Marine biology
rsbl.royalsocietypublishing.org
How do coral barnacles start their life in their hosts? Jennie Chien Wen Liu1,2, Jens Thorvald Høeg3 and Benny K. K. Chan1 1
Research Cite this article: Liu JCW, Høeg JT, Chan BKK. 2016 How do coral barnacles start their life in their hosts? Biol. Lett. 12: 20160124. http://dx.doi.org/10.1098/rsbl.2016.0124
Received: 14 February 2016 Accepted: 23 May 2016
Subject Areas: developmental biology, ecology Keywords: coral-associated barnacle, cyprid, metamorphosis
Author for correspondence: Benny K. K. Chan e-mail: [email protected]
Electronic supplementary material is available at http://dx.doi.org/10.1098/rsbl.2016.0124 or via http://rsbl.royalsocietypublishing.org.
Biodiversity Research Center, Academia Sinica, Taipei 115, Taiwan, Republic of China Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei 106, Taiwan, Republic of China 3 Marine Biology Section, Department of Biology, University of Copenhagen, DK-2100, Copenhagen, Denmark 2
BKKC, 0000-0001-9479-024X Coral-associated invertebrates are the most significant contributors to the diversity of reef ecosystems, but no studies have examined how larvae manage to settle and grow in their coral hosts. Video recordings were used to document this process in the coral barnacle Darwiniella angularis associated with the coral Cyphastrea chalcidicum. Settlement and metamorphosis in feeding juveniles lasted 8–11 days and comprised six phases. The settling cyprid starts by poking its antennules into the tissue of the prospective host (I: probing stage). The coral releases digestive filaments for defence, but tolerating such attack the cyprid penetrates further (II: battling stage). Ecdysis is completed 2 days after settlement (III: carapace detachment). The barnacle becomes embedded deep in the coral tissue while completing metamorphosis between 4 and 6 days (IV: embedding stage), but reappears as a feeding juvenile 8– 11 days after settlement (V: emerging stage; VI: feeding stage). Cyprids preferably settle in areas between the coral polyps, where they have a much higher survival rate than on the polyp surfaces.
1. Introduction Numerous invertebrates are associated with coral reef ecosystems, and many of these are symbiotic forms that contribute significantly to the overall biodiversity of the reef fauna [1]. Coral-associated barnacles (Thoracica: Pyrgomatidae) are particularly common and occur in almost all coral species worldwide [2]. These suspension feeding barnacles have their shell plates embedded in the coral skeleton (figure 1a,b), and the symbiotic relationship extends to the nutritional level with the zooxanthellae in the coral. The barnacles ingest organic matter released from the zooxanthellae, while the zooxanthellae in turn absorb ammonium released by the barnacles [3]. Knowing how coral-associated animals, such as barnacles, manage to settle in their potentially hostile hosts armed with nematocyst defences, is essential for understanding symbiotic relationships, and the patterns and processes that govern reef biodiversity. There are around 80 species of coral barnacles with varying degrees of host specificity, but difficulties in larval rearing and maintenance of corals for experiments have until now prevented such studies. Coral barnacles disperse by nauplius larvae, resembling those in other cirripedes (figure 1c), but the terminal cyprid stage is remarkably specialized (figure 1d,e,g,h). Their antennules have spear-shaped attachment organs (figure 1d,e,g,h), a marked contrast to the bell-shaped ones found in cyprids of other barnacles [4,5] (figure 1f ). We have successfully cultured larvae of the coral barnacle Darwiniella angularis (figure 1a,b), which is fairly host specific, being reported 75% on Cyphastrea chalcidicum, and 25% on Astreopora sp. in Taiwan [6]. Using video microscopy, we recorded for the first time the entire
& 2016 The Author(s) Published by the Royal Society. All rights reserved.
barnacle shell
(a)
2
(e)
(c)
rsbl.royalsocietypublishing.org
orifice
100 AS3
(d)
coral polyps
(b) AS2 cirri PS3
20
100
(g)
(f)
tae
y or
sensory setae
se
ns
se
t organ attachmen
AS 4
AS4 attachment organ
5
10
(h)
antennules
ds
opo
100
ac hor
t
Figure 1. Morphological features of Darwiniella angularis. (a) Darwiniella angularis (white arrows) associated with the coral Cyphastrea chalcidicum in Taiwan. (b) Magnified view of D. angularis showing cirri extended. (c) Nauplius larva (stage II). (d) Lateral view of a live cyprid. (e) Magnified view of cypris antennule. AS, antennulary segment; PS, postaxial seta (2, 3, 4—segments 2, 3, 4). (f ) In the free-living barnacle Balanus perforatus the antennules have a bell-shaped attachment organ with a circular attachment disc. Antennulary segment four (AS4) carries various sensory setae for substratum location. (g) The spear-shaped attachment organ of D. angularis cyprids differs extensively from the bell-shaped one (f ) in free-living barnacles. (h) Cyprid of D. angularis under SEM. Scale bar in micrometres. sequence of events from cyprid settlement in the coral host, to the appearance of a feeding juvenile barnacle.
