Phenotypic plasticity and morphological integration in a marine modular invertebrate

Abstract

Background: Colonial invertebrates such as corals exhibit nested levels of modularity, imposing a challenge to the depiction of their morphological evolution. Comparisons among diverse Caribbean gorgonian corals suggest decoupling of evolution at the polyp vs. branch/internode levels. Thus, evolutionary change in polyp form or size (the colonial module sensu stricto) does not imply a change in colony form (constructed of modular branches and other emergent features). This study examined the patterns of morphological integration at the intraspecific level. Pseudopterogorgia bipinnata (Verrill) (Octocorallia: Gorgoniidae) is a Caribbean shallow water gorgonian that can colonize most reef habitats (shallow/exposed vs. deep/protected; 1-45 m) and shows great morphological variation.

Results: To characterize the genotype/environment relationship and phenotypic plasticity in P. bipinnata, two microsatellite loci, mitochondrial (MSH1) and nuclear (ITS) DNA sequences, and (ITS2) DGGE banding patterns were initially compared among the populations present in the coral reefs of Belize (Carrie Bow Cay), Panama (Bocas del Toro), Colombia (Cartagena) and the Bahamas (San Salvador). Despite the large and discrete differentiation of morphotypes, there was no concordant genetic variation (DGGE banding patterns) in the ITS2 genotypes from Belize, Panama and Colombia. ITS1-5.8S-ITS2 phylogenetic analysis afforded evidence for considering the species P. kallos (Bielschowsky) as the shallow-most morphotype of P. bipinnata from exposed environments. The population from Carrie Bow Cay, Belize (1-45 m) was examined to determine the phenotypic integration of modular features such as branch thickness, polyp aperture, inter-polyp distance, internode length and branch length. Third-order partial correlation coefficients suggested significant integration between polypar and colonial traits. Some features did not change at all despite 10-fold differences in other integrated features. More importantly, some colonial features showed dependence on modular features.

Conclusion: Consequently, module integration in gorgonian corals can be shifted, switched or canalized along lineages. Modular marine organisms such as corals are variations on a single theme: their modules can couple or decouple, allowing them to adapt to all marine benthic environments.

Figure 1

Pseudopterogorgia bipinnata (Verrill). A. Shallow-water bushy morphotype (= P. kallos [Bielchowsky]) 2 m in water depth; B. Typical regularly branched morphotype (17 m). C. Long-branched deep morphotype (17 m). Detail of dry colonies below correspond to fragments of the same colonies above (Carrie Bow Cay, Belize, 2003).

Figure 2

Phylogenetic hypotheses of gorgonian corals (phylograms). A. Nuclear DNA (ITS1–5.8s-ITS2, 731 bp) maximum likelihood tree using the best-fit model (TVMef+G) selected by AIC. Arrows indicate the species ITS2 secondary structure with their different morphologies. B. Mitochondrial DNA (MSH1 776 bp), maximum likelihood tree using the best-fit model (K81uf+I) in PAUP* selected by Akaike Iinformation Criterion-AIC in Modeltest. Above node support values are from 100 bootstrap replicates (>75 only). Maximum parsimony analyses yielded the same results. Scale in substitutions per site. B = Bahamas, CBC = Carrie Bow Cay, Belize, I = Typical Intermediate morphotype, D = Deep water morphotype. SEM scale = 20 μm. Additional mtDNA sequences from [4].

Figure 3

Box plots from the distribution and variance of ten measurements per colony for Branch Internode (A), Branch length (B), Branch Thickness (C), Polyp aperture (D), and Interpolyp Distance (E). The median line is inside the 25th and 75th percentiles with external error bars at the 10th and 90th percentiles.

Figure 4

Graphical model for the morphological integration of traits in the phenotypic plasticity of Pseudopterogorgia bipinnata. Bold edges show strong bonds after using third-order PCC, dashed lines depict only background zero-order bivariate correlations existing among all contrast variates (Table 1). Branch Internode (I), Branch lenght (B), Branch Thickness (T) and Polyp aperture (A).