RESEARCH ARTICLE

[Link]

Synergistic Comprehensive Activation Methods for

Dual-Modality PDT and Hypoxia-Triggered Chemotherapy
Guided by NIR-II Imaging beyond 1700 nm in Deep Tumors
Haoran Chen, Shengnan Liu, Wang Wang, Fei Long, Qiqing Li, Dechao Gan, Xiaodan Li,
Bin Li, Xianggui Kong, Dabing Li, and Yulei Chang*

1. Introduction
Although photodynamic therapy (PDT) holds great promise for applications
in cancer treatment, it has limited effectiveness against deep hypoxic tumors. Photodynamic therapy (PDT) is highly fa-
Moreover, the lack of visualization guidance for precision theranostics vored for tumor therapy due to its non-
invasive nature, high spatiotemporal selec-
poses additional challenges, hindering its broader clinical adoption. By
tivity, and repeatability.[1] Traditional clin-
combining NIR-IIc (1800 nm) imaging guidance with internally and externally ically approved PDT, such as hematopor-
activatable dual-modality PDT and hypoxia-triggered chemotherapy, this phyrin derivative, primarily following the
study proposes a conceptual framework to overcome these limitations. This Type-II pathway, utilizes small molecule
approach involves the use of photoswitchable lanthanide-doped nanoparticles photosensitizer (PS) to convert O2 into
singlet oxygen (1 O2 ) under specific light
featuring Tm3+ -activated upconversion/downshifting emissions coupled with
excitation.[2] However, it has limited efficacy
carboxyl-terminated Ir(III) complex-based Type I/II photosensitizer to form when treating deep hypoxic tumors. Most of
a nanophotosensitizer. The findings demonstrate that this system enabled the PSs used in clinic can be activated under
NIR-IIc imaging guidance upon 808/980 nm excitation while selectively UV–vis light.[3] The irradiance and flux of
activating external PDT under 980 nm irradiation, thereby ensuring accurate the excitation light are significantly attenu-
therapy and minimizing phototoxicity risk. The Ir(III) complex conjugates with ated while passing through the tissue, often
falling below the threshold required for ef-
luminol to form a self-illuminating Type I/II photosensitizer, which can respond
fective tumor tissue ablation. Additionally,
to the elevated H2 O2 levels in the tumor microenvironment, effectively traditional Type-II PDT, whose efficiency is
catalyzing chemiluminescence-assisted PDT. Moreover, PDT aggravates completely dependent on the amount of O2
tumor hypoxia, which in turn activates the hypoxia-activatable prodrugs in the microenvironment, exacerbates hy-
like tirapazamine, resulting in a synergistic antitumor effect. With NIR-IIc poxia, further decreasing the efficacy of the
antitumor effect.[4] Moreover, efficient and
imaging-guided dual-modality PDT, this study introduces a groundbreaking
targeted PDT requires visual imaging guid-
approach that unites Type I/II PDT with chemotherapy, significantly ance to accurately locate the tumor site, de-
advancing the precise and effective treatment of deep hypoxic tumors. termine the optimal timing of photoirradia-
tion, and monitor the treatment response in
real time.[5] However, traditional PS-based
fluorescence imaging/diagnostic methods can harm healthy,
H. Chen, W. Wang, F. Long, Q. Li, D. Gan, B. Li, X. Kong, D. Li, Y. Chang
State Key Laboratory of Luminescence Science and Technology non-cancerous tissues and skin due to their risk of photosensi-
Changchun Institute of Optics tive toxicity. Therefore, it is the need of the hour to find more
Fine Mechanics and Physics effective strategies targeting deep tumors. One of the main
Chinese Academy of Sciences barriers to deep-seated tumor PDT is achieving effective light
Changchun 130033, P. R. China
E-mail: yuleichang@[Link] transmission, particularly for tumors of varying sizes and loca-
S. Liu tions. In recent times, significant progress has been made in
School of Pharmacy addressing this challenge, including the use of diverse excita-
Shenyang Pharmaceutical University tion light sources to enhance PDT efficacy.[6] In addition, X-ray
103 Wenhua Road Shenhe District, Shenyang, Liaoning 110016, P. R. China as an excitation source of deep PDT often faces biosafety issues
X. Li caused by ionizing radiation. Alternatives such as NIR light- and
Department of Respiratory Medicine
bio/chemiluminescence-driven PDT offer a promising and safer
the First Hospital of Jilin University
Changchun 130021, P. R. China solution for the treatment of deep tumors.[7]
Recently, the expanding of the NIR-II bioimaging window
The ORCID identification number(s) for the author(s) of this article (1000–2300 nm), especially beyond 1700 nm, led to the in-
can be found under [Link] troduction of the NIR-IIc (1700–2000 nm) and NIR-IId/NIR-
DOI: 10.1002/smll.202500553 III (2100–2300 nm) subwindows.[8] These subwindows provide

