Marine Ecology Progress Series 570:87

MARINE ECOLOGY PROGRESS SERIES

Mar Ecol Prog Ser

Vol. 570: 87–99, 2017

https://doi.org/10.3354/meps12114

Published April 27

INTRODUCTION

One key determinant of coral resilience is the abil-

ity to heal from wounds. In nature, corals sustain

lesions regularly, and from multiple causes, includ-

ing corallivory (Kaufman 1981, Jompa & McCook

2003, Jayewardene & Birkeland 2006, Rotjan and

Lewis 2008, Cole et al. 2008), algal abrasion (Coyer

et al. 1993, Grace 2004), sedimentation (Nugues &

Roberts 2003a,b), and hurricane activity (Bythell et

al. 1993a). Tissue regeneration (by which tissue has

re-grown and/ or polyp body plan is reimposed) and

full wound recovery (by which colony integrity is

restored, including calcification) are energetically

demanding (Oren et al. 2001, Henry & Hart 2005,

Jayewardene et al. 2009, Lenihan & Edmunds 2010).

As a result, wounding can lead to reductions in

colony fecundity (Oren et al. 2001, Rotjan & Dimond

2010) and growth (Meesters et al. 1994, Jayewardene

2010, Lenihan & Edmunds 2010). Furthermore, until

epithelial integrity is reestablished, wounded colo -

nies are left with patches of bare calcium carbonate

that are susceptible to overgrowth by benthic com-

petitors such as algae and sponges (Meesters et al.

1996, 1997, Diaz-Pulido & McCook 2002, Jompa &

McCook 2003, Rotjan & Lewis 2005). Such over-

growth may inhibit the recovery of coral tissue and

overall colony growth (River & Edmunds 2001).

Temperature and symbiosis affect lesion recovery

in experimentally wounded, facultatively symbiotic

temperate corals

E. M. Burmester

1,2,

, J. R. Finnerty

, L. Kaufman

, R. D. Rotjan

1,2,

Boston University Department of Biology, 5 Cummington Mall, Boston, MA 02215, USA

John H Prescott Marine Laboratory, Anderson-Cabot Center for Ocean Life, New England Aquarium, 1 Central Wharf,

Boston, MA 02110, USA

ABSTRACT: The health of most reef-building corals depends upon an intracellular symbiosis with

photosynthetic dinoflagellates of the genus Symbiodinium that is acutely sensitive to increasing

ocean temperatures. However, distinguishing the individual effects of both temperature and symbi-

otic state on coral health is difficult to investigate experimentally in most tropical corals because the

symbiosis is obligate. Here, we varied temperature (9, 18, 24°C) and symbiotic state (symbiotic,

aposymbiotic) in the facultatively symbiotic, temperate scleractinian coral Astrangia poculata to ex-

plore the individual impact of temperature and symbiosis on wound healing, an important compo-

nent of coral resilience, by determining wound size using calibrated photographs and

characterizing developmental stage through the healing process over time. Symbiotic corals

demonstrated a significant healing advantage over corals with lower densities of S. psygmophilum

(aposymbiotic state), regardless of temperature. In addition, overall recovery success of both symbi-

otic states increased with temperature. These data suggest that a functional symbiotic relationship

with S. psygmophilum promotes lesion recovery despite heterotrophic energy sources. Reductions

in healing rate and tissue cover near the wound site under cold temperatures suggest that wound

healing is compromised during the winter in these temperate corals. This study demonstrates that

supplemental energy sources from symbiosis, coupled with optimal growth conditions, promote

wound healing and may offer insight into factors enhancing wound recovery in tropical corals.

KEY WORDS: Coral · Recovery · Symbiosis · Temperature · Lesions

Resale or republication not permitted without written consent of the publisher

Mar Ecol Prog Ser 570: 87–99, 2017

Therefore, the pace of healing and regeneration is

critical to coral survival.

Healing rate can be influenced by a number of fac-

tors, including coral species (Bak & Steward-Van Es

1980), the size of the lesion (Meesters et al. 1996), and

the ratio of wound perimeter to surface area (Meesters

et al. 1997). Infrequent high-energy traumas tend to

cause larger injuries, whereas due to their greater fre-

quency, the fine-scale wounds that result from smaller

disturbances (such as fish bites) can account for a

greater loss of live tissue than large-scale wounds

(Hughes & Jackson 1985, Bythell et al. 1993b, Rotjan

& Lewis 2006). Therefore, it is particularly important

to understand the factors that impact how rapidly

corals heal from these small lesions.

Environmental stress also influences the rate of

healing and regeneration (Fisher et al. 2007). For

example, temperature-induced bleaching (Meesters

& Bak 1993, Rotjan et al. 2006) and high sedimenta-

tion (Meesters et al. 1994, Cróquer et al. 2002) impair

healing capability in corals. Because of their impor-

tance to coral resilience and sensitivity to environ-

mental conditions, wound recovery and regeneration

have been used as metrics to assess colony health in

the field (Meesters et al. 1994) and in the laboratory

(Work & Aeby 2010), and to forecast the ability of a

coral to survive prevailing environmental conditions

(Downs et al. 2005).

While lesion healing has been studied extensively in

the field in a variety of species (reviewed by Henry &

Hart 2005, and for example: Nagelkerken & Bak 1998,

Denis et al. 2011, Cameron & Edmunds 2014) as well

as across a range of intrinsic and extrinsic factors (e.g.

Van Veghel & Bak 1994, Nagelkerken et al. 1999,

Kramarsky-Winter & Loya 2000, Edmunds 2009), few

studies have directly studied the impact of symbiosis

on the healing process. In tropical corals, photosyn-

thetic endosymbiotic alga of the genus Symbiodinium

can provide up to 95% of the coral host’s energy (Mus-

catine 1990); however, environmental stress (such as

rising temperatures or changes in salinity) can decou-

ple this obligate symbiosis, resulting in ‘bleaching’

(Hoegh-Guldberg 1999, Rowan 2004). This bleached

state is not stable because of the resulting nutritional

deficiency, and those colonies that are unable to shift

or regain symbiotic partners will die (Grottoli et al.

2006, Baker et al. 2008). Because it is difficult to main-

tain tropical corals in a bleached state, and because

high temperatures result in bleaching, it is difficult to

experimentally distinguish the impacts of bleaching

and temperature on wound healing (Henry & Hart

2005), although this is of particular interest to coral

conservation (Edmunds & Lenihan 2010).

The northern star coral Astrangia poculata (= A.

danae; Peters et al. 1988) is an ideal organism to decou-

ple the effects of temperature and symbiosis be cause

its distribution spans a wide range of temperatures and

it exhibits a facultative symbiosis with Sym bio dinium.

