Visualizing Viral and Host Processes Using Fluorescence Microscopy

1. Introduction

Virus infection is a spatially and temporally organized process in which viral life-cycle progression, host-cell remodeling, and antiviral responses continuously interact to determine infection outcome [1,2,3]. Influenza A virus (IAV) provides a clear example of this spatial and temporal organization: its life cycle requires coordinated use of distinct cellular compartments, such as the endosomal compartment, the nucleus, intracellular trafficking networks, and the plasma membrane [4,5]. Importantly, infection progression and outcome are not uniform across infected cells. Even within genetically identical cell populations, individual cells can follow distinct trajectories due to differences in incoming virus particles, viral gene expression dynamics, host-cell state, and antiviral response activation [2,6,7,8]. As a result, measurements that average across populations can obscure cell-to-cell heterogeneity, including divergent infection trajectories and rare infection states. Recent advances in single-particle tracking, live-cell microscopy, multiplexed imaging of viral RNAs, super-resolution microscopy, and tissue-scale imaging have begun to fill important gaps between population-averaged descriptions of infection and the spatially and temporally organized events occurring in individual cells [9,10,11]. These imaging modalities reveal when and where viral and host processes occur, resolve heterogeneous infection trajectories, and allow multiple features of infection to be related within the same cell or tissue. However, insights obtained through imaging remain distributed across studies focused on different viruses, life-cycle stages, host processes, imaging modalities, and spatial scales. Here, we integrate diverse imaging studies to assess how imaging has collectively expanded our understanding of virus infection.

With a primary focus on IAV, this review draws on selected examples from other viruses that offer relevant methodological or biological insights and examines infection from three complementary perspectives: viral components and infection progression, virus-induced remodeling of host-cell organization, and antiviral responses. Together, these perspectives show how viral life-cycle progression, host-cell remodeling, and antiviral responses are interconnected and shape infection outcomes.

2. Imaging Viral Components

2.1. Visualizing Virion Attachment, Trafficking, and Membrane Fusion

IAV infection begins with attachment to susceptible cells. To establish productive attachment, virions must navigate extracellular environments while avoiding repeated engagement with decoy receptors [12,13,14]. HA binds terminal sialic acid residues on cellular and extracellular glycoconjugates, whereas NA cleaves these residues; accordingly, perturbation of either HA receptor binding or NA enzymatic activity reduces virion infectivity [15,16]. However, extracellular navigation is a spatial process, and endpoint measurements could not reveal how HA and NA together direct virion movement. Labeling virions with lipophilic membrane dyes enabled tracking of individual particles across sialylated surfaces and through complex extracellular environments (Figure 1A) [12,17]. Tracking single virions revealed that coordinated receptor binding by HA and receptor cleavage by NA promote persistent movement through sialylated environments by preventing repeated attachment to previously encountered receptors [12], likely facilitating successful engagement with susceptible cells. Following productive attachment, IAV particles are internalized by endocytosis, and subsequent acidification of endosomal compartments triggers fusion between viral and host membranes. Fluorescent labeling allowed visualization of virion attachment to the host cell plasma membrane and transport through the endosomal network, while self-quenching lipid dyes provided a direct readout of the timing and location of membrane fusion [18,19,20,21]. Combining single-virion tracking with detection of membrane fusion demonstrated that fusion does not occur immediately after uptake. Instead, virions undergo microtubule-dependent transport through the endosomal network toward the perinuclear region before endosomal acidification triggers membrane fusion [18]. Thus, single-virion tracking and membrane-fusion measurements established that successful entry depends on completion of multiple trafficking and maturation steps before genome release.

