Immune Cell Dynamics in Living Tissues: Insights From Intravital Microscopy

Abstract

Intravital microscopy has emerged as a pivotal platform for real-time visualization of immune responses at the cellular level, providing detailed insights into the dynamic behaviors of immune cells in both healthy and pathological conditions. However, linking these dynamic observations to heterogeneity at the single-cell level remains challenging because of the difficulty in identifying specific behavioral or functional states in vivo. Recent advances in microscopic techniques and image analysis have facilitated further investigation of immune cells in various organs. These advancements have not only improved the resolution of immune-cell behaviors but have also revealed novel therapeutic targets. In this review, we discuss the potential of intravital microscopy techniques and the recent advances in image-based analyses to obtain a comprehensive view of various immune cells under pathological conditions.

INTRODUCTION

Intravital microscopy (IVM) is a powerful and influential approach for studying immune responses that allows real-time imaging of immune-cell behavior within living organisms at high spatial and temporal resolutions (1,2). Unlike traditional in vitro systems, which often fail to fully capture the dynamics of tissue microenvironments, IVM allows examination of immune cells within their native physiological contexts, where the vasculature (3), extracellular matrix (ECM) composition (4), and local chemotactic gradients continually affect cellular behavior (5). Through real-time visualization, this technique has revealed previously underappreciated aspects of immune-cell biology, including dynamic migration patterns (6), intercellular communication (7), and rapid functional transitions that occur during immune activation, infection, and tissue injury (8). Collectively, these advances have shifted our understanding of immune responses from static model-based interpretations to dynamic, context-dependent processes occurring in living tissues.

Technological advances in microscopy are central to the expansion of IVM applications. The development of multiphoton (9) and spinning-disk confocal microscopy (10) has enabled high-resolution imaging of immune-cell dynamics across a wide range of organs, including the brain (11), liver (12), kidney (13), and skin (14,15). These improvements have facilitated deep-tissue visualization while minimizing phototoxicity and preserving physiological integrity (16), thereby allowing repetitive imaging of the same locations over time in living organisms (17). In parallel, improvements in fluorescent labeling techniques and genetically modified reporters have expanded the ability to distinguish between multiple immune-cell populations and simultaneously track their movements (18). Consequently, researchers can now monitor immune-cell migration, cell–cell interactions, and structural remodeling events over extended periods, thereby revealing immune processes that were previously difficult to observe using conventional imaging or ex vivo approaches.

The influence of IVM is particularly evident in studies of diverse immune-cell types, including neutrophils, monocytes, macrophages, T cells, and B cells (2). It has enabled the visualization of distinct cellular programs, ranging from the rapid chemotactic swarming of innate immune cells (5) to the stable synaptic contacts formed by adaptive lymphocytes (7), thereby revealing levels of complexity that remain invisible to static analyses. Beyond single-cell tracking, IVM has also allowed the observation of coordinated group behaviors and spatial organization within tissues, highlighting interactions that emerge only in intact biological environments (4,11). These findings have provided critical insights into the coordination of immune-cell activity to maintain tissue homeostasis and regulate inflammatory responses within structurally and functionally distinct tissue environments (19). Importantly, even within the same immune-cell population, individual cells can exhibit distinct patterns of migration, arrest, and cell–cell interactions, demonstrating that IVM can reveal dynamic behavioral states at the single-cell level.

Beyond advancing traditional immunology, IVM has also shown important therapeutic potential by enabling the direct visualization of immune dysregulation under pathological conditions, including infection (14), cancer (20), and sterile inflammation (21). Although IVM can effectively reveal how cells behave within living tissues, a major challenge is determining how these dynamic behaviors reflect distinct functional states at the single-cell level. Recent progress in IVM, fluorescent reporter systems, and quantitative image analysis has provided an important framework for linking in vivo cellular behavior with quantitative single-cell behavioral profiles. In this review, we summarize the recent advances in IVM, highlight methodological developments that have improved analytical depth, and outline future directions for image-based single-cell behavioral analyses of immune cells under both physiological and pathological conditions.

IVM OF NEUTROPHILS

Several neutrophil-enriched or neutrophil-targeting reporter mice have been used to visualize neutrophils in live murine models (Table 1). The Lyz2-enhanced green fluorescent protein (EGFP) reporter mouse, also known as LysM-EGFP, was generated by knock-in of EGFP into the endogenous Lyz2 locus to enable the labeling of myelomonocytic cells, with the highest expression observed in neutrophils and substantial expression in monocytes (22). The Ly6g-Cre-T2A-tdTomato reporter mouse primarily shows labeling of neutrophils, with no detectable expression in bone marrow progenitor populations, including granulocyte-monocyte progenitors (23). The S100A8 (MRP8)-Cre-ires/GFP reporter mouse also predominantly shows labeling of neutrophils with GFP (24,25). These reporter systems have enabled real-time tracking of neutrophil recruitment, vascular interactions, interstitial migration, and effector behaviors in vivo. However, their labeling specificity, signal intensity, and potential overlap with other myeloid populations differ among mouse lines. Thus, neutrophil behavior observed by IVM should be interpreted with careful consideration of reporter selection, labeling strategy, and imaging site, particularly when linking dynamic behaviors to specific neutrophil states or subsets.

Table 1. Reporter mouse lines used for intravital microscopy of immune cells.