2. Material and methods
(white LED lamps). Seawater was changed daily. Nauplii were fed with mixed green algae and diatoms (Tetraselmis chui, Chaetoceros muelleri, Isochrysis lutea, Skeletonema costatum). The cultures were maintained at 268C, with more than 80% of the larvae successfully developed into cyprids.
(a) Collection and larval culture
(b) Settlement and metamorphosis
Corals C. chalcidicum with adult D. angularis were collected by scuba divers in northeast Taiwan between 7 and 10 m. Coral pieces hosting barnacles (approx. 30 cm2) were maintained in 1 l beakers, using filtered seawater under a 10 L : 14 D cycle
Cyprids were exposed to coral pieces (approx. 5 cm2) and kept in aquaria (30 15 6 cm) with weak aeration. Among 735 cyprids, 85 settled on the coral hosts. Unsettled cyprids kept swimming and eventually died. The settlement location of
Biol. Lett. 12: 20160124
orifice 500
1000
AS4
PS2
shell
DObs pffiffiffiffiffiffiffiffi , a=n
0:5
where DObs is the mean observed nearest-neighbour distance; a is the area under study; and n is the total number of individuals. Rn ranges from 0 to 2.15, representing distribution patterns from clustered (less than 1), over random (1), to regular (more than 1).
3. Results Darwiniella angularis requires 8 –10 days to develop through six naupliar stages into cyprids. Video recordings of the entire metamorphosis process, from settled cyprids to feeding settlers, can be viewed in the electronic supplementary materials. Settlement and metamorphosis lasted 8– 11 days and comprised 6 distinct phases (figure 2a– l). When exposed to the coral, the cyprid first poked its antennules into the soft tissue and then pulled its entire carapace down into the coral (Phase I, probing stage, figure 2a,d). Subsequently, a battle started between coral and settled larva. The cyprid used its spear-shaped antennules to penetrate further into the coral, which released digestive filaments in defence. All the settling cyprids were capable of tolerating these attacks, and continued to penetrate further into the tissue (Phase II, battling stage, figure 2b,e). The cypris carapace detached 2 days after settlement, thus completing ecdysis (Phase III, carapace detachment, figure 2c,f ). Following detachment, the barnacle entered deeper inside the coral tissue, where hidden from view, it used another 4–6 days to complete metamorphosis (Phase IV, embedding stage, figure 2g,j). Individuals dissected free during this hidden phase were developing both an apical mantle aperture and mineral shell plates (figure 2j ). About 10 days after settlement, the early juvenile emerged on the coral surface (Phase V, emerging stage, figure 2h,k). Feeding started the following day (Phase VI, feeding settler, figure 2i,l). During settlement, the cyprids appeared to discriminate between the various locations on the coral. Among the 85 cyprids that settled on the coral, 62.59 + 6.22% (mean + 1s.d., n ¼ 3) settled on tissue areas between the polyps, 28.54 + 4.82% on adult barnacle shells, but only 8.87 + 1.69% on the polyps themselves (x 2 (2, N ¼ 85) ¼ 34.73, p , 0.001). Cyprids attached to polyps were all expelled during the battling to the embedding stage, subsequently dying, while more than 95% of those that settled in spaces between polyps survived into feeding barnacles (x 2 (2, N ¼ 85) ¼ 51.73, p , 0.001) Cyprids preferred to settle on corals hosting conspecifics. In the choice experiment, 23 of the 26 settled cyprids (88.4%)
4. Discussion Symbiosis is an increasingly important research area, especially in complex coral reef systems. This is the first study documenting how the larvae of coral-associated barnacles settle in live coral tissue and start their symbiotic life. The ability of the cyprids to tolerate the nematocyst defences of the host is apparently a key to their survival and subsequent embedding deep into the coral tissue, where they can complete metamorphosis unimpeded. An additional factor that aids the survival of the cyprids is their ability to preferentially settle away from the actual polyps (see results of x 2 settlement test). A high level of settlement site discrimination is known from other barnacles [7]. The preference for gregarious settlement in barnacles seems to also favour and enhance mating success. We therefore preliminarily conclude that the distributional pattern of barnacles on their coral host is a function of larval settlement site selection, conspecific settlement cues and post-settlement mortality. In trials where we exposed D. angularis cyprids to the nonhost coral Lepastrea transversa, the larvae only swam in the water column but never ‘walked’ on the coral surfaces. They even fled rapidly away if accidentally contacting the non-host coral. The ability of D. angularis cyprids to discriminate between coral species and select specific sites on the proper host highlights the importance of their antennular sensory organs for survival during settlement. Surprisingly, examination of scanning electron micrographs (SEM) of both D. angularis and other coral barnacles revealed that the cypris sensory setae are similar in structure and number to those found in more conventional species (figure 1g) [4]. Hence, settlement site specificity probably depends on neurological processes rather