Functional analysis of target of rapamycin in regulating mycelial growth and development in Penicillium italicum

Abstract

Citrus fruits are highly valued for their nutritional benefits and global economic significance. They are susceptible to many fungal pathogens during postharvest storage. Citrus blue mold, caused by the fungus Penicillium italicum, is a major citrus disease, leading to significant economic losses every year. P. italicum spreads rapidly after infection, and the molecular mechanisms regulating its growth and development are largely unknown. Target of rapamycin (TOR) is a central regulator of cell growth, metabolism, and stress responses in eukaryotes. In this study, we investigated the role of TOR in P. italicum using rapamycin, a specific TOR inhibitor. We found that P. italicum possesses a single TOR gene. Rapamycin treatment significantly inhibited mycelial growth of P. italicum by up to 50%, leading to small, twisted mycelia cells and disordered growth orientation. The distribution of mycelial layers was also affected by rapamycin treatment. In addition, rapamycin strongly suppressed conidia formation, while it did not influence conidial germination. These observations suggest that PiTOR regulates mycelial growth and differentiation, but not conidial germination in P. italicum. Transcriptomic analysis showed that rapamycin treatment led to 322 upregulated genes and 149 downregulated genes. Gene ontology (GO) annotation analysis revealed that many dysregulated genes were involved in membrane integrity, consistent with the observed morphological abnormalities. GO enrichment analysis further showed that downregulated genes were associated with cell wall components, whereas upregulated DEGs were linked to amino acid metabolic and catabolic processes, suggesting a crucial role of PiTOR in regulating cell growth and metabolism. This study enhances our understanding of TOR signaling in phytopathogenic fungi and facilitates the development of potential strategies for controlling citrus blue mold.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10123-026-00789-1.

Introduction

Citrus is an important fruit crop in the world, with an annual fruit production of more than 150 million tons worldwide (Andrade et al. 2023). Citrus fruits are popular for their taste and richness in nutrients and bioactive compounds (Shorbagi et al. 2022; Andrade et al. 2023). China has many citrus cultivars and has the largest cultivation area in the world (UNdata, https://data.un.org). For example, Citrus reticulata (mandarin) is hypothesized to be domesticated in South China, where it has been cultivated for more than 4,000 years (Chinese Citrus Society 2008; Wang et al. 2018). Citrus fruits are susceptible to fungal pathogens during transportation and postharvest storage. In fact, more than 20 types of postharvest fungal diseases have been reported to infect citrus, among which blue mold, green mold and sour rot are the most prevalent diseases (François et al. 2022). Citrus blue mold is caused by Penicillium italicum. P. italicum is a necrotrophic fungus that is responsible for 10–30% of postharvest citrus loss worldwide, with losses even reaching 30–50% in China (Chen et al. 2019). Although some studies have focused on the infection mechanisms of P. italicum, the molecular regulatory network for its growth and development remains largely unknown (Zhang et al. 2020; Li et al. 2022a, b).

Target of rapamycin (TOR) is a master regulator of cell growth, development, and metabolism in response to nutrient and stress cues in eukaryotes (Schmelzle and Hall 2000; Wullschleger et al. 2006; Liu and Xiong 2022). It is a conserved Ser/Thr protein kinase that regulates many essential developmental processes, from embryogenesis to senescence (Ren et al. 2013a; Brunkard 2020; Bjedov and Rallis 2020; Liu and Xiong 2022). TOR genes were first identified in Saccharomyces cerevisiae (budding yeast) through genetic screening of mutants resistant to rapamycin (Heitman et al. 1991; Kunz et al. 1993). Budding yeast, Schizosaccharomyces pombe (fission yeast), and many other fungi have two TOR paralogs (Heitman et al. 1991; Shertz et al. 2010; Ikai et al. 2011), whereas many other species, including mammals and many plant species, have only one TOR gene (Schmelzle and Hall 2000; Ren et al. 2013a; Liu and Xiong 2022). The TOR kinase is a large protein (~ 280 kDa) containing five conserved domains: a HEAT (Huntington, EF3A, ATM, TOR) repeat domain, a FAT (FRAP, ATM, and TRRAP) domain, an FRB (FKBP12 rapamycin-binding) domain, a kinase domain, and a FATC (FRAP, ATM and TRRAP C-terminal) domain (Schmelzle and Hall 2000; Wullschleger et al. 2006). TOR binds with other proteins to form two complexes—TOR complex 1 (TORC1) and TOR complex 2 (TORC2) to mediate nutrient and stress signals. The TOR Complex 1 (TORC1) mainly contains TOR, RAPTOR (KOG1 in budding yeast) and lethal with Sect. 13 protein 8 (LST8). TORC1 promotes cell growth by facilitating G1-to-S phase progression in eukaryotes (Barbet et al. 1996; Schmelzle and Hall 2000; Xiong et al. 2013), and controls ribosome biogenesis, transcription, translation, autophagy, stress responses, and metabolism (Schmelzle and Hall 2000; Wullschleger et al. 2006; Liu and Xiong 2022). In contrast, the TORC2 primarily contains TOR, RICTOR (AVO3 in budding yeast) and LST8, and regulates plasma membrane homeostasis and actin organization (Schmelzle and Hall 2000; Wullschleger et al. 2006; Thorner 2022). In budding yeast, TORC1 contains either ScTOR1 or ScTOR2, and TORC2 contains ScTOR2 rather than ScTOR1 (Heitman et al. 1991; Schmelzle and Hall 2000; Wullschleger et al. 2006). The fission yeast SpTOR2 is a specific subunit of TORC1, while SpTOR1 is a specific subunit of TORC2 (Weisman et al. 2007; Ikai et al. 2011). In mammals, the single TOR gene can form both TOR complexes (Schmelzle and Hall 2000; Wullschleger et al. 2006). By contrast, while the function of TORC1 has been widely reported in plants (Ren et al. 2013a; Xiong et al. 2013, 2016; Deng et al. 2020; Liu and Xiong 2022), orthologs of RICTOR have not been identified in plants.

