Analysis of COs and LCOs from fungal exudates
The list and sources of the 59 species of fungi and three species of oomycetes (Heterokontophyta) used in the study are presented in Supplementary Data 1–3. The fungal species examined are representatives of each sub-phyla within five phyla (out of eight phyla) of the Kingdom Fungi30. The absence of contaminants in the fungal and oomycete strains was systematically checked by PCR by using the specific primers ITS1F/ITS4 and fD1/rP231,32 (Supplementary Fig. 2). The inoculum type (cells, mycelium, spores, or zoospores), culture media and culture times used for each strain are indicated in Supplementary Data 1–3.
Fungi and oomycetes producing mycelia were pre-cultivated in Petri dishes on the solid media gelled with agar as indicated in Supplementary Data 233,34. Once the mycelium had covered the dish, plugs of the mycelium were transferred to Sylon-coated culture flasks35 or to 6.7 × 11.4 cm flat-bottom PYREX® flasks (Corning, Inc. Corning, NY), or for the Russulales to 25 × 95 mm flat-bottom culture tubes (PhytoTech), respectively, filled with 50 ml or 12 ml of the appropriate liquid medium to produce and collect exudates (Supplementary Data 2). In addition, using a separate experimental method, C. geophilum, G. stellatum, Lepidopterella palustris, Leptosphaeria maculans, S. sclerotiorum, and the species of Amanita, Hebeloma, and Paxillus were pre-cultivated as above but they were inoculated on a cellophane membrane laid on the solid medium. This membrane was used to transfer agar-free mycelium to Petri dishes filled with deionized sterile water or with liquid culture medium in order to produce and collect exudates (Supplementary Data 2).
For the anaerobic Neocallimastigomycetes, Neocallimastix californiae, Piromyces finnis, Anaeromyces robustus, and Caecomyces churrovis, 1 ml of fungal zoospores was used to inoculate 20 ml of modified minimal Medium C in a 60 ml borosilicate serum bottles containing 0.2 g switchgrass while sparging with CO236,37. Fungal cultures were incubated anaerobically 6 days before collecting the exudates.
For fungi (except AM fungi) producing cells, spores, or zoospores, 106 of these propagules were produced and collected according to published methods33,37,38,39,40,41,42,43,44,45,46,47,48. Propagules were inoculated directly in five independent Sylon-coated flasks with 50 ml liquid medium per species. AM cultures were propagated by in vitro mycorrhizal root organ cultures in solid M medium containing Phytagel (Sigma-Aldrich) and collected after solubilization of Phytagel39,49. Exudates from the AM fungal strains were collected from 10,000 spores germinating in 10 ml liquid medium for 10 days.
The various liquid media (broth or water), enriched with exudates, were filtered under sterile conditions through a 0.22 µm Millipore Express® PES membrane (MilliporeSigma, Darmstadt, Germany) prior to being analyzed in the bioassays.
One hundred to 400 ml of culture filtrates, depending on the fungal cultures, were extracted twice with butanol (1 : 1 v/v). The pooled butanol phases were washed with distilled water and evaporated under vacuum. The dry extract was re-dissolved in 4 ml water : acetonitrile (ACN) (1 : 1 v/v) and dried under nitrogen. This crude extract was resuspended in 1 ml of 20% ACN in water and separated on Hypersep C18 (500 mg, 3 ml, Thermo Fisher Scientific) by sequential elution with 3 ml each of 20%, 50%, and 100% ACN in water, respectively. The eluted samples were then dried under nitrogen. Occasionally, for further purification, the 50% eluate was resuspended in 75% ACN in water and separated on Chromabond HILIC (500 mg, 3 ml) by sequential elution with 3 ml each of 100%, 80 and 75% ACN in water. The eluates were then dried under nitrogen.
The presence of LCOs in filtered crude exudates (1× or 10×) or in the butanol fractions of media were assayed by root hair branching in V. sativa, which is induced by nsLCOs50, by root hair branching in M. truncatula accession Jemalong A17, which is induced by sLCOs, and by expression of MtENOD11 using the pENOD11:GUS transcriptional fusion in M. truncatula, which is also induced by (s)LCOs4,51.