than on morphological details in the setae. It would therefore be exciting to combine more refined settlement experiments with electrophysiological recordings from the larvae. Coral barnacles take longer to complete metamorphosis than any other thoracican barnacles studied (figure 2m,n). For example, the metamorphosis from initial attachment of the cyprid to shedding of the carapace lasted less than 1 day in the intertidal free-living Amphibalanus amphitrite, 1.5 days in the subtidal free-living Megabalanus rosa (figure 2m) and 3.5 days in the neustonic species Lepas sp. These three species commenced feeding shortly after carapace shedding [8,9]. By contrast, D. angularis completed ecdysis in 2 days but required another 8 days to become a feeding juvenile (figure 2n). This prolonged metamorphosis is most likely owing to the time used for battling with the coral host soon after settlement, and the need to complete the rest of the metamorphosis process hidden deep in the coral tissue. These events can be easily compared to parasitic barnacles (Rhizocephala), which must also combat host defences during settlement, thereafter needing an internal parasitic phase to establish full control of their host before they re-emerge [10]. Interestingly, both coral barnacles (80þ species) and rhizocephalans (250þ species) are major components to the overall cirripede diversity (ca 1200 species). Apparently,
3
Biol. Lett. 12: 20160124
Rn ¼
attached on coral pieces already hosting D. angularis (x 2 (1, N ¼ 26) ¼ 15.38, p , 0.001). The nearest-neighbour analysis showed that all three juvenile populations collected in the wild (Rn1,2,3) had a clustered distribution, with the barnacles concentrated in spaces between polyps (Rn1 ¼ 0.383, Rn2 ¼ 0.615, Rn3 ¼ 0.667).
rsbl.royalsocietypublishing.org
each cyprid (i.e. on polyps, between polyps, and on shells of live D. angularis on corals) was recorded using a video camera (Lumix GH1) under a stereomicroscope. A x 2-test was used to examine whether the numbers of settlers and post-settlement juveniles (numbers pooled from three pieces of coral in each of three aquaria, cyprids numbers in the three aquaria: 103, 291, 177) were similar in the three indicated locations above. In a separate aquaria, 164 cyprids were exposed to two coral pieces with and without D. angularis. An additional x 2-test was used to examine whether the settlers equally preferred the two pieces. The distribution pattern of D. angularis in the wild was examined by the nearest-neighbour analysis of 44 juveniles (shell diameter , 1 mm) located on the surface of three coral pieces. The nearest-neighbour index (Rn) is based on the formula:
4 phase II
(b)
probing stage
phase III
(c)
battling stage
carapace detachment separated epidermis
ca
rap
ac e
cyprid
100
100
digestive filaments
nt ce
me
(e)
(f)
1
10
cypris thorax
2 days primordial scutum
primordial tergum
cypr i
d apical
mantle coral antennule
100
phase IV
(g)
(h) phase V (lateral view)
embedding stage
100
(i) phase VI
emerging stage
embedding individual
feeding settler fused scutum and tergum
(top view) cirri of the feeding settler
digestive filaments
scutum and tergum underneath the coral tissue
shell
100
3 days
100
100
10 days
11 days apical
(j)
(k)
shell
cirri
base
Megabalanus rosa
basal
base
l
primordial scutum and tergum
(l)
shell
shel
(m)
base
100
100 fused scutum and tergum
100
100
ecdysis
nauplius eye compound eye
separated inclined from antennule epidermis the substratum antennular cuticle and compound eyes
Darwiniella angularis
(n)
sis
digestive filaments
co
cirri
ec dy
emerging individual
separated epidermis
ral
0h
7h
1.5 days
2 days
4 days
10 days
Figure 2. Metamorphosis in Darwiniella angularis. (a – l) Light and scanning electron microscopy of phases in settlement and metamorphosis. Phase I: probing stage (a,d). The cyprid pokes its spear-shaped antennules into the soft coral tissue. Phase II: battling stage (b,e). The coral hosts release digestive filaments in defence; 28 h post-settlement the cyprid epidermis has separated from the carapace cuticle and its tissues start to concentrate (b). Cyprid removed from the coral tissue (e), showing the tiny mass of cement secreted by the antennule (circle inset) for anchorage. Phase III: carapace detachment (c,f ). Two days postsettlement, cyprid carapace and thorax are shed, revealing body of the metamorphosing barnacle with primordial shell plates. Phase IV: embedding stage (g,j). The metamorphosing cyprid remains totally embedded in coral tissue for several days. Removing such individuals ( j ) reveals incipient formation of shell plates. Phase V: emerging stage (h,k). The metamorphosed settler re-emerges from the coral tissue. Phase VI: feeding settler (i,l). One day later, the juvenile barnacle starts cirral suspension feeding. Scale bar in micrometres. (m,n) Comparison of metamorphosis in Megabalanus rosa [8] and D. angularis. Although overall comparable, the metamorphosis differs in both specific events and timing. (m) In M. rosa, metamorphosis is completed in 1.5 days. (n) In D. angularis, the cyprid first pokes its antennules into the soft coral tissue. Metamorphosis proceeds while completely hidden in the coral tissue and is not completed until 10 days after attachment. (Online version in colour.)
