Coral Reefs DOI 10.1007/s00338-014-1153-2

REPORT

Crown-of-thorns starfish predation and physical injuries promote brown band disease on corals Sefano M. Katz • F. Joseph Pollock David G. Bourne • Bette L. Willis

•

Received: 12 October 2013 / Accepted: 4 April 2014 Ó Springer-Verlag Berlin Heidelberg 2014

Abstract Brown band (BrB) disease manifests on corals as a ciliate-dominated lesion that typically progresses rapidly causing extensive mortality, but it is unclear whether the dominant ciliate Porpostoma guamense is a primary or an opportunistic pathogen, the latter taking advantage of compromised coral tissue or depressed host resistance. In this study, manipulative aquarium-based experiments were used to investigate the role of P. guamense as a pathogen when inoculated onto fragments of the coral Acropora hyacinthus that were either healthy, preyed on by Acanthaster planci (crown-of-thorns starfish; COTS), or experimentally injured. Following ciliate inoculation, BrB lesions developed on all of COTS-predated fragments (n = 9 fragments) and progressed up to 4.6 ± 0.3 cm d-1, resulting in *70 % of coral tissue loss after 4 d. Similarly, BrB lesions developed rapidly on

experimentally injured corals and *38 % of coral tissue area was lost 60 h after inoculation. In contrast, no BrB lesions were observed on healthy corals following experimental inoculations. A choice experiment demonstrated that ciliates are strongly attracted to physically injured corals, with over 55 % of inoculated ciliates migrating to injured corals and forming distinct lesions, whereas ciliates did not migrate to healthy corals. Our results indicate that ciliates characteristic of BrB disease are opportunistic pathogens that rapidly migrate to and colonise compromised coral tissue, leading to rapid coral mortality, particularly following predation or injury. Predicted increases in tropical storms, cyclones, and COTS outbreaks are likely to increase the incidence of coral injury in the near future, promoting BrB disease and further contributing to declines in coral cover.

Communicated by Biology Editor Prof. Brian Helmuth

Keywords Coral disease  Brown band disease  Porpostoma guamense  Opportunistic pathogen  Crown-of-thorns starfish  Injury

S. M. Katz (&)  F. J. Pollock  D. G. Bourne Australian Institute of Marine Science, PMB 3, Townsville, QLD 4810, Australia e-mail: [email protected] F. J. Pollock e-mail: [email protected] D. G. Bourne e-mail: [email protected] S. M. Katz  F. J. Pollock  B. L. Willis ARC Centre of Excellence for Coral Reef Studies, School of Marine and Tropical Biology, James Cook University, Townsville, QLD 4811, Australia e-mail: [email protected] S. M. Katz  F. J. Pollock AIMS@JCU, James Cook University, Townsville, QLD 4811, Australia

Introduction Physical injuries provide open wounds on surface tissue for pathogens to invade and establish disease. Thus, on coral reefs, physical disturbances and predation are potential mechanisms enabling the development of coral diseases. The widespread impact of predation by Acanthaster planci (crown-of-thorns starfish; COTS) on coral communities has been known for over four decades (Endean and Stablum 1973; Kenchington 1978). COTS outbreaks may also regulate reef community structure through selective feeding preferences (Goreau et al. 1972), which can severely reduce rates of framework deposition and net reef accretion

123

Coral Reefs

(Goreau 1964). Recent studies using broad-scale surveys have recognised both COTS predation and cyclones as two of the dominant drivers of coral reef decline on the Great Barrier Reef (GBR) (Osborne et al. 2011; De’ath et al. 2012), but the role of these disturbances in disease initiation has received comparatively less attention. Cyclones and hurricanes impact corals through breakage, scarring, and dislodgement from hydro-mechanical forces, waterborne debris, and lowered osmotic pressure due to changing salinity (Goreau 1964; Done 1992). While storms damage reefs primarily through mechanical forces, COTS impact corals through chemical injury, digesting tissues to expose bare skeleton (Ormond et al. 1976). Even when coral mortality is not complete, COTS predation reduces the growth rate and fitness of corals through the removal of live tissue (reviewed in Rotjan and Lewis 2008). With both tropical cyclones and COTS outbreaks expected to increase in frequency and intensity in coming years (Fabricius et al. 2010; Knutson et al. 2010; Emanuel 2013), effective active and passive reef management approaches (e.g., Birkeland and Lucas 1990; GBRMPA 2004; Waterhouse et al. 2010) will increasingly depend upon a complete understanding of factors influencing coral health following physical damage and predation. Coral diseases can play an important role in shaping coral reef communities (Willis et al. 2004; Harvell et al. 2007), and disease impacts are predicted to increase with projected climate change (Hayes and Goreau 1998; Harvell et al. 2002; Willis et al. 2004). Field studies have shown that coral injury can lead to the initiation of coral disease (Page and Willis 2008) and have identified a strong association between physical damage and high coral disease prevalence (Lamb and Willis 2011). Although environmental stress and physical injury are predicted to increase the susceptibility of corals to infectious diseases, empirical evidence demonstrating these links is currently limited (Bruno et al. 2003; reviewed in Sokolow 2009). Brown band disease (BrB) is a virulent coral disease which manifests as a dense aggregation of ciliates that advances over the coral surface exposing underlying white skeleton (Willis et al. 2004). BrB was first identified in 2003 on northern and southern reefs of the GBR, Australia (Willis et al. 2004), but the disease has since been reported from Western Australian reefs (FJ Pollock, personal observation) and sites throughout the Indo-Pacific, including Indonesia (Nugues and Bak 2009), the Philippines (Raymundo et al. 2009), Guam (Myers and Raymundo 2009), Palau (Page et al. 2009), China (Qiu et al. 2010), Japan (Weil et al. 2012), India (Sukumaran et al. 2011), and East Africa (Harvell et al. 2007). The dominant BrB ciliate was initially identified from the coral Acropora muricata on the GBR as a protozoan scuticociliate, class Oligohymenophorea (Bourne et al. 2008). Later

123

morphological work on ciliates associated with corals in the genus Acropora in Guam identified the dominant ciliate as a member of the genus Porpostoma, naming it Porpostoma guamense (Lobban et al. 2011). Subsequent phylogenetic work clustered the dominant ciliate from A. muricata on the GBR in the genus Philaster, suggesting that the ciliate should be renamed as Philaster guamense (Sweet and Bythell 2012). Furthermore, Sweet and Bythell (2012) suggested that the dominant ciliate plays a secondary role in BrB pathogenesis, and hypothesised that a ciliate with a lower population density plays a more invasive role in BrB pathogenesis. Conflicting results from studies of BrB and uncertainty concerning the role of ciliates in BrB pathogenesis (Bourne et al. 2008; Lobban et al. 2011; Sweet and Bythell 2012; Nicolet et al. 2013) highlight the need for further detailed studies of coral susceptibility to infestations by BrB-associated ciliates. Accordingly, the objectives of this study were (1) to provide an understanding of the identity of the dominant ciliates associated with BrB lesions and their role in initiation and development of the disease and (2) to investigate the role of COTS predation injuries and physical injuries in the development of BrB.

