Cytoplasmic sequestering of a fungal stress-activated MAPK in response to a host plant phenolic acid

doi:10.21203/rs.3.rs-3197235/v1

Download PDF

Article

Cytoplasmic sequestering of a fungal stress-activated MAPK in response to a host plant phenolic acid

https://doi.org/10.21203/rs.3.rs-3197235/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The stress-activated MAPK Hog1 in Cochliobolus heterostrophus, a maize foliar pathogen, undergoes dephosphorylation upon exposure to ferulic acid (FA), a phenolic compound abundant in the host plant. Unlike its nuclear localization during osmotic stress, Hog1:GFP forms cytoplasmic foci in response to FA, indicating its sequestering. By using Hog1:GFP as an affinity purification bait, we isolated an FA-dependent sub-proteome from a subcellular fraction enriched with fluorescent foci. The identified proteins include RNA-binding proteins, translation initiation factors and mitochondrial proteins, suggesting the foci to be stress granules. Notably, the RRM and pumilio domain protein Puf2 was enriched, and Puf2:tdTomato formed foci partially colocalizing with Hog1:GFP foci. Hog1:GFP did not colocalize with peroxisomes. Mitochondria, however, which rapidly fragmented upon FA exposure, are partially overlapped with Hog1:GFP foci. The sequestering and dephosphorylation of MAPK Hog1 may collectively attenuate cell death induced by defense compounds released by the plant host.

Biological sciences/Microbiology/Cellular microbiology

Biological sciences/Cell biology/Cell signalling/Stress signalling

Pathogens detect host-derived molecules as signals, providing information on the presence of the host, or of impending stress generated by host defenses. Ferulic acid (FA), an abundant phenolic in maize, triggers rapid cell death in the fungal pathogen Cochliobolus heterostrophus, agent of southern corn leaf blight ¹. Phenolic acids are among barriers that plants deploy against pathogens. Fungal pathogens have evolved the ability to degrade diverse plant defense compounds, which drive selection of metabolic pathways in fungal genomes ^2,3. In Ceratocystidaceae genomes, catechol dioxygenases are correlated with pathogenicity ⁴. Phenolics are, nevertheless, carbon sources for fungal growth, particularly those with vicinal OH groups on the phenolic ring. Phenolics lacking this catechol structure are more toxic, less readily degraded, and perceived by the pathogen as a stress ⁵. The stress-activated MAPK ortholog ChHog1 (High Osmolarity Glycerol 1) is required for full tolerance of this stress in C. heterostrophus. Perhaps surprisingly, ChHog1 is dephosphorylated in response to FA ¹, unlike osmotic stress activation by dual phosphorylation. Hog1 dephosphorylation might mitigate sustained activation, which in yeast causes growth arrest or cell death ^6,7. Indeed, the fungicide fludioxonil kills fungal cells by Hog1 hyperactivation ^8–11. Dephosphorylation, by reducing signaling, could mitigate FA-promoted cell death. A stress-sensitive transcription factor (ChAP1) also participates in the FA response. Comparison of the transcriptomic signatures of wild-type and Chap. 1 mutants suggested that when stress overcomes the ability of the cell to respond, it initiates regulated cell death (RCD) ¹². To return to homeostasis, cells initiate macromolecular repair, adjustment of energy metabolism, growth arrest and, finally, RCD¹³. To investigate the cellular correlates of ChHog1 dephosphorylation, rather than phosphorylation as occurs for osmotic stress, we followed the localization of a C-terminal GFP fusion, discovering ChHog1 accumulation in cytoplasmic foci.

Hog1 localization in response to stress

To follow the cellular localization of C. heterostrophus Hog1 in response to FA stress, we replaced the resident copy with a C-terminal Hog1:GFP fusion by a double-crossover event, retaining the native upstream regulatory region. In the proteomic studies described below, peptides derived from the Hog1 coding sequence and those derived from GFP were detected at similar abundance, confirming that there was no expression of GFP independent of Hog1 (for example from an additional insertion and rearrangement event). The gene replacement did not affect growth, development or virulence (Figure S1A,B). The fusion protein appeared in immunoblots at the expected size for the fusion. Dual phosphorylation of the fusion protein in response to osmotic stress by 1 M sorbitol as well as dephosphorylation in response to 2 mM FA were similar to native WT Hog1: in high osmolarity conditions Hog1 was strongly phosphorylated within 10 min, and dephosphorylated within 10–20 min in exposure to FA as previously reported ¹⁴(Figure S1C). These experiments defined the time frame to test intracellular localization of Hog1:GFP. Hog1 translocated to the nucleus upon exposure to high osmolarity as expected, indicating a functional Hog1:GFP fusion protein (Fig. 1A,B). Surprisingly, upon exposure to FA, Hog1:GFP did not translocate to the nucleus, but rather accumulated in cytoplasmic foci (Fig. 1A,B).