Biol. Lett. 12: 20160124
18 h
coral
(d )
100
rsbl.royalsocietypublishing.org
phase I
(a)
solving the problem of settlement on such hostile hosts entails opportunities for evolutionary success.
final version and agree to be accountable for the work in the manuscript.
5
Ethics. This research complies with laws of the Taiwan, Republic of
Competing interests. No competing interests. Funding. Supported by the Senior Investigator Award to B.K.K.C. and
rsbl.royalsocietypublishing.org
China, regarding research. There is no need for ethical approval for experiments involving barnacle larvae in Taiwan. Data accessibility. Settlement experiment data and video of metamorphosis are provided in the electronic supplementary material.
Authors’ contributions. B.K.K.C. conceived the study; J.C.W.L., B.K.K.C. and J.T.H. acquired and analysed the data and all authors contributed to the preparation of the manuscript. All authors approved the
the Carlsberg Foundation (2013_01_0130) to J.T.H.
Acknowledgements. We thank Yao-Fung Tsao, Shu-Mei Lin, Wei-Peng Hsieh and Pei-Chen Tsai (Academia Sinica) for help with larval culture. Thanks to Brett Gonzales for editing the English of the manuscript.
1.
2.
3.
4.
Blackall LL, Wilson B, van Oppen MJH. 2015 Coral – the world’s most diverse symbiotic ecosystem. Mol. Ecol. 24, 5330–5347. (doi:10.1111/mec.13400) Tsang LM, Chu KH, Nozawa Y, Chan BKK. 2014 Morphological and host specificity evolution in coral symbiont barnacles (Balanomorpha: Pyrgomatidae) inferred from a multi-locus phylogeny. Mol. Phylogenet. Evol. 77, 11–22. (doi:10.1016/j.ympev.2014.03.002) Achituv Y, Brickner I, Erez J. 1997 Stable carbon isotope ratios in Red Sea barnacles (Cirripedia) as an indicator of their food source. Mar. Biol. 130, 243–247. (doi:10.1007/s002270050244) Brickner I, Høeg JT. 2010 Antennular specialization in cyprids of coral associated barnacles. J. Exp. Mar.
5.
6.
7.
Biol. Ecol. 392, 115– 124. (doi:10.1016/j.jembe. 2010.04.015) Chan BKK, Høeg JT, Kado R. 2014 Chapter 22 Thoracica. In Atlas of Crustacean Larvae (eds JW Martin, J Olesen, JT Høeg), pp. 116–121. Baltimore, MD: Johns Hopkins University Press. Chen YY, Lin HC, Chan BKK. 2012 Description of a new species of coral-inhabiting barnacle, Darwiniella angularis sp. n. (Cirripedia, Pyrgomatidae) from Taiwan. ZooKeys 214, 43. (doi:10.3897/zookeys.214.3291) Aldred N, Clare AS. 2009 Mechanisms and principles underlying temporary adhesion, surface exploration and settlement site selection by barnacle cyprids: a short review. In Functional
surfaces in biology (ed. SN Gorb), pp. 43 –65. The Netherlands: Springer. 8. Høeg JT, Maruzzo D, Okano K, Glenner H, Chan BKK. 2012 Metamorphosis in Balanomorphan, Pedunculated, and parasitic barnacles: a video-based analysis. Integr. Comp. Biol. 52, 337–347. (doi: 10.1093/icb/ics053) 9. Maruzzo D, Aldred N, Clare AS, Høeg JT. 2012 Metamorphosis in the cirripede crustacean Balanus amphitrite. PLoS ONE 7, e37408. (doi:10.1371/ journal.pone.0037408) 10. Høeg JT, Glenner H, Shields J. 2005 Cirripedia Thoracica and Rhizocephala (barnacles). In Marine parasitology (ed. K Rohde), pp. 154–165. Wallingford, UK: CABI Publishing.
