Gold nanoparticle-mediated metabolic engineering in Trichoderma longibrachiatum MD33 unveils a hybrid terpenoid-alkaloid pathway for enhanced dendrobine biosynthesis

Abstract

Background

Dendrobine, a neuroprotective and anticancer sesquiterpenic alkaloid, is primarily sourced from endangered Dendrobium orchids, posing sustainability challenges to its production. Endophytic fungi, such as Trichoderma longibrachiatum MD33, offer an alternative; however, unresolved biosynthetic pathways and low yields hinder industrial scalability. Enhancing fungal metabolism through nanotechnology could address these limitations; however, nanoparticle-mediated engineering remains unexplored for dendrobine biosynthesis. This study aimed to (1) optimize dendrobine production in T. longibrachiatum MD33 using gold nanoparticles (CH-AuNPs) functionalized with alkaloid precursors and (2) elucidate the biosynthetic pathway to enable targeted metabolic engineering. CH-AuNPs were chemically synthesized, functionalized with L-phenylalanine, L-tyrosine, and tyramine, and applied to fungal cultures at concentrations of 0.5–20.0 mg/L. Multi-omics analyses (transcriptomics, proteomics, and metabolomics) identified pathway enzymes, and oxidative stress markers and dendrobine yields were quantified.

Results

Dose-dependent CH-AuNP exposure (10.0 mg/L optimal) elevated dendrobine production by 63.7%, balancing pathway activation and oxidative stress. Multi-omics analysis revealed a hybrid terpenoid-alkaloid pathway, wherein sesquiterpene scaffolds from the mevalonate pathway merge with ornithine-derived piperidine moieties. This process is regulated by sesquiterpene synthases (TPS), cytochrome P450s (CYP71D1), and O-methyltransferases (COMT). Metabolomic analysis provided direct evidence for the rechanneling of nitrogen metabolism, with depletion of glutamate and ornithine pools and accumulation of polyamine pathway intermediates such as putrescine, supporting the transcriptional upregulation of ornithine decarboxylase (ODC). Mechanistically, low-to-moderate oxidative stress induced by CH-AuNPs activated redox-sensitive transcription factors and stress-responsive pathways, which in turn upregulated terpenoid and alkaloid biosynthesis genes. This controlled stress response enhanced precursor flux and enzyme activity, leading to increased dendrobine synthesis without triggering cellular damage in the cells. Concentrations > 10.0 mg/L suppressed metabolism owing to oxidative damage.

Conclusions

CH-AuNPs act as precision tools to upregulate dendrobine biosynthesis in T. longibrachiatum MD33, resolving the hybrid pathway and establishing this fungus as a sustainable production platform for dendrobine. The dose-dependent response highlights the dual role of nanoparticle-mediated engineering in metabolic enhancement and stress induction. This integration of nanotechnology and multi-omics bridges the critical gaps in fungal biotechnology, enabling scalable and eco-friendly alkaloid synthesis. Future applications include CRISPR-AuNP genome editing and bioreactor optimization, which will advance pharmaceutical and environmental biotechnologies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12934-026-02936-9.

Introduction

Dendrobine, a bioactive sesquiterpenoid alkaloid with a complex fused-ring structure, predominantly isolated from the medicinal orchid Dendrobium nobile [1–4], has garnered significant attention for its neuroprotective, anti-inflammatory, and anticancer properties [5, 6]. Traditionally extracted from plant tissues, dendrobine production remains constrained by the slow growth of Dendrobium species, seasonal variability, and destructive harvesting practices, which threaten biodiversity and limit commercial scalability [7, 8]. To address these challenges, microbial biosynthesis using endophytic fungi such as Trichoderma longibrachiatum MD33 has emerged as a sustainable alternative. This strain, isolated from D. nobile stems, demonstrates inherent dendrobine-producing capabilities, yet yields remain suboptimal due to incomplete metabolic pathway characterization and regulatory inefficiencies [9–12]. Thus, developing a cost-effective and scalable production platform is essential not only for scientific understanding but also for translating the potential of dendrobine into tangible therapeutic and commercial applications.

Despite advances in fungal metabolic engineering, critical gaps persist in understanding dendrobine biosynthesis [2, 13]. First, its hybrid terpenoid-alkaloid structure suggests a convergent pathway involving both mevalonate (MVA)-derived sesquiterpenoid precursors and alkaloid moieties derived from aromatic amino acids or polyamines [14]. However, the enzymatic machinery orchestrating this convergence—particularly cytochrome P450 monooxygenases (CYPs) and prenyltransferases—remains unannotated in Trichoderma species [15, 16]. Second, fungal systems often suffer from precursor scarcity, as critical intermediates like acetyl-CoA and aromatic amino acids are diverted toward primary metabolism under standard cultivation conditions [17, 18]. Third, while abiotic elicitors (e.g., jasmonic acid, heavy metals) can stimulate secondary metabolism, their broad-spectrum effects often disrupt cellular homeostasis, leading to growth inhibition or oxidative stress [5, 12, 19]. Specifically, oxidative stress, characterized by the accumulation of reactive oxygen species (ROS), can act as a double-edged sword: at low levels, it serves as a signaling cue to activate defense-related secondary metabolism, whereas at high levels, it causes cellular damage and metabolic shutdown [20]. Thus, a targeted strategy to enhance precursor flux while precisely modulating pathway genes is essential to unlock the full potential of fungal dendrobine biosynthesis.

Nanotechnology offers a transformative approach to metabolic engineering by enabling the targeted delivery of precursors and elicitors [13]. Gold nanoparticles (AuNPs) have unique advantages owing to their biocompatibility, tunable surface chemistry, and ability to act as both carriers and nano-elicitors [21, 22]. While AuNPs have been previously used to enhance secondary metabolite production in plant cell cultures (e.g., paclitaxel in Taxus) and bacterial systems, their application in fungal systems, particularly for the biosynthesis of hybrid terpenoid-alkaloids such as dendrobine, remains largely unexplored [5, 23]. Most existing studies have focused on either physical elicitation (e.g., ROS induction) or passive delivery of generic precursors, and few have combined rational precursor loading with multi-omics-driven pathway elucidation in a fungal host [19, 24]. This study introduces a precision nanoparticle platform in which AuNPs (CH-AuNPs) are specifically functionalized with biosynthetically relevant precursors (L-phenylalanine, L-tyrosine, and tyramine) to simultaneously address precursor scarcity and pathway regulation, a dual strategy not yet reported for dendrobine or most fungal alkaloids.

The choice of L-phenylalanine (L-Phe), L-tyrosine (L-Tyr), and tyramine as functionalization molecules is based on the predicted hybrid nature of dendrobine biosynthesis and the interconnected nature of aromatic amino acid metabolism [25, 26]. Based on prior phytochemical and genomic analyses of Dendrobium species, dendrobine is proposed to arise from the fusion of a sesquiterpenoid scaffold (derived from the mevalonate pathway) with an alkaloid moiety, likely originating from aromatic amino acid metabolism [14, 27, 28]. Although these three compounds are metabolically interconvertible in vivo, all three were deliberately co-loaded onto CH-AuNPs to address multiple potential bottlenecks simultaneously. L-Phenylalanine serves as a key entry point into the phenylpropanoid and shikimate pathways and can be converted to L-Tyrosine. L-Tyrosine is a direct precursor of tyramine and other phenolic amines implicated in alkaloid diversification [29, 30]. Tyramine is a known intermediate in plant alkaloid biosynthesis and may act as both a precursor and signalling molecule that modulates stress-responsive secondary metabolism [31, 32]. By providing all three, the aim was to (1) ensure an adequate supply of aromatic carbon skeletons (2), supplement the immediate precursor pool for nitrogen-containing moiety synthesis, and (3) potentially elicit stress-responsive pathways linked to alkaloid production, thereby covering multiple nodes in the proposed biosynthetic network and minimizing limitations that could arise if only a single precursor was supplied [33, 34].