Materials and methods Study site and coral sample collection Colonies of the coral Acropora hyacinthus were collected from sheltered fringing reefs (*2–5 m depth) at Lizard Island (14°400 S, 145°270 E) in the northern sector of the GBR in June and July 2012. All coral colonies were fragmented (*10 cm 9 10 cm) prior to experimentation. Healthy fragments were collected from colonies with no visible injuries, lesions, or associated corallivores (e.g., A. planci, Drupella spp., and Coralliophila spp.). COTSpredated colonies were identified by wide zones of recently exposed, white skeleton with sloughing tissue and thick layers of mucus adjacent to live tissue. The COTS-injured colonies selected displayed fresh scars, but substantial portions of colonies remained covered with live tissue. COTS were found in close proximity to all COTS-injured corals sampled. BrB-infected colonies were identified by a distinct, macroscopically visible brown band that bordered live tissue at the advancing lesion front and exposed white skeleton behind (Willis et al. 2004). Fragments of BrB-infected colonies were subsequently viewed under a dissecting microscope (Wild PhotoMakroskop M400, 209 magnification) and all possessed the characteristic BrB ciliate mat described previously (Bourne et al. 2008; Lobban et al. 2011; Sweet and Bythell 2012).

Coral Reefs

BrB ciliate collection BrB-infected fragments of A. hyacinthus were placed in beakers containing 1 lm filtered seawater (FSW), sealed and left at room temperature for *3 h, resulting in a dense mass of ciliates swimming in the water and aggregating at the surface. Water containing the ciliate mass was gently poured onto a 30-lm nylon mesh and gently washed from the mesh with FSW. This procedure was repeated twice to further wash the ciliates collected. Five 100 ll aliquots of washed ciliates for the final inoculum preparation were counted under a dissecting microscope (Wild PhotoMakroskop M400, 209 magnification), and by extrapolation, the total ciliate densities were determined for subsequent inoculation experiments. Two preparations of ciliates were created for experimental inoculation, one for the COTS experiment and one for the injury experiment. Ciliates harvested from coral fragments were used within 24 h of collection.

Susceptibility of COTS-predated and injured corals to BrB Healthy fragments of A. hyacinthus were harvested using bone cutters and placed in flow-through aquaria for 10 d. This recovery period enabled corals to heal potential entry wounds created in the initial fragmentation, and to acclimate to aquarium conditions prior to the start of experiments. Healing was observed on all coral fragments when inspected under a dissecting microscope (Wild PhotoMakroskop M400, 209 magnification). In addition, none of the characteristic BrB ciliates (i.e., P. guamense) were observed on healthy coral fragments. COTS-predated coral fragments were freshly collected and inspected for the occurrence of characteristic BrB ciliates at 10 random locations along lesion fronts on scarred fragments. For the injury treatment, healthy fragments of A. hyacinthus corals were fragmented to approximately 10 cm 9 10 cm, resulting in 165 ± 23 branches. Artificial lesions were then inflicted to simulate physical damage caused by storm damage or reef users (e.g., reef walkers, breakage from snorkelers and divers). On each fragment, five branch tips (*5 mm of branch length) were broken using sterile bone cutters. Additionally, a 5-mm lesion was artificially inflicted on one side of five replicate branches, and five 2 9 5 mm lesions were artificially inflicted at the base of each fragment using a sterile surgical scalpel. The initial fragmentation process resulted in approximately five additional basal wounds per fragment. No P. guamense ciliates were observed on any fragments with COTS

scars or artificially induced lesions at the commencement of the experiment. Coral fragments were randomly placed on plastic racks in flow-through aquaria with seawater filtered to 1 lm (volume = 60 l, flow = 60 l h-1), a constant air supply, and under-ambient light and temperature at Lizard Island Research Station (mean: 24.6 ± 0.28 °C (SD); maximum = 25.88 °C; minimum = 24.05 °C). For the COTS inoculation experiment, four experimental treatments were established including (1) healthy coral fragments, (2) healthy coral fragments inoculated with BrB ciliates, (3) coral fragments with COTS scars, and (4) coral fragments with COTS scars and inoculated with BrB ciliates. Each treatment comprised three replicate aquaria, each containing three coral fragments (n = 9 fragments per treatment). For each of the two inoculated treatments, 40 ml of the ciliate preparation, containing 2.86 ± 0.16 9 104 ciliates (mean ± SEM), was added to each aquarium using a 10-ml serological pipette to create densities of *0.5 ciliates ml-1. For the artificially injured corals, the same experimental design was employed to produce four treatments: (1) healthy coral fragments, (2) healthy coral fragments inoculated with BrB ciliates, (3) injured coral fragments, and (4) injured coral fragments inoculated with BrB ciliates. Ciliates were inoculated into the treatment aquaria at a concentration of 2.81 ± 0.07 9 104 ciliates per aquarium, creating a density of *0.5 ciliates ml-1. Water flow was stopped for 12 h following ciliate inoculation in all experiments. Ten replicate areas of 25 mm2 (determined by an eyepiece graticule calibrated with a stage micrometer) were inspected on each fragment every 24 h under a dissecting microscope (Wild PhotoMakroskop M400, 209 magnification) to quantify ciliate numbers. Coral fragments were handled gently to minimise movement and disturbance to ciliates. Areas were located randomly on healthy fragments and along lesion fronts of COTS-scarred and artificially injured corals. Corals were photographed every 12 h for the duration of the experiment. Linear progression rates (LPR) of lesions were measured using the image analysis software package ImageJ 1.45 s (Abra`moff et al. 2004). LPRs were determined by overlaying photos of the same coral fragment at different time points (using topographical landmarks), randomly selecting five points along a lesion front, and measuring the distance between points on the lesion front at time X and the corresponding points on the lesion front at time X ? 1. Areas of coral tissue loss were measured using ImageJ 1.45 s (Abra`moff et al. 2004). In order to calculate the relative percentage area of tissue loss, the sum of the areas of denuded skeleton on each fragment was divided by the area of the whole fragment.

