2‐Azaazulenium Hemicyanine Dyes for NIR‐II Multimodal Imaging and Combined Phototherapy

ABSTRACT

Hemicyanine dyes are well‐defined donor–π–acceptor molecules whose photophysical and biological properties can be precisely tuned through rational molecular design. However, efficient emitters operating in the second near‐infrared (NIR‐II) region remain scarce, hindering their broader biomedical applications. Here, we report a strategically engineered series of 2‐azaazulenium hemicyanines that achieve bright NIR‐II emission together with multifunctionality. Among them, SWJT‐OCH₃ integrates intense NIR‐II fluorescence with outstanding photothermal and photodynamic performance, enabling multimodal imaging and combined phototherapy when encapsulated into biocompatible nanoparticles (SWJT NPs). Furthermore, an activatable probe (SWJT‐Probe) derived from this platform allows highly sensitive and selective detection of peroxynitrite (ONOO⁻) both in vitro and in vivo. This work establishes 2‐azaazulenium hemicyanines as versatile molecular scaffolds and provides a promising design paradigm for multifunctional NIR‐II fluorophores with broad biomedical potential.

A strategically engineered class of 2‐azaazulenium hemicyanines is developed, featuring bright fluorescence in the second near‐infrared (NIR‐II) window and multifunctional performance. SWJT‐OCH₃ shows type I/II reactive oxygen species (ROS) generation and excellent photothermal activity. This platform enables fluorescence/photoacoustic/photothermal imaging, combined phototherapy, and selective peroxynitrite (ONOO⁻) detection.

1. Introduction

Cancer remains one of the most serious global health threats [1]. Advances in early screening, precision medicine, and immunotherapy have markedly improved its prevention and treatment, yet effective and minimally invasive therapies are still needed [2, 3, 4]. Phototherapy, including photodynamic therapy (PDT) and photothermal therapy (PTT), has emerged as a promising option due to its high selectivity, minimal invasiveness, and excellent biosafety [5, 6, 7, 8]. In PDT, photosensitizers activated by light produce reactive oxygen species (ROS) that oxidatively damage biomolecules, disrupting cell integrity and inducing apoptosis [9, 10, 11, 12]. PTT, meanwhile, converts light energy into heat via photothermal agents, enabling precise tumor ablation. However, the hypoxic tumor microenvironment weakens PDT, and heat‐shock protein upregulation reduces PTT efficacy [13, 14, 15]. Combining PDT and PTT, particularly under multimodal imaging guidance, such as fluorescence (FLI), photoacoustic (PAI), and photothermal (PTI) imaging, can provide complementary effects and enable real‐time monitoring, thereby enhancing therapeutic outcomes [16, 17, 18].

Among photosensitive materials, organic small‐molecule fluorophores stand out for their defined structures, biocompatibility, and favorable pharmacokinetics [19, 20, 21, 22]. Cyanine dyes, with high molar extinction coefficients and tunable properties, are widely used in imaging and therapy [23, 24, 25]. Hemicyanines, an asymmetric subclass, feature a donor–π–acceptor (D–π–A) structure consisting of a nitrogen‐containing heterocycle, π‐conjugated bridge, and aromatic donor [26]. Their modular architecture allows precise tuning of electronic and photophysical properties, making them excellent scaffolds for structure–property studies and functional probe development [27]. To date, most hemicyanine dyes have employed nitrogen‐containing heterocycles (e.g., indole, benzothiazole, or quino‐line) as electron acceptors, coupled with hydroxyl‐, alkoxy‐, or amino‐substituted aromatic donors [28, 29, 30]. Their structural diversity is typically achieved by extending π‐conjugation or modulating intramolecular electronic effects, both effective strategies for tailoring molecular behavior [31, 32, 33]. However, designing and synthesizing hemicyanines with tunable architectures and long‐wavelength emission remains challenging [34]. In 2012, Lin et al. introduced a novel class of near‐infrared (NIR) hemicyanines by reacting chlorine‐substituted cyanines with resorcinol to adjust their emission wavelengths [35]. Over the past decade, substantial progress has been made in NIR dyes derived from hemicyanine scaffolds [36, 37, 38, 39, 40, 41]. Nevertheless, while the NIR‐I window (700–900 nm) offers deeper tissue penetration and reduced background autofluorescence relative to the visible region, it remains limited by photon absorption, scattering, and shallow penetration [42, 43, 44]. By contrast, the NIR‐II region reduces photon–tissue interactions, significantly improving imaging depth, spatial resolution, and signal‐to‐background ratio (SBR) [45, 46, 47, 48, 49, 50, 51]. Yet, most hemicyanines still operate in the NIR‐I range, and efficient NIR‐II emissive analogues are exceedingly rare. Hence, developing novel hemicyanine dyes with bright NIR‐II emission and multifunctional integration is of considerable significance.