The root hair branching assays in V. sativa and M. truncatula used the method of Cope et al.9. Eight young seedlings (3–7 days old) were treated with the fungal exudates, with the same concentration of solvent (negative controls), or with Nod factors purified from Rhizobium leguminosarum biovar viciae or Sinorhizobium meliloti supernatant at a concentration of 10−8 M (positive controls). One milliliter of fungal crude exudates or 40 µl of butanol fractions were applied on each seedling primary root.
The MtENOD11 gene expression assay was performed as in Maillet et al.4. Two kinds of samples were tested: butanol extracts diluted 100 times in water and HILIC column fractions diluted 10 times. Forty microliters of these solutions were applied to the primary root of each seedling for 16 hours. Seven to ten seedlings were tested by sample and compared to mock treatment (0.005% EtOH in water or 5% ACN in water). Plants were stained for 6 h. An arbitrary scale was used to quantify GUS (beta-glucuronidase)-staining (Supplementary Fig. 8).
Standard LCO compounds (non-sulfated C16:0 LCO IV, sulfated C16:0 LCO IV, non-sulfated C18:1 LCO IV, sulfated C18:1 LCO IV) were synthesized at CERMAV (Grenoble, France) and were used at 10−5 M in ACN/water (1/1, v/v) to determine retention times and to optimize HPLC/QTRAP tandem MS detection by MRM4,12. The UltiMate 3000 HPLC system (Dionex Corporation) was equipped with an Acquity C18 reversed-phase column (2.1 × 100 mm, 1.7 µm, Waters Corporation). Samples of 10 µl were injected. The elution was done at a constant flow rate of 450 µl min−1 using solvent A, water:acetic acid (1000 : 1, v-v) and solvent B, ACN, as follows: 30% B for 1 min, followed by a 30–100 % B during 8 min, followed by isocratic elution with 100% B for 2 min. A QTRAP 4500 mass spectrometer (Applied Biosystems, Foster City, USA) equipped with an electrospray ionization source in the positive ion mode was used to analyze samples in the MRM mode or in the EMS–EPI mode (see below). For the MRM mode analyses, from the known substitutions and chitin lengths already described for Nod factor structures, we created a database of all possible combinations of structures, including new ones never described before, with their corresponding precursor proton adduct ion [M + H]+ and product B ions: in total, 76,386 precursor ions, 2,598,159 theoretical combined structures, and 358,473 MRM transitions (Supplementary Data 5). Given that the number of MRM transitions to be selected for each analysis must be reasonably low to ensure proper sensitivity, we have selected the most commonly described Nod-LCOs (corresponding to 990 MRM transitions). This highly sensitive, targeted, analytical approach was suitable for samples containing low concentrations of molecules. For samples with higher concentrations of molecules, full scan EMS–EPI analyses were performed. During EMS analysis, major precursor ions are selected automatically, and, after the collision, EPI analysis accumulates their product ions in the trapping module. From this data set, we selected only the precursor ions containing 3 to 6 GlcNAc. This more comprehensive mode could only be used with P. adelphus, P. involutus, and G. rosea LCO-rich samples.
Short COs were separated and analyzed using the same LC-MS system, equipped with an hypercarb column (5 μm, 2 × 100 mm; Hypercarb, Thermo). Samples of 10 µl were injected. The elution was done at a constant flow rate of 400 µl min−1 using solvent A, water : acetic acid (1000 : 1, v-v) and solvent B, ACN, as follows: 100% A for 1 min, then 100–50% A in 30 min then 50–0% A in 3 min. COs were identified in the MRM mode by monitoring the transitions from precursor proton adduct ion [M + H]+m/z 628 (CO3), 831 (CO4), 1034 (CO5), or 1237 (CO6) generating after collision-induced dissociation (CID) the common product B ion m/z 204, comparatively to standard solutions (10−7 M in water). The capillary voltage was fixed at 4500 V and the source temperature at 400 °C. Fragmentation was performed by CID with nitrogen at a collision energy of 22–54 V; declustering potential was 90–130 V, optimized for each synthetic molecule available. Data processing was performed using Analyst 1.6.1 software (AB Sciex).
Experiments with A. fumigatus
A. fumigatus strain Af293 was grown in standard 90 mm Petri dishes on solid glucose minimal medium (GMM) and placed in the dark at 37 °C for 48 h52. Ten milliliters of 80% Tween 20 (Acros Organics, New Jersey) in sterile MiliQ water were added to the dishes and agitated with a sterile L-shaped cell spreader (Thermo Fisher Scientific, Waltham, MA) to collect spores. The spore suspension was sterilely transferred to a 50 ml polypropylene sterile Falcon® Centrifuge Tubes (Corning, Corning, NY). The spore suspension was homogenized by vortexing at maximum speed, and a 1 : 10 dilution was prepared with sterile MiliQ water, which was used to count spores using a hemocytometer. Afterward, spore suspension was adjusted to 106 spores with 80% Tween 20 in sterile MiliQ water44.