2.2. Following Incoming Viral Ribonucleoprotein Complexes (vRNPs) from Fusion to Nuclear Import

The events linking fusion to nuclear delivery have remained difficult to study because incoming vRNPs are present in low copy numbers. Early immunofluorescence (IF) studies of vRNPs performed at a high multiplicity of infection (MOI) suggested that nuclear import is efficient and rapid [22]. Subsequent application of single-molecule fluorescence in situ hybridization (smFISH) enabled visualization of viral RNA segments during the earliest stages of infection, revealing vRNP subcellular localization, colocalization, and nuclear import [23]. However, because these experiments required high-MOI infections and relied on fixed-cell snapshots, neither approach could track individual incoming vRNPs or reveal the timing of the intermediate steps linking membrane fusion to nuclear delivery. An important advance came with quantum-dot labeling of the PA subunit of the viral polymerase, allowing the first direct visualization of individual incoming vRNPs in living cells [24]. Quantum-dot tracking of individual incoming vRNPs demonstrated that membrane fusion occurs predominantly near the nucleus, where individual vRNPs are released from the incoming bundle before rapidly being imported into the nucleus. Recently, live vRNP labeling using fluorescent nanobodies targeting NP within vRNP complexes (VISUN; vRNP imaging of unmodified negative-strand viruses) enabled visualization of all incoming vRNPs without viral engineering or in-virion labeling procedures (Figure 1B) [25]. VISUN confirmed rapid and complete post-fusion vRNP debundling and demonstrated that the majority of incoming vRNPs successfully reach the nucleus within minutes of membrane fusion. Thus, the rapid and efficient transition from membrane fusion to nuclear delivery does not pose a major barrier to productive infection.

2.3. Visualizing Viral Protein Expression and Localization

Viral gene expression results in the accumulation of viral proteins, which can be visualized by immunofluorescence and used as markers of infection progression. Beyond measuring protein abundance, fluorescence imaging revealed that many influenza proteins undergo characteristic changes in subcellular localization that reflect their changing functions throughout infection [26,27]. For example, observation of nuclear accumulation of M1 helped define its role in terminating viral replication and promoting nuclear export of vRNPs [28]. Following completion of genome replication, NP redistributes from the nucleus to the cytoplasm during vRNP nuclear export. Imaging of NP localization thus directly visualizes the transition from viral genome replication to cytoplasmic transport [29]. Taken together, viral infection is not a static process of protein production, but involves coordinated spatial redistribution of viral factors between cellular compartments as infection progresses. Live-cell imaging complements fixed-cell approaches by resolving the temporal dynamics of viral protein expression and localization. Approaches for tracking protein expression or localization in living cells broadly fall into two categories: viruses that express directly tagged viral proteins, and reporter viruses that express fluorescent or luminescent proteins as indirect readouts of viral gene expression (Figure 1C).

Direct visualization of IAV proteins was achieved using recombinant IAVs expressing genetically tagged PB2[30,31], PA [32], NS1[33,34], or structural proteins [35]. Because many RNA viruses, including IAV, tolerate only limited genetic perturbation, particularly large insertions encoding fluorescent proteins, recombinant fluorescent viruses require extensive validation and optimization to ensure that tagging does not compromise viral fitness or genetic stability [31]. Live tracking of fluorescently labeled polymerase-containing vRNPs revealed directed, microtubule-dependent transport through the cytoplasm following nuclear export (Figure 1E) [30,31,32]. Real-time visualization of viral structural proteins enabled direct observation of virion assembly and release at the plasma membrane [35]. Because fluorescence becomes detectable only after sufficient accumulation of the tagged protein, direct protein-labeling approaches are best suited for studying later stages of infection and are generally less informative for the earliest stages.

Reporter viruses expressing fluorescent or bioluminescent proteins can be used to monitor infection spread continuously in cultured cells and animal models, revealing how viral dissemination depends on viral genotype, host antiviral responses, and tissue context [33,36,37,38,39]. At the single-cell level, substantial variation in the onset and rate of viral gene expression has been shown among individual infected cells despite synchronous infection [37]. Multicolor reporter viruses enabled coinfection to be visualized within individual cells and functional interactions between coinfecting viruses to be monitored [40]. Interactions between coinfecting viruses can either inhibit or promote productive infection. For example, live imaging showed that primary infection progressively reduces susceptibility to secondary infection through superinfection exclusion [40]. Conversely, coinfecting viruses can complement incomplete viral genomes by supplying missing functions in trans, thereby supporting productive infection [41].

