DETECTION OF BIOSYNTHETIC GENE CLUSTERS FROM METAGENOME OF NATURAL HONEY COLLECTED IN VIETNAM

This study utilizes metagenomic analysis of *Apis cerana* honey from Vietnam's Northwest region to uncover a highly novel biosynthetic potential, identifying 366 gene clusters with over 83% lacking known database matches and highlighting a specific *Atlantibacter hermannii*-derived azole-containing RiPP cluster as a promising target for new antimicrobial discovery.

The Gist

Imagine you have a jar of natural honey. To most people, it's just a sweet treat or a soothing remedy for a sore throat. But to the scientists in this study, that jar is actually a hidden treasure chest filled with microscopic factories.

Here is the story of what they found, explained simply:

1. The Honey is a Busy City, Not Just a Sweet Jar

Think of the honey as a bustling city. Inside, there are millions of tiny bacteria living together. For years, we knew these bacteria helped make honey taste good and fight off bad germs, but we didn't know how they did it.

Usually, to study these bacteria, scientists try to grow them in a lab dish (like planting seeds in a garden). But here's the problem: 99% of these bacteria are "shy." They refuse to grow in a lab. It's like trying to study a wild animal by only looking at the ones that are willing to come into a zoo. You miss almost everything.

2. The "Digital X-Ray" (Metagenomics)

Instead of trying to catch the bacteria, the scientists used a high-tech "digital X-ray" called metagenomics.

• The Analogy: Imagine you have a giant pile of shredded documents from a library, but you don't know which book they came from. Instead of trying to glue them back together one by one, you use a super-computer to scan the text and figure out the stories hidden inside.
• The Result: They took DNA samples from honey collected in the mountains of Vietnam and scanned the genetic code of every bacterium in the jar, even the ones that can't be grown in a lab.

3. Finding the "Instruction Manuals" (BGCs)

Inside every bacterium, there are sets of instructions called Biosynthetic Gene Clusters (BGCs).

• The Analogy: Think of these as recipe books or instruction manuals for building special chemical weapons. These bacteria use these recipes to build "secondary metabolites"—which are fancy words for chemical compounds that kill other bacteria or fungi.
• The Discovery: The scientists found 366 different recipe books in the honey. That's a huge library of potential new medicines!

4. The "New Recipes" (Novelty)

Here is the most exciting part. The scientists compared these 366 recipes against the world's biggest library of known recipes (a database called MIBiG).

• The Result: Over 83% of the recipes were brand new. They didn't match anything humans had ever seen before.
• The Metaphor: It's like walking into a kitchen and finding 304 cookbooks that no one has ever written down before. They are completely new ways to cook up chemicals that could fight disease.

5. The "Golden Ticket" (The Specific Discovery)

Among all these new recipes, the scientists found one specific "Golden Ticket."

• The Character: They found a recipe hidden inside a bacterium called Atlantibacter hermannii.
• The Weapon: This recipe builds a molecule called an azole-containing RiPP.
• The Prediction: The computer predicted this molecule has a 74.5% chance of being a powerful antibiotic (a germ killer).
• Why it matters: We are currently losing the war against "superbugs" (bacteria that antibiotics can't kill). This new recipe from the honey might be the new weapon we need to win that war.

6. Why This Matters for You

• The Problem: Bacteria are becoming resistant to our current medicines. We are running out of ways to treat infections.
• The Solution: Nature is full of undiscovered medicines. This study shows that honey is a goldmine we haven't fully explored yet.
• The Future: Now that the scientists have found the "instruction manual" (the gene cluster), they can try to build the chemical in a lab to test if it really works. If it does, it could become a new medicine to save lives.

In a Nutshell

This paper is like a detective story where the detectives (scientists) went into a jar of honey, used a high-tech scanner to read the secret notes of invisible bacteria, and found hundreds of new blueprints for life-saving medicines. They found one blueprint in particular that looks like it could be a super-weapon against dangerous germs, proving that sometimes the best place to look for the future of medicine is right in your kitchen pantry.

Technical Summary

1. Problem Statement

Natural honey possesses well-documented antimicrobial properties, often attributed to bioactive secondary metabolites produced by its associated microbial communities. However, the biosynthetic capacity of these honey-associated microbiomes remains poorly characterized. Traditional culture-dependent methods fail to access the vast majority of uncultivable microorganisms, limiting the discovery of novel antimicrobial agents. With the rise of antimicrobial resistance (AMR), there is an urgent need to explore underutilized environmental niches, such as the honey microbiome, to identify new Biosynthetic Gene Clusters (BGCs) encoding novel secondary metabolites.

2. Methodology

The study employed a comprehensive metagenomic approach to analyze bacterial communities in natural honey collected from the Northwest mountainous region of Vietnam (specifically from Apis cerana hives).

