H3K4 methylation regulates development, DNA repair, and virulence in Mucorales

Abstract

Mucorales are basal fungi that opportunistically cause a potentially fatal infection known as mucormycosis (black fungus disease), which poses a significant threat to human health due to its high mortality rate and its recent association with SARS-CoV-2 infections. On the other hand, histone methylation is a regulatory mechanism with pleiotropic effects, including the virulence of several pathogenic fungi. However, the role of epigenetic changes at the histone level never has been studied in Mucorales. Here, we dissected the functional role of Set1, a histone methyltransferase that catalyzes the methylation of H3K4, which is associated with the activation of gene transcription and virulence. A comparative analysis of the Mucor lusitanicus genome (previously known as Mucor circinelloides f. lusitanicus) identified only one homolog of Set1 from Candida albicans and Saccharomyces cerevisiae that contains the typical SET domain. Knockout strains in the gene set1 lacked H3K4 monomethylation, dimethylation, and trimethylation enzymatic activities. These strains also showed a significant reduction in vegetative growth and sporulation. Additionally, set1 null strains were more sensitive to SDS, EMS, and UV light, indicating severe impairment in the repair process of the cell wall and DNA lesions and a correlation between Set1 and these processes. During pathogen-host interactions, strains lacking the set1 gene exhibited shortened polar growth within the phagosome and attenuated virulence both in vitro and in vivo. Our findings suggest that the histone methyltransferase Set1 coordinates several cell processes related to the pathogenesis of M. lusitanicus and may be an important target for future therapeutic strategies against mucormycosis.

Introduction

Mucorales are early diverging fungi that are considered saprophytic and globally distributed (Spatafora et al. 2016). However, certain species can evolve as pathogens by adapting to different conditions, significantly compromising human health. Mucormycosis is an invasive infection caused by this group of fungi, particularly species of the genera Rhizopus, Mucor, and Lichtheimia, which infect host tissues and can spread to other organs (Singh et al. 2009; Ibrahim 2011; Petrikkos et al. 2012; Baldin and Ibrahim 2017). Although it has traditionally been considered a disease exclusive to individuals with compromised immune systems resulting from cancer treatments, hematological disorders, and uncontrolled diabetes, it has also been detected in healthy individuals (Ibrahim and Kontoyiannis 2013; Prakash and Chakrabarti 2019; Gebremariam et al. 2019). Mucormycosis can be acquired through inhalation, ingestion, or traumatic inoculation of Mucorales spores and can manifest in various clinical forms such as pulmonary, cutaneous, gastrointestinal, and rhinocerebral (Richardson 2009; Walther et al. 2019; Skiada et al. 2020). Despite the development of new antifungal agents, the incidence of mucormycosis has significantly increased in recent years, with alarming mortality rates of up to 90% (Saegeman et al. 2010; Rammaert et al. 2012; Guinea et al. 2017; Prakash and Chakrabarti 2019; Skiada et al. 2020). In addition, most Mucorales species exhibit innate resistance and have evolved various multi-resistance pathways to commonly used therapeutic options, making mucormycosis a deadly disease (Spellberg et al. 2005; Caramalho et al. 2017; Vellanki et al. 2020).

M. lusitanicus is a suitable model organism for studying the molecular mechanisms underlying the physiology, development, pathogenesis, and virulence of Mucorales. The development of new molecular techniques and omics technologies has substantially contributed to identifying several pathways related to the virulence and antifungal resistance of Mucorales (Park et al. 2019; Vellanki et al. 2020; Lax et al. 2020). Furthermore, these mechanisms have facilitated a better understanding of the genetics and pathogenesis of Mucorales. In this context, the regulatory mechanisms based on histone post-transcriptional modifications have been scarcely studied in Mucorales (Navarro-Mendoza et al. 2023), raising the consideration of epigenetic modifications as a new avenue to unveil the particularities of the mucoralean biology.

Epigenetic mechanisms modify gene expression transiently by finely modulating the genomic structure without altering the DNA sequence. These dynamic mechanisms are crucial in regulating the cellular response to different extra- and intracellular signals (Allis and Jenuwein 2016; Qureshi and Mehler 2018). Major epigenetic modifications include post-translational modifications of histones, chromatin remodeling, RNA interference (RNAi), and DNA methylation (Allis and Jenuwein 2016; Qureshi and Mehler 2018; Xu et al. 2021). The amino-terminal ends of histones serve as targets for numerous covalent modifications, such as methylation, acetylation, phosphorylation, and ubiquitination which finely tune chromatin status and gene expression (Strahl and Allis 2000; Marmorstein and Trievel 2009; Bannister and Kouzarides 2011; Rothbart and Strahl 2014; Mushtaq et al. 2021). Histone methylation is a chemical modification that occurs at specific lysines and arginines of the different histones; among the most studied are those of lysines 4, 9, 27, and 36 on histone H3, as marks for binding reader proteins that modulate gene transcription (Cheng and Zhang 2007; Greer and Shi 2012; Black et al. 2012; Musselman et al. 2012; Zhang et al. 2021).

The biological role of histone methylation marks depends exclusively on the site and extent of methylation (Shilatifard 2008; Takahashi and Shilatifard 2010). Histone 3 lysine 4 (H3K4) methylation plays an important role in the transcriptional activation of genes related to development, pathogenicity factors, secondary metabolism, and DNA repair in several pathogenic fungi (Raman et al. 2006; Pham et al. 2015; Liu et al. 2015; Gu et al. 2017; Janevska et al. 2018; Zhou et al. 2021). H3K4 methylation is executed by the Set1 enzyme, a methyltransferase with a conserved SET domain initially identified in yeast (Tschiersch et al. 1994; Miller et al. 2001; Xhemalce et al. 2011). In yeast, Set1 acts as the catalytic component of the Complex Protein Associated with Set1 (COMPASS) for the different extents of H3K4 methylation (Krogan et al. 2002; Shilatifard 2006; Takahashi and Shilatifard 2010).