2.4. Imaging Viral RNA Synthesis and Genome Replication

Although viral protein expression reports on infection progression, it provides only an indirect view of the underlying transcriptional and replication program. IAV infection progression is governed by the synthesis of distinct viral RNA species, including viral mRNAs, full-length complementary RNAs (cRNAs) that serve as replication templates, and negative-sense genomic RNAs (vRNAs). Biochemical approaches established the average temporal order of viral transcription and genome replication [42,43,44,45,46]. RNA imaging introduced a complementary perspective by allowing viral RNAs to be detected directly in individual infected cells while retaining information about their intracellular localization (Figure 1D). Initial RNA FISH studies examined the spatial organization of the eight viral genome segments within virions and showed that individual virions generally package one copy of each genome segment [47]. Subsequent multiplexed RNA imaging addressed the dynamics of viral transcription and replication. For example, multiplexed error-robust FISH (merFISH) was used to simultaneously quantify all eight viral mRNAs in individual infected cells, revealing segment-specific differences in cytoplasmic mRNA abundance due to both differential transcription and nuclear export [48]. More recently, direct RNA padlock probing coupled with rolling-circle amplification and in situ sequencing (mudRapp-seq) enabled simultaneous visualization of viral mRNAs and genomic RNAs, allowing transcriptional output and genome replication to be quantified on a segment-by-segment basis [49]. The mudRapp-seq measurements revealed pronounced segment-specific differences in transcriptional output despite largely similar genome replication dynamics. Direct live imaging of viral RNA dynamics and trafficking remains technically challenging. Current live RNA imaging strategies (e.g., PP7/MS2 stem-loop systems, RNA aptamers, or CRISPR-dCas13 imaging) rely on large engineered RNA insertions or exogenous probe delivery [50,51] and are difficult to implement in IAV because of its compact NP-coated vRNP architecture, limited genetic flexibility, and stringent RNA packaging signals distributed throughout each genome segment [5,52].

2.5. Continuous Live Imaging of vRNPs

IAV genome segments can instead be visualized indirectly, by targeting the associated proteins within the vRNP complex. Earlier protein-labeling strategies enabled imaging of either incoming [24] or newly synthesized [30,31,32] vRNPs, but not continuous tracking throughout infection. More recently, imaging of cells expressing fluorescent nanobodies targeting NP (VISUN) enabled continuous visualization of vRNPs throughout infection and direct measurement of the timing of genome replication and nuclear export in living cells (Figure 1B) [25]. Because individual infection trajectories can be followed from entry onward, VISUN also makes it possible to identify and quantify abortive infections, providing a more complete picture of infection outcomes within a cell population. Finally, because VISUN does not require viral engineering, it enables direct quantitative comparison of the life cycles of genetically diverse influenza strains and clinical isolates.

2.6. Imaging Genome Assembly and Progeny Production

Following nuclear export, newly synthesized vRNPs are transported toward the plasma membrane, where progeny virions are assembled. However, how assembly of the eight distinct genome segments is coordinated during transport to budding sites remained unclear. Segment-specific RNA imaging demonstrated that individual genome segments progressively colocalize during cytoplasmic transport, while live-cell imaging of fluorescently labeled polymerase-containing complexes revealed organized, directional trafficking of vRNPs along microtubules. Together, progressive segment colocalization and organized, directional vRNP trafficking support a model in which genome assembly occurs through sequential interactions on membranous compartments prior to budding rather than exclusively at the plasma membrane [23,32]. Recently, the addition of fluorescent nanobodies targeting the HA protein to the extracellular environment has provided the first live readout of progeny virion budding (Figure 1F) [25]. Combining HA-nanobody-based detection of progeny virion budding with imaging of vRNP nuclear export demonstrated that progeny virion production follows vRNP nuclear export with minimal time delay. However, not all cells with observed vRNP nuclear export subsequently produce detectable progeny virions, revealing that budding is not a guaranteed next step after vRNP export and instead is an additional source of infection heterogeneity.

3. Imaging Virus-Induced Host-Cell Remodeling

Following viral processes through a host cell identifies where successive stages of infection occur but does not necessarily reveal how infection changes the cellular environment in which those stages proceed. Viral infection remodels host cells by redistributing molecular components, reorganizing organelles, constructing new reaction environments, and altering cellular and tissue architecture. Because many virus-induced cellular changes are defined by their spatial organization, temporal dynamics, or ultrastructure rather than molecular abundance alone, imaging is uniquely suited to study virus-induced host-cell remodeling. Different imaging modalities provide complementary views: fluorescence microscopy visualizes the distribution of selected molecules, live-cell imaging resolves their dynamics, electron microscopy reveals membrane architecture and ultrastructure, and tissue imaging relates cellular phenotypes to higher-order organization. Accordingly, conclusions about host-cell remodeling should reflect the specific cellular property that each imaging approach directly measures.

