Abstract
Ischemia-reperfusion injury (IRI) in different organs is a common pathophysiological basis in a variety of clinical critical diseases, such as myocardial infarction, cerebral stroke, and organ transplantation. The early diagnosis and real-time monitoring of IRI is unequivocally of great significance in revealing the underlying pathogenesis, evaluating tissue activity, and improving prognosis. Near-infrared (NIR) fluorescence imaging technology has displayed great potential in investigating IRI via the manner of in vivo real-time visualization. As the cornerstone involved in such a technology, fluorescent probes play a crucial role therein, and their design strategy and performance dominantly determine the imaging specificity, sensitivity, and evaluation ability. This minireview aims to outline recent advances in research dedicated to the diagnostic applications of NIR fluorescent probes via IRI imaging, with a focus on the pathophysiological mechanism of IRI and the key biomarkers that generally act as the basis of probe design. Additionally, we classify and summarize the imaging applications of fluorescent probes according to their involvement in various IRI models of important organs, including heart, brain, lung, liver, kidney, and limbs, and give an insight into the key to design rationale and imaging performance of the probes. Finally, we present a viewpoint towards the key challenges in NIR fluorescence imaging for mapping IRI and the future research regarding developing more versatile and more efficient NIR fluorescent probes for IRI clinical imaging applications.
0 Introduction
Ischemia-reperfusion injury (IRI) is a complex pathological process in which tissues or organs, after blood flow interruption leading to hypoxia and insufficient supply of metabolic substances, fail to restore normal cell function even after blood flow is restored, and instead suffer more severe structural damage and physiological dysfunction [1]. As a key pathological mechanism common to many clinical diseases, IRI involves a biological cascade reaction involving multiple levels and multiple signaling pathways, and is commonly seen in various medical scenarios such as myocardial infarction, stroke, organ transplantation, resuscitation after shock, and peripheral vascular surgery [1-2]. Although re-establishing blood flow perfusion is an indispensable treatment measure to save ischemic tissues, the reperfusion process itself may trigger or aggravate cell and tissue damage, presenting a typical "double whammy" characteristic [3]. This pathological process is usually divided into two successive but different phases − the ischemic phase and the reperfusion phase. The two phases show significant differences in energy metabolism regulation, ion balance maintenance, oxidative stress response, and intracellular signal transduction, which together promote the further development of tissue damage.
In recent years, with the rapid advancement of molecular imaging technology, near-infrared (NIR) fluorescence imaging has shown broad application prospects in IRI research due to its unique advantages. Compared with traditional imaging methods, this technology has significant features such as high spatiotemporal resolution, real-time dynamic monitoring of living organisms, and non-invasiveness [4]. In particular, fluorescence imaging in the near-infrared region I (NIR-I, 650−900 nm) and the near-infrared region II (NIR-II, 1000−1700 nm) provides a feasible technical path for high-definition imaging of deep tissues in living organisms because the fluorescence signal has stronger tissue penetration ability and less autofluorescence interference and light scattering degree in biological tissues within this wavelength range [5]. As the key core of this imaging technology, the performance of the fluorescent probe directly affects the sensitivity and accuracy of imaging. Therefore, designing and constructing efficient NIR fluorescent probes that can target and identify IRI-related specific biomarkers is of great significance for promoting early detection of IRI, pathological mechanism analysis, and treatment effect evaluation.
This review systematically summarizes the research progress of NIR fluorescent probes in imaging applications in the field of interstitial liver injury (IRI) , based on the fundamental pathological mechanisms of IRI. First, it comprehensively analyzes the key pathophysiological changes and related specific biomarkers in the IRI process (see Figure1) , providing a basis for the rational design of probes. Based on this, NIR fluorescent probes are divided into two main categories: non-responsive and responsive, and are analyzed and compared in depth from multiple dimensions, including molecular design principles, signal response mechanisms, and actual imaging performance. Finally, considering the current bottlenecks in research, it proposes forward-looking thoughts and suggestions on future technological optimization directions and potential application prospects, aiming to provide useful references for the development of IRI-related probes and imaging research.
Fig.1Types of IRIs and their pathological progress.
1 . Analysis of the pathological progression of IRI
1.1 Pathophysiological changes during ischemic period
This process is accompanied by the production and accumulation of a large amount of lactic acid, which leads to acidification of the intracellular environment, resulting in a decrease in pH value and inducing metabolic acidosis [6]. This acidic condition not only inhibits the activity of a variety of key enzymes, but may also interfere with the normal function of ion channels and exacerbate the imbalance of cell homeostasis. In addition, as the ATP level continues to decline, the activity of Na+/K+-ATPase, which depends on it for energy, is severely inhibited, causing cell membrane potential disorder and abnormal accumulation of sodium ions (Na+) in the cell. The resulting osmotic pressure change drives a large amount of water to enter the cell, causing cell swelling, which may eventually lead to cell structure rupture, i.e., the typical phenomenon of "cell edema" [7]. Meanwhile, under the condition of weakened transmembrane Na+ concentration gradient, the Na+/Ca2+ exchanger (NCX) changes from the normalcalcium efflux mode to reverse operation, which promotes abnormal influx of calcium ions (Ca2+) and becomes an early cause of intracellular calcium overload [8]. Although this process is relatively mild in the early stage of ischemia, it lays the groundwork for more severe calcium homeostasis disorder in the reperfusion stage. At the subcellular level, mitochondrial function is also significantly impaired [9]. Decreased electron transport chain (ETC) activity leads to gradual depolarization of mitochondrial membrane potential (ΔΨm) , further weakening ATP regeneration potential and increasing the possibility of opening mitochondrial permeability transition pore (mPTP) [10]. In addition, due to insufficient energy supply, cytoskeletal proteins lose stable support, microtubule and microfilament networks depolymerize, cell morphology maintenance ability decreases, cell membrane integrity is damaged, thereby accelerating the progression to irreversible damage.
1.2 Pathophysiological changes during reperfusion
Although the initial purpose of reperfusion is to restore the oxygen supply and metabolic homeostasis of the tissue, the process itself can induce a series of more severe secondary injuries. This phenomenon is called the“reperfusion paradox” [11]. Among them, the rapid outbreak of reactive oxygen species (ROS) is considered to be one of the key driving factors of IRI[10]. The main sources of ROS include: leakage of the mitochondrial electron transport chain due to abnormal electron transport in the early stage of oxygen restoration; after xanthine dehydrogenase is converted into xanthine oxidase, a large amount of superoxide anion () is generated in the process of catalyzing uric acid production; and the NADPH oxidase (NOX) family is activated and continuously generates active free radicals [12]. These highly reactive components accumulate rapidly, exceeding the clearance capacity of the cell's endogenous antioxidant defense system (such as superoxide dismutase, glutathione peroxidase, etc.) , thereby triggering a severe oxidative stress state. Excessive ROS can attack a wide range of biological macromolecules such as lipids, proteins and DNA, leading to lipid peroxidation of cell membranes, especially mitochondrial membranes, causing abnormal protein structure and function, and DNA strand breaks, which seriouslydamages the integrity and physiological function of cells [13].
More importantly, ROS can also act as an important signaling mediator, activating a variety of pro-inflammatory and pro-apoptotic signaling pathways [14]. For example, it can oxidatively modify IκB protein, causing it to degrade and release nuclear factor κB (NF-κB) , which then enters the cell nucleus, initiates the gene expression of various inflammatory factors such as TNF-α, IL-1β and IL-6, and triggers local inflammatory responses [15]. At the same time, damaged vascular endothelial cells upregulate the expression of adhesion molecules such as ICAM-1, VCAM-1 and selectins, promote the adhesion, migration and tissue infiltration of neutrophils, release proteases and additional ROS, form a self-amplified positive feedback loop, and further expand the extent of tissue damage [16-17].
In addition, the imbalance of calcium homeostasis during reperfusion also significantly aggravated cell damage. ROS-induced lipid peroxidation of cell membrane and mitochondrial membrane disrupts membrane integrity, leading to abnormal opening of voltage-gated calcium channels, receptor-operated calcium channels and storage-operated calcium channels, which promotes a large influx of extracellular Ca2+ [18]. At the same time, the calcium stores in the endoplasmic reticulum release excessive Ca2+ due to increased sensitivity to ryanodine receptors (RyRs) and inositol triphosphate receptors (IP3Rs) , causing a sharp increase in intracytoplasmic calcium concentration [19]. High levels of Ca2+ further activate calcium-dependent proteases such as calpain, phospholipase A2 and endonucleases, triggering cytoskeleton disintegration, membrane structure rupture and chromatin breakage, ultimately driving cells toward programmed or necrotic death.
Microcirculatory dysfunction is one of the key features that cannot be ignored in reperfusion injury. Although blood flow to large vessels is restored, some tissue areas still cannot achieve effective blood perfusion due to endothelial cell damage, microthrombus formation, and leukocyte adhesion obstructing the microvascular lumen. This phenomenon is clinically referred to as "no-reflow" [20-21]. This phenomenonseriously weakens the actual efficacy of reperfusion therapy and significantly affects the clinical prognosis of patients.