Mutations in Set1 in Saccharomyces cerevisiae suppress H3K4 methylation, repressing transcriptional activity, which results in growth abnormalities and a deficiency in the DNA repair (Nislow et al. 1997; Bryk et al. 2002; Santos-Rosa et al. 2002; Boa et al. 2003; Freitag 2017; Mushtaq et al. 2021). Similarly, Set1 disruption in plant pathogenic fungi, including several Fusarium species, Magnaporthe oryzae, and Aspergillus flavus, has been shown to cause loss of the H3K4 mark. This loss affects gene transcription, secondary metabolism, response to several stress types, hyphal growth, conidiation, appressorium formation, and virulence (Pham et al. 2015; Liu et al. 2015, 2020; Gu et al. 2017; Freitag 2017).

This H3K4 epigenetic mark also plays a crucial role in the pathogenesis of the human fungal pathogen Candida albicans. Deletion of set1 in this fungus results in total loss of H3K4 methylation, triggering hyperfilamentous growth in particular in vitro conditions, alterations of the cell surface, and reduced adherence to epithelial cells. Furthermore, murine model infectious assays showed that set1 mutants could not develop an infection, as high animal survival rates were observed, suggesting that set1 is critical for Candida pathogenesis through H3K4 methylation (Raman et al. 2006).

The lack of information about the role of histone modifications in the biology of Mucorales and in their ability to cause infection prompted us to analyze the function of set1 gene, which codes for a Set1-type methyltransferase in M. lusitanicus. The deletion of set1 confirmed the H3K4 methylation is strictly dependent on Set1 and revealed its crucial role in regulating development, DNA repair, and stress responses in M. lusitanicus. Furthermore, set1 is required for full virulence and proper polar growth after phagocytosis, as well as induced cell death of macrophages. These findings provide the first evidence of epigenetic modifications, specifically histone methylation, regulating the physiology and pathogenesis of M. lusitanicus.

Materials and methods

Fungal strains and growth conditions

M. lusitanicus strain MU402 (pyrGˉ, leuAˉ) (Nicolás et al. 2007) was employed as a recipient strain during the process of genetic transformation to generate the set1 mutants (Additional file 1: Table S1). The strain MU636 (leuAˉ) (Navarro-Mendoza et al. 2019), derived from MU402, was used as a wild-type strain for the different experiments throughout this research.

All the M. lusitanicus strains were grown at 26 °C on yeast peptone glucose agar plates (YPG; 3 g/L yeast extract, 10 g/L peptone, 20 g/L glucose, 15 g/l agar), pH 4.5, under illumination conditions for sporulation and colony growth measurement. The colonies grown after protoplast transformation and the spores recovered from macrophages lysates were plated on Minimal Medium with Casamino acids (MMC; 10 g/L casaminoacids, 0.5 g/L yeast nitrogen base without amino acids and ammonium sulfate, 20 g/L glucose, 15 g/L agar), adjusted to pH 3.2, and supplemented with niacin (1 mg/ml) and thiamine (1 mg/ml) to isolate homocaryotic strains and evaluate growth fitness, respectively. M. lusitanicus cultures grown on Yeast Nitrogen Base (YNB; 1.5 g/L ammonium sulfate, 1.5 g/L glutamic acid, 0.5 g/L yeast nitrogen base without amino acids and ammonium sulfate, 10 g/L glucose, and 15 g/L agar) with pH 3.0 at 26 °C, supplemented with niacin (1 mg/ml), thiamine (1 mg/ml), and leucine (20 mg/L) were performed to examine the effect of ethyl methanesulfonate (EMS), sodium dodecyl sulfate (SDS), ultraviolet light (UV), and hydrogen peroxide (H₂O₂). Cell cultures for the experiments on macrophage cell death and polarity index were performed in Leivobitz L-15 medium (Biowest, Minneapolis, MN, USA) at 37 °C, with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Gibco). Propidium iodide ReadyProbes reagent (Invitrogen) was used to stain macrophages and estimate cell death (2 drops per 10⁶ cells/ml).

Phylogenetic analysis and ortholog search

Proteomes of 23 representative species were retrieved from the Joint Genome Institute (JGI) Mycocosm genome portal (Grigoriev et al. 2014) and Uniprot (Bateman et al. 2021). Sequences of S. cerevisiae Set1, Cps15 (Sgh1), Cps25 (Sdc1), Cps30 (Swd3), Cps35 (Swd2), Cps40 (Spp1), Cps50 (Swd1) and Cps60 (Bre2) proteins were queried against the selected proteomes using iterative HMMER jackhmmer searches (E-value $\le$ 1 × 10–3) (v3.3.2) (http://hmmer.org). A reciprocal BLASTp search (v2.10.1) (Camacho et al. 2009) was conducted, and sequences that failed to produce a hit were discarded. An additional search using the Pfam-A database (Mistry et al. 2021) using HMMER hmmscan (v3.3.2) (http://hmmer.org) served to remove hits that lacked the SET domain (Set1 orthologs), WD/WD40 domains (Swd1, Swd2, and Swd3 orthologs), PHD-finger (Spp1 orthologs), Dpy-30 (Sdc1 orthologs), COMPASS Sgh1 (Sgh1 orthologs) and SPRY (Bre2 orthologs). A final list and a matrix including information about the presence or absence of the putative orthologs were generated (Additional file 2: Dataset S1). A Species tree was generated after analyzing the 23 proteomes with OrthoFinder (Inflation factor, -I 1.5) (Emms and Kelly 2019). We identified 37,040 protein orthogroups, of which 1200 had all species present. Sequences contained in single-copy orthogroups were aligned using MAFFT (Katoh and Standley 2013), and the phylogenetic species tree was obtained using RAxML (Stamatakis 2014) (PROTGAMMAWAGF substitution model) with 100 bootstrap replicates.

