Coral Reefs (2006) 25: 531–543 DOI 10.1007/s00338-006-0133-6
R EP O RT
Isabelle J. Domart-Coulon Æ Nikki Traylor-Knowles Esther Peters Æ David Elbert Æ Craig A. Downs Kathy Price Æ Joanne Stubbs Æ Shawn McLaughlin Evelyn Cox Æ Greta Aeby Æ P. Randy Brown Gary K. Ostrander
Comprehensive characterization of skeletal tissue growth anomalies of the finger coral Porites compressa Received: 16 July 2005 / Accepted: 3 June 2006 / Published online: 14 July 2006 Springer-Verlag 2006
Abstract The scleractinian finger coral Porites compressa has been documented to develop raised growth anomalies of unknown origin, commonly referred to as ‘‘tumors’’. These skeletal tissue anomalies (STAs) are circumscribed nodule-like areas of enlarged skeleton and tissue with fewer polyps and zooxanthellae than adjacent tissue. A field survey of the STA prevalence in Oahu, Kaneohe Bay, Hawaii, was complemented by laboratory analysis to reveal biochemical, histological and skeletal differences between anomalous and reference tissue. MutY, Hsp90a1, GRP75 and metallothionein, proteins known to be up-regulated in hyperplastic tissues, were over expressed in the STAs compared to adjacent normal-appearing and reference tissues. Histological analysis was further accompanied by elemental and micro-structural analyses of skeleton. Anomalous skeleton was of similar aragonite composition to adjacent skeleton but more porous as evidenced by an increased Communicated by Ecology Editor P.J. Mumby I. J. Domart-Coulon De´partement Milieux et Peuplements Aquatiques, UMR 5178 BOME, Museum National d’Histoire Naturelle, 57 rue Cuvier, C.P.51, 75005 Paris, France N. Traylor-Knowles Æ G. K. Ostrander Department of Biology, Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA E. Peters Tetra Tech, Inc., 10306 Eaton Place, Suite 340, Fairfax, VA 22030, USA E. Peters Registry of Tumors in Lower Animals, 22900 Shaw Road, Suite 107, Sterling, VA 20166-4311, USA D. Elbert Æ J. Stubbs Department of Earth and Planetary Sciences, Johns Hopkins University, Baltimore, MD 21218-2685, USA C. A. Downs Haereticus Environmental Laboratory, P.O. Box 92, Clifford, VA 24533, USA
rate of vertical extension without thickening. Polyp structure was retained throughout the lesion, but abnormal polyps were hypertrophied, with increased mass of aboral tissue lining the skeleton, and thickened areas of skeletogenic calicoblastic epithelium along the basal floor. The latter were highly metabolically active and infiltrated with chromophore cells. These observations qualify the STAs as hyperplasia and are the first report in poritid corals of chromophore infiltration processes in active calicoblastic epithelium areas. Keywords Hyperplasia Æ Coral disease Æ Skeletal tissue Æ Chromophore cells Æ Porites compressa
Introduction Skeletal anomalies of scleractinian corals have been reported in reefs around the world and have been described as tumors or outgrowths because of their shape. In general, the anomalies are round, circumscribed, and K. Price Æ S. McLaughlin NOAA NOS Cooperative Oxford Laboratory, 904 South Morris St., Oxford, MD 21654, USA E. Cox Æ G. Aeby Hawaii Institute of Marine Biology, P.O. Box 1346, Kaneohe, HI 96822, USA G. K. Ostrander Pacific Biosciences Research Center, University of Hawaii at Manoa, 41 Ahui Street, Honolulu, HI 96813, USA P. R. Brown Æ G. K. Ostrander Department of Comparative Medicine, Johns Hopkins University, Baltimore, MD, USA G. K. Ostrander (&) 2500 Campus Road, Hawaii Hall 211, Honolulu, HI 96822, USA E-mail:
[email protected] Tel.: +1-808-9567837 Fax: +1-808-9562751
532
raised areas on the coral that differ markedly in morphology from adjacent tissue, often having fewer polyps and symbiotic zooxanthellae. They were first recorded in a dry specimen of the deep-water coral Madrepora kauaiensis in 1965 and interpreted as a tumor resulting from somatic budding that produced one single very large (i.e., gigantic) abnormal polyp (Squires 1965). They have since been documented in more than 40 species of scleractinians, 5 species of gorgonians, and 1 hydrozoan in both the Caribbean and the Indo-Pacific (reviewed by Sutherland et al. 2004). Although frequently referred to in the literature as ‘‘tumors,’’ these lesions represent several different proliferative processes and most should not be interpreted as a tumor in the strict pathology definition of ‘‘neoplasm’’ (abnormal cells that proliferate and continue to grow after the initial stimulus ceases). For example, validating White’s 1965 suggestion, the cases of gigantism in Madrepora sp. have been re-qualified in 1996 as polyp hypertrophy to wall off endoparasitic petrarcid crustaceans, and re-interpreted as galls (Grygier and Cairns 1996). Montipora sp. skeletal growth anomalies have been linked to skeletal invasion by polychaetes (Wielgus and Glassom 2002). The ‘‘tumors’’ in sea fans and other gorgonian octocorals have been found to be cellular reactions to algal (Morse et al. 1977; Goldberg et al. 1984), or fungal (e.g., Aspergillus sp.) invasions of the stiff proteinaceous gorgonian axis, in which there is amoebocyte infiltration and hyperplasia of the axis epithelium, resulting in hypersecretion of gorgonin and melanization to wall off the algae or fungi (Petes et al. 2003). Scleroblasts subsequently encapsulate the coated algae and pigmented sclerite accumulation results in the development of purple nodules, in a localized physical response more akin to granulomas (inflammatory reactions to encapsulate parasites) than tumors (Goldberg et al. 1984; Smith et al. 1998; Alker et al. 2004). Among scleractinians, acroporids appear to be the most prone to developing skeletal anomalies (Cheney 1975; Bak 1983; Peters et al. 1986; Coles and Seapy 1998; Yamashiro et al. 2000), characterized by rapid uncoordinated growth of the tissue and skeleton growing over healthy tissue, with loss of coral polyp structure and symbiotic algae (zooxanthellae). Uncontrolled cell proliferation in the absence of an obvious stimulus (such as an animal or plant parasite) qualified them as true neoplasms (Peters et al. 1986). In addition, the Acropora sp. lesions prevented colony growth and branch extension (Cheney 1975; Bak 1983), with reduced mucous secretory cells in the epidermis so that the lesions were susceptible to ulceration and colonization of the skeleton by algae. Gonad development was prevented and the polyps in these lesions died, although the entire colony did not die. Biochemical studies have reported depletion of storage lipids in the Montipora informis skeletal anomalies, implying increased energy demand for sustained tissue growth (Yamashiro et al. 2001). These lesions have been attributed to a neoplastic proliferation of the calicoblasts, the cells that produce the skeleton.