To test the hypothesis that Hog1 could initially be translocated to the nucleus upon exposure to FA, and then aggregate in cytoplasmic foci, Hog1 localization was observed at different times. FA-induced recruitment to foci was also compared to high osmolarity and heat shock stress responses. Upon exposure to 1 M sorbitol in complete medium (CMS), Hog1:GFP localized to the nuclei within 10 minutes. Nuclear retention was transient and reversible: after 60 minutes in CMS, Hog1 dispersed from the nucleus to the cytoplasm and the pattern was indistinguishable from unstimulated controls. Heat shock initiated Hog1 translocation to the nucleus within 10 minutes, after which Hog1 started accumulating in cytoplasmic foci within 30 minutes, and by 60 minutes Hog1 was observed only within foci. 60 minutes after relieving the heat shock, Hog1 dispersed back to the cytoplasm. In contrast, exposure to 2.5 mM ferulic acid caused direct Hog1:GFP localization to bright cytoplasmic foci, starting within 10 minutes, and the pattern observed is very similar to that seen after heat shock (Figure S2). Caffeic acid, which differs from ferulic acid only in that there is an OH at position 3 instead of the methoxy group present in FA, did not induce RCD even at 5 mM ¹. Caffeic acid can be metabolized via the beta ketoadipate pathway but otherwise has similar properties to ferulic acid. Caffeic acid at 2.5 mM did not trigger Hog1:GFP localization to foci (Figure S2). Reaction to FA was highly robust and Hog1 remained in the foci throughout the 60 min stress, and even 60 minutes after relieving the stress by replacing the FA-containing medium with CM only. Hog1:GFP was resolved, partially, from the foci by 6 hours after removal of FA (Figure S3C). Exposure to FA after sorbitol-induced nuclear localization again resulted in formation of fluorescent foci, while exposure to sorbitol after ferulic acid did not relocalize Hog1:GFP from the foci to nuclei (Figure S3A,B). The response to FA is thus dominant over osmotic stress. Release of Hog1:GFP from the nuclei when osmotic stress is relieved is rapid (5 min, Fig. S5A). Sequential osmotic and FA stress resulted in a faster response to FA, with foci visible already after 5 min exposure to FA, while upon exposure to FA alone, foci begin to appear at 10 min and are fully apparent by 20–30 min. Loss of Hog1:GFP from the foci at long times (Figure S3C) could result from protein turnover of Hog1:GFP or of the foci as complete units, or release from the foci back to the cytoplasm. Decreased fluorescence intensity after 6 h recovery would be consistent with turnover of Hog1:GFP, however note that there is no detectable loss of Hog1:GFP on immunoblots up to 40 min (Figure S1C).

Proteomic characterization of Hog1:GFP foci

The similarity of the foci to stress granules (SG) seen during heat shock (Figure S2) suggested that Hog1 is localized to SG upon exposure to FA. To investigate the composition of the foci, we isolated a subcellular fraction by a differential centrifugation protocol designed to enrich SG ¹⁵. We refer to this fraction (Figs. 1C, 2B) as “crude SG” since it contains additional components: peroxisomes, mitochondria, and any other structures or complexes sedimenting under the same conditions.

To determine the relation of the Hog1:GFP foci to peroxisomes, a mCherry protein whose C-terminus is the peroxisomal targeting signal SKL was expressed under the constitutive fungal Pgpd1 promotor. The peroxisome reporter strain was exposed to FA. Marked spherical organelles that tend to be larger than the Hog1:GFP-labelled foci were observed in a pattern distinct from the Hog1:GFP foci (Figure S4A). Thus Hog1 is not sequestered to peroxisomes.

To follow the relationship between Hog1:GFP and mitochondrial dynamics upon exposure to FA stress, cells were exposed to 2.5 mM FA and then stained with Mitotracker™ Deep Red. FA treatment resulted in mitochondrial fragmentation. When cells were treated with DM or caffeic acid (CA), mitochondrial fragmentation was not observed even after 30 minutes. Hog1:GFP increasingly colocalized with fragmented mitochondria at 10, 20 and 30 min (Figure S4B).

The crude SG fraction appears, visually, enriched for Hog1:GFP fluorescence, containing fluorescent puncta that are of similar size (about 0.5 µm) to the cytoplasmic foci seen in the cell. There are more such fluorescent particles visible in fractions isolated after FA treatment than from controls (Fig. 1C,E). After FA, most of the Hog1 signal is found in the pelleted crude SG fraction, and it is dephosphorylated relative to the initial state (Figs. 1D,E). Proteomic analysis of this fraction (Fig. 2A, S5A) identified 523 proteins (out of 2511 identified by at least two peptides) that were significantly elevated in the FA foci fraction compared to the control (three biological repeats, 12 technical repeats), Table S2, Fig. 2A). The FA-enriched proteins include, as expected, many functional classes. These include known stress-granule-associated proteins, for example: RNA processing, P bodies and translation-related proteins and MCM complex proteins. RNA binding proteins include known stress-granule markers such as MCM proteins, Tif1, Ded1¹⁵ and Jsn1/Puf2¹⁶ (Table S2).

To better characterize Hog1 foci, we subjected the crude SG fraction from FA-treated cells of the GFP-tagged Hog1 strain, to co-immunoprecipitation (Co-IP) by anti-GFP magnetic beads (Fig. 2B). The same fraction collected from DMSO-treated cells (DMSO) was analyzed as a control for FA-specific enrichment. A crude SG fraction extracted from an untagged strain (C4 WT) exposed to FA served as an additional control to take into account non-specific interactions with the beads (Figure S5C). Isolated proteins (Bound, see titles, Fig. 2C) from three independent biological repeats were subjected to MS/MS analysis. Overall, 357 proteins were elevated by more than two-fold in the FA-treated cells compared to the control. In addition, 290 proteins were enriched in the Bound sample of Hog1:GFP compared to the Untagged (C4) strain. Ninety proteins passed a double-sieve selection of being enriched compared to both controls (Fig. 2D, S5B,D). For 57 of these 90, coIP gave stronger enrichment (FA vs control) than observed in the initial crude SG fraction (Figure S5D). Molecular Function Gene Ontology (GO) term enrichment analysis (using DAVID Version 6.8) among the 90 protein set revealed enrichment of RNA binding proteins, particularly RNA recognition motif (RRM) - containing; ATPase activity, 5-methyltetrahydropteroyltriglutamate-homocysteine S-methyltransferase activity, translation initiation factor activity and ATP binding (Fig. 2C-E, Table S2).

We used SGD BLASTP to find the Yeast homologs of the enriched proteins. Overall, 77 proteins have significant Yeast homologs. Twelve of those proteins (~ 15.5%) were previously reported to assemble into SG upon heat shock (HS) ¹⁵¹⁷. This is a significant enrichment (4.12 fold, hypergeometric p-value = 2.6e-5) compared to a random sample.

