# Chapter 114. Environmental and Community RNA: Microbiomes, Phages, Ecological Defense, and Surveillance

## Scope Note

This chapter treats RNA as a community-level signal, ecological interaction, defense component, and measurement target in host-associated and environmental systems. Its primary question is how mixed-community RNA measurements and RNA-linked phage or microbial processes support ecological and surveillance inference. Insect antiviral responses remain a prerequisite bridge rather than the organizing center. Extracellular host carriers, circulating biomarkers, and recipient-cell transfer mechanisms belong to [Chapter 107](chapter1102.md); plant immunity, organism-pair host-pathogen mechanisms, RNA virus replication, and general metatranscriptomic workflow depth retain their dedicated owners.

## Executive Summary

RNA operates at community scale in at least three different ways. First, bacteriophages and bacteria use RNA in reciprocal attack and defense. Phage infection is a transcriptional program: early phage RNAs redirect host polymerases, middle transcripts implement replication and immune evasion, and late transcripts produce virion parts and lysis functions. Bacterial defenses include RNA-guided systems such as CRISPR-Cas, RNA-containing retrons, and defense-associated reverse transcriptase systems. Recent DRT9 work shows that a bacterial noncoding RNA can act as a structural and regulatory component of an anti-phage reverse transcriptase complex, enabling an oligomeric transition coupled to phage defense. Other reverse transcriptase-linked defenses synthesize long poly(A)-rich cDNA products during anti-phage responses, and retrons can function in anti-phage defense. The phage-bacteria interface is therefore not simply DNA virus versus bacterial genome; it is a dense RNA and RNA-derived-DNA regulatory battlefield.

Second, host-associated microbiomes contain RNA networks that influence host physiology without reducing to a single "microbial transcriptome" value. Microbial mRNAs and small RNAs report metabolic state, stress, phage infection, and antibiotic exposure. Host microRNAs, extracellular vesicle RNAs, immune transcripts, and metabolite-responsive host RNA programs can correlate with microbial community composition. In inflammatory bowel disease, reviews describe bidirectional hypotheses in which host microRNAs influence bacterial gene expression and microbial metabolites influence host microRNA networks, while also emphasizing context dependence, biomarker instability, and translational limitations. In mice, paternal gut microbiome perturbation can alter the male reproductive system and sperm small RNA payloads, with offspring effects transmitted through gametes and rescued by microbiome recovery before conception. These examples are powerful, but they should not be overgeneralized into a rule that every extracellular RNA detected in a microbiome sample is a functional interspecies message.

Third, community transcriptomics and environmental RNA measurements provide evidence about mixed populations. Bulk metatranscriptomics identifies active pathways in communities but is composition-sensitive, RNA-stability-sensitive, and vulnerable to host contamination, rRNA domination, extraction bias, and database bias. Single-microbe RNA sequencing now begins to resolve transcriptional heterogeneity within microbiomes, including host-phage activity associations in human gut samples. Direct imaging can quantify phage-infected cells in the ocean; pelagiphage infection of SAR11 can coincide with ribosome-deprived infected cells and population-level shifts during blooms. Wastewater RNA sequencing extends these principles into public health, combining targeted viral surveillance with bulk metatranscriptomics to monitor pathogens, microbial communities, and antimicrobial resistance genes across geography and time.

## Concept Inventory

- **Phage host takeover:** the coordinated redirection of bacterial transcription, translation, nucleotide pools, membranes, and metabolism during phage infection. RNA is central because phage genes must be expressed in timed waves, host mRNAs may be degraded or selectively translated, and phage-encoded or host-encoded small RNAs can alter the infection outcome.
- **RNA-linked anti-phage defense:** a broad category that includes CRISPR RNAs, retron RNAs, defense-associated reverse transcriptase noncoding RNAs, and other RNA or RNA-derived molecules that detect, remember, signal, or execute defense against phage. The category is mechanistically heterogeneous; grouping these systems together is useful for comparison but can obscure differences between guide-RNA targeting, abortive infection, and enzymatic production of toxic or signaling nucleic acid products.
- **Microbiome RNA network:** the set of host, bacterial, archaeal, fungal, viral, and extracellular RNA molecules whose abundance, localization, sequence, modification, or packaging changes across a microbial community and its host environment. The term includes functional RNA regulators and measurement readouts. Detection of an RNA correlation alone does not establish cross-kingdom regulation.
- **Community transcriptomics:** RNA sequencing of mixed communities. It can describe which genes are expressed by which taxa or functions in a sample, but it must be separated from metagenomics, which measures DNA potential, and from 16S rRNA profiling, which measures marker-gene composition rather than RNA activity.
- **Environmental RNA:** RNA collected from environmental matrices such as water, sediment, air filters, wastewater, surfaces, or soil. Environmental RNA can provide a shorter-lived and potentially more activity-linked signal than environmental DNA, but RNA degradation, extracellular protection, sampling chemistry, and reverse-transcription bias strongly affect interpretation. Field-level eRNA discussion and later aqueous eRNA method studies therefore treat environmental RNA as promising evidence for recent biological activity, not as direct proof that an organism was alive at the sampling moment.

