MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 250: 117?124, 2003 Published March 26
INTRODUCTION
Coral reefs are typified and dominated by the mutu-
alism between cnidarian hosts (e.g. stony corals, octo-
corals, sea anemones) and dinoflagellate symbionts
(algae commonly called zooxanthellae). The algae re-
ceive the cnidarian?s nitrogenous wastes, and one of
the main benefits to the host is the translocation of
photosynthetically fixed substances from the zooxan-
thellae (Muscatine & Porter 1977). Coral reefs are
affected by environmental perturbations such as ele-
vated seawater temperatures and pathogens, which
have both increased in the last few decades (Rosen-
berg & Ben-Haim 2002). These perturbations often
cause coral bleaching (i.e. the loss of zooxanthellae
and/or zooxanthella pigments) (Glynn 1996).
The possibility of symbiont switching in coral-algal
mutualisms has been raised to explain the potential
adaptive value of coral bleaching (Buddemeier &
Fautin 1993, Ware et al. 1996). Under new environ-
mental conditions, a specific mutualistic pair, consist-
ing of an individual cnidarian genotype and its algal
genotype(s), may not work well as a unit. If symbiont
switching can occur, a new host-symbiont combination
might perform better in the new environment. If the
mutualistic unit is fixed, the cnidarian-zooxanthella
genotypic combination will have to adapt as a unit
to the new environment or perish. Understanding
whether individual hosts can change their individual
symbionts is therefore crucial for conservation ap-
proaches and for understanding the mutualisms as a
whole.
? Inter-Research 2003 ? www.int-res.com*Email: tlgoulet@olemiss.edu
Stability of an octocoral-algal symbiosis over time
and space
Tamar L. Goulet1, 2,*, Mary Alice Coffroth1
1Department of Biological Sciences, State University of New York at Buffalo, 661 Hochstetter Hall, Buffalo, New York 14260, USA
2Present address: Department of Biology, University of Mississippi, 524 Shoemaker Hall, Mississippi 38677, USA
ABSTRACT: In symbiosis, 2 taxonomically different organisms co-exist, each pursuing their own
agenda and yet, they are linked in one entity. A mutualistic symbiosis may break up if it is no longer
beneficial to either one of the partners. Changing needs over time or changing environmental condi-
tions may prompt symbiont switching. For example, corals may survive elevated temperatures by
switching their algal symbionts. If switching occurs, the new combination of host and symbiont geno-
types may perform better. Conversely, the partners may be fixed for life, with the degree to which the
mutualism responds to changing selection pressures dictated by the existing partners. Understand-
ing the genotypic dynamics of a mutualism is important for predicting the potential resilience of a
mutualism over time and in the face of environmental perturbations. Although mutualisms tend to be
characterized at the species level or higher, host-symbiont dynamics is an individual-level question,
requiring individual-level analysis. We used multilocus DNA fingerprinting to examine long-term
temporal and spatial symbiont change in the mutualism between the octocoral Plexaura kuna and its
algal symbionts (zooxanthellae). We monitored zooxanthella genotypes within a colony for up to
10 yr, among P. kuna clonemates, across different habitats and in colonies transplanted to novel envi-
ronments. In all instances, the prominent zooxanthella genotype within a P. kuna colony remained
unchanged although zooxanthella genotypes varied among genetically distinct P. kuna colonies.
Such tremendous temporal and spatial stability may occur in other coral hosts, influencing the
reaction and survival of mutualisms during environmental change.
KEY WORDS:  Symbiosis ? Zooxanthellae ? Corals ? Plexaura kuna ? Genotype ? Multilocus DNA
fingerprinting ? Stability
Resale or republication not permitted without written consent of the publisher
Mar Ecol Prog Ser 250: 117?124, 2003
Until recently, zooxanthella identity remained elu-
sive due to the paucity of differentiating morphological
characteristics. Rowan & Powers (Rowan 1991, Rowan
& Powers 1991, 1992), using sequence comparisons
and restriction fragment length polymorphisms (RFLP)
of DNA coding for small subunit ribosomal RNA
(SSuRNA), provided a relatively easy and attainable
division of zooxanthellae into genetically discernable
groups termed clades. Since then, researchers have
assigned zooxanthella cladal identity to symbionts in a
wide variety of hosts such as stony corals (Darius et al.
1998, Baker 1999, Loh et al. 2001, Rodriguez-Lanetty et
al. 2001, van Oppen et al. 2001, Diekmann et al. 2002,
LaJeunesse 2002), octocorals (Carlos et al. 1999, Gou-
let 1999, LaJeunesse 2002, Santos et al. 2002, Goulet &
Coffroth 2003), sponges (Hill & Wilcox 1998) and giant
clams (Baillie et al. 2000b).
