Holding time or fixative formulation has no obvious effect on histology of Porites evermanni and Montipora capitata

Abstract

Collection of coral for histologic examination requires holding of samples in seawater for a time before they are fixed for histologic processing. This could adversely affect the interpretation of morphologic changes during histologic examinations. We evaluated the microscopic morphology of Porites evermanni and Montipora capitata held (0–120 minutes) in seawater prior to fixation in Z-Fix formulated with raw or artificial seawater. We saw no evident effects of treatments on microscopic morphology. However, among 88 statistical comparisons, and after accounting for false discovery rate, holding time prior to fixation was associated with a significant increase in degree of mucosity of basal body walls.

Coral reefs are under multiple threats including climate change, land based pollution, and more recently disease. ¹¹ An example is the incursion of a mass mortality event attributed to stony coral tissue loss disease in Florida in 2014 affecting multiple species of corals on a broad geographic scale on reefs in the western North Atlantic Ocean. ⁶ Histology is playing a progressively more important role in aiding our understanding of potential causes of mortality in corals, including stony coral tissue loss disease. ⁴ However, the presence of lesions in “apparently normal” corals has raised concerns that processing or collection artifacts could influence or misguide interpretation of tissues at the microscopic level.^3,6 Potential artifacts could be incurred by the holding time prior to fixation. For example, in fish, gills and intestines are prone to rapid autolysis making it imperative that these organs are harvested as close to the time of death as possible to ensure maximum integrity for histologic examination. ¹ Commonly, diseased corals for histologic investigations are collected by scuba diving, requiring investigators to hold corals in collection containers in seawater from minutes to hours prior to fixation. ¹⁰ Moreover, there are variations on how fixatives are formulated, with some investigators using fixative diluted in raw seawater, whereas others use artificial seawater (ASW). Such holding times and fixative formulations could introduce microscopic morphologic changes in coral samples that could be interpreted as lesions. To address this, we tested fixative formulations with raw and ASW, varied holding times on Porites evermanni and Montipora capitata prior to fixation, and evaluated effects on microscopic morphology. We hypothesized that holding these 2 coral species in seawater for up to 2 hours prior to fixation would have no noticeable effects for routine diagnostics.

Coral colonies originated from the Division of Aquatic Resources Hawaii Coral Restoration Nursery located on Sand Island, Oahu, Hawaii (21.303773N, -157.89027W). Colonies were held in large (~10,000 L) tanks supplied with flow-through seawater with salinity ranging from 32 to 35 parts per thousand (ppt) filtered by sand prior to ingress into tanks. We chose P. evermanni and M. capitata, because they were available, and these genera dominate reefs in the Indo-Pacific thus making them suitable test subjects reflective of what investigators might encounter in the region. Apparently normal coral colonies growing in tanks and bereft of gross lesions were sectioned with a band saw in ~2- to 4-cm² fragments. Fragments were held outdoors in the shade at ambient temperature (27°C–28°C) in Whirl-Pak (Thermo-Fisher, Waltham, Massachusetts, USA) sterilized bags in seawater. At 0, 10, 30, 60, 90, and 120 minutes, 3 fragments from each species were placed in Z-Fix (Anatech, Sparks, Nevada, USA) diluted 1:5 either in raw seawater from flow-through tanks or ASW (Instant Ocean, Blacksburg, Virginia, USA) prepared according to manufacturer instructions with deionized water obtained from a Milli-Q IQ7000 system (Millipore, Burlington, Massachusetts, USA) and adjusted to 32 to 35 ppt salinity to simulate seawater conditions in tanks (Supplemental Figure S1). These holding times were chosen to simulate potential dive times and conditions routinely encountered when doing coral disease surveys with standard 80 ft ³ (2.3 m³) aluminum air cylinders. Three weeks post-fixation, corals were decalcified with Cal-Ex II (Fisher Scientific, Waltham, Massachusetts, USA), processed using standard histology methods, and stained with hematoxylin and eosin.

