Viral metagenomics of synanthropic urban bats: a surveillance strategy for uncovering potentially zoonotic viruses

This study presents a scalable, cost-effective metagenomic surveillance framework leveraging Brazil's existing rabies monitoring program to detect diverse zoonotic viruses in urban bats, including a novel filovirus, thereby validating a practical One Health model for epidemic preparedness in low- and middle-income countries.

The Gist

Imagine the city of São Paulo as a bustling, crowded party. In the corners of this party, there are tiny, furry guests: bats. Most people don't notice them, but these bats are like living libraries filled with thousands of different viruses. Some of these viruses are harmless to the bats but could be dangerous if they accidentally jump to humans (a "zoonotic spillover").

For years, scientists have been trying to peek inside these libraries to see what's written on the shelves, but the old methods were like trying to find a specific book by only looking at the title. You could only find books you already knew existed. If a brand-new, dangerous virus showed up, the old methods would miss it completely.

This paper describes a new, smarter way to scan these libraries using a technique called viral metagenomics. Here is the story of how they did it, explained simply:

1. The "Lost and Found" Strategy

Instead of going out into the wild to catch bats (which is hard, expensive, and tricky because bats are protected animals), the researchers used a clever shortcut. They tapped into the city's existing "Lost and Found" system for rabies.

Every year, citizens in São Paulo find bats that are sick, injured, or dead near their homes and call the authorities. The city collects these bats to check for rabies. The researchers realized: Why not use these same bats to look for other viruses? It's like checking the "Lost and Found" bin at an airport not just for lost wallets, but to see if anyone accidentally left behind a dangerous package.

2. The "Smart Filter"

They had 2,422 bats in their database. They couldn't sequence all of them (it would cost too much money and take too long). So, they built a digital sieve (an algorithm) to pick the best 150 bats.

• They picked bats from different neighborhoods to get a good geographic mix.
• They picked different species of bats.
• They picked bats found in different situations (some in houses, some on the street).

This ensured they weren't just looking at one tiny corner of the viral library, but getting a representative sample of the whole city.

3. The "Microscope on a Stick"

Once they selected the 150 bats, they took two specific parts from each: the lungs and the intestines.

• Why? Think of the lungs as the front door where respiratory viruses enter, and the intestines as the kitchen where food-borne viruses hang out.
• They used a Nanopore sequencer. Imagine this as a tiny, portable USB stick that can read the genetic code (DNA/RNA) of viruses as they pass through a microscopic hole, like reading a long string of beads one by one. It's cheap, fast, and doesn't need a massive, expensive laboratory.

4. The "Digital Detective Work"

The machine generated millions of lines of genetic code. The researchers then used powerful computer programs to act as digital detectives.

• They compared these lines against a massive database of known viruses.
• They filtered out the "noise" (like plant viruses the bat might have eaten in a bug, or harmless background noise).
• They kept only the "suspects" that looked like real viruses living inside the bat.

5. The Big Discoveries

The results were exciting. They found 98 different viral snippets from 12 different virus families.

• The "Usual Suspects": They found viruses related to Coronaviruses (like SARS), Paramyxoviruses (like Nipah), and others that are known to be dangerous.
• The "Smoking Gun": The most shocking find was a Filovirus (the family that includes Ebola). This was the first time a filovirus has ever been found in a bat in the Americas. It's like finding a rare, dangerous species of shark in a freshwater lake where no one thought sharks could exist.
• The "Red Herring": They also found viruses that likely came from the bat's diet (like viruses from insects or plants the bat ate), showing that the method is so sensitive it catches everything, even things that aren't actually infecting the bat.

6. Why This Matters (The "One Health" Lesson)

The paper concludes that this method is a game-changer for One Health—the idea that human health, animal health, and environmental health are all connected.

• Cost-Effective: It uses bats that are already being collected for free (for rabies checks), so they don't need a new, expensive program.
• Scalable: Because the equipment is portable and cheap, this method can be used in any country, even those with fewer resources (Low- and Middle-Income Countries).
• Early Warning: By finding these viruses before they jump to humans, we can prepare vaccines and treatments. It's like having a smoke detector that goes off when the first wisp of smoke appears, rather than waiting for the whole house to burn down.

The Bottom Line

This study proves that we don't need to build expensive new labs to hunt for the next pandemic. We just need to look a little closer at the animals we are already monitoring. By turning a routine rabies check into a high-tech virus hunt, the researchers in Brazil have created a blueprint for the rest of the world to catch dangerous viruses early, before they can cause a global crisis.

Technical Summary

1. Problem Statement

• Emerging Zoonotic Threats: Approximately 75% of emerging infectious diseases have zoonotic origins, with bats serving as significant reservoirs for diverse RNA viruses (e.g., Coronaviridae, Filoviridae, Paramyxoviridae).
• Limitations of Current Surveillance: Traditional surveillance methods (PCR, ELISA) are restricted to known pathogens, creating blind spots for novel or divergent viruses. Furthermore, active field sampling of bats is logistically difficult, ethically complex (due to legal protections), and costly.
• Underutilized Resources: Brazil's passive rabies surveillance program collects thousands of bat carcasses annually (approx. 3,000/year in São Paulo) via citizen reports. However, these specimens are typically discarded after rabies testing, representing a missed opportunity for broader viral discovery.
• The Gap: There is no integrated, cost-effective framework to leverage these existing passive surveillance samples for broad-spectrum viral metagenomics to detect emerging zoonotic threats.