• Data Source: Raw metagenomic reads from nine honey samples were retrieved from the NCBI Sequence Read Archive (SRA).
• Preprocessing & Assembly:
  • Taxonomic classification was performed using Kraken2.
  • Bacterial and unclassified reads were co-assembled into contigs using MEGAHIT (v1.2.9), retaining contigs $\ge$ 500 bp.
• Genome Reconstruction (MAGs):
  • Reads were mapped back to contigs using BBmap to generate coverage data.
  • Three binning tools (metaBAT2, SemiBin2, and Vamb) were used to group contigs into genome bins.
  • Bins were refined using DAS Tool to create an optimized set.
  • Quality Control: CheckM2 assessed bin quality. Only bins with completeness $\ge$ 50% and contamination $\le$ 10% were retained as Metagenome-Assembled Genomes (MAGs).
  • Taxonomic assignment was performed using GTDB-Tk and the Bin Annotation Tool (BAT) against the GTDB database (R220).
• BGC Detection & Analysis:
  • antiSMASH (v8.0.2) was used to detect BGCs in MAGs and unbinned contigs ($\ge$ 5,000 bp). Hybrid clusters were deconstructed and counted by individual component classes.
  • BiG-SCAPE (v2.0) constructed a similarity network against the MIBiG 4.0 database to group BGCs into Gene Cluster Families (GCFs).
  • NPBdetect predicted biological activities (antibacterial, antifungal, etc.).
  • NeuRiPP and RiPPMiner were used to screen for precursor peptides of Ribosomally synthesized and Post-translationally modified Peptides (RiPPs).
  • BLASTp was used for functional annotation against RefSeq and NCBI nr databases.

3. Key Results

A. Recovery of MAGs

• A total of 42 high-quality MAGs were recovered from the nine honey samples.
• These MAGs represented diverse taxonomic groups, including Atlantibacter, Klebsiella, Enterobacter, Apilactobacillus, Lactococcus, and Sphingomonas.
• Notably, Atlantibacter hermannii (vamb_1) was recovered with 100% completeness.

B. BGC Diversity and Novelty

• Total Count: 366 BGCs were identified across 38 compound classes.
• Dominant Classes: The most prevalent classes were Terpenes (16.7%), Terpene-precursors (15.8%), RiPP-like compounds, and Nonribosomal Peptide Synthetases (NRPS).
• Novelty: A striking 83% (304/366) of the identified BGCs lacked close matches in the MIBiG reference database, indicating a high degree of biosynthetic novelty.
• Gene Cluster Families (GCFs): The analysis identified 31 GCFs. Only 15 contained BGCs with MIBiG references. 276 BGCs were singletons (unique to the dataset).

C. Prioritized Candidate: GCF07 (Azole-containing RiPP)

The study focused on GCF07, an azole-containing RiPP family, due to its predicted bioactivity.

• Source: The cluster was found in a MAG assigned to Atlantibacter hermannii (vamb_1).
• Prediction: NPBdetect predicted a 74.5% probability of antibacterial activity for this cluster, the highest among the analyzed candidates.
• Precursor Peptide: The cluster contains a centrally located gene encoding a short peptide (56 amino acids, "Peptide V1").
  • Sequence: MSALRKCRMVEGQPGETAYLTANMSVLLRQADRFCCVTAAAFYRRRVLAIADRVTE
  • Homology: >98% similarity to hypothetical proteins in Klebsiella spp., though functionally uncharacterized.
  • Validation: Both NeuRiPP and RiPPMiner confirmed Peptide V1 as a RiPP precursor.
• Comparison: Other clusters in GCF07 (from Klebsiella pneumoniae, Enterobacter soli, and unbinned contigs) showed lower predicted antibacterial activity or appeared incomplete (containing only 3–5 genes).

4. Significance and Contributions

• Untapped Reservoir: The study establishes the Apis cerana honey microbiome from Vietnam as a rich, largely untapped reservoir for novel secondary metabolites, with over 80% of BGCs being novel to current databases.
• Methodological Validation: It demonstrates the efficacy of metagenomics and MAG recovery in bypassing the "unculturable" barrier to access the metabolic potential of complex microbial communities.
• Target for Drug Discovery: The identification of the Atlantibacter hermannii azole-containing RiPP BGC provides a concrete, high-priority target for heterologous expression and experimental validation. Azole-containing peptides are known for potent antibacterial and antifungal properties.
• Ecological Insight: The findings suggest that honey-associated bacteria contribute significantly to the antimicrobial properties of honey, potentially offering new avenues for combating multidrug-resistant pathogens.

5. Limitations and Future Directions

• In Silico Nature: All bioactivity predictions are computational; experimental validation (heterologous expression, purification, and bioassays) is required to confirm the chemical structure and efficacy of the compounds.
• Assembly Completeness: Some identified BGCs (e.g., in Enterobacter soli) were incomplete, likely due to limitations in short-read assembly. The authors suggest that long-read sequencing could improve BGC recovery.
• Silencing: Even if expressed, some BGCs may remain silent under laboratory conditions, necessitating specific activation strategies.

Conclusion: This research successfully bridges metagenomic mining with applied biotechnology, highlighting the Vietnamese honey microbiome as a promising source for the next generation of antimicrobial agents.