Slides were blinded and examined microscopically by 4 independent investigators (TMW, AW, CS, and MMD) to gauge any detectable morphologic changes in tissue appearance attributable to either type of fixative or time held in seawater prior to fixation. Then, various features were assigned numerical scores by 2 investigators (CS and AW) working separately. Once histology assessments were made and scoring assigned, slides were unmasked. In total, we examined 72 fragments comprising 36 M. capitata and 36 P. evermanni. The degree of mucus production (ie, mucosity) at surface (SBW) and basal body walls (BBW) were scored 1 to 3 as follows: 1—mild, 1% to 25% mucocytes distended with mucus, cell borders still distinct, or mucus covering 1% to 25% surface; 2—moderate, 26% to 75% mucocytes distended, cell borders sometimes indistinct, or mucus covering 26% to 75% surface; and 3—severe, >75% mucocytes distended with mucus, cell borders indistinct, or mucus covering >75% surface (Fig. 1a–c). Symbiodiniaceae degeneration and necrosis were rare in all fragments with less than 5% of observed Symbiodiniaceae; typically 0, never exceeding 2 per high-powered field (2.37 mm²), and thus categorized as present/absent. Symbiodiniaceae degeneration (Fig. 1d) was defined as endosymbiont swelling, vacuolation, and/or pallor sometimes accompanied by enlargement of the symbiosome. Symbiodiniaceae necrosis (Fig. 1d) was defined as endosymbiont shrinkage with distortion of spherical shape, cytoplasmic hypereosinophilia, nuclear pyknosis, and/or ghost cells.^3,6 Tissue necrosis was categorized as present/absent, and when present, the affected tissue type including SBW, BBW, or mesenterial filament (MF) was recorded. Tissue necrosis (Fig. 2a–b) was defined as loss of tissue architecture or loss of cellular detail with nuclear pyknosis, lysis, dissociation, or karyorrhexis and hypereosinophilia and homogenization of the cytoplasm.^3,6 Fragmentation (ie, breakages in tissues without cell lysis or loss of cellular detail) was categorized as present/absent. Endolithic algae and/or fungi were identified by their septate or non-septate branching filaments within coral skeletal spaces (Fig. 2c).

Figure 1.

Histologic mucosity scores (a–c) and histopathology observed in study corals (d). B, basal body wall; E, epidermis of surface body wall; G, gastrodermis of surface body wall; M, mesenterial filament. Hematoxylin and eosin. (a) Mucosity score 1 in Porites evermanni. Note isolated islands of mucocytes distended with mucus in basal body wall gastrodermis. The presence of melanin granules (arrowhead) is a normal finding in the species. (b) Mucosity score 2 in P. evermanni; as in (a) but increased to confluent mucus layers in basal body wall gastrodermis. (c) Mucosity score 3 in P. evermanni; note marked increased in the thickness of basal body wall gastrodermis with >75% of tissue comprising mucocytes distended with mucus. (d) Symbiodiniaceae degeneration and necrosis in Montipora capitata. A degenerate endosymbiont (arrow, upper inset) is swollen with vacuolated cytoplasm and surrounded by increased clear space indicating swelling of the symbiosome. A necrotic endosymbiont is shrunken and misshapen with a pyknotic nucleus (lower inset).

Figure 2.

Histopathology observed in study corals. B, basal body wall; M, mesenterial filament; G, gastrovascular cavity. Hematoxylin and eosin. (a) Basal body wall necrosis (N) in Montipora capitata as indicated by area of cellular dissociation and pyknotic nuclei. (b) Basal body wall necrosis (N) in M. capitata as indicated by an area of lost cellular detail containing remnant pyknotic nuclei and cytoplasmic debris. (c) Endolithic fungi or algae (E) in Porites evermanni. Hyphae with obscure septae are within decalcified skeletal spaces (S) surrounded by eosinophilic granular calicodermal cells (arrow) characteristic of Porites.

Because data did not fit assumptions of normality as evaluated by the Shapiro-Wilk test, nonparametric tests were used. For the 2 nonbinary measurements (degree of mucus production on SBW or BBW), we compared effects of observer (A and B) or species using Wilcoxon t-test, whereas effects of different holding times prior to fixation were compared using Kruskal-Wallis ANOVA. For binary measurements (presence/absence of Symbiodiniaceae degeneration and necrosis, tissue necrosis, or tissue fragmentation), we used logistic regression to compare observer, species, seawater type, and holding times prior to fixation along with all possible interactions. Level of significance for all statistical tests was 0.05, and this cutoff was adjusted to account for false discovery rate that may result from running multiple statistical tests on the same sample set. ² All statistical analyses were done with R. ⁸

Of the 88 statistical comparisons, only mucosity of BBW increased significantly with increased holding times prior to fixation; all other parameters were not significantly affected by the observer, holding time, fixative time, or coral species (Supplemental Table S1). Mucosity scores varied widely among samples but were not related to the type of seawater used for fixation (Supplemental Table S2). Symbiodiniaceae degeneration and necrosis were consistently observed (Supplemental Table S2) but were, in all cases, very rare with 0–2 per high-powered fields (2.37 mm²), affecting less than 5% of observed Symbiodiniaceae and typically identified in the SBW gastrodermis (Fig. 1d). Tissue fragmentation without necrosis of SBW, BBW, and MF was consistently noted at the margin of biopsies in P. evarmanni and occasionally in M. capitata (Supplemental Table S2). Tissue necrosis involving the margin of biopsies was often observed in both species but was presumptively attributed to an artifact of traumatic sampling and therefore not included in results. Necrosis not involving biopsy margins, which was rare (Supplemental Table S2) and observed only in M. capitata, involved all tissue types including SBW, BBW, and MF and was associated with endolithic organisms in one biopsy, but not in other affected biopsies. Endolithic microorganisms, predominantly fungal hyphae or algae, were identified in both species, variably represented across all biopsies, and infrequently associated with hyaline lamellae deposition or calicoblast hypertrophy. ³