2. Methodology

The study developed a scalable framework integrating passive surveillance data with nanopore metagenomics.

• Sample Selection & Sourcing:
  • Source: 2,422 bats collected via passive rabies surveillance in São Paulo State, Brazil, between October 2023 and February 2024.
  • Selection Algorithm: An objective decision-tree algorithm was used to select 150 representative specimens (6.19% of the total) to minimize sampling bias regarding geography, species, and time.
  • Inclusion Criteria: Specimens had to be taxonomically identified, negative for rabies, and not infants.
  • Tissues: Paired lung and intestine samples were selected as they represent key interfaces for respiratory and enteric viral shedding/spillover.
• Sequencing Strategy:
  • Platform: Portable Nanopore sequencing (MinION MK1C) using the SMART-9N protocol for unbiased cDNA synthesis.
  • Rationale: Chosen for its portability, low cost, and suitability for resource-limited settings, allowing integration into routine public health workflows without high-end infrastructure.
• Bioinformatics Pipeline:
  • Processing: Raw reads were basecalled, trimmed, and assembled de novo (using MEGAHIT, selected as the most sensitive assembler).
  • Filtering: Rigorous filters were applied to remove artifacts, including:
    • Minimum coverage thresholds.
    • Index-hopping filters (requiring >0.5% of max reads for a family in a run).
    • Secondary similarity searches against broader databases (NCBI nr) to exclude false positives.
    • Exclusion of contigs supported by fewer than two reads.
  • Phylogenetics: Maximum-likelihood trees were constructed using concatenated protein fragments to classify novel sequences.

3. Key Results

• Viral Discovery:
  • Scale: 34.3 million reads generated; 114 high-confidence viral contigs identified from 12 viral families.
  • Prevalence: 16% (24/150) of bats harbored viral hits in at least one tissue.
  • Tissue Specificity:
    • Lungs: Dominated by vertebrate-associated viruses, including Flaviviridae (Pegivirus/Hepacivirus), Paramyxoviridae (Morbillivirus), Coronaviridae (Alphacoronavirus), and Filoviridae.
    • Intestines: Dominated by Phenuiviridae (likely dietary/ectoparasite origin) and Rhabdoviridae.
• Key Viral Families Detected:
  • Zoonotic Relevance: Arenaviridae, Coronaviridae, Paramyxoviridae, Flaviviridae, Rhabdoviridae, and Filoviridae.
  • Novel Findings:
    • Filoviridae: Detection of an unclassified filovirus in a Nyctinomops laticaudatus (free-tailed bat) lung. Phylogenetic analysis placed it outside known Orthoebolavirus and Cuevavirus genera. This is the first reported detection of a filovirus in bats in the Americas.
    • Coronaviridae: A novel Alphacoronavirus in Molossus molossus.
    • Paramyxoviridae: A Morbillivirus closely related to Morbillivirus myotis, known to recognize human cell receptors.
    • Arenaviridae: A sequence highly similar to Mammarenavirus tacaribense.
• Ecological Insights:
  • The study highlighted a distinct separation of viral families between tissues, suggesting tissue tropism and dietary exposure (e.g., insect/plant viruses in intestines) drive virome composition.
  • 25% of infected bats had reported contact with humans or pets, underscoring spillover risk.

4. Key Contributions

• Operational Framework: Demonstrated that passive rabies surveillance samples can be effectively repurposed for viral metagenomics, creating a "low-cost, high-yield" surveillance model.
• Scalability for LMICs: Validated the use of portable Nanopore sequencing as a viable tool for viral discovery in low- and middle-income countries (LMICs) where high-throughput infrastructure is limited.
• Policy Impact: The framework was presented to the Pan American Health Organization (PAHO/WHO) and Brazilian health authorities in February 2025. It was formally recognized as a robust model for national expansion and recommended for inclusion in the National One Health Committee's agenda.
• Novel Pathogen Detection: Provided the first evidence of filoviruses in the Americas, expanding the known geographic range of this high-consequence viral family.

5. Significance

• Pandemic Preparedness: This study proves that "One Health" surveillance can be operationalized by leveraging existing animal health infrastructure. It shifts the paradigm from reactive outbreak response to proactive, pre-emptive viral discovery.
• Cost-Effectiveness: By utilizing specimens already collected for rabies and using portable sequencing, the cost per sample is significantly reduced compared to active field sampling campaigns.
• Early Warning System: The detection of high-consequence viruses (Filoviridae, Paramyxoviridae) in urban synanthropic bats highlights the immediate risk of spillover in densely populated areas like São Paulo.
• Future Directions: The authors recommend integrating targeted PCR, virus isolation, and receptor binding studies to confirm the biological properties of these novel viruses. The framework serves as a blueprint for global surveillance networks to monitor zoonotic threats in urban wildlife.

Limitations Noted:

• Variable time intervals between animal death and sample collection may have degraded RNA, affecting genome assembly completeness.
• Lack of PCR or culture-based confirmation for some novel sequences (though the filovirus finding was re-sequenced to confirm reproducibility).
• Passive surveillance introduces sampling bias (e.g., over-representation of bats found on the ground or in human dwellings).