To meet this challenge, a series of 2‐azaazulenium hemicyanine dyes (Br, OH, and OCH₃) were rationally designed and synthesized, achieving efficient NIR‐II fluorescence and laying the groundwork for multimodal imaging‐guided phototherapy (Figure 1a). Among these, SWJT‐OCH₃ exhibited absorption and intense NIR‐II emission, along with outstanding PDT and PTT performance. When encapsulated into biocompatible nanoparticles (SWJT NPs), the dye showed enhanced NIR‐II fluorescence, ROS generation with features of both type I and type II photodynamic processes, and a photothermal conversion efficiency (PCE) of 66.2%. SWJT NPs enabled high‐resolution lymph node imaging and multimodal tumor visualization, including FLI, PAI, and PTI. In vitro and in vivo studies further confirmed their robust PDT/PTT combined effects, effectively eradicating cancer cells (Figure 1b). Moreover, capitalizing on the multifunctional platform of 2‐azaazulenium hemicyanine, SWJT‐Probe was prepared, an activatable NIR‐II fluorescent probe capable of highly sensitive and selective detection of peroxynitrite (ONOO⁻) both in vitro and in vivo, successfully applying it to oxidative stress imaging in tumor models. Collectively, a new class of 2‐azaazulenium hemicyanine dyes with NIR‐II emission was developed, highlighting their promising applications in multimodal imaging (FLI/PAI/PTI), combined phototherapy, and selective ONOO⁻ detection. This study provides valuable molecular design insights for the development of multifunctional NIR‐II fluorophores with potential for clinical translation.

FIGURE 1.

(a) Chemical structures of hemicyanine‐based NIR fluorophores. (b) Schematic illustration of SWJT NPs for multimodal imaging–guided combined phototherapy.

2. Results and Discussion

2.1. Molecular Design and Synthesis

To achieve longer‐wavelength emission, a new class of D–π–A hemicyanine dyes was designed using a 2‐azaazulene (cyclohepta‐[c]pyrrole) moiety as the electron acceptor, with tailored donor structures to enable NIR‐II emission. The 2‐azaazulene acceptor features an extended non‐benzenoid π‐conjugated framework and pronounced acceptor character, which are expected to lower the electronic transition energy and thereby promote red‐shifted absorption and emission. Specifically, 2‐butyl‐1,3‐trimethylcycloheptane[c] pyrrole tetrafluoroborate was employed as the electron acceptor, a vinylene group served as the π‐bridge, and bromo‐, hydroxy‐, or methoxy‐substituted dihydroquinoline was selected as the electron donor to synthesize three target molecules: SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ (Figure 2a). A concise structural comparison with representative previously reported hemicyanine dyes is provided in Table S1.

FIGURE 2.

(a) The chemical structures of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃. (b), (c) Absorption and fluorescence spectra of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ (10.0 µM, λ _ex = 808 nm) in DMSO. (d) Relative fluorescence intensity (F/F₀) of DCFH at 525 nm in the presence of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃. (e) Relative absorbance (A/A₀) of DPBF at 410 nm after irradiation in the presence of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃. (f) Fluorescence intensity ratio (F/F₀) of DHR 123 for O₂ ^•− detection in the presence of SWJT‐Br, SWJT‐OH and SWJT‐OCH₃. (g) Photothermal heating curves of water, SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ (20.0 µM) in aqueous solution under 808 nm irradiation (1.0 W·cm⁻²). (h)–(j) Frontier molecular orbitals (HOMO/LUMO) and energy levels of (h) SWJT‐Br, (i) SWJT‐OH, and (j) SWJT‐OCH₃ determined by DFT/B3LYP method. (k)–(m) Singlet and triplet excited state geometries of (k) SWJT‐Br, (l) SWJT‐OH, and (m) SWJT‐OCH₃ by the TD‐DFT calculations. Data are shown as mean ± SD (n = 3).

To the best of our knowledge, no literature has yet reported a complete synthetic route for 2‐butyl‐1,3‐trimethylcycloheptane[c] pyrrole tetrafluoroborate, which may partly explain why 2‐azaazulenium cyanine dyes have remained underexplored [52, 53, 54, 55]. Herein we report the first peer‐reviewed synthesis and full characterization of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ (Scheme S1), building on our previous patent disclosure [56]. All unreported intermediates and final compounds were fully characterized by ¹H NMR, ¹³C NMR, high‐resolution mass spectrometry (HRMS), and high‐performance liquid chromatography (HPLC) (Figures S1–S18).