A. poculata is a temperate species whose native range

extends along the US east coast, from Florida and the

Gulf of Mexico to Rhode Island (RI) (Dimond & Car-

rington 2007, Thornhill et al. 2008). In nature, A. pocu-

lata may be found in sym bio sis with S. psygmophilum

(Lajeunesse et al. 2012), but coloniesmayalsobe found

in an aposymbiotic (relatively low density of S. psyg-

mophilum) state in the same habitat with symbiotic

colonies (Dimond et al. 2013). Even within a colony,

symbiont densities can vary markedly among polyps,

resulting in mixed or mottled phenotypes (Fig. 1). Re-

gardless of symbiont state, all colonies of A. poculata

rely heavily on heterotrophy as a source of energy

(Szmant-Froelich & Pilson 1980). The symbiotic state of

the colony is a function of zooxanthellae expulsion

rates, i.e. apo symbiotic colonies actively maintain a

very low density of S. psygmophilum through high ex-

pulsion rates (Dimond & Carrington 2007). Polyps of A.

poculata remain functionally aposymbiotic until S.

psygmo philum density reaches or exceeds 10

cells

−2

, after which polyps appear consistently brown

through out the body column (Dimond & Carrington

2007). As a result, these functionally aposymbiotic

colo nies differ significantly from bleached corals, as

the low density of Symbiodinium sp. within host cells is

(1) not indicative of stress and (2) not the result of a

breakdown in symbiotic pathways between alga and

host. Thus, unlike corals in obligate symbioses, quali-

tative differences in symbiont density exist as stable

states in nature. Additionally, the sympatric overlap of

different symbiont states provides a natural experiment

for exploring the relative roles of symbiont and host on

different biological functions. Previous studies have

exploited the facultative symbiosis of A. pocu lata to

investigate the effects of the coral−algal symbiosis on

coral health, including the effect of symbiont density

on coral nutrition (Szmant-Froelich & Pilson 1980,

1984), resistance to ocean acidification (Holcomb et

al. 2010, 2012), post-sedimentation re covery (Cohen et

al. 2002), calcification and meta bolism (Jacques &

Pilson 1980, Cummings 1983, Jacques et al. 1983), and

physical parameters of wound recovery (DeFilippo

et al. 2016).

The objective of this study was to determine the

effects of temperature and symbiont state on wound

recovery in coral colonies. We conducted controlled

laboratory experiments to investigate tissue recovery

in naturally occurring symbiotic (high density of S.

Burmester et al.: Temperature, symbiosis, and coral recovery

psygmo philum) and aposymbiotic (low density or

absence of S. psygmophilum) colonies of A. poculata

at 3 environmentally relevant temperatures charac-

teristic of winter (9°C), summer (18°C), and a temper-

ature above this coral’s natural range (24°C). Several

healing metrics indicate that symbiotic corals exhibit

significantly greater healing ability at 18 and 24°C.

MATERIALS AND METHODS

Collection and husbandry

Colonies exhibiting a range of symbiont densities

were collected from depths of 6 to 10 m at Fort

Wether ill State Park in Jamestown, RI (41° 28’ 40” N,

71° 21’ 34” W) from late spring through early fall

2014. As de scribed by DeFilippo et al.

(2016), speci mens were housed in a

flow-through aquarium system at the

New England Aquarium. Seawater

was filtered using a protein skimmer

and UV treatment, and water quality

was measured weekly. Tanks were

illuminated for 10 h d

−1

using T5 HO

fluorescent lighting fixtures (Hamilton

Techno logy, Aruba Sun T5-V Series).

Photosynthetically active radiation

(PAR) was kept constant at an average

of 37.5 ± 10.1 µmol m

−2

−1

. Corals

were fed daily with frozen copepods

(JEHM) directed to all polyps within

each colony using a turkey baster. All

colonies were acclimated to 18°C for

at least 2 wk prior to commencing

healing trials. Throughout the dura-

tion of the initial acclimation and the

experimental trials, all colonies were

cleaned weekly using a soft nylon

brush to remove algae and forceps to

remove epibionts (e.g. polychaetes).

Only bare skeletal regions (i.e. no live

tissue) were brushed in this manner to

avoid inflicting additional wounds or

tissue damage. Colonies were as -

signed symbiont state as described by

Dimond & Carrington (2007) and de -

monstrated by Dimond et al. (2013)

and DeFilippo et al. (2016), where

symbiont state was determined visu-

ally using color as a proxy for chloro-

phyll density. Additionally, to ensure

this method accurately represented

our system, a random subset of photos of 20 symbiotic

and 20 aposymbiotic colonies were analyzed for

approximated chlorophyll density using the protocol

of Dimond & Carrington (2007). On average, the ap -

proximated chlorophyll densities for symbiotic colo -

nies (0.45 ± 0.04 µg cm

−2

, mean ± SE) were signifi-

cantly higher than those of aposymbiotic colonies (0.13

± 0.02 µg cm

−2

; ANOVA, F

1,38

= 53.4058, p < 0.0001).

Experimental setup and temperature manipulation

Symbiotic and aposymbiotic colonies were sorted

into pairs for treatment. Colony pairs were randomly

placed into treatment groups, controlling for size

between treatments. Sizing was based on colony

mass, a measurement of whole colony mass after

Fig. 1. (A) Stages of wound healing as demonstrated in Astrangia poculata. a:

Experimentally-induced bare skeleton wound (w) from initial polyps (ip); b: no

demonstration of recovery with wound site covered in algae (a); c: formation of

undifferentiated tissue (ut) from wound edges (as shown) or from within cal-

ice; d: formation of tentacle nubs (tn); e: re-establishment of full polyp (fp) with

fully functional tentacles. (B) Symbiont states in A. poculata. f: Aposymbiotic

colonies appear white; g: mixed symbiont colonies have a range of symbiotic

(s) and aposymbiotic (ap) polyps; h: fully symbiotic colonies appear brown

Mar Ecol Prog Ser 570: 87–99, 2017

the removal of excess water from bare skeletal

regions. This measurement exhibited a high degree

of precision; 5 separate mass measurements were

performed on each of 37 colonies, and the mean

variation in measurements performed on the same

individual was 0.69%. Overall, average colony mass

was 6.32 ± 0.27 g (SE). Paired colonies were placed

adjacent to one another on a submerged platform in

a randomized fashion in order to control for tank

microclimate (lighting, flow, etc.). However, each

pair of colonies was kept at a distance of 6−10 cm to

avoid direct inter action. Three temperature treat-

ments were tested: 9°C (within the natural range for

winter), 18°C (within the natural range for summer),

and 24°C (outside the typical temperature range for

this site). Experimental temperatures were chosen

based on environmental data for the area (NOAA

Tides & Currents, Newport, RI: site 8452660), where

sea surface temperatures ranged from 0.5 to 23°C

during the 2014 calendar year. To account for sea-

sonality and to accommodate the limitations of our

system (i.e. tank space, ability to manipulate multi-

ple temperatures simultaneously), 9°C treatments

were performed in the winter months and 24°C

treatments were performed in the summer and early

fall. For continuity between seasonal experiments

and to control for captivity duration, 2 trials of the

18°C group were carried out: 1 in winter and 1 in

summer. Winter and summer 18°C trials were then

compared to determine any potential changes due

to season, captivity duration, or husbandry (there

were none). Therefore, at 18°C, 40 colonies of each

symbiont state were sampled; meanwhile, 20

colonies of each symbiont type were sampled at 9

and 24°C. At 18°C, 1 wounded aposymbiotic colony

was lost during routine tank cleaning (non-mortal-

ity), reducing the sample size of that group to 39

colonies (and a total of 159 colonies overall). Corals

in the 9 and 24°C tanks were subjected to a temper-

ature increase or decrease from 18°C at a rate of

1°C every 12 h. Colonies in these tanks were accli-

mated at their final experimental temperatures for

1 wk prior to wounding.

Wounding

All experimental corals were wounded in a consis-

tent fashion: a single polyp, centrally located in the

colony, was removed using a scalpel, and the calyx

was cleared of tissue using a Waterpik (Fig. 1) before

the immediate surrounding skeleton was filed to a

uniform basal skeletal height using a diamond-

coated file, followed by a final cleaning with a Water-

pik. These lesions were designed to mimic destruc-

tive forces that remove both tissue and skeleton (i.e.

predation, mechanical injury, storm damage, etc.),

but were inflicted in such a way to create as uniform a

wound as possible between replicates. Colony mass

for each colony was measured before and after

wounding to determine wound size (via change in

mass).