Co-enrichment of Hog1 with SG markers supports the hypothesis that Hog1 is sequestered in SG or related subcellular structures. Puf2 is an RNA-binding, pumilio-domain SG-associated protein whose abundance was strongly FA-dependent (Table S2). A predicted ortholog of yeast Puf2 was identified and expressed as a C-terminal fusion to tdTomato. The C-terminal tdTomato Puf2 fusion protein localized to foci upon exposure to FA. In live imaging, starting about 30–40 min after application of FA the Puf2:tdT fluorescence signal increased and localized to cytoplasmic foci (Fig. 3A). A subset, but not all, of Hog1:GFP labelled foci are co-localized with tdTomato:Puf2 (red channel, Fig. 3A,B, Supplementary Video 2). In fixed samples, two-fold higher intensity of the Puf2:tdTomato signal is evident at 25 min (Fig. 3B). The simplest interpretation is that upon exposure to FA, Hog1:GFP localizes to SG or another LLPS structure which then imports Puf2 or fuses with Puf2-containing SG. Formation of foci containing a known SG protein, Puf2 ¹⁶, is consistent with FA being a stress signal. In agreement with the time course (Fig. 3), deletion of Puf2 did not prevent formation of Hog1:GFP – labelled foci, implying that Puf2 is not essential for the sequestering of Hog1 (Fig. 4B). Co-immunoprecipitation (Co-IP) analysis of Hog1:GFP enriched for mitochondrial proteins, suggesting an association between Hog1 and mitochondria under FA-induced stress, consistent with colocalization (Figure S4B).

Relationship between dephosphorylation and foci formation

We then asked whether Hog1 dephosphorylation upon exposure of the cells to FA occurs before or after incorporation into cytoplasmic foci. Immunofluorescence images of fixed cells probed with antibody to dual phosphorylated Hog1 show a basal level of cytoplasmic and nuclear localization. Osmotic stress strongly increased dual phosphorylation combined with nuclear retention as shown by strongly fluorescent nuclei, particularly in appressoria (Fig. 4A). The basal level of dually phosphorylated Hog1 staining was not affected by exposure to DMSO solvent or CA as controls, but was reduced by FA. Dually phosphorylated Hog1 at the locations of the Hog1:GFP foci was not detectable by immunofluorescence, although this signal is just visible in immunoblots (Fig. 1D,E). Conversely, dephosphorylation might be required for incorporation into foci. This was tested by FA treatment of a PtcB deletion strain that expresses the Hog1:GFP fusion protein (PtcB is a Hog1 phosphatase ¹⁴. Formation of foci was normal in the PtcB deletion strain (Fig. 4B), so FA-induced dephosphorylation, which is absent or strongly reduced when PtcB is deleted, is not a prerequisite for localization of Hog1 to the foci. Likewise, lack of the phosphatase CDC14, also required for Hog1 dephosphorylation ¹⁴ does not affect foci formation (Fig. 4B). Immunoblot analysis showed that Hog1 is significantly dephosphorylated in the FA foci fraction as compared to the control (Fig. 1D,E).

Ferulic acid (FA), a phenolic acid abundant in maize, the host plant of C. heterostrophus, acts on the pathogen as a stress, causing rapid cell death with hallmarks of apoptotic-like RCD ¹. Although this remains to be tested in planta, high levels of FA in the maize host cell walls and leaf cuticle might contribute to the (paradoxical) extensive cell death observed in necrotrophic fungal pathogens like C. heterostrophus during their initial attack on the plant host, when one might have expected maximal metabolism and growth needed to support invasion ¹⁸.

Stressors generally activate MAPK P38 orthologs like Hog1 by dual phosphorylation and nuclear import. We found here that FA, instead, sequesters Hog1 to cytoplasmic foci. Hyperactivation of fungal Hog1 by stressors or fungicides such as arsenite or fludioxonil is the main route to cell death caused by these compounds ^{19 20}. Indeed, overexpression of the protein phosphatase MoPtp2 in the rice blast pathogen Magnaporthe oryzae lowers Hog1 phosphorylation and counteracts cell death caused by fludioxonil ²¹. Hyperactivation occurs by increased dual phosphorylation of the stress-activated MAPK. FA stress causes, instead, dephosphorylation of Hog1, which, we propose, would act to mitigate the cell-death-promoting activity of this plant defense compound. In the encounter between maize leaves and the fungus, FA is released from the cell wall and cuticle. In agreement with our model, exposure of the fungal hyphae to FA indeed results in dephosphorylation of Hog1 ¹⁴, Figure S1C). Dephosphorylation of Hog1/SakA/P38 is developmentally regulated in the initiation of Aspergillus nidulans spore germination ²².

In addition to dephosphorylation, a sensitive mechanism to dampen Hog1 activity and prevent hyperactivation could be sequestered to a compartment where it cannot interact with downstream partners or enter the nucleus. Hog1:GFP relocalizes to foci that appear similar to stress granules (SG), which are cytoplasmic mRNP granules formed from mRNAs stalled in translation initiation when inhibited by different chemicals or by stress responses. Other Liquid-Liquid Phase Separation (LLPS) compartments include P-bodies which are distinguished from SG by their bias towards mRNA decay components rather than translation initiation factors^{2324 25}. SGs have been proposed to intercept and sequester signaling components ²⁶. Exposure of Candida boidinii, Pichia pastoris and Schizosaccharomyces pombe to high-temperature stress resulted in colocalization of the CbHog1 with the stress granule marker Pab1 ²⁷. In other cases, Hog1 was found to affect cellular processes linked to SG formation, though indirectly ²⁸. It was also shown that SGs containing Hog1 are induced when yeast cells begin to enter stationary phase ²⁹. In mammalian cell lines, sequestering of the signaling scaffold protein RACK1 to SGs under some stresses inhibits P38 MAPK-mediated apoptosis ³⁰, precedent for our hypothesis that sequestering of Hog1 can dampen its activity. Sequestering of the mRNA components rather than (or in addition to) signaling proteins provides direct control of translation as shown recently in the Neurospora crassa circadian clock ³¹.