Generation of M. lusitanicus set1 knockout strains

Construction of recombinant fragment

To construct the recombinant fragment for set1 gene disruption, 1 kb sequences of upstream and downstream regions from set1 open reading frame (ORF) and 2 kb sequence of pyrG selection marker were amplified by PCR using appropriate primers (Additional file 1: Table S2). Later, resulting PCR products were subjected to overlapping PCR employing specific oligonucleotides (Additional file 1: Table S2) to generate the construct, which consists of the pyrG selectable marker surrounded by 5´and 3’ ends of the set1 gene. The PCR amplifications were performed with Herculase II fusión DNA Polymerase following supplier recommendation (Agilent, Santa Clara, CA, USA).

Genetic transformation of M. lusitanicus

The protoplast preparation was conducted to genetically transform the recipient strain MU402 (pyrGˉ, leuAˉ) according to the previously established protocol (Gutiérrez et al. 2011). The genetic transformation of the construct into strain MU402 (pyrGˉ, leuAˉ) was performed by the protoplasts electroporation to replace the set1 locus by double homologous recombination, as has been previously reported (Gutiérrez et al. 2011). Colonies developed from transformed protoplasts were transferred to a minimal medium MMC, pH 3.2, supplemented with niacin (1 mg/ml) and (1 mg/ml) by several growth cycles to select homokaryons. The integration of the construct into the target locus and homokaryosis was checked by amplifying the entire recombinant fragment by PCR using specific oligonucleotides (Additional file 1: Table S2).

Protein extraction and detection

To prepare cell extracts from wild-type and set1Δ strains, 100 mg of mycelium in 300 µl lysis buffer (8 M Urea, 5% w/v SDS, 40 mM Tris-ClH pH6.8, 0.1 mM EDTA, 0.4 mg/ml bromophenol blue, 1% β-mercaptoethanol, 5 mM PMSF, and 2% Protease Inhibitor Cocktail from Merck) were lysed with 0.5 mm Zirconia/Silica Beads in a FastPrep-24™ homogenizer, centrifuged at 20,000×g for 10 min at 4 °C, and the supernatant was collected. 30 µl of histone extracts were resolved on 10% SDS-PAGE and transferred to a nitrocellulose membrane using the Semi-Dry Electroblotting Unit System (Sigma-Aldrich). Histone H3 and modified residues were detected with antibodies against H3 (Abcam, ab176842), H3K4me1 (EpigenTek, C10005-1-3K4M), H3K4me2 (Abcam, ab176878), or H3K4me3 (EpigenTek, C10005-1-3K4T).

Phenotypic analysis of set1 knockout mutants

To test the sporulation and colony growth of set1 mutants (Additional file 1: Table S1), 500 fresh spores from each strain were inoculated in the center of solid YPG pH 4.5 and incubated at 26 °C for 72 h in 12 h dark–light cycles. The number of produced spores was estimated using a Neubauer chamber and normalized with the growth area corresponding to each M. lusitanicus strain. For radial growth, the colony diameter from fungal strains was registered each 24 h by 3 days.

Sensitivity testing

To determine the M. lusitanicus ability to respond to different chemical agents, preparations containing 10⁴, 10³, 10², and 10 freshly harvested spores were spotted on solid YNB media, pH 3.0, supplemented with SDS (0.002%), and EMS (0.05%). In addition, 200 fresh spores plated on YNB agar plates with pH 3.0 were irradiated with UV (10 mJ/cm²) at 254 nm.

To evaluate the fungal survival ability. All fungal cultures were incubated at 26 °C, and after 48 h, the survival percentage was calculated from the total number of cells obtained in control conditions.

In vitro host–pathogen interaction assays

For the host–pathogen interaction experiments, the fungal spores were challenged with mouse macrophages (J774A.1) according to previously established protocols (Pérez-Arques et al. 2019). In brief, the freshly harvested spores of the set1 mutants were added to cell cultures, in a proportion of 1.5 spores per macrophage, in Leibovitz L-15 media (Biowest, Minneapolis, MN, USA) amended with 10% FBS and 1% penicillin/streptomycin (Gibco) and incubated at 37 °C. After 30 min of co-incubation, non-phagocytized spores were removed, adding phosphate-buffered saline (PBS, 1X) to the cultures. After 5.5 h of interaction, pictures of germinated spores in the macrophages were taken to determine the polarity index using ImageJ software, as reported previously (Pérez-Arques et al. 2019). To evaluate the M. lusitanicus survival, the phagocytized spores were released by the addition of NP-40 (0.1%, Sigma Aldrich) to lyse cells. The recovered spores (500 spores) were plated on MMC plates with pH 3.2, placed at 26 °C, and after 48 h, the development of healthy colonies was visually inspected. Spores not confronted with macrophages and macrophage cultures served as control samples.