2.2. Photophysical, Photodynamic, and Photothermal Properties

The UV–visible absorption and fluorescence spectra of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ in various solvents were first examined (Figures 2b,c, and S19). Among the three dyes, SWJT‐OCH₃ displayed the highest absorption intensity at its maximum absorption wavelength and the strongest fluorescence at 904 and 1000 nm, attributed to the electron‐donating effect of the methoxy group. Next, the molar extinction coefficients (ε) and absolute fluorescence quantum yields (Φ_F) of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ are provided in Table S2. Correspondingly, SWJT‐OCH₃ exhibited the largest molar extinction coefficient (ε) and the highest Φ_F, consistent with enhanced light absorption and fluorescence efficiency.

Given their distinct optical behaviors, the photodynamic and photothermal characteristics of these compounds were systematically evaluated. The overall ability to generate ROS was assessed using 2′,7′‐dichlorodihydrofluorescein (DCFH) as a general probe. Under 808 nm laser irradiation (1.0 W·cm⁻²), all three compounds produced strong fluorescence signals, confirming efficient ROS generation. Notably, SWJT‐OCH₃ exhibited a slightly higher fluorescence enhancement, likely due to methoxy‐induced intramolecular charge transfer (ICT) and improved photophysical behavior (Figures 2d and S20). The production of singlet oxygen (¹O₂) was further evaluated with 1,3‐diphenylisobenzofuran (DPBF). After 5 min of 808 nm irradiation, the absorbance of SWJT‐OCH₃ at 415 nm decreased to 52%, a sharper decline than that observed for SWJT‐Br and SWJT‐OH (Figures 2e and S21). The relative singlet oxygen quantum yields (Φ) of the molecules were de‐termined using ICG as a reference and followed the order SWJT‐Br < SWJT‐OH < SWJT‐OCH₃ (Table S1). To probe type‐I ROS generation, dihydrorhodamine 123 (DHR 123) and 3,3′,5,5′‐tetramethylbenzidine (TMB) were used to detect superoxide anion (O₂ ^·−) and hydroxyl radicals (·OH), respectively. As shown in Figures 2f and S22, SWJT‐OCH₃ efficiently produced both O₂ ^·− and ·OH, supporting the involvement of radical‐related ROS. The photostability of the molecules was examined under continuous 808 nm laser irradiation using ICG as a reference, and SWJT‐OCH₃ displayed favorable photostability with negligible attenuation of its long‐wavelength absorption (Figure S23). Collectively, these results suggest that SWJT‐OCH₃ is capable of producing diverse ROS, implying the possible involvement of type‐I and type‐II related processes and underscoring its potential as a multifunctional and efficient photosensitizer.

In addition to photodynamic performance, the photothermal properties of the three dyes were examined. Upon 808 nm laser irradiation, the temperatures of aqueous solutions of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ increased to 43.2°C, 46.2°C, and 49.3°C, respectively (Figures 2g and S24). Remarkably, SWJT‐OCH₃ exhibited the largest temperature rise (ΔT = 27.0°C), which may stem from enhanced PCE induced by the methoxy substituent. To further investigate its photothermal regulation capacity, the temperature profiles and corresponding infrared thermal images of SWJT‐OCH₃ were recorded as functions of concentration and laser power density. At a constant power density of 1.0 W·cm⁻², the temperature elevation increased with concentration, demonstrating a clear concentration‐dependent photothermal effect (Figures S25 and S26). Moreover, after 5 min of irradiation, the solution temperature rose from 40.7°C to 52.6°C as the laser power density increased from 0.6 to 1.2 W·cm⁻² (Figures S27 and S28). The infrared thermal images corroborated this power‐dependent heating behavior. Together, these findings reveal that SWJT‐OCH₃ possesses excellent photophysical features along with efficient photodynamic and photothermal activities, highlighting its strong promise for phototherapeutic applications.