123

Coral Reefs

Ciliate choice experiment A choice experiment was performed to assess the habitat preference of P. guamense ciliates for healthy versus injured corals. Healthy fragments of A. hyacinthus were fragmented to approximately 4 cm 9 4 cm, resulting in 49 ± 3 branches, and left to heal and acclimate in the aquaria for 3 d. Artificial injuries were inflicted on five fragments. On each fragment, three branch tips (*5 mm of branch length) were broken using sterile bone cutters. A 5-mm lesion was artificially inflicted on one side of three replicate branches, and three 2 mm 9 5 mm lesions were artificially inflicted at the base of each fragment using a sterile surgical scalpel. To evaluate active migration of the ciliates, coral fragments were randomly assigned to five sets of paired 1–l plastic containers, so that each pair had one healthy and one artificially injured coral fragment. Each container was supplied with constant air set at a minimal flow rate in order to minimise turbulence that might affect the ciliates’ movement. The paired containers were connected using 50-ml Falcon tubes (SARSTEDT) inserted horizontally near the tops of containers, creating a permanent water bridge between each pair of containers, through which the ciliates could move and their movement could be observed. A small inoculation hole was created in the middle of the Falcon tubes equidistant from both fragments. Each pair of containers (n = 5) was inoculated with 0.49 ± 0.01 9 103 ciliates, creating a density of *0.5 ciliates ml-1 across the two containers. Movement of ciliates was recorded through unaided eye observations during the first 15 min following inoculation by scoring the number of ciliates that had migrated out of the water bridge and were moving towards a healthy or injured coral. Corals were inspected in their containers every 12 h for the duration of the experiment (36 h) under a dissecting microscope (Wild PhotoMakroskop M400, 209 magnification) to quantify and compare the number of ciliates on healthy fragments with the number of ciliates on lesions of injured corals. The experiment was terminated after 36 h to avoid a bias due to ciliate reproduction, which was previously observed to occur after *48 h. Corals were photographed every 12 h during the experiment. Percentage area of coral tissue loss was measured using ImageJ 1.45 s (Abra`moff et al. 2004), as described above. Ciliate observations During all experiments, ciliate behaviour (e.g., movement, reproduction), location, and morphological appearance (e.g., encystment, conjugation, binary fission) were monitored macroscopically and microscopically (Wild PhotoMakroskop M400, 209 magnification) every 12 or 24 h. For all microscopic observations, coral fragments were

123

handled with care to minimise movement and disturbance to the ciliates.

DNA extraction and PCR amplification To genetically identify the ciliates associated with the disease lesion, four fragments of infected corals from the COTS experiment containing the BrB lesion were snapfrozen in liquid nitrogen and subsequently stored at -80 °C. Total genomic DNA was extracted using a modified Phenol–Chloroform extraction protocol (Barengo et al. 2004). Nubbins 2 cm3 in size, each containing a lesion front, were vortexed with 500 ll of Milli-Q water. Next, 0.5 ml of lysis buffer (0.75 M Sucrose, 40 mM EDTA, 50 mM Tris, pH 8.3) was added and mixed before addition of 75 ll (100 mg ml-1) of lysosyme. Samples were incubated in a water bath at 37 °C for 1 h, frozen in liquid nitrogen, and thawed in a 65 °C water bath three times. Samples were cooled to room temperature, and 100 ll 10 % SDS and 200 ll Proteinase K (2 mg ml-1) was added before samples were incubated at 37 °C for 1 h. Samples were freezethawed three times before adding an equal volume of Phenol:Chloroform:Isoamylalcohol (25:24:1) and incubated shaking (200 RPM) at 37 °C for 10 min. Samples were centrifuged for 10 min at 14,000 rpm, and the supernatant was transferred to a fresh tube before adding an equal volume of Chloroform:Isoamylalcohol (24:1). The supernatant was transferred to a fresh tube after centrifugation for 5 min at 14,000 rpm and incubated on ice for 3 min. NaAc (50 ll, 3 M) was added, mixed, and then an equal volume of Isopropanol was added and mixed, following which samples were incubated at room temperature for 7 min. One ml of 70 % EtOH was added after centrifugation for 30 s at 14,000 rpm, the supernatant discarded, and the pellet airdried for 15 min after centrifugation for 5 min at 14,000 rpm. Pellets were resuspended in 30 ll Milli-Q water and stored at -20 °C until further use. Ciliate 18S rRNA genes were amplified using a nested PCR approach. The first amplification was carried out in a final volume of 50 ll containing 1 ll of DNA template, 1.5U Bio-X-Act Polymease (Bioline), 19 Reaction Buffer (Bioline), 4 mM MgCl2, 250 lM of dNTP mix (Bioline), 2.5 ll Hi-Spec additive (Bioline), and 10 lM of each primer. The primers used were Cil-F (50 -TGGTAGTGTATTGGACWACCA-30 ) and an equimolar mixture of three reverse primers, Cil-RI (50 -TCTGATCGTCTTTGATCCCTTA-30 ), Cil-RII (50 -TC TRATCGTCTTTGATCCCCTA-30 ), and Cil-RIII (50 -TCT GATTGTCTTTGATCCCCTA-30 ) (Lara et al. 2007). PCR conditions consisted of an initial denaturation at 94 °C for 5 min, followed by 35 cycles of 94 °C for 1 min, 65–55 °C (Touch Down of 10 cycles) for 1 min, 72 °C for 1 min, and a final elongation of 10 min at 72 °C. The second reaction

Coral Reefs

was carried out to amplify a 1:100 dilution from the first PCR using the forward primer Cil-F carrying a 36-bp GC clamp and the reverse primer CilDGGE-r (50 -TGAAAAC ATCCTTGGCAAATG-30 ) (Jousset et al. 2010). The rest of the mix was as described above. Amplification conditions for the second run consisted of an initial denaturation at 94 °C for 5 min, followed by 26 cycles of 94 °C for 1 min, 52 °C for 1 min, and 72 °C for 1 min with a final elongation of 10 min at 72 °C to reduce double bands in the DGGE patterns (Janse et al. 2004). All reactions were performed using Kyratec PCR SuperCycler SC200. PCR products were verified by agarose gel electrophoresis [1 % (w/v) agarose] with GelRed (Biotium) staining and visualised using a Vilber Lourmat ChemiSmart 3000 camera. Denaturing gradient gel electrophoresis (DGGE) Using the amplification products above, three replicates for each of the four samples were used to produce a DGGE profile using the INGENYphorU-system. PCR products were resolved on 6 % (w/v) polyacrylamide gel that contained a 30–37 % formamide (denaturant) gradient for 18 h at 60 °C and a constant voltage of 50 V. Gels were stained with SYBR Gold (Invitrogen). To minimise contamination, plugs from representative bands were excised using a 20to 200-ll micropipette tip with cut ends to identify the dominant DGGE bands across samples. Tips with plugs were placed in microtubes with a PCR master-mix and amplified using the second amplification schedule described above, with the forward primer Cil-F and reverse primer CilDGGE-r. PCR products were verified as described above and were sequenced (Macrogen Inc., Korea). Sequences were aligned in CLUSTALW (Tamura et al. 2007) using an IUB cost matrix with a gap open penalty of 15 and a gap extend penalty of 7. The bootstrap consensus tree (1,0009 re-sampling) was constructed in MEGA 4 using the Tamura 3-Parameter distance model with Toxoplasma gondii (X75429) as the out-group. Statistical analyses All statistical analyses were performed using Statistica (version 10, Statsoft, Tulsa, USA) and R: Statistical Computing Software (R Development Core Team 2013). Healthy control fragments did not display any signs of lesion formation, mortality, or P. guamense ciliate colonisation and were therefore excluded from subsequent analyses. Lesion progression rates (LPR) and areas of tissue loss in the COTS experiment were analysed with a repeated measures ANOVA (prior to analysis, no significant tank effect was detected using nested ANOVA). Data satisfied all assumptions of the repeated measures ANOVA

after they were log-transformed (log(x ? 1)), and a Tukey’s HSD post hoc test was subsequently performed. The relationship between LPR and ciliate counts was tested using Spearman’s rank order correlations test. COTS-predated fragments that were not inoculated did not display any P. guamense ciliates and were therefore excluded from the subsequent analysis. The mean numbers of P. guamense ciliates on COTS-predated fragments that were inoculated were compared at different time points using repeated measures ANOVA. Areas of tissue loss on injured fragments due to the inoculation of ciliates could not be compared using parametric statistics because data failed to meet assumptions of normality and homogeneity of variances. Therefore, areas of tissue loss were compared using Kruskal–Wallis ANOVA followed by multiple comparisons (no significant tank effect was observed using Kruskal–Wallis ANOVA). Injured fragments that were not inoculated did not display any signs of ciliate colonisation and were therefore excluded from subsequent analyses. The numbers of P. guamense ciliates on inoculated injured fragments were compared at different time points using repeated measures ANOVA. Ciliate density data derived from the choice experiment data were not normally distributed and their variance not homogenous; therefore, the Mann–Whitney U test (Mann and Whitney 1947) was employed.