Lesion recovery

Lesions were photographed using a Leica

M165FC stereomicroscope immediately after

wounding (Day 0) and at 10 time points post-

wounding (5, 10, 15, 20, 25, 30, 40, 50, 60, and 75 d).

Photographs from Day 0 were used to calibrate

post-wound photos to ensure consistent magnifica-

tion as well as colony angle and position. Photo-

graphs were scored for whether or not colonies

exhibited signs of healing (i.e. new tissue within the

wound site) as well as the developmental state of

new tissue (undifferentiated tissue, tentacle nubs, or

full polyp; Fig. 1A [panels c−e, respectively]). These

3 stages were chosen because they can be identified

unambiguously, and they represent important

developmental landmarks on the way towards full

healing, which begins with the formation of undif-

ferentiated tissue, followed by initiation of tentacle

formation, and concludes with the formation of a

full polyp that has the ability to feed (as evidenced

by tentacular contraction; Fig. 1A). Achievement of

this final developmental landmark was recognized

as the indicator of healing success. Wound surface

area was determined for Days 0 and 75 using hand-

drawn area tools that are part of the Leica M165FC

software. Three measurements were taken for each

photo, and the wound surface area was estimated as

their average. The percent change in wound surface

area was calculated as final wound area minus the

initial wound area divided by the initial wound

area.

Colony mass

Bare skeletal regions of each colony were tho -

rough ly cleaned, avoiding live tissue, with a soft

nylon brush and forceps prior to each measurement

of mass to prevent confounding colony growth with

algal growth. The difference between the initial post-

wound mass (Day 0) and final mass (Day 75) was

Burmester et al.: Temperature, symbiosis, and coral recovery

used to calculate a healing mass differential, i.e. the

gain or loss of mass over the healing process. We also

determined the mass of the wound itself by subtract-

ing the colony’s post-wound mass from its pre-wound

mass. Two specimens were removed from the analy-

sis (1 each from the 9°C symbiotic and aposymbiotic

groups) because their colony mass at the time of

wounding had not been accurately recorded.

Photosynthetic efficiency

In order to test for symbiont performance, photo-

synthetic efficiency (maximum quantum yield: F

)

was measured for each colony at 3 time points over

the course of the experiment (0, 30, and 60 d) using a

Walz JUNIOR-PAM pulse-amplitude modulated flu-

orescence meter. All readings were taken between

11:00 and 13:00 h on each day to reduce diel fluctua-

tions between readings across time points. As

described by DeFilippo et al. (2016), colonies were

acclimated to darkness for 30 min in a closed dark

container and then transferred individually to a glass

beaker with 5−7 cm of seawater stored within a dark

box to reduce light exposure during measurement.

For each measurement, polyps were first exposed to

6 s of far-red illumination to determine minimal fluo-

rescence while dark-adapted (F

). They were then

ex posed to 0.6 s of a saturating pulse (10 000 µmol

−2

−1

) to determine maximal fluorescence (F

). The

change in fluorescence between F

and F

(ΔF) was

divided by maximal fluorescence (F

) to calculate

maximum quantum yield (F

; Suggett et al. 2010).

Readings were taken with a light fiber held approxi-

mately 1 mm from the oral groove of 3 haphazardly

selected polyps per colony, and the resulting average

maximum quantum was used to represent the colony.

Documentation of altered symbiotic state

We visually inspected microscope photos from

Day 75 to qualitatively assess whether any of the

polyps within a 2-polyp radius of the wound had

undergone a change in symbiont density as evi-

denced by a shift in color. Shifts in symbiont density

were quantified as the percentage of polyps chang-

ing symbiont state between Day 0 and Day 75; only

polyp-level shifts from aposymbiotic to symbiotic

were observed. Such shifts became apparent when

individual polyps in a colony would develop clusters

of brown spots (i.e. symbiosomes) detectable under

the microscope.

Statistics

Healing initiation (proportion of colonies having

developed any of the 3 landmark stages of healing)

and healing success (proportion of colonies having

regenerated new full polyps) were analyzed using

generalized logistic mixed models (GLMMs) in the

lme4 package in R. Logistic models were compared

using Akaike’s information criterion (AIC) scores,

and a reduction in AIC of at least 2 was required to

accept a given model over another or to validate the

inclusion of a new variable in the model (Burnham &

Anderson 2002). In the case of 2 models whose AIC

scores differed by less than 2, the simpler model was

chosen (Burnham & Anderson 2002). Odds ratios

were generated using exponentiated estimates. The

effects of symbiont state and temperature on colony

mass and surface area were compared using Laplace-

approximated generalized linear mixed models on

the nlme package in R (R Core Team 2013, Pinheiro

et al. 2017). In order to test for the potential effect of

colony size and wound size on recovery and colony

mass, we included wound mass, initial mass, and

initial wound surface area as fixed effects in addition

to symbiont state and temperature in the analyses of

healing initiation, healing success, wound closure,

and colony mass. The impacts of added fixed, inter -

action, and random effects were determined using a

forward and reverse stepwise approach (using AIC

for logistic models and likelihood ratio tests for linear

models). Additionally, tank assignment was desig-

nated as a random effect in all generalized logistic

and linear models to control for potential conse-

quences of multiple colony pairs in the same tank.

In order to test for changes in photosynthetic effi-

ciency over time, maximum quantum yield was ana-

lyzed between symbiont states and temperatures

over time using a restricted maximum liklihood

(REML)-fitted GLMM in the lme4 package in R (R

Core Team 2013, Bates et al. 2015). Individual coral

identity was nested within tank in order to control for

changes in an individual across time points and to

account for potential confounds of tank housing.

Model selection criteria were based on reductions in

AIC in the same manner as logistic model selections.

Fisher’s exact tests and ANOVAs were used to

determine the effect of shifts in symbiont density on

the relative ratios of healing ability and stage, and

relative means of colony mass and surface area

recovery, respectively, for aposymbiotic corals at 18

and 24°C. No colonies at 9°C showed signs of shifts in

symbiont density, and this group was therefore ex -

cluded from this analysis.

Mar Ecol Prog Ser 570: 87–99, 2017

RESULTS

Healing initiation and success

Overall, there was a significant impact of symbiosis

on the healing process (Fig. 2). Adjusted for tank

grouping, symbiotic corals (40 out of 80 colonies)

were 2.4 times as likely as aposymbiotic (25/79)

corals to have initiated healing (developed new tis-

sue of any kind) at the wound site (Table 1), and 5.8

times as likely to have successfully completed heal-

ing (reached the full polyp stage) by the end of the

trial (15/80 symbiotic; 3/79 aposymbiotic; Table 2).

Healing initiation also increased with temperature

(Fig. 2): with random effects, the odds ratio of

exhibiting any healing at 24°C (23/40) was 4.7 times

greater than at 9°C (9/40) and 1.6 times greater than

at 18°C (33/79), while the odds ratio for corals at 18°C

was 2.9 times that of corals kept at 9°C. None of the

9°C colonies developed full polyps by the end of the

trial; however, among those colonies that did suc-

cessfully complete the healing process, the odds ratio

of complete healing at 24°C (10/40) was 3.2 times

greater than at 18°C (8/79). Consistent with these

observations, both symbiont state and temperature

were found to be significant predictors of healing ini-

tiation in a GLMM analysis (Table 3). However, nei-

ther the interaction between these 2 variables nor the

added fixed effects of mass, wound mass, and wound

surface area significantly improved the model (AIC

208.6, Table 3). Likewise, only symbiont state and

temperature were found to be significant predictors

of full polyp formation (i.e. healing success) in corals

at 18 and 24°C (AIC 68.64, Table 4).

