What makes a winner? Symbiont and host dynamics determine Caribbean octocoral resilience to bleaching

Abstract

Unlike reef-building, scleractinian corals, Caribbean soft corals (octocorals) have not suffered marked declines in abundance associated with anthropogenic ocean warming. Both octocorals and reef-building scleractinians depend on a nutritional symbiosis with single-celled algae living within their tissues. In both groups, increased ocean temperatures can induce symbiont loss (bleaching) and coral death. Multiple heat waves from 2014 to 2016 resulted in widespread damage to reef ecosystems and provided an opportunity to examine the bleaching response of three Caribbean octocoral species. Symbiont densities declined during the heat waves but recovered quickly, and colony mortality was low. The dominant symbiont genotypes within a host generally did not change, and all colonies hosted symbiont species in the genus Breviolum. Their association with thermally tolerant symbionts likely contributes to the octocoral holobiont's resistance to mortality and the resilience of their symbiont populations. The resistance and resilience of Caribbean octocorals offer clues for the future of coral reefs.

Fig. 1.. The temperature at the two sites where octocorals were monitored from May 2015 until August 2017.

CMF, red line; SC2, blue line. Note that temperature data are missing at SC2 from May 2016 until December 2016 due to the loss of a temperature logger.

Fig. 2.. Bleaching card scores from CoralWatch coral health chart for each species.

(A) Proportion of M. elongata colonies in each bleaching state. (B) Proportion of M. atlantica colonies in each bleaching state. (C) Bleached (left) and recovering (right) P. dichotoma with arrows indicating bleaching card score; colony heights, ~30 cm. (D) Proportion of P. dichotoma colonies in each bleaching state. Colony color was matched to a CoralWatch coral health chart with lighter colors indicating increased bleaching from white (totally bleached) to dark brown (not bleached).

Fig. 3.. Symbiont density over time for the three host species from May 2015 to August 2017.

M. atlantica, blue circle; M. elongata, red square; P. dichotoma, yellow triangle. Error bars are SD.

Fig. 4.. The proportion of P. dichotoma colonies that harbored solely Breviolum B1 or both Breviolum B1 and Cladocopium sp. over the course of the study.

Blue, colonies with only Breviolum B1 symbionts; blue and white hatching, colonies with both Breviolum B1 and Cladocopium sp. symbionts.

Fig. 5.. Maximum likelihood phylogenetic tree of symbionts from the three host species based on sequence variation in domain V of the cp-23S rDNA gene and the flanking region of the B7Sym15 microsatellite.

Branch lengths have been transformed into a cladogram, and support values (SH-aLRT/ultrafast bootstrap) are shown directly to the left of the node they refer to. Nodes have been colored according to SH-aLRT support values, with >90% support as blue, 85 to 90% as yellow, and <85% as red.

Fig. 6.. The number of observations where symbiont genotypes within a colony changed between sampling periods for each species.

(A) M. atlantica (blue), (B) M. elongata (red), and (C) P. dichotoma (yellow). Solid bar, no change between sampling periods; Hatched, symbiont genotypes changed between sampling periods.

Fig. 7.. Cell density by genotype for each species before (May 2015), during (September 2015), and after the bleaching event (November 2015).

(A) M. atlantica, (B) M. elongata, and (C) P. dichotoma. Colors: May 2015 (blue), September 2015 (yellow), and November 2015 (red). Note that M. atlantica was not monitored in May 2015. Letters on the x axis are arbitrary names signifying the MLG designation based on the presence/absence of a given allele for each sample.