Results Susceptibility of COTS-predated corals to BrB Inoculation of BrB ciliates into aquaria containing coral fragments with fresh COTS predation scars resulted in the rapid development of characteristic BrB disease lesions, followed by rapid coral tissue loss. Within 84 h following inoculation of a ciliate preparation containing 2.86 ± 0.16 9 104 cells, COTS-predated corals suffered significant (70.7 ± 7.2 %) tissue loss (Fig. 1). Mean percentage tissue loss on COTS-predated fragments was nearly ten-fold higher in ciliate-inoculated than in non-inoculated treatments (8.5 ± 5.6 % after 84 h) (F1,3 = 6.92, p \ 0.001). Linear progression rates of BrB lesions on inoculated COTS-predated corals increased more than sixfold over the course of the experiment, from 0.69 ± 0.23 cm d-1 in the first 24 h to 3.3 ± 0.8 cm d-1 at 48 h and 4.6 ± 0.3 cm d-1 at 72 h (F1,3 = 6.92, p \ 0.001; Fig. 1). Linear progression rates of lesions were more than 12-fold slower on non-inoculated COTS-predated corals (0.36 ± 0.36 cm d-1 at 72 h; Fig. 1). The density of ciliates at lesion fronts (i.e., ciliates mm-2) in the inoculated COTS treatment increased more than fourfold over the course of the experiment, from 0.9 ± 0.2 ciliates at 24 h to 2.9 ± 0.6 ciliates at 48 h and

123

Coral Reefs

4.2 ± 0.5 ciliates at 72 h (F2,16 = 23.63, p \ 0.0001). No significant differences were detected in linear progression rates and area of tissue loss among tanks within experimental treatments (F1,4 = 1.1, p [ 0.41). Linear progression rate of lesions was significantly correlated with the density of ciliates for inoculated, COTS-predated corals throughout the experiment (R2 = 0.89, p \ 0.05; Fig. 2). The experiment was terminated after 84 h due to total mortality of fragments from actively progressing lesions. No signs of BrB lesions were observed on any healthy fragments, either in treatments with or without ciliate inoculation, throughout the experiment. In addition, no distinctive BrB ciliates were observed on COTS-predated fragments in the absence of ciliate addition when examined under a dissecting microscope. However, small aggregations of other types of ciliates were observed behind lesion fronts, which were comprised of sloughed tissue and mucus originating from the COTS.

Area of tissue loss (%)

70 60

Tissue loss - Inoculated Tissue loss - Not inoculated LPR - Inoculated LPR - Not inoculated

5 4

50 3

40 30

2

20 1

10 0

0 0

12

24

36

48

60

72

Linear progression rate (cm/day)

80

84

Time after inoculation (hours)

7

Artificially injured corals inoculated with a preparation containing 2.81 ± 0.07 9 104 ciliates resulted in signs of BrB disease within 12 h, with ciliates aggregating on secreted coral mucus surrounding the lesions, and subsequently forming progressive lesion fronts. The average area of tissue loss differed significantly before and after ciliate inoculation (H2,27 = 19.14, p \ 0.01; Fig. 3). At 24 h, percentage area of tissue loss was 9.42 ± 3.91 %, whereas 48 h post addition of the ciliate preparation, percentage area of tissue loss increased threefold to 29.88 ± 4.38 %. By 60 h post addition of the ciliate preparation, the average area of tissue loss reached 37.88 ± 3.49 %. For injured corals that were not inoculated with the ciliate preparation, no signs of BrB lesions were observed throughout the experiment and ciliates were not seen at the lesion interface. Furthermore, these corals partially healed, significantly reducing the area of tissue loss caused by the artificially inflicted wounds by more than twofold, from 2.03 ± 0.2 % at the start of the experiment to 0.8 ± 0.25 % by 60 h after injury (H2,27 = 9.63, p \ 0.01). No significant differences were detected in area of tissue loss among tanks within experimental treatments (H2,9 = 1.07, p [ 0.59). The number of ciliates at the lesion front increased throughout the experiment and was correlated with higher areas of tissue loss on the coral fragments. For example, the number of ciliates increased significantly on infected corals in the period between 24 h and 48 h (1.6 ± 0.3 and 2.7 ± 0.1 ciliates per mm2, respectively; F1,8 = 72.96, p \ 0.0001). Ciliate choice experiment When a preparation of 0.49 ± 0.01 9 103 BrB ciliates was inoculated into a water bridge between healthy and

R2 = 0.89, p < 0.05

45

6

Area of tissue loss (%)

Linear progression rate (cm/day)

Fig. 1 Comparison of mean percentage area of tissue loss (± SEM) (columns), and mean linear progression rates (LPR) (±SEM) of lesions (dots) on inoculated (grey) and not inoculated (white) COTSpredated coral fragments (n = 9 fragments per treatment). Healthy corals are not presented as all values equal zero

Susceptibility of injured corals to BrB

5 4 3 2 1

40

Tissue loss - Inoculated Tissue loss - Not inoculated

35 30 25 20 15 10 5 0

0 0

1

2

3

4

5

6

Ciliate counts (ciliates/mm2)

Fig. 2 Relationship between linear progression rates of lesions and ciliate densities (recorded from 25 mm2 areas of lesion fronts) on coral fragments that had been collected with COTS scars and inoculated with ciliates, throughout the experiment

123

0

12

24

36

48

60

Time after inoculation (hours)

Fig. 3 Mean percentage area of tissue loss (±SEM) on inoculated (grey) and not inoculated (white) coral fragments with artificially inflicted injuries (n = 9 fragments per treatment). Healthy corals are not presented as all values equal zero

Coral Reefs Table 1 Average counts of total ciliates (±SEM) observed on injured and healthy coral fragments in the ciliate choice experiment at regular time intervals

Ciliate behaviour, morphology, and molecular identification

Time (h)

At lesion fronts, ciliates were observed to consume surface tissue layers, ingest zooxanthellae that appeared to remain intact, burrow into polyps, enter coelenterons through the mouths of polyps, move throughout tissue networks connecting polyps, and emerge feeding on mesenterial filaments. For both COTS-predated and artificially injured corals, microscopic observations 48 h after inoculation with ciliates revealed that many ciliates were either exhibiting pre-conjugating activity (e.g., slow moving, non-motile, bumping into each other) or were conjugated, indicating sexual reproductive activity. Other ciliates were observed to be in various stages of binary fission, an asexual reproductive activity more common to sexually immature ciliates (Ricci 1990). At late experimental time periods, when coral fragments had [95 % tissue loss, ciliates were encapsulated by a cyst wall and had reduced cell volumes. PCR amplicons of the 18S rRNA gene sequences derived from ciliates isolated from BrB lesions on inoculated, COTS-predated fragments formed four distinct, dominant bands on DGGE. When placed into a phylogenetic tree (Fig. 5), DNA sequences of the most prominent band that was present in all replicate samples clustered tightly with the previously described BrB-associated ciliate P. guamense Morph 2 (BrB Ciliate 99 % similarity, Sweet and Bythell 2012; Bourne et al. 2008). Three other sequences were retrieved and affiliated with the ciliates Cohnilembus verminus (DGGE B2, 98 % similarity), Holosticha diademata (DGGE B3, 98 % similarity), and Homalogastra setosa (DGGE B4, 95 % similarity; Fig. 5).