Target gene/promoter	Fluorescence	Labeled cells	Reference
LysM	GFP	Myeloid cells (neutrophils + monocytes/macrophages)	(22)
Ly6G	tdTomato	Neutrophils	(23)
MRP8	GFP	Mostly neutrophils	(24,25)
CX3CR1	GFP	Mainly monocytes/macrophages in the periphery and microglia in the CNS	(77)
CSF1R	EGFP	Monocytes, macrophages, and dendritic cells	(78)
F4/80	Reporter required	Mature tissue-resident macrophages (Kupffer cells, Langerhans cells)	(79)
CCR2	RFP	Inflammatory monocytes (Ly6Chigh), some subsets of dendritic cells	(80)
FOXP3	GFP	Regulatory T cells	(108,109)
CD4	Reporter required	T-cell lineage (CD4 and CD8)	(110)
CD8a	Reporter required	Cytotoxic T cells	(111)
CD19	Reporter required	B-lymphocytes	(112,113)
AID (Aicda)	Reporter required	Germinal-center B cells	(114)
Blimp-1	GFP	Plasma cells	(115)

CNS, central nervous system; RFP, red fluorescent protein.

Recruitment and vascular interactions

On the inflamed endothelium, the classical pathway of circulating neutrophil recruitment involves rolling, slow rolling, arrest, crawling, and transmigration (26). Rolling is primarily initiated by the interactions between selectins and fucosylated ligands such as P-selectin glycoprotein ligand-1 (PSGL-1), E-selectin ligand-1, and CD44 (27). Slow rolling and adhesion are mediated by β2 integrins such as αLβ2 and αMβ2 through their interactions with endothelial ligands, mainly intercellular adhesion molecule (ICAM)-1 and ICAM-2 (28). Crawling is largely regulated by interactions between neutrophil αMβ2 and ICAM-1-enriched endothelial tricellular junctions (29,30). The subsequent transmigration depends on homophilic interactions between adhesion molecules expressed on both neutrophils and endothelial cells (31), together with heterophilic interactions between junctional adhesion molecules and neutrophil integrins (32).

IVM has been widely used to elucidate the in vivo mechanisms of neutrophil recruitment under physiological and pathological conditions. Two-photon IVM in MRP8-Cre × mTmG reporter mice showed that inhibition of a disintegrin and metalloproteinase domain-containing protein 8 (ADAM8) impaired intrapulmonary neutrophil trafficking, whereas pharmacological ADAM8 inhibition reduced pulmonary infiltration, attenuated lung injury, and improved bacterial containment (33). Similarly, IVM identified several molecular regulators of neutrophil-vascular interactions, including PECAM-1 and CD99 during alveolar transmigration (34), endoplasmic reticulum protein 72 (ERp72)-mediated Mac-1 activation during adhesion and crawling (35), mast cell-derived IL-17A during transpericyte migration (36), and L-selectin/C5a-C5aR1 signaling during neutrophil trafficking to draining lymph nodes (37). These studies demonstrate how IVM can be used to visualize the sequential steps in neutrophil recruitment, from vascular adhesion and crawling to transendothelial migration and subsequent tissue entry (Fig. 1).

Figure 1. Integrative overview of immune-cell dynamics revealed by intravital microscopy. This figure summarizes how intravital microscopy reveals the dynamic behaviors of immune-cell populations in living tissues, including motility, morphology/behavioral states, cellular interactions, and their potential functional implications. For neutrophils, IVM enables real-time visualization of the sequential steps of neutrophil recruitment, including rolling, adhesion, crawling, and transmigration. After tissue entry, neutrophils can be tracked as they migrate through the interstitial spaces and exhibit swarming behavior. These behaviors are accompanied by morphological changes and interactions with endothelial cells, platelets, and invading pathogens. These observations provide insights into host defense, inflammatory amplification, vascular dysfunction, and tissue injury. For monocytes and macrophages, IVM reveals vascular patrolling, recruitment, tissue positioning, morphological transitions, and interactions with endothelial cells, neutrophils, stromal cells, apoptotic cells, and pathogens. These dynamic behaviors are associated with microbial and apoptotic-cell clearance, resolution of inflammation, vascular repair, and tissue remodeling or fibrosis. For lymphocytes, IVM enables visualization of trafficking, arrest, extravasation, tissue surveillance, behavioral transitions from motile to engaged or activated states, and interactions with endothelial cells, antigen-presenting cells, tumor or infected cells, B cells, and regulatory T cells. These observations provide insights into cytotoxic activity, immune regulation, immune memory, and humoral immune responses. Overall, IVM links dynamic cellular behavior within tissue microenvironments to potential functional consequences in health and disease.