In terms of cell death mechanisms, IRI involves the joint participation of multiple programmed cell death pathways. In addition to traditional apoptosis and necrosis, recent studies have found that pyroptosis − a cell death mechanism mediated by caspase-1/4/5/11 activation, accompanied by the release of a large number of pro-inflammatory factors and leading to cell rupture − plays an important role in the body's innate immune response [22]. Of particular note is ferroptosis, a newly recognized form of regulated cell death, which, due to its dependence on iron ion catalysis and its core characteristic of abnormal accumulation of lipid peroxides, has been shown to play a key regulatory role in IRI models of multiple organs such as the heart, brain, liver, and kidney [23]. Ferroptosis is mainly related to the decrease in glutathione peroxidase4 (GPX4) activity, the inhibition of the function of the cysteine/glutamate reverse transport system (System Xc-) , and the excessive oxidation of polyunsaturated fatty acids, and has now become an important potential target for intervening in the pathological process of IRI [24].
1.3 Key biomarker system
The aforementioned complex pathophysiological processes are accompanied by a series of dynamic biomolecular events, resulting in a wealth of biomarkers that can be used for detection and imaging. These biomarkers cover multiple dimensions, including but not limited to: reactive oxygen species (, H2O2, ONOO, HClO, HOBr, ·OH) [25], sulfur-containing signal molecules (GSH, Cys, Hcy, H2S, H2Sn) [26], gaseous signal molecules (NO, CO, SO2) [27–29], metal ions ( Fe2+, Ca2+) , pH value [30], viscosity [31], and redox state related indicators.
Highly selective and highly sensitive fluorescent probes designed based on these molecular characteristics can not only enable real-time monitoring of IRI processes under in vivo or in vitro conditions, but also provide strong evidence and technical support for mechanism analysis, efficacy evaluation, and new drug development. It isworth noting that IRI in different organs has its own unique biomarker expression profile. For example, the production of mitochondrial ROS is particularly significant in myocardial IRI [32], the changes in excitatory amino acids and neurotransmitters are more prominent in brain IRI [33], and the disorder of the cytochrome P450 system in liver IRI is characteristic [34]. This organ specificity provides an important basis for the development of targeted probes.
2 Diagnostic methods and techniques for IRI
Since IRI has a significant impact on the progression of tissue damage and the clinical prognosis of patients, its early detection and precise intervention are of particular importance in clinical practice. However, currently used imaging techniques, such as computed tomography (CT) and magnetic resonance imaging (MRI) , mainly rely on changes in anatomical structures for judgment. They often lack sufficient sensitivity and specificity for molecular-level changes in the early stage of IRI that have not yet shown obvious morphological abnormalities, such as the initiation of oxidative stress, the release of inflammatory factors, or fluctuations in cell membrane potential. At the same time, due to the limitation of temporal resolution, these methods are difficult to achieve continuous and dynamic monitoring [35]. On the other hand, although tissue biopsy can provide pathological evidence to a certain extent, its invasive operation may aggravate local tissue damage, and due to the limitations of sampling sites and tissue heterogeneity, it is difficult to comprehensively and accurately reflect the overall dynamics of the lesion [36]. Therefore, the development of in vivo detection methods with high spatiotemporal resolution, non-invasive characteristics, and the ability to track the dynamic evolution of key biomarkers in real time has become a core requirement for promoting the translation of basic research on IRI into clinical practice.
In recent years, optical molecular imaging technology has provided strong technical support for addressing the above challenges [37]. This technology has several significant advantages: (1) It has extremely high detection sensitivity and good real-time performance, and can capture the minute changes of target components in themicroenvironment to achieve dynamic visualization of physiological and pathological processes. (2) It has excellent spatial resolution and can accurately target specific subcellular structures (such as mitochondria, lysosomes, and Golgi apparatus) for localization imaging. (3) It is a non-invasive detection method, suitable for long-term tracing of live cells, isolated tissues, and live animal models. (4) It has flexible and diverse probe design: Through reasonable molecular structure regulation, highly selective response systems can be constructed for specific biomarkers (such as reactive oxygen species/sulfur species, metal ions, enzymes, gaseous signal molecules, etc.) , which significantly improves the targeting and accuracy of detection.
Fluorescence imaging techniques in the NIR-I and NIR-II windows have attracted much attention due to their excellent optical properties. Light in this band exhibits low scattering and strong penetration in biological tissues, enabling imaging of deep tissues at the millimeter or even centimeter level. At the same time, since the autofluorescence interference of organisms in this region is significantly reduced, the background signal is effectively suppressed, thereby significantly improving the signal-to-noise ratio and resolution of the image. These advantages provide important technical support for the dynamic and clear visualization of IRI-related pathological processes under in vivo conditions [38].
Against this backdrop, NIR fluorescent probes, as the core medium connecting molecular events and optical signals, directly determine the sensitivity, selectivity, and quantification capabilities of imaging through their design strategies and performance. An ideal IRI imaging probe should possess the following characteristics: (1) high molar extinction coefficient and fluorescence quantum yield; (2) good photostability and chemical stability; (3) suitable biodistribution and metabolic properties; (4) high selectivity and sensitivity to the target biomarker; (5) good biocompatibility and low toxicity. Current research is moving towards multifunctionality, intelligence, and clinical translation.
3 Classification of NIR fluorescent probes used for IRI imaging and diagnosis
Based on their response mechanisms, modes of action, and signal output characteristics to pathological microenvironment stimuli, NIR fluorescent probes can be systematically classified into two main types: non-responsive probes and responsive probes. These two types differ significantly in their molecular design principles, in vivo distribution behavior, imaging functional guidance, and clinical translational potential. However, in the research and monitoring of IRI, both types of probes leverage their respective strengths, complement each other, and together construct an important technical system for achieving dynamic visualization of disease processes.
3.1 Non-responsive fluorescent probes
These probes are not metabolically dependent on specific biochemical reactions or pathological factors. Their optical properties (such as excitation and emission wavelengths, fluorescence intensity, quantum yield, and photostability) exhibit good stability in various physiological and pathological environments, thus they are classified as "structure-guided" or "passively targeted" imaging agents. Their primary function is to visualize macroscopic anatomical structures and physiological changes at the tissue level, rather than capturing dynamic biological events at the molecular level. Typical applications include assessing blood perfusion, changes in microvascular permeability, lymphatic drainage capacity, and the degree of extracellular space expansion.
Normally, these probes are injected into the body via the vein and distributed to various tissues through blood circulation, and they can be locally aggregated by utilizing the unique vascular structure characteristics of the lesion area. This process mainly relies on the enhanced permeability and retention effect (EPR effect) [39]. During the IRI process, the ischemic stage is caused by energy deficiency and oxidative stress, resulting in endothelial cell damage and degradation of junction proteins (such as VE-cadherin and tight junction-related proteins) . After reperfusion, the inflammatoryresponse is further aggravated, and a large amount of pro-inflammatory factors such as TNF-α and IL-1β are released, which aggravates the destruction of the vascular endothelial barrier and significantly increases the permeability of capillaries [40]. At the same time, the lymphatic drainage function in the damaged area is restricted, forming a microenvironment state of "high exudation and low clearance", which makes it easy for nanoscale probes or macromolecular fluorescent markers to leak from the structurally abnormal blood vessel wall and accumulate in the lesion site for a long time, thus showing a signal contrast that is significantly higher than that of normal tissue in imaging.
To improve the spatial resolution and targeting accuracy of imaging, some non-responsive probes can be chemically modified and covalently linked to specific targeting ligands, thereby transforming them into "actively targeted" probes to achieve specific recognition of the damaged area. Commonly used targeting molecules include monoclonal antibodies (such as anti-ICAM-1, anti-VCAM-1, and anti-P-selectin antibodies) , which can selectively bind to adhesion molecules that are significantly upregulated on the surface of activated endothelial cells [41]; functional peptides such as cyclic arginine-glycine-aspartic acid (cRGD) peptides can target integrin αvβ3 receptors that are highly expressed during inflammatory responses and tissue repair [42]; in addition, small molecule ligands such as folic acid can be used to recognize folic acid receptors overexpressed on the surface of certain injury-related macrophages [43]. Through the above-mentioned ligand-mediated active targeting mechanism, not only is the selective enrichment ability of probes at the lesion site significantly enhanced and their residence time in the target area prolonged, but the contrast and biological specificity of imaging signals are also greatly improved. Therefore, these probes demonstrate superior performance in depicting the spatial distribution of IRI lesions, clearly defining lesion boundaries and their junction with normal tissue, and show broad application prospects in clinical application scenarios such as intraoperative real-time navigation, ischemia range assessment, and dynamic monitoring of treatment effects.
