In situ metabolomics for agricultural and microbiological applications
​Imaging mass spectrometry (IMS) has emerged as a transformative “in situ metabolomics” technology that enables direct visualization of specialized metabolites in intact biological systems with spatial and temporal resolution. Unlike conventional metabolomics, which typically extracts and averages chemical signals across a sample, IMS can reveal where metabolites are produced, how they spread, and when they accumulate during biological interactions—insights that are essential for understanding natural products in their native ecological and physiological context. Our laboratory develops and applies next-generation IMS workflows to illuminate the chemical mechanisms underlying microbial competition, plant–microbe interactions, and pathogen invasion.

​A major technical focus of our work is overcoming long-standing bottlenecks in small-molecule imaging. While matrix-assisted laser desorption/ionization (MALDI) IMS has been widely adopted, it can suffer from interference and reduced performance when analyzing low-mass metabolites.

To address these challenges, we have developed surface-assisted laser desorption/ionization (SALDI) IMS platforms that improve small-molecule detection and expand accessible chemical space, allowing more faithful mapping of microbial and plant metabolites in situ. We have also contributed to the broader methodological ecosystem that connects in situ chemical imaging with structure-centric annotation, including community infrastructure for tandem MS data sharing and molecular networking.
Picture

​Our IMS program is tightly linked to agricultural and microbiological applications, where spatial chemistry is often the missing layer needed to translate “candidate” discoveries into practical solutions. In crop protection, many promising biocontrol strains fail during field deployment because their functional molecules are not produced consistently, or because their metabolic behavior in complex environments is poorly understood. Our approach uses IMS to monitor biocontrol metabolite production and turnover directly at interaction zones, such as microbe–microbe boundaries, pathogen confrontation fronts, or plant-associated niches. This enables function-first evaluation of antagonistic microbes by verifying whether protective metabolites appear at the right place and time to suppress infection.

By combining technology development with application-driven questions, our long-term goal is to establish IMS as a routine, decision-enabling platform for agriculture and microbiology—one that connects molecules to mechanisms, and mechanisms to deployable strategies for crop disease control and microbial innovation.
New biocontrol agents for agricultural diseases control
​Banana Fusarium wilt, caused by Fusarium oxysporum f. sp. cubense tropical race 4 (Foc TR4), is one of the most destructive soil-borne diseases threatening global banana production. Once a field becomes infested, eradication is extremely difficult, and conventional control practices offer limited and often unsustainable protection. In our laboratory, we aim to develop next-generation, environmentally responsible strategies to control Fusarium wilt by leveraging the natural protective capacity of the rhizosphere microbiome and its specialized metabolites.

A central concept guiding our work is that some soils exhibit disease-suppressive properties, where resident microbial communities can naturally restrict pathogen establishment and reduce disease severity. Our research focuses on deciphering how beneficial microbes from disease-suppressive soils protect banana plants, with particular emphasis on microbial metabolic communication as an underexplored driver of biocontrol efficacy. Using an integrated multi-omics framework, we combine metagenomics, functional isolation, and untargeted metabolomics to identify “hub microbes” and the bioactive metabolites or biosynthetic gene clusters that underlie antagonism against Foc TR4 and promote host resilience. This approach allows us to move beyond taxonomic descriptions of the microbiome toward a function-first view of what makes a rhizosphere community protective.
圖片

Our program is organized around three synergistic research directions. First, we systematically discover and characterize hub microbes and their metabolites that suppress Foc TR4, integrating culture-dependent screening with sequencing-guided prioritization to link candidate strains to functional pathways and chemical signatures. Second, we investigate how beneficial microbes and their metabolites reprogram banana root immunity, using transcriptomics to map immune-related responses and pinpoint defense-associated pathways influenced by microbial inoculation. We couple these readouts with metabolomic profiling of banana roots to identify biochemical markers of enhanced resistance and defense priming. Third, we translate these mechanistic insights into practical solutions by designing and evaluating synthetic microbial consortia (SynComs) that mimic the protective functions of disease-suppressive soils. These consortia are tested through controlled pathogenicity platforms (including high-throughput banana–Foc TR4 assays) and greenhouse pot trials that monitor both disease suppression and plant growth promotion under field-relevant conditions.

By integrating mechanistic discovery with consortium engineering, our long-term goal is to establish scalable microbial solutions that reduce reliance on chemical pesticides while improving crop durability and sustainability. Ultimately, our work seeks to provide a blueprint for microbiome-informed management of Fusarium wilt and other soil-borne diseases across agriculture.