IVM also revealed disease-specific neutrophil recruitment patterns that contribute to vascular dysfunction and tissue injury. Four-dimensional confocal IVM in mice after transient middle cerebral artery occlusion showed that calcium release-activated channel regulator 2A enhances αMβ2^high migratory neutrophil behavior, thereby promoting neutrophil infiltration into the ischemic brain and contributing to brain injury (38). In permanent ischemic stroke, IVM further showed that motile neutrophils can obstruct brain capillaries, resulting in reversible capillary stalls and peri-ischemic flow stagnation (39). In other inflammatory settings, IVM revealed that aging exacerbates stroke-induced pulmonary intravascular neutrophil dysfunction (40), LPS-induced neuroinflammation promotes neutrophil recruitment and reverses transendothelial migration (41), senescent endothelial cells enhance (C-X-C) ligand 1 (CXCL1)-dependent neutrophil adhesion and vascular leakage (42), IL-36 signaling promotes neutrophil adhesion and coronary microvascular perfusion defects after myocardial ischemia-reperfusion injury (43), and CD11b inhibition improves microcirculatory flow in cisplatin-induced acute kidney injury (AKI) (44). Moreover, IVM studies in models of sickle cell disease (SCD), severe acute respiratory syndrome coronavirus 2 infection, and retinal uveitis as well as human-derived vascular models have demonstrated context- and organ-dependent neutrophil recruitment and adhesion associated with impaired perfusion, endothelial injury, inflammatory thrombosis, and vascular leakage (45,46,47,48). Overall, these findings suggest that neutrophil recruitment should be viewed not simply as a fixed sequence, but as a dynamic process shaped by the vasculature, inflammatory stimulus, disease stage, and local tissue architecture. However, many IVM studies continue to focus on representative recruitment events rather than quantitatively defining distinct recruitment states, limiting the scope to determine whether adhesion, crawling, or capillary stalling reflect a stable, functionally distinct state or a transient activation state. Addressing this limitation will require extending the analysis beyond vascular entry to include interstitial migration, swarming, and pathogen-directed behavior.

Interstitial migration, swarming, and pathogen interactions

After neutrophil recruitment and transmigration, IVM enabled the analysis of neutrophil behavior within interstitial and mucosal environments, including migration, swarming, and interactions with pathogens (Fig. 1). Neutrophil migration through the skin interstitium was visualized in LysM-Cre × mT/mG reporter mice (49). Parallel tracking of neutrophils and bacteria around transfemoral implants revealed neutrophil–bacteria contact, supporting the analysis of infection-associated swarming and pathogen-interaction dynamics (50). Confocal IVM of the colonic mucosa showed that biopsy-induced focal mucosal injury triggered robust neutrophil swarming, whereas dextran sodium sulfate-induced colitis led to neutrophil accumulation in the lamina propria, which was characterized by reduced motility and increased CXCR4 expression (51). In addition, spleen-derived type I interferons suppress ERK/leukotriene B4-dependent neutrophil swarming through interferon-α/β receptor signaling, thereby weakening neutrophil-mediated bacterial containment and promoting intracellular bacterial immune evasion (52). Helicobacter pylori infection of the gastric mucosa induces time-dependent neutrophil dynamics characterized by increased neutrophil influx and reduced motility during acute infection (53). IVM during Staphylococcus aureus infection showed that α-toxin impaired early neutrophil localization and clustering by disrupting neutrophil calcium signaling (54). Together, these studies showed that transmigrated neutrophils exhibit context-dependent migratory, swarming, and pathogen-interaction behaviors that can be resolved using IVM. However, altered motility or swarming should not be interpreted as protective because these behaviors may either promote microbial containment or amplify local tissue injury. Thus, IVM-based observations of migration and clustering should be interpreted together with functional outcomes, including pathogen clearance, tissue damage, and inflammatory amplification, as well as downstream effector functions, such as phagocytosis, ROS production, and neutrophil extracellular trap (NET) formation.

Effector functions: phagocytosis, ROS production, and NETosis

After recruitment, neutrophils perform their effector functions through phagocytosis, ROS production, and NET formation. They rapidly phagocytose opsonized microbes via phagocytic receptors, such as Fcγ receptors and C-type lectin receptors (55,56), and this step is followed by phagosome maturation and antimicrobial granule fusion (57,58), which contribute to microbial killing and tissue remodeling (59,60). This process is accompanied by the activation of the NADPH oxidase complex, which generates ROS, and myeloperoxidase (MPO)-dependent oxidative reactions that contribute to microbial killing (61,62,63). Additionally, neutrophils can release NETs, which are composed of extracellular chromatin structures containing neutrophil elastase, MPO, and other antimicrobial components (64). NET formation can be induced by inflammatory mediators and pathogen-associated molecular patterns and is supported by protein-arginine deiminase type-4-mediated histone citrullination, which allows neutrophils to immobilize pathogens within chromatin-based structures (65).

IVM has revealed how these effector functions occur in vivo under both inflammatory and infectious conditions. In the in vivo NETosis studies, lung IVM showed that platelet glycoprotein VI promotes platelet-neutrophil interactions, neutrophil adhesion/crawling, transmigration, and NETosis during early LPS-induced acute lung injury (66). IVM in SCD mice showed that neutrophils shed gasdermin D-dependent NETs in the liver microcirculation, which embolize the lungs and promote pulmonary vaso-occlusion and inflammatory lung injury (67). Additional IVM studies have linked NET formation to tissue injury in several inflammatory conditions, including reduced neutrophil migration and NET formation after lysophosphatidylcholine treatment in LPS-induced acute lung inflammation (68), perivascular NET release associated with microvasospasm and reduced pial arteriolar blood flow in subarachnoid hemorrhage-related cerebral ischemia (69), NOD-, LRR- and pyrin domain-containing protein 3-associated NETosis, IL-1β production, and blood–brain barrier disruption in LPS-induced neuroinflammation (70,71), and platelet-dependent NET release during severe influenza A infection (72).