3.2 Responsive fluorescent probes
Responsive fluorescent probes are a class of functional imaging agents constructed based on molecular engineering and chemical biology principles. Their core characteristic lies in their ability to selectively respond to changes in biochemical or physical parameters in specific pathological microenvironments and convert this stimulus into a detectable optical signal output. These probes typically contain three key components − a recognition moiety, a linker/spacer, and a fluorescent reporter group − which, through ingenious design, achieve response modes of "signal silencing-activation release" or "wavelength modulation-ratio output".
In the complex pathological process of IRI, responsive probes can specifically identify a variety of key biomarkers, including:
(1) Oxidizing species: In the process of IRI, the abnormal accumulation of oxidizing components is one of the core pathological mechanisms leading to tissue damage. These substances mainly include reactive oxygen species (ROS) and reactive nitrogen species (RNS) . ROS such as superoxide anion () , hydrogen peroxide (H2O2) , hydroxyl radical (•OH) , hypochlorous acid (HClO) , and lipid peroxide radical (LOO•) mainly originate from electron leakage in the mitochondrial electron transport chain, activation of xanthine oxidase, and upregulation of NADPH oxidase (NOX) . can rapidly combine with nitric oxide (NO) to generate highly cytotoxic peroxynitrite (ONOO-) ; while H2O2, catalyzed by ferrous ions (Fe2+) , is converted into highly destructive •OH via the Fenton reaction, leading to DNA breakage, protein oxidative modification, and lipid peroxidation. Reactive amino acids (RNS) are mainly represented by NO, nitrogen dioxide (NO2) , and ONOO-, especially ONOO-, which can induce nitration of protein tyrosine residues, interfering with normal mitochondrial function and promoting the activation of cell death pathways. Furthermore, under free radical attack, polyunsaturated fatty acids undergo a chain reaction of lipid peroxidation, producing cytotoxic secondary metabolites such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) . These substances not only disrupt the integrity of cell membrane structures but can also act as signaling molecules to activate inflammatoryresponses and programmed cell death pathways. This demonstrates that oxidizing species not only directly attack key biomolecules such as nucleic acids, proteins, and lipids, but also play a regulatory role in cell signal transduction, deeply participating in the development of IRI. Therefore, systematically monitoring their dynamic changes is of great value for a deeper understanding of disease mechanisms and the development of targeted intervention strategies.
(2) Reducing species: Reducing substances play a crucial role in maintaining cellular redox balance. Major endogenous reducing molecules include reduced glutathione (GSH) , thioredoxin (Trx) , hydrogen sulfide (H2S) and polysulfides (H2Sn) , cysteine (Cys) , homocysteine (Hcy) , and ascorbic acid, which exert cellular protective effects by directly scavenging free radicals, participating in antioxidant enzyme reactions, or regulating signal transduction pathways. During IRI, mitochondrial dysfunction and metabolic disorders disrupt the intracellular reducing environment, manifested as a decreased GSH/GSSG ratio, reduced Trx system activity, and decreased expression of H2S synthesis-related enzymes (such as CBS and CSE) , resulting in a significant weakening of overall antioxidant capacity. This decline in reducing power not only reduces the efficiency of ROS scavenging but may also exacerbate lipid peroxidation, protein damage, and DNA oxidation, thereby promoting the initiation of programmed cell death pathways such as ferroptosis and apoptosis. Therefore, reducing species are not only a core component in maintaining redox balance, but can also serve as an important biomarker for assessing the severity of IRI.
(3) pH changes: Dynamic fluctuations in tissue pH are an important physiological indicator reflecting metabolic disorders. In the early stages of ischemia, due to impaired oxidative phosphorylation, cells turn to anaerobic glycolysis, producing large amounts of lactic acid. Simultaneously, ATP hydrolysis releases H+, leading to intracellular acidosis. To cope with this change, the Na+/H+ exchanger (NHE1) is activated to expel protons, but due to weakened extracellular buffering capacity, an abnormal pH distribution pattern of "intracellular acidity and relatively alkaline extracellularity" is formed. During reperfusion, as blood circulation recovers, extracellular H+ is rapidlycleared, while NHE1 remains active, triggering intracellular alkalization, causing a reversal of the transmembrane pH gradient, which in turn promotes Ca²+ influx and calcium overload. In addition, an alkaline environment can increase the probability of opening the mitochondrial permeability transition pore (mPTP) and accelerate lipid peroxidation, thereby promoting the development of programmed cell death pathways such as ferroptosis. pH variation patterns differ across organs. pH fluctuations in myocardial tissue are closely related to arrhythmias, while in brain tissue they can inhibit neuronal electrical activity. This demonstrates that pH not only serves as an indicator of metabolic state but also directly participates in regulating various key injury mechanisms. Thanks to advancements in NIR fluorescent probe technology, pH-responsive probes constructed based on intramolecular charge transfer (ICT) or excited-state intramolecular proton transfer (ESIPT) mechanisms have enabled high-resolution ratio imaging of in vivo pH changes, providing an effective technical means for dynamic monitoring of IRI and evaluation of intervention strategies.
(4) Fluctuations in metal ion concentration: Abnormal fluctuations in intracellular metal ion concentration play a key driving role in oxidative stress and cell death processes, with imbalances in calcium (Ca²+) and iron (Fe²+/Fe³+) ions being particularly significant. In the early stages of ischemia, reduced ATP synthesis leads to decreased activity of energy-dependent Ca²+-ATPase, resulting in the inability of cytoplasmic Ca²+ to be effectively pumped out of the cell or stored in the endoplasmic reticulum, causing a gradual accumulation of intracellular Ca²+. During reperfusion, the generation of large amounts of reactive oxygen species (ROS) further damages the cell membrane and mitochondrial membrane structures, prompting abnormal opening of voltage-gated, receptor-regulated, and storage-operated calcium channels, causing a rapid influx of Ca²+ and calcium overload. The continuously elevated Ca²+ level can activate downstream effector molecules such as calpapsin, phospholipase A2, and endonucleases, triggering cytoskeleton disintegration, membrane system damage, and DNA breakage, ultimately driving cell necrosis or apoptosis. Meanwhile, the increase of free ferrous ions (Fe²+) can catalyze the conversion of H2O2 into highly reactivehydroxyl radicals (•OH) through the Fenton reaction, significantly exacerbating lipid peroxidation and triggering ferroptosis. During ischemia-reperfusion injury (IRI) , iron metabolism disorders manifest as accelerated ferritin degradation, upregulation of transferrin receptor expression, and decreased ferroportin levels, all contributing to the expansion of the intracellular free iron pool. Furthermore, zinc ions (Zn²+) and copper ions (Cu+/Cu²+) also participate in regulating redox states and the function of various enzymes, and may synergistically enhance oxidative damage under specific conditions. Therefore, metal ions are not only important cell signaling mediators but also serve as core hubs in oxidative damage and programmed cell death pathways, playing multiple pathological roles in ischemia-reperfusion injury, thus becoming key molecular targets requiring dynamic monitoring and targeted intervention.
(5) Changes in enzyme activity: During IRI, the activities of multiple key enzymes undergo significant and dynamic changes, deeply involved in core pathological mechanisms such as inflammatory response, oxidative stress, and cell death. During reperfusion, neutrophils and macrophages infiltrate damaged tissues, releasing myeloperoxidase (MPO) , which utilizes H2O2 and chloride ions to catalyze the generation of highly reactive hypochlorous acid (HClO) , thereby exacerbating protein oxidative modification and lipid peroxidation damage. At the same time, mitochondrial dysfunction prompts the release of cytochrome c into the cytoplasm, activating the caspase cascade reaction, in which caspase-3, as a key executive enzyme, is responsible for mediating the final implementation of DNA breakage and apoptosis. In addition, decreased lysosomal membrane stability leads to the leakage of cathepsin B and L (cathepsin B/L) , promoting mitochondrial outer membrane permeability in the local acidic environment, further amplifying cell damage signals. On the other hand, inducible nitric oxide synthase (iNOS) is highly expressed under the stimulation of inflammatory factors, continuously producing NO. This NO rapidly combines with superoxide anions () to generate highly oxidizing peroxynitrite (ONOO-) , triggering widespread protein nitration modifications and interfering with energy metabolism, leading to cellular dysfunction. Changes in the activity of these enzymesnot only directly reflect the progression of tissue damage but also possess high pathological specificity, thus serving as potential biomarkers and providing important evidence for real-time monitoring of intraepithelial neoplasia (IRI) , disease assessment, and evaluation of treatment efficacy.