Spores were germinated in GMM liquid broth supplemented with various LCOs, COs, and fatty acids at a final concentration of 10−8 M. All LCOs and COs stock solutions were in 0.005% aqueous ethanol. The LCOs used were as follows: sulfated C16:0 LCO, non-sulfated C16:0 LCO, sulfated C18:1 LCO, and non-sulfated C18:1 LCO. The COs used were CO4, CO5, and CO8 (IsoSep, Tullinge, Sweden). The fatty acids used were palmitic and oleic acids. The negative control for these analyses was 0.005% aqueous ethanol, the solvent in which the LCO and CO stocks were prepared. The spore concentration was adjusted to 106 spores per ml of medium. One milliliter of spore suspension with the treatment of LCOs or COs were distributed into two replicate wells of a sterile Costar® 24 clear wells round, flat-bottom plate (Corning, Corning, New York) and the cells were incubated at 37 °C for 3 h. They were then observed at 1 h intervals over 21 h using a Nikon Ti inverted microscope with a ×40 objective and ten pictures were taken for each well every hour. Over 200 spores were scored for germination in each well. After 12 h of incubation, the length of the germinating apical hypha and the number of secondary branches per apical hypha were scored for over 200 germinated spores per well. Four independent experiments were performed. No differences between experiments were observed. Dose–response experiments were carried out in the same way, except that the spores were treated with a range of concentrations of sulfated C16:0 LCO(s) from 10−6 to 10−13 M.
Spores of A. fumigatus were grown in GMM supplemented with either 10−8 M sulfated C16:0 LCO or 0.005% ethanol as a negative control. The density was adjusted to 106 spores per ml of medium and the cultures were maintained at 37 °C on a New Brunswick Scientific Excella E25 incubator shaker (Eppendorf, Hamburg, Germany) at 250 r.p.m. The spores were collected after 30 and 120 min by filtering the liquid broth through sterile cheesecloth. Spores were completely removed from the cheesecloth with a sterile spatula and placed into 1.5 ml FisherbrandTM Premium Microcentrifuge tubes (Thermo Fisher Scientific, Waltham, MA). Four independent cultures were replicated per treatment and time point. Immediately after spore collection, tubes were placed in liquid nitrogen for 10 min. The spores were ground to a fine powder in liquid nitrogen and transferred into 50 ml centrifuge tubes. Total RNA was extracted by using QIAzol Lysis Reagent (Qiagen, Hilden, Germany) according to the manufacturer’s instructions but with an additional phenol : chloroform : isoamyl alcohol (24 : 1 : 1) extraction step before RNA precipitation. For RNA sequencing (RNA-Seq), total RNAs were further purified by using the RNeasy Mini Kit (Qiagen). RNA samples were digested with DNase and stored at −80 °C for further use. A NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) was used to quantify and assess the purity of RNA. NanoDrop readings for samples were 112.24–491.44 ng µl−1.
Sixteen libraries of RNA-Seq single-end reads were prepared by using the TruSeq library preparation protocol and sequenced with an HiSeq 2500 sequencing system (Illumina, San Diego, CA). The 16 libraries corresponded to each of the four biological replicates for each of the four treatments. Read quality was assessed with FastQC 0.11.5. Read quality was excellent and adapter sequences were minimal, so reads were not trimmed. Paired-end reads were pseudo-aligned and quantified by using Kallisto 0.42.3 and the reference transcriptome of A. fumigatus Af293 (released 21 December 2012) downloaded from the Joint Genome Institute’s Genome Portal53,54. Bootstrap values were 100. Pairwise transcriptomic comparisons were made by using Sleuth version 0.30.055. We defined transcripts as differentially expressed if they had a false discovery rate (q-value) < 0.05, p-value < 0.01, and β-values < −0.4 or >0.4. GO-enrichment analysis for the A. fumigatus genome was carried by using the Gene ID. GO enrichment was performed using FungiDB56. Potentially regulated biochemical pathways were identified using the Search Pathway feature in KEGG Mapper57. The search mode was set to “Afm,” to specify A. fumigatus as our reference organisms. DEGs for the 30 and 120 mpi timepoints were entered as search objects. A list of objects was returned (Supplementary Data 8).