3.1. Visualizing Spatial Remodeling of Gene Expression

Cellular gene expression depends not only on molecular abundance, but also on localization of RNAs and RNA processing machinery. Immunofluorescence (IF) and single-molecule RNA FISH (smFISH) can determine whether selected proteins or RNAs remain dispersed or accumulate in compartments at the single-cell level, and have been particularly useful for studying IAV-induced host shutoff, a process characterized by extensive spatial reorganization of the cellular gene-expression machinery. Combined smFISH and immunofluorescence microscopy revealed depletion of cytoplasmic poly(A) RNA and nuclear accumulation of poly(A) RNA and poly(A)-binding protein 1 (PABPC1), while confocal fluorescence microscopy further showed dispersal of nuclear speckles, nuclear compartments involved in pre-mRNA processing [53,54,55]. Comparisons of cells infected with recombinant IAVs lacking expression of either NS1 or PA-X distinguished the individual spatial changes that together comprise host shutoff. Viral protein PA-X accounts for much of the cytoplasmic RNA loss, and NS1 compromises NXF1-dependent mRNA export. Consistently, nuclear PABPC1 accumulation depends on concerted PA-X and NS1 activity. Thus, combining imaging of host-shutoff-associated spatial changes with the use of engineered viruses that fail to express individual viral proteins reveals mechanistically distinct host-shutoff processes.

Although fixed-cell IF and smFISH reveal spatial reorganization of RNAs and RNA-processing machinery during infection, they do not report on RNA integrity, translation, or movement [56]. By combining live-cell imaging of viral RNA translation and replication dynamics with fixed-cell measurements of host protein synthesis, a study of coxsackievirus B3-infected cells showed that global host protein synthesis was already substantially reduced before replication of the incoming viral RNA, whereas viral translation efficiency remained stable throughout infection [57]. This illustrates how combining measurements of viral and host RNA dynamics with protein synthesis can provide functional information that complements mRNA localization measurements. Applied to influenza, similar approaches could directly link virus-induced changes in RNA localization to their consequences for protein synthesis.

3.2. Detection of Virus-Built Reaction Environments

IAV replication takes place in the nucleus and does not require formation of the membrane-associated replication organelles (ROs) characteristic of many cytoplasmic RNA viruses [5]. In contrast, diverse viruses such as flaviviruses, enteroviruses, and coronaviruses construct de novo membranous compartments dedicated to viral replication [58]. Imaging approaches developed to understand RO biogenesis have clarified their molecular composition, assembly dynamics, and membrane architecture. Fluorescence microscopy identified enriched viral and host components [58]. Live fluorescence imaging revealed the dynamics of RO formation [59], whereas electron tomography revealed highly organized, topologically complex membrane architecture [60]. Together, these observations showed how viruses construct and organize specialized replication environments. However, molecular enrichment, morphology, and exchange dynamics do not by themselves establish organelle function, which requires complementary functional evidence.

3.3. Visualizing Organelle Function and Membrane Topology

Infection also remodels pre-existing organelles, creating an important interpretive distinction between changes in morphology and changes in function. Mitochondria illustrate this distinction, since they may fragment, elongate, or redistribute during infection. Yet, these morphological changes do not by themselves establish altered mitochondrial function. Specialized live reporters and fluorescence biosensors can measure membrane potential or local biochemical activity, that are not apparent from mitochondrial shape or position. Live-cell imaging using mitochondrial functional reporters in IAV-infected cells linked expression of selected PB1-F2 variants to loss of membrane potential and organelle injury, but the magnitude and direction of IAV-associated mitochondrial changes depend on viral strain and experimental context [61,62,63].

Membrane topology presents a related but distinct imaging challenge: the distribution of labeled proteins does not define the membrane structures that contain them. During virion budding, for example, viral proteins accumulate at defined sites on the plasma membrane, but their localization alone does not reveal how the membrane is reshaped. Electron microscopy and tomography therefore complement fluorescence imaging by resolving distinct membrane architectures during IAV budding, including glycoprotein-associated membrane curvature, M1-dependent formation of filamentous virions, and M2 enrichment at budding necks associated with membrane scission [64,65,66]. Together, fluorescence microscopy, electron microscopy, and tomography combine molecular localization with direct visualization of membrane architecture to reveal membrane-remodeling events that cannot be inferred from protein localization alone.