(6) Physical parameters of the microenvironment: Viscosity, polarity, and mechanical tension, among other physical microenvironment parameters, have a significant impact on IRI. Ischemia induces energy metabolism disorders, leading to massive ATP consumption, damage to mitochondrial structure, and consequently, cytoskeleton rearrangement and slowed cytoplasmic flow, resulting in increased local microenvironment viscosity. High viscosity restricts the free diffusion of molecules, weakens intracellular signal transduction efficiency, and promotes the aggregation and deposition of misfolded proteins. Simultaneously, decreased fluidity of the cell membrane lipid bilayer and the occurrence of membrane phase transitions cause membrane structure to become rigid, interfering with the normal function of membrane-related proteins. At the polarity level, due to intensified lipid peroxidation and disruption of membrane integrity, the dielectric properties inside and outside the cell membrane become disordered, affecting the maintenance of transmembrane potential and the regulatory activity of ion channels. During the reperfusion phase, accompanied by enhanced inflammatory response, tissue edema occurs, interstitial pressure increases, and abnormal mechanical stress is generated, which can activate mechanosensitive ion channels such as Piezo1/2, induce Ca2+influx, and trigger calcium overload and related inflammatory signal cascades. In recent years, near-infrared (NIR) fluorescent probes constructed based on molecular rotor or fluorescence resonance energy transfer (FRET) mechanisms have enabled real-time, dynamic imaging monitoring of the viscosity, polarity, and mechanical properties of the in vivo microenvironment [44-45]. The visualization of these physical parameters not only helps to deepen the understanding of the pathological role of non-chemical microenvironmental disturbances during IRI, but also provides a theoretical basis and technical support for developing novel therapeutic strategies that target and regulate the physical microenvironment.
In summary, non-responsive probes focus on providing anatomical and physiological information, suitable for early hemodynamic assessment and structural localization; while responsive probes focus on revealing pathological events at the molecular level, enabling in-situ, real-time, and dynamic monitoring of key processes such as oxidative stress, calcium homeostasis imbalance, inflammatory activation, and programmed cell death. The development of NIR fluorescent probes is gradually driving a paradigm shift in IRI from "morphological observation" to "functional analysis" and even "mechanism elucidation". This review will systematically summarize the research progress in recent years in this field, focusing on probe design principles, biocompatibility optimization, multimodal integration strategies, and their application in animal models for IRI monitoring of important organs such as the myocardium, brain, liver, and kidneys, based on the above classification system. It will also discuss current challenges, such as long-term toxicity assessment, clinical translation pathways, and the establishment of standardized imaging protocols, aiming to provide theoretical basis and directional guidance for the future development of high-performance probes and their application in precision medicine.
4 Application of non-responsive NIR fluorescent probes in IRI imaging
Non-responsive probes have stable fluorescence signals, and their enrichment in damaged areas mainly depends on physical distribution or specific binding.
4.1 Non-targeted probes
These probes are typically small molecule dyes or nanoparticles that are enriched using the EPR effect unique to the IRI region. IRI can cause widening of the microvascular endothelial space, increased vascular permeability, and lymphatic drainage obstruction, thereby allowing nanoprobes of a specific size (usually 10–200 nm) to selectively extravasate and remain in the interstitial space.
4.1.1 Application in cardiac IRI imaging
Ji J J et al. [46] developed a novel NIR-II phototriggered dynamic covalent hydrogelsystem to address the limitations of current miRNA delivery systems. This system co-encapsulates miR-196c-3p mimics and BTN nanoparticles with photothermal conversion properties (see Figure2 (a) ) for cardiac IRI imaging. Utilizing the NIR-II fluorescence imaging capability of BTN, real-time non-invasive monitoring of the drug release process was achieved under 1064 nm laser excitation. This system provides a new, spatiotemporally controllable, non-invasive drug delivery and real-time monitoring scheme for treatment strategies such as improving cardiac function, reducing infarct size, and increasing survival rate in rats.
4.1.2 Application in brain IRI imaging
Gong X T et al. [47] first developed a near-infrared emitting semiconductor polymer nanoparticle (NESPN) (see Figure2 (b) ) and used it as a two-photon fluorescent probe for real-time high-resolution imaging of brain IRI. In view of the problems of limited sensitivity, insufficient penetration depth and easy photobleaching of organic dyes in existing stroke imaging technology, the research team prepared NESPN with a particle size of about 60 nm and a fluorescence quantum yield of 4.56% based on semiconductor polymers with donor-acceptor structure. The probe showed excellent two-photon absorption performance and photostability under 890 nm femtosecond laser excitation, and its fluorescence intensity decayed significantly less than that of green fluorescent protein under continuous irradiation. Experiments confirmed that NESPN can clearly display the microvascular structure of live mouse brain (resolution reaches the micrometer scale) and support continuous vascular imaging for up to 6 h. The probe has good neurocompatibility and has no significant effect on neuronal morphology and dendritic spine density. Further studies showed that NESPN can be used to dynamically monitor cerebral blood flow velocity and accurately locate the penumbra region in the cerebral ischemia model. Most importantly, the technology successfully achieved real-time visualization of the entire process of cerebral ischemia and reperfusion, accurately capturing the dynamic changes in blood flow interruption and recovery.
Bian H et al. [48] reported a novel high-performance NIR-II fluorescent dye family BM dye synthesized by a simple two-step method (see Figure2 (c) ) , whose representativemolecule BM3 has made significant breakthroughs in brightness and stability. In response to the problems of low brightness and complex synthesis of existing NIR-II fluorophores, the research team designed BM dyes with a rigid planar structure based on commercial raw materials. Among them, the molar extinction coefficient of BM3 in DMSO is 3.7×105 M-1·cm-1, and the fluorescence quantum yield is as high as 18.4%. Theoretical calculations show that its excellent performance is due to high oscillator strength, narrow absorption half-width and low electron-phonon coupling effect. The dye exhibits excellent photostability and chemical stability in physiological environment, and its aggregation behavior in aqueous phase can be effectively inhibited by surfactants. In in vivo imaging, BM3 can achieve high-resolution lymphatic vessel imaging at extremely low doses (75 pmol) and accurately distinguish between two lymphatic inflammation patterns induced by turpentine and Staphylococcus aureus. In the brain IRI model, BM3 penetrates the intact skull, clearly delineates the microvascular structure of the brain, and monitors the blockage and partial recovery of blood flow during ischemia and reperfusion in real time, revealing microcirculatory damage details that are difficult to capture by traditional imaging methods.
Fig.2(a) Preparation of NIR-II phototriggered hydrogel BTN and its real-time monitoring for cardiac IRI treatment [46]; (b) Spectroscopy of NIR-emitting semiconductor polymer nanoparticles (NESPN) and in vivo two-photon brain IRI vascular imaging [47]; (c) Spectroscopy of novel high-brightness and high-stability BM fluorophores and their application in brain IRI using NIR-II imaging [48].
4.1.3 Application in renal IRI imaging
Yi S et al. [49] developed an ultra-small gold nanocluster probe (see Figure3 (a) ) by adopting a dual-ligand stabilization strategy to address the problem of low renal clearance efficiency of traditional NIR-II fluorescent materials. The hydrodynamic diameter of the probe is precisely controlled within the range of 2.3 nm to 2.8 nm, which just meets the size requirements of glomerular filtration, thus achieving efficient renal clearance. Its quantum yield is 1.4% to 2.0%, which is nearly an order of magnitude higher than that reported previously. Under 1064 nm laser excitation, the probe exhibits excellent tissue penetration, spatial resolution and signal-to-noise ratio of fluorescence signal. In mouse models of renal IRI and unilateral ureteral obstruction, in vivo dynamic imaging can clearly track the entire process of the probe passing through the glomerulus, entering the collecting system and finally being discharged into the bladder, and can sensitively distinguish different degrees of renal dysfunction. Its diagnostic sensitivity is significantly better than the clinical routine serum creatinine and urea nitrogen indicators, providing a powerful tool for the early diagnosis of acute kidney injury.
Yao C et al. [50] started with the design of small organic molecules, aiming to solve the key problems of insufficient brightness and short blood circulation time of most NIR-II organic probes. They used atom transfer radical polymerization technology to synthesize a series of brush-shaped macromolecular probes with aza-BODIPY as the fluorescent core and polyethylene glycol chains grafted on the periphery, and screened out the high-performance FBP912 (see Figure3 (b) ) . The molar extinction coefficient of this probe is as high as 60 M-¹·cm-¹, and the overall brightness is about 10 times that of the previous renally clearable organic probes. It has both excellent photostability and resistance to reactive oxygen quenching. Its hydrodynamic diameter is about 4 nm, which is lower than the glomerular filtration threshold, and it achieves a renal clearance rate of 65% within 12 h. The unique brush conformation gives it a plasma half-life of up to 6.1 h, which is significantly better than conventional linear PEGylated probes. In the renal IRI model, the FBP912 achieved high signal-to-noise ratio imaging, with signal changes occurring earlier than traditional serological indicators. It can accurately reflect early renal function damage, providing a high-performance tool for dynamic non-invasive monitoring of kidney diseases and opening up new avenues for imaging diagnosis of diseases such as tumors.
Fig.3(a) Preparation, in vitro imaging, and high-resolution fluorescence imaging of renal scavenging gold nanoclusters with NIR-II excitation and emission [49]; (b) Structure, fluorescence spectrum, and renal IRI bioimaging of the NIR-II organic molecular probe FBP series [50].
