Parasite-host ecology: the limited impacts of an intimate enemy on host microbiomes

Abstract

Background: Impacts of biotic stressors, such as consumers, on coral microbiomes have gained attention as corals decline worldwide. Corallivore feeding can alter coral microbiomes in ways that contribute to dysbiosis, but feeding strategies are diverse - complicating generalizations about the nature of consumer impacts on coral microbiomes.

Results: In field experiments, feeding by Coralliophila violacea, a parasitic snail that suppresses coral growth, altered the microbiome of its host, Porites cylindrica, but these impacts were spatially constrained. Alterations in microbial community composition and variability were largely restricted to snail feeding scars; basal or distal areas ~ 1.5 cm or 6-8 cm away, respectively, were largely unaltered. Feeding scars were enriched in taxa common to stressed corals (e.g. Flavobacteriaceae, Rhodobacteraceae) and depauperate in putative beneficial symbionts (e.g. Endozoicomonadaceae) compared to locations that lacked feeding.

Conclusions: Previous studies that assessed consumer impacts on coral microbiomes suggested that feeding disrupts microbial communities, potentially leading to dysbiosis, but those studies involved mobile corallivores that move across and among numerous individual hosts. Sedentary parasites like C. violacea that spend long intervals with individual hosts and are dependent on hosts for food and shelter may minimize damage to host microbiomes to assure continued host health and thus exploitation. More mobile consumers that forage across numerous hosts should not experience these constraints. Thus, stability or disruption of microbiomes on attacked corals may vary based on the foraging strategy of coral consumers.

Keywords: Coral reefs; Coralliophila; Corallivore; Gastropod; Microbial interactions; Parasite-host interactions; Parasitism.

Fig. 1

Microbiomes of feeding scars. a Sampling schematic of scar locations from outplanted corals in our manipulative experiment and natural colonies in the field used for analyses (one-factor test, factor: treatment). b Microbiome composition (beta diversity) of scar samples among treatments. Letters in legend denote significant differences (p < 0.01). c Microbiome variability (beta dispersion) of scar samples among treatments (⍺ = 0.025). d Microbiome alpha diversity of scar samples among treatments (⍺ = 0.025). Analyses were performed following rarefaction (subsampling without replacement) to a uniform sequence count of 8013 sequences per sample

Fig. 2

Average microbial community composition. Data rarefied to 8013 sequences per sample for P. cylindrica outplants without snails (left, n = 8 [Basal], 7 [Distal]) or with snails (center, n = 43 [Scar], 36 [Basal], 39 [Distal]), and natural colonies with snails (right, n = 15 [Scar], 13 [Basal], 12 [Distal]). Taxa are grouped by family, with the ten most abundant families depicted separately. All other families were pooled and depicted as ‘Other’

Fig. 3

Microbiomes of scar, basal, & distal locations on outplanted corals with snails. a Sampling schematic of locations from corals in our manipulative experiment that were subjected to snail feeding used for analyses (one-factor test, factor: location). b Microbiome composition (beta diversity) of samples by location. Letters in legend denote significant differences (p < 0.01). c Microbiome variability (beta dispersion) of samples by location. Letters in legend denote significant differences (p < 0.01). d Microbiome alpha diversity of samples by location. Letters in legend denote significant differences (p < 0.01). Analyses were performed following rarefaction (subsampling without replacement) to a uniform sequence count of 8013 sequences per sample

Fig. 4

Microbiomes of scar, basal, & distal locations on natural colonies with snail feeding. a Sampling schematic of locations from natural coral colonies that were subjected to snail feeding used for analyses (one-factor test, factor: treatment). b Microbiome composition (beta diversity) of samples by location. Letters in legend denote significant differences (p < 0.01). c Microbiome variability (beta dispersion) of samples by location. d Microbiome alpha diversity of samples by location. Letters in legend denote significant differences (p < 0.01). Analyses were performed following rarefaction (subsampling without replacement) to a uniform sequence count of 8013 sequences per sample