Rapamycin is a specific TOR inhibitor. It is a macrolide compound purified from the bacterium Streptomyces hygroscopicus, which was isolated from Easter Island (Vézina and Kudelski 1975). Rapamycin is currently used as an immunosuppressive and antiproliferative agent (Saunders et al. 2001; Ferrer et al. 2011), and is considered as a potential anticancer and antiaging drug (Hua et al. 2019; Lee et al. 2024). It binds to FK506-binding protein of 12 kDa (FKBP12), which then interacts with the FRB domain of TOR to prevent the interaction between TOR and RAPTOR, resulting in the inhibition of TORC1 function (Heitman et al. 1991; Chiu et al. 1994; Sabatini et al. 1994; Aylett et al. 2016). In contrast, TORC2 is insensitive to rapamycin (Zheng et al. 1995; Loewith et al. 2002). The TORC2- specific subunit RICTOR occupies the FRB domain of TOR, blocking the binding of the rapamycin–FKBP12 complex to this domain (Sarbassov et al. 2004; Gaubitz et al. 2015). Since loss of TOR function leads to embryonic lethality, rapamycin has been widely used to decipher TOR function in many species.

In yeast, plants, and mammalian cells, TOR has been shown to regulate protein translation by phosphorylating S6K (the AGC kinase Sch9 in yeast) and other proteins, such as the Tap42–PP2A complex in yeast, 4E-BPs in mammals, and TAP46 in plants (Schmelzle and Hall 2000; Wullschleger et al. 2006; Liu and Xiong 2022). It also controls multiple steps in ribosome biogenesis, including ribosomal RNA transcription, translation of ribosomal proteins, and ribosome assembly in eukaryotes (Schmelzle and Hall 2000; Wullschleger et al. 2006; Eltschinger and Loewith 2016; Liu and Xiong 2022). The ATG1/ATG13 complex plays an essential role in initiating autophagy, which degrades damaged proteins and organelles for nutrient recycling. Nutrient-activated TORC1 can directly phosphorylate ATG1 (and ATG13) to inhibit autophagy initiation. Under nutrient deprivation conditions, TOR function is suppressed by AMPK, leading to the activation of the ATG1/ATG13 complex to trigger autophagy for nutrient recycling (Wullschleger et al. 2006; Suttangkakul et al. 2011; Eltschinger and Loewith 2016; Liu and Xiong 2022). TOR generally promotes anabolic metabolism but inhibits catabolic metabolism to facilitate cell growth and development. Nitrogen sources, such as nitrate and amino acids, serve as upstream signals to activate the TOR signaling pathway. At the same time, TOR also controls the assimilation and metabolism of these molecules (Schmelzle and Hall 2000; Wullschleger et al. 2006; Liu and Xiong 2022). In addition, TOR modulates cell wall formation and integrity. In budding yeast, both rapamycin and nutrient starvation can activate the cell wall integrity (CWI) pathway (Torres et al. 2002). In Arabidopsis, strong evidence indicates that the glucose–TOR pathway controls cell wall biosynthesis (Ren et al. 2013a; Xiong et al. 2013). Arabidopsis TOR kinase was reported to affect cell wall structure, possibly through ROL5 (REPRESSOR of LRX1) (Leiber et al. 2010).

Although rapamycin was first identified as an antifungal agent (Vézina and Kudelski 1975), TOR function in phytopathogenic fungi has not been widely investigated. Through the application of rapamycin, TOR knockdown lines, and knockout mutants of downstream targets, recent studies have shed light on TOR function in some phytopathogenic fungi. For example, rapamycin can strongly inhibit mycelial growth and sporulation in Botrytis cinerea, Verticillium dahliae, Fusarium oxysporum, Fusarium graminearum, and Aspergillus flavus, indicating the essential role of TOR kinases in these species (Yu et al. 2014; Li et al. 2019, 2021, 2023; Xiong et al. 2019; Wang et al. 2023). In addition, expressions of genes related to cell wall biosynthesis and integrity were deregulated when the TOR signaling pathway was suppressed in these species. In this study, we investigated the role of P. italicum TOR (PiTOR) using rapamycin. We found that P. italicum has a single TOR gene. Homologous genes encoding other essential components of TORC1 and TORC2 were also found in P. italicum, suggesting that TORC1 and TORC2 may exist in P. italicum. Rapamycin significantly inhibited its mycelial growth, differentiation, and sporulation. Conidial germination, however, was not affected by rapamycin. Transcriptomic analysis of RNA-Seq data showed that many genes were dysregulated after rapamycin treatment, many of which are related to the integral component of membrane, consistent with the observed morphological disorder after rapamycin treatment. These deregulated genes were enriched in multiple molecular processes, including metabolic and catabolic processes of cell wall components and amino acids. Our results suggest that PiTOR regulates mycelial growth and development by controlling membrane and cell wall development as well as multiple molecular processes. This study enhances our understanding of TOR signaling in phytopathogenic fungi and provides insights into potential strategies for controlling blue mold in citrus crops.

