Time Course Exo-Metabolomic Profiling in the Green Marine Macroalga Ulva (Chlorophyta) for Identification of Growth Phase-Dependent Biomarkers

Abstract

The marine green macroalga Ulva (Chlorophyta) lives in a mutualistic symbiosis with bacteria that influence growth, development, and morphogenesis. We surveyed changes in Ulva's chemosphere, which was defined as a space where organisms interact with each other via compounds, such as infochemicals, nutrients, morphogens, and defense compounds. Thereby, Ulva mutabilis cooperates with bacteria, in particular, Roseovarius sp. strain MS2 and Maribacter sp. strain MS6 (formerly identified as Roseobacter sp. strain MS2 and Cytophaga sp. strain MS6). Without this accompanying microbial flora, U. mutabilis forms only callus-like colonies. However, upon addition of the two bacteria species, in effect forming a tripartite community, morphogenesis can be completely restored. Under this strictly standardized condition, bioactive and eco-physiologically-relevant marine natural products can be discovered. Solid phase extracted waterborne metabolites were analyzed using a metabolomics platform, facilitating gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS) analysis, combined with the necessary acquisition of biological metadata. Multivariate statistics of the GC-MS and LC-MS data revealed strong differences between Ulva's growth phases, as well as between the axenic Ulva cultures and the tripartite community. Waterborne biomarkers, including glycerol, were identified as potential indicators for algal carbon source and bacterial-algal interactions. Furthermore, it was demonstrated that U. mutabilis releases glycerol that can be utilized for growth by Roseovarius sp. MS2.

Keywords: Maribacter; Roseovarius; Ulva; axenic culture; bioreactors; chemosphere; cross-kingdom cross-talk; metabolite profiling.

Figure 1

(A) Representative images of Ulva mutabilis (morphotype “slender”) are shown. Left: A five-week-old axenic culture (AC) with its aberrant morphotype and unusual cell wall protrusions (scale bar = 100 μm); center: Tripartite community (TC) of young germlings and the two bacterial strains Roseovarius sp. MS2 and Maribacter sp. MS6 with normal algal development (scale bar = 100 μm); and right: A mature specimen of U. mutabilis (scale bar = 1 cm); (B) Drawing of the 25 L bioreactor used for cultivation of U. mutabilis: (1) air inlet; (2) sterile HEPA-Vent (Ø = 50 mm, Whatman) filter; (3) air outlet; (4) sampling outlet; (5) bubbling tube (Duran glass, Ø = 4 mm); (6) sampling tube (Teflon, Ø 1 mm); and (7) hose clamp to control the sampling flow; (C) Experimental design: Three bioreactors were inoculated per treatment. Samples were taken from each bioreactor in triplicate for solid phase extraction (SPE). Each extract was then divided in half for gas chromatography-mass spectrometry (GC-MS) and ultra high performance liquid chromatography-mass spectrometry (UHPLC-MS) analyses. Three injections of each sample were used for UHPLC analysis. Upon identification of pairs of retention time and m/z, the data matrix obtained was transformed and standardized before multivariate analysis (PCoA/CAP) was conducted.

Figure 2

(A) Diagram summarizing all biotic and abiotic data (i.e., biological metadata) for Ulva’s tripartite community (TC), which have been collected from the onset of the algal culture (Day 0) until the mature specimen of U. mutabilis was ready for spontaneous gametogenesis, terminating the cultivation. The life cycle of U. mutabilis status was categorized according to the inducibility of the gametogenesis: non-inducible gametogenesis (N.I.G.), artificially inducible status (A.I.G.), spontaneously inducible status (S.I.G.) (mean values ± SD (n = 3)); (B) The same as (A), but U. mutabilis was grown under axenic conditions (AC); (C) Axenicity of the Ulva culture medium and presence of bacteria was proven by PCR of the 16S rRNA gene. An agarose gel is shown. Lanes 1–9: U. mutabilis inoculated with bacteria (TC) tested in three bioreactors on days 0 (1–3), 28 (4–6) and 49 (7–9); Lane 10: PCR-negative control; Lane 11: Control of Ulva culture medium (UCM) without Ulva and bacteria; Lanes 12–14: Axenic cultures (AC) tested on days 14 (12), 28 (13) and 49 (14); (D) Denaturing gradient gel electrophoresis analysis of PCR-amplified fragments of the bacterial 16S rRNA genes in the tripartite community on Day 49 (Lane 1) has been amplified to compare it with a mix of DNA standards derived from reference bacteria (Lane St).

Figure 3

Multivariate data analysis of the exo-metabolome of Ulva mutabilis within the tripartite community (TC) and grown under axenic conditions (AC) after GC-MS measurements of 43 samples: (A) the first two canonical axes of the canonical analysis of principle coordinates (CAP) analysis are plotted. CAP analysis demonstrates the separation based on the states of gametogenesis along with the growth phases of U. mutabilis: non-inducible gametogenesis (N.I.G., orange), artificially inducible status (A.I.G., turquoise) and spontaneously inducible status (S.I.G., light gray); (B) Scaled vectors of the metabolites (ID numbers) were significant (|r| > 0.3) for the separation of the groups. The numbers refer to the metabolites in the heat map (Figure 4). Inserts show images of Ulva’s growth during the three phases, N.I.G., A.I.G., and S.I.G., selected as a priori groups for CAP analysis (bar = 1 cm).