IVM has also clarified how neutrophils engage pathogens through phagocytosis and direct cell–pathogen interactions. In cutaneous Leishmania major infections, Ly6G facilitates early neutrophil extravasation and pathogen engagement, potentially through β2-integrin-related interactions (73). In lethal Pseudomonas aeruginosa pneumonia, low-dose LPS pre-exposure functionally enhanced neutrophil phagocytosis and galectin-3-associated antibacterial activity (74). Multiphoton IVM during S. aureus skin infection revealed that the loss of ArlRS or MgrA increases neutrophil motility, infiltration into bacterial aggregates, and direct neutrophil–bacteria engagement in vivo (75). For ROS-dependent behavior, bone marrow IVM showed that CXCR4-guided neutrophils mechanically plucked megakaryocyte proplatelets through NADPH oxidase 2/ROS-dependent signaling, thereby enhancing platelet production in thromboinflammatory cardiovascular disease (76). Collectively, these studies show that IVM can place neutrophil effector functions within their spatial and temporal tissue contexts rather than treating them as isolated biochemical events. Nevertheless, effector readouts such as NETosis, ROS generation, or phagocytosis are often inferred from imaging markers and therefore require complementary molecular or functional validation.

IVM OF MONOCYTES AND MACROPHAGES

Several reporter mouse lines with labeled monocytes and macrophages have been used to visualize these cells in living murine models (Table 1). The CX3CR1-GFP reporter mouse, which is generated by inserting EGFP into the endogenous Cx3cr1 locus, labels peripheral monocytes/macrophages as well as microglia in the central nervous system (77). The CSF1R-EGFP reporter line has been widely used to visualize monocytes, macrophages, and dendritic cells (78), whereas the F4/80-Cre line is commonly used to label mature tissue-resident macrophages, including Kupffer and Langerhans cells (79). The CCR2-red fluorescent protein reporter mouse is particularly useful for detecting Ly6C^high inflammatory monocytes, although some dendritic cell subsets may also be labeled (80). Together, these reporter systems have enabled real-time intravital tracking of monocyte and macrophage dynamics, providing insights into their in vivo behaviors such as migration, phagocytosis, efferocytosis, and tissue remodeling. Because these reporter systems often label overlapping myeloid compartments, IVM-based interpretation of monocyte and macrophage behavior requires careful distinction of circulating monocytes, tissue-resident macrophages, dendritic cells, and microglia. This issue is particularly important in studies where behavioral features such as patrolling, tissue positioning, or phagocytosis are used to infer lineage identity, activation state, or functional specialization.

Dynamic positioning and intercellular communication of monocytes/macrophages

Monocytes and macrophages are highly plastic innate immune cells whose phenotypes and functions are continuously shaped by tissue-derived signals, inflammatory mediators, and cell–cell interactions (81). These cells have traditionally been described using the M1/M2 polarization model, wherein M1-like macrophages are related to pro-inflammatory responses and pathogen clearance (82), while M2-like macrophages are linked to inflammation resolution, tissue repair, and remodeling (83). However, this binary classification provides only a simplified view of macrophage activation and does not fully explain the dynamic and context-dependent behaviors of monocytes and macrophages within living tissues. In vivo, monocytes and macrophages participate in vascular surveillance, inflammatory recruitment, tissue localization, and intercellular communication. Therefore, this section focuses on the directly observable in vivo behaviors of monocytes and macrophages, including cell positioning, vascular patrolling, recruitment, and cell–cell contacts, rather than on major histocompatibility complex (MHC)-dependent Ag-presentation processes.

Previous IVM studies have revealed that monocytes and macrophages exhibit diverse in vivo behaviors, including vascular patrolling, endothelial adhesion, inflammatory recruitment, tissue positioning, and cell–cell interactions (Fig. 1). Glomerular multiphoton IVM has shown that monocyte surveillance of glomerular capillaries supports neutrophil retention and activation, particularly through classical monocytes, during acute inflammation (84). IVM has further clarified how monocyte adhesion and recruitment are regulated by inflammatory mechanisms in inflamed vessels. The myeloid mineralocorticoid receptor promotes the slow rolling of monocytes and firm adhesion to inflamed mesenteric vessels by transcriptionally regulating PSGL-1 expression (85). Similarly, NET-associated histone H2A enhances monocyte adhesion and contributes to endotoxemia-accelerated atherosclerosis (86). In myocardial infarction and chronic kidney disease (CKD), surface-expressed IL-1α on monocytes promotes endothelial adhesion and leukocyte recruitment, thereby linking monocyte activation to inflammatory tissue injury (87). Collectively, these studies demonstrate that monocyte-endothelial interactions are controlled by context-specific molecular cues and can contribute to vascular inflammation across multiple disease settings. These observations suggest that monocyte recruitment is not merely a route of tissue entry, but also an active process that can shape endothelial activation, neutrophil retention, and subsequent inflammatory amplification. However, vascular adhesion alone does not determine whether recruited monocytes acquire inflammatory, reparative, or regulatory roles. Therefore, tracking the tissue localization and intercellular communication is necessary to understand the functional consequences of monocyte recruitment.