​​Our program is organized around three synergistic research directions. First, we systematically discover and characterize hub microbes and their metabolites that suppress Foc TR4, integrating culture-dependent screening with sequencing-guided prioritization to link candidate strains to functional pathways and chemical signatures. Second, we investigate how beneficial microbes and their metabolites reprogram banana root immunity, using transcriptomics to map immune-related responses and pinpoint defense-associated pathways influenced by microbial inoculation. We couple these readouts with metabolomic profiling of banana roots to identify biochemical markers of enhanced resistance and defense priming.

​Third, we translate these mechanistic insights into practical solutions by designing and evaluating synthetic microbial consortia (SynComs) that mimic the protective functions of disease-suppressive soils. These consortia are tested through controlled pathogenicity platforms (including high-throughput banana–Foc TR4 assays) and greenhouse pot trials that monitor both disease suppression and plant growth promotion under field-relevant conditions.

By integrating mechanistic discovery with consortium engineering, our long-term goal is to establish scalable microbial solutions that reduce reliance on chemical pesticides while improving crop durability and sustainability. Ultimately, our work seeks to provide a blueprint for microbiome-informed management of Fusarium wilt and other soil-borne diseases across agriculture.
Engineering biosynthetic enzymes for natural products synthesis
Bacterial conjugated polyynes are a rare but exceptionally potent class of natural products, distinguished by their highly unsaturated carbon frameworks with multiple conjugated triple bonds. Although only a small number of bacterial polyynes have been reported to date, they display broad bioactivities, including antifungal, antibacterial, and pesticidal effects, and often function as chemical “weapons” or signaling molecules that shape microbial interactions in complex ecosystems.

​Our laboratory investigates bacterial polyynes as a model system to understand how specialized biosynthetic enzymes generate structurally unusual metabolites, and how these metabolites can be engineered into sustainable platforms for biomedical and agricultural applications. A major focus of our work is decoding and reprogramming polyyne biosynthetic gene clusters (BGCs). Through phylogeny-guided genome mining and functional genomics, we previously identified a core polyyne BGC in Massilia sp. YMA4 (the mas cluster), which converts C16 fatty acid precursors into a family of polyynes, including massilins and collimonins. These pathways encode a compact enzymatic “assembly line” composed of an acyl-ACP ligase, multiple desaturase/acetylenase enzymes, an acyl carrier protein, a hydrolase, and accessory redox components. Importantly, tailoring enzymes such as the dioxygenase MasB can introduce key oxidative modifications that diversify the final chemical structures and frequently enhance bioactivity.
Image description
Beyond pathway discovery, we aim to engineer biosynthetic enzymes to expand the chemical space of polyynes. Polyyne formation relies on membrane-associated desaturase/acetylenase enzymes that are notoriously difficult to characterize and optimize, yet they govern critical features such as the number and position of triple bonds. We address this challenge through a combined strategy of (i) heterologous expression for pathway reconstruction, (ii) targeted enzyme swapping and modular co-expression to tune product profiles, and (iii) structure-guided functional validation of newly generated derivatives. Using engineered E. coli expression systems, we have reconstructed core polyyne production and demonstrated that introducing specific tailoring genes can convert a primary precursor into downstream, oxidized derivatives, providing a direct blueprint for pathway upgrading and diversification.

To accelerate discovery and engineering cycles, our current program integrates ancestral sequence reconstruction (ASR) and cell-free synthesis (CFS) as next-generation platforms for enzyme innovation. ASR enables us to “resurrect” ancestral versions of key tailoring enzymes (particularly dioxygenases related to MasB) that may exhibit improved stability, altered regioselectivity, or expanded substrate scope. In parallel, CFS bypasses cellular constraints and enables rapid, high-throughput testing of engineered enzyme variants and pathway combinations, with product readouts quantified by HPLC and mass spectrometry. Finally, for scalable biomanufacturing, we develop synthetic biology workflows for efficient multi-gene assembly and robust heterologous production, creating a practical route from enzyme discovery to translational supply.

By connecting genome mining, enzyme mechanism, and synthetic biology, our long-term goal is to establish bacterial polyynes as a versatile platform for designing new bioactive molecules and producing them sustainably at scale—turning rare microbial chemistry into deployable solutions for agriculture and human health.