Injured

0.25

Healthy

42 ± 5

43 ± 5

12

0.13 ± 0.04 9 103

0

24 36

0.21 ± 0.06 9 103 0.28 ± 0.06 9 103

0 0

At 0.25 h, numbers refer to ciliates that had migrated out of the water bridge and were moving towards a healthy or injured coral. At all subsequent time points, numbers refer to total ciliates observed over the whole healthy or injured coral fragments

30

Tissue loss - Injured Tissue loss - Healthy

Area of tissue loss (%)

25

20

15

10

5

0 0.25

12

24

36

Time after inoculation (hours)

Fig. 4 Mean percentage area of tissue loss (±SEM) on injured (grey) and healthy (white) coral fragments after ciliate inoculation into a water bridge between healthy coral fragments and physically injured coral fragments (n = 5 fragments per treatment). Note percentage area of tissue loss = 0 for all healthy fragments

artificially injured coral fragments (n = 5 experimental replicates), no significant differences in ciliate numbers were observed between healthy and injured corals (42 ± 5 and 43 ± 5 ciliates, respectively, p [ 0.99) during the first 15 min of the experiment. However, by 12 h, ciliates had migrated towards injured corals, resulting in signs of BrB lesions. Total ciliate numbers at lesion borders reached 0.13 ± 0.04 9 103 ciliates (Table 1), and the average percentage area of tissue loss in 12 h was 7.27 ± 1.72 % (Fig. 4). Ciliate densities at lesion fronts increased approximately sixfold throughout the experiment, reaching 0.28 ± 0.06 9 103 ciliates at lesion borders at 36 h (W = 15, z = 2.63, p \ 0.01; Table 1). Average percentage area of tissue loss reached 19.61 ± 8.25 % by the end of the experiment at 36 h (Fig. 4). No visible signs of BrB lesions were observed on healthy corals throughout the experiment, and ciliates were not observed on these corals.

Discussion This study demonstrates that COTS predation scars and physical injury provide an opportunity for the ciliate species P. guamense to enter compromised coral tissue, form BrB lesions, and ultimately kill the coral host. Following ciliate inoculation, health-compromised corals in all experiments quickly developed characteristic BrB disease lesions. In contrast, all healthy corals that were similarly inoculated remained healthy and ciliate-free. These results demonstrate that coral injuries, either through breakage or predation, allow P. guamense ciliates to opportunistically colonise the compromised coral host, proliferate, and systematically consume tissue from the infected coral. The susceptibility of corals to disease has previously been associated with physical injury (Willis et al. 2004; Page and Willis 2008; Raymundo et al. 2009), and significant increases in BrB prevalence have been recorded at heavily

123

Coral Reefs 69

BrB Ciliate (this study) KF359478 Sweet and Bythell 2012 M2 PWS HQ204546

82

Sweet and Bythell 2012 M2 Br B J N626269 Bour ne et al 2008 Br B Cil AY876050 Sweet and Bythell 2012 M1 Br B J N626268 99 10 0

Sweet and Bythell 2012 M1 PWS HQ204545

Qiu et al 2010 Cil2-2 HM030719 Qiu et al 2010 Cil2-1 HM030718 Philaster digitiformis

FJ648350

Qiu et al 2010 Cil1 HM030717 Parauronema longum AY212807

81

10 0

Philaster dicentrarchi

KC285109

Miamiensis avidus J N689230 DGGEB4 (this study) KF359481

99

AB505510

60 87

Homalogastra setosa GU590870 Cohnilembus verminus

98 10 0 96

HM236339

DGGEB2 (this study) KF359479 DQ504341 Urocentrum turbo EF114300 DGGEB3 (this study) KF359480 10 0

Holosticha diademata DQ059583

Trichodina sp. HM583859 62

S tentor coeruleus AF357145 Toxoplasma gondii X75429

0.02

Fig. 5 Neighbour-joining consensus tree of partial 18S rRNA gene sequences of four species of ciliates (highlighted in bold) identified on corals from the ‘COTS experiment’

impacted tourism sites (Lamb and Willis 2011). This highlights the need for a thorough in situ study assessing how frequently ciliates find and colonise injured corals. Once ciliates were established at lesion sites on our experimental corals, their densities increased rapidly (up to 4.2 ± 0.5 ciliates per mm2; F2,16 = 23.63, p \ 0.0001), correlating with elevated lesion progression rates (up to 4.6 ± 0.3 cm d-1; F1,3 = 6.92, p \ 0.001) and rapid loss of coral tissue, resulting in rapid colony-wide mortality in \84 h. Feeding scars and injuries offering the potential for invasion by opportunistic ciliates are likely to increase BrB prevalence in the coming decades, particularly given that the frequency, duration, and intensity of storms are predicted to increase with changing climate (Emanuel 2005; Abbs et al. 2006; Knutson et al. 2010; Emanuel 2013), and occurrences of COTS outbreaks will become more frequent (Fabricius et al. 2010). Outbreaks of COTS on the GBR have been and continue to be a source of open feeding scars that contribute to the widespread occurrence of injuries susceptible to BrB pathogenesis, further contributing to declines in coral cover in regions with high numbers of COTS. Recently, high BrB abundance has been observed in COTS outbreak areas (northern GBR sector; SM Katz, personal observation), highlighting the need for a focussed study to compare the prevalence of BrB between COTS outbreak and non-outbreak sites to more robustly document

123

the links between predation and BrB. Outbreaks of Drupella spp. have been documented to follow bleaching events (Antonius and Riegl 1998; Baird and Marshall 2002), and the snail’s predation may promote the onset of BrB by providing an entry point for the ciliates and by serving as a vector of ciliates to injured corals that may result in the onset of the disease (Nicolet et al. 2013). De’ath et al. (2012) highlighted that COTS outbreaks and tropical storm activity were the two largest contributors to coral cover declines on the GBR over the last 27 yr. However, other environmental factors such as thermal anomalies, high levels of ultraviolet radiation, changing osmotic pressure associated with flood plumes, and decreased water quality from nutrient enrichment, pollutants, and sedimentation are compromising coral health and increasing their susceptibility to disease (Goreau et al. 1998; reviewed in Rosenberg and Ben-Haim 2002; Bruno et al. 2003; Harvell et al. 2007; Miller et al. 2009). For example, coral bleaching is a significant cause of global coral mortality since the 1980s (Glynn 1993; Wilkinson and Souter 2008; Osborne et al. 2011; Dea’th et al. 2012) and mortalities may be partly a result of pathogenic infections following thermal stress (reviewed in Rosenberg and Ben-Haim 2002; Miller et al. 2009). Therefore, corals are becoming challenged by synergistic physical, biotic, and anthropogenic stressors, in addition to severe storms and outbreaks of COTS that may similarly provide an entry