In addition to characterizing vascular adhesion, IVM has revealed how monocytes and macrophage-lineage cells are repositioned within injured tissues and communicate with other immune or stromal cells. In carbon tetrachloride (CCl4)-induced liver fibrosis, splenic monocytes enhance hepatic CX3CR1⁺ cell motility and interactions, thereby promoting pro-inflammatory and profibrotic microenvironments (88). Consistently, IVM of acetaminophen-induced liver injury revealed persistent interactions among Ly6C⁺ monocytes, CD63⁺ endothelial cells, and Kupffer cells/macrophages, supporting a myeloid Mas-regulated inflammatory microenvironment (89). In irritant contact dermatitis, CX3CR1⁺Ly6C⁺ dermal monocytes accumulated around hair follicles and interacted with neutrophils, promoting transepidermal neutrophil migration (90). In corneal epithelial injury, spleen-derived Ly6C^high monocytes migrate to the injured corneal limbus, where sympathetic nerve-β2-adrenergic signaling regulates subsequent neutrophil recruitment (91). These studies highlighted the ability of monocytes and macrophages to occupy specific tissue niches and coordinate inflammatory cell recruitment through local intercellular communication.

IVM has also provided insights into the disease context-dependent positioning of monocytes, macrophages, and related myeloid populations during tissue injury, tissue repair, and cancer. In ischemia-reperfusion injury-induced AKI, entospletinib reduced monocyte recruitment and suppressed SYK-dependent CKD-associated macrophage activation, thereby resolving renal inflammation (92). CCR2-dependent monocyte recruitment is required for angiogenic vascular repair (93). In ischemic stroke, IVM revealed rapid microglial morphological changes, including enlarged cell bodies and dendrite loss, before substantial peripheral monocyte and neutrophil infiltration (94). In MB49 bladder cancer, IVM in CX3CR1-GFP mice showed that SYMPHONY therapy promotes the accumulation of CX3CR1⁺ monocyte/macrophage-lineage cells in distant tumors (95). Collectively, these observations suggest that IVM is particularly useful for defining the localization of monocytes and macrophage-lineage cells, their interactions with vascular and immune-cell compartments, and the dynamic regulation of these spatial and interactive behaviors during inflammation, vascular injury, fibrosis, stroke, and cancer. However, IVM alone often cannot determine whether these observed behaviors correspond to specific activation states, tissue origins, or molecularly distinct macrophage subsets, highlighting the need for complementary phenotypic and molecular analyses.

Phagocytosis, efferocytosis, and tissue remodeling

Macrophages support host defense by clearing microbes, cellular debris, and apoptotic cells through phagocytosis and efferocytosis (96). In addition to their clearance functions, macrophages regulate immune responses and tissue homeostasis by secreting cytokines and growth factors (97). Through this phenotypic plasticity, macrophages function as central regulators of tissue repair by coordinating the transition from inflammation to resolution as well as structural recovery (98). During the proliferative phase, macrophages often acquire repair-associated, M2-like features, leading to the secretion of anti-inflammatory mediators and growth factors that support angiogenesis and fibroblast proliferation (99). During the remodeling phase, macrophages maintain a precise equilibrium between ECM synthesis and degradation, facilitating the replacement of the provisional matrix with functional tissue (100). However, chronic or dysregulated tissue repair can promote pathological fibrosis because persistent profibrotic signaling drives excessive ECM accumulation and scarring (101). This framework highlights why macrophage behavior should be analyzed as a spatially and temporally organized process rather than as a fixed polarization state. IVM is particularly useful in this context because it can reveal when and where macrophages contact dying cells, pathogens, endothelial cells, or stromal compartments during the transition from inflammation to repair or fibrosis.

IVM studies have provided direct evidence that macrophage clearance contributes to the resolution of inflammation, antimicrobial defense, and tissue protection in vivo. In liver ischemia-reperfusion injury, IVM has shown that myeloid MAS signaling promotes MER proto-oncogene, tyrosine kinase-dependent macrophage efferocytosis of aged neutrophils, thereby attenuating neutrophil-driven liver inflammation (102). Additionally, in chronic liver injury, PD-L1 blockade restored the bacterial capture of F4/80⁺ liver macrophages (103). In urinary tract infection, CX3CR1⁺ suburothelial perivascular macrophages maintained vascular integrity, captured invading bacteria, and subsequently released macrophage extracellular traps (METs) that sequester pathogens within the urothelium, thereby preventing systemic dissemination (104). In the alveolar niche, monocyte-derived alveolar macrophages rapidly acquire alveolar migratory behavior and bacterial phagocytic capacity (105), and IVM further showed that alveolar macrophages patrol across alveoli via the pores of Kohn and rapidly capture inhaled bacteria (106). In addition to pathogen clearance and tissue protection, repair-associated monocyte/macrophage signaling can also promote tissue recovery. Mechanistically, Ly6C⁺ monocyte-derived IL-6 promotes VEGFA⁺ repair-associated microglia generation, supporting vascular repair and neurological recovery (107). Collectively, these studies demonstrate that IVM can directly visualize macrophage behaviors and functions that are difficult to infer from static tissue analysis, including bacterial capture, efferocytosis, MET release, alveolar patrolling behavior, and repair-associated interactions. However, phagocytosis, efferocytosis, and tissue remodeling are often context-dependent and may have protective or pathological consequences. Therefore, IVM-based observations of macrophage behavior should be interpreted together with functional outcomes, including microbial containment, inflammatory injury, vascular repair, fibrosis, and tissue recovery.

