Fig 2: Identification and distribution of Prorocentrum cf. balticum.a Maximum Likelihood phylogenetic tree showing alignment of the internal transcribed spacer (ITS) gene region indicating a clear differentiation of the P. cf. balticum strains (blue) from described Prorocentrum species, and the close relation to undefined strains. Accession numbers, species designation, and strain codes are provided for each sequence and values at the nodes represent Bayesian posterior probability and Maximum Likelihood bootstrap support. b, c Light microscope images of P. cf. balticum showing a cell under Differential Interference Contrast (DIC) (b) and peripherally located red-autofluorescent chloroplasts (c). d Scanning Electron Microscope (SEM) image of P. cf. balticum showing distinctive dual wing-like apical projections and unique large pores with emanating large spines (white arrows). Images b, c were taken with an Upright Fluorescence Microscope (Nikon Eclipse Ni, Japan), fitted with a Cy5 620/60 nm ex 700/75 nm em filter and a monochrome camera (Nikon DS-Qi2) under 1000× magnification with oil immersion; scale bar = 5 µm. Scale bar in d = 0.5 µm. e Global biogeography of P. cf. balticum produced using the relative abundance of 18S rDNA sequences from the Tara Oceans amplicon dataset.

Fig 3: PCAIs induce the degeneration of 3D spheroid cultures. Cells were seeded in Nunclon Sphera 96-well, U-shaped-bottoms microplates and allowed to grow and form compact spheroids overnight. They were then treated with the indicated concentrations of the respective PCAIs for 48 h. The effects of the PCAIs were captured using the Nikon Eclipse Ti 100 inverted microscope (4× magnification) with Nikon DS Qi2 camera. At higher concentrations of PCAIs, the spheroids disintegrate. The effect of the PCAIs on spheroid cultures was quantified by measuring the areas of the spheroid bodies using the NIS-Element software. The results are the mean ± SEM, n = 3. Statistical significance (* p < 0.05, ** p < 0.01 and *** p < 0.001) was determined using one-way ANOVA with post hoc Dunnett’s tests.

Fig 4: Glycosylation is critical to maintaining the aqueous solubility of Mg7a(A) Enzymatic deglycosylation of Mg7a. Time-course of deglycosylation of Mg7a, as measured by relative ion count of glycoforms at 1, 2 and 4 h of treatment with α-N-acetylgalactosaminidase. At 1 h, only 2% of detected peptide was still fully glycosylated, while 91% was fully deglycosylated and 7% was partially deglycosylated (i.e., Mg7a with the loss of 2 GalNAc moieties). At 2 h, no fully glycosylated peptide was detected, and 94% was fully deglycosylated.(B) Purification of deglyco-Mg7a by HPLC on a Gemini NX-C18 column (250 × 4.6 mm; particle size, 3 μm; pore size, 110 Å; Phenomenex) using a gradient of 5 to 50% solvent B (90% MeCN and 0.05% TFA) over 45 min at a flow rate of 1 mL/min. Mass of eluting peaks was confirmed by MALDI-MS (linear mode). Theoretical monoisotopic [M+1H]1+ of Mg7a (loss of 2 GalNAc) = m/z 6705; Mg7a (loss of 3 GalNAc) = m/z 6503. Trace from Mg7a (fully glycosylated), run under the same conditions, is shown for comparison.(C) Fibrillar precipitate formed upon reconstitution of deglycosylated Mg7a in water at a concentration of ∼100 μM. Image was acquired on a Nikon DS-Qi2 microscope; scale bar = 100 μm.(D) Activity of Mg7a and analogs on F11 cells. Data are expressed as mean ± SEM (n = 3).

Fig 5: Immunofluorescence staining of organs from 10-week-old female C57BL/6NRj mice. Lung, aorta, kidney and skin stained with Alexa Fluor 488-anti-heparan sulfate (green), Alexa Fluor 647-anti A1M (red) and Hoechst (blue). Images were obtained on a Widefield Epifluorescence Ti2 microscope equipped with a Nikon DS-Qi2 camera. Magnification 10x (left column) and 40x (right column). Scale bars, 100 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)