## What to Know Before Reading This Chapter

The reader should distinguish three uses of the word "RNA response." One use means a molecular defense response, such as a bacterial RNA-linked anti-phage system. A second use means a regulatory program, such as a phage transcriptional cascade or a host inflammatory transcriptome. A third use means a measurement response, such as wastewater RNA fragments increasing before clinical cases are detected. These are related because they all involve RNA, but they are not interchangeable.

The reader should also keep track of biological scale. A bacterial culture can reveal a DRT9 noncoding RNA mechanism, but an environmental microbial community adds strain structure, phage host range, nutrient limitation, and spatial heterogeneity. A wastewater RNA sample can detect viral variants and microbial signatures, but it is a mixed, degraded, flow-dependent composite of many contributors. Mechanistic claims should therefore state whether the evidence comes from purified biochemistry, genetics, cultured cells, animal models, clinical cohorts, environmental sampling, or computational association.

The chapter uses "community-level RNA defense" in a broad but disciplined sense. It includes defenses exchanged between bacteria and phages and ecological defense states inferred from shifts in active microbial populations. It does not imply that a microbial community has a single coordinated immune system. Insect-vector antiviral mechanisms matter here only when they change transmission ecology or the meaning of a surveillance signal; [Chapter 111](chapter1106.md) and [Chapter 113](chapter1107.md) own their molecular and organismal treatment.

## 114.1. Phage RNA programs, bacterial RNA-linked defense, and ecological host takeover

Bacteriophages are often introduced as DNA or RNA genomes inside protein capsids, but an infection becomes biologically real only when transcripts are made and translated. For a tailed double-stranded DNA phage, early RNAs may encode regulators that redirect bacterial RNA polymerase, inhibit host defense, or prepare DNA replication. Middle RNAs often expand replication and recombination functions. Late RNAs encode structural proteins, packaging proteins, and lysis functions. RNA phages and single-stranded DNA phages have different genome strategies, but the same principle holds: the temporal control of RNA synthesis and RNA stability determines whether infection progresses, stalls, or triggers defense.

Phage host takeover includes transcriptional takeover, translational takeover, RNA decay, nucleotide-pool manipulation, and membrane remodeling. Some phages encode anti-sigma factors, alternative sigma factors, antiterminators, or RNA polymerase-modifying proteins. Others encode RNases or recruit host RNases to degrade host transcripts while preserving phage RNAs. Phage transcripts can contain strong ribosome-binding sites, structured leaders, terminators, or anti-terminator-responsive regions that tune expression timing. Bacterial cells can counter by sensing infection-associated transcriptional stress, DNA damage, RNA-DNA hybrids, nucleotide depletion, or phage proteins. Because these processes are intertwined, "phage RNA" should not be treated as a passive copy of phage DNA. Phage RNAs are timed regulatory molecules and sometimes direct actors in bacterial-phage conflict.

![Figure 114.2. Phage RNA Takeover and Bacterial RNA-Linked Defense](../assets/figures/chapter1108_figure2.png)

**Figure 114.2. Phage RNA Takeover and Bacterial RNA-Linked Defense.** Bacteriophage infection depends on timed RNA expression and RNA-sensitive host physiology. Bacterial defenses use RNA as guides, templates, scaffolds, and regulators rather than relying on CRISPR alone.

Small RNAs can participate on both sides of phage infection. The chapter reference scaffold includes a 2024 Cell Host & Microbe study titled "Small RNAs direct attack and defense mechanisms in a quorum sensing phage and its host". This source is highly topical but is currently a manual-PDF/fulltext-available record with unknown reuse status, so this draft cites it for provenance but does not derive detailed prose from converted text. The conceptual lesson is that bacterial small-RNA regulation and phage regulatory programs can be coupled to quorum sensing, host state, and defense activation. Additional expert review should verify exact mechanisms before claim-level expansion.