Effect Odds ratio (95% CI)

Symbiont state 2.370 (1.205, 4.663)

9°C: 18°C 2.856 (1.033, 7.898)

9°C: 24°C 4.664 (1.459, 14.910)

18°C: 24°C 1.633 (0.6325, 4.215)

Table 1. Odds ratios (with 95% CI) for fixed effect comparisons

of healing ability of Astrangia poculata

Effect Odds ratio (95% CI)

Symbiont state 5.769 (1.346, 24.732)

Temperature 3.253 (0.927, 11.418)

Table 2. Odds ratios (with 95% confidence interval, fixed

effect comparisons of healing success (polyp formation) of

Astrangia poculata

Fig. 2. Proportion in each healing stage of symbiotic and

aposymbiotic colonies of Astrangia poculata maintained at 3

temperatures (9, 18, 24°C) 75 d post-wounding. Numbers in

each column represent total number of colonies at that heal-

ing stage. Cumulative areas in blue delineate colonies that

have demonstrated signs of healing, while bars in dark blue

indicate only those colonies that have achieved successful

healing through the formation of a complete polyp

Effect Estimate SE Z p

Rooted under aposymbiotic and 9°C conditions

Intercept −1.7812 0.4876 −3.653 0.0003**

Symbiont state (S) 0.8627 0.3452 2.499 0.0124*

18°C 1.0496 0.5189 2.023 0.0431*

24°C 1.5399 0.5929 2.597 0.0094*

Rooted under symbiotic and 18°C conditions

Intercept 0.1311 0.3193 0.411 0.6813

Symbiont state (A) −0.8627 0.3452 −2.499 0.0124*

9°C −1.0496 0.5189 −2.023 0.0431*

24°C 0.4903 0.4839 1.013 0.3109

Table 3. Laplace-approximated generalized mixed logistic

regression for healing ability of Astrangia poculata (AIC =

208.6). S: symbiotic; A: aposymbiotic. For fixed effects, sym-

bols indicate significance considered at **p < 0.001, *p < 0.05

Effect Estimate SE Z p

Intercept −2.4522 0.7477 −3.280 0.0010**

Symbiont state (S) 1.7525 0.7427 2.360 0.0183*

Temperature (24°C) 1.1796 0.6406 1.842 0.0655

Table 4. Laplace-approximated generalized mixed logistic

regression for healing success (rooted under aposymbiotic

and 18°C conditions) (AIC = 68.64). S: symbiotic. For fixed

effects, symbols indicate significance considered at **p <

0.001, *p < 0.05,

p < 0.01

Burmester et al.: Temperature, symbiosis, and coral recovery

Wound closure

On average, the inflicted wounds were (mean ± SE)

38.4 ± 1.2 mm

in area and ac counted for a loss in

mass of 0.14 ± 0.01 g. With the exception of symbiotic

corals housed at 24°C, colonies, on average, experi-

enced an in crease in wound surface area, represent-

ing an ex pansion of the wound site (Fig. 3). All (100%)

symbiotic colonies experienced a decrease in wound

size at 24°C, while none (0%) of the colonies (regard-

less of symbiont density) experienced such a decline

at 9°C. More symbiotic colonies (32.5%) saw reduc-

tions in wound size at 18°C than did aposymbiotic

colonies (18%). A similar proportion of aposymbiotic

colo nies were able to restore any lost surface area at

18°C (18%) and 24°C (15%). The proportion of tissue

loss was lower in symbiotic corals (group mean, M =

0.098) than apo symbiotic corals (M = 0.195). Tempera-

ture also had a significant effect on wound surface

area (Table 5), as corals in warmer water recovered a

higher proportion of tissue than those in cooler water

= −0.349, M

= −0.137, M

= 0.046). Again, sym-

biont state and temperature were significant contrib-

utors to the best mixed model (Table 5), although the

effect of temperature was reduced (p = 0.0918).

Colony mass

Except for the 24°C symbiotic group, colonies lost

mass (g) over time (Fig. 4). Symbiotic colonies (M =

−0.004 g) lost less mass than did aposymbiotic colo -

nies (M = −0.409 g). Similarly, corals kept at higher

temperatures experienced less tissue loss than

colonies at lower temperatures, with the colonies at

24°C experiencing a slight gain in mass on average

= −0.345 g, M

= −0.242 g, M

= 0.021 g; Fig. 4).

Both symbitont state and temperature had a signifi-

cant impact on colony mass, but there was no signifi-

cant interaction between these 2 variables (Table 6).

Colony mass, wound mass, and initial wound surface

area had no significant effects.

Photosynthetic efficiency

Regardless of symbiont state, measurable levels of

photosynthesis were observed in all treatment

groups (Fig. 5). Symbiotic colonies had higher values

for maximum quantum yield (F

), with the excep-

tion of symbiotic colo nies kept at 9°C (which demon-

strated photosynthetic efficiency levels similar to

those of aposymbiotic corals). Aposymbiotic colonies,

on the other hand, demonstrated consistently similar

values regardless of temperature. The data

show a slight significant decline in photochemical

efficiency for symbiotic corals over time, but a slight

significant increase for aposymbiotic corals. The pre-

ferred model included only symbiont state and the

interactions between symbiont state and tempera-

ture as well as symbiont state and time (Table 7).

Based on our model selection criteria (requiring a

reduction in AIC of at least 2 to permit the inclusion

of new fixed effects), there was little support for the

inclusion of other fixed effects (temperature and

time) or interactions of fixed effects (temperature/

time).

Fig. 3. Proportional change in wound surface area on As-

trangia poculata after 75 d normalized over the initial wound

surface area. Values were measured from aligned photo-

graphs by dividing the final wound surface area on Day 75

by the initial wound surface area on Day 0. Negative values

indicate an increase in wound size, while positive values

indicate a decline in wound size. Error bars are SE

Effect Estimate SE df t p

Rooted under aposymbiotic and 9°C conditions

Intercept −0.5382 0.1138 136 −4.7289 <0.0001**

Symbiont state (S) 0.3977 0.0670 136 5.9463 <0.0001**

18°C 0.0908 0.1347 19 0.6737 0.5086

24°C 0.2822 0.1588 18 1.7761 0.0918

Rooted under symbiotic and 18°C conditions

Intercept −0.0497 0.0860 136 −0.5774 0.5646

Symbiont state (A) −0.3977 0.0670 136 −5.9463 <0.0001**

9°C −0.0908 0.1347 19 −0.6737 0.5086

24°C 0.1914 0.1404 19 1.3631 0.1888

Table 5. Laplace-approximated generalized linear mixed model

for wound surface area (AIC = 218.8). S: symbiotic; A: aposymbi-

otic. For fixed effects, symbols indicate significance considered at

**p < 0.001, *p < 0.05,

p < 0.01

Mar Ecol Prog Ser 570: 87–99, 2017

Symbiont state switching

At 18 and 24°C, some polyps of aposymbiotic corals

were found to switch symbiont states, as indicated by

pockets of high Symbiodinium psygmophilum den-

sity along the body column or oral groove. This phe-

nomenon was significantly more likely in colonies at

24°C (15/20; 75%) than in colonies kept at 18°C

(18/39; 46.15%; 1-tailed Fisher’s exact test; n = 59,

p = 0.0319). However, among these corals, there was

no significant effect of symbiont switching on healing

initiation (2-tailed Fisher’s exact test; n = 59, p =

0.4232), colony mass (x

switch

= −0.1244 g; x

non-switch

−0.1313 g; n = 59, F

1,57

= 0.0075, p = 0.9310), or

wound closure (x

switch

= −0.441; x

non-switch

= −0.303; n =

59, F

1,57

= 1.1012, p = 0.2984). It should be noted that

no aposymbiotic colonies ever completely switched

states where every polyp in the colony transitioned

from white to brown.