2.3. Theoretical Calculations

To gain further insight into the photodynamic mechanism, detailed density functional theory (DFT) calculations and electrostatic potential (ESP) analyses were performed [57, 58, 59]. As shown in Figures 2h–j and S29, the molecular geometries of SWJT‐Br, SWJT‐OH, and SWJT‐OCH₃ were optimized, and their frontier molecular orbitals (HOMO/LUMO) and molecular electrostatic potential (MEP) maps were obtained. The DFT results showed dense electron‐cloud distributions and relatively narrow HOMO–LUMO gaps for all three dyes, suggesting potential for long‐wavelength absorption. The computed HOMO–LUMO gaps decreased progressively from 2.16 eV (SWJT‐Br) to 2.08 eV (SWJT‐OH) and 2.06 eV (SWJT‐OCH₃), consistent with the increasing electron‐donating ability of the substituents. Compared with SWJT‐Br and SWJT‐OH, SWJT‐OCH₃ exhibited a more confined electrostatic potential range (27.32–70.96 kcal·mol⁻¹), suggesting a more uniform charge distribution in both ground and excited states. The 2‐azaazulenium hemicyanine dyes may facilitate light harvesting and intramolecular charge transfer, accounting for their superior photophysical behavior. To clarify the role of intersystem crossing (ISC) in ROS formation, the singlet–triplet energy gaps (ΔE _ST) were calculated (Figure 2k–m). A smaller ΔE _ST favors ISC and enhances ROS production via both type‐I and type‐II pathways. SWJT‐OCH₃ exhibited ΔE _(S1–T1) = 0.72 eV and ΔE _(S1–T2) = 0.26 eV, values lower than those of SWJT‐Br (0.86 and 0.36 eV) and SWJT‐OH (0.80 and 0.30 eV). These theoretical findings confirm that the favorable electronic configuration of SWJT‐OCH₃ contributes to its outstanding photophysical and phototherapeutic performance.

2.4. Preparation and Characterization of SWJT NPs

To enhance the aqueous solubility and biocompatibility of SWJT‐OCH₃, nanoparticles (SWJT NPs) were prepared using a DSPE‐mPEG₂₀₀₀ nanoprecipitation method (Figure 3a). The encapsulation efficiency reached 31.9%. Compared with free SWJT‐OCH₃, SWJT NPs exhibited a markedly enhanced fluorescence intensity at 1020 nm and showed a maximum absorption wavelength of approximately 780 nm (Figure 3b). Dynamic light scattering (DLS) analysis revealed an average hydrodynamic diameter of 78.82 nm (Figure 3c), a size that may favor accumulation at tumor sites. Transmission electron microscopy (TEM) further confirmed a uniform, monodisperse spherical morphology (Figure 3c, inset). In addition, the zeta potential of SWJT NPs was approximately −25.33 mV, indicating excellent colloidal stability and potential biocompatibility (Figure S30).

FIGURE 3.

(a) Schematic illustration of the self‐assembly process of SWJT NPs from SWJT‐OCH₃ and DSPE‐mPEG₂₀₀₀. (b) The absorption and fluorescence spectra of SWJT‐OCH₃ and SWJT NPs in PBS buffer solution (10.0 µM, λ _ex = 808 nm). (c) DLS and TEM images of SWJT NPs. (d) Time‐dependent fluorescence spectra of SWJT NPs in the presence of DCFH under 808 nm irradiation (1.0 W·cm⁻²). (e) EPR spectra for the detection of ¹O₂ (TEMP), O₂ ^·− (BMPO), and ·OH (BMPO in the presence of SOD) under 808 nm irradiation. (f) Photothermal heating curves of different concentrations of SWJT NPs under 808 nm irradiation (1.0 W·cm⁻²). (g) Photothermal heating curves and (h) infrared thermal images of SWJT NPs (20.0 µM) irradiated with 808 nm lasers at different power densities. (i) Calculation of the photothermal conversion efficiency (η) of SWJT NPs. (j) The photothermal stability of SWJT NPs under 808 nm irradiation (1.0 W·cm⁻²) through five heating–cooling cycles.

The capacity of SWJT NPs for ROS generation and photo‐thermal conversion was systematically investigated. As shown in Figures 3d and S31, after 808 nm laser irradiation, the fluorescence intensity of DCFH increased nearly 50‐fold, demonstrating efficient ROS production. Meanwhile, the absorbance of DPBF at 415 nm decreased to approximately 20%, whereas the absorption band near 780 nm remained almost unchanged, indicating efficient ¹O₂ generation and excellent photostability of SWJT NPs (Figure S32a,b). In addition, positive responses from O₂ ^·− and ·OH indicators were also detected (Figure S32c,d), further confirming the generation of multiple ROS under 808 nm irradiation. To further identify the specific ROS species, electron paramagnetic resonance (EPR) spectroscopy was employed: TEMP was used as a trap for ¹O₂, BMPO for O₂ ^·−, and BMPO in the presence of superoxide dismutase (SOD) for ·OH. Distinct signals corresponding to ¹O₂, O₂ ^·−, and ·OH were clearly observed during 808 nm irradiation of SWJT NPs (Figure 3e), indicating that both type I and type II pathways coexist in the ROS generation process. These findings suggest that SWJT NPs can induce the production of multiple ROS via combined mechanisms. Beyond photodynamic activity, the photothermal properties of SWJT NPs were comprehensively examined. The temperature elevation of SWJT NP solutions was recorded under different concentrations and laser power densities (Figure 3f,g). PTI clearly revealed the corresponding thermal signals (Figures 3h and S33). The photothermal heating of SWJT NPs showed a positive correlation with both particle concentration and laser power density. Under 808 nm laser irradiation for 300 s, the temperature of SWJT NP solutions increased to 68.3°C as the concentration rose from 25.0 µg·mL⁻¹ to 150.0 µg·mL⁻¹. Similarly, in a 100.0 µg·mL⁻¹ solution, increasing the laser power from 0.6 W·cm⁻² to 1.2 W·cm⁻² elevated the temperature to 66.3°C. The photothermal conversion efficiency (η) of SWJT NPs reached 66.2%, indicating favorable NIR‐II photothermal performance. Moreover, SWJT NPs exhibited excellent photothermal stability (Figure 3i,j). Overall, the combined photodynamic and photothermal features observed in vitro demonstrate that SWJT NPs possess substantial potential for advanced biomedical applications.

