Bacterial diketopiperazines stimulate diatom growth and lipid accumulation

Abstract

Diatoms are photosynthetic microalgae that fix a significant fraction of the world's carbon. Because of their photosynthetic efficiency and high-lipid content, diatoms are priority candidates for biofuel production. Here, we report that sporulating Bacillus thuringiensis and other members of the Bacillus cereus group, when in co-culture with the marine diatom Phaeodactylum tricornutum, significantly increase diatom cell count. Bioassay-guided purification of the mother cell lysate of B. thuringiensis led to the identification of two diketopiperazines (DKPs) that stimulate both P. tricornutum growth and increase its lipid content. These findings may be exploited to enhance P. tricornutum growth and microalgae-based biofuel production. As increasing numbers of DKPs are isolated from marine microbes, the work gives potential clues to bacterial-produced growth factors for marine microalgae.

Figure 1

Bacillus cereus clade of bacteria stimulates P. tricornutum growth during co-culture. A, Bacillus cereus clade members, B. cereus sp. 6A1, B. thuringiensis israelensis (Bti), B. thuringiensis sp. 4A4, and B. thuringiensis sp. 4Q7 as well as Bacillus subtilis clade members, B. subtilis and B. amyloliquefaciens, are tested for their ability to stimulate P. tricornutum (Pt) growth. Pt cells (y-axis) were counted at Day 6 of co-culture. Each culture condition has three biological replicates. Error bars are ±1 SD (Standard Deviation). “*” represents significant difference from “Pt only” (two-tailed Student’s t test: P < 0.05). To control for carryover LB into the co-culture by the bacteria, LB of equivalent volume (100 µL) was added to the L1 medium in the “Pt only” control. B, A tube containing co-culture of Pt and Bti and a tube with Pt in axenic condition (both at Day 5). The brown color tube indicates higher Pt cell density. C, Growth curve of Pt when co-cultured with S. aureus, B. thuringiensis sp. israelensis, both bacteria combined, or under axenic condition (Pt only). The bacteria are added to the Pt culture at Day 1. The Pt cell number (Y-axis) was quantified at each day (X-axis). D, Microscopic images of Pt cells (left) and Pt cells in co-culture (right; both at Day 4). A chain of Bti spores is indicated by an arrow. Both images are at the same magnification; scale bar is 20 µm.

Figure 2

The mother cell lysate of Bti contains a heat-labile stimulator. A, Pt cell counts (left Y-axis) and Bti spore counts (right Y-axis) in co-cultures, showing correlation between diatom cells and bacterial spores. B, Pt cell counts in culture with Bti mother cell lysate (spores removed), the Bti spores, and Pt only, respectively. C, Pt cell counts in cultures containing sporulated Bti mother cell lysate, nonsporulating Bti cell lysate, and the axenic control (Pt only). Cell counts were done at Day 7 of co-culture. D, Pt cell counts in culture with Bti mother cell lysate, autoclaved B. thuringiensis mother cell lysate, or the axenic control (Pt only). Each culturing condition contained three replicates. Error bars are ±1 SD.

Figure 3

Identification of two DKPs as the growth simulators from B. thuringiensis. A, The growth enhancing effect of 1 was significant in Days 3 and 7 cultures. B, Significant stimulating effect of 2 was shown in Day 7 culture. Each culturing condition has three biological replicates. Error bars are ±1 SD. “*” indicates significant difference (two-tailed Student’s t test: P < 0.05) between each DKP-stimulated culture and the control. For the controls, MeOH/H₂O was added to the L1 medium at the same percentage as the DKP solution. C, Structures of cyclic-l-Pro-l-OMet (1) and cyclic-l-Val-ΔAla (2). D, Key COSY and HMBC correlations of 1 and 2.

Figure 4

The bacteria-derived DKPs (cyclic-l-Pro-l-OMet and cyclic-l-Val-ΔAla) stimulate Pt growth in a dose-dependent manner. A, Pt cell numbers when growing in culture containing different concentrations of cyclic-l-Pro-l-OMet (1) and cyclic-l-Val-ΔAla (2). Cell counts were done at Day 7, when effective stimulation was established for both DKPs. B, Solutions containing mixed pair of single l-amino acids (l-Pro/l-Met AA and l-Val/l-Ala AA) failed to stimulate Pt growth at different concentrations. Cell counts were done at Day 8 of culture. Each culturing condition has three biological replicates. Error bars are ±1 SD. C, Absolute quantification of cyclic-l-Pro-l-OMet (1) and D, Absolute quantification of cyclic-l-Val-ΔAla (2) in Bti, B. thuringiensis sp.4A4, B. thuringiensis sp.4Q7, B. thuringiensis sp.407. The calibration curve is also shown.

Figure 5

Stimulated Pt cells have higher lipid content and higher yield of beneficial fatty acids. A, Fatty acid fractions of 7-d-old Pt culture grown in L1 medium (Pt only) or L1 medium containing Bti lysate measured by GC–MS. Values represent normalized fatty acid production per cell of each culture as peak area ratio versus the internal standard. Cells of Pt only averaged 4.35E + 06 Pt cells per milliliter, whereas cells of Pt + Bti lysate averaged 1.34E + 07 cells per milliliter. B, DIC microscopy images (top) and fluorescent microscopic images (bottom) of BODIPY-stained Pt cells at stationary phase (12 d). All images are at the same magnification; scale bar is 10 µm. C, Relative Fluorescence/cell (Y-axis) is equal to Integrated Density (see “Methods” section) based on BODIPY staining of Pt cells at stationary phase. Black horizontal bars represent the average Integrated Density of cells in the experimental condition. N = 23 (Pt only), 30 cells (Pt + Bti lysate), and 47 (Pt + cyclic-l-Pro-l-OMet).