DISCUSSION

Overall, our data suggest that symbiotic colonies

exhibit a clear advantage over aposymbiotic colonies

with respect to single-polyp wound healing. At all

temperatures, a greater proportion of symbiotic co -

rals developed new tissue at the wound site and went

on to develop full polyps. Likewise, regardless of

temperature, symbiotic corals lost less mass and less

tissue area at the lesion site after wounding (to the

point of partially to fully closing lesions at 24°C).

Therefore, symbiotic corals proved more successful

at resisting wound expansion and achieving wound

closure for single polyp lesions. These results are

consistent with previous studies, which suggested that

symbiotic colo nies of Astrangia poculata may be able

to recover a higher proportion of multipolyp wounds

from small pockets of residual tissue (DeFilippo et

al. 2016). Additionally, wounding experiments on

another temperate and facultatively sym bio tic coral,

Oculina patagonica (Fine et al. 2001), demonstrated

a similar advantage to symbiosis, where by the per-

centage recovery of lesions was greater in un -

bleached colonies than partially bleached or fully

bleached (which did not recover at all) colonies (Fine

et al. 2002).

Effect Estimate SE df t p

Rooted under aposymbiotic and 9°C conditions

Intercept −0.2686 0.0569 133 −4.7214 <0.0001**

Symbiont state (S) 0.1073 0.0322 133 3.3317 0.0011*

18°C 0.1700 0.0673 19 2.5258 0.0206*

24°C 0.2364 0.0794 19 2.9784 0.0077*

Rooted under symbiotic and 18°C conditions

Intercept 0.0087 0.0426 133 0.2040 0.8387

Symbiont state (A) −0.1073 0.0322 133 3.3317 0.0011*

9°C −0.1700 0.0673 19 −2.5258 0.0206*

24°C 0.0664 0.0698 19 0.9516 0.3533

Table 6. Laplace approximated generalized linear mixed model for

growth (AIC = −8.535). S: symbiotic; A: aposymbiotic. Symbols

indicate significance considered at **p < 0.001, *p < 0.05

Fig. 4. Mean change in Astrangia poculata colony mass 75 d

after wounding. Values were calculated by subtracting the

final mass of each colony from its post-wound mass. Error

bars are SE

Fig. 5. Mean photosynthetic efficiency (F

) as measured

via pulse-amplitude modulated fluorescence (Walz JUNIOR-

PAM). Error bars are SE. For all 9°C and 24°C groups, N = 20;

for 18°C symbiotic corals, N = 40; and for 18°C aposymbiotic

corals, N = 39

Burmester et al.: Temperature, symbiosis, and coral recovery

The positive effect of symbiosis on healing ob -

served here could be attributed to a number of bene-

fits conferred to the corals by their symbiotic part-

ners, including (1) higher energy reserves and tissue

content at the time of wounding, (2) added energy

availability during healing due to active photosyn-

thesis, or (3) potential direct symbiont contribution to

the healing pathway (Fine et al. 2002). Be cause we

also found that aposymbiotic polyps can switch par-

tially or fully to a symbiotic state — and that these

switches did not significantly impact the colony’s

ability to heal or grow — it is possible that symbiotic

advantage accrues from a long-term association.

Alternatively, a related study on the facultatively

symbiotic O. patagonica (Fine et al. 2002) deter-

mined that carbon can be preferentially translocated

to recovering tissue from a distance of 4−5 cm away;

however, translocation only occurs in fully un -

bleached (and not partially bleached, 30−80%) colo -

nies. This suggests a bleaching threshold (or a mini-

mum density of Symbiodinium spp.) below which

colonial integration and resource translocation is dis-

rupted. There could also be an energetic cost to sym-

biont acquisition that overcomes the potential imme-

diate energy gain from photosynthesis (Hill & Hill

2012). In tropical corals, which rely more heavily on

photosynthesis to supply their nutritional needs, this

disparity in healing ability between the symbiotic

and aposymbiotic states could have significant im -

pacts on coral health, particularly after bleaching.

Even after returning to a full symbiont

load, the diminishment of energetic re -

serves that results from a prolonged period

of bleaching could undermine wound re -

covery for an extended period of time

(Meesters & Bak 1993). This could com-

pound the impact of bleaching and elevate

post-bleaching mortality rates.

While our study does not directly ad dress

the possible advantages of a facul tatively

symbiotic life history, it does raise some im-

portant questions about the costs, benefits,

and dynamics of the coral− algal symbiosis.

While symbiosis clearly en hanced healing

initiation, wound closure, and the develop-

ment of full polyps in the current study,

naturally occurring, apo symbiotic colonies

of A. poculata are abundant in the field

(Dimond et al. 2013). Furthermore, while

symbiont state can be manipulated, switch-

ing be tween symbiont states does not

appear to be common in nature (Dimond &

Carrington 2008). The prevalence of apo -

symbiotic colonies in nature therefore suggests a cost

to maintaining a high density of Symbiodinium that

outweighs the observed advantages of the symbiotic

state on colony health and recovery. Dimond et al.

(2013) suggested that this cost may be most pro-

nounced at cold, winter temperatures when colonies

are dormant. In winter, all colonies experience a net

loss of tissue, but aposymbiotic colonies lose less

tissue than symbiotic colonies (Dimond et al. 2013).

Thus, while symbiotic colonies have enhanced heal-

ing potential in summer, they may also need to com-

pensate for greater tissue loss during the winter.

Environmental temperature had a direct effect on

tissue loss and replacement in this experiment, re -

gardless of symbiotic state. At 9°C, a natural winter

temperature, both aposymbiotic and symbiotic colo -

nies experienced greater tissue loss at the wound site

and lost more total mass over time than at 18°C (a

temperature more typical of summer) or 24°C (a tem-

perature not typically encountered by A. poculata in

Rhode Island). Similarly, the smallest statistical dif-

ference between symbiotic and aposymbiotic colo -

nies occurred at 9°C, as colonies of both symbiont

states had low rates of wound recovery; these cold-

exposed colonies were less likely to develop new tis-

sue at the wound site, and no colonies were able to

successfully complete the regeneration process.

These results are consistent with previous studies

that found temperature to be a major driver of calcifi-

cation and growth in both obligate and facultative

Effect Estimate SE t

Rooted under aposymbiotic and 18°C conditions

Intercept 0.3292 0.0220 14.950*

Symbiont state (S) 0.1569 0.0228 6.877*

Symbiont state (A): Temperature (9°C) 0.0342 0.0370 0.925

Symbiont state (S): Temperature (9°C) −0.0926 0.0254 −3.652*

Symbiont state (A): Temperature (24°C) −0.0048 0.0328 −0.145

Symbiont state (S): Temperature (24°C) −0.0044 0.0260 −0.171

Symbiont state (A): Time 0.0012 0.0003 4.239*

Symbiont state (S): Time −0.0005 0.0003 −1.940*

Rooted under aposymbiotic and 9°C conditions

Intercept 0.3634 0.0318 11.414*

Symbiont state (S) 0.1569 0.0228 6.877*

Symbiont state (A): Temperature (18°C) −0.0342 0.0370 −0.925

Symbiont state (S): Temperature (18°C) 0.0926 0.0254 3.652*

Symbiont state (A): Temperature (24°C) −0.0390 0.0413 −0.942

Symbiont state (S): Temperature (24°C) −0.0881 0.0298 2.958*

Symbiont state (A): Time 0.0012 0.0003 4.239*

Symbiont state (S): Time −0.0005 0.0003 −1.940*

Table 7. REML-fitted generalized linear mixed model for photo -

synthetic efficiency (AIC = −678.4). S: symbiotic; A: aposymbiotic.

