Purified agar

Aspergillus minimal medium (AMM) agar media, described in a previous study with several modifications, was used [15 (link),16 (link)]. The media contained 1.5% purified agar (Oxoid, Basingstoke, UK), 10xSalt solution, 20% Glucose, 20% MgSO₄, trace elements, and water. This medium had glucose as the main carbon source, and sodium nitrate was added as a source of nitrogen. After five days of incubation at 37 °C, conidia were harvested in sterile water, and filtered through a 40-μm cell strainer (BD Biosciences, Heidelberg, Germany). Following vigorous vortexing, the concentration of conidia was determined using a Vi-CELL analyser (Beckman Coulter, Brea, CA, USA) or a hemocytometer.
Swollen conidia of A. fumigatus were obtained following the method described in the aforementioned published article [17 (link)]. Conidia were grown in liquid Sabouraud dextrose media for 3 h with shaking at 30 rpm. After swelling, the swollen conidia were washed and then re-suspended in RPMI 1640 Medium, GlutaMAX 25 mM HEPES media, and 10% FBS (R10 medium).

The viability of phototrophs in the microbial mat was tested in liquid and solid BG11 medium. About 0.5 g of sample previously homogenized was inoculated in duplicates in flasks with BG11 liquid medium. Another 0.5 g of the sample was hydrated in BG11 liquid medium and then aliquots were transferred in duplicates to BG11 medium plates with purified agar (Oxoid, Hampshire, United Kingdom). Flasks and plates were incubated for 2 months at 12°C with photosynthetically active radiation (PAR) of ~10 μmol photons m⁻² s⁻¹ under a photoperiod of 16 h, at 20°C with a constant PAR of ~7 and ~20 μmol photons m⁻² s⁻¹, and at 27°C with a constant PAR of ~3 μmol photons m⁻² s⁻¹. The different PAR intensities were achieved using a shadow mesh. Positive controls with Chrococcidiopsis sp. strain 029 (Billi et al., 2001 (link)) were included for the tests at 20°C and 27°C, and two negative controls with liquid and solid BG11 medium without sample were included in the experiments.
The viability of heterotrophs was tested in solid R2A and LB media. About 0.5 g of sample previously homogenized was hydrated in sterile PBS buffer (Biosolve Chimie, Dieuze, France) and then aliquots of 10-fold dilutions were transferred to R2A and LB agar media. Plates were incubated at 15°C, 20°C, and 25°C in parallel in dark conditions for 20 days. Three positive controls of Rhodococcus sp. JG3 (Goordial et al., 2015 (link)) and three negative controls without sample were included in the experiments.

To investigate whether C. albicans was able to grow with ferritin as the sole source of iron, we added 350 µM BPS to the SD agar medium (6.7 g/l yeast nitrogen base, YNB [Difco]; 20 g/l D-glucose; 20 g/l purified agar [Oxoid, http://www.oxoid.com]). Additionally, HEPES buffer (Sigma-Aldrich) was added to the medium as indicated and the pH was adjust to 7.4 using a 5 M NaOH stock solution (Roth). To prevent active acidification of the medium by the fungus, 20 g/l casamino acids (Difco) was used in place of D-glucose. The ferritin solution (ferritin from horse spleen [Sigma-Aldrich]) was diluted 1∶100 in a dilution buffer (5 mM HEPES; 0,1 M NaCl [Roth]) and passed through a Microcon YM-100 Centrifugal Filter Unit (Millipore, http://www.millipore.com). The retentate was collected in a fresh 1.5 ml microcentrifuge-tube and the original volume was adjusted with the dilution buffer. Afterwards, this ferritin solution was plated out on agar surfaces at indicated concentrations. To monitor the pH changes in the medium during C. albicans growth, the pH indicator bromocresol green (Sigma-Aldrich) was added to the medium at a concentration of 3.9 mg/l.