Biol. Lett. 12: 20160124
References
rsbl.royalsocietypublishing.org
How do coral barnacles start their life in their hosts? Jennie Chien Wen Liu1,2, Jens Thorvald Høeg3 and Benny K. K. Chan1 1
Research Cite this article: Liu JCW, Høeg JT, Chan BKK. 2016 How do coral barnacles start their life in their hosts? Biol. Lett. 12: 20160124. http://dx.doi.org/10.1098/rsbl.2016.0124
Received: 14 February 2016 Accepted: 23 May 2016
Subject Areas: developmental biology, ecology Keywords: coral-associated barnacle, cyprid, metamorphosis
Author for correspondence: Benny K. K. Chan e-mail: [email protected]
Electronic supplementary material is available at http://dx.doi.org/10.1098/rsbl.2016.0124 or via http://rsbl.royalsocietypublishing.org.
Biodiversity Research Center, Academia Sinica, Taipei 115, Taiwan, Republic of China Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei 106, Taiwan, Republic of China 3 Marine Biology Section, Department of Biology, University of Copenhagen, DK-2100, Copenhagen, Denmark 2
BKKC, 0000-0001-9479-024X Coral-associated invertebrates are the most significant contributors to the diversity of reef ecosystems, but no studies have examined how larvae manage to settle and grow in their coral hosts. Video recordings were used to document this process in the coral barnacle Darwiniella angularis associated with the coral Cyphastrea chalcidicum. Settlement and metamorphosis in feeding juveniles lasted 8–11 days and comprised six phases. The settling cyprid starts by poking its antennules into the tissue of the prospective host (I: probing stage). The coral releases digestive filaments for defence, but tolerating such attack the cyprid penetrates further (II: battling stage). Ecdysis is completed 2 days after settlement (III: carapace detachment). The barnacle becomes embedded deep in the coral tissue while completing metamorphosis between 4 and 6 days (IV: embedding stage), but reappears as a feeding juvenile 8– 11 days after settlement (V: emerging stage; VI: feeding stage). Cyprids preferably settle in areas between the coral polyps, where they have a much higher survival rate than on the polyp surfaces.
1. Introduction Numerous invertebrates are associated with coral reef ecosystems, and many of these are symbiotic forms that contribute significantly to the overall biodiversity of the reef fauna [1]. Coral-associated barnacles (Thoracica: Pyrgomatidae) are particularly common and occur in almost all coral species worldwide [2]. These suspension feeding barnacles have their shell plates embedded in the coral skeleton (figure 1a,b), and the symbiotic relationship extends to the nutritional level with the zooxanthellae in the coral. The barnacles ingest organic matter released from the zooxanthellae, while the zooxanthellae in turn absorb ammonium released by the barnacles [3]. Knowing how coral-associated animals, such as barnacles, manage to settle in their potentially hostile hosts armed with nematocyst defences, is essential for understanding symbiotic relationships, and the patterns and processes that govern reef biodiversity. There are around 80 species of coral barnacles with varying degrees of host specificity, but difficulties in larval rearing and maintenance of corals for experiments have until now prevented such studies. Coral barnacles disperse by nauplius larvae, resembling those in other cirripedes (figure 1c), but the terminal cyprid stage is remarkably specialized (figure 1d,e,g,h). Their antennules have spear-shaped attachment organs (figure 1d,e,g,h), a marked contrast to the bell-shaped ones found in cyprids of other barnacles [4,5] (figure 1f ). We have successfully cultured larvae of the coral barnacle Darwiniella angularis (figure 1a,b), which is fairly host specific, being reported 75% on Cyphastrea chalcidicum, and 25% on Astreopora sp. in Taiwan [6]. Using video microscopy, we recorded for the first time the entire
& 2016 The Author(s) Published by the Royal Society. All rights reserved.
barnacle shell
(a)
2
(e)
(c)
rsbl.royalsocietypublishing.org
orifice
100 AS3
(d)
coral polyps
(b) AS2 cirri PS3
20
100
(g)
(f)
tae
y or
sensory setae
se
ns
se
t organ attachmen
AS 4
AS4 attachment organ
5
10
(h)
antennules
ds
opo
100
ac hor
t
Figure 1. Morphological features of Darwiniella angularis. (a) Darwiniella angularis (white arrows) associated with the coral Cyphastrea chalcidicum in Taiwan. (b) Magnified view of D. angularis showing cirri extended. (c) Nauplius larva (stage II). (d) Lateral view of a live cyprid. (e) Magnified view of cypris antennule. AS, antennulary segment; PS, postaxial seta (2, 3, 4—segments 2, 3, 4). (f ) In the free-living barnacle Balanus perforatus the antennules have a bell-shaped attachment organ with a circular attachment disc. Antennulary segment four (AS4) carries various sensory setae for substratum location. (g) The spear-shaped attachment organ of D. angularis cyprids differs extensively from the bell-shaped one (f ) in free-living barnacles. (h) Cyprid of D. angularis under SEM. Scale bar in micrometres. sequence of events from cyprid settlement in the coral host, to the appearance of a feeding juvenile barnacle.
