672819 0.517699102 Montipora capitata MCNR02925 168 -15.82539219 -0.34041119 Montipora capitata MCNR14121 12 13.1961151 28.6810961 Montipora capitata MCNR14122 12 2.603376055 18.08835706 Montipora capitata MCNR14123 12 -9.272483459 6.212497542 Montipora capitata MCNR14124 12 5.464267178 20.94924818 Montipora capitata MCNR14125 12 26.09650813 41.58148913 Montipora capitata MCNR14922 168 10.27642249 25.76140349 Montipora capitata MCNR14923 168 -10.71998157 4.764999431 Montipora capitata MCNR14924 168 3.878556956 19.36353796 Montipora capitata MCNR14925 168 7.21805268 22.70303368 Montipora capitata MCBD02121 12 -12.16542598 3.319555017 Montipora capitata MCBD02122 12 -16.60246284 -1.11748184 Montipora capitata MCBD02123 12 -17.24461289 -1.75963189 Montipora capitata MCBD02124 12 -11.82334605 3.661634955 Montipora capitata MCBD02125 12 -16.96354721 -1.47856621

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Montipora capitata MCBD02921 168 7.600950392 23.08593139 Montipora capitata MCBD02922 168 -12.17349492 3.311486084 Montipora capitata MCBD02923 168 -16.31265767 -0.82767667 Montipora capitata MCBD02924 168 -13.62549962 1.859481382 Montipora capitata MCBD02925 168 -17.78764661 -2.30266561 Montipora capitata MCBD14042 4 -4.193900967 11.29108003 Montipora capitata MCBD14044 4 -4.564224473 10.92075653 Montipora capitata MCBD14045 4 -10.97109404 4.51388696 Montipora capitata MCBD14121 12 3.934147402 19.4191284 Montipora capitata MCBD14122 12 5.49262769 20.97760869 Montipora capitata MCBD14124 12 -2.377537208 13.10744379 Montipora capitata MCBD14243 24 -11.13116573 4.353815266 Montipora capitata MCBD14244 24 -8.234868515 7.250112486 Montipora capitata MCBD14245 24 -11.61038037 3.874600631 Montipora capitata MCBD14 1 average 24 3.934147402 0 19.4191284 0 Montipora capitata MCBD14 2 average 24 0.649363362 6.8494101 16.13434436 6.849410099 Montipora capitata MCBD14 3 average 24 -11.13116573 0 4.353815266 0 Montipora capitata MCBD14 4 average 24 -5.058876732 2.9598299 10.42610427 2.959829919 Montipora capitata MCBD14 5 average 24 -11.29073721 0.4520437 4.194243795 0.452043699 Montipora capitata MCBD14923 168 -11.18619038 4.298790621 Montipora capitata MCBD14925 168 -10.14072213 5.344258875 Montipora capitata MCND02121 12 -6.843182275 8.641798726 Montipora capitata MCND02122 12 -9.449791391 6.03518961 Montipora capitata MCND02123 12 -4.697619368 10.78736163 Montipora capitata MCND02124 12 -7.454325088 8.030655913

Enrichment Std Dev. Standard Chase above Enrichment δ13C VPDB Deviation Species Full Code interval Baseline above (‰) of δ13C (hours) δ13C VPDB Baseline δ13C VPDB (‰) (‰) VPDB (‰) Montipora capitata MCND02921 168 -8.642833218 6.842147782 Montipora capitata MCND02922 168 -10.2096578 5.275323197 Montipora capitata MCND02923 168 -9.916316204 5.568664797 Montipora capitata MCND02924 168 -9.712078367 5.772902634 Montipora capitata MCND02925 168 -10.81442651 4.67055449 Montipora capitata MCND14121 12 -2.827721665 12.65725934 Montipora capitata MCND14122 12 -3.253776424 12.23120458 Montipora capitata MCND14123 12 -2.814719994 12.67026101 Montipora capitata MCND14124 12 -8.895501525 6.589479476 Montipora capitata MCND14125 12 -0.122373961 15.36260704 Montipora capitata MCND14921 168 -9.195298681 6.28968232 Montipora capitata MCND14922 168 -6.75520579 8.72977521 Montipora capitata MCND14923 168 -12.0036697 3.481311301 Montipora capitata MCND14924 168 -9.611485086 5.873495914 Montipora capitata MCND14925 168 -11.8564906 3.628490401