3.4. Following Endomembrane Trafficking

To transport progeny vRNPs from the nucleus to the plasma membrane, IAV remodels the pre-existing Rab11-dependent recycling endosome network. This remodeling of the recycling endosomal compartment was characterized by studies using several complementary imaging approaches. Fixed-cell fluorescence showed that the normally dispersed Rab11 recycling endosomal network becomes enlarged and clustered during infection [67]. Using a vRNP reconstitution system, dual-color live imaging first identified vRNP cotrafficking with Rab11-positive membranes [68]. Light-sheet imaging later resolved Rab11-vRNA cotransport during infection and showed that vRNP-bearing carriers move more slowly, travel shorter distances, and arrest more frequently [69]. Fixed and live measurements therefore establish both redistribution of the recycling network and a change in its transport dynamics. Correlative light-electron microscopy localized vRNPs to infection-remodeled Rab11 carriers, directly connecting Rab11 redistribution with vRNP trafficking [70]. Live imaging showed that vRNP-containing inclusions deform, fuse, split, and exchange material (Figure 1E), while fluorescence recovery measurements supported rapid internal turnover [71]. The observed dynamics and rapid internal turnover of vRNP-containing inclusions are consistent with liquid-like behavior but can also arise from other binding and clustering mechanisms. Definitive assignment of the material state therefore requires complementary evidence [72].

As IAV reorganizes the Rab11 compartment, Rab11-dependent cargo recycling declines. Functional assays using fluorescent transferrin as recycling cargo demonstrated that Rab11-dependent cargo transport is attenuated during IAV infection. Fluorescence imaging of Rab11 redistribution, combined with functional recycling and binding measurements, supported a model in which vRNPs compete with Rab11-interacting proteins for Rab11 binding, thereby reducing cellular recycling and contributing to Rab11 redistribution [67]. A subsequent study combining fluorescence imaging with perturbational evidence showed that ATG9A promotes the release of vRNP-bearing Rab11 endosomes from microtubules and their organization into viral inclusions, indicating a regulated reorganization mechanism not fully explained by competition for Rab11 binding alone [73]. Correlative light-electron microscopy further showed that altered Rab11 carriers associate with remodeled endoplasmic reticulum, suggesting that IAV reorganizes a broader membrane network than previously appreciated [70]. The combined evidence therefore supports regulated remodeling of the recycling endosome compartment that extends beyond Rab11-binding competition alone. Overall, IAV redirects a constitutive recycling system into a slower, spatially concentrated transport environment that supports vRNP delivery while compromising host cargo recycling. This model was established by integrating spatial, dynamic, ultrastructural, and perturbational measurements rather than any single imaging approach.

3.5. Imaging of Intercellular and Tissue-Scale Remodeling

Virus infection remodels host tissues at multiple spatial scales, ranging from interactions between neighboring cells to changes in tissue architecture and function. Imaging has been essential for linking intercellular and tissue-level changes to virus dissemination and pathogenesis. Confocal and super-resolution imaging localized Rab11-associated vRNPs within actin-rich intercellular connections between infected and neighboring cells, demonstrating directed movement of viral material from infected cells toward neighboring cells [74]. Infectivity measurements confirmed that transfer through these structures seeds productive infection, identifying a potential route of cell-to-cell spread that complements extracellular virus release. Beyond intercellular communication, imaging revealed how infection alters tissue function. Live imaging of IAV-infected airway systems demonstrated infection-associated changes in ciliary activity [75]. Longitudinal imaging of SARS-CoV-2-infected airways linked mucociliary movement to the spatial pattern of viral spread, showing that tissue-level transport processes shape infection dissemination across the airway epithelium [76]. Infection-induced remodeling also changes tissue composition. For example, three-dimensional light-sheet imaging of Zika virus-infected cerebral organoids quantified the loss and redistribution of specific cell populations [77]. Similarly, imaging showed that IAV infection depletes ciliated cells and reduces airway epithelial thickness, thereby changing the cell-type composition of the airway epithelium [78]. Yet electrical measurements indicated preserved barrier function, demonstrating that tissue composition and function can be remodeled independently during infection. By linking infection to changes in intercellular communication, tissue organization, and function, imaging extends virology research beyond individual cells to higher-order processes that influence viral spread and disease progression.