Determination of cell death

For the cell death experiments, spore-macrophage interactions were prepared in the same conditions mentioned above (Pérez-Arques et al. 2019). After 24 h of co-incubation at 37 °C, propidium iodide (PI, Invitrogen) was added to the co-cultures to evaluate macrophage death, according to supplier recommendations. The images were taken with Texas Red and bright-field filters 30 min after PI application using the 20/0.8-A objective from a Nikon Eclipse 80i fluorescent microscope equipped with a Nikon DS-Ri2 camera. The images were processed into binary images and overlapped using ImageJ software. The cell death ratio was estimated by manually counting the PI-stained cells and the total number of macrophages in the microscope images.

Infection assays in Galleria mellonella

Sixth instar larvae of G. mellonella (SAGIP, Italy), weighing 0.3–0.4 g, were selected for experimental use (Fallon et al. 2012). 10⁶ spores in a volume of 20 µl were injected per larva through one of the hind pro-legs as described previously (Kelly and Kavanagh 2011). Larvae were incubated at 30 °C. Untouched larvae and larvae injected with sterile IPS (insect physiological saline (Lackner et al. 2019)) served as controls. G. mellonella survival was monitored every 24 h up to 144 h. For each test group, 20 larvae were used, and experiments were repeated at least twice. Survival curves were statistically analyzed by log-rank (Mantel-Cox) test, utilizing GraphPad Prism 8.0.2 software. P values ≤ 0.05 were considered statistically significant.

RNA-sequencing analysis

Total RNA was extracted from MU636 and set1 mutant strains growing in plates using the RNeasy Plant Mini Kit (Qiagen) and treated with DNase (Sigma, On-Column DNaseI treatment set). RNA integrity was quality-checked using the Bioanalyzer 2100 (Agilent), and samples were submitted to Novogene for library preparation and sequencing. Raw paired-end (PE) reads from M. lusitanicus RNA-seq datasets were quality-checked using FastQC v0.11.9 before and after removing adapter sequences with Trim Galore! V0.6.7. Pairs containing a read with a Phred quality score (q) ≤ 33 and/or a total length < 20 nt were removed from the analysis, as well as adapter sequences with an overlap ≥ 4 nt. The PE processed reads were aligned to M. lusitanicus v3.0 genome (herein Mucci3, available at https://mycocosm.jgi.doe.gov/Mucci3/Mucci3.home.html) using hisat2 v2.2.1 (Kim et al. 2019) with a maximum intron length of 500 nt. Individual count matrices were created from the Binary Alignment Maps (BAM) using featureCounts v2.0.3 (Liao et al. 2014), excluding multimapping reads (Supplementary data ‘Raw counts’). PE reads were specified (p) and counted as fragments (countReadPairs) at the gene level (t) with protein ID as attribute (g). Differential gene expression analysis was performed by DESeq2 v1.34.0 package (Love et al. 2014). Genes with a False Discovery rate (FDR) ≤ 0.05 and a log₂ fold change (log₂FC) ≤ 1 or ≥ 1 were considered differentially expressed genes (DEGs). The volcano plot of DEGs was generated by VolcaNoseR web app (Goedhart and Luijsterburg 2020).

Protein prediction and structural analysis

Structure prediction of the MlSac1 protein (ID 1537005) was performed using AlphaFold2 (Jumper et al. 2021) implemented with the freely available ColaFold pipeline (Mirdita et al. 2022). ColabFold was run with the default configuration, except for selecting template mode with pdb70 database, and the multiple sequence alignments were generated with MMseqs2 (Mirdita et al. 2019). MlSac1 model 1 was superimposed to ScSac1 (PDB 3LWT) by Matchmaker using ChimeraX (Pettersen et al. 2021).

Results

The genome of M. lusitanicus encodes Set1, an enzyme with histone methyltransferase activity

To identify the genes involved in H3K4 methylation, we inspected the proteomes of 20 representative species of the major fungal phyla using the S. cerevisiae COMPASS components as the query (Fig. 1a, Additional file 2: Dataset S1). Overall, the components of the COMPASS are highly conserved across fungi, with the only exception of the Cps15/Shg1, a non-essential component of this complex found exclusively in S. cerevisiae (Roguev et al. 2001), which is also absent in animals (Fig. 1a). All the other components are found in most of the fungal phyla, apart from the Cps25/Sdc1 component that is absent in two distant phylogenetic fungal groups: Zoopagomycota and Basidiomycota. The presence of all the components of the COMPASS in Mucorales suggests the importance of H3K4 methylation in this group of fungi. To understand the function of H3K4 in Mucorales biology, we focused on the set1 gene of M. lusitanicus, which encodes the key methyltransferase component of the COMPASS. The predicted amino acid sequence of the only one set1 homolog (JGI ID 1544069) of M. lusitanicus contains a putative SET domain, characteristic of known Set1 methyltransferases (Fig. 1b). Furthermore, phylogenetic analysis, comparing Set1 enzymes of distinct fungal species with M. lusitanicus methyltransferase, revealed a close clustering between M. lusitanicus methyltransferase and Set1 proteins from Apophysomyces ossiformis, C. albicans and S. cerevisiae (Fig. 1c and Additional file 1: Table S3). Taken together, these findings suggest that Set1 likely has a conserved activity in methylating histone 3 lysines 4, which may regulate a variety of cellular processes in M. lusitanicus, similar to the function of characterized methyltransferases Set1 in other fungi (Raman et al. 2006; Zhou et al. 2021). To test this hypothesis, we deleted the set1 gene in the strain MU402 strain (pyrGˉ, leuAˉ) using the strategy of replacement by double homologous recombination (Fig. 2a) (Nicolás et al. 2018). Endpoint PCR analysis of two candidate mutant strains, which derived from independent genetic transformations, showed that pyrG replaced the set1 gene in homokaryosis. This was confirmed by the observation of only a fragment of 5.1 kb corresponding to the mutant locus (Fig. 2b). These mutants were designated as set1-3∆ and set1-4∆ for further analysis (Additional file 1: Table S1).