The key unresolved questions addressed in this study are as follows: (1) How do functionalized AuNPs interact with fungal transporters to facilitate precursor internalization (2)? What transcriptional and metabolic shifts occur under nanoparticle-mediated stress, and how do they intersect with dendrobine biosynthesis (3)? What is the dose-dependent relationship between CH-AuNP-induced oxidative stress, its dual role in metabolic elicitation versus cytotoxicity, and the resulting dendrobine yield? Addressing these questions is critical for designing precision tools for microbial alkaloid production.

This study bridges these gaps by investigating gold nanoparticle-mediated metabolic engineering in Trichoderma longibrachiatum MD33. The central hypothesis posits that chemically synthesized AuNPs (CH-AuNPs) functionalized with dendrobine precursors (L-phenylalanine, L-tyrosine, and tyramine) enhance precursor bioavailability, upregulate terpenoid-alkaloid pathway genes, and induce a controlled oxidative stress response that further stimulates secondary metabolism, thereby optimizing dendrobine yields. A critical aspect of testing this hypothesis involved the quantitative confirmation of the co-loading of the three precursor molecules (L-phenylalanine, L-tyrosine, and tyramine) on the CH-AuNPs, providing a stoichiometric basis for subsequent dose-response studies and mechanistic interpretation. To test this hypothesis, CH-AuNPs were synthesized and characterized, and their dose-dependent effects on fungal metabolism were evaluated. Transcriptomic and metabolomic analyses were integrated to elucidate the underlying hybrid biosynthetic pathways. The integration of transcriptomic and metabolomic profiles was designed to generate a correlative, evidence-based model of the dendrobine biosynthetic pathway in T. longibrachiatum MD33, providing a testable framework for future functional studies.

Materials and methods

Synthesis and characterization of gold nanoparticles (CH-AuNPs)

Gold nanoparticles (CH-AuNPs) were chemically synthesized using trisodium citrate. Briefly, 20 mL of 1.0 mM chloroauric acid (HAuCl₄) was heated to boiling under continuous magnetic stirring (500 rpm). A 2 mL aliquot of 1% (w/v) trisodium citrate dihydrate solution was rapidly introduced, inducing a visible color transition from pale yellow to reddish violet, indicative of the formation of colloidal CH-AuNPs. The reaction mixture was maintained at a boiling temperature for 15 min to ensure complete reduction, followed by cooling to room temperature with continuous stirring (Supplementary Fig. S1A). To ensure colloidal uniformity and remove larger aggregates, the synthesized nanoparticles underwent sequential centrifugation: first at 2,000 rpm for 10 min to discard aggregates and obtain a monodisperse fraction, then at 10,000 rpm for 20 min, and finally at 15,000 rpm for 30 min to obtain the desired fraction.

The synthesized CH-AuNPs were characterized using UV-Vis spectrophotometry (Shimadzu UV-1800), scanning electron microscopy (SEM; Hitachi SU8000 FE-SEM), dynamic light scattering (DLS; Malvern Zetasizer Nano ZS), and zeta potential measurements. UV-Vis absorbance spectra (200–800 nm) were recorded, revealing a distinct localized surface plasmon resonance (LSPR) peak at 550 nm, indicating spherical nanoparticles with an average diameter of approximately 50 nm. SEM imaging at 110,000× magnification confirmed a monodisperse spherical morphology with a narrow size distribution (50 ± 5 nm). DLS measurements indicated a hydrodynamic diameter of 55 ± 3 nm and a polydispersity index (PDI) of 0.15, confirming the colloidal stability. Zeta potential measurements in 0.02 M NaCl (pH 7.0) demonstrated a surface charge of -32 ± 2 mV, reflecting strong electrostatic repulsion and minimal aggregation (Fig. 1A).

Fig. 1.

An integrated workflow for nanoparticle-mediated metabolic engineering of Trichoderma longibrachiatum MD33 for enhanced dendrobine biosynthesis: A Synthesis and characterization of CH-AuNPs (chemical reduction, UV-Vis/SEM, DLS/Zeta potential). B Functionalization with metabolic precursors (core structure, adsorption of Phe/Tyr/Tyramine, HPLC validation). C Fungal cultivation and CH-AuNP treatment (culture, dose-dependent addition, incubation). D Biological response and dendrobine production (molecular pathway upregulation, dose-response yield curve with optimal 10 mg/L dose)

Functionalization of CH-AuNPs with precursor compounds

Chemically synthesized gold nanoparticles (CH-AuNPs) were functionalized with three key alkaloid biosynthetic precursors, L-phenylalanine (L-Phe), L-tyrosine (L-Tyr), and tyramine, to enhance their role as targeted nano-elicitors in Trichoderma longibrachiatum MD33. Aqueous solutions of each precursor were prepared at a concentration of 1 mg/mL in distilled water. For functionalization, the CH-AuNP suspension was combined with the precursor solutions at a 1:1 (v/v) ratio. The mixtures were gently stirred at room temperature for 2 h to facilitate the adsorption of the precursor molecules onto the nanoparticle surfaces (Supplementary Fig. S1B). Following incubation, unbound precursors were removed by centrifugation at 10,000 rpm for 15 min. The resulting pellets were washed thrice with distilled water to ensure the complete removal of non-adsorbed compounds.

Successful loading was initially assessed using UV-Vis spectroscopy. To further confirm the presence and quantify the loading of all three precursors, given the similarity in their structures and UV absorption profiles, the functionalized CH-AuNPs were subjected to acid hydrolysis (6 M HCl, 110 °C, 24 h) to release the bound molecules. The hydrolysate was neutralized, filtered, and analyzed using High-Performance Liquid Chromatography (HPLC) on an Agilent 1260 Infinity II system equipped with a C18 column (4.6 × 250 mm, 5 μm). Isocratic elution was performed using a mobile phase of 10 mM ammonium acetate (pH 5.0) and methanol (95:5, v/v) at a flow rate of 1.0 mL/min. Detection was performed using a diode array detector (DAD) set at 210, 274, and 280 nm for L-Phe, L-Tyr, and tyramine, respectively. Peaks were identified by comparing the retention times of certified standards. This chromatographic analysis confirmed the co-loading of all three precursors in the liposomes. Quantification was performed using external calibration curves, and loading efficiencies were calculated as the mass of precursor adsorbed per mass of CH-AuNPs (w/w), yielding 82 ± 3% for L-Phe, 79 ± 4% for L-Tyr, and 85 ± 3% for tyramine. The corresponding molar ratios of the loaded precursors were approximately L-Phe: L-Tyr: tyramine = 1.0:0.8:1.1. Functionalized nanoparticles were stored at 4 °C in amber vials to prevent photodegradation and were used within one week for further studies (Fig. 1B).

Fungal cultivation and experimental treatment

The endophytic fungal strain Trichoderma longibrachiatum MD33 (NCBI accession: MN826683), originally isolated from Dendrobium nobile stems, was cultivated in potato dextrose (PD) broth (pH 6.5) at 28 °C with continuous shaking (120 rpm) for 120 h [7, 9]. For nanoparticle treatment, fungal cultures were inoculated into 1-liter flasks containing 500 mL PD broth supplemented under two conditions: (1) with precursor-loaded CH-AuNPs at final concentrations of 0.5, 1.0, 5.0, 10.0, and 20.0 mg/L; (2) a control receiving an equivalent volume of sterile distilled water. All treatments were performed in triplicates. The cultures were incubated for 45 d under the same conditions (28 °C, 120 rpm). Biological triplicates were independently maintained for each treatment and control group (Fig. 1C and D).