Bacteria possess many anti-phage systems, and several are RNA-linked. CRISPR-Cas immunity uses CRISPR RNAs to recognize complementary invader sequences. Restriction-modification and phosphorothioate systems are DNA-centered but can affect phage population dynamics and interact with broader defense islands; the chapter scaffold includes a 2024 Annual Review of Microbiology review on DNA phosphorothioate modification systems and associated phage defense systems. Retrons, long known as bacterial elements that produce multicopy single-stranded DNA from a noncoding RNA template by reverse transcription, are now also understood as anti-phage defense modules in some contexts. A landmark Cell paper reported that bacterial retrons function in anti-phage defense. This does not mean every retron is defensive; retrons are diverse, and their defense outputs depend on associated proteins and infection-triggered circuitry.

Defense-associated reverse transcriptase systems show how far RNA-linked anti-phage defense has expanded. In the DRT9 system, a reverse transcriptase protein and a noncoding RNA form a complex whose oligomeric state changes during activation. Cryo-electron microscopy and biochemical assays showed that the noncoding RNA helps assemble the anti-phage complex, and that substrate binding is associated with a transition from a tetrameric to a hexameric arrangement. Catalytic activity and the noncoding RNA are required for DRT9-mediated phage resistance in the tested system. A related Science paper reports bacterial reverse transcriptase synthesis of long poly(A)-rich cDNA for anti-phage defense. These studies support a broader principle: some bacterial defense systems use RNA not only as a guide or message but as a scaffold, template, and allosteric regulator.

The phrase "anti-defense" refers to phage mechanisms that evade, inhibit, or rewire bacterial defenses. Anti-CRISPR proteins are the most familiar examples, but phages can also avoid restriction by modifying DNA, encode proteins that inhibit abortive infection, alter transcription timing to outrun defense, or manipulate host metabolism so that defense activation is too late or too costly. RNA can be part of anti-defense when phage transcripts mimic host regulatory RNAs, sequester proteins, resist nucleases, or shift expression timing. The local automated bibliography is thin for phage anti-defense beyond RNA-linked defense systems, so the anti-defense treatment here is conceptual and should be expanded with dedicated anti-CRISPR and phage-host-takeover reviews before final publication.

There are two common misconceptions. The first is that bacterial defense is equivalent to CRISPR. CRISPR is important, but bacteria also use restriction-modification, abortive infection, toxin-antitoxin systems, retrons, DRTs, defense islands, cyclic nucleotide signaling, and many systems still being characterized. The second misconception is that phage takeover always means global host shutdown. Some phages do rapidly suppress host gene expression, but others exploit active host transcription, translation, and metabolism. In microbial communities, partial infection, lysogeny, chronic release, abortive infection, and failed adsorption all contribute to observed RNA profiles.

## 114.2. Microbiome RNA networks, host physiology, and cross-kingdom evidence

A microbiome RNA network includes RNA molecules made by microbes, RNA molecules made by the host in response to microbes, and extracellular RNAs that move through mucus, lumen, vesicles, diet, or environmental fluid. The gut is the best-studied host-associated example, but similar principles apply to skin, respiratory, reproductive, aquatic, and plant-associated microbiomes. The term "network" is useful because microbial metabolism, host immunity, epithelial turnover, and viral predation are coupled. It is also risky because network language can make correlations sound causal. A rigorous microbiome RNA claim must say which RNA is measured, where it is located, whether it is intracellular or extracellular, what organism produced it, and what perturbation changes it.

Microbial mRNAs report community function more directly than DNA. A gut metagenome can show that genes for butyrate production, bile acid transformation, antimicrobial resistance, or stress response are present. A metatranscriptome asks whether those genes are being transcribed at sampling time. This matters because taxa with similar gene content can occupy different physiological states. The smRandom-seq2 single-microbe RNA-seq study captured tens of thousands of individual microbes from human gut samples, assigned species identities, and identified transcriptional heterogeneity among taxa such as Prevotella, Roseburia, and Phascolarctobacterium. The same study developed a module to identify bacteria-phage transcriptional activity associations. This supports the idea that community RNA is not only a bulk average; it can resolve subpopulations and infection-associated activity inside a complex microbiome.

![Figure 114.3. Host-Microbiome RNA Network Hypotheses and Evidence Gates](../assets/figures/chapter1108_figure3.png)

**Figure 114.3. Host-Microbiome RNA Network Hypotheses and Evidence Gates.** Microbiome RNA networks include microbial state markers, host response RNAs, and possible extracellular RNA transfer. The evidence ladder prevents correlation from being mistaken for communication.