Experiments with C. glabrata
C. glabrata was grown overnight on yeast extract–peptone–dextrose medium (1% yeast extract, 2% peptone, 2% dextrose), supplemented with uridine (80 μg ml−1) on an orbital shaker at 200 r.p.m. and 30 °C. Ten microliters of the overnight culture were diluted 1 : 1000 in Dulbecco’s phosphate-buffered saline (without calcium or magnesium; HyClone Laboratories, Inc., Logan, UT) and counted by using a hemocytometer.
To initiate and develop biofilm production, RPMI 1640 medium (Thermo Fisher Scientific) is used for species of Candida58. Cells from an overnight culture were pelleted and resuspended (106 cells per ml) in RPMI 1640 medium, supplemented with various COs, LCOs and fatty acids at a final concentration of 10−8 M. The LCOs, COs, fatty acids, and negative controls used were the same as for the experiments with A. fumigatus. Three hundred microliters of cell suspension were distributed into each well of a sterile µ-Slide 8 Well chambered coverslip (Ibidi USA, Fitchburg, WI) and the cells were observed at 10-minute intervals over 12 hours at 37 °C with 5% CO2 in an INU series microscope incubator (Tokhi Hit, Shizuoka-ken Japan) attached to a Nikon TI2-E inverted microscope (Nikon, Louisville KY). Each well corresponded to one treatment and five pictures were taken for each well every 10 min. After 12 h, the total number of pseudohyphae per well was counted. Four independent experiments were performed. No differences between experiments were observed. Dose–response experiments were carried out in the same way, except that the cells were treated with a range of concentrations of sulfated C16:0 LCOs from 10−6 to 10−13 M.
Experiments with R. mucilaginosa
R. mucilaginosa strain was grown in 50 ml DifcoTM Dehydrated Culture Media: Potato Dextrose Broth (Thermo Fisher Scientific), in 125 ml flasks at 25 °C on an orbital shaker at 250 r.p.m. Cells were counted by using CountessTM Cell Counting Chamber slides and the Countess II Automated Cell Counter (Invitrogen, Carlsbad, CA).
The concentration of R. mucilaginosa cells was adjusted to 106 cells per ml of potato dextrose broth and various LCOs and COs were added to a final concentration of 10−8 M. The LCOs, COs and negative control were the same as for the experiments with A. fumigatus and C. glabrata above. Two hundred microliters of each mixture were distributed into the wells of a sterile Costar 96 Well flat-bottom plate (Corning, Corning, NY). The OD600 of each well was measured in 1 h intervals in a Cytation 5 Cell Imaging Multi-Mode Reader (BioTek Instruments, Winooski, VT) over 24 h at 25 °C with shaking at 0.5 r.p.m. The outer wells of the plates were filled with sterile Milli-Q water to prevent evaporation of the samples. After 24 h, the maximum V was analyzed to determine the final OD600 nm reading per treatment. Six technical replicates were carried out for each treatment and three independent experiments were performed. No differences between experiments were observed.
Prediction of proteins involved in LCO synthesis and LysM-containing proteins in fungi
The predicted number of genes encoding chitin synthases, chitin deacetylases, N-acyltransferases, and LysM-containing proteins was reported according to the respective references 59,60,61,62,63,64,65,66,67,68,69,70.
Statistical analyses were performed using RStudio (version 1.2.1335, RStudio Team 2015, Boston, MA) and GraphPad Prism software version 8.3.0 (GraphPad, San Diego, CA). One-way analysis of variance differences were considered significant when p < 0.05. For the A. fumigatus experiments, the Tukey’s single-step multiple comparison test was used to compare the different concentrations of C16:0 sLCOs and control, and the Dunnett’s pairwise test was used to compare the treatments (LCOs and COs) to the control. Statistically significant differences were based on p-values < 0.05. For the C. glabrata experiments, the Tukey’s single-step multiple comparison test was used to compare all treatments to each other, as the control had no pseudohyphae formation, and to compare the different concentrations of C16:0 sLCO and control. For the R. mucilaginosa experiment, Dunnett’s pairwise test was used to compare the treatments (LCOs and COs) to the control.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.