However, Work and Rameyer (2005) have found skeletal anomalies in Acropora cytherea that result from proliferation of the gastrodermal cells to form similar porous protuberant masses as gastrovascular canals increase and these acroporid lesions were characterized as hyperplasia of gastrovascular canals. Peters et al. (1986) interpreted lesions in which polyps and tissues appear normal but enlarged and having more cells and structures than adjacent tissue (e.g., Loya et al. 1984; Work and Rameyer 2005), as hyperplasms or areas of accelerated growth. Few of these latter lesions have been examined histologically, although cellular hyperplasia and hypertrophy have been noted. This type of skeletal anomaly has been described as a tumor in many species of encrusting or massive corals, including the family Poritidae (Loya et al. 1984; Hunter and Field 1997). In the massive Pavona clavus (Gateno et al. 2003), tumor-like anomalies were characterized morphologically by porous skeleton and changes in the corallite (i.e., the calcium carbonate skeletal cup elaborated by each individual polyp) and the presence of fewer polyps and symbiotic zooxanthellae per surface area. These ‘‘bumps’’ grew faster than the surrounding tissue and were assumed to be formed through accelerated cell growth. However, polyp structure was retained and apart from the reduction in zooxanthellae density, no cytological changes were reported (Gateno et al. 2003). Areas of increased growth were restricted on the colony and could not be transmitted by fusion between colonies (Gateno et al. 2003). Physiological mechanisms, cell types and potential pathogenic agents involved in sustained tissue growth have not been identified. In the absence of evidence for neoplastic cells, use of the term ‘‘hyperplasm’’ is questionable. Hence we propose to use the name skeletal tissue anomaly (STA) in place of ‘‘tumor’’ and this terminology will be used in the rest of this work. This study examined the STAs of the finger coral Porites compressa (Hexacorallia: Scleractinia: Fungiina: Poritidae) which dominate the Hawaiian reef communities along with Porites lobata. The P. compressa is a perforate stony coral composed of a colony of small individual polyps embedded 3 mm deep into the calcium carbonate exoskeleton and laterally connected through an extensive network of gastrovascular canals (Barnes and Lough 1992). Corallites are contiguous, 0.6 to 1.3 mm in diameter, and filled with 12 septae (vertical skeletal plates radiating from the corallite wall). The CaCO3 skeleton deposition is biologically controlled by the aboral tissue via its specialized calicoblastic epithelium. Kaneohe Bay, Oahu, Hawaii has a high occurrence of the STAs relative to the rest of the Hawaiian islands (G. Aeby, unpublished observations). The watershed contributing to Kaneohe Bay is about 14.7 km2, steeply sloped, and subject to intense rainfall which can rapidly impact near-shore water quality through resulting nutrient influxes of nitrogen, phosphorus and suspended inorganic and organic sediment (e.g., Ringuet and Mackenzie 2005). The corresponding land use includes
533
conservation, agriculture and low-density urbanization (Nichols et al. 1996). As such, significant anthropogenic impacts persist. For example, suspended particulate matter has recently been implicated in elemental transport of lead, zinc, copper, cobalt nickel and vanadium into the bay (De Carlo et al. 2004) and likely contributes to transport and deposition of other compounds. Links to known environmental stressors have not been established and the causative agent(s) for these lesions remains to be determined. Differences in the composition and abundance of microbial communities have recently been reported in the mucus of STA bearing colonies compared to healthy colonies of P. compressa from Kaneohe Bay (Breitbart et al. 2005). However, no pathogen was isolated. Abundant mucus production by corals of the genus Porites is part of a non-specific stress reaction and its chemical composition may support the growth of a different bacterial community, as a secondary effect to the stress due to the anomaly. The aims of this study were to determine the prevalence of STAs in P. compressa populations in Kaneohe Bay, Hawaii, and to characterize differences in the skeletal, cellular and biochemical composition of their tissues. In doing so it was hoped to gain insight into potential differentiation, proliferative, hyperplastic and/ or neoplastic processes that might be occurring.
Materials and methods Study area Fragments of P. compressa colonies bearing the STAs were collected along the reef crest off the Coconut Island, Kaneohe Bay, Oahu Island, Hawaii (21 26.000’N, 15747.000’W) at a depth of 0.5 to 1 m on four occasions between June 2003 and December 2004. A total of 22 samples were examined, each originating from an individual coral colony that displayed both anomalous and apparently healthy reference tissue. Histological analysis was conducted on 8 samples, petrographic sections on 4 samples and skeletal analysis on 10 samples. Biochemical analysis was conducted on 5 P. compressa specimens sampled from Kaneohe Bay at a depth of 1 m in February 2005 and included the STA tissue and normal-appearing tissue from the same colonies. To minimize variability due to the location of the tissue, only the dome-like tip (4 mm) of the finger-like extension from a colony was used as a sample for analysis for normal-looking tissue. The tops of bulbous protrusions that characterize the STA tissue were used as the source for sample collection, to maintain a similar consistency in exposure to incident light. Field surveys The prevalence and size of P. compressa STAs were recorded across the reef. Two 10-m transect lines were laid
parallel to the reef edge linearly separated by approximately 5 m. Along each line, a 10 m · 1 m belt transect was surveyed for the number of colonies of P. compressa and the number of P. compressa with STAs. The number of STAs was recorded, and the diameter of the largest and smallest STA on each colony measured to the nearest millimeter. Surveys were conducted on the reef flat, reef edge and reef slope. Transect lines between reef zones were parallel to each other and separated by approximately 3 m. The surface area covered by each STA was measured in specimens shipped to Johns Hopkins University with the aluminum foil technique (Marsh 1969). Briefly, aluminum foil was applied to the surface of the anomaly, notches were made in the foil to fit into crevices, and weights were measured. Weight was calibrated against the weight of a 100-cm2 piece of foil and the surface area of each anomaly was calculated. Histopathology The P. compressa colony fragments composed of anomalous and adjacent reference tissue were collected for histopathology in June 2003 (three fragments) and June 2004 (six fragments). The fragments were immediately fixed, respectively, in Helly’s fixative and in Z-Fix (Anatech, Battle Creek, MI, USA) and sent to the International Registry of Coral Pathology (IRCP) in Oxford, MA, USA for histopathological processing following the protocol of Peters et al. (2005) and archiving (IRCP accession nos. 71, 72, 73 and 192A,B, C, D, E, F). To investigate tissue structure, gonad development, cellular characteristics and facilitate the difficult interpretation of 2D analyses from the 3D coral tissue, serial histological sections were made for a total of eight specimens and stained in turn with Mayer’s hematoxylin and eosin (H&E), Cason’s trichrome, and Alcian Blue– Periodic Acid Schiff (AB/PAS) according to protocols detailed in Peters et al. (2005). One section of each stain was examined. Stained sections were submitted to the Registry of Tumors in Lower Animals (RTLA) in Reston, VA, USA for archiving. Tissue structure was described for the normal and STA portions of the specimens. Polyp length was indicated by the depth of the tissue layer, measured from the base of the polyp to the oral surface epithelium. Stage of gonad development within the coral mesenteries was recorded. The largest diameter of either spermaries (three specimens) or ova (five specimens) was measured in longitudinal sections of the polyp gastric cavity. Abundance and diameter of gastrovascular canals was recorded in the lower third of the polyp. Thickness of the aboral tissue bilayer lining the gastrovascular canals was measured. Cytological differences between the normal and STA portions of the specimens were recorded. In the gastrodermal epithelium of the upper polyp region, along the pharynx, zooxanthellae density was estimated by
534