Coral Reefs

site for BrB ciliates. The results of our current study, which highlight that injured corals are more susceptible to invasion by opportunistically pathogenic ciliates, suggest that the prevalence of BrB is likely to increase at regional scales in the coming decades during COTS outbreaks, extreme weather, and mass coral bleaching events. Predation pressures and injuries cause corals to invest in highly energetic defence mechanisms such as mucus secretion, expression of immune response cascades, and allelopathy (Bruno and Witman 1996; McCook et al. 2001), as well as repair mechanisms such as regeneration of somatic tissue and skeleton (Goss 1992; Palmer et al. 2011). A coral’s regeneration capacity is a function of species and lesion shape (Meesters et al. 1997) and determined by a multitude of responses to environmental parameters (Kramarsky-Winter 2004) that can adversely affect a coral’s fitness and capacity to withstand assaults from other challenges, such as ciliate pathogens (Kramarsky-Winter 2004; reviewed in Rotjan and Lewis 2008). However, the 10-d acclimatisation period in our study provided experimental fragments with sufficient time to heal from fragmentation injuries and to create a barrier against possible ciliate incursions, giving evidence that open wounds can heal and regenerate tissue within a few days. For example, Palmer et al. (2011) demonstrated healing of 1- to 2-cm wounds on the coral Porites cylindrica after only 48 h. If lesions are large (e.g., from COTS predation), recovery rates decrease as energy needed to repair the lesion front increases, playing an important role in the coral’s recovery potential (Oren et al. 1997). Evidence that lesion size declined over 60 h on artificially injured corals that were not inoculated with ciliates in our study (Fig. 3) provides further support for assertions that corals are capable of rapid tissue regeneration following wounding. However, our finding that ciliates invade injury sites rapidly, resulting in signs of BrB disease within 24 h and rapid tissue losses within 60 h (up to 65 % tissue loss), highlights the virulence of P. guamense when provided with an entry point. Ciliates at injury sites were able to overcome the coral’s defensive mechanisms, including extensive mucus secretion and the sweeping motions of nematocyst-covered mesenterial filaments, to form characteristic BrB disease lesions. Overall, injury to corals caused by predation, destructive storms, localised impacts by reef-based human activities, destructive fishing practices, and wide scale anthropogenic influences are likely to provide an underlying factor for infestations by opportunistic pathogens (Burge et al. 2013). Thus, there is need for in situ investigations to fully understand how frequently opportunistic, pathogenic ciliates can find and attack injured corals. The 18S rRNA gene sequences of the dominant ciliate associated with the BrB lesions in this study affiliated

closely with P. guamense, a ciliate previously found associated with BrB lesions (Bourne et al. 2008; Sweet and Bythell 2012). Although Sweet and Bythell (2012) suggested that P. guamense (Morph 2) ciliates had a secondary role in BrB pathogenesis, our study clearly shows that this ciliate is capable of rapidly establishing BrB lesions and causing extensive tissue loss. Also, although Sweet and Bythell (2012) described a ciliate with different morphological features related to a Philaster sp. (named Morph 1), present on all BrB disease fragments at their study on Heron Island (southern GBR), and thought to play the primary aetiological role in BrB pathogenesis, we found no evidence of Philaster sp. (Morph 1) ciliates in extensive microscopic investigations, and no sequences related to this ciliate were retrieved in DGGE profiling of the ciliate community. Three other ciliate-related sequences were retrieved and identified as C. verminus (DGGE B2), previously described as a commensal organism associated with molluscs (Xiaozhong et al. 1995), H. diademata (DGGE B3), which has been previously associated with BrB- and WS-infected corals (Sweet and Bythell 2012), and H. setosa (DGGE B4), a ciliate mainly found in soil (Foissner 1998; Fig. 5). The role of three of the four ciliates (closely related to C. verminus, H. diademata, and H. setosa) in BrB pathogenesis is currently unknown. However, microscopic behavioural observations suggest that they may play an opportunistic role, as they were observed to engulf Sybiodinium behind the trailing edge of the lesion front. P. guamense appears to be the primary ciliate responsible for progression of lesions and mortality on our experimental corals, as it was consistently found in large numbers directly at the lesion front actively consuming live coral tissue. Our finding that P. guamense actively migrates towards coral lesions when provided with the option of healthy or injured fragments implies that these ciliates are attracted to tissue sloughing from fresh coral injuries that may occur during other disease infections (Glynn 1993) and stressrelated necrosis (Peters 1984) causing tissue sloughing, although it is unclear whether some other signal from the wounded coral is also involved. It is unlikely that asexual reproduction (rather than active migration) caused the rapid increase in ciliate numbers on injured fragments, as no ciliate reproductive activity was observed within 36 h of the choice experiment. On average, approximately 270 of the 500 ciliates inoculated into each treatment replicate were observed on injured corals at 36 h, and no ciliates were observed on any of the healthy corals throughout the duration of the choice experiment. Furthermore, reproductive activity (e.g., slow movement and conjugation) was only observed to start 48 h after inoculation in other experiments. Reproduction has been observed to take up to 46 h in cultures of Scuticociliatea (Glauconema trihymene)

123

Coral Reefs

and only after the ciliates were well fed (Long and Zufall 2010). Thus, it is unlikely that the number of ciliates observed on injured corals could be accounted for by reproductive activity rather than active migration. The mode by which BrB ciliates target corals in situ is not understood, although different corallivores, including A. planci, have been suggested to act as disease transmission vectors (Nugues and Bak 2009; Chong-seng et al. 2011; Nicolet et al. 2013). For example, field observations suggest the potential for BrB disease to be spread between corals by corallivorous fish selectively feeding on disease lesions (Chong-Seng et al. 2011) and have shown that the corallivorous snail Drupella sp. can be a vector of BrB on the staghorn coral A. muricata (Nicolet et al. 2013). Observations that five out of eight table corals (Acropora cytherea) exhibited signs of BrB infection following COTS predation further link COTS and BrB (Nugues and Bak 2009). It was not possible to confirm in these studies, however, whether fish or COTS were the vector or whether ciliates infested the corals via a different mode. Although adult Drupella snails transmit BrB in manipulative experiments, they may only account for localised disease spread. If ciliates adhere to the mouthparts of corallivorous fish and then are deposited on subsequent feeding scars as the corallivore moves from one coral colony to the next, BrB may be spread more widely. However, further investigations are needed to examine the likelihood that ciliates are transmitted on mouthparts or within faecal deposits and to identify potential vectors. Results from the current study clearly demonstrate that ciliates have the ability to actively move towards and infect health-compromised corals, at least when corals are in close proximity. Furthermore, the ability of ciliates to remain dormant but viable for months in an encysted state (Fenchel 1990) may allow them to be transmitted between reefs and even regions by both biotic and abiotic vectors. This adaptation highlights the potential for BrB disease to impact reefs with high levels of health-compromised corals. Currently, the phylogenetic relationships among BrB ciliates from geographically disparate locations are unknown, but further studies could reveal patterns of ciliate translocation on regional and ocean-basin scales. This study conclusively demonstrates that COTS predation and physical injuries can promote rapid infestations of P. guamense ciliates, resulting in BrB disease lesions and rapid mortality of infected corals in aquaria-based studies. With COTS outbreaks, tropical storms, and cyclones all predicted to increase in tropical reef systems in coming years, these findings suggest that there will be associated increases in BrB prevalence on impacted reefs. BrB ciliates have the potential to actively detect and colonise compromised corals, rapidly proliferate to form dense ciliate aggregations, and progress across corals consuming tissue and causing extensive coral loss. The