IVM OF LYMPHOCYTES

This section primarily focuses on IVM studies of lymphocyte dynamics in peripheral and diseased tissues, including tumors, infectious and autoimmune conditions, and tissue-resident memory T-cell niches. To visualize lymphocytes in living murine models, several T- and B-cell reporter or Cre-driver mouse lines have been used in combination with fluorescent reporter systems (Table 1). For T lymphocytes, the Foxp3-EGFP reporter mouse allows direct visualization of Tregs through endogenous Foxp3 expression (108,109). CD4-Cre is commonly used for genetic manipulation or fluorescent labeling of most T-cell lineages by driving Cre recombinase expression from the CD4 locus during thymic development (110). Similarly, E8I-Cre predominantly mediates Cre activity in mature CD8⁺ cytotoxic T cells through the CD8α enhancer E8I (111). In B-lymphocytes, CD19-Cre mediates B-cell–specific genetic recombination beginning at the early stages of B-cell development and can be used to label most B-cell lineages (112,113). Aicda-Cre primarily drives Cre activity in activated or germinal-center B cells, allowing lineage tracing or conditional gene manipulation during Ab diversification processes, including somatic hypermutation and class-switch recombination (114). The Prdm1-GFP reporter mouse enables visualization of Blimp-1-expressing plasma cells during plasma cell differentiation, since Blimp-1 is a key transcriptional regulator of terminal B-cell differentiation (115). Together, these reporter and Cre-driver systems have enabled real-time intravital tracking of lymphocyte dynamics, providing insights into T-cell trafficking, cytotoxic and regulatory interactions, B-cell migration, B–T-cell interactions, and germinal-center–associated behaviors in vivo (Fig. 1). In comparison with innate immune cells, lymphocyte behavior is strongly shaped by Ag specificity, clonal expansion, differentiation state, and tissue residency, which are not always directly resolved by imaging alone. Therefore, IVM studies of T and B cells should be interpreted together with information regarding Ag recognition, activation status, and tissue niche organization when defining functional lymphocyte states in vivo.

T-cell trafficking, effector engagement, and regulatory dynamics

After thymic selection, mature naïve T cells circulate among the blood, lymphatic system, and secondary lymphoid organs until Ag encounter initiates activation (116,117). Ag presentation by dendritic cells, followed by TCR engagement, co-stimulatory signaling, and cytokine-mediated instructions, drives T-cell activation, clonal expansion, and functional differentiation (118,119,120,121,122,123,124). Activated CD4⁺ T cells differentiate into functionally specialized helper or regulatory subsets, thereby coordinating cytotoxic immunity, B-cell responses, and immune regulation (125). In parallel, CD8⁺ T cells acquire cytotoxic effector programs that support trafficking into inflamed tissues and tumor microenvironments and enable Ag-specific killing through perforin- and granzyme-dependent mechanisms (126,127,128,129). After immune resolution, a subset of Ag-experienced T cells persists as long-lived memory T cells, supporting rapid recall responses and tissue surveillance upon Ag reencounter. IVM studies in secondary lymphoid organs have revealed T-cell migration, Ag surveillance, and dynamic interactions with dendritic cells during activation (6,130).

IVM has enabled direct visualization of T-cell trafficking and tissue entry across tumor, autoimmune, and infectious disease contexts. From the perspective of tumor immunity, this approach has provided valuable insights into how T cells access tumor tissues. In melanoma brain metastases, T cells primarily undergo extravasation at ICAM-1-enriched peritumoral venous vessels (PVVs), identifying PVVs as a spatially specialized entry route into the brain tumor microenvironment (131). Similarly, in solid tumor models, MECA-79⁺ tumor-associated high endothelial venules functioned as specialized vascular sites for T-cell tethering, rolling, arrest, crawling, and extravasation, thereby contributing to increased tumor-infiltrating lymphocyte accumulation (132). IVM has also clarified how T-cell trafficking contributes to inflammatory tissue responses. In collagen-induced arthritis, fingolimod (FTY720) reduced CD4⁺ T-lymphocyte adhesion and recruitment to the peripheral vessels of inflamed joints (133). Malaria-specific T cells accumulated in cerebral vessels, followed by their subsequent entry into the brain parenchyma (134). Within tumor tissues, IVM has allowed direct analysis of how tumor-infiltrating T cells behave within local tumor niches. In liver metastasis, IVM has captured prolonged CTL-tumor cell contacts, CTL-induced Ca²⁺ influx, and caspase-3-dependent tumor cell apoptosis, showing that tumor cell death often follows cumulative interactions with multiple CTLs (135). In B-cell lymphoma, IVM revealed heterogeneous outcomes of CTL-tumor cell encounters, wherein irreversible tumor cell killing occurred only in a small fraction of interactions (136). In pancreatic cancer, IL-18R signaling constrained intratumoral CTL motility and Ag-specific target engagement in the tumor microenvironment (137). In addition to characterizing cytotoxic interactions, IVM has revealed regulatory T-cell dynamics within tumors. In experimental glioblastoma, IVM revealed progressive FoxP3⁺ Treg accumulation in the tumor and peritumor regions (138). Collectively, these studies revealed that T-cell function in the tumor microenvironment is not determined solely by the presence of T cells, but also by dynamic processes such as CTL-tumor cell contact, cumulative cytotoxic engagement, migration and arrest behaviors, and the local accumulation of immunoregulatory T-cell subsets. IVM has further shown that anticancer therapies can change intratumoral T-cell dynamics: carboplatin promotes CD8⁺ T-cell trafficking in epithelial ovarian cancer (139); CTLA-4 blockade induces tumor-associated Treg expansion in colorectal cancer (140); and PD-1 blockade activates CD8⁺ T cells, while indirectly supporting Treg accumulation through an IL-2/inducible T-cell costimulator (ICOS)-dependent CD8⁺ T-cell–Treg crosstalk in melanoma (141). Further IVM studies have shown that invariant natural killer T cell-mediated antitumor activity can be enhanced by therapeutic activation in colorectal cancer liver metastasis (142). These therapeutic IVM studies have also indicated that increased T-cell accumulation is not sufficient to infer effective antitumor immunity. Functional outcomes depend on whether T cells establish functional contact with tumor cells, maintain cytotoxic activity, and overcome local regulatory circuits such as Treg-mediated suppression. Thus, IVM is particularly valuable when trafficking, contact duration, calcium signaling, and target-cell death are analyzed together rather than as separate observations.

