Computational protocol: Relationship between Phylogeny and Immunity Suggests Older Caribbean Coral Lineages Are More Resistant to Disease

Similar protocols

Protocol publication

[…] The NCBI database has a significant number of sequences from most of the corals used in this project. The 28S rDNA region has been sequenced (as of September 2013) for 11 of the 14 species in this study (), and phylogenetic reconstructions showing similar topologies to other molecular markers, and divergence time estimation using fossils (Caryophyllia spp., Flabellum spp. and Dendrophyllidae), were available. Sequences for three species (Meandrina jacksoni GenBank KJ946355, Pseudodiploria strigosa KJ946354 and Siderastrea radians KJ946356) were generated in this project after extraction of DNA using a modified protocol from LaJeunesse et al. . A small fragment (∼3 mg) of skeleton and tissue was mixed with (of glass beads (∼200 µl, 1 mm, Ceroglass, Columbia, TN) and 600 µl of a cell lysis solution (0.2 M Tris, 2 mM EDTA, 0.7% SDS, pH 7.6) and shaken on a BioSpec (Bartlesville, OK) beadbeater for 100 seconds. Proteinase K (3 µl −20 mg/ml) was added and incubated at 65°C for 1 hour. The incubation was followed by protein precipitation with ammonium acetate (250 µl −9 M) and freezing at −20°C. The frozen extract was centrifuged (10,000 G for 15 minutes) and the supernatant removed, mixed with 600 µl of isopropanol (100%) and centrifuged (10,000 G for 5 minutes). The DNA pellet was washed with 70% ethanol, air dried, and resuspended in 75 µl of distilled water and stored at −20°C.The 28S rDNA region was amplified using the 28SROM.IFw (5′-GGCGACCCGCTGAATTCAAGCATAT-3′) and 28SDES.VRv 5′-GGTCTTTCGCCCCTATACTC-3′) primers . Reactions were performed using Perfect Taq Plus DNA Polymerase (5-Prime, Gaithersburg, MD) following the manufacturer recommended reaction composition on 2 µl of 1∶40 dilutions of the extracted DNA (final reaction volume 25 µl). Amplifications consisted of 35 cycles of 95°C, 52°C and 72°C steps, each for 30 seconds. Amplified products were cleaned with ExoSap (Affymetrix, Santa Clara, CA) and sequenced with the forward primer using Big Dye 3.1 terminator mix (Applied Biosystems/Life Technologies, Grand Island, NY) on an ABI Hitachi 3730XL genetic analyzer at UTA' Genomics Core Facility. DNA sequence chromatograms were reviewed and edited using Geneious Pro 5.0 . The resulting sequences were combined with those obtained from the NCBI data base (657 to 685 bp) and alignments were performed on ClustalW using a gap-opening penalty of 15 and a extension penalty of 6 . Phylogenies were constructed on MrBayes , using a general time-reversible model with gamma distributed rate heterogeneity (GTR+G) as substitution model, a chain length of 1,100,000 and 100,000 burn-in (phylogenies can be found in ).In order to determine the approximate age of the studied lineages, that is the extant species and their ancestors, we calculated the divergence times on the 28S rDNA based phylogeny (). Divergence times were determined with BEAST 1.7.5 , using a relaxed-clock uncorrelated lognormal allowing for nucleotide substitutions rates to vary between lineages . The tree prior used a Yule process and the model of substitution was set to gamma distributed rate heterogeneity as suggested by jModeltest 2.1.3 , with invariant sites (GTR+G+I). Node ages and posterior probabilities were estimated on a run with 10,000,000 generations and saving the topologies and parameters every 1,000 generations. In order to estimate the early divergence of the studied groups, calibrations were done using Dendrophyllidae (∼127 Mya; including sequences of Tubastrea coccinea, Cladopsamia gracilis, Leptosammia pruvoti, Endopachys grati, Enollopsamia rostrata and Balanophyllia spp.), Caryophyllia spp. (∼160 Mya) and Flabellum spp. (∼77.5 Mya) as suggested by Stolarski et al. . Phylogenetic reconstruction within BEAST was performed using Mean heights and node heights, a prior probability of 0.1 and 1,000,000 burn-in. [...] In order to test the hypothesis that closely related taxa have similar activity of constitutive immune components, the data was partitioned in three groups corresponding to each of the clades of interest: family, genus and species (). Results from the immune assays were averaged for each taxon on each level. Additionally, to obtain an integrated measure of immunity the first component scores from a principal component analysis (PC1) with all individual immune measures was used as an additional category. Principal components reduce dimensions and convert multiple variables into composite indicators , improving the analysis of immune capacity in relation to other biological/environmental parameters . The analyses were done in JMP 10.0.0 (SAS Institute, Cary, NC).In order to detect a phylogenetic signal, the immune data sets were compared against family, genera and species phylogenetic reconstructions (). The genus and family phylogenies were built with subsets of selected sequences from each group (underlined categories in ). For example, Pseudodiplora strigosa 28S sequence was used as the representative for the Pseudodiplora genus.The possibility of phylogenetic signals in coral constitutive immune levels was tested with Bloomberg's K and Moran's I measures as described by Gittleman and Kot and by Abouheif test . K assumes the data follows a Brownian motion model (BM) and compares the observed phylogenetic signal with that of the trait under the BM model. Higher K values for a particular trait represent a stronger phylogenetic signal, and zero values indicate no effect of phylogeny , . Moran's I (I) on the other hand, is a model-independent measure of autocorrelation in which the relation between the variation in the trait and the phylogenetic distance is established. In this method, the data is divided into the phylogenetic component and the trait component and correlarograms are built to determine the effect of ranks and distances . Lastly, the Abouheif test (A), modifies Moran's I to successfully detect phylogenetic signal of different traits on phylogenies with both low and high number branches. All phylogenetic signal tests were performed on R using geiger, carper, picante, adephylo and phylobase packages.A Spearman's rank order correlation index was used to determine correlations between each immune measure and estimates of number of diseases and prevalence (proportion of infected individuals in the population of a given species at a given time) for each taxonomic level (i.e. species, genus and family). Disease parameters (number of diseases and prevalence) were compiled from literature reporting epizootic events in the Caribbean from 1997 to 2005. Only manuscripts with species level resolution, and infected or diseased colonies specified as a subset of the total population surveyed were used so that number of diseases and prevalence estimates could be normalized across reports. Prevalence was estimated from the reports by combining data from all surveyed diseases (e.g. Black Band, White Band, White Plague, Yellow Band, Dark Spots, and growth anomalies) for each species. A total of 12 reports with appropriate information were found from different locations in the Caribbean, including Bermuda, Florida, Bahamas, Puerto Rico, Saint Croix, Bonaire, Yucatan-Mexico, Colombia and Venezuela () , , –. The geographic coverage and extent of the disease surveys in these reports provided a very robust data set for both number of diseases and disease prevalence, thus truly representing the disease dynamics across the region. Phylogenetic signal in the number of diseases and prevalence was estimated as described above. […]

Pipeline specifications

Software tools Geneious, Clustal W, MrBayes, jModelTest, PHYSIG, Picante, adephylo
Application Phylogenetics
Organisms Bacteria