Using the cladal level of resolution, Rowan & Knowlton
(1995), Rowan et al. (1997), Baker (2001) and Toller et
al. (2001a) documented that 10 coral species can host
several zooxanthella clades simultaneously. Following
environmental change, these corals displayed different
zooxanthella clades (Rowan et al. 1997, Baker 2001,
Toller et al. 2001b). It is unknown whether these clades
were previously present but undetected because they
were rare, or whether new clades entered the host from
the environment (Toller et al. 2001b, Hoegh-Guldberg et
al. 2002). Although the implication of zooxanthella cladal
turnover in a changing environment is intriguing, in fact,
such turnover appears to be the exception rather than
the rule. The majority of cnidarians (168 of 199 species)
host only 1 detectable zooxanthella clade (Rowan &
Powers 1991, Baker & Rowan 1997, Billinghurst et al.
1997, Bythell et al. 1997, Wilcox 1997, Darius et al. 1998,
Baker 1999, Goulet 1999, van Oppen et al. 2001, Diek-
mann et al. 2002), making cladal-level resolution not
adequate for resolving the dynamics of zooxanthellae,
either temporally or spatially, in the majority of these
mutualisms. Additional genetic techniques have been
utilized to elucidate variation within a clade (Baillie et
al. 1998, 2000a,b, Carlos et al. 2000, LaJeunesse 2001,
Santos et al. 2001); however, these techniques are still
too coarse to examine the individual level association
between a host genotype and its symbiont genotype(s).
We therefore chose multilocus DNA fingerprinting to
examine individual mutualisms. DNA fingerprinting of
total genomic DNA resolves between genetically dif-
ferent individuals (Jeffreys et al. 1985a,b) and has been
used for identifying distinct genotypes (Coffroth et al.
1992, Coffroth & Lasker 1998).
Here, we document the temporal and spatial dynam-
ics of zooxanthellae within the Caribbean octocoral
Plexaura kuna. As an adult, P. kuna hosts only zoo-
xanthellae belonging to Clade B (Rowan & Powers
1991, Goulet & Coffroth 2003). Using multilocus DNA
fingerprinting, we determined the zooxanthella geno-
typic identity in individual P. kuna colonies temporally
for up to a decade. Also, P. kuna is highly clonal,
frequently propagating asexually via fragmentation
(Coffroth & Lasker 1998). The presence of genetically
identical host colonies of varying ages allowed us to
extend the temporal analysis of within host zooxan-
thella dynamics across the decades of colony growth
and propagation. Spatially, we sampled clonemates
and determined the zooxanthella genotype(s) in a host
following a reciprocal transplant experiment.
MATERIALS AND METHODS
Temporal comparisons of zooxanthella genotypes.
First, we repeatedly sampled 21 individual Plexaura
kuna colonies from 3 reefs in the San Blas Islands,
Panama, over a maximum 10 yr period (1989 to 1998)
(Table 1). Second, we compared zooxanthella geno-
types in P. kuna clonemates from 12 genetically dis-
tinct P. kuna host clones. The P. kuna clones occurred
118
Table 1. Plexaura kuna. The similarity index (SI) of zooxan-
thella (Z) and host (H) DNA fingerprints from the same P.
kuna colony over a maximum 10 yr period. For Z versus Z
comparisons between years, the DNA fingerprints of zooxan-
thellae from all samples of a given colony were scored against
each other. For Z versus H comparisons, the zooxanthella
fingerprints were scored against the host. Samples were
collected in July or August 1989, 1991, 1993, 1994, 1995 and
1996; January and September 1997; and in April 1998.
Colonies were sampled at Korbiski (K), Sail Rock (S) and 
Tiantupo (Ti) reefs in the San Blas Islands, Panama
P. kuna Years sampled Maximum SI Z vs Z SI Z vs H
colony yr between
samples
K 522 89, 91, 95, 96, 98 10 1 0.16
K 523 89, 96, 98 10 1 0.43
K 525 89, 94, 96, 98 10 1 0.30
K 940 89, 91, 94, 98 10 1 0.39
S 1 91, 93, 94, 95, 98 8 1 0.00
S 11 91, 93, 98 8 1 0.00
S 26 95, 98 4 1 0.09
S 58 96, 98 3 1 0.00
S 63 96, 98 3 1 0.31
S 68 96, 98 3 1 0.18
S 73 96, 98 3 1 0.16
S 79 96, 98 3 1 0.09
S 84 96, 98 3 1 0.07
S 87 96, 98 3 1 0.07
Ti 549 93, 96, 98 6 1 0.06
Ti 579 94, 95, 96, 98 5 1 0.07
Ti 602 95, 96, 98 4 1 0.06
Ti 640 96, 98 3 1 0.06
Ti 647 96, 98 3 1 0.07
Ti 649 96, 98 3 1 0.06
Ti 655 96, 97, 97, 98 3 1 0.06
Goulet & Coffroth: Long-term zooxanthella stability
on 3 reefs in the San Blas Islands, Panama: Korbiski
reef (K, 2 clones), Sail Rock Reef (S, 4 clones), Tiantupo
Reef (Ti, 3 clones) and on a reef in the Florida Keys:
Three Sisters Reef (TS, 3 clones). The number of clone-
mates sampled from each clone ranged from 2 to 16
clonemates (Table 2). Samples from Clones D, H and T
from Three Sisters Reef included all colonies known to
belong to those clones (Coffroth & Lasker 1998).