Small 2025, 21, 2500553 2500553 (1 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

subcentimeter tissue information by reducing scattering and ficacy of the system. This multifaceted approach combined the
autofluorescence compared to visible light, thereby enhancing advantages of advanced NIR-II luminescence imaging, dual-
phototheranostic capabilities. For this purpose, lanthanide-doped modality PDT, and chemotherapeutic synergy, providing a so-
nanoparticles (LnNPs) emerged as a promising candidate for phisticated and highly targeted strategy to overcome the chal-
deep-depth luminescence imaging as they are among the few ma- lenges of deep hypoxic tumor therapy and push the boundaries
terials emitting beyond 1700 nm.[9] Examples of such lanthanides of precision oncology.
include Tm3+ , Ho3+ , and Dy3+ ions.[10] The “photoswitchable”
LnNPs were achieved through selective excitation regulation. 2. Results and Discussion
They were strategically engineered to separate upconversion (for
sensitizing PSs) from downshifting emissions (for NIR-II imag- Through the rational design of a core–shell–shell (CSS) nanos-
ing). As a result of coupling with the PSs, Forster resonance tructure (LiTmF4 :0.5%Er@LiYbF4 :0.5%Tm@LiYF4 ), this study
energy transfer (FRET) mechanism-based nanophotosensitizers proposed Tm3+ -activated photoswitchable LnNPs, featuring both
were formed, enabling precise NIR-II imaging-guided PDT upon UV–vis upconversion luminescence (UCL) and 1800 nm down-
external NIR excitation (EPDT) while minimizing the photosen- shifting luminescence (DSL) through selective excitation control
sitivity toxicity generally associated with diagnostic imaging.[11] (808/980 nm). First, the LnNPs were synthesized via a high-
Chemiluminescence resonance energy transfer (CRET) of- temperature co-precipitation method. Subsequently, a 980 nm
fers an optimal approach for deep-tissue PDT, eliminating single wavelength excited Yb/Tm shell (S1 ) was epitaxially grown
the need for external light excitation.[12] By utilizing the ele- on the 808 and 1208 nm dual wavelength excited LiTmF4 :0.5%Er
vated H2 O2 levels in the tumor microenvironment (TME), a core. This was followed by coating an inert shell LiYF4 (S2 ) to
chemiluminescence-triggered PS system is reported for self- suppress surface quenching, thereby enhancing overall lumines-
chemiluminescence-triggered PDT (CPDT) designed for effec- cence brightness. Additionally, the shell thickness was optimized
tive deep tumor ablation.[12d,13] However, the current CDPT to achieve brighter luminescence intensity. With shell thickness
nanoplatform is limited by its reliance on the visible emission (S1 +S2 ) of 8.94 nm × 5.01 nm, 9.74 nm × 5.55 nm, and 11.27 nm
of the small molecule PS. These emissions suffer from tissue in- × 6.83 nm, the resulting structures were termed CSS1, CSS2, and
terference, leading to subpar imaging quality and preventing the CSS3, respectively.
CDPT nanoplatform from achieving precise diagnosis and high- As shown in Figure S1 (Supporting Information), the trans-
definition imaging guidance for precision therapy. Therefore, mission electron microscopy (TEM) images revealed that the
integrating TME-mediated CPDT with NIR-II imaging-guided core had a size of 21.64 nm × 14.57 nm. The TEM images
EPDT overcomes the challenges of shallow tissue penetration of shown in Figure 1a and Figure S1 (Supporting Information)
UV–vis imaging in traditional CPDT, resulting in a more effica- show that the mean nanoparticle sizes of LnNPs were 39.52
cious and precise deep-tumor therapy. nm × 24.59 nm, 41.12 nm × 25.67 nm, and 44.19 nm ×
This study reports the design of a novel “Internal and External 28.23 nm, respectively (Figure 1b). CCS3 LnNPs, exhibiting up-
Cultivation” PDT strategy by constructing a liposomal nanoplat- conversion/downshifting luminescence performance, were se-
form (LnNPs-Ir/Lum-Ir/TPZ@Liposome, LLT@Lip). The plat- lected for further characterization (See the spectral characteri-
form integrated CPDT and EPDT guidance with Tm3+ -activated zation section below). The high-resolution TEM (HRTEM) mea-
1800 nm NIR-IIc imaging. Powerful PDT is known to rapidly surement revealed a d spacing of 0.406 nm for CSS3 (inset of
consume O2 , aggravating severe hypoxia in TME and reduc- Figure 1a). Meanwhile, the diffraction peaks of LiTmF4 :0.5%Er
ing PDT efficiency, particularly during the later stages of treat- core and CSS3 nanoparticles were consistent with the reference
ment. In contrast, Ir(III) complexes-based EPDT and CPDT patterns of LiYF4 , indicating that LnNPs were pure tetragonal
PSs are designed to operate via both Type I/II mechanisms, phase nanocrystals (Figure 1c). Furthermore, due to the similar
thereby maintaining therapeutic efficacy even under hypoxic contrast between LiTmF4 :0.5%Er core and LiYbF4 :0.5%Tm inter-
conditions.[14] The EPDT PSs (LnNPs-Ir) were prepared by cou- mediate layer, the high-angle annular dark-field scanning trans-
pling Tm3+ -activated upconversion/downshifting LnNPs with mission electron microscopy (HAADF STEM results) (Figure S2,
the Ir(III) complex, while the CPDT PSs (Lum-Ir conjugate) were Supporting Information) revealed clear boundaries between the
synthesized by coupling the amino group of luminol with the bright LiYbF4 :0.5%Tm shell and the dark LiYF4 shell. As shown
carboxyl groups of the Ir(III) complex through amide conden- in Figure 1d and Figure S3 (Supporting Information), the ele-
sation. The two PSs were then encapsulated in the hydropho- mental mapping image and linear scanning data of CSS3 con-
bic lipid layer, along with the hypoxia-activated chemotherapy firmed its elemental distribution and core–shell structure.
prodrug tirapazamine (TPZ) in the internal aqueous phase, to The DSL properties of the prepared materials were next in-
synergize with PDT (Scheme 1). The findings of this study re- vestigated. As shown in Figure 1e, the characteristic DSL peaks
vealed that the 1800 nm luminescence imaging-guided exoge- of Tm3+ were concentrated at 1800 nm (from 1600–2100 nm),
nous 808/980 nm photoswitchable EPDT significantly enhanced corresponding to the 3 F4 →3 H6 transitions. The enhanced cross-
the spatial and temporal resolution of tumor imaging during relaxation (CR) in high-doped Tm3+ between 3 H4 →3 F4 and
treatment. Simultaneously, the Lum-Ir conjugate (containing two 3
H6 →3 F4 increases the population at the 3 F4 energy level, thereby
luminol molecules) activated CPDT in situ, driven by the el- significantly enhancing the DSL intensity. Meanwhile, selec-
evated levels of H2 O2 in TME, facilitating Type-I/II PDT. No- tive doping of 0.5% Er3+ enhances the energy downward relax-
tably, the Ir(III) complex-based photocatalysis also activated TPZ. ation pathway, thereby boosting the downshifting emission of
Consequently, the combination of E/CPDT and hypoxia-activated Tm3+ (Figure S4, Supporting Information). This was attributed
chemotherapy markedly improved the antitumor therapeutic ef- to the larger absorption cross-section of Yb3+ in the shell, which