4.1.4 Application in limb IRI imaging
Li B et al.[51] reported an organic small molecule probe, LZ-1105, with long-circulating properties (see Figure4 (a) ) . The absorption and emission wavelengths of this probe both exceed 1000 nm, which falls within the NIR-II window. Its blood circulation half-life is as long as 3.2 h, which is significantly better than the clinically commonly used indocyanine green dye. This long-circulating property provides a sufficient timewindow for continuous observation of vascular physiological and pathological processes. Based on LZ-1105, researchers can not only obtain high-resolution, high signal-to-noise ratio static vascular anatomy images, but also monitor dynamic processes such as lower extremity IRI, carotid artery thrombolysis, and blood-brain barrier opening and repair in real time, which significantly expands the application scope of NIR-II imaging in vascular function research.
Jia Q et al. [52] designed an erbium-doped rare-earth nanoprobe − Er-DCNPs (see Figure4 (b) ) − based on an inorganic material system. Through Nd3+ sensitization and Ce3+ co-doping strategy, the downconversion luminescence efficiency of the probe at 1530 nm was significantly improved. In vivo animal model experiments showed that the probe has a tissue penetration capability of up to 7 mm and excellent spatial resolution. In a mouse peripheral artery disease model, Er-DCNPs can not only clearly display the fine anatomical structure of the lower limb blood vessels, but also assess the blood flow recovery and monitor the thrombolysis process in real time and quantitatively, and successfully achieve non-contact measurement of key physiological parameters such as heart rate and respiratory rate, showing broad potential in the integration of diagnosis and treatment.
Fig.4(a) Spectral diagram of LZ-1105, an organic NIR-II molecule with a long blood half-life, and in vivo dynamic vascular NIR-II imaging [51]; (b) Dynamic imaging of vascular perfusion and fluorescence spectrum of erbium-based lanthanide nanoprobes emitting at 1500 nm [52].
4.2 Targeted probes
To improve imaging specificity and signal-to-noise ratio, targeted probes actively identify and bind to biomarkers (such as adhesion molecules, enzymes, and receptors) overexpressed in the IRI lesion area by coupling with specific ligands (such as antibodies, peptides, and small molecules) , thereby achieving precise localization of the damaged area.
4.2.1 Application in cardiac IRI imaging
Ferroplasmosis, an iron-dependent programmed cell death mechanism, has been shown to play a crucial role in myocardial ischemic inflammatory resorption (IRI) . Yang W et al. [53] constructed a multimodal imaging platform targeting ferroptosis (see Figure5) . The team designed multifunctional nanoprobes CCINPs, the core of which is superparamagnetic cubic iron oxide nanoparticles (for MRI imaging) , and the surface is modified with indocyanine green (for NIR fluorescence imaging) , cell-penetrating peptides, and targeting peptides targeting transferrin receptor 1 (upregulated expression in ferroptosis cardiomyocytes) , which significantly enhances the targeting and uptake efficiency of damaged cardiomyocytes. In a mouse model of myocardial IRI induced by ligation of the left anterior descending coronary artery, the probes achieved, for the first time, in vivo and quantitative visualization of the ferroptosis process in myocardial tissue through MRI/NIR-FL dual-modal imaging, providing a new tool for the precise diagnosis and efficacy evaluation of cardiovascular diseases.
4.2.2 Application in brain IRI imaging
The presence of the blood-brain barrier severely limits the brain delivery efficiency of most probes. (Seyda) MZ et al. [54] developed a delivery system based on a "platelet hitchhiking" strategy (see Figure6 (a) ) . They prepared fucosaccharide-modified lipid nanoparticles loaded with the NIR dye IR780 (constituting the T-IR780 probe) or the therapeutic drug triiodothyronine. Fucosaccharide can specifically bind to P-selectin, which is highly expressed on the surface of platelets, thereby enabling the nanoparticles to efficiently cross the damaged blood-brain barrier and accumulate in brain IRI lesions by using platelets as a carrier. In vivo NIR-II imaging showed that after intravenous injection of T-IR780, the signal of the probe in the ischemic brain region peaked at 2 h and was significantly higher than that in the non-targeted control group. Ex vivo organ imaging further confirmed its specific aggregation ability in the brain. This bio-inspired strategy provides a new idea for brain-targeted delivery.
Furthermore, the brain's lymphoid system plays a crucial role in the clearance of metabolic waste, and its dysfunction is closely related to brain IRI. B et al. [55] usedNIR-II nanoprobes (quantum dots and BSA@IR-780) to simulate cerebrospinal fluid circulation by injecting the cerebellomedullary cistern, and combined with NIR-II imaging technology to non-invasively and dynamically monitor changes in lymphoid system function under anesthesia and brain IRI conditions (see Figure6 (b) ) . This technology can penetrate the intact scalp and observe the flow of cerebrospinal fluid along the perivascular space in real time at high resolution, providing an in vivo research method for understanding the mechanisms of central nervous system diseases such as stroke.
Fig.6(a) Brain IRI imaging by binding platelets to penetrate the blood-brain barrier and targeting ischemic areas of the brain [54]; (b) Brain IRI imaging using quantum dots and BSA@IR-780 [55].
4.2.3 Application in lung IRI imaging
He X et al. [56] constructed a therapeutic upconversion nanoplatform − USDPFs (see Figure7 (a) ) . The platform uses core-shell upconversion nanoparticles as the core, coated with mesoporous silica, modified with the photosensitizer γ-oxo-1-pyrenebutyric acid and loaded with the anti-inflammatory drug dexamethasone, and finally sealed the pores with fluorescein isothiocyanate-labeled β-cyclodextrin. Under 980 nm laser excitation, UCNPs convert the near-infrared light that can penetrate deep tissues into local ultraviolet-visible light, triggering the decomposition of the photosensitizer and controlling the release of dexamethasone, while realizing opticalimaging of deep tissues. In a mouse lung IRI model, NIR imaging after intravenous injection showed that the probe was specifically enriched in the lungs, realizing simultaneous treatment and imaging, demonstrating the application potential of upconversion materials in precision medicine of thoracic surgery.
4.2.4 Application in liver IRI imaging
Li C et al. [57] synthesized a novel organic NIR-II imaging agent − Y6CT nanoparticles (see Figure7 (b) ) . Through ingenious molecular design, the nanoparticles, using strong donor-receptor interactions between and within Y6CT molecules, can generate high-brightness NIR-II fluorescence under white LED excitation commonly used in clinical laparoscopy. By comparing NIR imaging of livers with intact vascular structures and livers with damaged blood vessels after IRI, the sensitive imaging ability of Y6CT nanoparticles for liver IRI was clearly demonstrated, providing a promising tool for real-time assessment of liver perfusion and damage extent during surgery.
Fig.7(a) USDPF nanoparticles used for lung IRI upconversion imaging [56]; (b) Y6CT nanoparticles used for liver IRI imaging [57].
4.2.5 Application in renal IRI imaging
Zhu D et al. [58] developed a NIR-II light-triggered intelligent delivery system − PLK3-LIP (see Figure8 (a) ) − for imaging and treatment of renal IRI. They modified the surface of liposomes encapsulating PLK3 kinase inhibitors with NIR-II dye FD-1080as a light absorber. Under 1080 nm laser irradiation, the photothermal effect generated by the dye can cause the liposome membrane to rupture, precisely controlling the release of drugs at the site of kidney injury, while using the NIR-II window for high-penetration imaging. In a mouse model of renal IRI, after intravenous injection of PLK3-LIP, the specific enrichment of the probe in the ischemic kidney can be dynamically observed through real-time NIR fluorescence signals (the signal peaks at 12 h) , thereby realizing the visualization of the damaged area. This study is the first to combine NIR-II light-triggered release with targeted delivery, providing a new strategy for the precise diagnosis and treatment of renal IRI.
4.2.6 Application in limb IRI imaging
Ji A et al. [59] developed a white LED-excited NIR-II imaging system (see Figure8 (b) ) . Its core innovation is the design and synthesis of a blue-shifted NIR-II organic dye with a donor-acceptor-donor structure, TPA-TQT. The absorption spectrum of this dye effectively overlaps with the emission spectrum of a400−700 nm white LED, and it can be safely and cost-effectively excited by white light to produce strong fluorescence of 1000-1400 nm. Compared with ICG, TPA-TQT has better photostability and is suitable for long-term dynamic imaging. In a mouse hind limb IRI model, the system clearly displayed the hind limb microvascular network at a high resolution of 103 μm and successfully monitored the dynamic process of inflammatory cell infiltration after reperfusion. Quantitative analysis further showed that the imaging signal intensity was linearly correlated with the degree of tissue damage, showing potential for quantitative assessment. This study provides a safer, more economical and portable NIR-II imaging scheme through dye innovation and system optimization, opening up new prospects for real-time monitoring during clinical surgery (especially for endoscopic applications) .