2.5. Antitumor Performance of SWJT NPs in Living Cells

Building on their photodynamic and photothermal capabilities, the therapeutic effect of SWJT NPs was subsequently evaluated at the cellular level under 808 nm irradiation. Cell viability was assessed using the CCK‐8 assay, which revealed high survival rates of HeLa cells treated with SWJT NPs within the concentration range (0–150 µg·mL⁻¹), indicating negligible cytotoxicity (Figures 4a and S34). Then, the antitumor effects of SWJT NPs were investigated at the cellular level. As shown in Figure 4a, after co‐incubation with SWJT NPs (0–150 µg·mL⁻¹), HeLa cells were subsequently divided into three groups to evaluate the effects of PDT and PTT, including cells irradiated at 4°C for PDT evaluation, cells pretreated with N‐acetylcysteine (NAC, a ROS scavenger) before irradiation for PTT assessment, and cells receiving the combined therapy of PDT and PTT. The PDT, PTT, and PDT+PTT groups exhibited dose‐dependent cytotoxicity, with half‐maximal inhibitory concentrations (IC₅₀) of 174.7, 90.7, and 64.9 µg·mL⁻¹, respectively (Figure S35). Notably, at 150 µg·mL⁻¹, cell survival rates were approximately 53% and 23% for the PDT and PTT groups, respectively, while the combination therapy reduced survival to 7%, highlighting a pronounced combined effect.

FIGURE 4.

(a) Cell viability of HeLa cells incubated with different concentrations of SWJT NPs with or without 808 nm laser irradiation (mean ± SD, n = 6). (b) Intracellular ROS and O₂ ^·− generation in SWJT NPs treated HeLa cells. (c) Confocal images of differently treated HeLa cells after co‐staining SWJT NPs with calcein‐AM/PI. (d) Mitochondrial membrane potential of HeLa cells detected by JC‐1 after different treatments. (e) Flow cytometry analysis of HeLa cells apoptosis after different treatments. Irradiation conditions: 808 nm laser, 1.0 W·cm⁻², 5 min. Green channel: λ _ex = 488 nm and λ _em = 500–550 nm; Red channel: λ _ex = 561 nm and λ _em = 600–650 nm.

Then, 2',7'‐dichlorodihydrofluorescein diacetate (DCFH‐DA) was used to assess the overall intracellular ROS generation induced by SWJT NPs under 808 nm irradiation, while dihydroethidium (DHE) was employed for specific detection of O₂ ^·− production (Figure 4b). To comprehensively evaluate the phototherapy effect of SWJT NPs under different conditions, the cells were divided into six groups: PBS, PBS + laser, SWJT NPs, SWJT NPs + laser + 4°C (PDT), SWJT NPs + laser + NAC (PTT), and SWJT NPs + laser (PDT + PTT). As shown in Figure 4b, in the presence of DCFH‐DA or DHE, the PDT, PTT, and PDT + PTT groups showed green fluorescence or red fluorescence, respectively. The fluorescence intensity of the PDT + PTT group was the strongest, while no fluorescence was observed in the control groups (PBS, PBS + laser, SWJT NPs). To assess whether ROS‐associated effects were retained under reduced irradiance, we performed low‐power studies at 808 nm (0.1–0.3 W·cm⁻², 5 min). SWJT NPs (100 µg·mL⁻¹) induced a power density–dependent reduction in HeLa cell viability and a concomitant increase in intracellular ROS (DCFH‐DA), both of which were attenuated by NAC (Figure S36). These results proved that SWJT NPs could efficiently induce intracellular ROS generation under laser irradiation.