Significance (*) assumed from t-values (df = 452) at p < 0.05

Mar Ecol Prog Ser 570: 87–99, 2017

zooxanthellate and azooxanthellate corals (Jacques

et al. 1983, Miller 1995, Marshall & Clode 2004, Ed -

munds 2005, Dimond et al. 2013). However, it should

be noted that growth and regeneration may be unre-

lated or even competing life traits, particularly in

times of stress (Denis et al. 2013). The drop in healing

initiation, healing completion, and colony mass likely

derives from a significant decline in photosynthetic

efficiency with de creasing temperature as well as

a potential de crease in oxidative respiration below

11.5°C (Jacques et al. 1983). Previous studies (Jac -

ques et al. 1983, Dimond et al. 2013) have noted a

state of metabolic dormancy in A. poculata below

10°C, characterized by the retraction of polyps into

their calices and the inability to feed. Interestingly, in

our study, all colonies at 9°C were actively feeding

over the course of the experiment, potentially ex -

plaining the ability of some colonies to exhibit some

healing despite their potentially reduced metabolism

and lack of energetic input from photosynthesis.

Additionally, while 9°C represents a typical winter

temperature for this habitat, it is far from the lowest

temperature encountered by wild colonies of A. poc-

ulata (0.3°C, NOAA Tides & Currents, Newport, RI:

site 8452660). Therefore, as documented by Dimond

et al. (2013), it is possible that quiescence could have

a significant and inverse impact on healing with

regards to symbiont state than observed in this study.

In contrast to 9°C, colonies at 24°C demonstrated

the greatest wound recovery, despite experiencing a

temperature outside the typical range of their natural

habitat (which ranged from 0.3 to 23°C in 2014 [the

year of this study]). Therefore, for this coral, these

data support that survivability and, therefore, habitat

range are more likely being limited by exposure to

cold, winter temperatures than prolonged exposure

to very warm summer temperatures (Dimond et al.

2013).

One hypothesis for the observed increase in heal-

ing and survivability by symbiotic corals is increased

photosynthetic efficiency. Over time, there was an

increase in photosynthetic efficiency in aposymbiotic

colonies. Surprisingly, there was an inverse relation-

ship with temperature (i.e. corals at 9°C experienced

the greatest rise in photosynthetic efficiency). How-

ever, this is very likely due to the fact that, in colder

temperatures, colonies tended to lose more live tissue

and, subsequently, bare skeletal scars were covered

in adventitious algae. Additionally, the limited Gain

settings on the JUNIOR-PAM may obscure fluores-

cence values when chlorophyll densities are low.

Cold-treated symbiotic corals exhibited photosyn-

thetic efficiency levels similar to aposymbiotic corals,

but again, it is difficult to determine how much of this

photochemical activity was generated by S. psygmo -

philum versus other potentially non-symbiotic photo-

synthetic organisms. S. psygmophilum is a cold-toler-

ant species whose photosynthetic efficiency peaks

between 18 and 25°C and reaches a minimum (but

does not cease altogether) at 10°C (Thornhill et al.

2008). Following this pattern, symbiotic corals at 18

and 24°C demonstrated higher levels of photosyn-

thetic activity, although warm-treated corals (24°C)

seemed to experience a reduction in photosynthetic

efficiency over time.

A. poculata is a gonochoristic coral (Peters et al.

1988), and production of gametes occurs at warmer

temperatures. Given the cost of producing gametes,

and the generally greater cost of producing eggs rel-

ative to sperm (Holcomb et al. 2012), it is possible

that corals at warmer temperatures might experience

a trade-off between gamete production and healing,

and this trade-off might differ between males and

females. However, because gametes can only be

identified and quantified using destructive methods,

we were unable to investigate the potential impact of

gamete production on recovery without producing

additional wounds, which would have confounded

our experiments. We hypothesize that (1) wounds

may confer a loss in fecundity, (2) gamete production

may incur a cost to lesion recovery, or (3) reproduc-

tive dynamics promote a synergistic loss for both tis-

sue recovery and fecundity (Rinkevich 1996).

Wound healing is a dynamic process impacted by a

number of different environmental and biological

factors. In addition to symbiotic state and tempera-

ture, lesion healing is affected by the perimeter to

surface area ratio of the wound (Van Woesik 1998) as

well as by the size of the lesion (Rotjan & Lewis 2008)

and the size of the colony (Meesters et al. 1996).