4. Heterogeneity of Antiviral Responses

Virus-induced host-cell and tissue remodeling occurs alongside antiviral responses that can redirect infection trajectories. Cells detect virus infection through pattern-recognition receptors activated by pathogen-associated molecular patterns (PAMPs), including viral nucleic acids with aberrant structures or subcellular localization. PAMP recognition triggers antiviral responses, most notably interferon (IFN) production and the expression of hundreds of interferon-stimulated genes (ISGs) [79], while viruses counteract PAMP-induced antiviral responses by concealing viral PAMPs, sequestering replication intermediates, and disrupting antiviral signaling pathways [80]. At the single-cell level, the balance between antiviral signaling and viral evasion produces heterogeneous response activation even under uniform infection conditions [2,6], reflecting both virus-intrinsic differences in genome integrity, gene expression, or particle composition and host-cell differences in the abundance and activity of sensing components [6,81,82]. Cell-to-cell heterogeneity in antiviral responses is evident during IAV infection. IAV evades antiviral responses in most infected cells [6,83], and the PAMPs that trigger immunity remain incompletely defined. Full-length vRNPs, defective genomes, and aberrant replication products have all been implicated as potential triggers of innate immune activation [84,85,86]. Imaging provides three complementary perspectives on cell-to-cell heterogeneity in antiviral response activation. First, time-resolved imaging that follows infection progression and host responses in individual cells identifies which viral life-cycle trajectories are associated with antiviral response activation [87]. Second, imaging the spatial organization of viral components and host factors reveals how antiviral signaling is initiated, amplified, or disrupted [88]. Third, localizing the effects of interferon-stimulated restriction factors identifies the viral life-cycle stages at which antiviral responses inhibit infection [89].

4.1. Relating Infection Trajectories to Antiviral Response Activation

Because population-level measurements average across infected cells, they can obscure antiviral responses confined to a responding minority. Imaging at single-cell resolution can identify responding cells and relate antiviral response activation to infection progression within the same cell. By preserving spatial context, imaging can additionally relate antiviral responses in infected cells to responses in neighboring bystander cells. For example, live-cell tracking of IAV replication, combined with fixed-cell analysis of antiviral response activation, showed that IFN-λ induction occurred predominantly in cells lacking functional NS1 expression while exhibiting high levels of vRNP replication [25]. Studies of other viruses have have extended time-resolved analysis of infection and antiviral response activation to individual infection trajectories and multicycle spread. The use of recombinant vesicular stomatitis virus (VSV) expressing fluorescent reporters, together with cells carrying an IFIT2-promoter-driven fluorescent reporter, enabled live visualization of viral spread and antiviral response activation across infected cell layers [90]. These measurements showed that viral spread was slower when antiviral signaling remained intact during multicycle infection. A complementary live-cell study of encephalomyocarditis virus (EMCV) infection combined virus imaging with an IFIT1 reporter, allowing simultaneous monitoring of viral replication and antiviral response activation in individual cells [87]. Analysis of individual infection trajectories revealed that cells with early, low-level viral replication were the most likely to activate antiviral signaling. The combined single-cell observations of EMCV infection and antiviral responses support a model in which antiviral signaling is favored when early viral replication generates sufficient PAMPs before viral immune-evasion mechanisms become fully established. Together, these studies show how single-cell imaging can resolve the temporal relationship between viral replication, antiviral signaling, viral immune evasion, and the resulting effect on virus spread.