Results

The TOR and FKBP12 protein sequences are highly conserved in P. italicum

To identify the sequence structure of the PiTOR protein, the TOR protein sequences of S. cerevisiae and B. cinerea were used to search for homologous sequences in P. italicum using both NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and EnsemblFungi (http://fungi.ensembl.org/Multi/Tools/Blast) (Ballester et al. 2015). The BALST results showed that P. italicum has a single TOR gene (PITC_004820), encoding a protein of 2,384 amino acids (GenBank ID: KGO76948). The PiTOR amino acid sequence shares 61.76% identity with that of B. cinerea TOR (BcTOR). The PiTOR coding region is 7,269 base pairs long, containing three exons and two introns (Fig. 1a). We also identified the TOR genes in Penicillium digitatum and Geotrichum candidum, two fungal species that also cause postharvest diseases in citrus (Fig. 1b). PiTOR shares 99.12% and 51.40% identity with PdTOR and GcTOR, respectively. Phylogenetic analysis was conducted using TOR protein sequences of P. italicum, P. digitatum, G. candidum, and some other species with reported TOR functions (Wullschleger et al. 2006; Ren et al. 2013b; Xiong et al. 2016, 2019; Brunkard 2020; Deng et al. 2020; Liu and Xiong 2022). The analysis revealed that PiTOR was closely related to PdTOR and BcTOR (Fig. 1b), suggesting that PiTOR is evolutionarily conserved and it may have functions similar to those of BcTOR. In addition, other components of the TORC1 and TORC2 complexes were also found in the P. italicum genome (Table 1), implying that both TORC1 and TORC2 may exist in P. italicum.

Fig. 1.

Protein sequence analysis of PiTOR and PiFKBP12. (a) The gene structure of PiTOR. Numbers above the boxes indicate distances from the start codon. (b) Phylogenetic analysis of TOR proteins from P. italicum (PITC_004820), P. digitatum (PDIP_64280), G. candidum (BN980_GECA12s01055g) and other species. The scale bar indicates substitutions per amino acid. (c) Phylogenetic analysis of PiFKBP12 in P. italicum and other species. The scale bar indicates substitution per amino acid. (d) Multiple sequence alignment of FKBP12 protein sequences of P. italicum and other species. Numbers on the right side indicate the distance from the first amino acid. Asterisks indicate the hydrophobic amino acids that constitute drug-binding sites in the H. sapiens FKBP12. (e) Multiple sequence alignment of the FRB domain of TOR proteins in P. italicum and other species. Numbers on the right side indicate the distance from the first amino acid

Table 1.

The putative components of TOR complexes in P. italicum

	H. sapiens	S. cerevisiae	B. cinerea	P. italicum
TORC1	mTOR	TOR1/2	TOR	TOR (PITC_004820), KGO76948
	RAPTOR	KOG1	RAPTOR	RAPTOR (PITC_036020), KGO78234
	LST8	LST8	LST8	LST8 (PITC_038420), KGO66962
	PRAS40	-	-	-
	DEPTOR	-	-	-
	-	TOC89		-
TORC2	mTOR	TOR1/2	TOR	TOR (PITC_004820), KGO76948
	RICTOR	AVO3	RICTOR	RICTOR (PITC_004700), KGO76936
	LST8	LST8	LST8	LST8 (PITC_038420), KGO66962
	SIN1	AVO1	SIN1	SIN1 (PITC_016560), KGO73900
	DEPTOR	-	-	-
	PRR5	BIT61	-	-
	-	AVO2	-	-
FKBP12	FKBP12	FKBP12	FKBP12	FKBP12 (PITC_065960), KGO70915

“-” indicates that the homologous protein is not identified. The Gene ID and the GenBank ID are shown for homologous proteins in P. italicum

As rapamycin binds to FKBP12 to inhibit TOR function, we identified the P. italicum FKBP12 (PiFKBP12) protein sequence using a similar BLAST method. The PiFKBP12 coding sequence has 781 base pairs long, containing 6 exons and 5 introns. It encodes a protein of 122 amino acids. Phylogenetic analysis indicated that PiFKBP12 is closely related to the B. cinerea FKBP12 (Fig. 1c). Multiple sequence alignment showed that FKBP12 protein sequences are highly conserved across different species (Fig. 1d). In particular, the hydrophobic amino acids that constitute drug-binding sites in the H. sapiens FKBP12 protein are also conserved in PiFKBP12 (Van Duyne et al. 1991; Duyne et al. 1993), implying that rapamycin can potentially bind to PiFKBP12. As the rapamycin-FKBP12 complex interacts with the FRB domain of TOR to inhibit its function, we further analyzed the FRB domain sequence of PiTOR. Multiple sequence alignment showed that its FRB domain shares high sequence similarity with those of other species (Fig. 1e), implying that the rapamycin–FKBP12 complex can inhibit PiTOR activity.