counting the number of zooxanthellae per 1.2 mm length of gastrodermis. In the calicoblastic epithelium overlying the skeleton, the size of calicoblasts was measured, the number of chromophore cells was counted per 1.5 mm length of calicoblastic epidermis, and local alterations of the aboral tissue in relation to changes in cell type and composition were recorded. To investigate autofluorescence of coral cell types and detect fungal hyphae in coral subskeletal space, we used Calcofluor White M2R (CW) (Sigma) which brightens the autofluorescence of chitin component of fungal wall (Rohringer et al. 1977). Unstained, formalin-fixed, histological sections 4–6 lm thick were incubated for 1 h in the dark at room temperature with CW 0.003% in PBS 50 mM pH 7.4 then rinsed and nuclei were counterstained for 20 min with 10 lg ml 1 Hoechst 34580 nuclear stain. Negative control sections were prepared without CW brightener, with and without Hoechst 34580 counterstain, to evaluate autofluorescence of unstained, formalin-fixed material. Sections were coverslipped with Permafluor and viewed on a Zeiss LSM510 Meta confocal laser scanning microscope using a blue diode laser at 405 nm, and Argon laser lines at 488 and 543 nm. Images from the same optical field were obtained (1) in Differential Interference Contrast (DIC) at 543 nm excitation; (2) at 405 nm excitation Band Pass 420–480 nm emission for nuclear DNA; and (3) at 488 nm excitation Band Pass 540–590 nm emission for CW. To compare relative fluorescence intensities in control and CW stained tissue all images obtained were captured at the same optical thickness (1 lm optical sections) and same detector gain. Macroscopic analysis of skeletal architecture and elemental composition Live specimens were bleached overnight in 10% Clorox to remove tissue, rinsed in double-distilled water (ddH2O) and air-dried for analysis of skeletal architecture and elemental composition. Corallite diameter (the distance between two opposite theca walls of one polyp) and coenosteum length (the distance separating two adjacent corallites) were measured both in the anomaly and in the reference skeleton with a binocular fitted with a micrometer. The growing tips of adjacent branches were used as reference. Petrographic thin sections were prepared from tissuecovered, epoxy-embedded skeleton of four coral fragments and polished to the approximate thickness of 30 lm. The thickness of septae, i.e., vertical projections from the corallite wall, was recorded in cross-sections of septae, perpendicular to the growth axis, with a Nikon optical microscope fitted with a micrometer, both in the anomaly and in the adjacent reference area. The dried skeleton of three specimens (anomalies 1.4, 2.7, and 4.6 cm in diameter) was longitudinally cut, parallel to the growth axis, using a rock saw with a diamond-edged blade. Skeletal slabs were sanded down
to 3–4 mm thickness using a circular polishing wheel. X-radiographs of each coral slab were made at the Johns Hopkins Hospital Radiology Lab using a Siemens Mobilett II machine, operating at 57 KvP and 2.5 mAs, a film focus distance of 1 m and exposure time of about 1/120 s to reveal density and growth banding patterns. The X-radiographs were digitized by high resolution scanning (1,200 dpi SSAO) and converted to positive images. Grayscale values were analyzed with the Scion image analysis software (Scion Corp.) and density profiles were plotted along transects across the STA and the adjacent reference skeleton. Each doublet made of the juxtaposition of one low-density (LD) and one highdensity (HD) band defined a cycle of vertical extension, according to Chalker et al. (1985). Distance across the doublet defined the linear growth rate. Crystal structures of reference adjacent and anomalous skeletal carbonates were determined by selected area electron diffraction (SAED) using a Philips CM300FEG transmission electron microscope (TEM). Samples were extracted using a carbide-tipped probe and gently crushed in liquid nitrogen with an agate mortar and pestle. The crushed grains were suspended in ethanol and dispersed on holey carbon–copper grids. Multiple diffraction patterns were collected and indexed to determine crystal identity. Chemical compositions of Ca, Mg, Fe, Mn and Sr in skeletal fragments were determined in petrographic sections by electron microprobe analysis using a JEOL JXA-8600 SuperProbe with Advanced Microbeam Automation System. Analysis used X-ray, wavelengthdispersive spectroscopy (WDS) with carbonate standards from the Smithsonian National Museum collection. These standards were: USNM-136321 calcite, USNM-10057 dolomite, USNM-R2460 siderite, NMNH-14727 rhodochrosite and USNM-10065 strontianite. To minimize volatilization of CO2 during standardization and analysis, the microprobe was operated at 15 kV, 20 nA with a modestly spread probe diameter of 5–10 microns. Count times were 30 s for Ca, 60 s for other elements and were selected to provide analytical precision of less than ± 1%. Four replicate coral colonies were examined, for a total of 29 STA areas and 35 reference skeleton chemical measurements. ELISA measurement of hyperplasia-associated proteins Samples were assayed according to methods adapted from Downs (2005). P. compressa disks were ground frozen to a fine dust in a pre-chilled mortar and pestle using liquid nitrogen. About 100 ll of frozen sample powder was placed in locking 1.8 ml microcentrifuge tubes along with 1,400 ll of a denaturing buffer consisting of 2% SDS, 50 mM Tris–HCl (pH 7.8), 15 mM dithiothreitol, 10 mM EDTA, 3% polyvinylpolypyrrolidone (wt/vol), 0.005 mM salicylic acid, 0.001% (v/v) dimethyl sulfoxide, 0.01 mM AEBSF, 0.04 mM bestatin, 0.001 E-64, 2 mM phenylmethylsulfonyl fluoride,
535
2 mM benzamide, 5 lM a-amino-caproic acid, and 1 lg 100 ll 1 pepstatin A. Samples were vortexed for 15 s, heated at 93C for 6 min with occasional vortexing, and then incubated at 25C for 10 min. Samples were subject to centrifugation (13,500g for 8–10 min) and the middlephase supernatant was aspirated and placed in a new tube. The sample supernatant was subjected to a protein concentration assay by the method of Ghosh et al. (1988). Twenty-five nanograms of total soluble protein (TSP) from each sample were applied in triplicate onto a Nunc Maxisorp 96-well microplate, and allowed to adsorb at 25C in a humidified chamber for 14 h. Unadsorbed material was aspirated using a Biotek 96-well microplate washer, and the microplate was incubated for an hour with a blocking buffer consisting of 7% defatted milk powder in Tris-buffered saline (TBS: 50 mM Trizma base and 13.7 mM sodium chloride with a final pH of 8.2). The plates were then washed three times and incubated for 1 h at room temperature with one of the four antibodies diluted in blocking buffer. Antibodies were AB-Hsp90a1 (antibody against invertebrate Hsp90a1), AB-GRP75-INV (antibody against invertebrate Grp75/mortalin), AB-MET1-INV (antibody against metallothionein Type 1, invertebrates) and AB-MUTY-INV (antibody against MutY DNA glycosylase). Microplates were washed three times in TBS using the microplate washer, and then incubated for 1 h with donkey anti-rabbit Fab fragment conjugated to horseradish peroxidase (ImmunoJackson Labs, Maine, USA). The microplates were again washed three times in TBS, after which a luminal/hydrogen peroxide solution was added simultaneously to all wells on each plate, and the microplate was read with a Biotek fluorometer/luminometer microplate reader. All samples were assayed in triplicate with intra-specific variation per plate of less than 4% across the entire 96-well microplate. An eightpoint calibration curve corresponding to each antibody was plated in triplicate for each microplate.
Data from ELISA were tested for normality using the Kolmogorov–Smirnov test (with Lilliefors’ correction) and for equal variance using the Levene Median test. If the data were normally distributed and homogeneous, a one-way analysis of variance (ANOVA) was employed. When data did not meet the homogeneity of variances requirement for one-way ANOVA, a Kruskal–Wallis One-Way Analysis of Variance on Ranks was used. When significant differences were found among treatment means, the difference between each of the groups was tested using the Tukey–Kramer Honestly Significant Difference (HSD) method, the Dunn’s post hoc test, or the Holm–Sidak test as an exact alpha-level test (Sokal and Rohlf 1995). All data are reported as means ± standard error.
Results Field survey The overall prevalence of STAs in P. compressa colonies at the Coconut Island site, Kaneohe Bay, Hawaii, was 21.7 ± 8.3% (mean ± SE). Across all reef zones, 73 colonies with the STAs were assessed. The reef slope had fewer colonies with the STAs compared to the reef flat and edge (Fig. 1) and the STA prevalence was higher in shallow areas compared to deeper areas. Most affected colonies had multiple STAs, with larger colonies having more lesions, but there was a high variability and no clear trend. The number of STAs per colony varied widely, ranging from 1 to 116, with an overall average number of 13.7 (SE ± 2.4). The size of STAs observed in situ ranged from 0.3 to 14 cm in diameter. The STAs analyzed in the laboratory ranged in size from 1.4 to 5.3 cm in diameter (mean = 2.5 cm), corresponding to surfaces from 7 to 88 cm2, with a mean of 19.6 cm2. Ulceration of the center of anomalies and fouling with filamentous algae was sometimes detected, but during
Statistical analysis 40
Average Prevalence (%)
Histological and skeletal measurements were recorded from 4 to 10 distinct coral colonies (specimen). Paired data points were taken for each specimen both in anomalous and adjacent reference tissue. Each data point was the average of quadruplicate measurements taken in random fields with a graduated micrometer. Difference between mean characteristics of three colonies collected in June 2003 and five colonies collected in June 2004 was tested with a one-way ANOVA and was small enough to be non-significant, and so data from these two sets were pooled. To correct for the small sample size, a non-parametric Wilcoxon Signed-rank t test was used for matched pairs at the p < 0.05 significance level (JMP Jumpstart 5.0 software). Results of the microprobe analysis were statistically analyzed with a one-way ANOVA at the p < 0.05 level of significance.
