INTRODUCTION
Fungal secondary metabolites (SMs) are diverse organic compounds that are not essential for growth or reproduction but provide ecological advantages to the organism. These compounds help deter predators, inhibit competitors, and facilitate symbiotic relationships, enhancing the organism’s survival, reproduction, and adaptability under diverse environmental conditions (
1–4).
Among the different SMs produced by fungi, mycotoxins are a significant group with a profound impact on the health of all living organisms, that is, microbes, plants, animals, and humans. They can contaminate food and feed, being produced by phytopathogenic fungi, which can lead to severe health issues in humans, like cancer, liver damage, and immunosuppression (
5). Fungi exhibit remarkable resilience against self-intoxication by their own mycotoxins, an essential trait for their survival and ecological success. Mechanisms underlying this resistance include (i) efficient efflux pumps that secrete toxins from cells (
6); (ii) enzymatic detoxification pathways (
7); (iii) duplicated or resistant target enzymes (
8); and (iv) subcellular separation of the toxic product (
9). The biosynthetic gene clusters (BGCs) responsible for the production of mycotoxins often carry adjacent non-biosynthetic genes that are relevant for conferring self-protection (
10).
The sphingolipid biosynthesis inhibitors fumonisins are a group of mycotoxins produced primarily by
Fusarium and
Aspergillus species (
11). They are polyketide-derived aminopentol compounds with a linear C
20 backbone and two tricarballylic esters. The best studied are B-series fumonisins (FBs), FB
1, FB
2, FB
3, and FB
4, which are predominantly produced by the corn-infecting fungus
Fusarium verticillioides. As demonstrated in mice experiments, the most toxic FB analog is FB
1, which is the final biosynthetic product accumulating in greatest abundance in
F. verticillioides (approximately 80%) (
12–14). The fumonisin BGC (
FUM) includes 16 genes (
Fig. 1A) (
12), expressed under the control of the Zn(II)
2Cys
6-type transcription factor Fum21 (
15).
FB
1 biosynthesis begins with the formation of the polyketide backbone by the polyketide synthase Fum1 (
16) (
Fig. S1). Next, the aminotransferase Fum8 catalyzes the condensation of the polyketide with
l-alanine (
17). Further modifications are performed by Fum6 (
18), Fum13 (
19), and Fum2 (
20). The tricarballylic acid side chains are synthesized by Fum7 and Fum10 (
21), and are attached to the backbone by Fum14 (
22). FB
1 biosynthesis ends with the hydroxylation at the C-5 position by Fum3 (
23) (
Fig. S1).
FB
1 disrupts sphingolipid metabolism by inhibiting ceramide synthase (CerS), a key enzyme in ceramide biosynthesis (
Fig. 1B) (
24). Because FB
1 is structurally similar to sphinganine, a substrate in ceramide biosynthesis, it competitively inhibits the enzyme by irreversibly binding to the active site of CerS (
25). The inhibition of CerS by sphinganine analog mycotoxins triggers apoptosis, which is caused by an accumulation of sphingoid bases, rather than reduced sphingolipid biosynthesis itself (
26). Targeted application of sphingolipid biosynthesis inhibitors, however, has promising potential as treatment against severe human diseases, such as cancer, schizophrenia, Alzheimer’s, multiple sclerosis, and diabetes (
27–29).
Ceramides constitute the backbone of complex sphingolipids. In filamentous fungi, ceramide biosynthesis begins with the condensation of
l-serine with palmitoyl-CoA by a serine palmitoyltransferase to form 3-ketosphinganine, which is further reduced by a ketoreductase to sphinganine. At this point, ceramide biosynthesis branches (
30). Sphinganine is converted by CerS to dihydroceramide, and subsequently to ceramide by a desaturase. Alternatively, sphinganine is converted to phytosphingosine by a C4-hydroxylase and to phytoceramide by CerS (
Fig. 1B). Ceramides are vital components of cell membranes and signaling molecules involved in apoptosis, cell differentiation, and proliferation (
31). This disruption thus contributes to various diseases in humans and animals (
32–34).