Rapamycin inhibits mycelial growth and sporulation of P. italicum

To investigate PiTOR function, 5µL of P. italicum spore suspension (1 × 10⁶ spores/mL) were inoculated onto Potato Dextrose Agar (PDA) plates containing different concentrations of rapamycin. As rapamycin was dissolved in dimethyl sulfoxide (DMSO), plates containing the same volume of DMSO were used as controls. After 5 days post-inoculation (dpi) (Fig. 2a and b) or 7 dpi (Fig. 2c and d), large fungal colonies formed on the control plate. Light blue, fluffy mycelia were developed in the center (layer I), surrounded by three visible concentric mycelial layers. The fluffy center was directly surrounded by a light blue, powdery layer (layer II) containing a large number of conidia, which was enclosed by a white, fluffy ring (layer III). A grey layer (layer IV) was formed at the outermost region, consisting of numerous outgrown hyphae. Colony size was smaller after rapamycin treatment, even at a concentration of 1 nM (1 nmol/L), indicating that P. italicum is sensitive to rapamycin (Fig. 2a, b). Rapamycin inhibited P. italicum growth in a dose-dependent manner when the concentration increased from 1 nM to 1 µM. Interestingly, the inhibition was not further enhanced with higher rapamycin concentrations (> 1 µM) (Fig. 2c, d), suggesting that rapamycin can only partially suppress PiTOR function.

Fig. 2.

Rapamycin inhibits mycelial growth and sporulation of P. italicum. (a–b) Colony morphology (a) and colony diameter (b) of P. italicum after 5 days post-inoculation (dpi) on PDA plates containing DMSO or 1 nM, 10 nM, 100 nM, 1 µM rapamycin (RAP). Plates containing the same volume of DMSO were used as controls. (c–d) Colony morphology (c) and colony diameter (d) of P. italicum after 7 dpi on PDA plates containing DMSO or 1 µM, 5 µM, 10 µM rapamycin. Four colony layers can be observed in the enlarged images in (c): (I) the light blue, fluffy layer in the center; (II) the light blue, powdery layer containing a large number of conidia; (III) the white, fluffy ring; (IV) the grey layer at the outermost region. (e) Microscopic images of mycelia, conidia, and conidiophores of DMSO or 1 µM rapamycin treatment after 5 dpi. (f–h) Morphology (f), cell length (g), and cell width (h) of hyphae after 8 h of incubation in Potato Dextrose Broth (PDB) medium with DMSO or 1 µM rapamycin. Scale bars represent 1 cm in (a) and (c) and 50 μm in (e) and (f). Bar plots are presented as means ± SEM (standard error of the mean). P values indicate the Student’s t-test

In addition to mycelial growth, mycelial morphology was also dramatically affected. As the rapamycin concentration increased, the mycelial center gradually turned white, accompanied by an increase in layer III and a decrease in layer II (Fig. 2a). Strikingly, the outermost gray layer (layer IV) was barely observed with 1 µM rapamycin treatment (Fig. 2a). These morphological changes demonstrate that rapamycin treatment strongly hampered mycelial growth, development, and cell differentiation. To examine in detail how rapamycin suppresses mycelial growth, microscopic images were taken from DMSO-treated control plates and plates containing 1 µM rapamycin. Hyphae in layer IV grew outward smoothly and uniformly on control plates. By contrast, hyphae were small and twisted with 1 µM rapamycin treatment (Fig. 2e), consistent with the observation that the size of layer IV was drastically reduced after rapamycin treatment (Fig. 2a). Furthermore, the density of conidia and the size of conidiophores were significantly decreased in the layer II after rapamycin treatment (Fig. 2e), consistent with the colony morphological changes.

To address the direct effect of rapamycin on hyphal growth, hyphae were incubated with either DMSO or 1 µM rapamycin in Potato Dextrose Broth (PDB) liquid medium for only 8 h. While hyphae grew radially outward in the DMSO control, hyphae were twisted and showed disorganized growth upon rapamycin treatment (Fig. 2f). The cell length was significantly reduced upon rapamycin treatment, indicating that PiTOR primarily promotes cell elongation. To our surprise, cell width increased rather than decreased with the short-term rapamycin treatment (Fig. 2h), which may represent a compensatory response to reduced cell elongation. This phenotype is quite different from that observed on plates, where hyphal cell width was noticeably smaller after 5 days of rapamycin treatment (Fig. 2e), suggesting that inhibition of PiTOR suppresses cell growth in different ways between short-term and long-term periods. Altogether, these results indicate that PiTOR is crucial for mycelial growth and development.

Rapamycin does not affect conidial germination of P. italicum

TOR has been shown to be essential for conidial germination in V. dahlia and F. oxysporum, but not in B. cinerea or F. graminearum (Yu et al. 2014; Li et al. 2019, 2021; Xiong et al. 2019). To investigate whether PiTOR regulates conidial germination, conidial germination rate was measured on PDA plates containing either DMSO or 1 µM rapamycin. Conidia started to germinate after 6 h of incubation, and approximately 90% of them germinated after 14 h. There was no significant difference in germination rate between the control and rapamycin-treated samples (Fig. 3a, b), indicating that rapamycin does not affect conidial germination of P. italicum. These results suggest that PiTOR does not control conidial germination. However, compared with the control, hyphal length was significantly shorter after 24 h of rapamycin treatment (Fig. 3c, d), further indicating that PiTOR is pivotal for hyphal growth.