Figure 4

Heat map of the intensities of exo-metabolites, which were correlated with the CAP axis (|r| > 0.3) and contributed to the classification of the growth phases (N.I.G., A.I.G., and S.I.G), based on GC-MS analysis. Metabolites were extracted from Ulva growth medium of the tripartite community (TC) or axenic culture (AC). Relative intensities were given in the range from 0 to 250,000. Smaller values were represented by light gray and higher values by dark grey/black boxes. Metabolites marked with a “?” had reverse match between 800 and 700, and marked with “??” a reverse match of below 700. The color code refers to the three states of the gametogenesis: non-inducible gametogenesis (N.I.G., orange), artificially inducible status (A.I.G., turquoise) and spontaneously inducible status (S.I.G., light gray). Representative ion trace chromatograms of the two biomarkers, ID #83 and ID #142, for phases A.I.G. and S.I.G. over the time course, are presented.

Figure 5

Multivariate data analysis of the exo-metabolome of Ulva mutabilis within the tripartite community (TC) and grown under axenic conditions (AC) as detected by UHPLC-ESI-ToF-MS. (A) The first two canonical axes of the CAP analysis are plotted. CAP analysis demonstrates the separation based on the states of gametogenesis along with the growth phases of Ulva mutabilis: Non-inducible gametogenesis (N.I.G., orange), artificially inducible status (A.I.G., turquoise) and spontaneously inducible status (S.I.G., light gray); (B) Scaled vectors of the metabolites (m/z) are presented significant for the separation. For better visualization, vectors of the metabolites were plotted according their retention times: (i) metabolites eluted between 0.50 and 2.00 min; (ii) 2.00 and 3.80 min; and (iii) 3.80 and 7.00 min. Here, the numbers correspond to the m/z [M+H⁺] values of the compounds, which can be also found in the heat map (Figure 6).

Figure 5

Multivariate data analysis of the exo-metabolome of Ulva mutabilis within the tripartite community (TC) and grown under axenic conditions (AC) as detected by UHPLC-ESI-ToF-MS. (A) The first two canonical axes of the CAP analysis are plotted. CAP analysis demonstrates the separation based on the states of gametogenesis along with the growth phases of Ulva mutabilis: Non-inducible gametogenesis (N.I.G., orange), artificially inducible status (A.I.G., turquoise) and spontaneously inducible status (S.I.G., light gray); (B) Scaled vectors of the metabolites (m/z) are presented significant for the separation. For better visualization, vectors of the metabolites were plotted according their retention times: (i) metabolites eluted between 0.50 and 2.00 min; (ii) 2.00 and 3.80 min; and (iii) 3.80 and 7.00 min. Here, the numbers correspond to the m/z [M+H⁺] values of the compounds, which can be also found in the heat map (Figure 6).

Figure 6

Heat map of the intensities of exo-metabolites, which were correlated with the CAP axis (|r| > 0.3) and contributed to the classification of the growth phases (N.I.G., A.I.G., and S.I.G), based on UHPLC-ESI-ToF-MS analysis. Metabolites were extracted from Ulva growth medium of the tripartite community (TC) or axenic culture (AC). Relative intensities were given in the range from 0 to 2000. Smaller values were represented by light gray and higher values by dark grey/black boxes. The color code refers to the three states of the gametogenesis: non-inducible gametogenesis (N.I.G., orange), artificial inducible status (A.I.G., turquoise) and spontaneously inducible status (S.I.G., light gray).

Figure 7

Utilization of glycerol as carbon source for Roseovarius sp. MS2. (A) Determination of glycerol in Ulva culture medium (UCM) of Ulva mutabilis under axenic conditions on Day 14 by comparison (B) with a reference standard upon derivatization with N-methyl-N-(trimethylsilyl)trifluoroacetamide. Total ion current chromatograms from a GC-MS analysis are presented (A,B). The mass spectrum (inset (A)) and the structure of the identified tris(trimethylsilyl) ether of glycerol (B) are shown. The molecular ion M^+● is not visible. (C) Glycerol was tested as the sole carbon source for Roseovarius sp. MS2 in UCM without Ulva. Growth curves of Roseovarius sp. MS2 in UCM with (white circle) and without (black circle) supplement of 1% glycerol (v:v) are plotted (mean values ± SD (n = 3)).

Figure 7

Utilization of glycerol as carbon source for Roseovarius sp. MS2. (A) Determination of glycerol in Ulva culture medium (UCM) of Ulva mutabilis under axenic conditions on Day 14 by comparison (B) with a reference standard upon derivatization with N-methyl-N-(trimethylsilyl)trifluoroacetamide. Total ion current chromatograms from a GC-MS analysis are presented (A,B). The mass spectrum (inset (A)) and the structure of the identified tris(trimethylsilyl) ether of glycerol (B) are shown. The molecular ion M^+● is not visible. (C) Glycerol was tested as the sole carbon source for Roseovarius sp. MS2 in UCM without Ulva. Growth curves of Roseovarius sp. MS2 in UCM with (white circle) and without (black circle) supplement of 1% glycerol (v:v) are plotted (mean values ± SD (n = 3)).