Spatial comparisons of zooxanthella genotypes.
Since clonemates are exposed to different microhabi-
tats on the same reef, Plexaura kuna clones also pro-
vided an environmental comparison. In the TiA and
TiB clones on Tiantupo Reef, we found the clonemates
that were farthest apart (13 and 22 m, respectively)
and sampled all clonemates along the transect be-
tween these maximal distances. To determine whether
zooxanthellae within small colonies differed from large
ones, we measured the height of P. kuna colonies (cm)
found along the longest axis of the TiA and TiB clones.
To further test for habitat influence on the zooxan-
thella complement within a host, we conducted a reci-
procal transplant experiment on 2 reefs in the San Blas
Islands, Panama. We placed transplants in 3 natural
habitats for Plexaura kuna, Tiantupo backreef shallow
(TiS <5 m), Sail Rock forereef shallow (SS <5 m) and
Sail Rock forereef deep (SD >10 m). We also placed
transplants in 2 extreme habitats where Plexaura kuna
is not found (although other octocorals inhabit the
area). These were the very deep forereef of Sail Rock
(SV >20 m,) and the deep backreef of Tiantupo (TiD
>10 m). From the natural habitats, we sampled a mini-
mum of 3 colonies, 1 each from 3 P. kuna clones. In
addition, 2 colonies were sampled from 1 TiS clone
(TiC) and 1 SD clone (Si dp3). We placed a 15 cm
branch fragment from each source site on modified
cement flats (Kim & Lasker 1997), randomly choosing
fragments and their position on the flat. Since different
areas of a P. kuna colony have the same prominent
zooxanthella genotype (Goulet & Coffroth 1997, 2003),
all fragments initially had the same prominent zoo-
xanthella genotype. Fragments were transplanted in
August 1996 and were sampled in January 1997, Sep-
tember 1997 and finally 20 mo later, in April 1998. To
document that zooxanthellae respond to this level of
environmental change, zooxanthella density within a
0.5 cm piece of P. kuna tissue, 1.0 cm from the tip, was
counted using a hemocytometer. Six replicate cell
counts were conducted. Each source site was analyzed
separately. Zooxanthella counts were compared with a
2-way ANOVA (Sokal & Rohlf 1981).
DNA fingerprinting analysis. DNA extraction fol-
lowed the protocol of Coffroth et al. (1992). For each
host genotype, at least 1 zooxanthella DNA sample was
further cleaned by putting the zooxanthella pellet
through a 20 and 80% Percoll step gradient (Stochaj &
Grossman 1997). For DNA fingerprinting, the bacterio-
phage, M13, was used as a probe in a chemilumines-
cence hybridization procedure (Boehringer Mannheim)
and the bands visualized with Lumi-Phos 480 (Life
Codes). Since Percoll-cleaned zooxanthellae were run
side by side to the zooxanthella and host lanes of the
same Plexaura kuna colony, bands could confidently be
assigned to the host or symbiont. We conducted pair-
wise comparisons of fragment patterns only on samples
run on the same gel. DNA fingerprints were scored us-
ing bands between 3 and 23 kb, which  then had their
similarity index (SI) calculated (SI = 2 ? No. of shared
bands/[No. of bands in Individual A + No. of bands in
Individual B]; Lynch 1988). SI ranges from 0, for geneti-
cally different individuals (i.e. no bands shared), to
1, for genetically identical individuals (i.e. all bands
shared). Genetically different individuals, however,
may share some bands due to co-migration and back-
ground levels of population band-sharing.
The DNA fingerprinting technique examines the
whole genome and provides a composite picture if
multiple genotypes are present, detecting the most
abundant genotypes. For example, Goulet & Coffroth
(2003) demonstrated that the DNA fingerprinting tech-
nique detected a zooxanthella genotype if it consti-
tuted 10% or more of the sample. Those genotypes
that may occur in low abundance do not provide
enough DNA for detection with this technique. We
therefore refer to the genotype visualized in the DNA
fingerprint as the prominent zooxanthella genotype.
119
Table 2. Plexaura kuna. The similarity index (SI) of zooxan-
thella (Z) and host (H) DNA fingerprints obtained from P.
kuna clones. The DNA fingerprints of zooxanthellae from all
sampled clonemates (n) of a clone were scored against each
other. For Z versus H comparisons, the zooxanthella DNA fin-
gerprints from a particular host were scored against the DNA
fingerprint of the host. Colonies were sampled at Korbiski (K),
Sail Rock (S) and Tiantupo (Ti) reefs in the San Blas Islands,
Panama, and Three Sisters reef (TS) in the Florida Keys.