Fig. 3.

Germination and elongation of P. italicum conidia under rapamycin treatment. (a) Conidial germination on PDA plates containing DMSO or 1 µM rapamycin (RAP) at different time points (2–14 h). (b) Conidial germination rate on PDA plates with DMSO or 1 µM rapamycin at different time points (2–16 h). (c–d) Hyphal morphology (c) and hyphal length (d) after 24 h of incubation on PDA plates containing DMSO or 1 µM rapamycin. Scale bars represent 50 μm in (a) and (c). The P value in (d) indicates the Student’s t-test

Rapamycin affects the expression of genes related to membrane integrity, molecular binding, and metabolic processes in P. italicum

To further decipher the function of PiTOR at the molecular level, we performed genome-wide expression profiling analysis using RNA sequencing (RNA-Seq) in mycelia after DMSO or rapamycin treatment. Principal component analysis (PCA) revealed that the RNA-Seq data from all replicates clustered together within each treatment, indicating a high degree of similarity between replicates (Fig. 4a). Compared with the DMSO treatment, 471 differentially expressed genes (DEGs) were identified after rapamycin treatment (Fig. 4b), including 322 upregulated DEGs and 149 downregulated DEGs (Table S1). Gene ontology (GO) annotations analysis revealed that these DEGs were involved in many different pathways (Fig. 4c). In the top 20 GO terms, DEGs were mainly related to molecular binding in the category of “molecular function”, such as nucleic acid binding, coenzyme binding, nucleotide binding, nucleoside phosphate binding, anion binding, and cation binding. In the category of “biological process”, DEGs were grouped into metabolic processes of many substances, including nucleobase-containing compound, cellular amino acid, cellular aromatic compound, cellular nitrogen compound, organic cyclic compound, organonitrogen compound, organic acid, macromolecule and heterocycle. DEGs were also related to cellular biosynthetic process and organic substance biosynthetic processes in the category. We also found that the expression of the ribosomal protein L44e was decreased while no ribosomal protein showed increased expression (Fig. S1), supporting the function of TOR in promoting ribosome biogenesis. These results suggest that PiTOR regulates metabolism and biosynthesis (Wullschleger et al. 2006; Li et al. 2021; Liu and Xiong 2022; Wang et al. 2023). Moreover, the GO term “integral component of membrane” in the category of “cellular component” was associated with the largest numbers of both upregulated DEGs (66 genes) and downregulated DEGs (44 genes) (Table S2), strongly suggesting that PiTOR is crucial for membrane integrity. Consistent with this, hyphal cell morphology was markedly affected after rapamycin treatment (Fig. 2f).

Fig. 4.

RNA-Seq analysis of gene expressions after rapamycin treatment. (a) Principal component analysis of RNA-Seq data from different replicates. (b) The volcano plot showing differentially expressed genes (DEGs) after rapamycin treatment. Upregulated DEGs were shown in red, and downregulated DEGs were shown in blue. (c) Gene ontology (GO) annotations analysis of upregulated DEGs (red) and downregulated DEGs (blue) after rapamycin treatment. The top 20 GO terms are shown. (d) The expression levels of PiTOR and PiLST8 in RAN-Seq and qRT-PCR analysis. (e) The expression levels of selected DEGs related to the GO term “integral component of membrane” in RAN-Seq and qRT-PCR analysis. D, DMSO. R, rapamycin

The activity of the TORC1 complex was directly suppressed by rapamycin. The expression of the PiTOR, PiRAPTOR and PiLST8 were not affected in our RNA-Seq analysis. qRT-PCR analysis of PiTOR and PiLST8 was performed and further confirmed that 1 µM rapamycin did not affect the expression of TORC1 (Fig. 4d). Interestingly, the expression levels of PiTOR and PiLST8 were also not significantly affected even by 5 µM rapamycin treatment, further confirming that rapamycin mainly affects PiTOR activity rather than its expression (Fig. 4d). This result is also consistent with our inhibition assay, in which rapamycin at concentrations higher than 1 µM did not enhance the inhibition (Fig. 2c). We also detected the expression levels of 4 upregulated DEGs and 4 downregulated DEGs related to the GO term “integral component of membrane” by qRT-PCR. Their expression changes were very similar to our RNA-Seq data, except for one gene that showed a weak reduction of expression in qRT-PCR analysis (Fig. 4e). Taken together, the qRT-PCR analysis of these genes is consistent with our RNA-Seq data.

GO enrichment analysis of DEGs

To further analyze the RNA-Seq data, GO enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed. In the top 20 enriched GO terms for downregulated DEGs, most of them were related to metabolic, catabolic, and biosynthetic processes (Fig. 5a). Genes associated with cell wall components were enriched, including hemicellulose, xylan, cell wall macromolecule, and cell wall polysaccharide. This result suggests that rapamycin severely affected the cell wall development in P. italicum (Table S3). Similar enrichment of DEGs in cell wall-related processes upon rapamycin treatment was also observed in Arabidopsis, tomato, B. cinerea, and V. dahliae (Ren et al. 2013a; Xiong et al. 2016, 2019; Li et al. 2019), suggesting that TOR has a conserved function in promoting cell wall development in eukaryotes.

Fig. 5.