Dendrobine extraction and quantification

Post-incubation, dendrobine was extracted from fungal metabolites using liquid-liquid extraction. The culture filtrate was mixed 1:1 (v/v) with chloroform, agitated at 180 rpm for 12 h, and centrifuged to isolate the organic phase of the filtrate. The chloroform layer was concentrated via rotary evaporation (75 °C) and reconstituted in 5 mL methanol for analysis. Quantification was performed using a Thermo Scientific Dionex UltiMate 3000 UHPLC system coupled to a Q-Exactive Orbitrap HF-X mass spectrometer. Separation was performed on a 150 × 2.1 mm column (1.9 μm particles) with a mobile phase of 0.1% formic acid and acetonitrile (95:5, v/v) at a flow rate of 0.3 mL/min. The detection parameters included a spray voltage of 3.5 kV, capillary temperature of 350 °C, and m/z range of 80–1200. Dendrobine was identified by matching its retention time, molecular weight (264.19 Da), and fragmentation patterns to those of a certified reference standard (> 99% purity, Chengdu DeSiTe Biological Technology, Chengdu, China).

Sample Preparation and experimental design

For transcriptomic and metabolomic analyses, Trichoderma longibrachiatum MD33 cultures were divided into two experimental groups: an untreated control (CK) and a treatment group exposed to precursor-loaded gold nanoparticles (CH-AuNPs) at concentrations of 0.5, 1.0, 5.0, 10.0, and 20.0 mg/L. Each group included three biological replicates, designated as CK1–CK3 (control) and NP05_1–NP20_3 (treated, with numerical suffixes indicating CH-AuNP concentrations). Following incubation, the fungal biomass was harvested via centrifugation (8,000 ×g for 10 min at 4 °C). The pellet was washed thrice with ice-cold phosphate-buffered saline (PBS; 20 mL per wash, 8,000 ×g, 5 min) to remove residual culture medium. Excess moisture was gently absorbed using sterile filter paper, and the pellet was flash-frozen in liquid nitrogen (5 min) to quench metabolic activity. The samples were wrapped in aluminum foil, stored on dry ice (-70 °C), and transported to Kegene Co. Ltd. (Shandong, China) under cryogenic conditions to ensure RNA and metabolite extraction integrity.

Transcriptomic and metabolomic profiling

For transcriptomic analysis, total RNA was extracted using TRIzol reagent, and the quality was assessed using NanoDrop spectrophotometry and Agilent Bioanalyzer, followed by sequencing on an Illumina NovaSeq 6000 system (150 bp paired-end reads). Raw reads were trimmed (Trimmomatic), mapped to the T. longibrachiatum genome (HISAT2), and analyzed for differential gene expression (DESeq2: |log2 fold change| >1, adjusted p < 0.05), with functional enrichment performed via topGO (Gene Ontology) and KOBAS (KEGG pathways). For metabolomics, metabolites were extracted from fungal biomass using methanol/acetonitrile/H₂O (2:2:1, v/v/v), centrifuged (14,000 ×g, 20 min, 4 °C), and vacuum-dried. Reconstituted samples were analyzed by UHPLC-Q-Exactive Orbitrap MS (Thermo Vanquish) equipped with a HILIC column (2.1 × 100 mm, 1.7 μm), and the data were processed using XCMS and the metabolite annotation and pathway mapping software, CAMERA, for peak alignment.

Statistical analysis

Data normalization and multivariate analysis (principal component analysis [PCA] and orthogonal partial least-squares discriminant analysis [OPLS-DA]) were conducted using the R package ropls. Differentially expressed metabolites (DEMs) were identified using variable importance in projection (VIP > 1) and Student’s t-test (‘p’ <0.05). The model robustness was validated using 7-fold cross-validation and permutation testing. The results are presented as the mean ± standard deviation (SD) of triplicate experiments. For dose-response analysis of dendrobine production, one-way ANOVA followed by Tukey’s post-hoc test was performed to determine significant differences between the treatment groups (p < 0.05). Data are presented as mean ± standard deviation (SD) of three independent biological replicates per group.

Results

Characterization of CH-AuNPs using UV-Vis spectrophotometer, scanning electron microscopy (SEM) and zeta potential analysis

UV-Vis spectrophotometry provides critical insights into the optical properties of Au nanoparticles, which reflect their particle size, shape, and concentration. In this analysis, the absorbance values were measured across a range of wavelengths to observe the plasmon resonance peak, a characteristic feature of CH-AuNPs arising from the collective oscillations of electrons on the particle surface when exposed to light. The absorbance values recorded at various wavelengths were 0.649 at 414 nm, 0.596 at 450 nm, 0.612 at 500 nm, 0.729 at 550 nm, 0.689 at 600 nm, 0.671 at 650 nm, and 0.639 at 700 nm. Notably, the highest absorbance peak at 550 nm (0.729) indicates the localized surface plasmon resonance (LSPR) peak, which is typically centered around 520–580 nm for gold nanoparticles, depending on their size and distribution. This LSPR peak signifies the presence of spherical Au nanoparticles, as shifts in this peak can occur with changes in particle morphology; larger particles or those with non-spherical shapes typically exhibit red-shifted peaks. SEM analysis revealed that the CH-AuNPs exhibited a uniform spherical morphology with minimal aggregation. The nanoparticles were homogeneously distributed across the substrate, as shown by grayscale imaging. Based on the micron marker (500 nm) and pixel size (0.902 nm), the average diameter of the CH-AuNPs was calculated to be 50 ± 5 nm. High-resolution imaging confirmed smooth nanoparticle surfaces and a narrow size distribution, consistent with successful chemical synthesis. The working distance (8.8 mm) and accelerating voltage (3.0 kV) were optimized to achieve a high spatial resolution, enabling the precise visualization of individual nanoparticles. These results validate the efficacy of the synthesis protocol for producing monodisperse spherical CH-AuNPs suitable for downstream applications. DLS measurements indicated a hydrodynamic size of 55 ± 3 nm and a PDI of 0.15, confirming the monodispersity of the NPs. Zeta potential analysis revealed that the CH-AuNPs exhibited a surface charge of -32 ± 2 mV under neutral pH conditions, indicating excellent colloidal stability. This value, which is below the threshold of -30 mV, indicates a strong electrostatic repulsion between the nanoparticles, confirming their high colloidal stability and resistance to agglomeration. Triplicate measurements demonstrated consistent reproducibility, with a standard deviation of < 5%, validating the homogeneity of the synthesized nanoparticles. The negative zeta potential was consistent with the chemically synthesized layer derived from the trisodium citrate reductant used during synthesis. Combined with UV-Vis spectrophotometry and SEM observations of monodisperse, non-agglomerated nanoparticles, these results confirmed the successful synthesis of stable CH-AuNPs suitable for fungal metabolic engineering (Fig. 1A).