Clones sampled from K, Ti and TS reefs occurred in depths
<5 m. From S reef, Clone sh2 was found in <5 m depth, while
Clones dp1, dp2 and dp3 occurred between 10 and 15 m
Reef Clone n SI of Z vs Z SI of Z vs H
K KA 130 1 0.16
KC 4 1 0.30
S sh2 2 1 0.31
dp1 2 1 0.16
dp2 3 1 0.09
dp3 3 1 0.07
Ti TiA 160 1 0.07
TiB 130 1 0.06
TiC 9 1 0.06
TS D 4 1 0.00
H 2 1 0.20
T 2 1 0.18
Mar Ecol Prog Ser 250: 117?124, 2003
RESULTS
Temporal comparisons of zooxanthella genotypes
Each genetically distinct host colony displayed a
unique zooxanthella DNA fingerprint, termed the
prominent genotype (Goulet & Coffroth 1997, 2003),
which did not change throughout the sampling period
(Fig. 1, Table 1). The low SI values in the zooxanthella
versus host comparisons demonstrate that the DNA
fingerprinting technique resolves between zooxan-
thella and host DNA (Fig. 1, Table 1).
When we compared clonemates within 12 geneti-
cally distinct Plexaura kuna clones, we found that,
without exception (irrespective of colony size), all
clonemates of the same clone exhibited the exact same
zooxanthella DNA fingerprint (Fig. 1, Table 2). On the
other hand, zooxanthella DNA fingerprints did vary
between different P. kuna clones (Fig. 1, Table 3).
Spatial comparisons of zooxanthella genotypes
Some of the sampled clonemates of Plexaura kuna
clones were found on the same reef in habitats varying
in depth and exposure to currents and wave action,
and separated by as much as 75 m (Coffroth & Lasker
1998). Despite these environmental differences, all
clonemates from a given clone shared the same unique
zooxanthella DNA profile (Table 2). When P. kuna
fragments were transplanted to different reefs and dif-
ferent depths, transplants exhibited changes in zoo-
xanthella density and branch morphology. Fragments
from shallow depths, transplanted to another reef
and/or deeper water, had significantly fewer zooxan-
thellae per cm2 compared to the control transplant
maintained at the original location in shallow water
(Fisher?s protected least significant differences [PLSD],
p < 0.05; Fig. 2). Fragments of shallow forereef colonies
transplanted to backreef habitats (shallow and deep),
as well as fragments transplanted from shallow to deep
in the backreef habitat, were significantly smaller in
diameter after 20 mo (Fisher?s PLSD, p < 0.05). The
120
Fig. 1. Plexaura kuna. Autoradiograph of DNA fingerprints of
P. kuna hosts (H) and their zooxanthellae (all other lanes).
Lanes 1 to 6, counting from left to right, are of zooxanthellae
and the host colony (Lane 4) belonging to Clone KF sampled
repeatedly from 1989 to 1998 as follows: July or August 1989,
1991, 1995 and 1996; and April 1998. Lanes 7, 8 and 9 to 11
are of zooxanthellae and hosts from Clones KA and KC,
respectively, from Korbiski Reef. Lanes 12 to 16, 17 to 20 and
21 to 24 are of zooxanthellae and a representative host from
clonemates of the clones TiC, TiB and TiA, respectively, all
from Tiantupo Reef. Zooxanthella samples further cleaned 
with a percoll step gradient are denoted by a P
Table 3. Plexaura kuna. Comparison of zooxanthella DNA
fingerprints of genetically distinct P. kuna colonies. Pairwise
comparisons were only performed on samples run side by
side on the same DNA fingerprinting gel. Colonies were sam-
pled at Korbiski (K), Sail Rock (S) and Tiantupo (Ti) reefs in
the San Blas Islands, Panama, and Three Sisters reef (TS) in
the Florida Keys. SI: similarity index
Reef Clones compared SI
K KA vs KC 0.31
KA vs KF 0.21
KB vs KC 0.05
KC vs KD 0.29
S sh1 vs sh2 0.48
sh2 vs sh3 0.15
dp1 vs dp2 0.11
dp2 vs dp3 0.24
1 vs 11 0.28
11 vs 26 0.14
Ti TiA vs TiB 0.28
TiB vs TiC 0.55
TiC vs TiA 0.27
K vs Ti KC vs TiC 0.24
TS D vs H 0.19
H vs T 0.10
Ti vs TS TiB vs T 0.07
KF KA KC TiC TiB TiA
89 91P 95 H 96 98 1 H 1 2 H 1 2 H 3 4 1
2
P H 3 1
2
P 3 4
Goulet & Coffroth: Long-term zooxanthella stability
zooxanthella DNA fingerprints however, were identi-
cal across all 5 transplants of a given colony and iden-
tical to the zooxanthella DNA fingerprint from the orig-
inal colony, both at the start, middle and end of the
experiment. Thus, exposing the host to different envi-
ronments, including different reefs and depths, did not
alter the prominent zooxanthella genotype.