Based on the literature, hemicyanine and cyanine dyes are well targeted to mitochondria due to their positive charge [60, 61]. Here, we investigated whether SWJT NPs affected mitochondrial membrane potential (MMP) using JC‐1 staining. In the polarized state of mitochondria, JC‐1 aggregates into J‐aggregates and emits red fluorescence, while in the depolarized or damaged state, it remains a monomer and emits green fluorescence. As shown in Figure 4d, HeLa cells in the control groups (PBS, PBS + laser, SWJT NPs) retained red fluorescence, indicating normal MMP. In contrast, HeLa cells treated with SWJT NPs and then exposed to 808 nm laser for 5 min showed pronounced green fluorescence, indicating depolarization of mitochondrial membrane potential. A key indicator of cell damage and apoptosis is the depolarization of mitochondrial membrane potential, and the Annexin V‐FITC/PI Assay Kit can be used to analyze the cell death mechanisms in depth. As shown in Figure 4e, cells were treated with SWJT NPs and 808 nm laser irradiation. Flow cytometry (FCM) analysis revealed that about 48.10% of the cancer cells were successfully induced to apoptosis, while only slight apoptosis and necrosis were observed in the other groups. Consistent with the FCM results, Western blot analysis revealed a decrease in pro‐caspase‐3, pro‐caspase‐9, and pro‐PARP in the SWJT NPs + laser group, accompanied by increased levels of the corresponding cleaved active forms (cleaved caspase‐3, cleaved caspase‐9, and cleaved PARP) (Figure S37). Taken together, SWJT NPs induced tumor cell death through the combined effects of PDT and PTT, showing significant in vitro antitumor activity.

2.6. Multimodal Imaging Performance of SWJT NPs

In view of the promising anti‐tumor efficacy of SWJT NPs in cellular assays, a systematic study was conducted on their multimodal imaging properties to evaluate their potential for in vivo applications. Recognizing the pivotal role of the lymphatic system in tumor metastasis, the fluorescence imaging capability of SWJT NPs in lymph nodes and lymphatic vessels was first evaluated. As shown in Figures 5a and S38, after subcutaneous injection of SWJT NPs near the footpads, fluorescence imaging enabled clear visualization of the popliteal (ROI‐1) and sciatic (ROI‐2) lymph nodes, as well as the connecting lymphatic vessels (ROI‐3). Within 2 h, the fluorescence signals in ROI‐1 and ROI‐2 exhibited negligible attenuation, indicating stable and long‐lasting fluorescence of SWJT NPs (Figure 5b). Meanwhile, the high SBRs of the five primary lymphatic vessels measured in ROI‐3 confirmed the efficient mapping of lymphatic drainage pathways (Figure 5c).

FIGURE 5.

(a) Time‐resolved NIR‐II fluorescence imaging of the popliteal (ROI‐1) and sciatic (ROI‐2) lymph nodes, along with the connecting lymphatic vessels (ROI‐3). (b) Quantitative analysis of fluorescence intensities in ROI‐1 and ROI‐2 at the indicated times. (c) Fluorescence intensities and corresponding SBRs of five lymphatic vessels. (d) PA images at different times following the intravenous injection of SWJT NPs. (e) Quantified PA signal intensities at the tumor site corresponding to (d). (f), (g) Temperature elevation curves at the tumor site and corresponding representative infrared thermal images of HeLa tumor‐bearing mice during continuous 808 nm irradiation (1.0 W·cm⁻²), measured 8 h after intravenous injection of SWJT NPs and saline. Data are presented as mean ± SD (n = 3).

Given the favorable lymphatic imaging performance, the feasibility of SWJT NPs for multimodal tumor imaging was subsequently evaluated using a subcutaneous HeLa tumor model. NIR‐II fluorescence imaging showed that upon tail vein injection of SWJT NPs into HeLa tumor‐bearing nude mice, the tumor region exhibited a progressively enhanced FLI signal, which peaked at 8 h and remained clear and detectable for up to 36 h. This time‐dependent accumulation confirmed the effective passive tumor targeting and accumulation of SWJT NPs (Figures S39 and S40). Ex vivo imaging of major organs at 36 h post‐injection further confirmed the effective and sustained accumulation of SWJT NPs in tumor tissue (Figure S41). Meanwhile, pronounced fluorescence signals were observed in the liver and spleen, indicating these organs are the primary organs involved in the metabolism of SWJT NPs [62]. Moreover, in vivo fluorescence imaging after intratumoral injection showed sustained accumulation of SWJT NPs in tumor tissue for up to 36 h, which was further confirmed by ex vivo imaging of major organs (Figures S42–S44). As expected, the photoacoustic imaging results were consistent with the fluorescence imaging data from the intravenous injection model, showing a similar in vivo distribution pattern and corresponding ex vivo organ distribution (Figures 5d,e, and S45). The optimal tumor accumulation window of SWJT NPs was determined through in vivo FLI and PAI analyses. Accordingly, PTI was performed at this time point to ensure efficient photothermal conversion and clear tumor contrast. The infrared thermal images and the corresponding temperature curves over time revealed the temperature of the tumor site rapidly increased from 37.3°C to 54.6°C within 6 min of 808 nm irradiation (Figure 5f,g). The rapid and significant thermal elevation confirmed that SWJT NPs had excellent photothermal conversion efficiency in vivo, reaching the therapeutic temperature range required for effective tumor ablation. In contrast, the control group treated with saline showed a negligible temperature elevation under the same irradiation conditions. These results showed that SWJT NPs had prolonged tumor retention, enabling complementary multimodal imaging through NIR‐II FLI, PAI, and PTI.