2. Material and methods
(white LED lamps). Seawater was changed daily. Nauplii were fed with mixed green algae and diatoms (Tetraselmis chui, Chaetoceros muelleri, Isochrysis lutea, Skeletonema costatum). The cultures were maintained at 268C, with more than 80% of the larvae successfully developed into cyprids.
(a) Collection and larval culture
(b) Settlement and metamorphosis
Corals C. chalcidicum with adult D. angularis were collected by scuba divers in northeast Taiwan between 7 and 10 m. Coral pieces hosting barnacles (approx. 30 cm2) were maintained in 1 l beakers, using filtered seawater under a 10 L : 14 D cycle
Cyprids were exposed to coral pieces (approx. 5 cm2) and kept in aquaria (30 15 6 cm) with weak aeration. Among 735 cyprids, 85 settled on the coral hosts. Unsettled cyprids kept swimming and eventually died. The settlement location of
Biol. Lett. 12: 20160124
orifice 500
1000
AS4
PS2
shell
DObs pffiffiffiffiffiffiffiffi , a=n
0:5
where DObs is the mean observed nearest-neighbour distance; a is the area under study; and n is the total number of individuals. Rn ranges from 0 to 2.15, representing distribution patterns from clustered (less than 1), over random (1), to regular (more than 1).
3. Results Darwiniella angularis requires 8 –10 days to develop through six naupliar stages into cyprids. Video recordings of the entire metamorphosis process, from settled cyprids to feeding settlers, can be viewed in the electronic supplementary materials. Settlement and metamorphosis lasted 8– 11 days and comprised 6 distinct phases (figure 2a– l). When exposed to the coral, the cyprid first poked its antennules into the soft tissue and then pulled its entire carapace down into the coral (Phase I, probing stage, figure 2a,d). Subsequently, a battle started between coral and settled larva. The cyprid used its spear-shaped antennules to penetrate further into the coral, which released digestive filaments in defence. All the settling cyprids were capable of tolerating these attacks, and continued to penetrate further into the tissue (Phase II, battling stage, figure 2b,e). The cypris carapace detached 2 days after settlement, thus completing ecdysis (Phase III, carapace detachment, figure 2c,f ). Following detachment, the barnacle entered deeper inside the coral tissue, where hidden from view, it used another 4–6 days to complete metamorphosis (Phase IV, embedding stage, figure 2g,j). Individuals dissected free during this hidden phase were developing both an apical mantle aperture and mineral shell plates (figure 2j ). About 10 days after settlement, the early juvenile emerged on the coral surface (Phase V, emerging stage, figure 2h,k). Feeding started the following day (Phase VI, feeding settler, figure 2i,l). During settlement, the cyprids appeared to discriminate between the various locations on the coral. Among the 85 cyprids that settled on the coral, 62.59 + 6.22% (mean + 1s.d., n ¼ 3) settled on tissue areas between the polyps, 28.54 + 4.82% on adult barnacle shells, but only 8.87 + 1.69% on the polyps themselves (x 2 (2, N ¼ 85) ¼ 34.73, p , 0.001). Cyprids attached to polyps were all expelled during the battling to the embedding stage, subsequently dying, while more than 95% of those that settled in spaces between polyps survived into feeding barnacles (x 2 (2, N ¼ 85) ¼ 51.73, p , 0.001) Cyprids preferred to settle on corals hosting conspecifics. In the choice experiment, 23 of the 26 settled cyprids (88.4%)
4. Discussion Symbiosis is an increasingly important research area, especially in complex coral reef systems. This is the first study documenting how the larvae of coral-associated barnacles settle in live coral tissue and start their symbiotic life. The ability of the cyprids to tolerate the nematocyst defences of the host is apparently a key to their survival and subsequent embedding deep into the coral tissue, where they can complete metamorphosis unimpeded. An additional factor that aids the survival of the cyprids is their ability to preferentially settle away from the actual polyps (see results of x 2 settlement test). A high level of settlement site discrimination is known from other barnacles [7]. The preference for gregarious settlement in barnacles seems to also favour and enhance mating success. We therefore preliminarily conclude that the distributional pattern of barnacles on their coral host is a function of larval settlement site selection, conspecific settlement cues and post-settlement mortality. In trials where we exposed D. angularis cyprids to the nonhost coral Lepastrea transversa, the larvae only swam in the water column but never ‘walked’ on the coral surfaces. They even fled rapidly away if accidentally contacting the non-host coral. The ability of D. angularis cyprids to discriminate between coral species and select specific sites on the proper host highlights the importance of their antennular sensory organs for survival during settlement. Surprisingly, examination of scanning electron micrographs (SEM) of both D. angularis and other coral barnacles revealed that the cypris sensory setae are similar in structure and number to those found in more conventional species (figure 1g) [4]. Hence, settlement site specificity probably depends on neurological processes rather