Fig 2.

Antibiotic exposure reduced bacterial load and host fitness. (A) Colony forming unit counts (CFUs, cells mL⁻¹) significantly differed across treatments and sampled time points (two-way ANOVA P < 0.001). Aiptasia exposed to antibiotics experienced a significant (Tukey HSD adjusted P < 0.001) reduction in bacterial load during treatment on days 41 and 55, and during the first phase of recovery on day 61. CFU counts returned to control levels on day 76 (day 0: n = 3, otherwise n = 5; day 48: CFU counts were excluded due to plate contamination). (B) Total protein concentrations differed significantly across time (two-way ANOVA P = 0.01) and experimental treatments (two-way ANOVA P < 0.001) between treated anemones and the control group. Boxplots of total protein (µg mL⁻¹) measured from each polyp (n = 5), revealed that anemone biomass significantly declined (Tukey HSD adjusted P < 0.001) initially during the priming phase when individuals were held in FASW in comparison to the control and remained significantly reduced in concentration in comparison to the control throughout the remainder of the experiment (Tukey HSD adjusted P < 0.001). (C) Exposure to antibiotics significantly reduced the asexual reproduction rate of individual anemones (n = 6) during treatment and recovery (Kruskal-Wallis P < 0.001). Boxplots of asexual reproduction rate as measured by number of observed pedal lacerates per individual per week showed that both antibiotics halted pedal laceration rate until the final recovery time point, when anemones held in ABS2 started to recover their capacity to undergo asexual reproduction. (D) Algal cell densities (cell density µg protein⁻¹) differed significantly across time and experimental treatments between treated anemones and the control group (n = 5) (two-way ANOVA P < 0.001). Boxplots of normalized symbiont density show that FASW reduced algal cell density during the priming phase. While anemones in ABS2 recovered their algal populations during treatment, anemones held in ABS1 had reduced algal population until the final recovery period. See Table S1 for all model statistics and Table S2 for means and standard errors.