We saw no clear difference between the morphology of control corals (those fixed immediately) and those held up to 120 minutes in seawater nor was there any evident difference between fragments stored in Z-Fix made with artificial versus raw seawater. Symbiodiniaceae degeneration and necrosis have been associated with thermal bleaching and other coral diseases ³ and are frequently evaluated by diagnosticians. Although Symbiodiniaceae degeneration and necrosis were observed in an increasing proportion of P. evarmanni biopsies with increasing holding times (Supplemental Table S2), we did not observe an increase in the proportion of individually affected Symbiodiniaceae within biopsies, which remained at low levels or 0–2 per high-powered fields (2.37 mm²) for all biopsies. Increased numbers of necrotic symbionts above such background levels would be of more diagnostic value when evaluating coral histology. ³ Symbiodiniaceae degeneration and necrosis were observed in all M. capitata biopsies regardless of the holding time or type of seawater used for fixation (Supplemental Table S2), indicating these conditions had no major effect on Symbiodiniaceae histology. Given that there is regular turnover of endosymbionts in scleractinian corals, finding degenerating endosymbionts in tissue sections at low levels is expected. ⁵

Tissue necrosis in corals is associated with tissue-loss diseases, tissue invasion by endolithic microorganisms, and growth anomalies, ³ and is therefore highly scrutinized by diagnosticians. However, tissue necrosis not related to sampling artifacts at biopsy margins was not observed in P. evarmanni and appeared to be observed in decreasing frequency in M. capitata biopsies with increased holding times (Supplemental Table S2). The practical significance of these findings indicates there is no major effect of holding time on tissue necrosis. Tissue fragmentation without necrosis is often observed in coral biopsies with unknown significance. Given this feature was noted in all P. evarmanni biopsies (Supplemental Table S2), this could reflect a species-specific response to injury (ie, sampling). There was no clear pattern of increasing frequency of this feature in M. capitata with increasing holding times nor was there a major difference between those fixed with artificial versus raw seawater, further supporting fragmentation is likely not related to these aspects of sample handling in M. capitata. To reduce the possibility of misinterpretation of necrosis and fragmentation at a tissue margin, the application of surgical ink on biopsy margins prior to decalcification can microscopically highlight areas prone to developing microscopic artifacts. ⁹

An increasing degree of mucus production in the BBW with increasing holding times was the only association of statistical significance. Given that we adjusted P values for false discovery rate, this association is likely real. However, these findings highlight the differences between diagnostic and quantitative pathology. From a practical standpoint, it is unlikely that a diagnostician examining tissues would note these differences in scores if applied as used in this assessment. The degree of mucus production has been associated with coral diseases and in response to irritation, ³ and there are likely species-specific differences in using this as a defense mechanism. Although is it unclear why a difference was observed in BBW mucus production but not in SBW mucus production, this finding further supports the limited effect of holding time on degree of mucus production.

Altogether the histologic abnormalities documented in this study were mild and nonspecific changes that at a low level can be considered “background” findings in apparently healthy corals. For example, endolithic organisms including algae, fungi, and cyanobacteria are believed to play an important role in the nitrogen cycle and help with recovery in bleached coral, ⁷ although an overgrowth of endolithic fungi is associated with necrosis and fragmentation of coral tissue for Siderastrea. ¹² Low number of ciliates may be seen in a healthy and diseased coral as opportunistic feeders. ³ Granular amoebocyte infiltration in P. evarmanni is thought to be part of their immune system and can be observed in both injured and apparently healthy fragments. ³ In our study, we observed 0 to 2 cells of Symbiodiniaceae degeneration and necrosis per high-power field (2.37 mm²) in healthy fragments, which might be reflective of the host-algae endosymbiont population regulation pathway. ⁵ No other microscopic lesions documented in diseased scleractinian corals were identified, ³ including those considered nonspecific responses to injury such as vacuolation of gastrodermal epithelial cells. Therefore, histologic changes interpreted to represent various disease processes in corals are unlikely influenced by fixation methods. We conclude that for these 2 coral species, holding in seawater at 24°C to 28°C while protected from sun exposure prior to fixation for up to 120 minutes will likely not lead to artifacts that could confound microscopic interpretation. This protocol might be applicable to test effects of sample processing on histology for other coral species, for instance, those from the western Atlantic Ocean.

Supplemental Material

sj-pdf-1-vet-10.1177_03009858241309403 – Supplemental material for Holding time or fixative formulation has no obvious effect on histology of Porites evermanni and Montipora capitata

Supplemental material, sj-pdf-1-vet-10.1177_03009858241309403 for Holding time or fixative formulation has no obvious effect on histology of Porites evermanni and Montipora capitata by Thierry M. Work, Chutimon Singhakarn, Amy Webb, Norton Chan and Michelle M. Dennis in Veterinary Pathology