35 30 25 20 15 10 5 0 reef flat
reef edge
reef slope
reef zones
Fig. 1 Prevalence of Porites compressa skeletal tissue anomalies (STAs) in the reef zones
536
the 2003–2004 18-month study period P. compressa colony survival was not affected by growth anomalies. Histopathology Polyps were twice as tall in the STA versus the adjacent reference area as, indicated by the increase in the distance from the base of the polyp to the oral surface epithelium from 2.9 ± 0.3 to 6.3 ± 2.3 mm (p = 0.004, n = 9) (Table 1). Gonads were only partially developed in the anomaly and often absent in the central, thickest, fastest growing region. Diameters of existing ova (five specimens) and spermaries (three specimens) were similar and pooled into gonad diameter, which was reduced by 23% in the STA compared to reference tissue, respectively, 106.3 ± 13.1 versus 134.9 ± 15.0 lm (p = 0.0008, n = 8). Gastrovascular canals, connecting adjacent polyps through the colony, were more abundant and larger in anomaly than in reference areas and were filled with mucus: their size (diameter) increased by 27%, at 160 ± 9.9 lm in the STA versus 126.1 ± 21.8 lm in the adjacent reference tissue (p = 0.008, n = 7). The aboral tissue layer was 76% thicker in anomalous versus adjacent reference tissue, measuring, respectively, 51.4 ± 7.3 lm in the STA versus 29.1 ± 5.7 lm in the adjacent reference area (p = 0.008, n = 8). Mesoglea thickness did not differ between anomalous and reference tissue, as revealed by the Heidenhain’s staining results (data not shown). Large intercellular spaces were detected in the basal floor of abnormal polyps, indicating local detachment of the aboral gastrodermis from the calicoblastic epithelium. Thickening of the aboral tissue was associated with changes in cellular composition and activity. Within anomalies, the zooxanthellae were present in the gastrodermal epithelium of the upper polyp, but their density was reduced by approximately 48% compared to reference tissue, at 44 ± 10 versus 84 ± 21 zooxanthellae per 1.2 mm length of oral gastroderm (p = 0.016, n = 7). Mucus was more abundant in the anomalous pharynx and gastrovascular canals, and the oral surface epithelium of abnormal polyps was partially lysed. Bacterial aggregates were visible both in reference and anomalous tissue in the mucus and within the oral
gastrodermal epithelium, but their abundance was increased in the mucus of anomalous polyps. Cytological changes were detected in the calicoblastic epithelium of anomalous tissue, in the lower third region of abnormal polyps and especially along their basal floor (Fig. 2). Cell clusters rich in acidophilic granules were distributed along a gradient, their number and size increasing from reference to the STA area. Within these clusters, the shape of calicoblast cells changed from flattened to cuboidal, their thickness more than doubled from 4.8 ± 2.7 to 12.5 ± 8.3 lm (p = 0.008, n = 6), and their acidophilic granule content (typical of the Porites genus, E. Peters unpublished observation) increased. Pseudopodia or filopodia were recorded to extend into the subskeletal space. These calicoblastic morphological changes indicated areas of increased metabolic activity. Along the abnormal polyp basal floor, the calicoblast epithelium was locally infiltrated with chromophore cells which have been described as yellow–brown pigmented cells, with a protean, amoeboidal shape and granulous content, characteristic of the Porites genus, (Duerden 1903). In reference tissue, the chromophore cells were primarily distributed in clusters in the oral gastrodermal epithelium with only a few single cells isolated in the aboral tissue. In contrast, in anomalous tissue, the chromophore cells were concentrated in pockets of five to eight cells in the aboral tissue, especially along the basal floor of abnormal polyps. Their density was significantly increased, from 19 ± 5 per 1.5 mm length of calicoblast epithelium in reference tissue to 68 ± 35 in the anomaly tissue (p = 0.031, n = 6) and their distribution within the aboral tissue shifted from equal apportioning between gastrodermal and calicoblastic epithelia (50–50%) to preferential localization in the calicoblastic epithelium (33–66%) of the anomaly. Abundant fungal hyphae, stained with AB/PAS, were detected in the sub-skeletal space underneath the calicoblastic epithelium of coral anomalies in decalcified, agar-enrobed paraffin sections. Their presence was confirmed within skeletal septae and along the basal floor of the polyps in the skeleton of nondecalcified petrographic sections. Although no quantitative trend was established, fungal hyphae were locally associated with areas of increased activity in the calicoblastic epithelium. In confocal microscopy (Fig. 3) the calicoblast epithelium of P. compressa was seen as an
Table 1 Tissue and cellular characteristics of Porites compressa skeletal tissue anomalies (STA)
Tissue
Cell
Characteristics
STA
Polyp length (mm) Gonad Diameter (lm) Gastrovascular canal diameter (lm) Aboral epithelia bilayer thickness (lm) Number of Zooxanthellae in oral gastroderm/ 1.2 mm Number of Chromophore cells in aboral tissue/1.5 mm Size of calicoblasts (lm)
6.3 106.3 160 51.4 44 68 12.5
± ± ± ± ± ± ±
2.3 13.1 9.9 7.3 10 35 8.3
Reference
p value
2.9 134.9 126.1 29.1 84 19 4.8
0.004 0.008 0.008 0.008 0.016 0.031 0.008
± ± ± ± ± ± ±
0.3 15.0 21.8 5.7 21 5 2.7
(n (n (n (n (n (n (n
= = = = = = =
9) 8) 7) 8) 7) 6) 6)
Data represent paired measurements taken in STA and adjacent reference tissue of individual colony fragments. Gonad diameter was pooled from measurement of ova (n = 5) and spermaries (n = 3). Means ± standard error
537 Fig. 2 Infiltration of chromophore cells in the calicoblastic epithelium of growth anomalies of Porites compressa, stained with Hematoxylin–Eosin. a Polyp basal floor in control adjacent tissue. b STA lesions. Arrows point to the local increased eosin pigmentation and thickness of the calicoblastic epithelium in STAs. c Aboral tissue in control. d STA: calicoblast morphological changes from flattened to cuboı¨ dal, with high eosinophil granule content and filopodial extensions into the skeleton; infiltration with chromophore cells. (sk skeleton, gc gastrovascular canal, sp spermaries, zoox zooxanthellae; cal calicoblast; ga aboral gastroderm; mg mesoglea; ch chromophore cells)