IVM has also clarified Ag-dependent T-cell activation and memory-cell behavior in the context of autoimmune and infectious diseases. In experimental autoimmune encephalomyelitis, two-photon IVM revealed microbiota- and MHC II-dependent calcium signaling in encephalitogenic CD4⁺ T cells, linking intestinal T-cell activation to Th17-like polarization (143). In addition to acute activation, IVM has been used to visualize tissue-specific memory T-cell positioning and surveillance after infection. These studies showed CD8⁺ memory T-cell positioning within monocyte-rich foci in the spleen, a secondary lymphoid organ (144), motile surveillance by corneal tissue-resident memory T (Trm) cells (145), and Ag-dependent clustering of lung CD103⁺ Trm cells near reinfected sites (146). Overall, these findings show that IVM can visualize not only T-cell entry and cytotoxic engagement but also regulatory interactions, Ag-dependent activation, and memory T-cell surveillance within distinct tissue microenvironments. Importantly, T-cell behavior also shapes humoral immunity by regulating the spatial positioning and duration of helper interactions with B cells. Therefore, IVM studies of lymphocytes should not be limited to cytotoxic or regulatory T-cell responses but should also consider how T-cell positioning and dwell time influence B-cell activation and germinal-center organization.

B-cell migration, B–T-cell interactions, and humoral immune niches

B cells mature completely in the bone marrow before entering secondary lymphoid organs, where mature naïve B cells occupy follicular zones and detect native Ags through B-cell receptors (147,148). Upon Ag engagement, activated B cells internalize and process Ags before presenting Ag-derived peptides to helper T cells, which provide signals that promote their migration into the germinal centers (149). Within germinal centers, B cells proliferate and undergo Ab diversification under the guidance of T follicular helper cells (150,151). Selected B cells become Ab-secreting plasma cells that populate the lymphoid tissues and bone marrow to maintain humoral immunity (152,153). In parallel, a subset of selected cells become memory B cells, supporting long-term immune memory and rapid recall responses upon Ag re-exposure (154). IVM studies in secondary lymphoid organs have also revealed B-cell migration, B–T-cell interactions, and dynamic organization within germinal centers (155,156).

Intravital and high-resolution imaging studies have revealed the mechanism by which B cells migrate, interact with T cells, and occupy specialized lymphoid or vascular niches in vivo. In secondary lymphoid organs, lysine deficient protein kinase 1 regulates B-cell interactions with cognate CD4⁺ T cells, thereby supporting efficient B–T-cell interaction (157). In the dura mater, spatial imaging has identified dura-associated lymphoid tissue containing B- and T-cell aggregates that support Ag sampling, germinal-center B-cell formation, and local humoral immune responses (158). Beyond secondary lymphoid and lymphoid-like niches, imaging studies have also clarified how tissue-specific microenvironments regulate B-cell development and immune function. In the bone marrow, von Hippel Lindau deletion in Dmp1-expressing osteolineage cells has been shown to alter the vascular architecture and impair B-cell lymphopoiesis (159). In the lungs, transitional type 2 B cells marginate within pulmonary capillaries through CD49e and CXCL13-CXCR5 signaling, thereby restraining excessive neutrophil vascular inflammation through LXA4-mediated regulation (160). Collectively, these studies indicate that imaging approaches can define not only B-cell migration and B–T-cell interactions, but also the tissue niches that support B-cell development, humoral immunity, and vascular immune regulation. Accordingly, future studies combining intravital tracking with Ag specificity, clonal information, and Ab-production readouts are required to determine whether distinct B-cell behaviors represent functionally meaningful humoral states. This gap highlights the need for advanced imaging and computational approaches that can extend local observations to quantitative single-cell behavioral analyses.