Precursor loading evaluation by UV-visible spectrophotometry

The successful adsorption of L-phenylalanine, L-tyrosine, and tyramine onto chemically synthesized gold nanoparticles (CH-AuNPs) was confirmed using a combination of UV-visible spectrophotometry and High-Performance Liquid Chromatography (HPLC). Initial UV-Vis analysis showed characteristic absorption bands corresponding to the aromatic and functional groups of each precursor: L-phenylalanine exhibited a peak at 210 nm, L-tyrosine at 274 nm, and tyramine at 280 nm (Fig. 1). To unambiguously confirm co-loading and quantify the proportion of each precursor, functionalized CH-AuNPs were hydrolyzed, and the released molecules were analyzed using HPLC. Chromatographic separation successfully resolved L-Phe (retention time, Rt = 6.2 min), L-Tyr (Rt = 8.7 min), and tyramine (Rt = 12.4 min) based on comparisons with the authentic standards. Quantification via external calibration curves confirmed robust loading efficiencies of 82 ± 3%, 79 ± 4%, and 85 ± 3% for L-Phe, L-Tyr, and tyramine, respectively. The relative molar ratios of the loaded precursors were approximately L-Phe: L-Tyr: tyramine = 1.0:0.8:1.1. These findings demonstrate the effective and quantifiable functionalization of CH-AuNPs with multiple biosynthetic precursors at a defined ratio, which is crucial for interpreting the subsequent dose-dependent enhancement in dendrobine yield and understanding the role of nanoparticles as a multi-precursor co-delivery vehicle. These findings demonstrate the effective and quantifiable functionalization of CH-AuNPs with multiple biosynthetic precursors, confirming their suitability for targeted co-delivery in metabolic engineering applications (Fig. 1B).

Concentration-dependent effects of CH-AuNPs on dendrobine production in trichoderma longibrachiatum (MD33)

The dose-dependent effects of CH-AuNPs on dendrobine production in T. longibrachiatum MD33 are summarized in Fig. 2A and B. In untreated control cultures (n = 3 biological replicates), the mean dendrobine concentration was 177.74 ± 8.92 ng/ml (mean ± SD). Exposure to precursor-loaded CH-AuNPs elicited a stronger, concentration-dependent increase. At 0.5 mg/L CH-AuNPs, dendrobine production increased to 202.72 ± 10.14 ng/ml (p < 0.05 compared to control). This trend continued at 1.0 mg/L (228.72 ± 11.44 ng/ml, p < 0.01) and 5.0 mg/L (248.33 ± 12.42 ng/ml, p < 0.01). Maximum enhancement was observed at 10.0 mg/L CH-AuNPs, with dendrobine concentration reaching 290.90 ± 14.55 ng/ml, representing a 63.7% increase relative to the control (p < 0.001). However, at 20.0 mg/L, dendrobine production declined significantly to 249.74 ± 12.49 ng/ml (p < 0.01 compared to 10.0 mg/L), indicating a toxicity threshold (Fig. 1D). Statistical analysis confirmed significant differences between the treatment groups (one-way ANOVA, F (5, 12) = 45.32, p < 0.0001). These results demonstrate a nonlinear hormetic relationship between CH-AuNP concentration and dendrobine biosynthesis, with 10.0 mg/L emerging as the optimal dose.

Fig. 2.

Impact of chemically synthesized gold nanoparticles (CH-AuNPs) on Dendrobine Production in Trichoderma longibrachiatum MD33: LC-MS quantitative analysis and chromatographic profiling. A: Dendrobine concentration (ng/ml, calculated from LC-MS peak area integration) in Trichoderma longibrachiatum MD33 cultures treated with gold nanoparticle-precursor compounds (CH-AuNPs). Data represent the mean ± SD of biological replicates, illustrating nanoparticle-induced changes in dendrobine yield. B: Representative dendrobine LC-MS chromatogram (total ion current or extracted ion chromatogram) showing the retention time and peak intensity from a CH-AuNP-treated MD33 culture, confirming the identity and consistent elution of dendrobine under the optimized analytical conditions

CH-AuNPs drive transcriptional and metabolic reprogramming toward dendrobine biosynthesis

Transcriptional and metabolic profiling revealed that precursor-loaded gold nanoparticles (CH-AuNPs) specifically reprogrammed T. longibrachiatum MD33 metabolism to favor dendrobine biosynthesis and its precursors. Transcriptome-wide PCA (Fig. 3A; Supplementary Table S1-S3) showed a clear dose-dependent separation from the control, with the 10.0 mg/L treatment group exhibiting the most distinct profile, correlating with the peak dendrobine yield (Fig. 2A, Table S4). Correlation heatmaps (Fig. 3B) confirmed dose-responsive divergence, with high-dose groups (NP10–NP20) exhibiting minimal correlation with the controls (0.08–0.17). The pie chart findings demonstrate a dose-dependent transcriptional disruption in Trichoderma longibrachiatum MD33, where higher concentrations of CH-AuNPs progressively diminish the correlation with controls, indicating profound metabolic reprogramming at elevated nanoparticle doses (Fig. 3C). Analysis of differentially expressed genes (DEGs) at this optimal dose pinpointed key upregulations proposed to be within the hybrid biosynthetic pathway (Fig. 3D and F; Table S4). Venn analysis (Fig. 3E) identified 290 shared DEGs across treatments (e.g., oxidative stress pathways; Table S4) and concentration-specific responses, including 78 unique DEGs at 10.0 mg/L that were associated with dendrobine biosynthesis (Table S4). A heatmap of the shared DEGs (Fig. 3F; Table S4) revealed hierarchical clustering by dose, with sustained upregulation of stress-responsive and secondary metabolite genes (e.g., terpenoid enzymes) at higher elicitor concentrations. Collectively, CH-AuNPs induced concentration-dependent transcriptional reprogramming, peaking at 10.0 mg/L, which optimized pathways for dendrobine synthesis while suppressing growth-related processes. Functional enrichment analysis revealed that optimal CH-AuNP dosing (10.0 mg/L) triggered a specific metabolic shift towards terpenoid and glutathione-associated pathways, redirecting carbon flux from primary growth to enhance sesquiterpenoid scaffold synthesis for dendrobine production. At this dose, a balanced oxidative stress response further optimized precursor availability and enzyme activity, whereas higher concentrations (≥ 20.0 mg/L) induced cytotoxic pathways that suppressed the biosynthesis machinery (Fig. 4A-B and Table S5).

Fig. 3.

Transcriptome-wide profiling of CH-AuNP-induced transcriptional remodeling in Trichoderma longibrachiatum MD33. PCA Principal Component Analysis, DEGs Differentially Expressed Genes, CH-AuNPs precursor-loaded gold nanoparticles, CK control group. A: PCA of transcriptome data (PC1:35.0% variance; PC2:20.7% variance); B: Heatmap of pairwise sample correlations; C: Pie chart illustrating transcriptional similarity clusters; D: Bar plots of DEGs (up/downregulated) across CH-AuNP concentrations, E: Venn diagram of overlapping/concentration-specific DEGs; F: Heatmap of shared DEG expression patterns

Fig. 4.

Functional enrichment analysis highlighting dendrobine-relevant pathways in Trichoderma longibrachiatum MD33 under CH-AuNP treatment. (GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; DEGs: Differentially Expressed Genes; CH-AuNPs: precursor-loaded gold nanoparticles; CK: Control Group). A: GO enrichment analysis of shared DEGs, highlighting specific biological processes linked to secondary metabolite biosynthesis. B: KEGG pathway enrichment of shared DEGs, emphasizing pathways directly relevant to dendrobine synthesis, including terpenoid backbone biosynthesis, sesquiterpenoid and alkaloid metabolism, and related metabolic shifts

Most notably, genes encoding enzymes of the mevalonate (MVA) pathway and downstream terpenoid biosynthesis were significantly induced (Table S6). This included a + 2.5-fold increase in 3-hydroxy-3-methylglutaryl-CoA reductase (HMGR), a critical rate-limiting enzyme for sesquiterpenoid scaffold production, and a + 3.1-fold upregulation of farnesyl pyrophosphate synthase (FPPS). Concurrently, a sesquiterpene synthase (TPS) gene, proposed to cyclize farnesyl pyrophosphate (FPP) into the dendrobane skeleton, showed a + 3.0-fold increase. For the alkaloid moiety, components of the ornithine/putrescine pathway were activated, with ornithine decarboxylase (ODC) upregulated by + 1.8-fold, linking nitrogen metabolism to alkaloid precursor formation. Furthermore, a cytochrome P450 monooxygenase (CYP71D1) and an O-methyltransferase (COMT), whose significant co-expression suggests a potential role in the functionalization and maturation of the hybrid molecule, were among the most strongly upregulated genes (+ 4.0-fold and + 3.3-fold, respectively). The specific catalytic roles of CYP71D1 and COMT in dendrobine biosynthesis were inferred from their coordinated expression with pathway genes and homology to enzymes in related pathways.