2.7. SWJT‐Probe for ONOO^‒ Detection and NIR‐II Imaging

Encouraged by the excellent multimodal imaging capabilities of SWJT NPs, the imaging of key biomarkers related to oxidative stress in the visualized tumor microenvironment had been studied. Among the biomarkers, ONOO⁻ has attracted widespread attention owing to its critical involvement in oxidative stress‐related pathologies. Therefore, a NIR‐II fluorescent small‐molecule probe, designated as SWJT‐Probe, was rationally designed and synthesized using the 2‐azaazulenium hemicyanine skeleton. The synthetic procedure was shown in Scheme S2, and the structure of SWJT‐Probe was verified by detailed ¹H NMR, ¹³C NMR, HRMS, and HPLC analyses (Figures S46–S49).

The optical sensing performance of the SWJT‐Probe toward ONOO⁻ was further investigated. As shown in Figure 6a, upon the addition of ONOO⁻ to the solution containing SWJT‐Probe, a distinct absorption peak emerged at 615 nm, accompanied by a clear color change from pale yellow to light blue. The fluorescence intensity at 979 nm gradually increased with the increase of ONOO⁻ concentration, which was 15‐fold stronger than that of the SWJT‐Probe alone (Figure 6b). The calibration curve related to fluorescence intensity and ONOO⁻ concentration was established, and the limit of detection (LOD) was calculated to be 31.69 nM by the 3σ/k method (Figure S50a). As shown in Figures 6c and S50b, the fluorescence responses of SWJT‐Probe to other analytes under the same condition were then measured. The analytes include metal ions (K⁺, Na⁺, Cu^{2 +}, Mg^{2 +}), inorganic anions (SO₄ ²⁻, HSO₄ ⁻), reactive sulfur species (HS⁻, S₂ ⁻, Cys, Hcy, GSH), and representative ROS (ClO⁻, H₂O₂, O₂ ^·−, ·OH). Only ONOO⁻ caused significant fluorescence enhancement at 979 nm, revealing the excellent selectivity of SWJT‐Probe. The proposed reaction pathway between SWJT‐Probe and ONOO⁻ was illustrated in Figure 6d, and the proposed mechanism was validated by HPLC and HRMS analysis (Figures S51 and S52). After adding ONOO⁻ to the probe solution, a new chromatographic peak was observed, accompanied by a signal of m/z = 466.2709, consistent with the calculated mass of the proposed product SWJT‐OH.

FIGURE 6.

(a) Absorption spectra of SWJT‐Probe (10.0 µM) in the absence and presence of ONOO⁻ (10.0 µM) in PBS buffer (pH = 7.4, containing 30% MeOH and 1% Tween‐80). (b) Fluorescence emission spectra of SWJT‐Probe (10.0 µM) in the presence of ONOO⁻ with different concentrations (0‐10.0 µM) (λ _ex = 808 nm). (c) Fluorescence responses of SWJT‐Probe (10.0 µM) toward various analytes (10.0 µM). (d) The mechanism of SWJT‐Probe for the detection of ONOO⁻. (e) In vivo fluorescence imaging of mice after administration of SWJT‐Probe under three conditions: healthy, tumor‐bearing, and tumor‐bearing with NAC pretreatment. (f) Quantitative analysis of fluorescence intensities corresponding to (e). Data are presented as mean ± SD (n = 3).

The SWJT‐Probe was tested for biological toxicity by the CCK‐8 assay, which exhibited negligible toxicity in living cells (Figure S53). Then, the potential of SWJT‐Probe for NIR‐II fluorescence imaging of ONOO⁻ in vivo was evaluated (Figure 6e). Strong fluorescence signals were observed at the tumor site in HeLa tumor‐bearing mice, while only weak signals appeared in the healthy control group. Notably, the group pretreated with NAC significantly reduced the fluorescence intensity within tumor regions. As shown in Figure 6f, quantitative analysis further confirmed that the in vivo fluorescence activation of SWJT‐Probe correlated with ONOO⁻ levels, thereby allowing the probe to selectively detect ONOO⁻ in tumors. The SWJT‐Probe exhibited sensitive and selective detection of ONOO⁻ in vitro and successfully visualized its dynamic fluctuations in the tumor microenvironment in vivo by NIR‐II fluorescence imaging.