GO enrichment analysis of DEGs. (a) The top 20 enriched GO terms of downregulated DEGs after rapamycin treatment. (b) The top 20 enriched GO terms of upregulated DEGs after rapamycin treatment. Red bars indicate the -log10 values of adjusted P values corresponding to the upper x-axis. Blue dots indicate the numbers of genes corresponding to the lower x-axis

In the top 20 enriched GO groups for upregulated DEGs, “oxidoreductase activity” was the most enriched GO term (Fig. 5b and Table S3). In addition, many GO groups were associated with metabolic and catabolic processes. For instance, DEGs were enriched in amino acid-related processes, including cellular amino acid metabolic and catabolic processes, alpha-amino acid metabolic processes, and the glutamine family amino acid metabolic process (Fig. 5b and Table S3). Glutamine has been shown to activate TOR to mediate cellular metabolism (Tanigawa et al. 2021). Conversely, TOR has been found to regulate glutamine metabolism in Arabidopsis and mammals (Csibi et al. 2013; Ingargiola et al. 2023). Our results imply that this regulation might be conserved in fungi. Moreover, KEGG pathway enrichment analysis also revealed that DEGs were associated with multiple metabolic and biosynthetic processes (Fig. S2 and S3). Combined together, these results suggest that PiTOR modulates membrane integrity, cell wall development, and multiple metabolic and catabolic processes.

Discussion

PiTOR regulates mycelial growth and development

In budding yeast and fission yeast, two TOR paralogs exist and can form TORC1 and TORC2 (Heitman et al. 1991; Shertz et al. 2010; Ikai et al. 2011). In contrast, mammals and many plants have only one TOR gene (Zheng et al. 1995; Kim et al. 2002; Shertz et al. 2010; Ren et al. 2013a; Xiong et al. 2016; Liu and Xiong 2022). Similar to B. cinerea, we identified only one TOR gene in P. italicum, whose amino acid sequence shares 61.76% identity with BcTOR (Fig. 1a) (Xiong et al. 2019). Besides B. cinerea and P. italicum, it was shown that Magnaporthe oryzae, F. graminearum, A. flavus, V. dahlia and S. sclerotiorum also have only one TOR gene, while F. oxysporum has two TOR genes (Yu et al. 2014; Marroquin-Guzman and Wilson 2015; Li et al. 2019, 2021, 2023; Jiao et al. 2023). The two TOR paralogs in budding yeast and fission yeast were considered to result from the whole-genome duplication events and independent segmental gene duplication events, respectively (Shertz et al. 2010). Gene duplication of TOR may drive subfunctionalization of paralogs (Shertz et al. 2010). Similarly, duplication of TOR in some filamentous fungi might promote survival in certain environments.

Other core subunits of TORC1 or TORC2 (RAPTOR, RICTOR, and LST8) were also identified in P. italicum (Table 1), suggesting that both TORC1 and TORC2 may exist in P. italicum. The highly conserved FRB domain and FKBP12 sequence of P. italicum further suggest that PiTOR may be sensitive to rapamycin (Fig. 1d, e). Consistently, we found that very low concentrations of rapamycin could inhibit mycelial growth and sporulation in P. italicum (Figs. 2 and 3c and d). Mycelial cell length was significantly reduced by both short-term and long-term rapamycin treatment, indicating that PiTOR promotes cell elongation.

Rapamycin can severely suppress mycelial growth of some filamentous fungi, such as B. cinerea, V. dahliae, F. oxysporum and F. graminearum (Yu et al. 2014; Li et al. 2019, 2021; Xiong et al. 2019). In A. flavus, 30 ng/mL rapamycin strongly inhibited mycelial growth, but the inhibition was not further enhanced even when the rapamycin concentration reached 100 ng/mL (Li et al. 2023). Here we also found that rapamycin only partially inhibits mycelial growth in P. italicum (Fig. 2a-d). One possibility is that the rapamycin-PiFKBP12 complex binds to the FRB domain of PiTOR with lower affinity compared to that in some other fungi. However, this cannot fully explain the observation, since a very high concentration of rapamycin (10 µM) also did not block mycelial growth of P. italicum. Alternatively, the rapamycin-PiFKBP12 binding on the FRB domain may only partially inhibit RAPTOR binding or TOR function in P. italicum. This is partially supported by observations in Arabidopsis, where high concentrations of rapamycin cannot fully block seedling growth of ScFKBP12 transgenic lines (Ren et al. 2013a).

PiTOR regulates mycelial differentiation but not conidial germination

P. italicum formed four different layers on PDA plates (Fig. 2). Rapamycin treatment strongly affected the distribution of these layers, suggesting that inhibition of PiTOR disturbed the differentiation of mycelial cells. The TOR pathway has been reported to regulate cell differentiation in animals, plants, and yeast. For example, TOR can modulate trophectoderm differentiation in the mouse blastocyst and the timing of neuronal cell differentiation in Drosophila melanogaster (Martin and Sutherland 2001; Bateman and McNeill 2004). TOR is linked to the formation and maintenance of the shoot apical meristem and root meristem in Arabidopsis (Montané and Menand 2013; Pfeiffer et al. 2016). In fission yeast, TOR controls cell differentiation and sexual development in response to starvation conditions (Matsuo et al. 2007; Weisman et al. 2007). Our results suggest that TOR also contributes to cell differentiation in P. italicum.