Metabolomic analysis supported this transcriptional shift (Fig. 5A–F; Table S7, S8). Notably, although L-Phe, L-Tyr, and tyramine are known to be interconvertible, the treatments led to distinct perturbations in their related metabolic pools. A measurable depletion of intracellular glutamic acid and specific shifts in ornithine and polyamine-related metabolites were observed (Fig. 5D and E), rather than a uniform accumulation of all three supplemented precursors. This pattern suggests that fungal metabolism actively channels the supplied precursors toward the nitrogen-containing arm of dendrobine biosynthesis rather than simply accumulating them, and implies that the biosynthetic route may preferentially utilize a specific branch of this interconnected network under elicitation conditions. This targeted reprogramming, which concurrently upregulates both the terpenoid and putative alkaloid-forming branches while inducing metabolomic shifts indicative of precursor channeling, provides a direct, multi-omics supported explanation for the 63.7% enhancement in dendrobine yield at 10.0 mg/L CH-AuNPs (Figs. 2 and 6). Metabolomic analysis provided direct evidence for the rechanneling of nitrogen metabolism. At the optimal elicitation dose (10.0 mg/L CH-AuNPs), a significant depletion of intracellular L-glutamic acid (log₂FC = -0.36, p < 0.05) and ornithine-related metabolites was observed, concurrent with an accumulation of downstream polyamine pathway intermediates, including putrescine (log₂FC = + 0.55, p < 0.05) and N-acetylputrescine. This pattern indicates an active draw on the ornithine pool, consistent with the transcriptional upregulation of ornithine decarboxylase (ODC), and confirms the reallocation of nitrogen flux toward the formation of putative alkaloid precursors (Table S7).

Fig. 5.

Metabolic profiling of CH-AuNP-induced dose-dependent perturbations in Trichoderma longibrachiatum MD33. CH-AuNPs Precursor-Loaded Gold Nanoparticles, DEMs Differentially Expressed Metabolites, PCA Principal Component Analysis, KEGG Kyoto Encyclopedia of Genes and Genomes, CK Control Group; Z-Score Standardized metabolite expression. A: PCA of metabolic profiles (PC1:28.6% variance; PC2:12.1% variance), showing dose-dependent clustering; B: Bar plots of DEM counts (up/downregulated) across CH-AuNP concentrations; C: Venn diagram of shared/unique DEMs among treatment groups; D: Heatmap of DEM expression trends, highlighting metabolic suppression; E: Z-score analysis of conserved metabolite changes across multiple doses; F: KEGG pathway impact analysis, emphasizing disrupted amino acid, energy, and secondary metabolism

Fig. 6.

Proposed hybrid terpenoid-alkaloid biosynthetic pathway for dendrobine in T. longibrachiatum MD33 and the elicitation mechanism of CH-AuNPs. A: Proposed Biosynthetic Route: The pathway integrates the mevalonate (MVA)-derived sesquiterpenoid scaffold with an ornithine-derived piperidine moiety. Key proposed steps include: (1) MVA pathway leading to Farnesyl Pyrophosphate (FPP); (2) Cyclization of FPP by a Sesquiterpene Synthase (TPS) to form the dendrobane core; (3) Biosynthesis of the piperidine moiety from ornithine via Ornithine Decarboxylase (ODC); (4) Conjugation of the two moieties by a putative Prenyltransferase (PTR); (5) Hydroxylation by a Cytochrome P450 monooxygenase (CYP71D1); and (6) O-Methylation by an O-Methyltransferase (COMT) to yield dendrobine. The final chemical structure of dendrobine is shown. B: CH-AuNP Elicitation Mechanism: Dose-dependent effect of precursor-loaded CH-AuNPs on the pathway: Low/Optimal doses (0.5–10.0 mg/L) enhance precursor delivery and upregulate pathway genes (HMGR, FPPS, TPS, ODC, CYP71D1, COMT) via a controlled oxidative stress response, leading to increased dendrobine synthesis. High doses (≥ 20.0 mg/L) cause excessive oxidative stress, disrupt metabolism, and inhibit production. AcCoA Acetyl-CoA, FPP Farnesyl Pyrophosphate

Proposed dendrobine biosynthesis pathway in trichoderma longibrachiatum MD33

The integration of transcriptomic (Table S4-6) and metabolomic analyses (Table S7-8) enabled the construction of a proposed nanoparticle-mediated biosynthetic pathway model for dendrobine in Trichoderma longibrachiatum MD33, integrating the chemical structure of dendrobine with key enzymatic steps, as depicted in Fig. 6A and B. Gold nanoparticles (AuNPs), synthesized via trisodium citrate reduction of chloroauric acid (HAuCl₄), formed spherical, monodisperse particles (~ 50 nm) with a localized surface plasmon resonance (LSPR) peak at 550 nm wavelength. The functionalization of CH-AuNPs with the precursors L-phenylalanine (L-Phe), L-tyrosine (L-Tyr), and tyramine was confirmed via UV-Vis spectrophotometry, with characteristic adsorption peaks at 210, 274, and 278 nm, respectively. These CH-AuNPs facilitated efficient precursor delivery into fungal cells via endocytosis or amino acid transporters (e.g., AAT1), followed by pH-triggered release into the acidic vesicles.

Upon cytosolic release, L-Phe was metabolized by phenylalanine ammonia-lyase (PAL, upregulated 2.5-fold at 10 mg/L CH-AuNPs) to form trans-cinnamic acid. This entry into the phenylpropanoid pathway was consistent with the observed transcriptional response. The metabolic fate of phenylpropanoid intermediates can involve degradation and conversion to acetyl-CoA via β-oxidation and other catabolic routes, ultimately supplying carbon to the tricarboxylic acid (TCA) cycle and the mevalonate (MVA) pathway. This indirect route, rather than the direct entry of phenylpyruvate into the TCA cycle, is a more accurate representation of fungal aromatic amino acid metabolism and aligns with the potential of these precursors to contribute to the acetyl-CoA pool required for sesquiterpenoid scaffold biosynthesis. Concurrently, L-Tyr was decarboxylated by tyrosine decarboxylase (TDC, enriched at 10 mg/L) to form tyramine, which modulates stress signaling and feeds into ornithine metabolism for alkaloid precursor synthesis. The MVA pathway, driven by the upregulation of HMGR (3-hydroxy-3-methylglutaryl-CoA reductase; +2.5-fold) and FPPS (farnesyl pyrophosphate synthase; +3.1-fold), converted acetyl-CoA into farnesyl pyrophosphate (FPP). Metabolomic data revealed the activation of the glyoxylate shunt under TCA cycle suppression, rerouting acetyl-CoA to sustain terpenoid flux (Fig. 4A-B; Table S4).