The biosynthesis of sphingolipids is compartmentalized. The first enzymatic reactions, leading up to ceramides, are catalyzed in the endoplasmic reticulum (ER), while complex sphingolipids are further synthesized in ER-derived vesicles and the Golgi (
35,
36). Recently, Fum3 (the enzyme catalyzing the last reaction in FB
1 biosynthesis) was shown to be localized in the cytosol in
F. verticillioides (
8,
37). This finding suggests a subcellular separation of the toxin FB
1 from its target CerS in the ER, contributing to self-protection.
The
FUM cluster in
F. verticillioides entails five non-biosynthetic genes adjacent to one another (
Fig. 1A),
FUM15 to
FUM19. Fum19 is an ATP-binding cassette (ABC) transporter that acts as a repressor of the
FUM gene cluster, regulating the levels of intracellular and secreted FB
1 (
8). Fum17 and Fum18 are two of five CerS homologs in
F. verticillioides, both co-localizing with ceramide biosynthesis in the ER. In particular, Fum18 was shown to confer resistance to FB
1 self-toxicity (
8). While FB
1 has antifungal activity, specifically against FB nonproducers,
F. verticillioides is more resistant to externally added FB
1 (
38).
Despite all the functional analyses of
FUM cluster genes for over two decades, the function of
FUM15 and
FUM16 has remained unclear so far. Also, the direct effect of
FUM15-FUM18 on ceramide biosynthesis has not yet been elucidated. Fum15 is a P450 monooxygenase, while Fum16 displays structural similarities with long-fatty-acid-CoA ligases (
12). Our hypothesis was that they, too, may play a role in self-protection.
In this work, we uncovered the involvement of Fum15 and Fum16 in self-protection against FB1 in F. verticillioides, and additionally, in the budding yeast Saccharomyces cerevisiae as a heterologous host. Using an in vitro enzyme assay, we showed that Fum16 is a functional palmitoyl-CoA ligase, while the presence of the P450 monooxygenase Fum15 reduces extracellular levels of FB1.
DISCUSSION
Mycotoxin-producing organisms evolved several strategies to overcome self-toxicity, like the duplication of target enzymes, efficient toxin secretion mechanisms, enzymatic detoxification, and pathway compartmentalization (
6–9). In our previous work, we identified the function of three genes in the
FUM cluster of
F. verticillioides that are located in a subcluster relevant for self-protection,
FUM17,
FUM18, and
FUM19 (
8). In the present study, we expanded on this knowledge by postulating the role of two additional genes,
FUM15 (P450 monooxygenase) and
FUM16 (palmitoyl-CoA ligase), and discovered that they, too, carry out self-protective functions (
Fig. 1B). We are therefore presenting evidence for the existence of a five-gene subcluster
FUM15-19 specifically dedicated to self-protection against the sphingolipid biosynthesis inhibitor FB
1 in
F. verticillioides.
The production of fungal SMs is tightly controlled not only on a temporal level but also spatially (
46). In our previous work, we showed that the CerS homologs Fum17 and Fum18 are compartmentalized and co-localize with ceramide biosynthesis in the ER, while the final biosynthetic step to FB
1 by Fum3 is performed in the cytosol (
8). Here, we demonstrated an equivalent localization for Fum15 and Fum16 in the ER. While certain P450 monooxygenases have been observed in various cellular compartments, they are typically located in the ER (
47) where they associate with NADPH-cytochrome P450 reductases for electron supply (
48). In contrast to that, long-fatty-acid-CoA ligases are predominantly found in peroxisomes (
49). The localization of Fum16 to the ER thus offered an initial indication of its role in ceramide biosynthesis.