autofluorescent cell monolayer due to the UV autofluorescence of the numerous granules in their cytoplasm (Fig. 3a), both in CW stained and unstained control sections. In the STA lesion, in contrast to the flattened calicoblasts of control adjacent tissue, the calicoblasts were cuboidal with a basally located nucleus (near the thin mesoglea) almost masked by the increased content in autofluorescent granules. Apical filopodial projections filled with autofluorescent granules were confirmed extending into the subskeletal space. Fungal hyphae, which chitin wall was enhanced by the CW fluorescent brightener, were abundant in the subskeletal space, in close proximity to the cuboidal calicoblast cells (Fig. 3b). The chromophore cells had a diffuse autofluorescence at 488 nm excitation, BP 540–590 emissions in control unstained sections, which significantly increased when they were stained with CW and became patchy in the STA lesion compared to adjacent reference tissue (Fig. 3). The chromophore cells were observed clustering above and into the STA calicoblast epithelium, locally disrupting the thin mesoglea separating aboral gastroderm from the calicoblasts. Skeletal architecture and composition There was a small but statistically significant 20% increase in corallite diameter in the STA compared to adjacent reference area, from 1.32 ± 0.14 to 1.48 ± 0.16 mm (Table 2). Within the abnormal coral-
lite, organization of skeletal elements was retained and typical of the Porites genus, with all 12 septae present, although the columella (trabecular central axial structure) was wider and more spongy. The coenosteum (extra-thecal skeleton separating adjacent corallite walls) increased in length from 0.2 ± 0.1 mm in reference skeleton to 0.8 ± 0.1 mm in anomaly, leading to a reduction in the density of corallites per surface area. The skeletal septae were much thinner and longer in the STA than in adjacent reference corallites. Measurements confirmed a significant 40% reduction in septal thickness in abnormal versus reference corallites (Table 2) at, respectively, 56.6 ± 7.3 versus 95.8 ± 4.2 lm (p < 0.01, n = 4). Centers of calcification were visible both in reference and anomalous septae (dark dots in Fig. 3a) and indicated an upward skeletal extension. The decreased thickness of anomalous septae indicated that lateral extension of skeleton was limited. In X-radiographs of three skeletal slabs with anomalies, skeletal density was lower in the anomaly than in the adjacent reference area, indicating a more porous skeleton (Fig. 4). Relative density was about 40% lower in the STA than in the reference skeleton, reflecting the measured decrease in septae thickness. Density was patchy in the anomaly, whereas the succession of highand low-density bands defined a clear growth-banding pattern in reference skeleton, and indicated that the STA growth occurred throughout several cycles. Grayscale density profiles across the STA and reference areas from anomalies 1.4, 2.7, and 4.6 cm in diameter revealed that
538
Fig. 3 Fluorescent calicoblast and chromophore cells, in close association with fungal hyphae in the skeletal space. a–c Confocal optical sections of control tissue. d–f STA lesion. Counterstained with Hoechst 34580 nuclear stain (a, d: Excitation: 405 nm—Emission: band pass 420–480 nm), stained with Calcofluor White (CW) chitin brightener (b, e: Excitation: 488 nm—Emission: band pass 540–590 nm) and corresponding bright field images (c, f). Ultra
violet autofluorescence of granules in calicoblast cells (cal) was confirmed in control sections without CW (data not shown). The fluorescence at 488 nm of chromophore cells (ch) (enhanced by CW) changes from diffuse in control area to concentrated in granules in the STAs, where the chitin wall of fungal filaments (f) is detected in the skeletal space, close to the calicoblastic epithelium. (sk skeleton)
anomaly corresponded to skeletal growth, respectively, over 1 cycle (7.5 mm), 2 cycles (16 mm) and 4 cycles (35 mm). Vertical extension rate, calculated from the distance across doublets of high-density and low-density bands, was 5 to 6 mm/cycle in adjacent reference areas versus 8 to 9 mm/cycle in the STAs, corresponding to 49% increase in the rate of vertical extension. The STA growth was initiated during the deposition of one highdensity band, which was continuous between anomaly and adjacent areas. Thickening of the high-density band was observed at the base of the anomaly, and the STA’s hemispherical growth seemed to radiate from a few
abnormal polyps. The 49% increase in vertical extension rate approximately compensated for the 40% decrease in skeletal density, indicating a stable calcification rate (vertical extension rate · skeletal density) between anomaly and reference area. The aragonitic nature of calcium carbonate crystals was confirmed both in the STA and in the reference skeleton by electron diffraction spectra as all samples examined indexed as the aragonite polymorph of CaCO3 (data not shown). Results of the microprobe analysis revealed no significant differences between the elemental composi-
Table 2 Skeletal characteristics of Porites compressa skeletal tissue anomalies (STA)
Micro-architecture Growth
Skeletal characteristics
STA
Corallite diameter (mm) Coenosteum length (lm) Septae thickness (lm) Vertical extension rate (mm year 1)
1.48 0.8 56.6 8.8
± ± ± ±
0.16 0.1 7.3 0.3
Reference
p value
1.32 0.2 95.8 5.9
0.002 0.008 0.004 0.039
± ± ± ±
0.14 0.1 4.2 1.1
(n (n (n (n
= = = =
10) 8) 4) 3)
Corallite diameter and coenosteum length were measured on bleached skeleton from individual colony fragments. Vertical extension rate was calculated from gray-scale image analysis of digitized X-radiographs of skeletal slabs. Septae thickness and elemental composition of CaCO3 were measured in skeleton petrographic thin sections via ion microprobe analysis. Means ± standard error
539 Fig. 4 Skeletal architecture of Porites compressa STA. a Macrophotography of skeleton. b X-radiograph of skeletal slab through STA and control adjacent branches, with corresponding density profiles (d) relative density along transect distance (mm). d Petrographic section across the septae, with c detail of STA, and e of reference skeleton (filled inverted triangle represents center of calcification)
tion of anomalous skeletal carbonate and the nearby healthy skeleton (one-way ANOVA, p > 0.05) (Table 2). Mg/Ca and Sr/Ca ratios were similar, respectively, at 0.0049–0.0052 and 0.0076–0.0079 in both reference and STA skeleton. ELISA measurement of hyperplasia-associated proteins The GRP75 levels were significantly higher (p < 0.05) in the STA tissue (385.7 ± 25 fm/ng/TSP) compared to the adjacent normal-appearing tissue (315.7 ± 29.7 fm/ ng/TSP) from the same colony (Fig. 5). MutY was significantly higher (p < 0.05) in the STA tissue (9.3 ± 1.9 fm/ng/TSP) compared to the normal appearing tissue from the same colony (4.9 ± 1.1 fm/ ng/TSP (Fig. 5). Similarly, Hsp90 was significantly higher in the STA tissue (722 ± 80 fm/ng/TSP) compared to the normal appearing tissue (500 ± 76 fm/ng/ TSP) (Fig. 5). Finally, metallothionein was significantly higher (p < 0.05) in the STA tissue (137.3 ± 20.8 fm/ ng/TSP) compared to the normal appearing tissue (71.3 ± 20.6 fm/ng/TSP) from the same colony (Fig. 5).