FPP underwent cyclization via a significantly upregulated sesquiterpene synthase (TPS, + 3.0-fold) to presumably form the dendrobane skeleton, which was subsequently hydroxylated by co-induced cytochrome P450 monooxygenases (CYP71D1, + 4.0-fold). Parallelly, ornithine-derived putrescine, synthesized via upregulated ornithine decarboxylase (ODC, + 1.8-fold) and supported by metabolomic perturbations in the ornithine/glutamate pools, formed the proposed piperidine alkaloid moiety. A putative prenyltransferase (PTR) is suggested to catalyze this conjugation, based on the co-induction of genes in related transferase families and the requirement for this biochemical step. This was followed by O-methylation via the strongly upregulated COMT (+ 3.3-fold) to produce the final dendrobine structure. The ABCG1 (+ 1.5-fold) transporter mediates dendrobine secretion, preventing feedback inhibition (Fig. 3D-F; Table S4).

Dose-dependent analyses revealed an optimal dendrobine yield at 10 mg/L CH-AuNPs, where transcriptional activation of HMGR, FPPS, TPS, CYP71D1, and COMT peaked, alongside redox homeostasis via D-galactarate (ROS scavenger) and glutathione (Table S6). Higher concentrations (≥ 20 mg/L) induced oxidative stress, suppressed arginine biosynthesis, and activated apoptosis, thereby diminishing precursor availability and dendrobine production (Fig. 6A and B). In conclusion, CH-AuNPs enhanced dendrobine biosynthesis by bypassing fungal uptake barriers, upregulating terpenoid-alkaloid pathway genes, and balancing oxidative stress. This systematic pathway provides a blueprint for precision metabolic engineering in fungal biotechnology, enabling the sustainable production of high-value sesquiterpenoid alkaloids.

Discussion

Characterization of CH-AuNPs using UV-vis, SEM, and zeta potential analysis

The synthesis of monodisperse, spherical chemical-synthesized gold nanoparticles (CH-AuNPs) with a localized surface plasmon resonance (LSPR) peak at 550 nm aligns with established protocols for colloidal CH-AuNPs [35]. The observed size (~ 50 nm), stability (zeta potential: -32 ± 2 mV) and low polydispersity (PDI: 0.15) are consistent with previous studies demonstrating that trisodium citrate reduction produces nanoparticles ideal for biological applications due to their biocompatibility and electrostatic repulsion [36, 37]. However, while the SEM-confirmed morphology and narrow size distribution validate synthesis reproducibility, discrepancies in LSPR peak positions compared to smaller CH-AuNPs (~ 520 nm for 20 nm particles) highlight the size-dependent optical properties critical for nanoparticle design [38, 39]. The stability of CH-AuNPs in fungal media, as evidenced by zeta potential, mirrors findings in bacterial systems but contrasts with reports of nanoparticle aggregation in plant cell cultures, underscoring the importance of host-specific compatibility [40].

Quantitative confirmation of precursor loading and its implications for yield optimization

The successful loading of L-phenylalanine, L-tyrosine, and tyramine onto the CH-AuNPs was confirmed using a two-tiered analytical approach. Beyond initial UV-Vis characterization, the use of HPLC to quantify the acid-liberated precursors provided definitive confirmation of co-loading and yielded precise loading efficiencies (82–85%) and, importantly, the molar ratio (L-Phe: L-Tyr: tyramine ≈ 1.0:0.8:1.1). This chromatographic validation addresses a key methodological consideration for multi-precursor functionalization studies, where UV spectroscopy alone may be insufficient to distinguish or quantify closely related molecules [41, 42]. This defined stoichiometry is likely pivotal for the observed 63.7% enhancement in dendrobine yield at the optimal dose. The co-delivery of all three precursors at this specific ratio, rather than a single precursor or non-specific loading, may optimally address the requirements of the proposed hybrid pathway [43, 44]. L-Phe and L-Tyr supply carbon skeletons and entry points into aromatic metabolism, whereas tyramine contributes to nitrogen-containing moiety synthesis and stress signaling [45, 46].

The metabolic interconvertibility of L-Phe, L-Tyr, and tyramine raises a pertinent question regarding which of these serves as the actual precursor for dendrobine biosynthesis. While the current multi-omics data cannot definitively pinpoint a single “true” precursor, the observed co-upregulation of genes related to aromatic amino acid metabolism (e.g., PAL, TDC) and the specific depletion of intracellular glutamate/ornithine pools suggest that the fungus actively processes these supplied compounds for alkaloid formation [47, 48]. The use of all three precursors in combination was a deliberate strategy to: (1) overcome potential substrate specificity or uptake limitations, as different transporters or enzymes may prefer one substrate over another; (2) ensure metabolic flexibility, as the fungus can utilize the most readily available or energetically favorable route under nanoparticle-induced stress; and (3) maximize precursor flux by supplying multiple entry points into the proposed hybrid pathway. Future studies employing isotope-labeled precursors (e.g., ¹³C-L-Phe vs. ¹³C-L-Tyr) in the presence of CH-AuNPs will be essential to trace the exact metabolic flow and identify the primary precursor(s) incorporated into the dendrobine skeleton. Nevertheless, the significant yield enhancement achieved with the triple-precursor-loaded CH-AuNPs supports the utility of this combinatorial approach in bypassing metabolic bottlenecks and stimulating the entire biosynthetic network [49, 50].

Furthermore, this quantitative loading data provides a direct basis for future yield-optimization studies. The identified molar ratio (1.0:0.8:1.1) serves as a starting point; subsequent studies could systematically vary the loading proportions of these precursors on CH-AuNPs to explore whether alternative stoichiometries further enhance dendrobine production or improve the metabolic balance. Such optimization would be a logical extension of the current work, leveraging established synthesis and quantification protocols without requiring new techniques [51]. Additionally, while the present study focused on precursor-loaded CH-AuNPs, future experiments should include parallel control groups (nanoparticle-only and precursor-only) to fully decouple their individual contributions and elucidate the synergies.

Concentration-dependent effects of CH-AuNPs on dendrobine production

The dose-dependent enhancement of dendrobine yield, peaking at 10 mg/L precursor-loaded CH-AuNPs (63.7% increase), must be interpreted in the context of the nanoparticle’s dual role, which parallels findings in Taxus cell cultures, where gold nanoparticles amplified paclitaxel production at optimal doses through elicitation [52]. The decline at 20 mg/L aligns with toxicity thresholds observed in other fungi [40], where high CH-AuNPs concentrations induced oxidative stress and apoptosis [53]. This non-linear response mirrors the “hormesis” effect-low doses stimulate metabolism, while high doses overwhelm detoxification systems-a phenomenon well-documented in nanoparticle-fungal interactions [54]. However, the absence of dendrobine suppression at intermediate doses (5 mg/L) contrasts with microbial systems, where even moderate precursor loaded nanoparticle levels favor upregulation of secondary metabolism, highlighting fungal resilience [55].

Notably, the observed 63.7% increase in dendrobine content with 10.0 mg/L CH-AuNPs is mechanistically linked to this combined effect. At this optimal concentration, CH-AuNPs generated moderate levels of reactive oxygen species (ROS), which acted as signaling molecules to activate redox-sensitive transcription factors (e.g., Yap1 and Atf1) and MAPK cascades. These regulators, in turn, upregulate genes involved in terpenoid and alkaloid biosynthesis (HMGR, TPS, CYP71D1), while also enhancing the production of antioxidant metabolites, such as glutathione and D-galactarate, to maintain cellular redox homeostasis. This balanced stress response redirects carbon flux from primary growth pathways to secondary metabolite synthesis, thereby boosting dendrobine production without causing significant growth inhibition [56–58].