2.8. Combined Phototherapy of SWJT NPs

Based on the effective in vitro antitumor activity and outstanding multimodal imaging capabilities of SWJT NPs, a subcutaneous HeLa tumor model was established in BALB/c nude mice to evaluate their in vivo antitumor efficacy (Figure 7a). The mice were randomly divided into four treatment groups according to the established tumor model and received treatment with PBS, PBS + laser, SWJT NPs, and SWJT NPs + laser. As shown in Figure 7b, no significant body weight changes were observed in any group throughout the treatment period. As shown in Figure 7c, during the 15‐day treatment period, the tumor volumes of the control groups (PBS, PBS + laser, and SWJT NPs) increased rapidly. In contrast, the SWJT NPs + laser group exhibited significant tumor suppression with no evidence of recurrence, demonstrating effective tumor ablation. After treatment, tumors were removed and weighed to evaluate the therapeutic efficacy. The ex vivo tumor weights and corresponding tumor images indicated that the SWJT NPs + laser group exhibited significantly greater tumor suppression effects than the control groups (Figure 7d,e).

FIGURE 7.

(a) Schematic illustration of the timeline for in vivo experiments. (b) Changes in mice body weight and (c) tumor volume throughout the treatment process. (d) Weight and (e) digital photographs of tumor removed 14 days after treatment (mean ± SD, n = 5), ***p <0.001. (f) Histological analysis of tumor sections by H&E, TUNEL, CD31, and Ki67 staining offer different treatment interventions. (g) Quantitative analysis of the positive rates of TUNEL, CD31, and Ki67 stained cells corresponding to (f).

To further elucidate the underlying mechanism of SWJT NPs mediated in vivo tumor eradication, histological and immunohistochemical analyses were conducted. H&E staining of the major organs (including the heart, liver, spleen, lungs, and kidneys) showed no discernible histopathological abnormalities in all treatment groups, confirming the high biocompatibility and good biosafety of SWJT NPs (Figure S54). As shown in Figure 7f, tumor sections of the SWJT NPs + laser group exhibited extensive histopathological disruption in H&E stains, compared with other groups. TUNEL assays, which detect DNA strand breaks as a hallmark of apoptosis, further confirmed the significant increase in apoptotic cells in the SWJT NPs + laser group, validating the enhanced apoptosis of tumor cells induced by phototherapy. In addition, immunohistochemical results revealed that the combination of SWJT NPs and 808 nm laser treatment significantly reduced CD31positive microvessel density and Ki‐67 proliferation index, indicating significant suppression of both tumor angiogenesis and cell proliferation (Figure 7f,g). Overall, SWJT NPs showed potent and precise combined PDT‐PTT therapeutic effects under 808 nm irradiation, resulting in highly effective ablation of cervical cancer.

3. Conclusion

In summary, we report 2‐azaazulenium hemicyanine dyes as a new class of D–π–A NIR‐II fluorophores with tunable structures and multifunctional phototherapeutic potential. Rational hybridization of 2‐azaazulene with dihydrothiazine afforded SWJT derivatives, among which SWJT‐OCH₃ shows bright NIR‐II emission, robust ROS generation, and outstanding photothermal efficiency. Encapsulation with DSPE‐mPEG₂₀₀₀ yielded biocompatible SWJT NPs with ROS generation capability, a photothermal conversion efficiency of 66.2%, and high‐resolution multimodal tumor imaging (FLI/PAI/PTI). Under 808 nm irradiation, SWJT NPs achieved potent combined PDT/PTT against cervical tumors, while an SWJT‐based probe enabled sensitive ONOO⁻ detection in vivo. This work broadens the design space of hemicyanine‐based NIR‐II platforms for imaging, therapy, and activatable sensing.

Author Contributions

Qiang Wang: methodology. Juyoung Yoon: writing – review and editing, project administration. Hai Xu: writing – original draft, project administration, resources. Yu Peng: writing – review and editing, writing – original draft, project administration. Min Peng: writing – original draft, data curation, project administration, methodology. Ya‐Wen Wang: project administration, writing – review and editing, writing ‐ original draft. Rong‐Dan Xiao: methodology. Rui Tan: methodology, supervision. Bing‐Ya Wang: methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: Experimental procedures, structural characterization data, and additional figures and schemes (PDF).

Dedicated to the 130th Anniversary of Southwest Jiaotong University and the 140th Anniversary of Ewha Womans University.

Data Availability Statement

The data that supports the findings of this study are available in the supporting information of this article.