Fig. 1

Components of the COMPASS complex. a Conservation of the COMPASS complex in fungi and animals. b Schematic representation of M. lusitanicus Set1 containing the putative SET domain, which includes catalytic residue, SAM binding site, and active site with the corresponding amino acids. c Phylogenetic analysis of Set1 homologs (Additional file 1: Table S3) from M. lusitanicus (Ml) and other fungal species such as Rhizopus microsporus (Rm), Aspergillus rouxii (Ar), Apophysomyces ossiformis (Ao), S. cerevisiae (Sc), C. albicans (Ca), Aspergillus nidulans (An), Fusarium graminearum (Fg) and Neurospora crassa (Nc) conducted with the MEGA version X software using the neighbour-joining method. A. thaliana (At) was included as an outgroup. The phylogenetic tree shows the bootstrap values in each node from 1000 replicates. The conserved SET domain was determined by Pfam (http://pfam.xfam.org/)

Fig. 2

Functional analysis of the gene set1 of M. lusitanicus. a Scheme of mutant locus after set1 locus replacement on the genome of wild-type strain by the pyrG construct by homologous recombination. The recombination sites on the genome are marked with dashed lines. The binding sites of oligonucleotides used to check genic deletion and the size of amplified fragments are denoted on the diagram. b PCR products of set1 locus of wild-type strain MU636 (WT) and selected set1 mutants were amplified with the primers Set1-UF and Set1-DR (Additional file 1: Table S2) and separated by electrophoresis. The expected band (black arrows) for the mutant and wild-type locus corresponded to 5.1 kb and 3.1 kb, respectively. c H3K4 methylation in the wild-type strain (WT) and set1 mutants was determined by western blot analysis by using specific antibodies against H3K4 mono- (me1), di- (me2) and trimethylation (me3). As a control, S. cerevisiae protein extract was included. The protein samples were also incubated with anti-histone H3 antibody as a loading control

To determine if Set1 is the methyltransferase responsible for H3K4 methylation, we used the western blotting approach and fungal mono-, di-, and trimethylated forms of H3K4-specific antibodies to determine H3K4 methylation levels in each mutant. In these western blottings, H3K4 monomethylation (me1), dimethylation (me2) and trimethylation (me3) were not detected in the two set1-independent mutants (Fig. 2c). These results demonstrated the essential role of Set1 in H3K4 methylation in M. lusitanicus.

set1 is involved in growth, asexual sporulation, cell wall integrity and DNA-repair

To investigate the role of set1 on the physiology and development of M. lusitanicus, we performed a phenotypic analysis of set1 mutants. Deletion of the set1 gene has a negative impact on the growth of M. lusitanicus, as both set1 mutants displayed a smaller colony diameter compared with the wild-type strain MU636, measured at 24-h intervals up to 72 h (Fig. 3a and b). Moreover, the lack of set1 reduced the production of asexual spores in both mutants, generating less than half the spores/cm² compared to the WT (P = 0.005) (Fig. 3c). These mutants also showed increased sensitivity to cell wall stress induced by the presence of SDS. We dropped appropriate spore concentrations (10⁴, 10³, 10², and 10 spores) of each strain on YNB plates containing 0.002% SDS. The ability of the strains to form colonies in the presence of SDS was used to determine their sensitivity levels. Our results showed that both set1 mutants were highly sensitive to SDS compared to the wild-type strain, with complete inhibition of growth at low spore concentrations (10³, 10², and 10 spores) (Fig. 4a). This phenotype was further supported by the lower percentage of set1 mutant spores that form colonies in medium seeded in plates with SDS (Fig. 4b, Additional file 1: Fig. S1). These results suggest that the activity methyltransferase of Set1 is important for appropriate mycelial growth, sporulation process, and integrity of the cell wall of M. lusitanicus.

Fig. 3

The deletion of set1 alters colony growth of M. lusitanicus. a Fungal colonies of the wild-type strain MU636 (WT) and set1 deletion mutants (set1-3∆ and set1-4∆) grown on YPG medium plates pH 4.5 at 26 °C for 3 days under constant illumination. b Graph displays the radial growth of the WT and set1∆ strains measured every 24 h for 72 h. The line bars in each time point of the growth kinetic represent the standard error from the three biological experiments. c) Production of vegetative spores in set1 mutants and WT on solid YPG pH 4.5

Fig. 4

set1 knockout strains are sensitive to SDS and EMS. a, b Growth of wild-type strain MU636 (WT) and set1∆ strains from different concentrations of spores spotted on YNB medium, YNB medium amended with SDS (0.002%) or EMS (0.05%). c, d Survival rate of M. lusitanicus strains after treatment with SDS (0.002%) or EMS (0.05%), respectively. The cultures were placed at 26 °C by 48 h. Data were analyzed by two-way ANOVA and asterisks above the charts indicate significant differences (****P < 0.0001)

Since Set1 has been involved in DNA repair in S. cerevisiae (Faucher and Wellinger 2010), we evaluated the involvement of M. lusitanicus set1 in DNA repair by analyzing the sensitivity of set1∆ strains to the alkylating agent EMS and UV light, which induces dimerization of DNA bases (Kielbassa et al. 1997; Douki and Cadet 2001). Spores of the set1∆ strains either grown in the presence of 0.05% EMS or exposed to a UV pulse (10 mJ/cm2) developed fewer colonies and exhibited drastically reduced survival levels compared to the wild-type strain (Fig. 4b and d, Additional file 1: Fig. S1), suggesting that H3K4 methylation plays an important role in DNA damage repair in M. lusitanicus.