The superior yield enhancement (63.7%) observed with the optimal CH-AuNP loading dose (10.0 mg/L) underscores its advantages over traditional abiotic elicitors or nanoparticle-only approaches. Unlike broad-spectrum agents such as jasmonic acid or heavy metal ions (e.g., Co²⁺, Cd²⁺), which non-specifically induce stress and often impair growth and primary metabolism, CH-AuNPs function as precision tools [12]. Multi-omics data revealed that CH-AuNPs at 10.0 mg/L specifically upregulated terpenoid-alkaloid genes (HMGR, TPS, CYP71D1) while concurrently delivering precursors (L-Phe, L-Tyr), a dual mechanism absent in conventional elicitors. In contrast, studies using cobalt chloride in the same fungal system reported significant growth retardation, a less pronounced increase in dendrobine yield, and a more generalized oxidative stress response [59, 60]. Furthermore, heavy metals often cause persistent toxicity and bioaccumulation, posing environmental and safety concerns for humans and animals. CH-AuNPs, synthesized using biocompatible chemicals, offer a more sustainable and controllable alternative. Their well-defined dose-response curves and clear toxicity thresholds (≥ 20.0 mg/L) enable precise metabolic tuning without the irreversible damage characteristic of metal ion toxicity. This targeted approach, which combines elicitation with precursor supply, positions CH-AuNPs as a next-generation strategy for the sustainable bioproduction of high-value alkaloids [61, 62].

Transcriptional and metabolic reprogramming elucidates the hybrid pathway

The dose-dependent transcriptional and metabolomic shifts provide compelling, multi-layered evidence for nanoparticle-mediated activation of the proposed hybrid pathway. The significant upregulation of HMGR and FPPS at the optimal elicitor dose (10.0 mg/L) was accompanied by metabolomic data suggesting a rerouting of acetyl-CoA flux, directly addressing a classic bottleneck in microbial terpenoid production—insufficient flux through the MVA pathway. This mirrors successful strategies in engineered yeasts and plants, where boosting these early steps is crucial for sesquiterpenoid yield [63, 64]. The concurrent induction of a specific TPS gene and the activation of the ornithine decarboxylase (ODC) pathway—the latter corroborated by measurable depletion of intracellular glutamic acid and perturbations in ornithine-related metabolites—strongly supports the model of a convergent biosynthesis supplying both core scaffolds [65]. Simultaneously, the activation of the ornithine decarboxylase (ODC) pathway indicates a microbial strategy for generating the piperidine moiety. Metabolomic data showing perturbations in ornithine/glutamate metabolism corroborate this transcriptional signal, indicating a reallocation of nitrogen metabolism toward alkaloid precursor synthesis. The striking co-upregulation of CYP71D1 and COMT further highlights these modification enzymes as critical, coordinated nodes in the proposed late-stage assembly. Thus, the integrated omics data demonstrate that CH-AuNPs elicit a precise and coordinated response across distinct but complementary metabolic branches, forming a coherent evidence-based model for dendrobine biosynthesis [66, 67]. This focused reprogramming, diverting carbon from the TCA cycle via the glyoxylate shunt toward acetyl-CoA and redirecting nitrogen flux via ornithine, effectively creates a metabolic “funnel” that enhances the supply of both core precursors required for dendrobine assembly [68, 69]. This discrepancy may be attributed to differences in antioxidant capacity and nitrogen metabolism between fungi and plants, highlighting the need for host-specific metabolic engineering strategies [70]. The metabolomic data provided direct support for the transcriptional activation of the ornithine/putrescine pathway. At the optimal CH-AuNP dose, a measurable depletion of intracellular glutamate and ornithine was observed alongside an increase in putrescine and N-acetylputrescine levels. This pattern of precursor consumption and product accumulation is consistent with an active flux through the ODC-catalyzed pathway and strongly supports the model of nitrogen reallocation for alkaloid precursor synthesis. These targeted shifts in nitrogen metabolism, while not necessarily among the largest global fold-changes, form a coherent biochemical narrative that directly substantiates the proposed hybrid biosynthetic route.

Proposed dendrobine biosynthesis pathway

The hybrid terpenoid-alkaloid pathway proposed here is supported by the following correlative evidence from multi-omics analysis: (1) coordinated upregulation of early MVA pathway genes (HMGR, FPPS) and a specific sesquiterpene synthase (TPS); (2) concurrent induction of the ornithine/putrescine pathway (ODC, DAO) alongside metabolomic perturbations in relevant nitrogen metabolites; (3) strong co-expression of late-stage modification enzymes (CYP71D1, COMT); and (4) a redirection of central carbon metabolism observed in the metabolome, aligning with the precursor demands of the proposed route. Although this multi-omics approach provides strong correlative evidence, functional validation is required for definitive proof of the results. The integrated data presented here offer a robust and testable model that shares logical parallels with other hybrid metabolite pathways [71–73]. The reliance on ABC transporters for dendrobine secretion aligns with findings in Bromus japonicus, though fungal-specific transporters like ABCG1 may require CRISPR-based optimization to enhance yields [74]. While the multi-omics approach applied provides strong correlative evidence for the involvement of HMGR, TPS, CYP71D1, and COMT in dendrobine biosynthesis, it is acknowledged that the functional annotations for the putative CYP71D1 and COMT genes are primarily based on homology to sequences from plants or other fungi and their significant co-expression patterns under elicitation. Such experiments were beyond the scope of the current study, which focused on establishing a nanoparticle-mediated elicitation platform and initial pathway elucidation. The lack of direct enzymatic validation is a recognized limitation of this study. Future studies should prioritize these functional approaches to confirm the proposed hybrid terpenoid-alkaloid pathway [2, 75–77].

Cytotoxic and environmental safety considerations of CH-AuNPs

The dose-dependent effects observed in this study highlight the dual nature of CH-AuNPs as metabolic elicitors and potential cytotoxic agents. While the optimal concentration (10.0 mg/L) significantly enhanced dendrobine yield, higher doses (≥ 20.0 mg/L) induced metabolic suppression, oxidative stress, and activation of apoptotic pathways (Figs. 3 and 5; Supplementary Table S4, 6, and 8). This toxicity threshold aligns with the established “hormesis” principle in nanotoxicology, where low doses stimulate biological responses, while excessive concentrations overwhelm cellular defense mechanisms [78]. The transition from beneficial stress signaling to cytotoxic damage is governed by cellular redox balance. At optimal doses, ROS levels are maintained within the signaling range by antioxidant systems (e.g., glutathione and superoxide dismutase). At supra-optimal doses, the antioxidant capacity is exceeded, leading to oxidative damage to lipids, proteins, and DNA, ultimately triggering programmed cell death [79, 80].

The specific cytotoxic mechanisms at high doses likely involve ROS overproduction, disruption of the TCA cycle and arginine biosynthesis, and subsequent cellular damage, as evidenced by the metabolomic and transcriptomic data [81]. However, the potential for nanoparticle accumulation and long-term ecological impact requires consideration for large-scale applications [82]. The observed fungal growth inhibition and metabolic shutdown at high concentrations suggest a self-limiting effect in contaminated environments; however, the persistence of AuNPs necessitates further investigation. Future studies should prioritize life cycle assessments and the development of biodegradable nanocarriers or recovery systems to mitigate environmental release. For biomedical or controlled bioprocessing applications, the established nontoxic optimal dose (10.0 mg/L) provides a safe window, underscoring the importance of precise dosage in nanobiotechnology to harness the benefits while minimizing risks [83–85].