4.2. Imaging Spatial Mechanisms of Antiviral Response Activation

Bulk measurements of antiviral signaling components do not reveal where viral PAMPs, host sensors, and signaling molecules assemble during innate immune activation. Imaging preserves this spatial context, allowing the localization of viral and host signaling components to be related to signaling outcomes [91,92]. The nonstructural protein 1 (NS1) is the principal IAV innate immune antagonist and interferes with several components of antiviral signaling [80,93]. Consistently, cells infected with NS1-deficient virus are more likely, but not guaranteed, to activate immune responses [6,94]. Immunofluorescence microscopy revealed that NS1 suppresses IFN expression by inhibiting nuclear accumulation of interferon regulatory factor 3 (IRF3), directly linking altered protein localization to impaired antiviral signaling [95]. Complementary work showed that cytosolic NS1 binds the E3 ubiquitin ligase TRIM25 to inhibit the RIG-I ubiquitination required for effective signaling to IRF3[96]. Super-resolution imaging has further been used to examine the spatial organization of TRIM25 in NS1-containing cytosolic aggregates, suggesting a spatial mechanism by which NS1 disrupts RIG-I signaling [97]. Immunofluorescence imaging of antiviral stress granule (avSG) markers also showed that NS1 inhibits the formation of avSGs [98,99], dynamic RNA-protein assemblies that concentrate antiviral signaling molecules such as TRIM25, RIG-I, and PKR. Together, visualization of IRF3 nuclear accumulation, TRIM25 sequestration in NS1-containing aggregates, and antiviral stress-granule formation reveals how NS1 disrupts the localization and assembly of antiviral signaling components.

4.3. Imaging Stage-Specific Mechanisms of Antiviral Restriction

Antiviral restriction factors impair productive infection, but the stage of the viral life cycle at which restriction occurs and the underlying molecular mechanism are often unclear. Imaging can localize these stage-specific effects, providing insight into how ISGs suppress virus replication. This is well illustrated by the interferon-induced transmembrane protein IFITM3, which restricts IAV infection [89]. Single-particle live-cell imaging of fluorescently labeled virions was used to determine which stage of entry is targeted by IFITM3. IFITM3 overexpression did not prevent lipid mixing between viral and endosomal membranes but blocked release of viral contents into the cytosol. Distinguishing lipid mixing from cytosolic content release localized the IFITM3-mediated block between hemifusion and fusion pore formation [100]. A second important interferon-stimulated restriction factor in IAV infection is human MxA, which imposes a particularly strong barrier to many avian IAVs. Several studies connect MxA activity to the viral NP protein [101,102], but place MxA-mediated restriction at different stages of infection. Imaging implicated MxA in blocking nuclear accumulation of incoming vRNPs [103]. More recent work further implicated human MxA in the cytoplasmic sequestration of newly synthesized vRNPs, preventing their trafficking to the plasma membrane [104]. The reported effects at entry and during late-stage vRNP trafficking indicate that MxA may act at multiple stages of infection, with the dominant restriction point potentially depending on viral and cellular context. More generally, imaging can identify the stage of the viral life cycle perturbed by antiviral restriction by comparing the localization and trafficking of viral components in restricted and unrestricted infections.

5. Concluding Remarks

Throughout this review, we describe imaging not as a single methodology, but as a collection of complementary approaches that reveal distinct aspects of virus infection (Table 1). Fluorescence microscopy resolves the spatial organization of viral and host components, live-cell imaging establishes the temporal order of infection events, RNA imaging indirectly measures viral transcription and genome replication, and tissue-scale imaging relates intracellular infection to changes in tissue organization and function. Together, these approaches extend population-averaged measurements by enabling direct observation of individual infection trajectories and relationships between viral and host processes that would otherwise remain obscured.

Imaging provides direct evidence for the measured observable, but not necessarily for the biological mechanisms underlying those observations. For example, colocalization does not establish molecular interaction, morphology does not define function, and RNA localization does not imply translation. Likewise, fluorescence measurements alone cannot resolve membrane topology or molecular mechanism. Interpreting imaging data therefore requires matching biological conclusions to the observable being measured and supporting mechanistic models with complementary functional evidence. Because different imaging approaches measure complementary observables, integrating these measurements can provide a more complete picture than any single approach alone. Measuring complementary parameters is particularly informative because viral and host-cell processes are themselves highly interconnected. Virus infection can remodel host cells, while host-cell remodeling and antiviral responses in turn influence the progression and outcome of infection. Yet specialized imaging approaches often examine viral replication, host-cell remodeling, antiviral signaling, and tissue organization separately. Combining measurements of these processes within the same cells or tissues would allow their spatial and temporal relationships to be examined directly. Such integrated measurements could link early intracellular events to subsequent productive or nonproductive outcomes, identify candidate causal relationships that can be evaluated using complementary evidence, and provide a more integrated understanding of how viral and host processes together shape infection outcome.