In this work, the involvement of FUM15 and FUM16 in protection against FB1 was demonstrated both in the native host F. verticillioides and in the heterologous host S. cerevisiae. Deletions of either FUM15, FUM16, or FUM18 in an FB1-overproducing strain of F. verticillioides resulted in growth inhibition, both on solid medium and in liquid culture. On the opposite, constitutive heterologous expression of these genes in S. cerevisiae conferred protection against FB1. FUM18 appears to be the non-biosynthetic gene conferring the highest degree of protection in the F. verticillioides experimental setup. This was confirmed by observing no difference in the growth inhibition phenotype both when deleting FUM18 by itself and in combination with the other non-biosynthetic genes FUM15-19. In contrast, in the S. cerevisiae system, it was FUM15 expression that resulted in the highest survival, with a growth phenotype almost comparable to the untreated S. cerevisiae WT. While all three genes were expressed to similar strength from the overexpression promoter, we cannot exclude differences in the resulting protein levels as we did not perform codon optimization. Alternatively, it might be due to variations in enzyme activities between the different host organisms.
Subsequently, we observed that the deletion of either
FUM15,
FUM16, or
FUM18 affects ceramide homeostasis in
F. verticillioides. In
S. cerevisiae, in the absence of FB
1, the expression of
FUM15 did not alter the levels of ceramide intermediates, which suggests that it does not have a direct enzymatic function in ceramide biosynthesis. Conversely,
FUM16 expression in untreated
S. cerevisiae cultures was sufficient to influence ceramide intermediate biosynthesis. There was no effect of
FUM16 expression on FB
1 levels in
S. cerevisiae, which was consistent with findings for the
FUM16 deletion mutant of
F. verticillioides (this study;
41), and supported the hypothesis that it plays a direct role in supplementing the ceramide pathway. This could be unequivocally verified by an
in vitro enzymatic assay, where we demonstrated that Fum16 is a functional palmitoyl-CoA ligase (
Fig. 1B). Recently, we described a similar case of enzymatic supplementation of sphingolipid biosynthesis serving as a self-protection mechanism in
Aspergillus. We explored the interactions between sphingolipid metabolism and sphingofungin biosynthesis in
Aspergillus fumigatus, with the latter being a sphingolipid biosynthesis inhibitor acting on SPT (
37). We demonstrated that the biosynthetic enzyme SphA (aminotransferase) plays a dual role in both sphingofungin and sphingolipid biosynthesis as it performs SPT activity in the presence of the serine and palmitoyl-CoA precursors (
50).
From an evolutionary perspective,
FUM15 seems to play an important role. The fumonisin BGC exhibits significant variation among black aspergilli (
51), which are reported to primarily produce the FB analogs FB
2 and FB
4 (precursor of FB
2), but not FB
1 (
52). Among the non-biosynthetic genes identified for self-protection in
Fusarium (
FUM15-19), only
FUM15 is present in the
FUM cluster of both fumonisin-producing and nonproducing strains of
A. niger (
53). Furthermore, in nonproducing isolates of
Aspergillus luchuensis,
Aspergillus brasiliensis, and
Aspergillus tubingensis, the partial
FUM cluster exclusively retained
FUM1 and
FUM15 (
51). Our expression data revealed that
FUM15 was not under the tight control of the transcription factor Fum21, similar to what was reported for
A. niger previously (
54).
Our data do not provide evidence for a direct role of Fum15 in ceramide biosynthesis. Instead, we suggest that it chemically modifies and thereby detoxifies FB
1 due to the fact that
FUM15 deletion in
F. verticillioides resulted in increased extracellular FB
1 levels, while the opposite was observed for the heterologous expression of
FUM15 in FB
1-supplemented
S. cerevisiae. P450 monooxygenases have been largely reported to be involved in the detoxification of mycotoxins through biotransformation and chemical modification (
55). In
F. fujikuroi, P450 monooxygenase(s) were shown to be involved in the detoxification of fusaric acid (
56) and gibepyrone A (
40). Other examples are the hydroxylation and epoxidation of aflatoxin B
1 (
7,
55), as well as the hydroxylation of deoxynivalenol (
7). It is apparent that other cluster-independent mechanisms cooperate to confer protection against self-toxicity by FB
1 in
F. verticillioides since the deletion of the complete subcluster
FUM15-19 still displayed viability. These are likely cluster-independent ABC transporters compensating for the loss of
FUM19 (
8). Additionally, a gene directly adjacent to the
FUM cluster,
FvZBD1, is postulated to act as a genetic repressor of fumonisin production (
57).