Small 2025, 21, 2500553 2500553 (2 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Scheme 1. a) Schematic diagram of constructing an antitumor PDT/chemotherapy synergistic nanoplatform (LLT@Lip) and b) its antitumor therapeutic
mechanism in the deep hypoxic tumor. The LLT@Lip theranostic nanoplatform, responsive to external and internal comprehensive photoexcitation, was
developed using highly efficient LnNPs-Ir and Lum-Ir PSs, providing real-time, high-definition, and deep-depth imaging guidance.

absorbed 980 nm photons, followed by the energy transfer from blue UCL is observed upon 808 nm excitation during NIR-II
the 2 F5/2 level of Yb3+ to the 3 H5 level of Tm3+ level, resulting imaging. In contrast, both NIR-IIc luminescence and UV-blue
in more intense DSL upon 980 nm irradiation (Figure S5, Sup- UCL can be performed simultaneously after photoswitching to
porting Information). Figure 1e illustrates the energy transfer 980 nm excitation. Since CCS3 demonstrated optimal spectral
processes and mechanisms involved in upconversion and down- properties, these photoswitchable LnNPs could be highly effec-
shifting emissions. Notably, the 1800 nm DSL intensity of CSS2 tive as light transducers for deep tissue luminescence imaging
was found to be similar to CSS3 but significantly stronger than and PDT guidance, in accordance with the principles of FRET.
CSS1 upon 808 nm irradiation (Figure 1f). This difference was By coupling CSS3 LnNP with Ir(III) complex (LnNP-Ir)
attributed to the fact that the shell thickness of CSS2 inhibited through ligand exchange, which ensured the shortest energy
surface quenching more effectively compared to CSS1. transfer distance, the EPDT nanophotosensitizer was obtained.
The UCL performance of the as-prepared photoswitchable In a previous study, we synthesized and characterized this Ir(III)
LnNPs was also investigated. Upon 980 nm excitation, CSS3, complex PS with Type-I and II photochemical reactions.[15] The
with the thickest LiYbF4 :0.5%Tm luminescent layer, exhibited the results revealed a large spectral overlap of the absorption spec-
strongest UCL intensity. As shown in Figure 1g, characteristic tra of Ir(III) complex PS with the emission spectra of LnNPs,
UCL peaks were observed at 346, 360, 452, and 482 nm. The high which forms donor- acceptor pairs of FRET. Nonradiative energy
concentration of Tm3+ doping in the core leads to intense CR, is transferred from the excited LnNPs to Ir(III) complex, which
causing rapid energy transfer to the low-lying energy levels. Con- is sensitized to emit fluorescence and generate ROS.[16] The as-
sequently, very little energy is transferred to the LiYbF4 :0.5%Tm prepared EPDT PS (LnNP-Ir) demonstrated a high energy trans-
shell for UCL emission when excited at 808 nm. Thus, no UV- fer efficiency of 76% at 450 nm, with the emission at 350 nm