The lack of set1 reduces virulence of M. lusitanicus

Set1 enzymes have been involved in the virulence of a few fungal pathogens belonging to the Ascomycota phylum (Raman et al. 2006; Pham et al. 2015; Liu et al. 2015, 2020; Gu et al. 2017; Freitag 2017), but its role has not been analyzed in other fungal groups. Therefore, we assessed the role of set1 in the pathogenesis of M. lusitanicus using the Galleria mellonella infection model (Maurer et al. 2019). Healthy larvae of G. mellonella were injected with spores of the wild-type strain or the set1 mutant strains and their survival was monitored daily for 6 days. Intriguingly, both set1 deletion mutants exhibited reduced virulence compared to the wild-type strain (P = 0.0348 and P = 0.0160, respectively) (Fig. 5), suggesting that the Set1 enzyme plays a critical role in the pathogenicity of M. lusitanicus.

Fig. 5

set1 is essential for M. lusitanicus virulence. Virulence of M. lusitanicus strains in G. mellonella model host inoculated with 10⁶ spores of the WT strain (MU636) and set1 mutants through the last pro-leg into the hemocoel. Percentage survival values were plotted for 6 days. Untouched larvae and larvae injected with IPS were used as control. Survival curves were statistically analyzed by log-rank (Mantel-Cox) test, utilizing GraphPad Prism 8.0.2 software. The difference in survival between set1 mutants and wild-type strain was statistically significant (P values ≤ 0.05)

To investigate the contribution of Set1 protein to the ability of M. lusitanicus to survive phagocytosis during interaction with macrophages, we challenged spores of set1-3∆ and set1-4∆ mutants with mouse macrophages for 5.5 h of co-culture and subsequently plated them on MMC agar plates. Similar to the WT strain, both set1 mutants retained the ability to counteract the cytotoxic environment during phagocytosis, as evidenced by the lack of damage in the grown colonies (Additional file 1: Fig. S2a). This finding suggests that the regulatory mechanism governing M. lusitanicus survival during macrophage phagocytosis does not depend on the Set1 enzyme. However, we observed that phagocytized spores from set1 deletion strains developed a shorter germ tube after 5.5 h of interaction with macrophages (Additional file 1: Fig. S2b). In contrast to the wild-type strain, the polarity rate, calculated from the length of the emerged hyphae and spore width, was significantly lower in the set1-3∆ and set1-4∆ mutants in comparison to the wild-type strain (Fig. 6a), indicating that Set1 is required for appropriate germ tube development during phagocytosis.

Fig. 6

set1 disruption results in a reduction of polarity index and ability to induce macrophage death. a Polarity index of the wild-type MU636 (WT) and set1 deletion strains determined from phagocytosed spores by mouse macrophages (J774A.1) after 5.5 h of incubation. b Images of live-cell microscopy of mouse macrophages (Φ) co-cultured for 24 h with spores of WT and set1∆ strains. The dead macrophages were stained with PI. Macrophage non-interacting with spores served as a control. c Macrophages cell death at 24 h of interaction with M. lusitanicus spores estimated by determining percentage of PI-stained macrophages in the fluorescent images. The charts display means ± SD based on three biological repetitions. The statistical differences were obtained by one-way ANOVA and indicated by asterisks (****P < 0.0001)

The decreased polar growth during interaction with macrophages could potentially impact M. lusitanicus pathogenesis. To test the role of Set1 in macrophage cell death, we performed experiments on the interaction between set1 mutant spores and mouse macrophages. Cell death was determined 24 h later by adding propidium iodide (PI), which stains dead cells or those with a damaged cell membrane. We observed fewer dead macrophage cells stained with PI in cultures with set1-3∆ and set1-4∆ mutants compared to the wild-type strain (Fig. 6b). Furthermore, the percentage of macrophage cell death, calculated from PI-positive cells and the total number of cells, was remarkably lower in the strains lacking the set1 gene compared to the WT strain (Fig. 6c). These results indicated that M. lusitanicus mediated killing of macrophages is regulated by the Set1 protein, likely through the methylation of H3K4.