Discussion The results indicate that shallow areas of the reef had a higher prevalence of P. compressa anomalies than the deeper areas. This is contrary to a recent assessment of the STA frequencies in the massive P. clavus community along the Pacific Coast of Costa Rica which exhibited an increasing gradient from shallow to deeper water (Gateno et al. 2003). In the absence of experimental data with in situ ultra violet (UV) sensors, the role proposed by Coles and Seapy (1998) of irradiance and UV-B
penetration in shallow reef areas as an environmental factor in the formation of scleractinian skeletal anomalies remains controversial. The linear extension rate for P. compressa was previously measured at approximately the same site (Point Reef, 21 26.18’N, 15747.56’W, Kaneohe Bay), along with temporal reference obtained by field staining with alizarin red: it was 13.4 mm over 6 months and did not vary with depth between 1.7 and 8.3 m (Grottoli 1999). These records indicate that the STAs characterized in this study corresponded to growth for less than 18 months, with an average of approximately 6 months (as calculated from the mean 2.5 cm STA diameter). Anomalous skeleton was 40% more porous, supported by thinning of anomalous skeletal septae by 40%. Reduction in skeletal density has been linked to incomplete septal thickening in the massive brain coral Diplora labyrinthiformis (Cohen et al. 2004). It was attributed to increased rate of tissue growth for unchanged rate of calcification, since the same quantity of calcium carbonate deposited along an increased surface of calicoblast tissue would result in skeletal extension without densification. The results of this study indicate that P. compressa STAs are areas of more active vertical extension of skeleton, which did not thicken, with unchanged calcification rate. The aragonite nature of calcium carbonate crystals was conserved in anomalies, as reported in Madrepora sp. skeletal anomalies which were qualified as gigantism due to hypertrophied polyps (Squires 1965). Skeletal elemental composition did not differ significantly between anomaly and adjacent reference areas on the P. compressa colony, in contrast to results in P. clavus skeletal ‘‘tumors’’ by Gateno et al. (2003) who reported higher Mg/Ca ratio and calcification of a higher magnesium aragonite. Both this study and Gateno’s
540
MutY
Hsp90
15
800 fmole Hsp90/ng TSP
fmole MutY/ng TSP
Fig. 5 Enzyme-linked immunosorbent assay (ELISA) analysis of proteins commonly associated with hyperplasia (MutY, Hsp90a1, GRP75 and metallothionein 1). Abnormal: sample from skeletal tissue anomaly (STA). Normal: sample from normal-appearing tissue adjacent to STA
10
5
0 Abnormal
400 200 0 Abnormal
Normal
Normal
Metallothionein I
Grp 75
200
400 fmole/ng TSP
fmole Grp75/ng TSP
600
300 200 100 0
150 100 50 0
Abnormal
chemical analyses were conducted at a microstructural resolution which could not discriminate between the centers of calcification and the radiating aragonite fibers, although recent evidence indicates that their elemental compositions differ significantly (Cuif and Meibom, personal communication). This low resolution could introduce a bias in the chemical results. According to Barnes and Lough (1992), skeletal extension accommodates tissue growth, and the depth of Porites sp. skeleton occupied by tissue reflects the rate of tissue growth. Polyp structure was retained throughout the anomaly, but the polyp density per surface area was reduced as the coenosteum, i.e., the skeletal distance separating adjacent polyps, was increased by a factor of 3. Abnormal polyps were quite large, only 20% wider but twice the length of reference polyps due to additional aboral tissue mass, localized in the lower third polyp area. This spacing of polyps, in conjunction with increased vertical extension explains the hemispherical shape of the anomalies: through successive somatic budding of one or few abnormal polyps, according to the polyp-oriented modeling of coral growth (Merks et al. 2004). The enlarged gastrovascular canal system in the STA lesions may improve flow of nutrients from the gastric cavity and the connecting polyps towards the calicoblastic epithelium, which might help to sustain increased tissue growth and skeletogenesis. Indeed, gastrovascular circulation in the branched scleractinian coral Acropora cervicornis (Gladfelter 1983) and in the soft octocoral Parerythropodium fulvum (Gateno et al. 1998) provides transport for nutrients and freely circulating cells. In the stony coral Favia favus, colony integration, i.e., translocation of resources from sites of acquisition to sites of maximal demand, is required for wound-healing by newly formed tissues (Oren et al. 2001).
Normal
Abnormal
Normal
The histological results indicated partial gonad development in the abnormal polyps, especially in the central, fastest-growing area of anomalies, suggesting a drain of energy resources primarily allocated to tissue growth at the expense of reproduction. In Caribbean sea fan octocorals, nodule growth in response to infection with fungi or protists has also been linked to localized suppression of reproduction (Petes et al. 2003). Yamashiro et al. (2001) have recorded decreased levels of storage lipid in tumors from M. informis compared to normal tissue, and interpreted this as increased energy demand for the tumor synthesis and calcification. A long-term study of affected P. compressa colonies would be important for assessing the effect of STAs on colony reproductive output and mortality, and for determining whether there is a critical surface/size beyond which anomalies stop growing. Abundant mucus and bacterial aggregates were observed within the gastric cavities of abnormal polyps, supporting the recent description by Breitbart et al. (2005) of increased mucus production and increased biomass of associated bacteria in skeletal tumors on P. compressa from Kaneohe Bay. However, the bacterial aggregates were restricted to the upper polyp area as is usually the case in normal poritid tissue, and were not detected in the calicoblastic epithelium. These observations do not support the role of a bacterial infectious agent in the STA lesions. In abnormal polyps swelling of the aboral tissue layer was observed, as well as thickening due to cytological changes in the calicoblast epithelium, especially in the polyp basal floor. The calicoblast cell shape shifted from flattened to cuboidal, their acidophilic granule content increased, and apical filopodial projections were detected, extending into the subskeletal space. These changes, taken together, indicate a differentiated and
541
highly metabolically active status of anomalous calicoblasts. Although formalin-fixed coral tissue sections can be prone to artifacts (autofluorescence, distortion of the subskeletal space), similar observations of columnarshaped calicoblasts with long filamentous apical projections have been reported in sections of undercalcified cryofixed and freeze-substituted hermatypic Galaxea fascicularis and Acropora formosa corals as well as the ahermatypic Tubastrea faulkneri coral, extending into the skeleton in areas of highest skeletal extension (Marshall and Wright 1993). These authors noted small granules, autofluorescent under UV illumination, in the apical region of calicoblast cells, close to the skeleton, and suggested they were associated with skeletal organic matrix deposition, without being able to determine whether the extensions were cytoplasmic, or part of an organic matrix. Observations from this study show that these projections are cytoplasmic in nature and that autofluorescent granules correlate with the eosin positive granules characteristic of calicoblasts of the Porites genus (E. Peters, unpublished observation). This morphology suggests involvement in rapid calcification, supporting the accelerated vertical extension rate observed in the STA lesions. Infiltration was observed in the anomalous aboral tissue of the STAs with chromophore cells, amoeboidshaped and pigmented, forming clusters preferentially distributed in the calicoblast epithelium. These cells seemed to migrate from the gastrodermal epithelium and intercalate between the calicoblasts, locally causing the separation of both epithelia. An inflammatory response is a process characterized by swelling and cell infiltration and occurs when injury or infection damages tissue. Corals are thought to have proficient defense mechanisms against injury and disease, although few components of their innate immune systems have been characterized to this day (reviewed by Mullen et al. 2004). The mucociliary system of scleractinians plays an important role and amoebocytes have been reported to migrate from uninjured to injured tissue and assist in encapsulation and wound healing (Goldberg et al. 1984; Meszaros and Bigger 1999). However, in this study very few amoebocytes were detected in the thin mesoglea of P. compressa. The infiltration of the chromophore cells, of unclear biochemical content and physiological function, within the active calicoblast epithelium of STA lesions are of particular interest. The combination of tissue swelling, cytological changes and infiltration of specific cell types are characteristic features of an inflammatory-like reaction, although the reason for such a reaction in the P. compressa STAs is still unknown. The four proteins measured, MutY, Grp75, Hsp90a1, and metallothionein, are often altered in hyperplastic and tumorigenic tissues in mammals and invertebrates (Wadhwa et al. 2002). The MutY is an evolutionarily conserved DNA glycosylase that functions to repair oxidized DNA bases, and is usually up-regulated in response to oxidative stress and in tissues that have a high tolerance to oxidative stress. The hyper-accumulation of
MutY in the STA tissue compared to reference tissue indicates that (1) the STA tissue may be responding to an oxidative stress event, consistent with an inflammatory-like hypothesis or a photo-oxidative stress; (2) the STA tissue has an increased capacity to deal with DNA damage lesions; and/or (3) the STA tissue, like hyperplasia in mammals, may have an altered transcriptionalregulatory profile. The Hsp90a1 contributes to cell cycle regulation, growth, development, evolution, apoptosis, cancer, and stress (Rutherford and Lindquist 1998; Holt et al. 1999; Caplin et al. 2003). The Hsp90 can also be up-regulated in hyperplastic tissue and in cells undergoing a high level of mitogenesis. Hyper-accumulation of the Hsp90 in the STA tissue compared to reference, likely indicates either a stressed physiological condition as a result of protein denaturing stress (e.g., heat stress, oxidative stress) or a hyper-mitogenic/cytokinetic physiological condition. The GRP75 (glucose-regulated protein 75) is a Hsp70 homologue chaperonin that is localized to the mitochondria. This protein functions to mature newly imported protein and may renature stress-denatured proteins. Modulation of the GRP75 protein levels often reflects mitochondrial protein metabolic activity, which in turn, can respond directly to changes in mitochondrial metabolic activity. This protein is often up-regulated in tumorigenic and hyperplastic tissue, most likely as a result of increased mitochondrial metabolic demands (Srokowski et al 2004). A significant increase of the GRP75 in the STA polyps indicates a difference in mitochondrial protein metabolic condition, although the factor responsible for this is speculative. The metallothionein examined in this study is a cysteine-rich protein homologous to the invertebrate and vertebrate Type 1 form. It localizes to the intermembrane space of the mitochondria and can regulate oxidative phosphorylation (Simpkins et al. 1998; Ye et al. 2001). Studies showing the interaction of zinc, metallothionein, and mitochondrial function further support the role of metallothionein as a regulator of cellular energy production and redox state (Maret 2000, 2003;Coyle et al. 2002). The metallothioneins have been demonstrated to hyper-accumulate in response to bacterial infection, exposure to some types of mitochondrial inhibitors (i.e., pesticides), oxidative stress, developmental changes, and growth factors (Kondoh et al. 2003, Cherian et al. 2003; Haq et al. 2003; Regala and Rice 2004). The hyper-accumulation of metallothionein indicates a possible ‘‘forcing function’’ on mitochondrial condition, which is consistent with GRP75 expression in the STA polyps. The histological observations provided no evidence of any infectious agents within the anomalous, fastgrowing tissue of P. compressa STAs. Although a viral etiology cannot be excluded, as viral particles have been observed inside zooxanthellae symbionts of Pavona danae (Wilson et al. 2005). Likewise, molecular analysis of tissue may reveal endogenous somatic mutations in the abnormal polyps causing increased cell proliferation.