Appendix D: Average Lipid δ13C Enrichment per Chase Interval

Mean Enrichment Recovery Chase interval above Species Treatment Interval Standard Error (hrs) Baseline (monhts) δ13C VPDB (‰) Porites compressa BR 1 24 49.3425 18.4075 Porites compressa BR 1 168 49.9265 13.7638 Porites compressa BR 11 24 144.7455 49.3892 Porites compressa BR 11 168 66.2277 13.2836 Porites compressa BD 1 24 4.3723 1.5074 Porites compressa BD 1 168 -0.3463 0.5211 Porites compressa BD 11 24 11.631 1.2047 Porites compressa BD 11 168 9.7822 0.6026 Montipora capitata BR 1 24 2.2888 0.4904 Montipora capitata BR 1 168 -0.0784 0.5564 Montipora capitata BR 11 24 42.235 13.4866 Montipora capitata BR 11 168 25.9985 3.6917 Montipora capitata BD 1 24 0.5251 1.2161 Montipora capitata BD 1 168 5.0253 4.6212 Montipora capitata BD 11 24 10.9055 3.0661 Montipora capitata BD 11 168 4.8215 0.5227 Porites compressa NR 1 24 63.1855 6.3824 Porites compressa NR 1 168 80.3775 11.3262 Porites compressa NR 11 24 42.4156 14.307 Porites compressa NR 11 168 26.9307 10.1559 Porites compressa ND 1 24 29.8814 2.0756 Porites compressa ND 1 168 7.5334 3.8389 Porites compressa ND 11 24 17.1881 4.8936 Porites compressa ND 11 168 10.045 0.8515 Montipora capitata NR 1 24 3.1438 0.799 Montipora capitata NR 1 168 0.8294 0.7811 Montipora capitata NR 11 24 23.1025 5.8636 Montipora capitata NR 11 168 18.1482 4.6484 Montipora capitata ND 1 24 8.3738 0.9783 Montipora capitata ND 1 168 5.6259 0.3564 Montipora capitata ND 11 24 11.9022 1.4399 Montipora capitata ND 11 168 5.6006 0.9675

BR= Bleached, rotifer label, BD= Bleached, DIC label, NR= Control, rotifer label, ND=

Control, DIC label 48

Appendix E: Average Percent Contribution of Heterotrophic and Photoautotrophic

C to Coral Lipids

Recovery Species Treatment Chase (hours) Lipid% DIC Std error Lipid% rot Std error (months) Porites compressa Control 1 24 67.4196 16.9959 32.5804 6.8851 Porites compressa Control 1 168 68.6614 15.3003 31.3386 19.3449 Porites compressa Bleached 1 24 69.4166 3.7772 30.5834 3.0198 Porites compressa Bleached 1 168 82.758 10.7268 17.242 1.8627 Porites compressa Control 11 24 81.8512 13.3643 18.1488 4.142 Porites compressa Control 11 168 87.3246 16.3268 12.6754 2.0213 Porites compressa Bleached 11 24 76.1378 17.5856 23.8622 7.7536 Porites compressa Bleached 11 168 85.5511 30.5809 14.9752 3.9596 Montipora capitata Control 1 24 66.1869 11.6331 33.8131 10.138 Montipora capitata Control 1 168 75.6737 21.4448 24.3263 6.5491 Montipora capitata Bleached 1 24 60.5774 6.1866 39.4226 5.9454 Montipora capitata Bleached 1 168 74.7854 24.8222 25.2146 8.1804 Montipora capitata Control 11 24 89.9049 18.6884 10.0951 4.6368 Montipora capitata Control 11 168 89.4685 31.1408 10.5315 3.0129 Montipora capitata Bleached 11 24 81.7042 3.8406 18.2958 4.3829 Montipora capitata Bleached 11 168 87.2353 20.2868 12.7647 4.7461

Appendix F: Lipid Extraction Method Comparison

Over the course of this project, two methods of lipid extraction were compared and tested for validity. Initially corals were airbrushed and total tissue was separated into endosymbiont and animal fractions. Lipids were then extracted using 2:1

Dichloromethane (DCM): Methanol. The lipid extractions were performed as stated in the above methods section, only DCM was used in place of Chloroform during all steps.

This method is published in “Lipid Analysis: Isolation, Separation, Identification and

Lipidomic Analysis” by Christie and Han, and has been used on other organisms in the past. In preparation for isotopic analysis, subsamples of total lipid extracts were dried down in tin capsules at 50°C under high-purity N2 gas to a constant weight and analyzed for δ13C as described in Grottoli and Rodrigues (2011). Each sample was combusted in a

Costech elemental analyzer (EA) and the resulting CO2 gas was automatically analyzed for δ13C with a Finnigan Delta IV stable isotope ratio mass spectrometer via a Finnigan

ConFlow III open-split interface. After analysis the data were reviewed, and many of the

DCM extracted samples had low %C values (between 0-30%). Since lipids are mainly C, values of 60% or higher were expected. Thus, the samples were not entirely lipid. In addition, the δ13C enrichment values were lower than that expected for natural abundance values (around -18‰), indicating a blank carbon contamination. Also, due to the separation of animal and endosymbiont fractions, the amount of signal that was picked up during analysis was very low, and it seemed likely that water background levels were causing drawdown of enrichment values (making the samples appear more depleted than they actually were). These issues lead us to rethink the method of lipid extraction.