Based on the results of this study, we conclude that the non-biosynthetic genes,
FUM15-19, in the
F. verticillioides FUM cluster compose a subcluster of contiguous genes, which are able to confer self-resistance against the toxic effects of FB
1.
FUM16,
FUM17, and
FUM18 do so by supplementing ceramide biosynthesis as palmitoyl-CoA ligase and CerSs, respectively (this study;
8)).
FUM19 encodes an ABC transporter that acts as a repressor of the
FUM gene cluster (
8), while
FUM15 is possibly involved in chemical detoxification of FB
1. Addressing mycotoxin contamination is crucial for maintaining a safe and sustainable food supply, highlighting the importance of continued research and monitoring (
1). In particular, future identification of these self-protection subclusters could uncover mycotoxins with currently unknown biological functions and potentially assist in developing plant cultivars with enhanced resistance.
MATERIALS AND METHODS
General molecular methods
For amplification of desired DNA fragments from template DNA, the Phusion Flash High-Fidelity PCR Master Mix (Thermo Fisher Scientific, Dreieich, Germany) was used. All primers used for PCR amplification, diagnostic PCR, and Southern blots are listed in
Table S1. For DNA enzymatic digestions, restriction enzymes by New England Biolabs (Frankfurt am Main, Germany) were used. DNA extraction from agarose gel was performed using the GeneJET Gel Extraction Kit (Thermo Fisher Scientific, Dreieich, Germany).
Vector assembly was achieved using either
S. cerevisiae transformation-associated recombination (TAR) (
58) or Gibson assembly (
59) and checked with restriction digestion and sequencing. Plasmid DNA purification was performed using the NucleoSpin Plasmid Mini Kit (Macherey-Nagel, Düren, Germany).
Plasmid construction
Promoter exchange of
F. verticillioides FUM21 with the inducible
tetON promoter (
Fig. S7) started out by amplifying upstream and (intragenic) downstream sequences (relative to the start codon) with the primer pairs fum21_5F/nat1_fum21_5R and TETon_FUM21_3F/TetON_FUM21_3R2, respectively.
tetON (
39) was amplified from pYES2-ptrA-TetON (
37) with nat1_tetON_F2/gpda_for_vv.
natR was amplified from pZPnat1 (GenBank accession no.
AY631958.1) with hph_trpC_F/nat1_R. The obtained fragments were cloned into the BamHI/HindIII-digested shuttle vector pYES2 (Life Technologies, Darmstadt, Germany) using Gibson assembly.
Deletion vectors harbored ~1 kb upstream and downstream flanks of the gene of interest, as well as a deletion cassette. Flanks were amplified using primer pairs 5F/5R and 3F/3R (
Table S1), as well as
F. verticillioides genomic DNA.
FUM8,
FUM15,
FUM16,
FUM18, and
FUM15-19 were deleted by exchange with the hygromycin B resistance gene under the control of
A. nidulans PgpdA (
Fig. S7). For that purpose,
hphR was amplified from pUC-hph (
60) with gpda_for_vv/Hph_Rev_VV2. The complete vectors were assembled under the use of a linearized (HindIII/XbaI) pYES2, applying TAR cloning.
Microscopy vectors were obtained by fusing the genes of interest to the constitutive
A. nidulans promoter
PoliC and C-terminal fusion to
GFP (
Fig. S8). To this end, the insert was cloned into NcoI-digested pNDH-OGG (
61), applying TAR cloning.
S. cerevisiae expression vectors (
Fig. S9) were constructed by amplifying the inserts from
F. verticillioides cDNA. The inserts were then cloned into the amplified (primer pair TEF_Rv/pYes2_cyc1T_Fw) backbone pYES2::
Ptef1 (
62) using Gibson assembly.