123

resulting pressure of extensive, widespread outbreaks of coral diseases and in particular BrB, in combination with existing pressures, including temperature anomalies, eutrophication, and ocean acidification, will further exacerbate coral cover declines on tropical reefs. This study underscores the urgent need for further detailed studies of the biology, behaviour, chemical attractants, vectors, and reservoirs of BrB pathogens, to be able to understand and manage this emerging threat on coral reefs. Acknowledgments The authors acknowledge Naohisa Wada, Manuela Giammusso and the staff of Orpheus Island Research Station for their technical assistance in the pilot study, and to Liam Zarri and the staff of Lizard Island Research Station for their technical and logistical support. The authors also acknowledge Jean-Baptiste Raina for his assistance with statistical analyses, and Emmanuelle Botte, Jason Doyle, Kathleen Morrow and Andrew Muirhead for their laboratory assistance and advice at the Australian Institute of Marine Science. The authors also thank two anonymous reviewers for their comments that improved the article. This work was funded by a Lizard Island Research Foundation Fellowship awarded to F.J. Pollock for study at Lizard Island Research Station, a facility of the Australian Museum, and by funding to B.L. Willis through the ARC Centre of Excellence for Coral Reef Studies.

References Abbs DJ, Aryal S, Campbell E, McGregor J, Nguyen K, Palmer M, Ragter T, Watterson I, Bates B (2006) Projections of extreme rainfall and cyclones. A report to the Australian Greenhouse Office, Canberra Abra`moff MD, Magalha˜es PJ, Ram SJ (2004) Image processing with Image. J. Biophotonics International 11:36–42 Antonius A, Riegl B (1998) Coral diseases and Drupella cornus invasion in the Red Sea. Coral Reefs 17:48 Baird AH, Marshall PA (2002) Mortality, growth and reproduction in scleractinian corals following bleaching on the Great Barrier Reef. Mar Ecol Prog Ser 237:133–141 Barengo NC, Benito JM, Abate CM (2004) Molecular characterization of microbial communities from marine environments. In: Spencer JFT, Ragout de Spencer AL (eds) Environmental microbiology, methods and protocols. Humana Press Inc., New Jersey, pp 51–74 Birkeland C, Lucas JS (1990) Acanthaster planci: major management problems of coral reefs. CRC Press, Boca Raton, p 257 Bourne DG, Boyett HV, Henderson ME, Muirhead A, Willis BL (2008) Identification of a ciliate (Oligohymenophorea: Scuticociliatia) associated with brown band disease on corals of the Great Barrier Reef. Appl Environ Microbiol 74:883–888 Bruno JF, Witman JD (1996) Defense mechanisms of scleractinian cup corals against overgrowth by colonial invertebrates. J Exp Mar Biol Ecol 207:229–241 Bruno JF, Petes LE, Harvell CD, Hettinger A (2003) Nutrient enrichment can increase the severity of coral diseases. Ecol Lett 6:1056–1061 Burge CA, Kim CJS, Lyles JM, Harvell CD (2013) Special issue oceans and humans health: the ecology of marine opportunists. Microb Ecol 65:869–879 Chong-Seng KM, Willis BL, Cole AJ, Pratchett MS (2011) Selective feeding by coral reef fishes on coral lesions associated with brown band and black band disease. Coral Reefs 30:473–481

Coral Reefs De’ath G, Fabricius KE, Sweatman H, Puotinen M (2012) The 27– year decline of coral cover on the Great Barrier Reef and its causes. Proc Natl Acad Sci USA 109:17995–17999 Done TJ (1992) Effects of tropical cyclone waves on ecological and geomorphological structures on the Great Barrier Reef. Cont Shelf Res 12:859–872 Emanuel K (2005) Increasing destructiveness of tropical cyclones over the past 30 years. Nature 436:686–688 Emanuel K (2013) Downscaling CMIP5 climate models shows increased tropical cyclone activity over the 21st century. Proc Natl Acad Sci USA 110:12219–12224 Endean R, Stablum W (1973) A study of some aspects of the crownof-thorns starfish (Acanthaster planci) infestation of reefs of Australia is Great Barrier Reef. Atoll Res Bull 167:1–62 Fabricius KE, Okaji K, De’ath G (2010) Three lines of evidence to link outbreaks of the crown-of-thorns seastar Acanthaster planci to the release of larval food limitation. Coral Reefs 29:593–605 Fenchel T (1990) Adaptive significance of polymorphic life cycles in Protozoa: responses to starvation and refeeding in two species of marine ciliates. J Exp Mar Biol Ecol 136:159–177 Foissner W (1998) An updated compilation of world soil ciliates (Protozoa, Ciliophora), with ecological notes, new records, and descriptions of new species. Eur J Protistol 34:195–235 Glynn PW (1993) Coral reef bleaching: ecological perspectives. Coral Reefs 12:1–17 Goreau TF (1964) Israel south red sea expedition, 1962, reports: 2. On the predation of coral by the spiny starfish Acanthaster planci (L.) in the southern Red Sea. Bull Sea Fish Res Stat Haifa 35:1–5 Goreau TF, Lang JC, Graham EA (1972) Structure and ecology of the Saipan reefs in relation to predation by Acanthaster planci (Linnaeus). Bull Mar Sci 22:113–152 Goreau TF, Cervino J, Goreau M, Hayes R, Hayes M, Richardson L, Smith G, DeMeyer K, Negelkerken I, Garzon-Ferrera J, Gil D, Garrison G, Williams EH, Bunkley-Williams L, Quirolo C, Patterson K, Porter JW, Porter K (1998) Rapid spread of diseases in Caribbean coral reefs. Rev Biol Trop 46:157–171 Goss RJ (1992) The evolution of regeneration: adaptive or inherent? J Theor Biol 159:241–260 Great Barrier Reef Marine Park Authority (2004) Great barrier reef marine park zoning plan 2003. Harvell CD, Mitchell CE, Ward JR et al (2002) Climate warming and disease risks for terrestrial and marine biota. Science 296:2158–2162 Harvell CD, Jorda´n-Dahlgren E, Merkel S, Rosenberg E, Raymundo L, Smith G, Weil E, Willis BL (2007) Coral disease, environmental drivers, and the balance between coral and microbial associates. Oceanography 20:172–195 Hayes RL, Goreau NI (1998) The significance of emerging diseases in the tropical coral reef ecosystem. Rev Biol Trop 46:173–185 Janse I, Bok J, Zwart G (2004) A simple remedy against artifactual double bands in denaturing gradient gel electrophoresis. J Microbiol Methods 57:279–281 Jousset A, Lara E, Nikolausz M, Harms H, Chatzinotas A (2010) Application of the denaturing gradient gel electrophoresis (DGGE) technique as an efficient diagnostic tool for ciliate communities in soil. Sci Total Environ 408:1221–1225 Kenchington R (1978) The crown-of-thorns crisis in Australia: a retrospective analysis. Environ Conserv 5:11–20 Knutson TR, McBride JL, Chan J, Emanuel K, Holland G, Landsea C, Held I, Kossin JP, Srivastava AK, Sugi M (2010) Tropical cyclones and climate change. Nature Geosci 3(3):157–163 Kramarsky-Winter E (2004) What can regeneration processes tell us about coral disease? In: Rosenberg E, Loya Y (eds) Coral health and disease. Springer, Berlin, pp 217–230