Proteomic analysis showed significant enrichment of RNA binding proteins and translation initiation factors in Hog1:GFP foci (Fig. 2E). The yeast orthologs of two of the RRM domain RNA binding proteins, MRN1 and NRP1, were previously found in the SG proteome in two independent studies ^{15 17}. Moreover, Hog1:GFP foci content is significantly enriched with 12 proteins (including the above two) previously associated with SGs. To date, the best-understood mechanism for SG formation involves the stress-induced phosphorylation of the translation initiation factor 2 alpha subunit (eIF2α), which subsequently hampers translation initiation ³². Notably, eIF2α (N4X6B8) exhibited enrichment in Hog1:GFP foci Co-IP.

An enriched protein in a pelletable fraction following FA treatment and in Hog1:GFP foci Co-IP is Puf2, a SG-associated RNA-binding protein (Fig. 2C, Table S2). Previous studies showed that Puf2 exhibits translational repressor activity³³. This raises the possibility that Puf2 represses translation of specific mRNAs associated with Hog1 FA stress-dependent structures. We therefore expressed a fluorescent Puf2 fusion protein construct replacing the resident Puf2 gene, and found that by 30 min exposure to FA, fluorescence signals of Hog1-containing and Puf2-containing granules overlap extensively though not entirely (Fig. 3). Incomplete co-localization may reflect the different time course for granule formation and incorporation of Hog1 (Fig. 3), or heterogeneity in the granule population, with some Hog1:Gfp-containing granules enriched for Puf2 and some not. Considering the dynamic nature of LLPS, either or both explanations are plausible. If sequestering of Hog1 acts to dampen its hyperactivation under stress, this raises the question of how sequestering and dephosphorylation are integrated. We found that the FA signal is dominant (in the biochemical sense) to osmotic stress: exposure to sorbitol could not release Hog1 from the granules, while after exposure to FA, Hog1 exited the nucleus and accumulated in granules, even under osmotic stress. Furthermore, Hog1 is sequestered primarily in its dephosphorylated state (Figs. 1D and 4A,B). This suggests that dephosphorylation of activated Hog1 might trigger its incorporation into SG. The mechanism, however, is likely more complex, because distinct signaling hierarchies could lead to granule formation and Hog1 sequestering. Indeed, Puf2 deletion mutants still sequester Hog1 (Fig. 4C). Likewise, mutants lacking phosphatases PtcB or CDC14, required for rapid dephosphorylation of Hog1, still sequester Hog1 (Fig. 4C).

FA induced rapid mitochondrial fragmentation within 10 min, as seen for example in Aspergillus fumigatus following oxidative stress³⁴. The pattern of Hog1:GFP foci is distinct from fragmented mitochondria, though both fluorescence signals are abundantly distributed and there is some overlap. The dynamic nature of LLPS and the observation that they can contain key components of signaling pathways, forms a cyclic relationship in which SG or other LLPS can alter signaling and protein methylation, phosphorylation, and glycosylation which in turn influence SG assembly ^{25,35 35,36} Conversely, formation or disassembly of stress-induced condensates may be regulated by kinases as in the case of Grb7 phosphorylation in carcinoma cells ²⁴. In Schizosaccharomyces pombe, the NDR/LATS kinase Orb6 phosphorylates the RNA-binding protein STS5 under non-stress conditions. Dephosphorylation of STS5 under stress conditions permits stress granule formation ^37,38.

Given the key role of MAPK import to the nucleus in eukaryotic cells in general, we expect that signal-dependent sequestration of the stress-activated Hog1 (P38 ortholog) kinase, as found here, will have implications well beyond the fungal kingdom, in normal developmental cascades and in disease.

C. heterostrophus strains

Strains used in this study and oligonucleotide sequences are given in Table S1. Deletions were made using either the split marker protocol or via transformation with a plasmid derived from pUCATPH ³⁹containing the 5’ and 3’ flanks of the gene of interest flanking the hygromycin phosphotransferase (Hyg) expression cassette. The amino acid sequence the ScPUF2 RNA-binding protein gene retrieved from the Uniprot database (http://www.Uniprot.org/) and C. heterostrophus homologs of these genes were identified using BLASTP searches of the JGI database Home - Cochliobolus heterostrophus C4 v6.0 (doe.gov) ^40,41. DNA fragments of these genes were amplified using the oligonucleotides given in Table S1. GFP tdTomato and mCherry tagging protein sequences were retrieved from Addgene (https://www.addgene.org/), their sequence was optimized for C. heterostrophus codon usage using the IDT (https://eu.idtdna.com) codon optimization tool according to the genome of the related Dothideomycete Alternaria alternata, and the optimized genes were synthesized at IDT (https://eu.idtdna.com) synthesis gBlocks services, with appropriate restriction sites. Bait protein and tag sequences were restriction digested and ligated into pNG1 containing a neomycin/G418 resistance cassette (kindly provided by Prof. B. Gillian Turgeon, Cornell University) (for the ChHog1:GFP protein) or into pUCATPH containing a hygromycin B resistance cassette³⁹ (for the ChPuf2::tdTomato protein). A Pgpd1:mCherry:SKL fusion protein was expressed from a synthesized sequence and used to label peroxisomes.