than on morphological details in the setae. It would therefore be exciting to combine more refined settlement experiments with electrophysiological recordings from the larvae. Coral barnacles take longer to complete metamorphosis than any other thoracican barnacles studied (figure 2m,n). For example, the metamorphosis from initial attachment of the cyprid to shedding of the carapace lasted less than 1 day in the intertidal free-living Amphibalanus amphitrite, 1.5 days in the subtidal free-living Megabalanus rosa (figure 2m) and 3.5 days in the neustonic species Lepas sp. These three species commenced feeding shortly after carapace shedding [8,9]. By contrast, D. angularis completed ecdysis in 2 days but required another 8 days to become a feeding juvenile (figure 2n). This prolonged metamorphosis is most likely owing to the time used for battling with the coral host soon after settlement, and the need to complete the rest of the metamorphosis process hidden deep in the coral tissue. These events can be easily compared to parasitic barnacles (Rhizocephala), which must also combat host defences during settlement, thereafter needing an internal parasitic phase to establish full control of their host before they re-emerge [10]. Interestingly, both coral barnacles (80þ species) and rhizocephalans (250þ species) are major components to the overall cirripede diversity (ca 1200 species). Apparently,
3
Biol. Lett. 12: 20160124
Rn ¼
attached on coral pieces already hosting D. angularis (x 2 (1, N ¼ 26) ¼ 15.38, p , 0.001). The nearest-neighbour analysis showed that all three juvenile populations collected in the wild (Rn1,2,3) had a clustered distribution, with the barnacles concentrated in spaces between polyps (Rn1 ¼ 0.383, Rn2 ¼ 0.615, Rn3 ¼ 0.667).
rsbl.royalsocietypublishing.org
each cyprid (i.e. on polyps, between polyps, and on shells of live D. angularis on corals) was recorded using a video camera (Lumix GH1) under a stereomicroscope. A x 2-test was used to examine whether the numbers of settlers and post-settlement juveniles (numbers pooled from three pieces of coral in each of three aquaria, cyprids numbers in the three aquaria: 103, 291, 177) were similar in the three indicated locations above. In a separate aquaria, 164 cyprids were exposed to two coral pieces with and without D. angularis. An additional x 2-test was used to examine whether the settlers equally preferred the two pieces. The distribution pattern of D. angularis in the wild was examined by the nearest-neighbour analysis of 44 juveniles (shell diameter , 1 mm) located on the surface of three coral pieces. The nearest-neighbour index (Rn) is based on the formula:
4 phase II
(b)
probing stage
phase III
(c)
battling stage
carapace detachment separated epidermis
ca
rap
ac e
cyprid
100
100
digestive filaments
nt ce
me
(e)
(f)
1
10
cypris thorax
2 days primordial scutum
primordial tergum
cypr i
d apical
mantle coral antennule
100
phase IV
(g)
(h) phase V (lateral view)
embedding stage
100
(i) phase VI
emerging stage
embedding individual
feeding settler fused scutum and tergum
(top view) cirri of the feeding settler
digestive filaments
scutum and tergum underneath the coral tissue
shell
100
3 days
100
100
10 days
11 days apical
(j)
(k)
shell
cirri
base
Megabalanus rosa
basal
base
l
primordial scutum and tergum
(l)
shell
shel
(m)
base
100
100 fused scutum and tergum
100
100
ecdysis
nauplius eye compound eye
separated inclined from antennule epidermis the substratum antennular cuticle and compound eyes
Darwiniella angularis
(n)
sis
digestive filaments
co
cirri
ec dy
emerging individual
separated epidermis
ral
0h
7h
1.5 days
2 days
4 days
10 days
Figure 2. Metamorphosis in Darwiniella angularis. (a – l) Light and scanning electron microscopy of phases in settlement and metamorphosis. Phase I: probing stage (a,d). The cyprid pokes its spear-shaped antennules into the soft coral tissue. Phase II: battling stage (b,e). The coral hosts release digestive filaments in defence; 28 h post-settlement the cyprid epidermis has separated from the carapace cuticle and its tissues start to concentrate (b). Cyprid removed from the coral tissue (e), showing the tiny mass of cement secreted by the antennule (circle inset) for anchorage. Phase III: carapace detachment (c,f ). Two days postsettlement, cyprid carapace and thorax are shed, revealing body of the metamorphosing barnacle with primordial shell plates. Phase IV: embedding stage (g,j). The metamorphosing cyprid remains totally embedded in coral tissue for several days. Removing such individuals ( j ) reveals incipient formation of shell plates. Phase V: emerging stage (h,k). The metamorphosed settler re-emerges from the coral tissue. Phase VI: feeding settler (i,l). One day later, the juvenile barnacle starts cirral suspension feeding. Scale bar in micrometres. (m,n) Comparison of metamorphosis in Megabalanus rosa [8] and D. angularis. Although overall comparable, the metamorphosis differs in both specific events and timing. (m) In M. rosa, metamorphosis is completed in 1.5 days. (n) In D. angularis, the cyprid first pokes its antennules into the soft coral tissue. Metamorphosis proceeds while completely hidden in the coral tissue and is not completed until 10 days after attachment. (Online version in colour.)