Finally, the bacterial overexpression vector was obtained by amplifying an N-terminally truncated version of FUM16 from F. verticillioides cDNA. The insert was then cloned into the BamHI/HindIII-digested vector pMALC2HTEV (Addgene no. 75286) using Gibson assembly.
F. verticillioides transformation and analysis of transformants
Protoplast transformation of
F. verticillioides was carried out as described elsewhere (
63). Here, 20–30 µg of the circular vector was transformed to achieve overexpression of GFP-tagged
FUM15 and
FUM16, while PCR-amplified fragments were transformed for inducible expression of
FUM21, and for gene deletions. Transformants of
tetON::FUM21 were maintained on complete medium (CM) plates (
64) containing 200 µg/mL nourseothricin (Jena Bioscience, Jena, Germany). This strain was then used as background for the single-gene deletions, and the transformants were grown on CM plates containing both hygromycin B (200 µg/mL, InvivoGen Europe, Toulouse, France) and nourseothricin (200 µg/mL). Homologous recombination of the flanks and absence of untransformed nuclei were tested by diagnostic PCR, while Southern blot experiments excluded additional ectopic integration events. Thus, 2–10 independent transformants were verified for
tetON::FUM21,
tetON::FUM21/Δ
fum8,
tetON::FUM21/Δ
fum15,
tetON::FUM21/Δ
fum16,
tetON::FUM21/Δ
fum18, and
tetON::FUM21/Δ
fum15-19 mutants (
Fig. S7). The microscopy strain
FUM18::DsRed (
8) was maintained on CM plates containing 200 µg/mL nourseothricin and was used as background for the transformation of
FUM15::GFP and
FUM16::GFP. The transformants were grown on CM plates containing both hygromycin B (200 µg/mL) and nourseothricin (200 µg/mL). The ectopic integration of the plasmids was verified by diagnostic PCR (
Fig. S8).
E. coli media and growth conditions
For cloning, E. coli DH5α cells were incubated on Luria-Bertani (LB) plates supplemented with 60 µg/mL carbenicillin (Carl Roth, Karlsruhe, Germany) and incubated at 37°C overnight. In order to isolate plasmids, single colonies were inoculated in LB medium with carbenicillin and incubated at 37°C and 180 rpm overnight. For protein production, E. coli BL21 (DE3) competent cells were transformed with the appropriate plasmid, grown overnight (37°C, 180 rpm) in LB medium with kanamycin (25 µg/mL, Carl Roth, Karlsruhe, Germany).
S. cerevisiae media and growth conditions
S. cerevisiae BY4741 (Euroscarf, Oberursel, Germany) was used as a background strain and is thus referred to as ScWT. General maintenance of the generated S. cerevisiae mutants was performed on solidified synthetic defined medium lacking uracil (SD-Ura). S. cerevisiae cells for plate assays were pre-cultured in liquid SD-Ura and shaken overnight in 100 mL flasks with baffles at 30°C and 180 rpm. All cultivations for growth curves, survival assays, as well as the analysis of ceramide intermediates were carried out in SD-Ura at 30°C. For gene expression analysis, 20 mL SD-Ura in 100 mL flasks with baffles was inoculated with a dense overnight culture to an OD600 of 0.2 and grown out to an OD600 of 1 before harvesting and freeze-drying.
F. verticillioides media and growth conditions
F. verticillioides M-3125 (
65) was used as parental FvWT strain for the analysis. General maintenance of fungal strains was performed on solidified CM, with or without appropriate selection (200 µg/mL nourseothricin, 200 µg/mL hygromycin B). For the cultivation of strains in liquid culture, 100 mL of Darken pre-culture (
66) in 300 mL Erlenmeyer flasks was inoculated with a mycelial plug and shaken for 3 days at 180 rpm and 28°C. For the main culture, 500 µL of the pre-culture was transferred to 100 mL of synthetic ICI medium (Imperial Chemical Industries, Ltd., London, UK) (
67) supplemented with 6 mM glutamine (Gln) and shaken under the same conditions for 2 days. Gene expression of the
FUM cluster under the
tetON promoter was induced via the addition of 50 µg/mL doxycycline hyclate (Applichem, Darmstadt, Germany), and the cultures were shaken under the same conditions for an additional 5 days for FB
1 and ceramide intermediate analyses.