542
The potential for a cause-and-effect relationship between the STAs and anthropogenic compounds persists and warrants further investigation. The close association of subskeletal fungal hyphae with active calicoblast areas within the P. compressa STA lesions is interesting to note and provides yet another factor possibly involved in the etiology of these STAs. Endolithic skeleton-boring filamentous fungi have been shown to be closely associated with the calicoblastic epithelium of scleractinians, and are especially abundant in poritids (Le Campion-Alsumard et al. 1995; Ravindran et al 2001). In histological sections of the lower third area of abnormal P. compressa polyps, a parallel was observed between the proximity of subskeletal fungal hyphae to the calicoblastic epithelium and its thickening and infiltration with chromophore cells. In vitro exposure to fungal hyphae stimulate skeletogenic calicoblastic cells isolated from branching Pocillopora damicornis coral (Domart-Coulon et al. 2004). Past episodes of soil erosion may have transported opportunistic terrestrial fungi into Kaneohe Bay, Hawaii, which interacted with P. compressa and other corals and favored the development of STAs in P. compressa. Preliminary hypotheses to investigate in the future include testing whether localized, increased deposition of skeleton acts as a physical barrier to contain internal boring microorganisms, or whether decreased density and increased porosity of anomalous skeleton favors opportunistic colonization by microborers. Additional studies will also be necessary to define the physiological function of the chromophore cells typical of the Porites genus, which have been shown here for the first time to be associated with skeletal tissue anomalies in Scleractinians. The histological, biochemical, and skeletal characterization of P. compressa STAs converge to indicate localized sustained tissue growth, and vertical extension of low-density skeleton, which did not thicken. The abnormal polyps were characterized by an increased mass of differentiated calicoblastic epithelium, thickened along the polyp basal floor and infiltrated with chromophore cells. The cause(s) for and specificity of this tissue growth remains to be determined, but our observations qualify the poritid STAs as hyperplasia. Acknowledgements We thank Laurie Pipitone, and Carolyn Magee of the Johns Hopkins Medical Institutions Radiology Research Laboratory for access to and help with X-radiography, Dr. Erik Scully of Towson University for advice on the statistical analysis of data, Jon Wingerath of the Smithsonian Institution for aid in the preparation of the petrographic sections, Michael McCaffery and the staff of the Johns Hopkins University Integrated Imaging Center for access and assistance with the Zeiss Axiovert microscope. Special thanks are due to the National Oceanic and Atmospheric Administration (NOAA) Oxford Cooperative Laboratory for access to the histology lab and to the Registry of Tumors in Lower Animals for access to the specimen collection. Anonymous reviewers are thanked for their comments which helped to improve the manuscript. This work was funded in part by a grant from NOAA (NA16OP2920).
References Alker AP, Kim K, Dube DH, Harvell DC (2004) Localized induction of a generalized response against multiple biotic agents in Caribbean sea fans. Coral Reefs 23:397–405 Bak RPM (1983) Neoplasia, regeneration and growth in the reefbuilding coral Acropora palmata. Mar Biol 77:221–227 Barnes DJ, Lough JM (1992) Systematic variations in the depth of skeleton occupied by coral tissue in massive colonies of Porites from the Great Barrier Reef. J Exp Mar Biol Ecol 159:113–128 Breitbart M, Bhagooli R, Griffin S, Johnston I, Rohwer F (2005) Microbial communities associated with skeletal tumors on Porites compressa. FEMS Microbiol Lett 243:431–436 Caplin AJ, Jackson S, Smith D (2003) Hsp90 reaches new heights. EMBO Rep 4:126–130 Chalker B, Barnes D, Isdale P (1985) Calibration of X-ray densitometry for the measurement of coral skeletal density. Coral Reefs 4:95–100 Cheney D (1975) Hard tissue tumors of scleractinian corals. In: Hildemann WH, Benedict AA (eds) Immunologic phylogeny. Plenum, New York pp 77–87 Cherian MG, Jayasurya A, Bay BH (2003) Metallothioneins in human tumors and potential roles in carcinogenesis. Mutat Res 533(1–2):201–209 Cohen AL, Smith SR, McCartney MS, van Etten J (2004) How brain corals record climate: an integration of skeletal structure, growth, and chemistry of Diploria labyrinthiformis from Bermuda. Mar Ecol Prog Ser 271:147–158 Coles SL, Seapy DG (1998) Ultra-violet absorbing compounds and tumorous growths on acroporid corals from Bandar Khayran, Gulf of Oman, Ocean Indian. Coral Reefs 17:195–198 Coyle P, Philcox JC, Carey LC, Rofe AM (2002) Metallothionein: the multipurpose protein. Cell Mol Life Sci 59:627–647 De Carlo EH, BeltranVL, Tomlinson MS (2004) Composition of water and suspended sediment in streams of urbanized subtropical watersheds in Hawaii. Appl Geochem 19:1011–1037 Domart-Coulon I, Sinclair C, Hill R, Tambutte´ S, Puverel S, Ostrander GK (2004) A basidiomycete isolated from the skeleton of Pocillopora damicornis (Scleractinia) selectively stimulates short-term survival of coral skeletogenic cells. Mar Biol 144:83–592 Downs CA (2005) Cellular diagnostics and its application to aquatic and marine toxicology. In: Ostrander GK (ed) Techniques in aquatic toxicology, vol 2. CRC, Boca Raton, pp 181– 208 Duerden JE (1903) West Indian Madreporian Polyps. In: Memoirs of the National Academy of Sciences, Vol. VIII. Washington Government Printing Office, p 439 and plate IV Gateno D, Israel A, Barki Y, Rinkevich B (1998) Gastrovascular circulation in an octocoral: evidence of significant transport of coral and symbiont cells. Biol Bull 194:178–186 Gateno D, Leon A, Barki Y, Rinkevich B (2003) Skeletal tumor formations in the massive coral Pavona clavus. Mar Ecol Prog Ser 258:97–108 Ghosh S, Gepstein S, Heikkala JJ Dumbroff BG (1988) Use of a scanning densitometer or an ELISA plate reader for measurement of nanogram amounts of protein in crude extracts from biological tissue. Anal Biochem 169:227–233 Gladfelter EH (1983) Circulation of fluids in the gastrovascular system of the reef coral Acropora cervicornis. Biol Bull 165:619– 636 Goldberg WM, Makemson JC, Colley SB (1984) Entocladia endozoica sp nov., a pathogenic Chlorophyte: structure, life history, physiology, and effects on its coral host. Biol Bull 166:368–383 Grottoli AG (1999) Variability of stable isotopes and maximum linear extension in reef–coral skeletons at Kaneohe Bay, Hawaii. Mar Biol 135:437–449 Grygier MJ, Cairns SD (1996) Suspected neoplasms in deep-sea corals (Scleractinia: Oculinidae: Madrepora spp.) reinterpreted as galls caused by Petrarca madreporae n. sp. (Crustacea: Ascothoracida: Petrarcidae). Dis Aquat Org 24:61–69