Generation and Transformation of Fungal Protoplasts

Plasmid DNA was linearized and introduced into the fungi by PEG – Ca⁺⁺ mediated protoplast transformation ⁴². For the establishment of the HGP513 strain, C. heterostrophus race T strain C4 was grown in 100 mm Petri dishes with CMX (as CM, except glucose was replaced by D-xylose) medium, at 22°C for about 2 weeks, in a 12 h light/12 h dark regime. These plates were used for preparation of protoplasts. Spores were inoculated to 150 mL CM and cultured in a rotary shaker for 18 h at 30°C, 200 rpm. The fungal suspension was centrifuged at 8,000 rpm for 10 min (Sorvall, SLA1500 rotor). The resulting mycelium sediment was suspended with 70 ml of enzyme-osmoticum solution (see (Turgeon et al. 2010)(Turgeon et al. 2010)(Turgeon et al. 2010)) and divided into five 125 ml Erlenmeyer flasks followed by incubation for 2.5 h at 30°C, 70 rpm. The released protoplasts were filtered through gauze fabric. The filtrate was then centrifuged at 4500 rpm for 5 min at 4°C. The protoplast pellets were washed with 10 mL cold 0.7 M NaCl, followed by a second wash in 10 mL STC. The resulting protoplasts were suspended in 200 µl of STC, counted and adjusted to a concentration of 10⁸ cells/mL. Of these, 10⁷ protoplasts were used for each transformation. The protoplasts were incubated on ice with 25 µg of DNA for 12 min (control group contained no DNA yet was subjected to the same procedure). 60% w/v MW 3,350 PEG (Sigma) was added in three aliquots of 200, 200, and 800 µL each and then diluted with 1 mL STC and plated for regeneration. A 1% agar overlay containing 1 mg/ml G418 (Gibco) (Neo) was added after 18 h for selection. For protoplast viability determination, one of the control plates was overlayed with agar without Neo. All additional strains were established on the basis of the HGP513 strain using hygromycin B (HYG) 100 µg/ml as a secondary selection ⁴².

Single-spore isolation

Newly formed colonies after transformation were transferred to 10 mL CMX-Neo 1 mg/ml (for the HGP513 strain) or CMX-Neo 1mg/ml – Hyg 100 µg/ml (for all additional strains) plates. Mature spores were collected, diluted in sterile water so that up to 10 spores were dispersed on a 25ml CMX-Neo 1mg/ml (for the HGP513 strain) or CMX-Neo 1mg/ml – Hyg 100 µg/ml (for all additional strains) plates allowing new colonies to be formed - each from a single spore.

Genomic Verification

The replacements of genes by the fusion construct (for the HGP513; HGP513-PUF2-tdTomato strains) were verified by PCR. Each fusion was tested with a primer upstream to the gene paired with a primer corresponding to the end of the tag (e.g. Hog1-Up Stream & GFP-EcoRI rv). The replacements of genes by Hygromycin resistance cassette (for the HGP513-ΔchPtcB, HGP513-ΔchCDC14 and HGP513-ΔchPuf2 strains) were verified by PCR in 2 ways. First, with primers designed from the gene sequence (ORF) verifying the deletion of the gene, and second, with primers upstream and downstream of the gene’s flanks used to construct the knockout mutant, and primers within the hygromycin phosphotransferase coding sequence, verifying that the hygromycin resistance cassette was inserted in the correct location in the genome.

Stress induction

C. heterostrophus cells expressing the above fusion proteins were grown on selective plates as indicated for 2 weeks. Spores were collected with sterile water, filtered through sterile gauze and 2*10⁵ cells in 25 ml sterile water were incubated in a 150 mm plate, containing microscopy cover slips, for 5–6 h until spores attached to the surface and germinated. Then, the water was replaced with complete medium (CM) and spores were allowed to germinate and grow for 16 hr. Osmotic stress was applied by the replacement of the CM with either new CM (control) or CMS (CM supplemented with 1 M sorbitol). for the indicated time at RT. FA was appliedor by the replacement of the CM with either DM (CM supplemented with the same volume of DMSO, the solvent for ferulic acid stock, as the FA volume added) or FA (CM supplemented with 2.5 mM ferulic acid) for the indicated time at RT. For the heat shock (HS) treatment the CM medium was replaced by CM preheated to 45°C and incubated for 10, 30 or 60 min at 45°C. Cells were then fixed in 4% paraformaldehyde in PBST for half an hour, washed with PBST and imaged. Live imaging was also performed for both GFP and tdTomato constructs.

Immunohistochemistry and confocal fluorescence microscopy

Cells were induced as indicated and then rinsed three times with PBST and their cell wall was permeabilized by digestion with enzyme solution consisting of 67.5 mg/ml glucanase (VinoTaste, Novo, Denmark), 22 mg/ml Driselase (Sigma) and 2 mg/ml chitinase (Sigma) in PBS, on ice for 15 min. Cells were rinsed 3 times with PBST and a second permeabilization was done by adding 200 µl 0.1% Triton X-100 in 0.1% sodium citrate onto the slide, for 2 min. Cells were rinsed again and blocked with 3% BSA in PBST. Dually phosphorylated Hog1 was detected by p-p38 antibody (Cell Signaling, T180-Y182, 9211S) (1st Ab 1:400, 1 h) and anti-rabbit Alexa Fluor™ 568 (2nd Ab 1:1000, 1 h). Mounting was done using DAPI Fluoromount-G (SouthernBiotech CAT NO 0100 − 20). To visualize mitochondria, samples were stained with 1:20000 Mitotracker™ Deep Red FM (M46753, Thermo-Fisher) in CM for 5–30 min. Samples were imaged by Spinning Disk confocal (Yokogawa CSU-W1) microscope (Nikon) with a CFI PLAN APOCHROMAT 100x OIL objective (N.A-1.45). DAPI, FITC (GFP), Alexa Fluor™ 568 and iRFP720 signals were captured by exciting with laser lines: 405, 488, 561 and 640 nm, respectively. Emission signals were collected with sCMOS camera (Photometrics, PRIME –BSI) with 95% QE by using NIS software. Images were collected in Z-stack mode with 0.2–0.4 µm steps through 8–15 µm. Deconvolution was done using NIS-elements for images of strains carrying GFP constructs only. For images of strains carrying both GFP and tdTomato by NIS Denoise AI algorithm was applied. Data presented were obtained with the same imaging settings namely laser power, light path and camera exposure time from at least three independent experiments, all of which showed similar effects.