F. verticillioides growth assay on plates
F. verticillioides strains were first incubated on CM plates with an appropriate resistance marker at 30°C for 7 days from which spores were harvested and counted using a Cell Counter (Beckmann Coulter, Krefeld, Germany). 10 µL containing 10
4 spores was spotted on the middle of fresh CM plates with and without supplementation of doxy (10–50 µg/mL). Each analyzed strain was spotted in triplicates. Plates were incubated for 3–6 days at 30°C. The colony diameters on plates containing doxy were adjusted by considering the growth of the strains in the absence of the inducer. The inhibition rate was then calculated with a formula: inhibition rate (IR) = (
C −
T)/
C × 100 (
68), where
C (control) represents the average growth of the strain without doxy and
T (treated) represents the growth of the respective strain and replicate with doxy. For gene expression analysis, strains were grown for 3 days on CM plates with or without 25 µg/mL doxy. Cells were covered with a layer of cellophane to enable harvesting and subsequent freeze-drying of the mycelium.
Gene expression analysis via quantitative PCR
Fungal cultures were prepared as described above. Induction of 2-day-old ICI/6 mM Gln cultures with 10 µg/mL FB
1 for 2 h was done previously, and expression analysis was essentially performed as previously described (
8). Reactions were run on an Agilent Mx3000P qPCR System with the respective Agilent plastics (Santa Clara, CA, USA). Expression of the genes of interest and of the constitutively expressed reference genes (
FVEG_07930/
FvACT and
YFL039C/
ScACT1 [
69] encoding actin) was determined in triplicates with the primers listed in
Table S1. Expression relative to actin was calculated using the ∆Ct method (
70).
Confocal microscopy
Microscopy experiments were performed on an Axio Observer Spinning Disc Confocal Microscope (Carl Zeiss, Jena, Germany) using 63×/1.2 oil or 100×/1.4 oil objectives with a numerical aperture (NA) value of 0.55. Laser lines of 488 nm and 561 nm were used for fluorophore excitation. Analysis was done with the ZEN 2.6 software, and in postprocessing, microscopy images were adjusted for brightness using the ImageJ software (
https://imagej.nih.gov [
71,
72]). Fungal hyphae were grown out from 10
4 conidia in 300 µL of ICI/6 mM Gln as adherent cultures in 8-well dishes (ibidi, Gräfelfing, Germany) at 30°C for 16 h.
Fum16 expression and purification
In order to render Fum16p soluble for bacterial expression, purification, and
in vitro activity assay, its predicted N-terminal transmembrane domain had to be truncated; this corresponds to amino acids 1–88. Fum16p was expressed in
E. coli BL21 (DE3) as a fusion construct N-terminally tagged with MBP under the control of an inducible T7 promoter (
Fig. S6). A pre-culture was incubated overnight in LB medium containing 25 µg/mL kanamycin (37°C, 180 rpm). Then, 50 mL of Terrific Broth (TB) medium (
73) was inoculated with the pre-culture to an OD
600 of 0.1. The main culture was incubated (37°C, 180 rpm) until an OD
600 of 0.5 was reached. Next, the temperature was lowered to 18°C, expression was induced after 1 h with 1 mM isopropyl β-
d-1-thiogalactopyranoside (Carl Roth, Karlsruhe, Germany), and the culture was incubated overnight. The cell pellet was harvested by centrifugation (4,000 ×
g, 4°C, 20 min) and stored at −20°C until further processing.