Host microRNAs are another bridge between host physiology and microbiomes. MicroRNAs are short regulatory RNAs, typically about 22 nucleotides, that guide Argonaute proteins to target RNAs and alter RNA stability or translation. Host epithelial cells and immune cells can release extracellular vesicles containing small RNAs. Some studies propose that host-derived microRNAs enter bacterial cells and regulate bacterial gene expression, while microbial metabolites and pathogen-associated molecules alter host microRNA expression. A 2026 review on microRNA-microbiome networks in inflammatory bowel disease presents this bidirectional framework but also emphasizes methodological inconsistency, patient heterogeneity, temporal variability, and limited therapeutic translation. This is the correct teaching stance: cross-kingdom RNA regulation is plausible and supported in specific experimental systems, but it is not a universal explanation for all microbiome-disease associations.

The epigenome-microbiome axis broadens the same concept. A Trends in Microbiology review in the local bibliography focuses on host regulation of gut microbiota through epigenome-microbiome interactions. In this chapter, the relevant RNA point is that microbial metabolites such as short-chain fatty acids, bile acid derivatives, indoles, and inflammatory ligands can change host transcription and post-transcriptional regulation. Some effects pass through chromatin and transcription factor pathways; others involve host microRNAs, long noncoding RNAs, circular RNAs, RNA-binding proteins, or RNA decay. When a host transcriptome changes after microbiome perturbation, the mechanism may be direct sensing of microbial molecules, indirect immune activation, nutrient changes, tissue damage, or altered cell composition.

Disease association studies show the promise and limitations of integrated microbiome-host transcriptomics. A Hashimoto's thyroiditis study integrated gut metagenomic data with peripheral blood miRNA and mRNA data and reported microbial species and host RNA signatures that improved classification of early disease in a small case-control design. This is useful as a method example but not sufficient to prove causation in the gut-thyroid axis. The study itself notes small sample size, sex restriction, lack of disease controls, and the need for animal or transplantation experiments to test causality. Similar caution applies to cancer and inflammatory disease reviews that connect microbiome composition and microRNAs.

One of the more mechanistically provocative examples comes from the paternal gut microbiome. In mice, induced paternal gut dysbiosis changed testicular physiology, metabolite profiles, and sperm small RNA payloads. Offspring of dysbiotic fathers showed increased risk of growth restriction and mortality, and the effect was transmitted through gametes in in vitro fertilization experiments. Recovery of the paternal microbiome before conception rescued the offspring phenotype. This example links microbiome perturbation to germline RNA payloads and offspring physiology, but it should not be generalized to humans without evidence. It shows that microbiome state can affect host RNA-containing reproductive material in a controlled mouse model.

The boundary case for this section is extracellular RNA detected in feces, mucus, plasma, or environmental fluid. Some extracellular RNA is packaged in vesicles or ribonucleoprotein particles and may be protected. Some is released by dying cells or lysed microbes. Some may be dietary or environmental contamination. Some may be reverse-transcription or mapping artifact. Before interpreting extracellular RNA as communication, an experiment should test uptake, sequence specificity, dose, localization, dependence on RNA biogenesis machinery, and functional rescue or loss-of-function effects. This standard is demanding, but without it cross-kingdom RNA claims become too easy to overstate.

## 114.3. Community and environmental transcriptomics: activity, infection, and ecological inference

Community transcriptomics measures RNA from mixed populations. In infection biology, this can mean host-pathogen dual RNA-seq, microbiome metatranscriptomics, vector-pathogen transcriptomics, or multi-kingdom RNA-seq that includes host, bacteria, viruses, fungi, parasites, and phages. In ecology, it can mean RNA from plankton, soil, sediments, biofilms, wastewater, or animal-associated communities. The central promise is functional inference: RNA can show active pathways, stress states, and infection dynamics that DNA alone cannot. The central difficulty is compositional ambiguity: a transcript can increase because a taxon became more abundant, because the same cells expressed the gene more strongly, because RNA became more stable, or because extraction and mapping changed.