However, the influences of these physical parameters

may vary by species and developmental timing (Bak

1983, Meesters et al. 1992, 1997, Oren et al. 1997,

Van Woesik 1998). One previous study found that

‘Phoenix Effect’ recovery, whereby tissue is regener-

ated from within the calice rather than across wound

edges, is the primary mode of recovery in A. poculata

(DeFilippo et al. 2016). A similar pattern of recovery

was observed in the experimentally wounded colo -

nies in our trials. The wounds inflicted in this experi-

ment most closely resembled those created through

colony breakage or corallivory, where regions of both

skeleton and tissue are removed. In the field, colo -

nies of A. poculata are most often abraded by foliose

algae. Colony morphology is significantly impacted

(flattened) by repeat interactions in regions with high

Burmester et al.: Temperature, symbiosis, and coral recovery
macroalgal density (Grace 2004). While no predators
have yet been characterized in the field for this coral,
its ability to recover from such extremes (deep and
well cleared of residual tissue) demonstrates its util-
ity as a laboratory model for these studies.
Here, we have demonstrated that wound recovery
is affected by symbiont state and temperature in a
facultatively symbiotic coral. Comparable experi-
ments cannot be conducted in tropical corals because
elevated temperatures induce bleaching, and tropi-
cal corals cannot exist in a stable aposymbiotic state.
By allowing us to decouple 2 key parameters that are
conflated in tropical corals, A. poculata can provide
unique insight into how particular biological and
environmental states or stressors may impact coral
health. While the study of temperate corals offers
limited direct comparisons to tropical corals in their
metabolism and physiology, it is likely that core
mechanisms of the molecular stress response
machinery are deeply conserved. Additionally, we
can exploit the wide environmental tolerances of this
temperate coral to investigate how a range of stres-
sors and their intensities can affect coral health under
non-lethal conditions. Indeed, determining the thres -
h old for lethal stress is a critical area of study; using
temperate corals in controlled laboratory studies may
yield valuable insights into the effects of synergistic
stressors, and the relative role of symbiosis.
Acknowledgements. This work was funded by the PADI
foundation, the Boston University Marine Program, and the
New England Aquarium. We thank Kiki Ballotti, Georgie
Burruss, Nicholas Lawrence, Samantha Pelletier, Aaron Pil-
nick, and Ryan Schosberg for assistance with photography
and data collection. We are also grateful to Tasia Blough,
Julio Camperio-Ciani, Gregory Coote, Lukas DeFilippo,
Corbin Kuntz, Georgia Luddecke, Katrina Malakhoff, Jessie
Matthews, and Matthew Tohl for their roles in the hus-
bandry and maintenance of coral specimens. We appreciate
the input of the anonymous reviewers that contributed to the
improvement of this manuscript.
LITERATURE CITED
Bak RPM (1983) Neoplasia, regeneration and growth in the
reef-building coral Acropora palmata. Mar Biol 77: 
221−227
Bak RPM, Steward-Van Es Y (1980) Regeneration of super-
ficial damage in the scleractinian corals Agaricia agari -
cites f. purpurea and Porites astreoides. Bull Mar Sci 30: 
883−887
Baker A, Glynn PW, Riegl B (2008) Climate change and coral
reef bleaching:  an ecological assessment of long-term
impacts, recovery trends and future outlook. Estuar
Coast Shelf Sci 80: 435−471
Bates D, Maechler M, Bolker B, Walker S (2015) Fitting
l-inear mixed-effects models using lme4. J Stat Softw 67:
1–48
Burnham KP, Anderson DR (2002) Model selection and mul-
timodel inference:  a practical information-theoretic ap -
proach, 2nd edn. Springer, New York, NY
Bythell JC, Bythell M, Gladfelter EH (1993a) Initial results of
a long-term coral reef monitoring program: impact of
Hurricane Hugo at Buck Island Reef National Monu-
ment, St. Croix US Virgin Islands. J Exp Mar Biol Ecol
172:171–183
Bythell JC, Gladfelter EH, Bythell M (1993b) Chronic and
catastrophic natural mortality of three common Carib-
bean reef corals. Coral Reefs 12: 143−152
Cameron CM, Edmunds PJ (2014) Effects of simulated fish
predation on small colonies of massive Porites spp. and
Pocillopora meandrina. Mar Ecol Prog Ser 508: 139−148
Cohen AL, Owens KE, Lane GD, Shimizu N (2002) The
effect of algal symbionts on the accuracy of Sr/Ca pale-
otemperatures from coral. Science 296:331–333
Cole AJ, Pratchett MS, Jones GP (2008) Diversity and func-
tional importance of coral-feeding fishes on tropical coral
reefs. Fish Fish 9:286–307
Coyer JA, Ambrose RF, Engle JM, Carroll JC (1993) Interac-
tions between corals and algae on a temperate zone
rocky reef:  mediation by sea urchins. J Exp Mar Biol Ecol
167: 21−37
Cróquer A, Villamizar E, Noriega N (2002) Environmental
factors affecting tissue regeneration of the reef-building
coral  Montastraea annularis (Faviidae) at Los Roques
National Park, Venezuela. Rev Biol Trop 50: 1055−1065
Cummings C (1983) The biology of Astrangia danae. PhD
dissertation, University of Rhode Island, Kingston, RI
DeFilippo L, Burmester EM, Kaufman L, Rotjan RD (2016)
Patterns of surface lesion recovery in the northern star
coral Astrangia poculata. J Exp Mar Biol Ecol 481: 15−24
Denis V, Debreuil J, De Palmas S, Richard J, Guillaume
MMM, Bruggemann JH (2011) Lesion regeneration
capacities in populations of the massive coral Porites
lutea at Réunion Island:  environmental correlates. Mar
Ecol Prog Ser 428: 105−117
Denis V, Guillaume MMM, Goutx M, de Palmas S and oth-
ers (2013) Fast growth may impair regeneration capacity
in the branching coral Acropora muricata. PLOS ONE 8: 
e72618
Diaz-Pulido G, McCook LJ (2002) The fate of bleached
corals:  patterns and dynamics of algal recruitment. Mar
Ecol Prog Ser 232: 115−128
Dimond J, Carrington E (2007) Temporal variation in the
symbiosis and growth of the temperate scleractinian
coral Astrangia poculata. Mar Ecol Prog Ser 348: 161−172
Dimond J, Carrington E (2008) Symbiosis regulation in a fac-
ultatively symbiotic temperate coral:  zooxanthellae divi-
sion and expulsion. Coral Reefs 27: 601−604
Dimond J, Kerwin AH, Rotjan R, Sharp K, Stewart FJ,
Thornhill DJ (2013) A simple temperature-based model
predicts the upper latitudinal limit of the temperate coral
Astrangia poculata. Coral Reefs 32: 401−409
Downs CA, Woodley CM, Richmond RH, Lanning LL, Owen
R (2005) Shifting the paradigm of coral-reef ‘health’
assessment. Mar Pollut Bull 51: 486−494
Edmunds PJ (2005) The effect of sub-lethal increases in tem-
perature on the growth and population trajectories of
three scleractinian corals on the southern Barrier Reef.
Oecologia 146: 350−364
Edmunds PJ (2009) Effect of acclimatization to low tempera-
97