DISCUSSION
In a mutualistic association, a host and its symbiont
associate, to the benefit of both partners. Changing
needs, occurring over the host?s lifetime or because of
environmental change, may trigger symbiont switch-
ing. Our findings demonstrate a remarkable stability of
the mutualism between the Caribbean octocoral Plex-
aura kuna and its zooxanthellae. A P. kuna colony
exhibited the same prominent zooxanthella genotype
for as long as 10 yr. During this 10 yr period, elevated
water temperatures occurred in 1995 (CARICOMP
1997), causing several corals to bleach (Rowan et al.
1997). In P. kuna, a reduction in zooxanthella numbers
of >95% results in only a slight lightening of the colony
(S. R. Santos & M. A. Coffroth unpubl. data). Therefore,
although P. kuna can lose algae, visibly it does not look
bleached. Even with a potential reduction in zooxan-
thella numbers during the elevated temperature epi-
sode, the 8 colonies sampled before and after 1995
retained their unique zooxanthella DNA fingerprint.
Since we did not sample immediately after the rise in
temperature, zooxanthella turnover could have oc-
curred and then reverted to the original prominent
zooxanthella genotype. This scenario is not probable
because a P. kuna colony sampled at short intervals
(6 mo) exhibited the same zooxanthella DNA finger-
print. Therefore, in long-term monitoring of zooxan-
thella genotypes in the same individual host colony,
the zooxanthella prominent genotype did not change.
The clonal nature of Plexaura kuna allowed us to
further extend our temporal comparison, following the
symbiosis potentially throughout a host genet?s life.
Since different areas of a P. kuna colony have the same
prominent zooxanthella genotype (Goulet & Coffroth
2003), a branch broken from a colony (i.e. a new clone)
initially would have the same prominent zooxanthella
genotype as the parent colony. Small-sized P. kuna
clonemates could have different energetic demands
and allocations that might favor different symbionts.
Different P. kuna clonemates, situated in different
microhabitats, could have different requirements from
their symbionts. If different host-symbiont genotypic
combinations could occur, we would expect to see
them in some P. kuna clonemates. Yet, when compar-
ing P. kuna clonemates, ranging in size and in micro-
habitat on the same reef, the prominent zooxanthella
genotype within a P. kuna clone was always identical
between clonemates.
The zooxanthella genotypic stability exhibited in
Plexaura kuna cannot be attributed to a lack of zoo-
xanthella variability in zooxanthella populations at
large, since zooxanthella DNA fingerprints differed
between P. kuna clones (Table 3). Furthermore, other
cnidarian species on the same reefs harbor multiple
zooxanthella clades (Rowan & Knowlton 1995, Rowan
et al. 1997, Baker 2001, Toller et al. 2001a), demon-
strating the diversity of zooxanthella clades available
on these reefs. On a given reef, the host-symbiont
combination may be optimal given the zooxanthella
genotypes available. In addition, the environmental
variability on a single reef may not be great enough to
select among symbiont genotypes. The transplant
experiment, however, exposed P. kuna to new habitats
and environmental conditions.
121
Fig. 2. Plexaura kuna. Zooxanthella density of P. kuna frag-
ments transplanted to various depths within and among reefs.
Each bar represents the mean + SD of 3 colonies transplanted
from the source site (a, b or c) to a transplanted site (Ti =
Tiantupo; S = Sail Rock) of different depths (identified by fol-
lowing letters after site abbreviations: S = shallow; D = deep;
V = very deep) sampled 20 mo after the transplant. C = the
original colonies. *: a significant difference (p < 0.0001)
between zooxanthella numbers in the transplant at the origi-
nal site and the other treatments, as determined by a Fisher?s 
PLSD post-hoc test (Sokal & Rohlf 1981)
Mar Ecol Prog Ser 250: 117?124, 2003
The transplant experiment reported here placed
Plexaura kuna colonies on different reefs at different
depths, including an extreme depth, where P. kuna
is not normally found, thereby potentially exposing
P. kuna to a new pool of zooxanthellae. Transplanting
P. kuna did affect features of the host, as seen by a
morphological change and, in some cases, affected
zooxanthella densities, demonstrating both a host and
symbiont response to the treatments and the success of
the transplant experiment. The zooxanthella DNA fin-
gerprints however, were identical across all 5 trans-
plants of a given colony and to the zooxanthella DNA
fingerprint from the original colony. Furthermore,
since the transplants were sampled throughout the
year, August, January, September and April, no sea-
sonal change in zooxanthella genotype was observed.
