Fungal mycelia and bacterial thiamine establish a mutualistic growth mechanism

Heat maps of B. subtilis instantaneous velocity analyzed by track the position of each cell moving along hyphae from Video 1. We tracked the positions of each cell moving along hyphae and from the instantaneous position of the center of mass of a cell. We generated heat maps of the instantaneous velocity, where darker colors represent slower speeds and lighter colors represent higher velocities, respectively. The heat maps indicate that there is a weak oscillation in the instantaneous velocity over time (Fig 1C). The reason for the oscillations that we see in the instantaneous velocity of bacteria moving along hyphae may be the result of stick-slip motion mediated by the flagella. Because of the low water content and narrow gaps between hyphae and agar, the cells may momentarily become wedged in tight gaps; however, the flagellar motors may exert sufficient force to free the stuck cells.

(A) Time-lapse images of B. subtilis (green) monoculture for 60 s on the minimum agar media. Scale bar: 50 μm. (B) A. nidulans hyphae (DIC) surrounded by moving B. subtilis (green) from Video 4. (C) Image sequence of B. subtilis flagella mutant (Δhag) flow (arrows) along A. nidulans hyphae. Kymograph of the Δhag along the hyphae to the tip (white arrows) from Video 5. Total 10 min. Scale bar: 50 μm.

(A) Time-lapse proliferation of B. subtilis Δhag (green) and A. nidulans (DIC) after 17 h of co-culture. Scale bar: 100 μm. (B) Bright field (left) and green fluorescent (right) images of colonies of B. subtilis (168; WT or Δhag) mono- or co-culture with A. nidulans. Aerial growing hypha at the middle of colony disturb to detect the fluorescent signals in the co-culture. Scale bar: 5 mm. (C) Time-lapse proliferation of B. subtilis (3610; WT) mono- or co-culture with A. nidulans, and B. subtilis (Δhag) monoculture. Scale bar: 100 μm.

Each microorganism was cultured in minimum medium with indicated carbon source for 5 d. Extracellular hydrophobic metabolites in culture supernatant were extracted with acidified ethyl acetate and analyzed by LC-PDA-ESI/MS. Contour maps were constructed from the absorption intensity obtained by LC-PDA analysis. In each monoculture with different carbon source, we see little change in the EHM profiles of B. subtilis. On the other hand, the profiles of A. nidulans are affected by the different carbon sources, which consistent with the previous reports about the regulation of secondary metabolism. EHM in co-culture are similar to those in B. subtilis monoculture regardless of carbon source. The fungal and bacterial cells co-cultured as described above for 5 d in six-well plates. Supernatant of mono- or co-culture was collected by centrifugation at 15,000g for 10 min, and acidified by 1/100 volume of 2 M HCl. Equal volume of ethyl acetate was added to the acidified supernatant, stirred for 1 h and centrifuged at 1,000g for 10 min. The ethyl acetate fraction was collected to other tubes, and lyophilized. Resulting pellet was dissolved in 95% methanol and analyzed by LC-PDA-ESI/MS (LCMS-8030; Shimadzu) equipped with a 150 × 4.6-mm Purospher Star RP-18 column (particle size, 5 μm; Millipore-Merck). The initial mobile phase was solvent A: solvent B = 98:2 (solvent A, 0.1% formic acid; solvent B, acetonitrile), increased to 80% for x min and maintained at that ratio for another 5 min. UV/Vis spectra was monitored by SPD-M30A (Shimadzu). Mass spectra were acquired in the positive mode of LCMS-8030 with the following conditions: capillary voltage, 4.5 kV; detection range, m/z 50–600; desolvation line, 250°C; heat block, 400°C; nebulizer gas, 3 liters/min; drying gas, 15 liters/min.

(A) We co-cultured an A. nidulans ΔthiA strain with three non-motile B. subtilis strains as follows. MotB; H+-coupled MotA-MotB flagellar stator. FliG; flagellar motor switch protein, physically transduces force from MotA to the rotation of FliF. FliM; flagellar motor switch protein, part of the basal body C-ring. The ΔthiA fungal colony shows a severe growth defect on the plate without thiamine, which is recovered by adding thiamine. The growth defect of ΔthiA is not recovered by co-culture with the non-motile mutants as co-culture with wild-type B. subtilis. Fungal colonies of A. nidulans (ΔthiA) monoculture or co-cultivated with B. subtilis (ΔmotB, ΔfliG, or ΔfliM) on minimal medium with/without thiamine grown for 2 d at 30°C. Each deletion strain expressing ZsGreen. Because flagella are required for bacterial dispersal on the hyphae, the non-motile mutants grow at the center of the fungal colony but do not reach the periphery of the fungal ΔthiA colony (a, right). (B) The area of fungal colonies is measured by ImageJ software. Error bar: SD, n = 3, ***P ≤ 0.001, *P ≤ 0.05. (C) The amount of thiamine in supernatant in the monoculture of B. subtilis non-motile mutants by LC-MS-MRM analysis. OD₆₀₀ are indicated. B. subtilis non-motile mutants synthesize and secrete thiamine in the medium. (B, C) The growth effect on the fungal ΔthiA colony depends on the secretion amount in the four non-motile B. subtilis strains (B, C). (D) CFU on LB (dilution 10⁻⁶) in B. subtilis Δthi monoculture, B. subtilis Δthi + A. nidulans wild-type co-culture, B. subtilis Δthi + A. nidulans ΔthiA co-culture in 200 ml minimal medium with 0.4g shaking at 30°C for 2 and 3 d. CFU in B. subtilis Δthi + A. nidulans wild-type is higher than B. subtilis Δthi monoculture, whereas that in B. subtilis Δthi + A. nidulans ΔthiA co-culture is comparable with B. subtilis Δthi monoculture, suggesting a bi-directional thiamine transfer between B. subtilis and A. nidulans.

(A) Colonies of the B. subtilis reporter strain in monoculture and co-culture with A. nidulans on the minimal medium with/without thiamine grown for 2 d at 30°C. The images are constructed by 10 × 10 tiling of 500 × 500 μm image. Scale bar: 500 μm. (B) The B. subtilis reporter strain at hyphal middles and hyphal tips. Scale bar: 20 μm. Ratio of signal intensity, mScarlet-1/ZsGreen. Error bar: SD, n = 75, **P < 0.01.