Scaling up this nanoparticle-mediated system to industrial bioreactors presents several challenges, including maintaining nanoparticle stability under high-shear conditions, ensuring uniform dispersion in large volumes, and managing potential nanoparticle aggregation over extended cultivation periods. Additionally, the cost-effective synthesis and functionalization of CH-AuNPs must be optimized to be economically viable. To address these issues, future strategies could involve the use of immobilized nanoparticle systems or encapsulation within biocompatible matrices to enhance their stability and reusability [86, 87]. Furthermore, integrating real-time monitoring and feedback control systems within bioreactors could allow for the dynamic adjustment of nanoparticle dosage, optimizing dendrobine yield while minimizing oxidative stress. Such process optimizations would be directly informed by the dose-response data and oxidative stress thresholds established in this study. While this study demonstrates the potential of CH-AuNPs to enhance dendrobine biosynthesis in T. longibrachiatum MD33, future studies should focus on scaling the process to bioreactor systems to evaluate their long-term stability, reproducibility, and industrial feasibility. Advances in continuous fermentation and downstream processing are essential to translate this proof-of-concept into a sustainable and industrially feasible platform for producing dendrobine [7, 88].

Economic feasibility and application potential of the CH-AuNP strategy

Translating this nanoparticle-mediated strategy from proof-of-concept to an industrially viable process necessitates a critical evaluation of its economic feasibility against the high-value applications of dendrobine in the pharmaceutical industry. The primary cost driver is the synthesis and functionalization of the CH-AuNPs. Although chemical synthesis using trisodium citrate is relatively inexpensive, scaling up while maintaining monodispersity and precise functionalization requires further optimization [8, 89]. However, the significant (63.7%) yield enhancement at low nanoparticle concentrations (10 mg/L) suggests a favorable input-output ratio, where a modest investment in nano-elicitors could generate substantially more high value products. Furthermore, the potential for nanoparticle recovery, reuse, or the use of immobilized nanoparticle systems within bioreactors could drastically reduce long-term material costs. The integration of this elicitation strategy with existing fermentation infrastructure presents a path for process intensification, potentially reducing overall production time and resource consumption compared to plant extraction or less efficient microbial cultures [13, 90].

The economic rationale is strongly supported by the potential application of dendrobine. As a sesquiterpenoid alkaloid with demonstrated neuroprotective and anticancer activities, dendrobine holds significant value in pharmaceutical markets for neurodegenerative diseases (e.g., Alzheimer’s and Parkinson’s diseases) and in oncology. Its anti-inflammatory properties suggest potential applications in nutraceuticals and cosmetics. Beyond their effects on human health, dendrobine and related alkaloids have demonstrated insecticidal and antifungal activities, indicating their potential applications in green agrochemicals [5, 27, 91]. The current reliance on endangered Dendrobium orchids renders the supply chain unsustainable, ecologically damaging, and expensive. Therefore, establishing a reliable, scalable, and sustainable microbial production platform using endophytic fungi, such as T. longibrachiatum MD33, enhanced by precision nano-elicitation, addresses a critical supply bottleneck. This strategy aligns with the principles of bioeconomy, aiming to replace inefficient plant extraction with controlled fermentation, ultimately enabling the broader research, development, and commercialization of dendrobine-based products. Future techno-economic analyses based on scaled-up bioreactor data will be crucial for precisely quantifying the cost-benefit advantages of this strategy [10, 28, 71, 92].

Conclusion

While multi-omics data strongly support the involvement of key enzymes (HMGR, TPS, CYP71D1, and COMT) in the proposed hybrid terpenoid-alkaloid pathway, it is acknowledged that direct experimental validation through gene knockout, in vitro enzyme inhibition or activity assays, or isotope-labelling studies would further solidify these mechanistic claims. These approaches represent critical directions for future studies. Integrating such functional validation with the nanoparticle-mediated strategy described here could provide a comprehensive framework for pathway elucidation and the scalable production of dendrobine and related alkaloids. Importantly, this study provides a clear mechanistic link between CH-AuNP-induced oxidative stress and enhanced dendrobine biosynthesis, establishing a strongly supported multi-omics-derived model for its hybrid biosynthetic pathway in T. longibrachiatum MD33. At an optimal concentration (10.0 mg/L), CH-AuNPs generated sub-toxic levels of ROS that acted as a signaling cue to upregulate terpenoid and alkaloid pathway genes while simultaneously delivering key precursors to overcome metabolic bottlenecks. This dual role of CH-AuNPs as both elicitors and carriers highlights their potential as precision tools for metabolic engineering in non-model fungi. By potentially reducing the reliance on expensive and unsustainable plant extraction, this strategy could improve the economic feasibility of dendrobine production for its promising applications in neuroprotection and oncology, respectively. Unlike conventional elicitors, CH-AuNPs act as dual-function agents, delivering precursors and eliciting stress-responsive pathways, a strategy that could revolutionize the production of high-value plant-derived alkaloids in microbial systems. This approach circumvents unsustainable plant extraction and aligns with the principles of green chemistry by minimizing solvent and waste usage. This approach circumvents unsustainable plant extraction and aligns with the principles of green chemistry and cost-effective biomanufacturing by minimizing solvent use, waste generation, and environmental impact.

The dose-response relationships and pathway insights established in this study lay a solid foundation for future bioprocess optimization and techno-economic assessment. For instance, the identified optimal CH-AuNP concentration (10.0 mg/L) and key regulatory genes (e.g., HMGR and CYP71D1) offer direct targets for further metabolic engineering or fermentation control aimed at maximizing yield and economic viability. Looking ahead, the multi-omics dataset generated here provides a valuable resource for guiding targeted genetic interventions, such as CRISPR-based editing of pathway genes, which could be informed by the transcriptional profiles observed under optimal elicitation. Similarly, understanding oxidative stress thresholds could inform the design of fed-batch or continuous bioreactor processes that maintain ROS levels within a stimulatory range. These potential applications are grounded in the empirical data presented in this study and represent logical extensions of the current work. By merging nanotechnology with synthetic biology, this study lays the foundation for a new era of sustainable biomanufacturing, in which fungi serve as eco-friendly biofactories for producing pharmaceuticals, nutraceuticals, and agrochemicals.

Furthermore, although the combinatorial provision of L-Phe, L-Tyr, and tyramine proved effective in enhancing dendrobine yield, future studies should employ isotopic tracing to delineate the precise metabolic flow and identify the primary precursor(s) funneled into the alkaloid moiety under nanoparticle elicitation. While multi-omics data strongly support their involvement in the proposed hybrid terpenoid-alkaloid pathway, direct experimental validation through gene knockout, knockdown, heterologous expression, or in vitro enzyme assays would further solidify the mechanistic claims. Future studies should prioritize the construction of targeted gene-edited strains (e.g., via CRISPR-Cas9) in T. longibrachiatum MD33 or heterologous expression in model systems combined with in vitro enzyme activity assays to confirm the catalytic functions of these enzymes. Such approaches would not only validate the roles of these genes in dendrobine biosynthesis but also enable precise metabolic engineering for yield optimization. Integrating the functional validation suggested here with the nanoparticle-mediated strategy could provide a comprehensive framework for pathway confirmation and scalable production.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Author contributionsConceptualization: SS, AJ, JC, and QG; data analysis: SS and AJ; original draft preparation: SS and AJ; editing: JC, QG; visualization: SS, and AJ; and supervision: QG. All authors have read and agreed to the published version of this manuscript.

Funding

Guizhou Science and Technology Corporation Platform Talents Fund (Grant No.: [2017]5733-001 and CK-1130-002), High Level Specialists Project Fund of Zunyi Medical University (1092/2000006)F-1092, the National Natural Science Foundation of China (82373981).

Data availability

All data generated or analysed during this study are included in this article and/or its supplementary.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.