Fig.8(a) PLK3-LIP for imaging and treatment of renal IRI [58]; (b) TPA-TQT for imaging of mouse hindlimb IRI under white light excitation [59].
Non-responsive NIR fluorescent probes provide powerful in vivo visualization tools for IRI research through passive targeting (EPR effect) or active targeting (specific molecular recognition) strategies. Current developments focus on improving probe brightness, stability, and biosafety, and driving their evolution towards therapeutic integration, multimodal imaging, and clinical applicability. Although challenges remain in targeting efficiency and clinical translation, these probes have significantly deepened our understanding of the pathological mechanisms of IRI and show broad application prospects in the field of precision medicine.
5 Application of responsive NIR fluorescent probes in IRI imaging
Responsive probes are the mainstay of IRI molecular imaging. They can convert invisible biomolecular events into detectable optical signals, enabling the "reporting" of specific pathological processes.
5.1 Oxidizing species response probes
Oxidative stress is a core component of IRI, and probe research targeting various ROS/RNS is particularly extensive.
5.1.1 Application in brain IRI imaging
In the detection of peroxynitrite (ONOO-) , Shang J et al. [60] developed a hydrazine-based recognition group-based probe TJO (see Figure9 (a) ) . It emits fluorescence at 730 nm with a Stokes shift of 167 nm and exhibits high selectivity, rapid response (within2 min) , and high sensitivity (detection limit 91 nM) for ONOO-, successfully achieving dynamic imaging of ONOO-in a mouse model of brain IRI. Xie C et al. [61] designed the probe SWJT-37 (see Figure9 (c) ) , which recognizes superoxide anion () using the trifluoromethanesulfonate group, producing near-infrared fluorescence at 667 nm. Its detection limit is 3.88 nM, fluorescence is enhanced 51-fold, and it has good blood-brain barrier penetration. It has been applied to endogenous imaging in a rat model of brain IRI. More groundbreakingly, Xu L et al. [62] constructed a NIR-II molecule, CR-OH, with a redshifted emission wavelength of 1026 nm by molecularly engineering a classic DCF dye, and developed an activation probe CR (see Figure9 (b) ) based on this. They achieved NIR-II fluorescence imaging of endogenous in a brain IRI model for the first time. This study provides an advanced molecular tool for elucidating the mechanism of oxidative stress in stroke.
Fig.9(a) Real-time monitoring of ONOO- in a mouse model of cell and brain IRI using hydrazine-based NIR fluorescent probe TJO [60]; (b) response mechanism of molecularly engineered 2′, 7′-dichlorofluorescein and in vivo NIR II brain IRI fluorescence imaging [62]; (c) activated NIR fluorescent probes penetrate the blood-brain barrier to achieve selective tracing of brain IRI [61].
5.1.2 Application in renal IRI imaging
To address the issues of poor biocompatibility and slow metabolism of traditional NIR-II probes, Zeng C et al. [63] developed a novel NIR-II fluorescence/photoacoustic dual-modal imaging probe − PEG3-HC-PB (see Figure10 (b) ) . Based on the heptamethrin structure, the probe effectively segments the hydrophobic region by introducing three PEG3 fragments into the conjugated backbone, significantly improving water solubility and renal clearance capacity (size less than 6−8 nm) and reducing non-specific protein adsorption. Its recognition mechanism depends on H2O2 triggering the conversion of phenylboronic acid groups to phenol, restoring NIR-II fluorescence (peak at 950 nm) and enhancing photoacoustic signal (peak at 830 nm) . In a mouse model of acute kidney injury induced by drugs or ischemia-reperfusion, the probe achieved high signal-to-noise ratio, real-time three-dimensional multispectral photoacoustic and NIR-II fluorescence dual-modal imaging of H2O2 in the kidney region.
Xie X et al. [64] focused on monitoring oxidative stress at the organelle level and developed the mitochondrial-targeting probe Mito-NIRHP (see Figure10 (a) ) . This probe utilizes the nucleophilic attack of H2O2 on α-ketoamides to induce Baeyer–Villiger rearrangement and hydrolysis, releasing the fluorophore Cy-NH2, achieving a fluorescence-on response at wavelengths of 670 nm/704 nm. Mito-NIRHP has high specificity for H2O2, is not affected by other ROS/RNS or biothiols, has a detection limit of 26 nM, and a response time of approximately 10 min. Guided by the mitochondrial-targeting group, this probe was successfully used to monitor fluctuations in endogenous H2O2 in cellular mitochondria and revealed a significant increase in H2O2 in the mitochondria of the damaged area in a mouse kidney IRI model.
Fig.10(a) The response mechanism of the α-ketoamide-based H2O2 NIR fluorescent probe Mito-NIRHP and its specific detection in a live model of renal IRI [64]; (b) The renal clearance probe PEG3-HC-PB was used to detect renal IRI by biomarker activation through NIR-II fluorescence and photoacoustic imaging [63].
5.1.3 Application in liver IRI imaging
Song D et al.[65] constructed a reversible probe for the NIR-IIb region based on rare earth-doped nanoparticles and molybdenum-based polyoxometalate clusters − REPOMs (see Figure11 (a) ) . This probe utilizes the valence state change between Mo5+/Mo6+ to achieve a reversible response to ROS and glutathione cycling, and achieves ratiometric signal output through an absorption-competitive induced luminescence mechanism. REPOMs exhibit good stability, selectivity and reversibility both in vitro and in vivo, and can be used to monitor the dynamic fluctuations of ROS during hepatic IRI in real time, and provide an evaluation basis for ischemic preconditioning and drug intervention.
Chen J et al. [66] designed an AIE-active nanoprobe BTPE-NO2@F127 based on a benzothiadiazole core (see Figure11 (b) ) . The probe uses nitrophenoxyacetamide as a recognition and quenching group, which breaks down under the action of H2O2 to release the chromophore BTPE-NH2 with strong NIR-II fluorescence (950−1200 nm) and photoacoustic signal. The probe has good water dispersibility, biocompatibility andhigh signal-to-noise ratio, and has been successfully applied to dual-modal imaging of various inflammatory models such as interstitial cystitis, drug-induced liver injury and liver IRI, achieving precise localization of lesion areas.
Fig.11(a) Real-time tracking of liver IRI using reversible NIR-II fluorescent redox probes REPOMs [65]; (b) Hydrogen peroxide-activated nanoprobe BTPE-NO₂@F127 was used to diagnose liver IRI by multispectral photoacoustic tomography and NIR-II fluorescence imaging [66].
5.2 Reducible species response probes
The dynamic balance of the intracellular antioxidant system plays an important role in IRI, and its dysregulation is an important marker of the damage process.
5.2.1 Application in cardiac and pulmonary IRI imaging
Luo X et al. [67] constructed a near-infrared light-activated dual-response nanoprobe − UCNP@mSiO2@SP-NP-NAP (see Figure12 (a) ) − for the simultaneous detection of hydrogen polysulfides and sulfur dioxide during myocardial IRI. This probe, based on upconversion nanoparticles, can convert 980 nm light into ultraviolet light to activate photoresponsive dyes, thereby achieving fluorescence detection of H2S2 and SO2, respectively. This system has good light control reversibility and high spatiotemporal resolution, with detection limits of 0.24 μM and 0.18 μM for SO2 and H2S2 in vitro, respectively. In cell and mouse I/R models, this probe successfully monitored the upregulation of H2S2 and SO2 levels, and confirmed their potential role in alleviating myocardial IRI by inhibiting the key ferroptosis protein GPX4.
Jiang L et al. [68] designed an iridium complex-based NIR probe, NIR-Ir-BDS (see Figure12 (b) ) . This probe uses 2, 4-dinitrobenzenesulfonamide as a recognition group and can release strongly NIR-emitting NIR-Ir-BH upon cleavage by biothiols. The probe possesses a large Stokes shift (172 nm) , an emission wavelength of 667 nm, a detection limit of 16.7 nM, and a rapid response (within 20 min) . Its cationic properties and lipid solubility enable it to specifically target mitochondria. In a ferroptosis-mediated lung IRI model, this probe enabled the first dynamic imaging of mitochondrial biothiols in mouse lung tissue. The results showed that mitochondrial biothiols were significantly depleted with prolonged ischemia, and intervention with ferroptosis inhibitors could reverse this process.
Fig.12(a) NIR-activated nanoprobes simultaneously detect the roles of H2S2 and SO2 in cardiac IRI [67]; (b) Novel NIR iridium (III) complex probes are used for precise imaging of changes in mitochondrial biothiols in ferroptosis-mediated lung IRI [68].
5.2.2 Application in brain IRI imaging
In the field of brain IRI research, Yang Y et al. [69] first constructed a self-eliminationchemistry-based NIR-HMPC probe (see Figure13 (a) ) . This probe utilizes a "thiol-chromene" click reaction to trigger a self-elimination process, releasing the NIR fluorophore NIR-OH, and possesses rapid response (within minutes) , high sensitivity (detection limit for Cys is 0.39 μM) , and mitochondrial targeting capability. This probe not only successfully monitored the dynamic changes of thiols during oxidative stress and apoptosis at the cellular level, but also achieved preliminary dynamic imaging of thiol levels in brain tissue in a brain IRI mouse model, providing an effective means to explore the temporal changes of thiol metabolism during brain I/R.