ADVANCED APPROACHES FOR EXTENDING IVM AND IMAGE ANALYSIS

IVM is a powerful platform for real-time monitoring of the in vivo behavior of various cell types. However, its application is limited by its restricted imaging depth, limited field of view, and optical accessibility (1,2,16,17). Recent advances in multiphoton microscopy, including two-photon and three-photon imaging systems, have improved deep-tissue visualization in comparison with conventional confocal microscopy (9,16,161,162), although imaging depth varies with tissue type and optical properties, fluorophore characteristics, labeling density, and imaging or detection conditions (163,164). These advances have enabled imaging across the entire depth of mouse lymph nodes and in previously inaccessible regions of tumor and bone tissues, thereby expanding the range of organs and cellular behaviors that can be examined in vivo (165,166). Despite these technical advances, IVM generally visualizes cellular dynamics within defined local regions, such as imaging windows or exposed tissues, rather than across entire organs. To complement these local intravital observations, tissue clearing serves as an ex vivo approach that enables organ-scale spatial mapping beyond the optically accessible fields captured by IVM. By rendering intact tissues optically transparent while preserving their three-dimensional architecture, these clearing approaches enable the visualization of immune-cell distribution, vascular architecture, and lesion-wide tissue organization (167,168,169). When combined with light-sheet or other volumetric imaging platforms, tissue clearing can reveal whether the local behaviors visualized by IVM are reflected in broader spatial patterns of immune-cell accumulation or tissue organization (170,171). Thus, tissue clearing does not replace IVM but complements it by placing dynamic, locally observed immune-cell behaviors within a broader anatomical and pathological context. As summarized in Fig. 2, advances in fluorescent labeling, deep-tissue imaging, and quantitative analysis have expanded the technological and analytical scope of IVM.

Figure 2. Technological and analytical advances in intravital microscopy. This schematic summarizes recent methodological and analytical advances in intravital microscopy. Fluorescent labeling strategies, including fluorescent Abs and reporter mice, improve the simultaneous visualization of multiple cell populations in vivo. Advances in multiphoton IVM, including two-photon and three-photon microscopy, have increased imaging depth and extended IVM to deeper tissue regions, with the indicated imaging depths representing examples from mouse brain imaging. Overall, these advances support four-dimensional image acquisition in living tissues, enabling single-cell tracking and extraction of dynamic cellular features. Subsequent quantitative analyses, including dimensionality reduction, clustering, and heatmap-based profiling, allow the characterization of single-cell behavioral profiles and distinct behavioral states.

Recent advances in IVM have led to the development of single-cell behavioral analysis, which enables the generation of quantitative behavioral profiles in vivo (172). In this analysis, quantitative tracking of individual cells over time enables the conversion of dynamic features, including motility, dwell time, arrest behavior, directional persistence, morphological changes, spatial positioning, and cell–cell interactions, into measurable behavioral profiles. These imaging-derived profiles can then be analyzed using computational approaches, including dimensionality reduction and clustering, to define distinct behavioral states within inflamed or diseased tissues. Methodologically, automated analysis platforms such as ACME have further improved this approach by enabling unbiased segmentation and feature extraction from four-dimensional IVM datasets, thereby increasing the scale and reproducibility of single-cell behavior analysis (173). For example, four-dimensional live imaging combined with high-dimensional morphokinetic analysis generates behavior-based maps that can distinguish immune-cell populations and reveal injury-associated neutrophil behavioral states (172). Disease-focused studies have also shown that single-cell behavioral analyses can link dynamic immune-cell states to pathological outcomes. In ischemic stroke, four-dimensional confocal IVM with quantitative single-neutrophil tracking revealed highly migratory αMβ2^high neutrophil states associated with neutrophil brain infiltration (38). In inflammatory injury models, circadian regulation through CXCR4 signaling repositioned neutrophils within injured tissues, reducing collateral tissue damage and illustrating how spatial behavior can determine inflammatory outcomes (174). On a broader scale, open IVM resources, such as Immunemap, allow large datasets of immune-cell behaviors to be compared across various imaging conditions, supporting the standardized and reproducible analysis of immune-cell motility patterns (175). Overall, these studies demonstrate that IVM-based behavioral analysis can define behavioral states by integrating molecular marker expression with dynamic features, including how individual cells move, arrest, interact, reposition, and transition within living tissues, thereby generating quantitative single-cell behavioral profiles.

CONCLUDING REMARKS

IVM has enabled real-time visualization of cellular behavior within living tissues under physiological and pathological conditions, supporting a dynamic interpretation of immune function in tissue microenvironments (Fig. 1). Recent advances in multiphoton microscopy have expanded imaging depth and optical access, enabling repeated observation of immune-cell behavior within previously difficult-to-access tissues. In parallel, reporter and Cre-driver systems, together with multicolor labeling strategies, have improved the simultaneous tracking of distinct cell populations and their interactions in vivo. However, reporter specificity, signal overlap, and differences in labeling efficiency remain important considerations when distinguishing individual cell populations. Furthermore, quantitative single-cell behavioral analysis represents an important future direction, since it can convert dynamic behavioral features into quantitative single-cell behavioral profiles. Nevertheless, complementary molecular and functional validation remains necessary to determine the biological significance of these imaging-derived patterns. In the future, integrating IVM-based behavioral analyses with spatial, molecular, and functional outcome measurements may provide a more comprehensive framework for defining immune-cell heterogeneity and identifying dynamic, context-dependent immune behaviors that contribute to disease progression or therapeutic response.

Abbreviations

ADAM8

metalloproteinase domain-containing protein 8

AKI

acute kidney injury

CKD

chronic kidney disease

EGFP

enhanced green fluorescent protein

ICAM

intercellular adhesion molecule

IVM

intravital microscopy

MET

macrophage extracellular trap

MPO

myeloperoxidase

NET

neutrophil extracellular trap

PSGL-1

P-selectin glycoprotein ligand-1

PVV

peritumoral venous vessel

SCD

sickle cell disease

Trm

tissue-resident memory T