SG production from induced samples and confocal fluorescence microscopy

Cultures were grown as described above in 150 mm plates. Exposure to 2.5 mM FA or DMSO (solvent-only control) was performed as described above. Cells were scraped immediately, and flash frozen in liquid nitrogen. Native protein extraction was performed using native buffer: 50mM Tris HCl pH 7.4, 100 mM potassium acetate, 2 mM magnesium acetate, 0.5 mM DTT, 0.5% NP40, 0.5 mM EDTA, 50 mM NaF, 50 mM β glycerophosphate, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate, aprotinin (Sigma A6279) 1:500, 0.005 mM Pepstatin A (P5318), 5 mM Leupeptin (L2884), 5 mM 110-Phenanthroline (131377), as described previously(Jain et al. 2016)(Jain et al. 2016)(Jain et al. 2016) with slight modifications. Briefly, 800 mg cells were ground in liquid nitrogen and then suspended in cold native buffer in a 1.5:1 ml:mg tissue ratio, vortexed vigorously for 1 min and let sit on ice for 3 min, repeating 3 times. Crude SG fraction was isolated by sequential centrifugation: nuclear fraction and debris were removed by 2 centrifugations of 850 g, 4°C, 2’ and additional 1500g, 4°C, 5’ centrifugation. The supernatant was centrifuged at 18000 xg, 4°C, 15’ after which the pellet contained the crude SG fraction. The total SG fraction was resuspended in 100 µl of native buffer and cleared again from residual debris by centrifugation at 1000 x g, 4°C, 2 min. Supernatant containing cleared SG was collected for further analysis by confocal microscopy, proteomic analysis and immunoblot analysis (Fig. 2B, 1C).

Confocal microscopy was performed on a 5 µl sample, imaged with a Spinning Disk confocal (Yokogawa CSU-W1) microscope (Nikon) by CFI PLAN APOCHROMAT 100x oil objective (NA-1.45). Excitation for the FITC (GFP) channel was with the 488 nm laser. Emission signal was collected with a sCMOS camera (PRIME –BSI) with 95% QE by using NIS software. Data presented were obtained with identical imaging settings from at least three independent experiments, all of which showed similar trends. Quantification of maximal intensity projection images was performed using IMARIS software (Oxford Instruments).

SG Fraction Co-Immunoprecipitation

The crude stress granule fraction was loaded on GFP-Trap_MB (ChromoTek, gtmb) and rotated at 4°C for 2 hr. Samples were washed three times with lysis buffer (50mM Tris HCl pH 7.4, 100 mM potassium acetate, 2 mM magnesium acetate, 0.5 mM DTT, 0.5% NP40, 0.5 mM EDTA, 50 mM NaF, 50 mM β glycerophosphate, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate, aprotinin (Sigma A6279) 1:500, 0.005 mM Pepstatin A (P5318), 5 mM Leupeptin (L2884), 5 mM 110-Phenanthroline (131377)). Samples were eluted with 40 µl 2X Laemmli sample buffer.

Proteolysis and mass spectrometry analysis

60 µl of the SG fraction samples was mixed with 15 µl of 5X Laemmli sample buffer and loaded on a 4–15% Mini-PROTEAN® TGX™ Precast Protein Gel (BioRad, cat # 4561084). The proteins in the gel were reduced with 2.8 mM DTT (60°C for 30 min), modified with 8.8 mM iodoacetamide in 100 mM ammonium bicarbonate (in the dark, room temperature for 30 min) and digested in 10% acetonitrile and 10 mM ammonium bicarbonate with modified trypsin (Promega) at a 1:50 enzyme-to-substrate ratio, overnight at 37°C. Second digestion was performed with 1:100 enzyme-to-substrate ratio, 4 hr at 37°C. The resulting tryptic peptides were desalted using C18 tips (Glygen^™ C18 TOP-TIP), dried, and re-suspended in 0.1% formic acid. First biological repeat, including 3 technical repeats, run on a Q Exactive mass spectrometer (Thermo) in a positive mode using repetitively full MS scan followed by collision induces dissociation (HCD) of the 10 most dominant ions selected from the first MS scan. The peptides were resolved by reverse-phase chromatography on 0.075 X 300-mm fused silica capillaries (J&W) packed with Reprosil reversed phase material (Dr Maisch GmbH, Germany). The peptides were eluted with linear 120 minutes gradient of 5 to 28% 15 minutes gradient of 28 to 95% and 25 minutes at 95% acetonitrile with 0.1% formic acid in water at flow rates of 0.15 µl/min. Second biological repeat, including 3 technical repeats, run on a Q Exactive HF mass spectrometer (Thermo) in a positive mode using repetitively full MS scan followed by collision induces dissociation (HCD) of the 20 most dominant ions selected from the first MS scan. The peptides were resolved by reverse-phase chromatography on 0.075 X 300-mm fused silica capillaries (J&W) packed with Reprosil reversed phase material (Dr Maisch GmbH, Germany). The peptides were eluted with linear 120 minutes gradient of 6 to 30% 15 minutes gradient of 30 to 95% and 15 minutes at 95% (80% acetonitrile with 0.1% formic acid in water) at flow rates of 0.15 µl/min. The mass spectrometry data from all repeats was analyzed using the MaxQuant software 1.5.2.8 (Mathias Mann's group) vs the Cochliobolus heterostrophus entry of the Uniprot database with 1% FDR (false discovery rate). The data were quantified by label free analysis using the same software. Statistical analysis of the identification and quantitation of results was done using Perseus 1.5.2.4 software (Mathias Mann's group). Statistically differential proteins were defined as: at least two peptides, p-value < 0.05, log fold change > 1.