Biol. Lett. 12: 20160124
18 h
coral
(d )
100
rsbl.royalsocietypublishing.org
phase I
(a)
solving the problem of settlement on such hostile hosts entails opportunities for evolutionary success.
final version and agree to be accountable for the work in the manuscript.
5
Ethics. This research complies with laws of the Taiwan, Republic of
Competing interests. No competing interests. Funding. Supported by the Senior Investigator Award to B.K.K.C. and
rsbl.royalsocietypublishing.org
China, regarding research. There is no need for ethical approval for experiments involving barnacle larvae in Taiwan. Data accessibility. Settlement experiment data and video of metamorphosis are provided in the electronic supplementary material.
Authors’ contributions. B.K.K.C. conceived the study; J.C.W.L., B.K.K.C. and J.T.H. acquired and analysed the data and all authors contributed to the preparation of the manuscript. All authors approved the
the Carlsberg Foundation (2013_01_0130) to J.T.H.
Acknowledgements. We thank Yao-Fung Tsao, Shu-Mei Lin, Wei-Peng Hsieh and Pei-Chen Tsai (Academia Sinica) for help with larval culture. Thanks to Brett Gonzales for editing the English of the manuscript.
1.
2.
3.
4.
Blackall LL, Wilson B, van Oppen MJH. 2015 Coral – the world’s most diverse symbiotic ecosystem. Mol. Ecol. 24, 5330–5347. (doi:10.1111/mec.13400) Tsang LM, Chu KH, Nozawa Y, Chan BKK. 2014 Morphological and host specificity evolution in coral symbiont barnacles (Balanomorpha: Pyrgomatidae) inferred from a multi-locus phylogeny. Mol. Phylogenet. Evol. 77, 11–22. (doi:10.1016/j.ympev.2014.03.002) Achituv Y, Brickner I, Erez J. 1997 Stable carbon isotope ratios in Red Sea barnacles (Cirripedia) as an indicator of their food source. Mar. Biol. 130, 243–247. (doi:10.1007/s002270050244) Brickner I, Høeg JT. 2010 Antennular specialization in cyprids of coral associated barnacles. J. Exp. Mar.
5.
6.
7.
Biol. Ecol. 392, 115– 124. (doi:10.1016/j.jembe. 2010.04.015) Chan BKK, Høeg JT, Kado R. 2014 Chapter 22 Thoracica. In Atlas of Crustacean Larvae (eds JW Martin, J Olesen, JT Høeg), pp. 116–121. Baltimore, MD: Johns Hopkins University Press. Chen YY, Lin HC, Chan BKK. 2012 Description of a new species of coral-inhabiting barnacle, Darwiniella angularis sp. n. (Cirripedia, Pyrgomatidae) from Taiwan. ZooKeys 214, 43. (doi:10.3897/zookeys.214.3291) Aldred N, Clare AS. 2009 Mechanisms and principles underlying temporary adhesion, surface exploration and settlement site selection by barnacle cyprids: a short review. In Functional
surfaces in biology (ed. SN Gorb), pp. 43 –65. The Netherlands: Springer. 8. Høeg JT, Maruzzo D, Okano K, Glenner H, Chan BKK. 2012 Metamorphosis in Balanomorphan, Pedunculated, and parasitic barnacles: a video-based analysis. Integr. Comp. Biol. 52, 337–347. (doi: 10.1093/icb/ics053) 9. Maruzzo D, Aldred N, Clare AS, Høeg JT. 2012 Metamorphosis in the cirripede crustacean Balanus amphitrite. PLoS ONE 7, e37408. (doi:10.1371/ journal.pone.0037408) 10. Høeg JT, Glenner H, Shields J. 2005 Cirripedia Thoracica and Rhizocephala (barnacles). In Marine parasitology (ed. K Rohde), pp. 154–165. Wallingford, UK: CABI Publishing.
Biol. Lett. 12: 20160124
References
