Fungal strains and routine maintenance
Trichoderma atroviride IMI 206040 (TaWT), Trichoderma reesei QM6a (Tr) and Trichoderma virens Gv29-8 (Tv) were used as Trichoderma wild type strains to study the effect of Phi in this work. Rhizoctonia solani AG5 was used as the phytopathogenic fungi for antibiosis and confrontations experiments. All the strains were maintained and propagated using potato dextrose agar (PDA, DIFCOÔ, USA) at 28° C. All the strains were kindly provided by Dr. Alfredo Herrera-Estrella from the Centro de Investigacion y de Estudios Avanzados in Mexico.
To express the ptxD gene in T. atroviride, two constructs were generated using the plasmid pCB1004 as backbone , which contained an independent hygromycin B resistance-cassette. The coding region of the ptxD gene from Pseudomonas stutzeri WM88 was used as reference to design and synthesize a DNA sequence (GenBank accession number MN434083) optimized according to the nuclear codon usage of T. atroviride using the GenScript’s OptimumGeneTM platform. In the first construct, the ptxD coding sequence, linked to the 5’UTR and 3’UTR of the cel7a (cbh1) gene of T. reesei , was placed under control of the ccg6 promoter from T. atroviride (GenBank accession number MK887357) and the blu17 terminator. The same arrangement was used in the second construct, except that the ccg6 promoter was replaced by the pki1 promoter. The constructs were cloned between the KpnI and XbaI sites in the multiple cloning site of pCB1004.
Genetic transformation of T. atroviride was carried out according to Herrera-Estrella . Briefly, transformed cells were spread onto PDA plates containing 100 ug×mL-1 hygromycin B as selective agent. In the darkness Trichoderma grows indefinitely as mycelium, whereas light triggers conidiation. Therefore, to favor the production of spores from the transformed cells, plates were incubated in the presence of light until transformants appeared. At least 50 transformants were picked out for each construct, and 15 of them were subjected to five rounds of single spore isolation until stable lines were obtained. Some stable lines were further characterized.
Transcriptomic data analysis
To analyze the expression pattern of ccg6 gene, T. atroviride RNA-seq data were downloaded from the European Nucleotide Archive (https://www.ebi.ac.uk/ena/) and non-wild-type samples were discarded. The following RNA-seq datasets were used for this purpose: (1) PRJNA575031: PRJNA575031: SRR10207228, SRR10207229, SRR10207220, SRR10207224, SRR10207225; (2) PRJNA310123: SRR2231021, SRR2239861, SRR2239862, SRR2230025, SRR2239855, SRR2239865, SRR2239856, SRR2239866, SRR2239857, SRR2239867, SRR2239868, SRR2239859, SRR2226780, SRR2226739, SRR2239800, SRR2239801, SRR2239792, SRR2239802, SRR2239803, SRR2239794, SRR2239804, SRR2239795, SRR2239796, SRR2239798, SRR2239799 ; (3) PRJNA336223: SRR4445669, SRR4445663, SRR4445668, SRR4445662, SRR4445664, SRR4445670, SRR4445667, SRR4445673, SRR4445666, SRR4445672, SRR4445665, SRR4445671; (4) PRJNA476116: SRR7343320, SRR7343321, SRR7343322, SRR7343323, SRR7343324, SRR7343325, SRR7343338, SRR7343339, SRR7343340, SRR7343341, SRR7343342, SRR7343343 ; (5) PRJNA508370: SRR8280322, SRR8280323, SRR8280324, SRR8280325, SRR8280326, SRR8280327, SRR8280334, SRR8280335, SRR8280336, SRR8280337, SRR8280338, SRR8280339, SRR8280346, SRR8280347, SRR8280348. A T. atroviride transcripts file was created with gffread (http://ccb.jhu.edu/software/stringtie/gff.shtml) using as inputs the T. atroviride IMI 206040 (GCA_000171015) TRIAT_v2.0 genome fasta and gff3 files available at EnsemblFungi, and fastq files were aligned and quantified using kallisto (version 0.46.1) . Expression data was gene level-summarized using tximport , and estimated counts were normalized using the GeTMM (Gene length corrected Trimmed Mean of M-values) method  to create the corresponding gene expression barplots. GeTMM is a normalization method combining gene-length correction with the normalization procedure TMM .