Immunoblots

Frozen mycelium of induced samples was ground in liquid nitrogen, suspended in an SDS buffer containing 5% SDS, 100 mM Tris pH 8 and 10 mM DTT, heated for 5’ at 95°C and sonicated (Microson XL-Misonic, 30s, level 5). The extract was centrifuged (10000 xg, 10’, RT) and the supernatant acetone-precipitated. The resolved protein pellet was solubilized by urea buffer containing 8 M urea, 100 mM ammonium bicarbonate and 10 mM DTT. Protein concentration was determined using Bradford assay and 30 µg of each extract was loaded on a 10% SDS-PAGE. For the SG fraction, 20 µl was mixed with 5X Laemmli sample buffer and loaded on 10% SDS-PAGE. Transfer was performed using Trans Blot Turbo protein transfer kit & system (BioRad). Blots were probed with either custom produced ChHog1 antibody (https://www.biomatik.com/) or p-p38 antibody (Cell Signaling, T180-Y182, 9211S) and imaged using Fusion Pulse (VILBER).

Acknowledgements: R.Z. gratefully acknowledges graduate fellowships from the Irwin and Joan Jacobs Graduate School, Technion and ISEF. This study was funded by Israel Science Foundation grants 420/17 and 927/22. Microscopy at the Life Sciences & Engineering center was funded in part by The Russell Berrie Nanotechnology Institute, Technion. B.A.H. is incumbent of the Feinberg Academic Chair.

Data availability statement: All relevant data are included in the manuscript and supplementary files. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [1] partner repository with the dataset identifier PXD044457.