Bacterial cells were thawed, resuspended in buffer A (20 mM Tris, 0.2 M NaCl, pH 8.0) containing 0.5 mM 4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF) and EDTA-free protease inhibitor cocktail (Roche, Grenzach-Wyhlen, Germany), and lysed using sonication (Sonopuls 2070, Bandelin, cycle 6, 75% intensity, 2 × 2 min) on ice. The protein preparations were centrifuged (16,000 ×
g, 4°C, 15 min), and the supernatants were applied to an MBP-Trap HP 1 mL column connected to an Aekta FPLC system (both GE Healthcare, Münich, Germany). After washing with 25 mM imidazole, the proteins were eluted with 500 mM imidazole using buffer B (20 mM Tris, 0.2 M NaCl, 10 mM maltose, pH 8.0). Protein-containing fractions were analyzed by Coomassie-stained SDS-PAGE, pooled, and concentrated via an Amicon filter (10 kDa cutoff) to 500 µL. This was applied to a Superdex 200 Increase 10/300 (Cytiva, Freiburg im Breisgau, Germany), and the protein was further purified by size-exclusion chromatography. Again, protein-containing fractions were analyzed (
Fig. S6), pooled, flash-frozen in liquid nitrogen, and stored at −80°C. Protein concentrations were determined using the Bradford assay.
Fum16 in vitro activity assay
The reaction was performed in a 100 µL volume containing 100 mM HEPES (pH 7.5), 0.5 mM TCEP, 2.5 mM MgCl
2, 0.5 mM ATP, 0.5 mM CoA, 5 mM serine, 0.5 mM palmitic acid, 20 µg/mL FadD (purified in references
37,
74), 100 µg/mL SPT (purified in reference
74), and 100 µg/mL Fum16p. The reaction was incubated overnight at 37°C and extracted with 50% (vol/vol) methanol and filtered through a 0.2 µm PTFE filter (Chromafil, Macherey-Nagel, Düren, Germany). The filtrate was then subjected to HPLC-HRMS analysis. The negative control did not contain any SPT enzyme.
S. cerevisiae cell viability assay
FB1 growth inhibition of S. cerevisiae strains expressing F. verticillioides genes FUM8, FUM15, FUM16, and FUM18 was evaluated by monitoring the OD600 every 15 min for up to 24 h. The experiments were conducted in an Infinite M200 plate reader (Tecan, Crailsheim, Germany) with either sterile black 96-well plates (BRAND plates; VWR, Darmstadt, Germany), or sterile black 24-well plates (ibidi, Gräfelfing, Germany). Incubation was performed with SD-Ura at 30°C. Each well contained either 500 µg/mL FB1 (Cayman Chemicals, Ann Arbor, MI) or water as a control, and 0.1 OD600 cells diluted in SD-Ura.
FB1 measurement for F. verticillioides liquid cultures
The supernatant was separated from mycelium through Miracloth and further clarified via centrifugation. Then, 100 µL of the clarified supernatant was mixed with 100% (vol/vol) MeOH and 1% (vol/vol) naringenin as an internal standard (1 mg/mL in methanol; Sigma-Aldrich, Steinheim, Germany) to give 1 mL (thereby diluted 1:10) and filtered through a 0.2 µm PTFE filter.
Filtered mycelia were washed with water, lyophilized, and weighed. Approximately 100 mg of dry mycelium was powdered using liquid nitrogen and resuspended in 1 mL MeOH:CHCl3 (1:1, vol/vol). The mycelium suspension was vigorously shaken (1,400 rpm) at 40°C overnight. Cell debris were pelleted (16,000 × g, 4°C, 15 min), and the supernatants were transferred to a fresh tube. The pellets were resuspended again in 1 mL MeOH:CHCl3 (1:1, vol/vol) and shaken for 1 h. The process was repeated two more times. The combined supernatants were collected, evaporated using SpeedVac, and resuspended overnight in 100% (vol/vol) MeOH. The samples were then filtered with a 0.2 µm PTFE filter prior to HPLC-HRMS analysis.