Analysis of the ccg6 promoter and CCG6 protein
Multiple alignments of protein sequences were performed using MAFFT (version 7)  and the results visualized using the software BoxShade . Phylogenetic tree was generated using the neighbor-joining method and 100 bootstrap resampling using Phylo.io . NCBI accession numbers used in the analyses are as follows: Trichoderma atroviride IMI 206040 (T. atroviride), XP_013943686.1; Trichoderma gamsii (T. gamsii), XP_018661560.1; Trichoderma asperellum CBS 43.97 (T. asperellum), XP_024758524.1; Trichoderma arundinaceum (T. arundinaceum), RFU72315.1; Trichoderma harzianum CBS 226.95 (T. harzianum), XP_024772586.1; Trichoderma virens Gv29-8 (T. virens), XP_013960948.1; Trichoderma reesei QM6a (T. reesei); XP_006961490.1; Trichoderma citrinoviride (T. citrinoviride); XP_024751409.1; Hirsutella minnesotensis 3608 (H. minnesotensis), KJ271071.1; Tolypocladium ophioglossides CBS 100239 (T. ophioglossides), POR39196.1; Neursopora crassa OR74A (N. crassa), XP_960686.2; Fusariun avenaceum (F. avenaceum), KIL96403.1; and Fusarium fujikuroi IMI 58289 (F. fujikuroi), XP_023430618.1. Amino acid sequences of the CCG6 proteins from different Trichoderma species were analyzed using InterProScan  and SignalP version 5.0 servers . Analyses of the ccg6 promoter sequences of T. atroviride, T. reesei, and T. asperellum to identify DNA motifs were performed using the MEME Suite searching for motifs with minimum and maximum width site of 6 and 50 nucleotides, respectively . Specific DNA motifs were then analyzed in Tomtom  using YEASTRACT database to identify matches with reported motifs.
Analysis of the presence of the ptxD gene by PCR
Genomic DNA was isolated from mycelial tissue using an urea-based method. Briefly, mycelial tissue was harvested, frozen and ground with liquid nitrogen and 600 uL of urea buffer [7 M urea (Sigma-Aldrich, Saint Louis, MO 63103, USA), 0.3 M NaCl (Sigma-Aldrich, Saint Louis, MO 63103, USA), 0.02 M EDTA (Sigma-Aldrich, Saint Louis, MO 63103, USA), 0.05 M Tris-HCl pH 8 (Sigma-Aldrich, Saint Louis, MO 63103, USA)] were added to 100-150 mg of ground tissue, mixed toughly and incubated at room temperature for 30 minutes. Samples were centrifuged at 10000 rpm and the supernatant mixed with one volume of isopropanol. Samples were mixed by inversion, incubated for 5-10 minutes at room temperature, and then centrifuged for 5 minutes at 10,000 rpm. The pellet was washed using ethanol 70% and DNA resuspended in sterile deionized water. Genomic DNA was quantified in a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA) and its integrity was verified through agarose gel electrophoresis (1%). Possible RNA contamination from DNA samples was removed with RNase A (InvitrogenÔ, PureLinkÔ, RNase A, Van Alley Way Carlsbad, CA, USA) following manufacturer’s instructions. Later, 500 ng of genomic DNA were used as template to amplify the ptxD gene using Taq DNA Polymerase (InvitrogenÔ, Van Alley Way Carlsbad, CA, USA) according to manufacturer’s instructions. The following conditions were used to amplify the ptxD gene: 95°C for 3 minutes, 95°C for 30 seconds, 65°C for 30 seconds, 72°C for 1 minute, 72°C for 5 minutes and finally held at 4°C. Primers used for ptxD amplification were TaptxDF 5’ ATGCTGCCTAAGCTTGTC 3’ and TaptxDR 5’ TCAGCAGGCGGCAGGCTC 3’. PCR product size (840 bp) was verified in agarose gel electrophoresis (1%) using SYBR-Safe DNA Gel Stain (InvitrogenÔ, Van Alley Way Carlsbad, CA, USA) in order to determine the presence of the transgene.