Shalaby, S., Larkov, O., Lamdan, N.-L., Goldshmidt-Tran, O., Horwitz, B.A.: Plant phenolic acids induce programmed cell death of a fungal pathogen: MAPK signaling and survival of Cochliobolus heterostrophus. Environ. Microbiol. 18, 4188–4199 (2016)
Gluck-Thaler, E., Slot, J.C.: Specialized plant biochemistry drives gene clustering in fungi. ISME J. 12, 1694–1705 (2018)
Gluck-Thaler, E., Vijayakumar, V., Slot, J.C.: Fungal adaptation to plant defences through convergent assembly of metabolic modules. Mol. Ecol. 27, 5120–5136 (2018)
Soal, N.C., Coetzee, M.P.A., van der Nest, M.A., Hammerbacher, A., Wingfield, B.: D. Phenolic degradation by catechol dioxygenases is associated with pathogenic fungi with a necrotrophic lifestyle in the Ceratocystidaceae. G3 (Bethesda) 12, (2022)
Shalaby, S., Horwitz, B.A., Larkov, O.: Structure–Activity Relationships Delineate How the Maize Pathogen Cochliobolus heterostrophus Uses Aromatic Compounds as Signals and Metabolites. Mol. Plant Microbe Interact. 25, 931–940 (2012)
Yaakov, G., Bell, M., Hohmann, S., Engelberg, D.: Combination of Two Activating Mutations in One HOG1 Gene Forms Hyperactive Enzymes That Induce Growth Arrest. Mol. Cell. Biol. 23, 4826–4840 (2003)
Vendrell, A., et al.: Sir2 histone deacetylase prevents programmed cell death caused by sustained activation of the Hog1 stress-activated protein kinase. EMBO Rep. 12, 1062–1068 (2011)
Yoshimi, A., Kojima, K., Takano, Y., Tanaka, C.: Group III histidine kinase is a positive regulator of Hog1-type mitogen-activated protein kinase in filamentous fungi. Eukaryot. Cell. 4, 1820–1828 (2005)
Izumitsu, K., Yoshimi, A., Tanaka, C.: Two-component response regulators Ssk1p and Skn7p additively regulate high-osmolarity adaptation and fungicide sensitivity in Cochliobolus heterostrophus. Eukaryot. Cell. 6, 171–181 (2007)
Furukawa, K., Randhawa, A., Kaur, H., Mondal, A.K., Hohmann, S.: Fungal fludioxonil sensitivity is diminished by a constitutively active form of the group III histidine kinase. FEBS Lett. 586, 2417–2422 (2012)
Bohnert, S., et al.: Fungicide resistance toward fludioxonil conferred by overexpression of the phosphatase gene MoPTP2 in Magnaporthe oryzae. Mol. Microbiol. 111, 662–677 (2019)
Simaan, H., et al.: The AP-1-like transcription factor ChAP1 balances tolerance and cell death in the response of the maize pathogen Cochliobolus heterostrophus to a plant phenolic. Curr. Genet. 66, 187–203 (2020)
Kültz, D.: Defining biological stress and stress responses based on principles of physics. 350–358 doi: (2020). 10.1002/jez.2340
Zuchman, R., Koren, R., Horwitz, B.A.: Developmental roles of the hog1 protein phosphatases of the maize pathogen Cochliobolus heterostrophus. J. Fungi. 7, 1–20 (2021)
Jain, S., et al.: ATPase-Modulated Stress Granules Contain a Diverse Proteome and Substructure. Cell. 164, 487–498 (2016)
Hsiao, W.-Y., Wang, Y.-T., Wang, S.-W.: Fission Yeast Puf2, a Pumilio and FBF Family RNA-Binding Protein, Links Stress Granules to Processing Bodies. Mol. Cell. Biol. 40, (2020)
Kuechler, E.R., Budzyńska, P.M., Bernardini, J.P., Gsponer, J., Mayor, T.: Distinct Features of Stress Granule Proteins Predict Localization in Membraneless Organelles. J. Mol. Biol. 432, 2349–2368 (2020)
Shlezinger, N., et al.: Anti-apoptotic machinery protects the necrotrophic fungus botrytis cinerea from host-induced apoptotic-like cell death during plant infection. PLoS Pathog 7, (2011)
Kojima, K., et al.: Fungicide activity through activation of a fungal signalling pathway. Mol. Microbiol. 53, 1785–1796 (2004)
Lee, J., Levin, D.E.: Intracellular mechanism by which arsenite activates the yeast stress MAPK Hog1. Mol. Biol. Cell. 29, 1904–1915 (2018)
Bohnert, S., Neumann, H., Thines, E., Jacob, S.: Visualizing fungicide action: an in vivo tool for rapid validation of fungicides with target location HOG pathway. Pest Manag Sci. 75, 772–778 (2019)
Jaimes-Arroyo, R., et al.: The SrkA kinase is part of the SakA Mitogen-activated protein kinase interactome and regulates stress responses and development in Aspergillus nidulans. Eukaryot. Cell. 14, 495–510 (2015)
Sheth, U., Parker, R.: Targeting of Aberrant mRNAs to Cytoplasmic Processing Bodies. Cell. 125, 1095–1109 (2006)
Buchan, J.R., Parker, R.: Eukaryotic Stress Granules: The Ins and Outs of Translation. Mol. Cell. 36, 932–941 (2009)
Protter, D.S.W., Parker, R.: Principles and Properties of Stress Granules. Trends Cell. Biol. 26, 668–679 (2016)
Kedersha, N., Ivanov, P., Anderson, P.: Stress granules and cell signaling: more than just a passing phase? Trends Biochem. Sci. 38, 494–506 (2013)
Shiraishi, K., Hioki, T., Habata, A., Yurimoto, H., Sakai, Y.: Yeast Hog1 proteins are sequestered in stress granules during high-temperature stress. J. Cell. Sci. 131, (2018)
Yahya, G., et al.: Stress granules display bistable dynamics modulated by Cdk. J. Cell Biol. 220, (2021)
Nostramo, R., Xing, S., Zhang, B., Herman, P.K.: Insights into the role of P-bodies and stress granules in protein quality control. Genetics. 213, 251–265 (2019)
Arimoto, K., Fukuda, H., Imajoh-Ohmi, S., Saito, H., Takekawa, M.: Formation of stress granules inhibits apoptosis by suppressing stress-responsive MAPK pathways. Nat. Cell. Biol. 10, 1324–1332 (2008)
Castillo, K.D., et al.: A circadian clock translational control mechanism targets specific mRNAs to cytoplasmic messenger ribonucleoprotein granules. Cell. Rep. 41, (2022)
Anderson, P., Kedersha, N.: Stressful initiations. J. Cell. Sci. 115, 3227–3234 (2002)
Reynaud, K., McGeachy, A.M., Noble, D., Meacham, Z.A., Ingolia, N.T.: Surveying the global landscape of post-transcriptional regulators. Nat. Struct. Mol. Biol. 2023, 1–13 (2023). 10.1038/s41594-023-00999-5
Ruf, D., Brantl, V., Wagener, J.: Mitochondrial Fragmentation in Aspergillus fumigatus as Early Marker of Granulocyte Killing Activity. Front. Cell. Infect. Microbiol. 8, (2018)
Youn, J.Y., et al.: Properties of Stress Granule and P-Body Proteomes. Mol. Cell. 76, 286–294 (2019)
Buchan, J.R., Parker, R.: Eukaryotic Stress Granules: The Ins and Outs of Translation. Mol. Cell. 36, 932–941 (2009)
Chen, C., Pino, M.R., Haller, P.R., Verde, F.: Conserved NDR/LATS kinase controls RAS GTPase activity to regulate cell growth and chronological lifespan. Mol. Biol. Cell. 30, 2598–2616 (2019)
Tams, R.N., Chen, C., Nuñez, I., Haller, P.R., Verde, F.: The Orb6-Sts5 Axis Regulates Stress Granule Formation and Heat Stress Response in Fission Yeast. bioRxiv (2021).
02.26.432566
(2021) doi:10.1101/2021.02.26.432566
Lu, S., et al.: Tagged mutations at the Tox1 locus of Cochliobolus heterostrophus by restriction enzyme-mediated integration. Proc. Natl. Acad. Sci. U S A. 91, 12649–12653 (1994)
Grigoriev, I.V., et al.: MycoCosm portal: Gearing up for 1000 fungal genomes. Nucleic Acids Res. 42, D699–D704 (2014)
Ohm, R.A., et al.: Diverse Lifestyles and Strategies of Plant Pathogenesis Encoded in the Genomes of Eighteen Dothideomycetes Fungi. PLoS Pathog. 8, e1003037 (2012)
Turgeon, B.G., Condon, B., Liu, J., Zhang, N.: Protoplast transformation of filamentous fungi. Methods Mol. Biol. 638, 3–19 (2010)

There is NO Competing Interest.

SupplementaryVideo1.mp4
Supplementary Video 1
TableS2.xlsx
SupplementaryVideo2.mp4
Supplementary Video 2
TableS1.xlsx
TableS3.xlsx
SupplementaryFigures.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Cytoplasmic sequestering of a fungal stress-activated MAPK in response to a host plant phenolic acid

Status:

Version 1

Abstract

Figures

Introduction

Results

Hog1 localization in response to stress

Proteomic characterization of Hog1:GFP foci

Relationship between dephosphorylation and foci formation

Discussion

Methods

C. heterostrophus strains

Generation and Transformation of Fungal Protoplasts

Single-spore isolation

Genomic Verification

Stress induction

Immunohistochemistry and confocal fluorescence microscopy

SG production from induced samples and confocal fluorescence microscopy

SG Fraction Co-Immunoprecipitation

Proteolysis and mass spectrometry analysis

Immunoblots

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1