FB1 measurement for S. cerevisiae liquid cultures
The whole content from each well of the cell viability assay plate (see above) was centrifuged (10,000 × g, 4°C, 5 min) and 100 µL supernatant was mixed with 100% (vol/vol) MeOH to give 1 mL and filtered through a 0.2 µm PTFE filter (thereby diluted 1:10).
Extraction of ceramide intermediates from F. verticillioides and S. cerevisiae
For F. verticillioides, the same extraction method was used as described above for FB1 extraction from mycelium. For S. cerevisiae cells, the obtained pellet (10,000 × g, 4°C, 5 min) was washed twice with 900 µL water and resuspended in 1 mL MeOH:CHCl3 (1:1, vol/vol). The suspension was mixed with 0.5 mm glass beads, and the cells were lysed using a SpeedMill Plus (Analytic Jena, Jena, Germany, two cycles of 1 min). Cell debris were pelleted (10,000 × g, 4°C, 10 min), and the supernatant was collected. The pellet was resuspended again in 1 mL MeOH:CHCl3 (1:1, vol/vol) and lysed once more. The process was repeated once more. The combined supernatants were collected, evaporated using SpeedVac, and resuspended overnight in 100% (vol/vol) MeOH. The samples were then filtered with a 0.2 µm PTFE filter for HPLC-HRMS analysis.
HPLC-HRMS analysis
HPLC-HRMS analysis was performed using an LC-MS system consisting of a Q-Exactive Plus Hybrid Quadrupole Orbitrap mass spectrometer using electrospray ionization and a Dionex UltiMate 3000 UHPLC system (Thermo Fisher Scientific, Dreieich, Germany). Sample separation via HPLC was performed with a Kinetex C
18 column (2.1 × 150 mm, 2.5 µm, 100 Å, Phenomenex) at a flowrate of 0.3 mL/min and column oven at 40°C. For analysis of FB
1 and ceramide intermediates from both
F. verticillioides and
S. cerevisiae, an injection volume of 3 µL was set. The following elution gradient was used for FB
1 detection (solvent A, H
2O plus 0.1% [vol/vol] HCOOH; solvent B, acetonitrile plus 0.1% [vol/vol] HCOOH): 5% B for 0.5 min, 5 to 97% B in 11.5 min, and 97% B for 3 min. The following gradient was used for ceramide intermediates detection: 5% B for 0.5 min, 5 to 97% B in 54.5 min, and 97% B for 3 min. Raw LC-MS data were analyzed using XCalibur (Thermo Fisher Scientific, Dreieich, Germany) with a mass resolution of 10 ppm (
Fig. S3 and S10). FB
1 peak areas were normalized against the internal standard to account for variability between runs. FB
1 and ceramide intermediate levels were related to the dry weight of the
F. verticillioides cultures, which were performed in biological triplicate. The standard curve was prepared by injecting 3 µL of the following FB
1 concentrations prepared in ethanol in triplicate: 0.1 µg/mL, 0.5 µg/mL, 1 µg/mL, 10 µg/mL, and 100 µg/mL.
ACKNOWLEDGMENTS
We thank Daniela Hildebrandt and Ahsan Ullah (Leibniz-HKI) for excellent technical assistance.
This work was supported by the German Research Foundation (DFG grant 453246485 to V.V. and S.J.) and by the Free State of Thuringia and the European Social Fund (project SphinX, 2017FGR0073) to V.V. Work in S.J.’s laboratory is supported by the Free State of Thuringia and the European Social Fund Plus (project FusInfect, 2022FGR0007).
The authors declare no competing interests. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Conceptualization, V.V. and S.J.; methodology, F.G., K.J., Y.H., S.H., S.J., and V.V.; investigation, F.G., K.J., Y.H., S.H., and S.J.; writing – original draft, F.G.; writing – editing, S.J.; writing – review, all authors; funding acquisition, V.V. and S.J.; supervision, V.V. and S.J.