To analyze expression levels of the ptxD constructs, TaWT and transgenic strains were cultured on PDA plates overlaid with a sterile cellophane membrane and incubated for 3 days at 28°C. Then, mycelia were harvested, frozen and ground with liquid nitrogen. Total RNA isolation was carried out using PureZOLTM (BIORAD, Hercules, CA, USA) isolation reagent, according to manufacturer’s instructions. Total RNA concentration was determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA) and its integrity and quality was verified through agarose gel electrophoresis (1.5%). Afterwards, 1 µg of total RNA was treated with DNase I (InvitrogenÔ, Van Alley Way Carlsbad, CA, USA) to remove any possible genomic DNA and cleaned up using the RNeasy® Mini Kit (Qiagen) according manufacturer’s instructions. Briefly, total RNA was reversed-transcribed to cDNA with SuperScript III Reverse Trancriptase (InvitrogenÔ, Van Alley Way Carlsbad, CA, USA), according to manufacturer’s instructions. Oligonucleotides used for ptxD amplification were: TaptxDF 5’-ATGCTGCCTAAGCTTGTC-3’ and TaptxDR 5’-TCAGCAGGCGGCAGGCTC-3’. Real Time qPCR was performed in quadruplicates for each sample using a 7500 Real Time System and the SYBR® Green Master Mix (Applied Biosystems, USA), according to the manufacturer’s instructions. T. atroviride glyceraldehyde-3-phosphate dehydrogenase (gpd, Id 297741) gene was used as reference and was amplified in parallel with the ptxD gene. Primers used to amplify the gpd gene were gpdF 5’-GTGCTGCCCAGAACATCATCC-3’ and gpdR 5’-TGGCGGTAGGGACACGAATG-3’. Standard curves were obtained using five-serial-dilutions for ptxD and gpd genes. The data were analyzed with the 2-ΔΔCt method to determine the expression of the ptxD gene.
To analyze the expression levels of the ccg6 gene, TaWT was cultured for 4 days at 28°C in darkness on liquid VMM supplemented with 2 % glucose, sucrose and mannitol as the carbon source, and under carbon (0.26%) and nitrogen (0.003 mM) starvation, and in PDB as control. Additionally, TaWT was cultured on both liquid PDB and VMM at 28°C in darkness, and mycelium collected 2, 4 and 6 days after inoculation. Mycelia were harvested, frozen and ground with liquid nitrogen and total RNA was isolated using PureZOLTM (BIORAD, Hercules, CA, USA) and processed as described above. However, real Time qPCR was performed in triplicates for each sample using the Q-qPCR instrument (Quantabio, Germany) and the PerfeCTa® SYBR® Green FastMix® (Quantabio, Germany), according to the manufacturer’s instructions. T. atroviride gpd (Id 297741), RpS (Id 297741), and elF-4 (Id 301614) genes were used as qRT-PCR expression controls. Primers used to amplify ccg6, gpd, and elF-4 control genes were: ccg6F5’-CGACACACCTCGCCAATATAC-3’, ccg6R5’-GTAGCGCATCTTCTCGTG-3’; gpdF5’-GCTGCCGATGGTGAGCTCAAGGG-3’ and gpdR5’-GAGGTCGAGGACACGGCGGGA-3’, and elF-4F5’-GTCCAACTACGATGAGACTGTC-3’ and elF-4F 5’-TCGTGGCCCTTGATAACAG-3’, respectively. The data were analyzed with the 2-ΔΔCt method to determine the expression levels of the ccg6 gene.