The team then developed the probe DCI-Ac-Py (See Figure13 (b) ) [70]. The probe uses pyridine carboxylate as a recognition group, which triggers ester bond cleavage and intramolecular cyclization under the action of biothiols, achieving a fluorescence-on response at 713 nm. The detection limits for Cys, GSH and Hcy are27 nM, 55 nM and 74 nM, respectively. The probe has good blood-brain barrier penetration ability, can monitor the fluctuation of biothiols in the brain in real time at the in vivo level, and found that its changes are closely related to the activation of the NF-κB signaling pathway, providing a new perspective for understanding the molecular mechanism of brain I/R.
Fig.13(a) Visualization of thiol flux during mouse brain IRI using a self-destructingNIR fluorescent probe triggered by the thiol-chromene "click" reaction [69]; (b) The fluorescent probe DCI-Ac-Py was used to investigate the role of biothiols in brain IRI-related signaling pathways [70].
5.2.3 Application in liver IRI imaging
Ye M et al. [71] designed a NIR upconversion luminescent nanosystem based on upconversion nanoparticles (see Figure14 (a) ) for carbon monoxide detection. The system consists of NaYF4: Yb/Tm UCNPs modified with CyOND dye and PdCl2. It utilizes the luminescence resonance energy transfer mechanism to trigger NIR upconversion luminescence enhancement in the presence of CO. This probe has the advantages of high selectivity, high tissue penetration and low background interference. It has been successfully applied to the visualization of CO biosignals in live cells, zebrafish and mouse liver IRI models, and the protective effect of CO in liver IRI has been verified.
Xiong S et al. [72] developed an H2S-activated NIR fluorescent probe, hCy-H2S (see Figure14 (b) ) . This probe uses DNP as its recognition unit and cleaves to release the strong NIR fluorophore hCy-MP under the action of H2S. It has high sensitivity (detection limit of 0.16 μM) , rapid response, and good biocompatibility. The probe is preferentially enriched in the liver and can be used to image H2S fluctuations in real time and in situ in cellular and mouse liver IRI models, revealing its key role in oxidative stress relief.
The two studies above developed NIR optical probes with high spatiotemporal resolution targeting CO and H2S, two gaseous signaling molecules with cytoprotective effects, respectively. These probes enabled dynamic, in-situ imaging during liver IRI, providing important technical support for understanding their molecular mechanisms and early diagnosis.
Fig.14(a) Deep imaging of liver IRI using CO-activated upconversion luminescent nanoparticles [71]; (b) Fluorescent probe hCy-H2S used for in situ tracking of liver IRI [72].
5.3 Application of enzyme-responsive probes in liver and lower extremity IRI imaging
Activation of specific enzymes provides highly specific targets for IRI imaging. Hong SJ et al. [73] constructed a dual-enzyme activated NIR fluorescent probe, QN-NIR (see Figure15 (a) ) , for the early detection of liver ischemia. This probe integrates an hNQO1-sensitive TLQ group and an NTR-sensitive NB group. Fluorescence quenching is deactivated only when the two enzymes coexist, generating708 nm NIR emission with high signal-to-noise ratio and low background interference. In a mouse liver ischemia model, the fluorescence of this probe increased significantly in the ischemic region, and the signal dropped after reperfusion, indicating that it can specifically identify the ischemic stage.
Zhang Y et al. [74] developed a reversible aryl azo NIR probe, HDSF (see Figure15 (b) ) , for in vivo cyclic hypoxia imaging. This probe achieves a reversible fluorescence switching response (emission wavelength of 705 nm) in hypoxia-noroxic cycles byintroducing a trifluoromethyl stable reduction intermediate to block the irreversible cleavage of N–N bonds. HDSF has good biocompatibility, mitochondrial targeting ability and high selectivity, and has been successfully used for real-time hypoxia dynamics imaging in mouse solid tumor and limb IRI models.
The probes described above expand the application of NIR imaging in IRI research from two perspectives: "dual-enzyme synergistic detection" and "reversible identification of cyclic hypoxia".
Fig.15(a) In vivo monitoring of hepatic IRI using the dual-enzyme activated NIR fluorescent probe QN-NIR [73]; (b) Hypoxia imaging of lower extremity IRI using the reversible azo NIR probe HDSF [74].
5.4 Application of pH-responsive probes in brain IRI imaging
The acidic microenvironment of ischemic regions is a key pathological feature and important imaging target of ischemic inflammatory resection (IRI) . Cheng Y et al. [75] constructed a pH-responsive multifunctional nanosystem (see Figure16 (a) ) for the integrated diagnosis and treatment of acute ischemic stroke. This system utilizes the acidic environment of the ischemic area to achieve targeted and controlled release of rapamycin, while integrating Gd3+ and Ce6 molecules to endow it with acid-enhanced MRI and NIR fluorescence dual-modal imaging capabilities. In a rat middle cerebral artery occlusion model, the system significantly improved neurological function scores, infarct volume, and inflammation levels.
Fig.16(a) pH-responsive multifunctional nanoparticles for imaging of acute ischemic stroke [75]; (b) pH-responsive NIR liposome probe BOD@Lip for real-time objective assessment of ischemic stroke [76].
Yao S et al. [76] developed a pH-responsive NIR fluorescent liposome probe, BOD@Lip (see Figure16 (b) ) , for real-time assessment of the degree of cerebral ischemia. This probe uses Aza-BODIPY as its fluorescent core, exhibits enhanced fluorescence under acidic conditions, and can cross the damaged blood-brain barrier and accumulate in ischemic areas. This study established a quantitative correlation between fluorescence signal intensity and neurological deficit scores, and validated its assessment accuracy using multimodal techniques.
The above studies demonstrate the broad application prospects of pH-responsive nanomaterials in the diagnosis and treatment of stroke, and provide new ideas for precision medicine of ischemic stroke.
5.5 Application of other microenvironment parameter response probes in IRI imaging
The physical parameters of the cellular microenvironment (such as viscosity and polarity) undergo dramatic changes during intracellular resorption (IRI) and play a pivotal role in connecting molecular events with cell fate. As key bridges linking molecular events and cell fate, they are becoming an important breakthrough forunderstanding IRI mechanisms and developing new strategies.
Hu W et al. [77] reported a NIR fluorescent probe with a large Stokes shift (290 nm) − XZTU-Mito-Vis (see Figure17 (a) ) . It is based on the intramolecular proton transfer mechanism in the excited state and is used to monitor changes in mitochondrial viscosity. The probe exhibits a418-fold increase in fluorescence when viscosity increases, with an emission wavelength at 715 nm, and achieves mitochondrial-specific targeting through a pyridinium salt structure. In oxygen-glucose deprivation/reperfusion and middle cerebral artery occlusion models, XZTU-Mito-Vis accurately captured the increase in mitochondrial viscosity caused by oxidative stress and verified the neuroprotective effect of apocynin in inhibiting inflammatory factors (TNF-α, IL-1β) and alleviating cerebral infarction. This probe combines blood-brain barrier penetration ability with high-resolution imaging performance, providing a novel molecular tool for early diagnosis and treatment assessment of stroke.
In liver IRI studies, Liu J et al. [78] developed a viscosity-activated NIR-II fluorescent probe, NP-V (see Figure17 (b) ) , for precise navigation of liver IRI. This probe enhances structural rigidity and introduces flexible conjugated bonds to achieve a13-fold increase in fluorescence intensity in a high-viscosity environment, while also exhibiting good photostability and biocompatibility. NP-V specifically targets lysosomes, and a significant increase in lysosomal viscosity (2.9-fold) was successfully monitored in the HIRI model, further revealing the key role of the ROS–MDA–cathepsin B signaling pathway in mediating viscosity changes. In live mice, NP-V achieved high signal-to-background ratio (SBR = 3.7) NIR-II imaging of the liver lesion area and guided the precise surgical resection of the lesion tissue, with its effectiveness being pathologically verified. This study is the first to establish lysosomal viscosity as a reliable biomarker for HIRI, providing a new strategy for intraoperative navigation and mechanistic research.
Ai Y et al.[79] designed a NIR probe targeting lipid droplets—A-01 (see Figure17 (c) ) , which achieves a highly sensitive response to polarity changes through anintramolecular charge transfer mechanism. The emission wavelength of this probe redshifts by 140 nm when polarity is enhanced, the fluorescence intensity decreases by 14 times, and it is not sensitive to viscosity interference. A-01 specifically labels lipid droplets in HeLa cells, and a decrease in cell polarity was detected in all four IR models, indicating that IRI leads to lipid droplet microenvironment disorder. Further studies found that low concentrations of lead-based Polygonatum can significantly alleviate the polarity changes and ROS increases caused by IR, demonstrating its protective effect. In a mouse hind limb IR model, A-01 successfully imaged the dynamic changes in polarity, and the fluorescence signal was enhanced by two times in the IR limb, providing a visualization tool for evaluating the role of natural drugs in the prevention and treatment of IRI.