Although rapamycin inhibited conidiophore size and sporulation of P. italicum, rapamycin did not affect its conidial germination, suggesting that PiTOR is not involved in conidial germination (Fig. 3), despite the fact that conidial germination is also triggered by nutrient signals. Rapamycin was shown to suppress spore germination in budding yeast, F. oxysporum and A. flavus, but not in F. graminearum and B. cinerea (Yu et al. 2014; Xiong et al. 2019; Alfatah et al. 2021; Li et al. 2021). According to these reports, the function of TOR in spore germination does not appear to be conserved. Spore germination is a multi-step process, and the mechanisms of spore germination may differ between species (Sephton-Clark and Voelz 2018). Thus, the nutrient-activated TOR pathway may be important for spore germination in some fungi but not in others.

PiTOR regulates cell membrane- and cell wall-related processes

In filamentous fungi, inhibition of the TOR pathway, either by TOR inhibitors or through deletion of downstream targets, impairs cell elongation and triggers noticeable hyphal morphological changes (Yu et al. 2014; Li et al. 2019, 2021; Xiong et al. 2019). Consistently, we found that rapamycin dramatically affected the hyphal morphology of P. italicum in both short-term (8 h) and long-term (5 days) treatment. GO annotations analysis revealed that the GO term “integral component of membrane” was associated with the largest group of DEGs (Fig. 4 and Table S2). GO enrichment analysis further showed that some downregulated DEGs were associated with cell wall components in P. italicum. These results demonstrate that TOR inhibition perturbs cell membrane- and cell wall-related processes, leading to morphological abnormalities in P. italicum.

In budding yeast, the TORC2–Ypk1/2 signaling controls sphingolipid biosynthesis, and TORC2 is considered as a main regulator of plasma membrane homeostasis and CWI (Niles et al. 2014; Thorner 2022). However, the fact that both rapamycin and nutrient starvation can activate the CWI pathway indicates that TORC1 also modulates cell wall development in budding yeast (Torres et al. 2002). In Arabidopsis, the glucose–TOR pathway controls cell wall biosynthesis (Ren et al. 2013a; Xiong et al. 2013). Inhibition of the TOR pathway also affected the expression of many cell wall-related genes in tomato, B. cinerea, A. flavus, V. dahliae and P. italicum (Xiong et al. 2016, 2019; Li et al. 2019, 2023), indicating that TOR has a conserved function in regulating cell wall development in plants and fungi.

PiTOR regulates amino acid-related processes

In addition to cell wall-related processes, enriched GO terms and enriched KEGG pathways are related to many metabolic and catabolic processes (Fig. 5 and Fig. S2), including some amino acid-related processes. This result suggests that PiTOR is involved in amino acid metabolism. Amino acids are crucial nitrogen-containing upstream signals of the TOR pathway (Liu and Xiong 2022). Glutamine is the most abundant amino acid in many eukaryotic cells and can activate TORC1 to modulate cellular metabolism (Tanigawa et al. 2021). Inversely, TOR has been shown to regulate glutamine metabolism in Arabidopsis and mammals (Csibi et al. 2013; Ingargiola et al. 2023). In our RNA-Seq data, upregulated DEGs were enriched in the glutamine family amino acid metabolic process, although the number of DEGs was not high. This finding implies that TOR also mediates glutamine metabolism in P. italicum.

Further research is needed to address PiTOR function through knockout or knockdown lines, as well as to identify downstream components of the TOR pathway in P. italicum. Moreover, as P. italicum and P. digitatum are sister species and their TOR proteins share high sequence identity, it is possible that PdTOR has functions similar to those of PiTOR. Therefore, P. digitatum may also be sensitive to rapamycin. Rapamycin has the potential to be used to control both citrus green mold and citrus green mold.

Materials and methods

Fungal strain and culture conditions

The P. italicum strain (SHBCC D23009) was ordered from the Shanghai Bioresource Collection Center. Fungi were inoculated on PDA plates and cultured at 28 ℃ in the dark for 7 days. Spores were then collected with water and stored at 4 ℃. Spores were cultured at 28 ℃ for each experiment.

Sequence blast, alignment and phylogenetic analysis

Protein sequences of TORC1 subunits, TORC2 subunits and FKBP12 in H. sapiens, D. melanogaster, (A) thaliana, C. reticulata, S. lycopersicum, S. tuberosum, S. cerevisiae, and (B) cinerea were downloaded from NCBI (https://blast.ncbi.nlm.nih.gov) (Wullschleger et al. 2006; Ren et al. 2013b; Xiong et al. 2016, 2019; Brunkard 2020; Deng et al. 2020; Liu and Xiong 2022). TOR protein sequences of S. cerevisiae and B. cinerea were used to identify TOR proteins in P. italicumin, P. digitatum, and G. candidum through BLAST in both NCBI and EnsemblFungi (http://fungi.ensembl.org). Protein sequences of TORC1 subunits, TORC2 subunits and FKBP12 of S. cerevisiae and B. cinerea were used to identify their homologues in P. italicumin. Multiple sequence alignment was performed using the MUSCLE tool with the Snapgene software. Phylogenetic trees were constructed using the Neighbor-Joining method with the MEGA11 software (Tamura et al. 2021). The percentages of replicate trees in which the associated taxa clustered together following the bootstrap test (1000 replicates) are shown next to the branches.