PTXD enzymatic activity
The enzymatic activity of PTXD was determined using total protein cell extracts. Flasks with 50 mL of Vogel’s Minimal medium (VMM)  supplemented with 36 mM Pi (KH2PO4) for WT strain and 1 mM Phi (KH2PO3) for transgenic strains, were inoculated with 1x107 conidia and cultured at 200 rpm, 28°C for 7 days. Mycelial growth was harvested and resuspended in 5 mL of resuspension buffer [50 mM MOPS (Sigma-Aldrich, Saint Louis, MO 63103, USA) pH 7.5, 25 mM NaCl (Sigma-Aldrich, Saint Louis, MO 63103, USA), 1 mM EDTA (Sigma-Aldrich, Saint Louis, MO 63103, USA), and 50 µL of Protease Inhibitor Cocktail (Sigma-Aldrich, Saint Louis, MO 63103, USA)] and cells were lysed by ultrasonication on ice using an Ultrasonic Processor (Cole-Parmer, Vernon Hills, IL, USA) with 40% amplitude, 2 cycles of 4 minutes with short bursts of 10 seconds followed by intervals of 5 seconds for cooling on ice. Then, samples were centrifuged at 9000 rpm at 4°C for 20 minutes. The resulting supernatant was applied to a 10K MWCO dialysis tubing (SnakeSkin Dialysis Tubing, ThermoFisher Scientific, Rockford, IL, USA) and incubated at 4°C for 24 hours. Samples were quantified using Quick StartTM Bradford Dye Reagent (BIORAD, Hercules, CA, USA) according to manufacturer’s instructions in a 96 well clear microtiter plate (Corning Inc, USA). Extracts were adjusted to the same protein concentration (0.5 mg·mL-1). Determination of PTXD enzymatic activity was performed using a fluorescence-based method for NADH detection. Each reaction was started in individual wells of a 96 well black microtiter plate by adding 100 µL assay mix to give a final concentration of 50 mM MOPS, 0.5 mM Na3PO3 (Sigma-Aldrich, Saint Louis, MO 63103, USA) pH 7.0, 0.75 mM NAD+ (Sigma-Aldrich, Saint Louis, MO 63103, USA), and 25 µg·µL-1 of protein extract per well. The microtiter plate was incubated at 37°C for 1 hour in the darkness. The reaction product was quantified with a fluorescence reader Fluoroskan AscentTM FL (Thermo Fisher Scientific, Vantaa, Finland) at 340 nm excitation and 460 nm emission wavelengths for NADH detection.
TaWT and transgenic strains were cultured on PDA plates overlaid with a sterile cellophane sheet, inoculated with 1x105 conidia suspension, and incubated at 28°C for 48 hours. The cellophane sheet together with the fungal colony was removed from cultures; plates were then inoculated with R. solani AG5 mycelial plugs. The plates were incubated at 28°C for 72 hours in total darkness and then photographed to determine fungal colony growth.
Mycoparasitism of TaWT and transgenic strains was evaluated against R. solani AG5. Mycelial plugs from R. solani AG5, taken from the edge of actively growing colonies of fresh fungal cultures and 1x105 conidia suspension from TaWT and transgenic strains were cultured on PDA, placed approximately 5 cm from each other. Plates were incubated at 28°C for 96 hours in total darkness and then photographed to determine fungal colony growth.