The transplant experiment demonstrated that not only
temporally but spatially exposing the host to different
environments, including different reefs and depths,
did not alter the prominent zooxanthella genotype.
The stasis of the prominent zooxanthella genotype
implies the presence of a mechanism that maintains a
specific symbiont, once it has entered the host. Primary
polyps of Plexaura kuna can initially acquire multiple
zooxanthella genotypes, including zooxanthellae be-
longing to different clades (Coffroth et al. 2001). After
initial acquisition, however, only Clade B zooxanthel-
lae were detected in all P. kuna colonies sampled
(Goulet 1999, Goulet & Coffroth 2003), even among
colonies as small as 0.5 cm in height (Coffroth et al.
2001). The prominent zooxanthella genotype may
competitively exclude other genotypes attempting to
enter the association. Fitt (1985) found that zooxan-
thella strains differed in their growth rate, and in a host
infected with 2 strains, faster growing strains outgrew
slower growing strains. Schoenberg & Trench (1976)
demonstrated that under normal conditions, resident
zooxanthella strains grew faster in their host compared
to introduced, foreign zooxanthella strains. Alterna-
tively, the host may be the one to selectively retain its
symbionts. At present, there are no data to evaluate
whether either the host or symbiont is maintaining
control over the symbiosis genotypic composition.
The lack of symbiont change is consistent with theo-
ries explaining the establishment of zooxanthella pop-
ulations in hosts with horizontal transmission (Wilcox
1997, Genkai-Kato & Yamamura 1999). In systems
where symbionts are acquired every generation, bene-
volent symbionts (those that are not detrimental to the
host) will remain if symbionts are acquired only once in
a life cycle and the acquisition stage is sensitive to
selection. ?Selfish symbionts?, symbionts that increase
their own growth at the expense of the host, will be
selected against (Wilcox 1997). Limiting symbiont
genotypes within a host also avoids competition among
symbiont genotypes that may lower host fitness (Frank
1996).
Using multilocus DNA fingerprinting, we demon-
strated that in the octocoral Plexaura kuna, the zoo-
xanthella genotypes are stable both temporally and
spatially. If the persistence of a fixed host-symbiont
pairing seen in the P. kuna-zooxanthella symbiosis is
representative of the other 84.4% of cnidarian mutu-
alisms that normally harbor a single zooxanthella
clade, then our findings have implications for predic-
tions of the response of reef corals to global warming
and for reef conservation measures. For example, the
symbiosis stability may indicate that reefs are more
resilient to global warming than predicted. On the
other hand, and of more concern, this stability may
represent the inability of the host to switch its zooxan-
thellae. If this were the case, then a bleached coral
could only recuperate through re-population by the
remaining algal population and not by acquiring zoo-
xanthellae from the environment. The question then
becomes not whether symbiont switching is possible,
but how far existing symbiotic combinations can be
pushed before symbiosis de-coupling occurs. Deter-
mining the degree of elasticity of a symbiosis will then
aid in predicting survival of specific host-symbiont
combinations during environmental changes brought
on by global warming or direct human influence.
Acknowledgements. We thank E. Beiring, A. Dwileski-Pfister,
D. Goulet, N. Knowlton, H. Lasker, R. Rowan, S. Santos,
T. Swain, A. Verde, the Kuna Indians and the Republic
of Panama. The comments of R. Buchholz, D. Goulet, T.
Threlkeld and anonymous reviewers greatly improved the
manuscript. This work was supported by: Lerner Gray Fund
for Marine Research; Houston Underwater Club; Mark Dia-
mond Research Fund; PADI Foundation; Sigma Xi Society;
and a Smithsonian Tropical Research Institute, short-term fel-
lowship to T.L.G.; and the National Undersea Research Cen-
ter at University of North Carolina at Wilmington and
National Science Foundation OCE-9530057 to M.A.C.
LITERATURE CITED
Baillie BK, Monje V, Silvestre V, Sison M, Belda-Baillie CA
(1998) Allozyme electrophoresis as a tool for distinguish-
ing different zooxanthellae symbiotic with giant clams.
Proc R Soc Lond Ser B 265:1949?1956
Baillie BK, Belda BCA, Maruyama T (2000a) Conspecificity
and Indo-Pacific distribution of Symbiodinium genotypes
(Dinophyceae) from giant clams. J Phycol 36:1153?1161
Baillie BK, Belda BCA, Silvestre V, Sison M, Gomez AV,
Gomez ED, Monje V (2000b) Genetic variation in Symbio-
dinium isolates from giant clams based on random-
amplified-polymorphic DNA (RAPD) patterns. Mar Biol
136:829?836
Baker AC (1999) The symbiosis ecology of reef-building
corals. PhD thesis, University of Miami, Coral Gables, FL
Baker AC (2001) Reef corals bleach to survive change. Nature
411:765?766
122
Goulet & Coffroth: Long-term zooxanthella stability
Baker AC, Rowan R (1997) Diversity of symbiotic dinoflagel-
lates (zooxanthellae) in scleractinian corals of the Carib-
bean and Eastern Pacific. Proc 8th Int Coral Reef Symp,
Panama 2:1301?1306
Billinghurst Z, Douglas AE, Trapido-Rosenthal HG (1997) On
the genetic diversity of the symbiosis between the coral
Montastraea cavernosa and zooxanthellae in Bermuda.