Samples were reairbrushed and lipids were extracted using 2:1 Chloroform:

Methanol in the exact method outlined in this document. This method is standard and has been used for lipid extraction in previous publications (Rodrigues and Grottoli 2007,

Rodrigues and Grottoli 2008, Rodrigues and Grottoli 2011). Again, in preparation for isotopic analysis, subsamples of total lipid extracts were dried down in tin capsules at

13 50°C under high-purity N2 gas to a constant weight and analyzed for δ C as described in

Grottoli and Rodrigues (2011). Each sample was combusted in a Costech elemental

13 analyzer (EA) and the resulting CO2 gas was automatically analyzed for δ C with a

Finnigan Delta IV stable isotope ratio mass spectrometer via a Finnigan ConFlow III open-split interface. The results of these analysis showed that %C was in an acceptable range (>50%). The C signal picked up by the machine was much higher, and the δ13C enrichment values did not show the effects of drawdown as they did in the previous method. It was decided that this second run produced more reliable and realistic data, so only lipids extracted using the 2:1 Chloroform: Methanol method were used in the thesis.

A comparison was done examining the effect of method on δ13C enrichment. It is clear that there are differences between average lipid δ13C enrichment values based on which method was used for lipid extraction under many treatments and chase intervals

(Table 4). Results of the ANOVA show that lipid extraction method has a significant effect on δ13C enrichment values in P. compressa, but not M. capitata (Table 5, 6).

Further tests are needed to examine the exact significance of lipid extraction method, but it is likely that, at least for corals, extracting lipids with 2:1 Chloroform: Methanol is more representative than extracting with 2:1 DCM: Methanol.

Table 4: Average lipid δ13C enrichment values for Montipora capitata and Porites compressa using two different extraction methods (DCM: Methanol or

Chloroform:Methanol).

Chase Extraction interval 13C Species Treatment Label Method (hours) Mean δ (‰, VPDB) Std Error Montipora capitata Bleached DIC DCM 24 -13.75531 1.341024 Bleached DIC DCM 168 -19.24339 1.43832 Bleached DIC Chloroform 24 -14.959879 1.216133 Bleached DIC Chloroform 168 -10.45967 4.62116 Bleached Rotifer DCM 24 -18.120644 0.875675 Bleached Rotifer DCM 168 -23.842091 0.805354 Bleached Rotifer Chloroform 24 -13.196202 0.490424 Bleached Rotifer Chloroform 168 -15.563359 0.556436 Control DIC DCM 24 -0.145089 3.535457 Control DIC DCM 168 -14.393587 1.527929 Control DIC Chloroform 24 -7.11123 0.978262 Control DIC Chloroform 168 -9.859062 0.356352 Control Rotifer DCM 24 -16.193784 1.216536 Control Rotifer DCM 168 -18.759459 1.245747 Control Rotifer Chloroform 24 -12.341167 0.799041 Control Rotifer Chloroform 168 -14.655575 0.781083 Porites compressa Bleached DIC DCM 24 -14.850367 2.207032 Bleached DIC DCM 168 -20.166625 1.972937 Bleached DIC Chloroform 24 -10.321931 2.119268 Bleached DIC Chloroform 168 -15.853104 0.521071 Bleached Rotifer DCM 24 2.644451 7.387039 Bleached Rotifer DCM 168 -18.419896 1.654778 Bleached Rotifer Chloroform 24 33.835683 18.407508 Bleached Rotifer Chloroform 168 34.419684 13.763786 Control DIC DCM 24 8.436258 4.712213 Control DIC DCM 168 -14.710387 2.440618 Control DIC Chloroform 24 14.374559 2.075575 Control DIC Chloroform 168 -7.973384 3.838897 Control Rotifer DCM 24 -2.036469 4.132022 Control Rotifer DCM 168 -18.965349 2.396501 Control Rotifer Chloroform 24 47.678734 6.382386 Control Rotifer Chloroform 168 64.870721 11.326209

Table 5: Porites compressa method comparison. Analysis of variance (ANOVA) of the difference in lipid δ13C enrichment values for corals extracted with DCM: Methanol and

Chloroform: Methanol.

Source DF Sum of Squares F p Model 5: 211 30751.3708 7.19 <0.0001 Method 1 16511.64104 19.31 <0.0001 Geno 4 12999.22371 3.80 0.053

Table 6: Montipora capitata method comparison. Analysis of variance (ANOVA) of the difference in lipid δ13C enrichment values for corals extracted with DCM: Methanol and

Chloroform: Methanol.

Source DF Sum of Squares F p Model 5: 18 14.19381763 0.77 0.5854 Method 1 0.05320724 0.01 0.9060 Geno 4 14.18956082 0.97 0.4581