These three studies, focusing on intraepithelial neoplasia (IRI) in different tissues − liver, lower limb, and brain − developed NIR fluorescent probes targeting physical microenvironment parameters such as viscosity and polarity. These probes not only achieved high-contrast imaging and precise intervention of the lesion area but also revealed the crucial roles of oxidative stress, lipid metabolism disorders, and lysosomal dysfunction in IRI. These probes provide a powerful optical platform for mechanistic research, early diagnosis, and drug evaluation of IRI, demonstrating broad prospects for clinical translation.
Fig.17(a) The large Stokes shift NIR fluorescent probe XZTU-Mito-Vis was used to accurately monitor oxidative stress and early diagnosis of stroke by detecting mitochondrial viscosity [77]; (b) The lysosomal viscosity-activated NIR II fluorescent probe NP-V was used for precise navigation of hepatic IRI [78]; (c) The lipid droplet-targeting NIR probe A-01 was used to evaluate the preventive effect of prunella vulgaris in IRI [79].
Responsive NIR fluorescent probes enable multidimensional, high spatiotemporal resolution dynamic imaging of the injury process by specifically recognizing key biomolecules (such as ROS/RNS, enzymes, and gaseous signaling molecules) and microenvironmental parameters (such as pH, viscosity, and polarity) during the intraoperative renal inflammatory response (IRI) process. These probes not only deepen our understanding of the molecular mechanisms of IRI but also demonstrate significant advantages in precision diagnosis, intraoperative navigation, and treatment assessment. Although challenges remain regarding probe specificity, in vivo stability, and clinical translation, with continuous innovation in molecular design and imaging technologies, responsive NIR probes are expected to provide stronger technical support for theprecision diagnosis and treatment of IRI.
6 Challenges and outlook
Despite the remarkable progress made by fluorescent probe technology in the research and clinical application of IRI and its enormous translational potential, we must be soberly aware that there is still a challenging road ahead from basic laboratory research to widespread clinical application.
6.1 Major challenges currently faced
6.1.1 The clinical translation challenges of probes
Currently, the vast majority of reported IRI fluorescent probes remain in the validation stage using cell and rodent (mouse, rat) models. Their clinical application faces multiple obstacles:
(1) Biosafety and toxicity: The long-term toxicity, immunogenicity, in vivo distribution, metabolic pathways and clearance mechanisms of probe components (such as Ir and Pd in metal complexes, organic dyes and their metabolites) have not been systematically and comprehensively evaluated, which is the primary hurdle for clinical translation.
(2) Lack of data in large animal models: Small animal models such as mice differ significantly from humans in terms of physiological structure, metabolic rate, and tissue penetration depth. The lack of validation data on efficacy and safety in large animal models such as pigs, dogs, or non-human primates is a key issue in translational research.
(3) Unclear regulatory approval path: As a novel in vivo diagnostic agent, fluorescent probes have different regulatory classifications, approval processes and standards than traditional drugs or imaging contrast agents. It is necessary to communicate with regulatory agencies (such as the National Medical Products Administration, the US FDA, etc.) in the early stage and establish a corresponding evaluation system.
6.1.2 Optimization potential of probe performance
(1) Specificity and cross-reactivity: Although probe design strives for specificity, in the complex in vivo microenvironment, some probes may still cross-react with structurally similar molecules, producing false positive signals. For example, borate ester probes for detecting H2O2 may also respond to other peroxides.
(2) Stability and photobleaching: Some organic fluorophores are prone to photobleaching under continuous light, which leads to signal attenuation and affects the accuracy of quantitative analysis and the feasibility of long-term observation.
(3) Quantitative analysis capabilities need to be improved: Most probes provide semi-quantitative signals that are either "on" or "enhanced". How to achieve absolute quantification and accurately reflect the true concentration of biomolecules in the body remains a challenge.
6.1.3 Limitations in understanding the complexity of IRI
(1) Incomplete coverage of biomarkers: Current probes mainly focus on ROS, RSS and a few gas molecules. However, IRI involves many other key processes, such as specific inflammatory factors (IL-1β, IL-18) , pyroptosis-related proteins (GSDMD) , autophagic flux, endoplasmic reticulum stress, etc., and excellent probes targeting these targets are still relatively scarce.
(2) Lack of visualization tools for multi-organ interactions: IRI usually manifests as a systemic process, and IRI in a single organ may trigger damage to distant organs (e.g., intestinal IRI leading to lung injury) . Currently, there is a lack of imaging techniques and probes that can simultaneously and in vivo study the correlation of molecular events between multiple organs.
6.2 Future development direction and outlook
To overcome the above challenges, future research should focus on the following cutting-edge directions.
6.2.1 Promoting clinical translation and practical application
(1) Emphasize biocompatibility design: In the early stages of probe molecule design, priority should be given to using materials with better biocompatibility, such as developing all-organic, water-soluble NIR dyes, or using materials that have been approved for clinical use (such as ICG, certain nanocarriers) to construct probe platforms.
(2) Conduct systematic preclinical studies: strictly follow drug development standards, verify the effectiveness of the probe in large animal IRI models, and complete comprehensive pharmacokinetic and toxicological studies.
(3) Explore clinical application scenarios: In the early stage, we can target minimally invasive or open scenarios such as intraoperative navigation. For example, we can use probes to assess graft viability in real time during organ transplantation surgery and determine the viable area of brain tissue or myocardium after thrombosis resection, so as to realize its initial clinical value.
6.2.2 Developing high-performance novel probes
(1) Long wavelength and NIR-II probes: Vigorously develop probes with excitation and emission wavelengths in the NIR-I and even NIR-II regions. NIR-II imaging has deeper tissue penetration, lower autofluorescence and higher spatial resolution, and is the future trend of in vivo imaging.
(2) Ratio-type and self-calibrating probes: Vigorously develop ratio-type probes, which quantify by measuring the ratio of fluorescence intensity at two different wavelengths. This can effectively eliminate the influence of environmental factors such as uneven probe concentration distribution and light intensity fluctuations, and significantly improve the accuracy and reliability of quantification.
(3) Activable smart probes: Design "AND" logic gate probes that are activated by multiple biological signals (such as ROS+pH, or enzyme activity+viscosity) . These probes will only emit strong signals when multiple conditions are met simultaneously in the specific microenvironment of IRI, which can greatly improve the specificity andaccuracy of detection and reduce the possibility of false positive interference.
6.2.3 Expanding multi-mode imaging and technology fusion
(1) Multimodal imaging probes: Develop probes that combine fluorescence imaging with other clinically mature imaging technologies (such as magnetic resonance imaging, computed tomography (CT) , ultrasound imaging, and photoacoustic imaging) . For example, the UCL nanosystem developed by Ye M et al. in 2022 is essentially a platform that integrates upconversion luminescence and potential photoacoustic imaging capabilities. This fusion combines the high sensitivity of fluorescence with the high penetration depth and clear anatomical structure of other modes, enabling comprehensive assessment of IRI from the molecular to the anatomical level.
(2) Integration with artificial intelligence: Using artificial intelligence and deep learning algorithms to automatically analyze, segment and quantify complex fluorescence imaging data, and to discover feature patterns that are difficult for the human eye to recognize, thereby achieving automation and precision in early diagnosis, classification, prognosis prediction and efficacy evaluation of IRI.
6.2.4 Deepen the research on IRI mechanisms
(1) Develop novel biomarker probes: Develop specific probes for key molecular events in novel cell death pathways such as ferroptosis, pyroptosis, and NETosis to visualize the spatiotemporal dynamics of these processes in IRI and provide tools for developing new targets.
(2) Multi-organ in vivo dynamic monitoring system: Using implantable optical fiber or wearable fluorescence imaging equipment, combined with highly specific probes, long-term and continuous dynamic monitoring of molecular events in multiple organs of free-moving animals can be achieved, thereby revealing the systemic pathophysiological network of IRI.
7 Conclusions
NIR fluorescent probes, acting as powerful "optical scalpels", are profoundly changingour understanding of intraepithelial neoplasia (IRI) . Non-responsive probes allow us to precisely delineate the anatomical extent of damage; responsive probes provide insight into the underlying molecular storms of IRI. From non-targeted to targeted, from single-response to multiple-response, from NIR-I to NIR-II, each advancement in probe technology provides us with more acute insights into understanding and managing IRI. While the road ahead remains challenging, with the deep integration of chemistry, biology, materials science, and medicine, we have reason to believe that higher-performance, more intelligent NIR fluorescent probes will play an indispensable role in the precise clinical diagnosis and personalized treatment of IRI in the near future.