**Table 114.1. Community RNA Methods, Outputs, and Interpretation Artifacts.** Community RNA methods differ in input material, resolution, and artifact profile. A method that supports taxonomic or functional inference may still be insufficient for causal interpretation.

| Method | Main input | Main output | Strongest inference | Common artifact | Best validation partner |
| --- | --- | --- | --- | --- | --- |
| **Bulk metatranscriptomics** | Mixed community total RNA, usually after rRNA depletion or capture. | Taxon-assigned transcripts and pathway-level expression profiles. | Active genes, stress states, and functional pathway shifts in a community. | Composition shifts, rRNA depletion bias, host RNA dilution, extraction bias, and database bias. | Spike-ins, metagenomics, absolute RT-qPCR, and time series. |
| **Dual RNA-seq** | Host-pathogen or host-microbiome RNA from the same infected tissue or culture. | Parallel host and microbial transcriptomes. | Reciprocal response timing and pathways linked to infection or dysbiosis. | Host RNA swamping microbial reads; cell composition and pathogen load confounding. | Pathogen load assays, histology or FISH, perturbation, and cell-type resolution. |
| **Small RNA sequencing** | Size-selected small RNAs from host, vector, virus, or microbial samples. | miRNAs, siRNAs, piRNA-like reads, bacterial sRNAs, and short RNA fragments. | Candidate small-RNA pathway activation or regulatory RNA production. | Degradation fragments mistaken for regulators; ligation, size-selection, and source biases. | Argonaute or pathway perturbation, RNA immunoprecipitation, northern blot, and RT-qPCR. |
| **Single-microbe RNA-seq** | Individual microbial cells captured by droplet or random-priming workflows. | Cell-level taxonomic labels, expression states, and phage-activity associations. | Transcriptional heterogeneity within taxa and possible bacteria-phage activity links. | Uneven cell lysis or capture, barcode mixing, rRNA dominance, and incomplete reference databases. | Imaging, culture, spacer or prophage evidence, and matched bulk metatranscriptomics. |
| **Direct-geneFISH or RNA/DNA FISH** | Fixed cells or environmental samples probed in situ. | Spatial counts of target genes, RNAs, phage-positive cells, and rRNA state. | Direct localization of infection-associated or ribosome-deprived cell states. | Probe specificity limits, autofluorescence, thresholding, segmentation, and detection limits. | Sequencing, qPCR, replicate probes, and microscopy controls. |
| **Wastewater targeted amplicon sequencing** | Concentrated wastewater RNA amplified for known pathogen or variant targets. | Target presence, mutation frequencies, variant mixtures, and site-time trends. | Population-level surveillance of known pathogens or variants. | Primer dropout, inhibitors, low coverage, variant deconvolution error, and flow effects. | Recovery spikes, clinical data, flow or fecal normalization, and replicate sampling. |
| **Wastewater or environmental shotgun RNA-seq** | Total RNA from wastewater, water, sediment, biofilm, air, or surfaces. | Broad taxonomic, viral, antimicrobial-resistance, and pathway signals. | Untargeted surveillance of active community signatures and candidate threats. | RNA stability bias, matrix inhibitors, host or environmental contamination, and reference bias. | Targeted RT-qPCR or amplicons, site metadata normalization, culture or imaging, and repeated sampling. |

The evidence ladder begins with sampling and preservation. RNA degrades quickly, but degradation is not uniform across RNA classes, organisms, compartments, or environmental matrices. rRNA is abundant and can dominate libraries. mRNA is more informative for gene activity but is usually less abundant and less stable. Small RNAs require size-specific extraction and library preparation. Viral RNA may be protected in capsids, vesicles, or particles, while naked RNA may decay rapidly. Host RNA can overwhelm microbial RNA in tissue samples. These biases mean that community transcriptomic methods need field blanks, collection and extraction controls, spike-ins, documented preservation timing, rRNA-depletion or capture decisions, and versioned taxonomic databases. Extracellular persistence and carrier protection are prerequisites for interpreting an environmental signal; carrier biology itself belongs to [Chapter 107](chapter1102.md).

The basic metatranscriptomic workflow is conceptually simple but practically fragile: collect and preserve a mixed sample, lyse its different cell types, deplete abundant RNA or enrich targets, construct a library, sequence, quality-filter, map or assemble, assign taxa, and summarize genes or pathways. Each step changes the answer. Gram-positive bacteria, Gram-negative bacteria, fungi, spores, viruses, and host cells lyse with different efficiencies. Depletion probes may work for some taxa and fail for others. Short reads may not distinguish strains, phage-host boundaries, or recently transferred genes. Zhang et al. (2021) and Tyagi and Katara (2024) review these constraints and reinforce that RNA abundance combines transcription, RNA stability, organism abundance, extraction efficiency, and database coverage.