Mar Ecol Prog Ser 570: 87–99, 2017
ture and reduced light on the response of reef corals to
elevated temperature. Mar Biol 156: 1797−1808
Edmunds PJ, Lenihan HS (2010) Effect of sub-lethal damage
to juvenile colonies of massive Porites spp. under con-
trasting regimes of temperature and water flow. Mar Biol
157: 887−897
Fine M, Zibrowius H, Loya Y (2001) Oculina patagonica:  a
non-lessepsian scleractinian coral invading the Mediter-
ranean Sea. Mar Biol 138: 1195−1203
Fine M, Oren U, Loya Y (2002) Bleaching effect on regener-
ation and resource translocation in the coral Oculina
patagonica. Mar Ecol Prog Ser 234: 119−125
Fisher EM, Fauth JE, Hallock P, Woodley CM (2007) Lesion
regeneration rates in reef-building corals Montastraea
spp. as indicators of colony condition. Mar Ecol Prog Ser
339: 61−71
Grace S (2004) Ecomorphology of the temperate scleractin-
ian Astrangia poculata:  coral-macroalgal interactions in
Narragansett Bay. PhD dissertation, University of Rhode
Island, Kingston, RI
Grottoli AG, Rodrigues LJ, Palardy JE (2006) Heterotrophic
plasticity and resilience in bleached corals. Nature 440: 
1186−1189
Henry L, Hart M (2005) Regeneration from injury and
resource allocation in sponges and corals — a review. Int
Rev Hydrobiol 90: 125−158
Hill M, Hill A (2012) The magnesium inhibition and arrested
phagosome hypotheses:  new perspectives on the evolu-
tion and ecology of Symbiodinium symbioses. Biol Rev
Camb Philos Soc 87: 804−821
Hoegh-Guldberg O (1999) Climate change, coral bleaching
and the future of the world’s coral reefs. Mar Freshw Res
50: 839−866
Holcomb M, McCorkle D, Cohen AL (2010) Long-term
effects of nutrient and CO
2
enrichment on the temperate
coral Astrangia poculata. J Exp Mar Biol Ecol 386: 27−33
Holcomb M, Cohen A, McCorkle D (2012) An investigation of
the calcification response of the scleractinian coral Astran-
gia poculata to elevated pCO
2
and the effects of nutrients,
zooxanthellae and gender. Biogeosciences 9: 29−39
Hughes TP, Jackson JBC (1985) Population dynamics and life
histories of foliaceous corals. Ecol Monogr 55: 141−156
Jacques TG, Pilson MEQ (1980) Experimental ecology of the
temperate scleractinian coral Astrangia danae I. Partition
of respiration, photosynthesis and calcification between
host and symbionts. Mar Biol 60:167–178
Jacques TG, Marshall N, Pilson MEQ (1983) Experimental
ecology of the temperate scleractinian Astrangia danae: 
effect of temperature, light intensity and symbiosis with
zooxanthellae on metabolic rate and calcification. Mar
Biol 76: 135−148
Jayewardene D (2010) Experimental determination of the
cost of lesion healing on Porites compressa growth. Coral
Reefs 29: 131−135
Jayewardene D, Birkeland C (2006) Fish predation on
Hawaiian corals. Coral Reefs 25: 328
Jayewardene D, Donahue MJ, Birkeland C (2009) Effects of
frequent fish predation on corals in Hawaii. Coral Reefs
28: 499−506
Jompa J, McCook LJ (2003) Coral-algal competition:  macro-
algae with different properties have different effects on
corals. Mar Ecol Prog Ser 258: 87−95
Kaufman L (1981) There was biological disturbance on
Pleistocene coral reefs. Paleobiology 7: 527−532
Kramarsky-Winter E, Loya Y (2000) Tissue regeneration in
the coral Fungia granulosa:  the effect of extrinsic and
intrinsic factors. Mar Biol 137: 867−873
Lajeunesse TC, Parkinson JE, Reimer JD (2012) A genetics-
based description of Symbiodinium minutum sp. nov. and
S. psygmophilum sp. nov. (Dinophyceae), two dinoflagel-
lates symbiotic with cnidaria. J Phycol 48: 1380−1391
Lenihan HS, Edmunds PJ (2010) Response of Pocillopora
verrucosa to corallivory varies with environmental condi-
tions. Mar Ecol Prog Ser 409: 51−63
Marshall AT, Clode P (2004) Calcification rate and the effect
of temperature in a zooxanthellate and an azooxanthel-
late scleractinian reef coral. Coral Reefs 23: 218−224
Meesters EH, Bak RPM (1993) Effects of coral bleaching on
tissue regeneration potential and colony survival. Mar
Ecol Prog Ser 96:189–198
Meesters EH, Bos A, Gast GJ (1992) Effects of sedimentation
and lesion position on coral tissue regeneration. Proc 7th
Int Coral Reef Symp 2: 681−688
Meesters EH, Noordeloos M, Bak RPM (1994) Damage and
regeneration:  links to growth in the reef-building coral
Montastrea annularis. Mar Ecol Prog Ser 112: 119−128
Meesters EH, Wesseling I, Bak RPM (1996) Partial mortality
in three species of reef-building corals (Scleractinia) and
the relation with colony morphology. Bull Mar Sci 58: 
838−852
Meesters EH, Pauchli W, Bak RPM (1997) Predicting regen-
eration of physical damage on a reef-building coral by
regeneration capacity and lesion shape. Mar Ecol Prog
Ser 146: 91−99
Miller MW (1995) Growth of a temperate coral:  effects of
temperature, light, depth, and heterotrophy. Mar Ecol
Prog Ser 122: 217−225
Muscatine L (1990) The role of symbiotic algae in carbon
and energy flux in reef corals. In:  Dubinsky Z (ed) Coral
reefs:  ecosystems of the world, Vol 25. Elsevier, New
York, NY, p 75−87
Nagelkerken I, Bak RPM (1998) Differential regeneration of
artificial lesions among sympatric morphs of the Carib-
bean corals Porites astreoides and  Stephanocoenia
michelinii. Mar Ecol Prog Ser 163: 279−283
Nagelkerken I, Meesters EH, Bak RPM (1999) Depth-related
variation in regeneration of artificial lesions in the Carib-
bean corals Porites astreoides and  Stephanocoenia
michelinii. J Exp Mar Biol Ecol 234: 29−39
Nugues MM, Roberts CM (2003a) Partial mortality in mas-
sive reef corals as an indicator of sediment stress on coral
reefs. Mar Pollut Bull 46: 314−323
Nugues MM, Roberts CM (2003b) Coral mortality and inter-
action with algae in relation to sedimentation. Coral
Reefs 22: 507−516
Oren U, Benayahu Y, Loya Y (1997) Effect of lesion size and
shape on regeneration of the Red Sea coral Favia favus.
Mar Ecol Prog Ser 146: 101−107
Oren U, Benayahu Y, Lubinevsky H, Loya Y (2001) Colony
integration during regeneration in the stony coral Favia
favus. Ecology 82: 802−813
Peters EC, Cairns SD, Pilson MEQ, Wells JW and others
(1988) Nomenclature and biology of Astrangia poculata
(= A. danae, = A. astreiformis) (Cnidaria:  Anthozoa). Proc
Biol Soc Wash 101: 234−250
Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team
(2017) nlme:  linear and nonlinear mixed effects models.
R package version 3.1-130. https: //CRAN.R-project. org/
package=nlme
R Core Team (2013) R:  a language and environment for sta-
98

Burmester et al.: Temperature, symbiosis, and coral recovery
tistical computing. R Foundation for Statistical Comput-
ing, Vienna
Rinkevich B (1996) Do reproduction and regeneration in
damaged corals compete for energy allocation? Mar Ecol
Prog Ser 143: 297−302
River GF, Edmunds PJ (2001) Mechanisms of interaction
between macroalgae and scleractinians on a coral reef in
Jamaica. J Exp Mar Biol Ecol 261: 159−172
Rotjan RD, Dimond JL (2010) Discriminating causes from
consequences of persistent parrotfish corallivory. J Exp
Mar Biol Ecol 390: 188−195
Rotjan RD, Lewis SM (2005) Selective predation by parrot-
fishes on the reef coral Porites astreoides. Mar Ecol Prog
Ser 305: 193−201
Rotjan RD, Lewis SM (2006) Parrotfish abundance and
selective corallivory on a Belizean coral reef. J Exp Mar
Biol Ecol 335: 292−301
Rotjan RD, Lewis SM (2008) Impact of coral predators on
tropical reefs. Mar Ecol Prog Ser 367: 73−91
Rotjan RD, Dimond JL, Thornhill D, Leichter J, Helmuth B,
Kemp D, Lewis SM (2006) Chronic parrotfish grazing
impedes coral recovery after bleaching. Coral Reefs 25: 
361−368
Rowan R (2004) Coral bleaching:  thermal adaptation in reef
coral symbionts. Nature 430: 742
Suggett DJ, Borowitzka MA, Prášil O (eds) (2010) Chloro-
phyll  a fluorescence in aquatic sciences:  methods and
applications. Developments in applied phycology, Vol 4.
Springer Science+Business Media, New York, NY
Szmant-Froelich A, Pilson MEQ (1980) The effects of feed-
ing frequency and symbiosis with zooxanthellae on the
biochemical composition of Astrangia danae Milne
Edwards and Haime, 1849. J Exp Mar Biol Ecol 48: 
85−97
Szmant-Froelich A, Pilson MEQ (1984) Effects of feeding
frequency and symbiosis with zooxanthellae on nitrogen
metabolism and respiration of the coral Astrangia pocu-
lata. Mar Biol 81: 153−162
Thornhill DJ, Kemp DW, Bruns BU, Fitt WK, Schmidt GW
(2008) Correspondence between cold tolerance and tem-
perate biogeography in a western Atlantic Symbio-
dinium (Dinophyta) lineage. J Phycol 44: 1126−1135
Van Veghel MLJ, Bak RPM (1994) Reproductive characteris-
tics of the polymorphic Caribbean reef building coral
Montastrea annularis. III. Reproduction in damaged and
regenerating colonies. Mar Ecol Prog Ser 109: 229−233
Van Woesik R (1998) Lesion healing on massive Porites spp.
corals. Mar Ecol Prog Ser 164: 213−220
Work TM, Aeby GS (2010) Wound repair in Montipora capi-
tata. J Invertebr Pathol 105: 116−119
99
Editorial responsibility: Peter Edmunds, 
Northridge, California, USA
Submitted: September 28, 2016; Accepted: March 9, 2017
Proofs received from author(s): April 8, 2017