543 Haq F, Mahoney M, Koropatnick J (2003) Signaling events for metallothionein induction. Mutat Res 533:211–226 Holt SE, Aisner DL, Baur J, Tesmer VM, Dy M, Ouellette M, Trager J, Morin GB, Toft DO, Shay JW, Wright WE, White MA (1999) Functional requirement of p23 and Hsp90 in telomerase complexes. Genes Dev 13:817–826 Hunter C, Field SN (1997) Characterization of tumors in Porites corals. Am Zool 37:55 Kondoh M, Imada N, Kamada K, Tsukahara R, Higashimoto M, Takiguchi M, Watanabe Y, Sato M (2003) Property of metallothionein as a Zn pool differs depending on the induced condition of metallothionein. Toxicol Lett 30(142):11–18 Le Campion-Alsumard T, Golubic S, Priess K (1995) Fungi in corals: symbiosis or disease? Interaction between polyps and fungi causes pearl-like skeleton biomineralization. Mar Ecol Prog Ser 117:137–147 Loya Y, Bull G, Pichon M. (1984) Tumor formations in scleractinian corals. Helgol Meeresunters 37:99–112 Maret W (2000) The function of zinc metallothionein: a link between cellular zinc and redox state. J Nutr 130:1455–1458 Maret W (2003) Cellular zinc and redox states converge in the metallothionein/thionein pair. J Nutr 133:1460S–1462S Marsh J (1969) Primary productivity of reef-building calcareous red algae. Ecology 51:255–263 Marshall AT, Wright OP (1993) Confocal laser scanning light microscopy of the extra-thecal epithelia of undecalcified scleractinian corals. Cell Tissue Res 272:533–543 Merks RMH, Hoekstra AG, Kaandorp JA, Sloot PMA (2004) Polyp oriented modelling of coral growth. J Theor Biol 228:559–576 Meszaros A, Bigger C (1999) Qualitative and quantitative study of wound healing processes in the coelenterate Plexaurella fusifera: spatial temporal and environmental (light attenuation) influences. J Invertebr Pathol 73:321–331 Morse DE, Morse ANC, Duncan H (1977) Algal ‘‘tumors’’ in the Caribbean sea fan, Gorgonia ventalina. Proc 3rd Int Coral Reef Symp 1:623–629 Mullen K, Peters E, Harvell D (2004) Coral resistance to disease. In: Rosenberg E, Loya Y (eds) Coral disease and health, Springer, Berlin Heidelberg New York, pp 377–399 Nichols W, Shade P, Hunt C (1996) Summary of the O’ahu, Hawai’i, Regional Aquifer-System Analysis. U.S. Geological Survey Professional Paper 1412-A Oren U, Benayahu Y, Lubinevsky H, Loya Y (2001) Colony integration during regeneration in the stony coral Favia favus. Ecology 82:802–813 Peters EC, Halas JC, McCarty HB (1986) Calicoblastic neoplasms in Acropora palmata, with a review of reports on anomalies of growth and form in corals. J Natl Cancer Inst 76:895–912 Peters EC, Price KL, Borsay-Horowitz DJ (2005) Histological preparation of invertebrates for evaluating contaminant effects. In: Ostrander GK (ed) Techniques in aquatic toxicology, vol 2. CRC, Boca Raton, pp 653–686 Petes LE, Harvell CD, Peters EC, Webb MAH, Mullen KM (2003) Pathogens compromise reproduction and induce melanization in Caribbean sea fans. Mar Ecol Prog Ser 264:167–171
Regala RP, Rice CD (2004) Mycobacteria, but not mercury, induces metallothionein (MT) protein in striped bass, Morone saxitilis, phagocytes, while both stimuli induce MT in channel catfish, Ictalurus punctatus, phagocytes. Mar Environ Res 58:719–723 Ringuet S, Mackenzie FT (2005) Controls on nutrient and phytoplankton dynamics by storm runoff events, southern Kaneohe Bay, Hawaii. Estuaries In press Ravindran J, Raghukumar C, Raghukumar S (2001) Fungi in Porites lutea: association with healthy and diseased corals. Dis Aquat Org 47:219–228 Rohringer R, Kim WK, Samborski DJ, Howes NK (1977) Calcofluor-optical brightener for fluorescence microscopy of fungal plant parasites in leaves. Phytopathology 67:808–810 Rutherford SL, Lindquist S (1998) Hsp90 as a capacitor for morphological evolution. Nature 396:336–342 Simpkins C, Balderman S, Mensah E (1998) Mitochondrial oxygen consumption is synergistically inhibited by metallothionein and calcium. J Surg Res 80:16–21 Smith GW, Harvell CD, Kim K, Ritchie KB, James SC, Buchan KC (1998) Cellular events occurring during the pathogenesis of aspergillosis of Gorgonia species. In: Proceedings of the 7th Symposium on the Natural History of the Bahamas. Rev Biol Trop 46:205–208 Sokal RR, Rohlf FJ (1995) Biometry: the principles and practice of statistics in biological research, 3rd edn. W. H. Freeman and Co, New York Squires D (1965) Neoplasia in a Coral?. Science 148:503–505 Srokowski T, Pfeifer JD, Li J, Olson LM, Radar JS (2004). Expression and localization of GRP75 in human epithelial tumors and normal tissues. Appl Immunohistochem Mol Morphol 12:132–138 Sutherland KP, Porter J W, Torres C (2004) Disease and immunity in Caribbean and Indo-Pacific zooxanthellate corals. Mar Ecol Prog Ser 266:273–302 Wadhwa R, Taira K, Kaul SC (2002) An Hsp70 family chaperone, mortalin/mthsp70/PBP74/GRP75: what, when, and where? Cell Stress Chaperones 7:309–316 White P (1965) Abnormal Corallites. Science 150:77–78 Wielgus J, Glassom D (2002) An aberrant growth form of red sea corals caused by polychaete infestations. Coral Reefs 21:315– 316 Wilson WH, Dale AL, Davy JE, Davy SK (2005) An enemy within? Observations of virus-like particles in reef corals. Coral Reefs 24:145–148 Work TM, Rameyer RA (2005) Characterizing lesions in corals from American Samoa. Coral Reefs 24:384–390 Yamashiro H, Yamamoto M, van Woesik R (2000) Tumor formation on the Coral Montipora informis. Dis Aquat Org 41:211–217 Yamashiro H, Hirosuke O, Onaga K, Iwasaki H, Takara K (2001) Coral tumors store reduced levels of lipids. J Exp Mar Biol Ecol 265:171–179 Ye B, Maret W, Vallee BL (2001) Zinc metallothionein imported into liver mitochondria modulates respiration. Proc Natl Acad Sci USA 98:2317–2322