In host-pathogen settings, community RNA can reveal reciprocal responses. A rainbow trout study infected fish with infectious hematopoietic necrosis virus and profiled skin immune RNA plus skin microbiota. Viral infection upregulated antiviral genes early, later coincided with stronger antibacterial immune signatures, and was associated with shifts in skin microbial community composition, including increases in opportunistic taxa. This is not an insect-vector study, but it is useful as a community infection example because it connects viral burden, host transcriptome, mucosal damage, and microbial dysbiosis. It also illustrates a frequent interpretation caveat: viral infection may cause tissue damage that permits bacterial shifts, bacterial shifts may intensify inflammation, or both may be linked through a third variable such as host condition.

In marine ecology, direct imaging can connect phage infection to community dynamics. A study of pelagiphage infection in SAR11 bacteria used fluorescence in situ hybridization approaches to visualize phage-infected cells. During a phytoplankton bloom, up to 19 percent of SAR11 cells were phage-infected, coinciding with a large reduction in SAR11 abundance. The study also described ribosome-deprived but phage-positive cells and detected phage-infected SAR11 and these ribosome-deprived cells across ocean transects. The key RNA concept is that loss or depletion of detectable rRNA can be an infection-associated cellular state, not merely a sequencing artifact. The interpretation remains open: ribosome loss could reflect phage exploitation, host stress, defense, or late infection physiology.

Single-microbe RNA sequencing pushes community transcriptomics beyond averaged samples. The smRandom-seq2 method introduced optimized random priming, droplet barcoding, and computational modules for taxonomic assignment, bacterial expression analysis, and phage activity association in human gut microbiome samples. It recovered single-microbe transcriptional profiles and detected host-phage activity associations. The method is important because microbial communities are made of cells, not only taxa. Two cells assigned to the same species can differ in stress response, mobile element expression, phage exposure, metabolic pathway use, or antibiotic resistance state. However, single-microbe RNA-seq is still technically demanding, database-dependent, and biased toward cells whose RNA is captured efficiently.

Community transcriptomics can also support parasitology and vector ecology, although the local chapter source list has limited direct material. A Plasmodium vivax review discusses insights from genomics, transcriptomics, and proteomics, but this is background for pathogen-omics rather than a direct source for community-level RNA defense. Sepsis biomarker reviews include circulating microRNAs, long noncoding RNAs, and circular RNAs as host-response markers, but they are not direct sources for environmental or microbiome RNA networks. These records are therefore treated as background or quarantined in the audit unless a later revision needs clinical biomarker comparison.

![Figure 114.4. From Community or Environmental RNA Sample to Evidence-Graded Claim](../assets/figures/chapter1108_figure4.png)

**Figure 114.4. From Community or Environmental RNA Sample to Evidence-Graded Claim.** Community and environmental transcriptomics become interpretable when field controls, persistence and transport, assignment methods, organism abundance, and validation routes are matched to detection, activity, functional, or causal claims.

The best practice for community transcriptomics is to separate four claims. A detection claim says that sequence evidence for an organism or virus is present. An activity claim says that an organism, infection, or pathway is transcriptionally active. A functional claim assigns the transcripts to a biological process but does not directly measure pathway flux. A causal claim says that a transcript or pathway changes an infection, physiological, ecological, or public-health outcome. Sequencing alone can strongly support detection and can support activity or functional inference when controls, reference databases, organism abundance, and RNA persistence are adequate. Causality usually requires perturbation, time series, spatial localization, culture or model systems, biochemical assays, or independent validation.

## 114.4. Wastewater and environmental RNA surveillance: detection, normalization, and public-health evidence

Biosurveillance uses biological measurements to detect, monitor, and interpret health or ecological threats. RNA-based biosurveillance includes targeted detection of known pathogens, untargeted metatranscriptomics, variant tracking, antimicrobial resistance expression analysis, environmental biodiversity monitoring, and early warning of community-level changes. Wastewater surveillance is the most visible public-health example because wastewater aggregates RNA and other molecules from many individuals while preserving temporal and geographic information.

Wastewater RNA sequencing can be targeted or broad. Targeted workflows amplify specific pathogens, such as SARS-CoV-2, and estimate variant frequencies from mutation patterns. Broad workflows use shotgun RNA sequencing to profile bacteria, viruses, fungi, antimicrobial resistance genes, and functional pathways. A Miami-Dade County wastewater study integrated targeted and bulk RNA sequencing across 2,238 samples collected from 2020 to 2022. Targeted sequencing tracked SARS-CoV-2 variants across space and time and corresponded with clinical positives from university and hospital populations. Bulk metatranscriptomics detected microbial taxa associated with human-associated and environmental sources, enteric pathogens, viral diversity, and antimicrobial resistance gene enrichment in hospital wastewater that correlated with antibiotic prescriptions.