Evaluation of the capacity of Trichoderma wild type strains to metabolize phosphite
To evaluate the capacity of TaWT, Tr, and Tv of growing using Phi as the P source, we used a modified VMM recipe, in which the source of P, monopotassium phosphate, was replaced by monopotassium phosphite. VMM is commonly used to cultivate Neurospora crassa, however, it has been also successfully used to grow Trichoderma species including TaWT, Tr, Tv, and T. asperellum [40, 41, 42, 43]. VMM was prepared as a 50X solution and used as a 1X solution and supplemented with glucose 2% and Bacto Agar (DIFCOÔ, USA) 1.5% when needed. The P sources were monopotassium phosphate (Pi, KH2PO4, Sigma-Aldrich, USA) or monopotassium phosphite (Phi, KH2PO3, Wanjie International Co., Limited, China), as established for each experiment. A 1x105 conidia suspension from TaWT was inoculated in Petri dishes containing VMMV supplemented with 36 mM Pi and incubated at 28°C for 36 hours. For experiments in which mycelium plugs were used as inoculum, mycelial plugs taken from the actively growing colony were placed in new Petri dishes with PDA, VMM without a P source and VMM supplemented with Pi (36 mM) or Phi (1, 2, 3, 4, or 5 mM) at the P sources. Plates were incubated at 28°C for 8 days in total darkness, and then colony growth was evaluated. Similar experiments were performed but using TaWT conidia (1x105) as inoculum in solid VMM media. Additionally, experiments in glass flaks containing liquid VMM media without a P source, supplemented with 36mM Pi, and 1, 2, 4, and 5 mM Phi, and inoculated with 1x105 TaWT conidia suspension were also performed (28°C, 200 rpm, 8 days, constant light). TaWT conidia were obtained from PDA cultures by exposing them to constant white light during 96 hours. Conidia were collected by scraping the culture surface with sterile water and counted in a Neubauer chamber.
Experiments to evaluate the capacity of Tr and Tv to metabolize Phi, were performed as mentioned before in VMM solid media, but both mycelium plugs and conidia inoculum were produced from fresh cultures grown on VMM with Pi (10 mM). Inoculum were placed in Petri dishes containing PDA, VMM media without a P source, supplemented with 36 mM Pi, and 0.25, 0.5, 0.75, 1, and 2 mM Phi. Colony growth was determined by measuring colony area (cm2) using ImageJ 2.0.0-rc-43/1.50e .
Growth of transgenic strains using phosphite as the sole phosphorus source
To determine the ability of transgenic strains to grow using Phi as sole P source, transgenic and TaWT strains were cultured in flasks containing 30 mL of VMM supplemented with 1, 2, 4, and 5 mM Phi. Each glass flask was inoculated with 1x107 conidia suspension from each Trichoderma strain. VMM media with Pi (36 mM) and without a P source were used as controls. Flasks were incubated at 28°C and 200 rpm for 7 days. Afterwards, biomass produced for each Trichoderma strain (dry weight, DW) was determined by filtration of the culture with a sterile Whatman filter paper and dry weight determined.
Competition experiments were performed at 28ºC and 200 rpm in 50 ml glass flasks containing 25 ml of VMM without a P source, supplemented with 10 mM Pi or 4mM Phi as P source. 1x108 conidia of transgenic strain ccg6OPT-3 were co-cultivated with 50 μL inoculum of a culture of E. coli DH5a. For the preparation of the inoculum of E. coli DH5a, 50 µL of glycerol stock was activated with 500 μL of SOC medium, incubated at 37ºC and 200 rpm during 2 hours. Monocultures were performed in the same conditions. At the end of one week of cultivation, mycelia were harvested by filtration with a sterile Whatman filter paper and dry weight determined. CFU were determined in the filtrate.
Adjustments on brightness, sharpness, and contrast were applied to photographs of Petri dishes and glass flasks in figures 3, 4, 5, 6, S2, S3, S4, S5, S7, and S8 to improve image quality and visibility of the mycelium. We did not obscure or eliminate any information present in the originals.
Sequences of the ccg6 promoter and the ptxD gene codon-optimized for T. atroviride were submitted to the GenBank database under accession numbers MK887357 and MN434083, respectively.
Statistical analyses were performed using a one-way ANOVA and Tukey HSD test (p< 0.05).