Proc 8th Int Coral Reef Symp, Panama 2:1291?1294
Buddemeier RW, Fautin DG (1993) Coral bleaching as an
adaptive mechanism. BioScience 43:320?326
Bythell JC, Douglas AE, Sharp VA, Searle JB, Brown BE
(1997) Algal genotype and photoacclimatory responses of
the symbiotic alga Symbiodinium in natural populations of
the sea anemone Anemonia viridis. Proc R Soc Lond Ser B
264:1277?1282
CARICOMP (Caribbean Coastal Marine Productivity) (1997)
Studies on Caribbean coral bleaching, 1995. Proc 8th Int
Coral Reef Symp, Panama 1:673?678
Carlos AA, Baillie BK, Kawachi M, Maruyama T (1999) Phylo-
genetic position of Symbiodinium (Dinophyceae) isolates
from tridacnids (Bivalvia), cardiids (Bivalvia), a sponge
(Porifera), a soft coral (Anthozoa), and a free-living strain.
J Phycol 35:1054?1062
Carlos A, Baillie B, Maruyama T (2000) Diversity of dinofla-
gellate symbionts (zooxanthellae) in a host individual. Mar
Ecol Prog Ser 195:93?100
Coffroth MA, Lasker HR (1998) Population structure of a
clonal gorgonian coral: the interplay between clonal re-
production and disturbance. Evolution 52:316?330
Coffroth MA, Lasker HR, Diamond ME, Bruenn JA, Berming-
ham E (1992) DNA fingerprints of a gorgonian coral: a
method for detecting clonal structure in a vegetative
species. Mar Biol 114:317?325
Coffroth MA, Santos SR, Goulet TL (2001) Early ontogenetic
expression of specificity in a cnidarian-algal symbiosis.
Mar Ecol Prog Ser 222:85?96
Darius HT, Dauga C, Grimont PAD, Chungue E, Martin PMV
(1998) Diversity in symbiotic dinoflagellates (Pyrrhophyta)
from 7 scleractinian coral species: restriction enzyme ana-
lysis of small subunit ribosomal RNA genes. J Eukaryot
Microbiol 45:619?627
Diekmann OE, Bak RPM, Tonk L, Stam WT, Olsen JL (2002)
No habitat correlation of zooxanthellae in the coral genus
Madracis on a Cura?ao reef. Mar Ecol Prog Ser 227:
221?232
Fitt WK (1985) Effect of different strains on the zooxanthella
Symbiodinium microadriaticum on growth and survival of
their coelenterate and molluscan hosts. Proc 5th Int Coral
Reef Congr, Tahiti 6:131?136
Frank SA (1996) Host-symbiont conflict over the mixing of
symbiotic lineages. Proc R Soc Lond Ser B 263:339?344
Genkai-Kato M, Yamamura N (1999) Evolution of mutualistic
symbiosis without vertical transmission. Theor Popul Biol
55:309?323
Glynn PW (1996) Coral reef bleaching: facts, hypotheses and
implications. Global Change Biol 2:495?509
Goulet TL (1999) Temporal and spatial stability of zooxanthel-
lae in octocorals. PhD thesis, State University of New York
at Buffalo
Goulet TL, Coffroth MA (1997) A within colony comparison
of zooxanthellae genotypes in the Caribbean gorgonian
Plexaura kuna. Proc 8th Int Coral Reef Symp, Panama 2:
1331?1334
Goulet TL, Coffroth MA (2003) Genetic composition of zoo-
xanthellae between and within colonnies of the octocoral
Plexaura kuna based on small subunit rDNA and multi-
locus DNA fingerprinting. Mar Biol 142:233?239
Hill M, Wilcox T (1998) Unusual mode of symbiont repopula-
tion after bleaching in Anthosigmella varians: acquisition
of different zooxanthellae strains. Symbiosis 25:279?289
Hoegh-Guldberg O, Jones RJ, Ward S, Loh W (2002) Ecology:
is coral bleaching really adaptive? Nature 415:601?602
Jeffreys AJ, Wilson V, Thein SL (1985a) Hypervariable ?mini-
satellite? regions in human DNA. Nature 314:67?73
Jeffreys AJ, Wilson V, Thein SL (1985b) Individual-specific
?fingerprints? of human DNA. Nature 316:76?79
Kim K, Lasker HR (1997) Flow-mediated resource competition
in the suspension feeding gorgonian Plexaura homomalla
(Esper). J Exp Mar Biol Ecol 215:49?64
LaJeunesse TC (2001) Investigating the biodiversity, ecology,
and phylogeny of endosymbiotic dinoflagellates in the
genus Symbiodinium using the ITS region: in search of a
?species? level marker. J Phycol 37:866?880
LaJeunesse TC (2002) Diversity and community structure of
symbiotic dinoflagellates from Caribbean coral reefs. Mar
Biol 141:387?400
Loh WKW, Loi T, Carter D, Hoegh-Guldberg O (2001) Genetic
variability of the symbiotic dinoflagellates from the wide
ranging coral species Seriatopora hystrix and Acropora
longicyathus in the Indo-West Pacific. Mar Ecol Prog Ser
222:97?107
Lynch M (1988) Estimation of relatedness by DNA finger-
printing. Mol Biol Evol 5:584?599
Muscatine L, Porter J (1977) Reef corals: mutualistic sym-
bioses adapted to nutrient-poor environments. BioScience
27:454?460
Rodriguez-Lanetty M, Loh W, Carter D, Hoegh-Guldberg O
(2001) Latitudinal variability in symbiont specificity within
the widespread scleractinian coral Plesiastrea versipora.
