Sample collection

To assess the effect of the environment on sponge biochemical and energetic content, individuals (n = 2-21) of Agelas conifera, Agelas tubulata, Amphimedon compressa, Aplysina cauliformis, Niphates amorpha, Niphates erecta, and Xestospongia muta were collected at 15 m depth from 2-4 sites at each of 4 locations across the broader Caribbean Basin: Belize (May 2017), Curaçao (March 2017) , Grand Cayman (January 2018), St. Croix (August 2018). Although not every species was found in every region, target species were generally common and were selected to represent a range of antipredator chemical defenses (defended, undefended, variably defended), relative microbial abundance (HMA=high microbial abundance, LMA=low microbial abundance) and relative abundance of photoautotrophic symbionts (high, intermediate, low).

Tissue samples were excised from individual sponges in situ using scissors, placed into individual resealable plastic bags, kept submerged in seawater in a shaded cooler, and returned to shore where they were frozen at -20°C for transport. In the lab, sample wet mass and volume were recorded, and samples were freeze dried to determine dry mass, and ground to a powder.

Proximate Biochemical Analysis

The proximate biochemical composition (PBC) of sponge tissue was quantified as described in Clayshulte Abraham et al. (in press). Briefly, carbohydrates were extracted from 10 mg of ground freeze-dried tissue in 5% trichloroacetic acid (TCA) and concentration was determined using the phenol-sulfuric acid method in microplate format, as described in Masuko et al. (2005). Absorbance was measured using a BioTek Synergy HT Multi-Detection Microplate Reader. The concentration of carbohydrates in samples was calculated relative to a glucose standard curve.

Protein was extracted from ground freeze-dried tissue in 1 M sodium hydroxide (NaOH) and the soluble protein concentration was analyzed using the Bradford Method (Bradford, 1976). Absorbance was measured using an Eppendorf Biophotometer. Soluble protein concentration in sponge samples was calculated relative to a standard curve using Bovine Serum Albumin (BSA).

Lipids were extracted using a modified version of the protocol described by Freeman et al. (1957). Briefly, ground freeze-dried tissue was sonicated, chloroform:methanol solution, and filtered into a conical tube containing distilled water. The organic layer was then pipetted into a pre-weighed vial, and this process was repeated three times. The organic solvent was then evaporated via vacuum centrifugation, and the final mass of dry lipid was recorded.

Ash was measured using methods in McClintock et al. (1991). Briefly, ground freeze-dried sponge tissue was placed into a pre-weighed foil pan, ashed at 500°C in a muffle furnace for 5 h, and the final mass of ash was recorded.

Refractory material was calculated by subtracting the combined masses of all of the measured components (carbohydrates, soluble protein, lipids, and ash) from the total sponge tissue mass to obtain ash-free dry weight (AFDW).

The proportion of each biochemical constituent was calculated by dividing each constituent’s mass by the sample’s AFDW. The total sponge tissue energetic content was calculated by multiplying the proportional dry mass of each biochemical constituent by the kilojoule (kJ) coefficients detailed in Gnaiger & Bitterlich (1984). For purposes of energetic calculations, refractory material was assumed to consist predominantly of insoluble protein.

To quantify the concentration of total sponge extract obtained from each sample, ground freeze-dried sponge tissue was extracted in methanol:methylene chloride and sonicated. This was repeated twice, and the three extracts were combined in a pre-weighed vial. The solvent was removed via vacuum centrifugation to yield extract dry mass. Natural extract concentrations were calculated by based on initial volume to dry mass relationships for each sample.