Genes regulated by set1

We performed a transcriptomic analysis to explore the effect of the lack of H3K4 methylation on the gene expression profile of M. lusitanicus. Messenger RNA was isolated and deep sequenced from the wild-type control and the set1 mutant strain after 24 h of growth on solid rich (YPG, yeast extract peptone glucose) medium. Transcriptomic analysis revealed a limited effect of H3K4 methylation on gene expression profiles in these growth conditions (Fig. 7a, Additional file 3: Dataset S2). A total of 403 genes were differentially expressed in the set1 mutant strain: 343 up-regulated genes and 60 down-regulated genes. This was a surprising result as H3K4 methylation is considered an epigenetic mark related to actively expressed genes in fungi (Raman et al. 2006; Pham et al. 2015; Liu et al. 2015; Gu et al. 2017; Janevska et al. 2018; Zhou et al. 2021) and suggests that in M. lusitanicus, this regulation may affect other transduction pathways regulators that control the expression of the up-regulated genes. Unfortunately, more than half of the genes (243/403) lacked EuKaryotic Orthologous Group (KOG) annotation, which hampered our understanding of their contribution to the phenotypes observed. The annotated genes were mainly related to metabolism (94/160), such as Cu²⁺/Zn2⁺ superoxide dismutase SOD1 (ID 1458748), Cytochrome b5 (ID 1535516), Chitinase (ID 1597095), which are involved in the dismutation of reactive oxygen species (ROS), cellular detoxification processes, and chitin decomposition, respectively. Highlighting the presence of a putative phosphoinositide phosphatase (ID 1537005) among top down-regulated DEGs, which shares a 41% of identity with S. cerevisiae Sac1 (ScSac1). The blast analysis of this putative M. lusitanicus Sac1 (MlSac1) revealed that both were the best reciprocal hits in S. cerevisiae proteome, and the superimposition of MlSac1 predicted structure with ScSac1 suggested that they were true homologs (Fig. 7b). Sac1 protein plays an essential role in the vesicle trafficking of S. cerevisiae being involved in protein secretion and cell wall maintenance (Schorr et al. 2001). This phosphoinositide phosphatase modulates phosphatidylinositol 4-phosphate concentration gradients driving the membrane homeostasis (Del Bel and Brill 2018). The down-regulation of Mlsac1 due to set1 mutation could unbalance the secretory pathways of M. lusitanicus, affecting the exportation of virulence factors and the proper composition of the cell membrane and cell wall.

Fig. 7

Genes regulated by set1. a Volcano plot of DEGs identified between a set1 mutant and a wild-type strain. The red dots denote up-regulated genes, the blue dots denote down-regulated genes, and the gray dots denote the genes with expression changes or significance below threshold (See material and methods). b Superimposition of ScSac1 (PDB 3LWT) in red and the predicted structure for MlSac1 in blue. The rmsd and the number of aligned residues are indicated

Discussion

Identifying novel regulators involved in the pathogenesis of Mucorales is crucial and could help to develop specific antifungal treatments for early mucormycosis therapy. Set1 is a histone methyltransferase of the COMPASS protein complex that catalyzes the addition of methyl groups to lysine 4 on histone 3 via its SET domain (Roguev et al. 2001; Takahashi and Shilatifard 2010). Generally, H3K4 methylation has been associated with the transcriptional activation of genes involved in various biological events in eukaryotes (Nakayama et al. 2001; Boa et al. 2003; Peters et al. 2003; Zhang et al. 2009; Pham et al. 2015). In several fungal species, the Set1 protein is required for cellular responses to stressful environments and pathogenesis (Raman et al. 2006; Pham et al. 2015; Liu et al. 2015; Zhou et al. 2021). However, the role of the methyltransferase Set1 in Mucorales has not been investigated. In this study, we characterized the function of the unique set1 gene in M. lusitanicus, providing new insights into the regulatory mechanisms of morphogenesis and pathogenesis in this fungus at the epigenetic level. The set1 gene of M. lusitanicus encodes a putative histone methyltransferase that shares high similarity to well-characterized Set1 of S. cerevisiae and C. albicans. Set1 deletion in C. albicans causes growth defects, reduces adherence to epithelial cells, and suppresses histone 3 lysine 4 (H3K4) methylation, leading to attenuated virulence (Raman et al. 2006). Similarly, in S. cerevisiae, disruption of SET1 results in a loss of H3K4 methylation and a reduction in gene transcription (Boa et al. 2003). Several phytopathogenic fungi have reported similar findings (Pham et al. 2015; Liu et al. 2015; Gu et al. 2017; Zhou et al. 2021). The amino acid sequence of M. lusitanicus Set1 harbors the SET domain characteristic of known histone methyltransferases Set1 (Fig. 1b), which is essential for H3K4 methylation and supports several biological events observed in various eukaryotic organisms. As an indication of transcriptional activation, H3K4 methylation is a conserved epigenetic modification found from yeast to humans. It is widely believed that H3K4 methylation is catalyzed by the methyltransferase Set1/COMPASS (Shilatifard 2012). In the phytopathogenic fungus Fusarium graminearum, FgSet1 is responsible for H3K4 mono-, di-, and trimethylation (Liu et al. 2015), whereas Set1 of Aspergillus flavus (AflSet1) only produce di- and trimethylation (Liu et al. 2020). To investigate the biological role of Set1 in histone methylation in the fungus M. lusitanicus, western blotting analysis was performed in this study. The results of the immunoblotting analysis showed that similar to FgSet1, Set1 primarily regulates the mono-, di-, and trimethylation of H3K4 (Fig. 2c). In contrast to AflSet1, however, Set1 is also responsible for H3K4 monomethylation in M. lusitanicus. These findings indicate that the methylation activity of Set1 is relatively conserved across different fungal species but shows functional differences related to the number of methylations.

In M. lusitanicus, as in other filamentous fungi and yeast, Set1 is associated with vegetative growth. The absence of Set1 resulted in a significant decrease in colony growth (Fig. 3a and b), indicating that Set1 is involved in this process. Moreover, our data suggest that the pathways regulating sporulation are also linked to Set1, as mutants and wild-type strains showed differences in spore production (Fig. 3c). In several phytopathogenic fungi species, Set1 is essential for spore formation (Janevska et al. 2018), similar to the case in M. lusitanicus, indicating functional conservation among different species within the fungal kingdom.

Little information exists on the interplay between Set1, H3K4 methylation, and stress signal response. Our genetic replacement experiments demonstrated that the set1 gene is required for M. lusitanicus to respond to cell wall-disrupting agents. The set1 deletion mutants were highly sensitive to SDS, indicating that the cell wall stress signaling pathway is compromised in the absence of Set1, which ultimately compromises cell survival (Fig. 4a and c). In Fusarium verticillioides, Set1 mediates the phosphorylation of the mitogen-activated protein kinase Mpk1 and Hog1 orthologs for preserving the cell wall integrity and responding to stress signals (Gu et al. 2017). Therefore, it is plausible that the deregulation of this process in M. lusitanicus results from the lack of the set1 gene.