Mar Biol 138:1175?1181
Rosenberg E, Ben-Haim Y (2002) Microbial diseases of corals
and global warming. Environ Microbiol 4:318?326
Rowan R (1991) Molecular systematics of symbiotic algae.
J Phycol 27:661?666
Rowan R, Knowlton N (1995) Intraspecific diversity and eco-
logical zonation in coral-algal symbiosis. Proc Natl Acad
Sci USA 92:2850?2853
Rowan R, Powers DA (1991) A molecular genetic classification
of zooxanthellae and the evolution of animal-algal sym-
biosis. Science 251:1348?1351
Rowan R, Powers DA (1992) Ribosomal RNA sequences and
the diversity of symbiotic dinoflagellates (zooxanthellae).
Proc Natl Acad Sci USA 89:3639?3643
Rowan R, Knowlton N, Baker A, Jara J (1997) Landscape eco-
logy of algal symbionts creates variation in episodes of
coral bleaching. Nature 388:265?269
Santos SR, Taylor DJ, Coffroth MA (2001) Genetic com-
parisons of freshly isolated versus cultured symbiotic dino-
flagellates: implications for extrapolating to the intact
symbiosis. J Phycol 37:900?912
Santos SR, Taylor DJ, Kinzie RAI, Hidaka M, Sakai K, Coffroth
MA (2002) Molecular phylogeny of symbiotic dinoflagel-
lates inferred from partial chloroplast large subunit (23S)-
rDNA sequences. Mol Phylogenet Evol 23:97?111
Schoenberg DA, Trench RK (1976) Specificity of symbioses
between marine cnidarians and zooxanthellae. In: Mackie
GO (ed) Coelenterate ecology and behavior. Plenum
Press, New York, p 423?432
Sokal RR, Rohlf FJ (1981) Biometry, 2nd edn. WH Freeman,
San Francisco
Stochaj WR, Grossman AR (1997) Differences in the protein
profiles of cultured and endosymbiotic Symbiodinium sp.
(Pyrrophyta) from the anemone Aiptasia pallida (Antho-
zoa). J Phycol 33:44?53
123
Mar Ecol Prog Ser 250: 117?124, 2003
Toller WW, Rowan R, Knowlton N (2001a) Zooxanthellae
of the Montastraea annularis species complex: patterns
of distribution of 4 taxa of Symbiodinium on different
reefs and across depths. Biol Bull (Woods Hole) 201:
348?359
Toller WW, Rowan R, Knowlton N (2001b) Repopulation of
zooxanthellae in the Caribbean corals Montastaea annu-
laris and M. faveolata following experimental and dis-
ease-associated bleaching. Biol Bull (Woods Hole) 201:
360?373
van Oppen MJH, Palstra FP, Piquet AMT, Miller DJ (2001)
Patterns of coral-dinoflagellate associations in Acropora:
significance of local availability and physiology of Sym-
biodinium strains and host-symbiont selectivity. Proc
R Soc Lond Ser B 268:1759?1767
Ware JR, Fautin DG, Buddemeier RW (1996) Patterns of coral
bleaching: modeling the adaptive bleaching hypothesis.
Ecol Model 84:199?214
Wilcox TP (1997) The evolution of algal-invertebrate sym-
bioses. PhD thesis, University of Houston, TX
124
Editorial responsibility: Otto Kinne (Editor), 
Oldendorf/Luhe, Germany
Submitted: November 6, 2002; Accepted: November 21, 2002
Proofs received from author(s): March 10, 2003