> **Box 114.1. Environmental RNA Is an Activity and Surveillance Clue, Not Direct Proof**
>
> - Detection is not diagnosis.
> - Trend is not source attribution.
> - eRNA is not automatic proof of viability, local origin, or current activity.
> - Cells, virions, vesicles, ribonucleoproteins, particles, and cold matrices can protect RNA; water, aerosols, animals, and wastewater can transport it.
> - Field blanks, transport controls, preservation timing, and matrix-specific decay measurements are part of the evidence.
> - Flow, rainfall, solids, inhibitors, and population size change wastewater signal.
> - Variant deconvolution depends on coverage, mutation sets, and mixtures.
> - Fine-grained sampling increases governance and privacy burdens.
> - Action thresholds require validation and communication plans.

Environmental RNA, often abbreviated eRNA, extends RNA surveillance beyond sewers. Water, sediment, soil, biofilm, air, and surface samples can contain RNA from local organisms and communities. Compared with environmental DNA, environmental RNA is often proposed as a more activity-linked or viability-linked signal because RNA can degrade faster and can reflect recent transcription. Cristescu's 2019 environmental RNA perspective frames this promise for biodiversity science, and later methods and degradation studies show why the interpretation must remain matrix-specific. RNA can persist if protected in cells, spores, virions, vesicles, ribonucleoproteins, extracellular polymeric material, mineral surfaces, or cold environments, and it can move from its production site through water flow, aerosols, animal movement, or wastewater networks. Different RNA classes decay at different rates. Sampling, filtration, storage, extraction, reverse transcription, and inhibitors can all distort the signal. Field blanks, transport controls, and matrix-specific persistence experiments are therefore biological evidence, not optional cleanup.

Public-health interpretation requires normalization. A wastewater pathogen signal can rise because infections increased, because flow decreased, because solids capture improved, because the population contributing to a site changed, because inhibitors changed extraction efficiency, or because variant deconvolution changed. Controls can include recovery spikes, human fecal markers, flow data, rainfall data, site metadata, sequencing depth, and replicate sampling. Privacy also matters. Building-level wastewater can guide interventions, but fine-grained sampling can reveal health information about small groups. RNA biosurveillance therefore has technical, ethical, and governance dimensions.

![Figure 114.5. From Population Shedding to a Normalized Wastewater RNA Signal](../assets/figures/chapter1108_figure5.png)

**Figure 114.5. From Population Shedding to a Normalized Wastewater RNA Signal.** A wastewater RNA signal is transformed between community shedding and public-health inference. Sewer decay, dilution, solids partitioning, sampling, recovery, inhibition, and contextual denominators must be considered before recovered target RNA supports a normalized population trend or variant mixture; the result is not an individual diagnosis.

The strongest use of RNA biosurveillance is not to replace clinical testing, ecological fieldwork, or mechanistic biology. Its strength is early, repeated, population-level signal detection. Wastewater sequencing can detect variants that individual testing misses. Environmental RNA can indicate recent activity of organisms that are difficult to observe. Metatranscriptomics can identify functional responses before community composition changes are obvious. The limitation is that surveillance signals are hypotheses until validated. A biosurveillance dashboard should distinguish detection, trend, attribution, and action threshold.

## Experimental Foundations and Evidence Standards

The methods in this chapter sit on an evidence spectrum. Biochemistry and structural biology can define a molecular mechanism, as in DRT9 cryo-electron microscopy and reverse transcription assays. Culture and organism-pair experiments can connect RNA responses to infection or fitness but may omit community structure. Community sequencing can capture mixed activity but requires computational assignment. Environmental surveillance can scale across populations but depends on sampling chemistry, field controls, transport, persistence, and normalization. In vector-borne surveillance, viral RNA detection must also be separated from infectious virus and transmission competence; the mechanistic evidence needed for that distinction belongs to [Chapter 111](chapter1106.md) and [Chapter 113](chapter1107.md).

A recurring artifact is compositional confounding. If a bacterial species doubles in abundance, every transcript from that species can appear higher even if per-cell expression is unchanged. If rRNA depletion works better in one sample than another, mRNA proportions can shift without a biological change. If host cells lyse more in inflamed tissue, host RNA can dilute microbial RNA. If a phage kills ribosome-rich cells, the remaining community RNA pool can appear to lose host rRNA even when infection is the driver. Strong studies use time series, microscopy, spike-ins, absolute quantification, and orthogonal assays to separate abundance, expression, and degradation.