Moreover, the repair of distinct types of DNA lesions was also impaired in M. lusitanicus. EMS reagent generates alkylated bases, which are removed from DNA by base excision (Gocke et al. 2009). The set1 mutant strains were more susceptible to EMS (Fig. 4b and d), supporting the notion that the repair mechanism of damaged nucleotides is deficient in these mutants due to the deletion of the set1 gene. Interestingly, these mutants also lose the ability to survive the genotoxicity of UV irradiation, as observed in our experiments (Additional File 1: Fig. S1). Applying UV light induces the formation of thymine dimers in DNA, and inaccurate repair by base excision leads to cell death (Hoeijmakers 2001). These findings suggest that the Set1 protein plays an important role in different repair pathways, including cell wall and DNA lesion repair, and may regulate the expression of genes related to cell damage repair through its methyltransferase activity.

In contrast to the sensitivity of the Set1-deficient strains to chemical compounds that induce cell damage, these strains do not exhibit impairment in their response to survive phagocytosis during interaction with mouse macrophage cells. Within the phagosome, fungal spores are subjected to a stressful environment, including nutrient deprivation, antimicrobial peptides, acidification, and oxidative stress, which are known to affect their survival (Pérez-Arques et al. 2019; Nicolás et al. 2020). Interestingly, the colonies of the set1 mutants presented healthy development after being challenged with macrophages (Additional File 1: Fig. S2), indicating that distinct defense strategies operate correctly to resist the macrophage attack. Previous studies have demonstrated the ability of some Mucorales species to germinate inside macrophages as a survival process and induction of cell death (Lee et al. 2015; López-Muñoz et al. 2018). In this sense, the set1 knockout strains displayed late germination supported by lower values of polarity index during the macrophage phagocytosis (Fig. 6a), suggesting an important role of the Set1 enzyme in the timely development of germ tube inside the phagosome. The set1 removal probably leads to a breakdown in the regulatory pathways of germination of M. lusitanicus in response to macrophages in vitro. M. lusitanicus strains impaired in germination or with shortened polar growth are known to be less virulent (Pérez-Arques et al. 2019). After analyzing the virulence in vitro of set1 mutants, we found that they present attenuated virulence due to the lower number of PI-stained macrophages observed in the fluorescent images (Fig. 6b), indicating a minor percentage of cell death. Thus, although the mutants can resist phagocytosis, they cannot kill macrophages at the same level as the wild-type strain. The attenuated virulence in vitro of these mutants was supported by infection experiments in vivo in G. mellonella, where lower mortality rates were observed in strains lacking the set1 gene (Fig. 5). These findings highlight the contribution of Set1 to the pathogenesis and virulence of M. lusitanicus. Previous studies have demonstrated that this fungus activates cell death by activating specific apoptosis-related genes in macrophages (Pérez-Arques et al. 2019). Therefore, the deregulation of this process could occur in the set1 mutants and partially explain the reduced virulence. However, additional studies are required to demonstrate this hypothesis. Similar to our results, Set1 plays an essential role in virulence in several species of phytopathogenic fungi, and strains lacking the set1 gene lose the ability to infect maize plants and defects in the toxin biosynthesis (Gu et al. 2017).

Transcriptomic analysis of set1 mutants revealed the genes repressed and activated by this gene (Fig. 7). The most striking result of this analysis was the larger number of genes that were repressed compared to those that were activated, which contrasts with the general activation of gene expression seen in other fungi. There are two potential explanations for these findings. Firstly, Set1 primarily activates gene expression, as seen in other organisms, but many activated genes act as repressors of different gene pathways. Alternatively, the less likely hypothesis is that the methylation of H3K4 by Set1 has a dual function, as observed with H3K9 methylation. Previous studies showed that the trimethylation of H3K9 is associated with transcriptional repressed loci, whereas dimethylation is associated with transcriptional activation (Liu et al. 2015). Another interesting result from the transcriptomic analysis was the high number of genes regulated by Set1 that are involved in metabolism. This finding could be linked to the observed decrease in virulence both in vitro and in vivo. A metabolic outfit affected by the lack of Set1 could result in a delay in germ tube growth. This delay, in turn, would represent a disadvantage for the fungus during interaction with phagocytic cells, which could explain the overall decrease in virulence observed in set1 mutant strains.

Conclusions

The knowledge on the role of epigenetic modification in regulating gene expression in early diverging fungi is scarce, despite they represent an important fraction of the fungal kingdom. The order Mucorales, which causes the lethal infection known as mucormycosis, is not exception. This work represents the first investigation into the role of the methylation of lysine 4 on histone 3 (H3K4) in a mucoralean fungus. This was accomplished by the generation of deletion mutants in the set1 gene, which encodes the specific H3K4 methyltransferase, as confirmed by analysis of histone H3 methylation in the mutants. Furthermore, phenotypic analyses of these mutants suggest that H3K4 methylation regulates physiology, development, cell wall integrity, and DNA repair. Additionally, our findings indicate that it also contributes to the virulence of M. lusitanicus, as strains lacking the set1 gene exhibited shortened polar growth within the phagosome and attenuated virulence both in vitro and in vivo. These results provide the first indication of the involvement of post-transcriptional modifications of histones in the virulence of early-diverging fungi.