Lamb JB, Willis BL (2011) Using coral disease prevalence to assess the effects of concentrating tourism activities on offshore reefs in a tropical marine park. Conserv Biol 25:1044–1052 Lara E, Berney C, Harms H, Chatziotas A (2007) Cultivationindependent analysis reveals a shift in ciliate 18S rRNA gene diversity in a polycyclic aromatic hydrocarbon-polluted soil. FEMS Microbiol Ecol 62(3):365–373 Lobban CS, Raymundo L, Montagnes DJS (2011) Porpostoma guamense n. sp., a philasterine scuticociliate associated with brown-band disease of corals. J Eukaryot Microbiol 58:103–113 Long H, Zufall RA (2010) Diverse modes of reproduction in the marine free-living ciliate Glauconema trihymene. BMC Microbiol 10:108 Mann HB, Whitney DR (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 18:50–60 McCook LJ, Jompa J, Diaz-Pulido G (2001) Competition between corals and algae on coral reefs: a review of evidence and mechanisms. Coral Reefs 19:400–417 Meesters EH, Pauchli W, Bak RPM (1997) Predicting regeneration of physical damage on a reef-building coral by regeneration capacity and lesion shape. Mar Ecol Prog Ser 146:91–99 Miller J, Muller E, Rogers C, Waara R, Atkinson A, Whelan KRT, Patterson M, Witcher B (2009) Coral disease following massive bleaching in 2005 causes 60% decline in coral cover on reefs in the US Virgin Islands. Coral Reefs 28:925–937 Myers R, Raymundo L (2009) Coral disease in Micronesian reefs: a link between disease prevalence and host abundance. Dis Aquat Organ 87:97–104 Nicolet KJ, Gardiner NM, Hoogenboom MO, Pratchett MS, Willis BL (2013) The corallivorous invertebrate Drupella aids in transmission of brown band disease on the Great Barrier Reef. Coral Reefs 32:585–595 Nugues MM, Bak RPM (2009) Brown-band syndrome on feeding scars of the crown-of-thorn starfish Acanthaster planci. Coral Reefs 28:507–510 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 Ormond RFG, Hanscomb NJ, Beach DH (1976) Food selection and learning in the crown-of-thorns starfish, Acanthaster planci (L). Mar Behav Physiol 4:93–105 Osborne K, Dolman AM, Burgess SC, Johns KA (2011) Disturbance and dynamics of coral cover on the Great Barrier Reef (1995–2005). PLoS One 6(3):e17516 Page CA, Willis BL (2008) Epidemiology of skeletal eroding band on the Great Barrier Reef and the role of injury in the initiation of this widespread coral disease. Coral Reefs 27:257–272 Page C, Baker DM, Harvell CD, Golbuu Y, Raymundo L, Neale SJ, Rosell KB, Rypien K, Andras J, Willis BL (2009) Influence of marine reserves on coral disease prevalence. Dis Aquat Org 87:135–150 Palmer CV, Traylor-Knowles NG, Willis BL, Bythell JC (2011) Corals use similar immune cells and wound-healing processes as those of higher organisms Plos One 6(8):e23992 Peters EC (1984) A survey of cellular reactions to environmental stress and disease in Caribbean scleractinian corals. Helgol Mar Sci 37:113–137 Qiu DJ, Huang LM, Huang H, Yang JH, Lin SJ (2010) Two functionally distinct ciliates dwelling in Acropora corals in the South China Sea near Sanya, Hainan Province, China. Appl Environ Microbiol 76:5639–5643 R Development Core Team (2013) R: A language and environment for statistical computing

123

Coral Reefs Raymundo LJ, Halford AR, Maypa AP, Kerr AM (2009) Functionally diverse reef-fish communities ameliorate coral diseases. Proc Natl Acad Sci USA 106:17067–17070 Ricci N (1990) The behaviour of ciliated protozoa. Anim Behav 40:1048–1069 Rosenberg E, Ben-Haim Y (2002) Microbial diseases of corals and global warming. Environ Microbiol 4:318–326 Rotjan RD, Lewis SM (2008) Impact of coral predators on tropical reefs. Mar Ecol Prog Ser 367:73–91 Sokolow S (2009) Effects of a changing climate on the dynamics of coral infectious disease: a review of the evidence. Dis Aquat Org 87:5–18 Sukumaran S, George RM, Vinod K, Sobhana KS, Naomi TS, Manisseri MK (2011) Temporal patterns in biodiversity and health status of reef corals of Palk Bay. Indian J Fish 58:73–77 Sweet M, Bythell J (2012) Ciliate and bacterial communities associated with White Syndrome and Brown Band Disease in reef-building corals. Env Microbiol 14:2184–2199 Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0. Mol Biol Evol 24(8):1596–1599

123

Waterhouse J, Grundy M, Gordon I, Brodie J, Eberhard R, Yorkstone H (2010) Managing the catchments of the great barrier reef. In: Ferrier RC and Jenkins A (eds) Handbook of catchment management. Wiley-Blackwell, pp 351–376 Weil E, Irikawa A, Casareto B, Suzuki Y (2012) Extended geographic distribution of several Indo-Pacific coral reef diseases. Dis Aquat Org 98:163–170 Wilkinson C, Souter D (2008) Status of Caribbean coral reefs after bleaching and hurricanes in 2005. Global Coral Reef Monitoring Network, and Reef and Rainforest Research Centre, Townsville, p 152 Willis BL, Page CA, Dinsdale EA (2004) Coral disease on the Great Barrier Reef. In: Rosenberg E, Loya Y (eds) Coral health and disease. Springer, Berlin, pp 69–104 Xiaozhong H, Kuidong X, Weibo S (1995) The commensal ciliate in mantle cavity of Argopecten. I. Morphological studies on Cohnilembus verminus. J Ocean Univ China 26:184–188