Another recurring artifact is overmapping. Short reads from conserved genes can map to the wrong taxon or phage. Viral databases are incomplete and unevenly curated. Phage host assignment from sequence alone is uncertain, especially for uncultured viruses. Single-microbe and metatranscriptomic methods that infer host-phage associations are powerful, but high-confidence assignments require read purity, sufficient viral coverage, alignment thresholds, and ideally direct validation by imaging, culture, spacer matches, prophage context, or strain-resolved assemblies.

## Biological Contexts and Cross-Chapter Boundaries

This chapter emphasizes phages, microbiomes, and environmental communities, but each topic connects to other parts of the book. Insect-vector antiviral RNAi, sequence-independent double-stranded-RNA responses, and the difference between cell-culture restriction and whole-vector transmission competence are treated in [Chapter 111](chapter1106.md) and [Chapter 113](chapter1107.md). RNA virus genome strategies, replication, recombination, and quasispecies are treated in [Chapter 115](chapter1109.md)-[Chapter 118](chapter1112.md). Extracellular host carrier composition and recipient-cell transfer mechanism belong to [Chapter 107](chapter1102.md). Bacterial small RNAs, CRISPR RNAs, riboswitches, and RNA-guided nucleases are treated in earlier bacterial and RNA-interference chapters. Chemistry, library construction, computational assignment, and workflow benchmarking for metatranscriptomics are revisited in the methods and informatics chapters. The purpose here is to show how those measurements and mechanisms become community-scale biological evidence.

The most important boundary is causality. Vector antiviral mechanisms can change transmission, but a vector-associated RNA signal alone does not establish transmission competence. Phage RNA programs can drive host takeover, but phage ecology is not only transcript timing. Microbiome RNA correlations can nominate regulatory axes, but host physiology is not determined by RNA networks alone. Wastewater RNA can guide public health, but it cannot by itself diagnose individual disease. These boundaries should be kept visible because community-level RNA science is especially prone to attractive but overgeneralized narratives.

## Recent Consensus

Recent consensus supports several points. RNA-linked bacterial anti-phage defenses are broader than CRISPR and include retrons and defense-associated reverse transcriptase systems. Microbiome-host RNA associations are real enough to motivate integrated study, but many cross-kingdom RNA claims remain context-dependent and technically challenging. Community transcriptomics and wastewater RNA sequencing are practical tools for ecological and public-health surveillance when organism abundance, RNA persistence, field controls, normalization, reference bias, and privacy are treated as part of the scientific method.

## Open Questions, Controversies, Deprecated Models, and Common Misconceptions

Open questions:

- Which RNA measurements best distinguish active infection from persistent, transported, or protected viral material in vector-associated and environmental samples? Mechanistic vector competence and antiviral-pathway tests belong to [Chapter 111](chapter1106.md) and [Chapter 113](chapter1107.md); this chapter owns the ecological and surveillance inference.
- How do phage and bacterial small RNAs shape infection outcomes in natural communities? There are strong mechanistic examples of RNA-linked defense, but community-level prevalence, host range, anti-defense countermeasures, and fitness costs remain incompletely mapped. Defense-associated reverse transcriptase systems are especially active areas because new mechanisms are still being discovered.

Controversies:

- Cross-kingdom microRNA communication remains controversial. Some experimental systems support host microRNAs entering bacteria and changing bacterial transcripts or growth, and disease reviews synthesize possible bidirectional networks. But extracellular RNA stability, uptake efficiency, dose, sequence specificity, and reproducibility remain sources of dispute. The safe statement is that cross-kingdom RNA regulation exists in some contexts and is not yet a universal explanatory framework for microbiome physiology.

Common misconceptions:

- "Metatranscriptomics directly measures what the community is doing." Metatranscriptomics measures recoverable RNA molecules after sampling, extraction, library construction, sequencing, and computational assignment. It can infer activity, but only through controls and appropriate normalization.
- "Environmental RNA proves that an organism is alive at sampling time." Environmental RNA often implies a more recent or active signal than DNA, but RNA can persist when protected, and different matrices change decay rates. eRNA should be interpreted probabilistically unless paired with independent evidence.
- "Wastewater surveillance is only pathogen detection." Wastewater RNA can track variants, enteric viruses, microbial community signals, antimicrobial resistance gene expression, and broader public-health indicators, but ethical and technical constraints increase as sampling becomes more local.