Rapamycin treatment

Rapamycin (S1039) was ordered from Selleck Chemicals. It was dissolved in DMSO to make the 10 mM (10 mmol/L) mother solution. Different volumes of Rapamycin solution were added into PDA media to make final concentrations (1 nM, 10 nM, 100 nM, 1 µM, 5 µM, 10 µM). DMSO were added to different groups so that they contained the same volume of DMSO. The same volume of DMSO was also added in the control group. To check colony size and mycelial development, 5 µl of spore suspension was added on PDA plates containing either DMSO or rapamycin. Spores were cultured for 5 days in Fig. 2a and b, and spores were cultured for 7 days in Fig. 2c and d. To check hyphal development and hyphal cell morphology, a small amount of spores were added into PDB media and were cultured for 7 days at 28 ℃ with 200 rpm in a shaker. Colonies were then transferred into new PDB media containing DMSO or 1 µM rapamycin, and were cultured for 8 more hours. At least three biological replicates were performed for each treatment. Mycelial and spore images were taken with a Leica DM6000B microscope.

Germination experiment

5 µl of spore suspension was added on PDA plates containing DMSO or 1 µM rapamycin. Plates were then cultured at 28 ℃ in the dark. Widefield images were taken with the Leica DM6000B microscope every two hours. Spores with hyphae were accounted as germinated spores.

RNA library construction and sequencing

Spores were initially cultured in PDB media for 7 days. Following this, the mycelia were transferred into fresh PDB media containing either DMSO or 1 µM rapamycin and cultured for 8 h. The mycelia were then collected for RNA extraction. 4 independent replicates were performed for each treatment. RNA extraction and RNA-Seq analysis were carried out by Majorbio Bio-pharm Biotechnology Company (Shanghai, China). Four independent biological replicates were performed for each treatment. Total RNA was extracted using TRIzol^® Reagent, and its quality was assessed with a 5300 Bioanalyzer (Agilent) and quantified using the ND-2000 (NanoDrop Technologies). RNA purification, reverse transcription, library construction, and sequencing were performed at Majorbio Bio-pharm Biotechnology Company (Shanghai, China) following the manufacturer’s guidelines. The RNA-seq transcriptome library was prepared using Illumina^® Prep, Ligation (San Diego, CA). Sequencing was conducted on the NovaSeq X Plus platform (PE1500). The assembled transcripts were compared against the NCBI protein nonredundant (NR), Clusters of Orthologous Groups of proteins (COG), and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases using Diamond. Proteins with the highest sequence similarity to the transcripts were identified to retrieve their functional annotations, with a typical E-value cut-off of less than 1.0 × 10^− 5. Raw sequencing data are available in NCBI with Bioproject No. PRJNA1232385.

Gene expression analysis of RNA-Seq data

Differential expression analysis was performed using the DESeq2 (Love et al. 2014). Genes with expression fold change > 2 were considered as differentially expressed genes (DEGs). RNA-Seq analysis including PCA analysis, GO annotations analysis, GO enrichment analysis and KEGG pathway enrichment analysis were performed with the Majorbio online platform (https://cloud.majorbio.com/). Goatools (https://github.com/tanghaibao/GOatools) and Python scipy (https://scipy.org/install/) were used for enrichment analysis of GO and KEGG pathway, respectively. The total number of upregulated DEGs and downregulated DEGs in each GO term was used to select the top 20 GO terms in GO annotation analysis. The display order of these GO groups was based on the number of upregulated DEGs in each group. p-adjust was used to select both the top 20 enriched GO terms and the top 20 enriched KEGG pathways.

qRT-PCR analysis

To confirm our RNA-Seq data, qRT-PCR analysis was performed for several genes. A small amount of spores were added into PDB media and were cultured for 7 days at 28 ℃ with 200 rpm in a shaker. Colonies were then transferred into new PDB media containing DMSO, 1 µM rapamycin or 5 µM rapamycin, and were cultured for 8 more hours. RNA was extracted with Omega Fungal RNA Kit (R6840-01), and cDNA was synthesized with PrimeScriptTM RT reagent Kit with gRNA Eraser (Perfect Real Time). qRT-PCR analysis was performed with Roche LightCycler 96 Instrument. β-Actin was used as control. Primers are listed in Table S4.

Measurement, figures and statistics

The colony diameter was measured with rules. Hyphal cell length and cell width were measured with the software ImageJ. The two-tailed Student’s t-test was used for statistical analysis. Bar plots are made with Excel and presented as means ± SEM (standard error of the mean). P-value < 0.05 was considered significant. Combined figures were made with the software Illustrator.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

The research funding was acquired by Tingting Chen. Experiments were designed by Tingting Chen and Kai Wang. Material preparation, data collection and analysis were performed by Tingting Chen, Hang Chen, Lan Ge, Xiaoke Fu, Tingting Hu and Kai Wang. The first draft of the manuscript was written by Tingting Chen, Hang Chen and Jing Zheng. The manuscript was revised by Kai Wang. All authors read and approved the final manuscript.

Funding

This research was funded by Guiding Science and Technology Plan Projects of Meishan Science and Technology Bureau, grant number 2023KJZD163.

Data availability

The data presented in this study are available on request to the corresponding author. Raw sequencing data are available in NCBI with Bioproject No. PRJNA1232385.

Declarations

Competing interests

The authors declare no competing interests.