Zebrafish-based identification of the antiseizure nucleoside inosine from the marine diatom Skeletonema marinoi

Abstract

With the goal of identifying neuroactive secondary metabolites from microalgae, a microscale in vivo zebrafish bioassay for antiseizure activity was used to evaluate bioactivities of the diatom Skeletonema marinoi, which was recently revealed as being a promising source of drug-like small molecules. A freeze-dried culture of S. marinoi was extracted by solvents with increasing polarities (hexane, dichloromethane, methanol and water) and these extracts were screened for anticonvulsant activity using a larval zebrafish epilepsy model with seizures induced by the GABAA antagonist pentylenetetrazole. The methanolic extract of S. marinoi exhibited significant anticonvulsant activity and was chosen for bioassay-guided fractionation, which associated the bioactivity with minor constituents. The key anticonvulsant constituent was identified as the nucleoside inosine, a well-known adenosine receptor agonist with previously reported antiseizure activities in mice and rat epilepsy models, but not reported to date as a bioactive constituent of microalgae. In addition, a UHPLC-HRMS metabolite profiling was used for dereplication of the other constituents of S. marinoi. Structures of the isolated compounds were elucidated by nuclear magnetic resonance and high-resolution spectrometry. These results highlight the potential of zebrafish-based screening and bioassay-guided fractionation to identify neuroactive marine natural products.

Fig 1. Larval locomotor activity (expressed in actinteg units) of S. marinoi extracts tested on zebrafish larvae.

Hexanic extract (SM01), dichloromethane extract (SM02), methanolic extract (SM03) and aqueous extract (SM04) were tested at 10, 30 and 100 μg/mL concentrations. PTZ was used at 20 mM as the proconvulsant agent. Controls are described as “-” for the vehicle control and “+” for the positive control. Data are represented as Actinteg mean ± SEM (n ≥ 5). Statistical analysis was performed by one-way ANOVA with Dunett’s test to compare samples with positive control, with P values of <0.001 (***), <0.0001 (****).

Fig 2. ELSD chromatograms for the methanolic extract before and after enrichment, and its fractionation by preparative chromatography.

(A) UHPLC chromatogram of the crude methanolic extract exhibiting mainly unretained polar metabolites. (B) UHPLC metabolite profiling after enrichment by VLC to remove very polar residues. (C) Preparative chromatography for the fractionation of the enriched extract highlighting the isolation of constituent 1.

Fig 3. Effects of inosine on larval locomotor activity after 30 min tracking period (expressed in actinteg).

Inosine (1) was tested at 10, 50 and 100 μg/mL. PTZ was used at 20 mM as proconvulsant agent. Controls are indicated as “-” for the vehicle control and “+” for the positive control. Data are represented as Actinteg mean ± SEM (n ≥ 3). Statistical analysis was performed by one-way ANOVA with Dunett’s test to compare samples with positive control, with P values of <0.05 (*), <0.01 (**).