Small 2025, 21, 2500553 2500553 (3 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Figure 1. a) TEM and HRTEM images of LiTmF4 :0.5%Er@LiYbF4 :0.5%Tm@LiYF4 (CSS3). b) Size statistics of CSS1, CSS2 and CSS3 LnNPs. c) XRD
patterns of LiTmF4 :0.5%Er core and CSS3 LnNPs. d) Scanning transmission electron microscopy (STEM) and corresponding elemental (Er, Tm, Yb, Y)
mappings of CSS3 LnNPs. e) Energy level diagram of CSS3 LnNPs. f) DSL and g) UCL emission spectra of CSS3 and corresponding LnNP-Ir (CSS3-Ir),
along with the absorption spectrum of the Ir(III) complex.

almost entirely quenched. Additionally, the emission of the Ir(III) activation properties (Figure S6a, Supporting Information). Un-
complex at 615 nm also confirmed that the energy transfer was surprisingly, the UV–vis absorption spectra of Lum-Ir conjugate
effective. exhibited characteristic absorption peaks of Ir(III) complexes and
Although the LnNP-Ir PSs hold great potential for the precise luminol (Figure S6b, Supporting Information). Upon 380 nm
diagnosis and treatment of tumors in deep tissues, NIR-triggered excitation, the photoluminescence (PL) spectra of the conjugate
EPDT still has limitations in overcoming the challenge of pene- demonstrated strong fluorescence emission of luminol at 454 nm
tration depth. Fortunately, CPDT can address this limitation by and Ir(III) complexes at 615 nm. When H2 O2 oxidized the lu-
stimulating PSs through in situ chemical reactions. Since CPDT minol moiety in Lum-Ir, the significant spectral overlap between
can be performed at any depth, it has great potential in the an- the luminol emission and Ir(III) complex absorption ensured ef-
titumor therapy of deep tumors. Luminol is a classical chemi- ficient energy transfer, leading to ROS production (Figure S7,
luminescent substrate that can rapidly react with H2 O2 to form Supporting Information). The luminescence resonance energy
an excited intermediate of 3-aminophthalic acid. When this in- transfer ratio was determined to be 11.49 by a method reported
termediate returns to the ground state, it emits 425 nm lumines- previously.[17] Since the TME is weakly acidic, the luminescence
cence, indicative of its potential as a spontaneous activation PS. properties of Lum-Ir were studied under different pH conditions.
Therefore, luminol was coupled with Ir(III) complexes to form a As shown in Figure S6c (Supporting Information), the lumines-
self-luminous CPDT photosensitizer (Lum-Ir) with in situ tumor cent signals of Lum-Ir conjugate, both with or without 510 nm

Small 2025, 21, 2500553 2500553 (4 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Figure 2. a) TEM image of the LLT@Lip. Scale bar is 200 nm. b) DLS and zeta potential of LLT@Lip over 15 days. c,d) Chemiluminescence images
of LLT@Lip in different H2 O2 and nanoplatform concentrations. e) Consumption curves of singlet oxygen indicators DPBF with different penetration
depths when incubated with LLT@Lip under 980 nm irradiation. f) Fluorescence spectra of SOSG for different treatment conditions. g) ESR spectra of
DMPO after incubation with LLT@Lip, LnNP-Ir@Lip, and Lum-Ir@Lip under various 980 nm light and H2 O2 conditions. h) Fluorescence spectra of
DHR123 under different conditions. i) ABDA degradation curves under different conditions. j) Ir(III) complex reacts with NADH and TPZ in solution.
When Ir(III) complex, LnNP-Ir, and Lum-Ir were exposed to 380 nm light, 980 nm light, and H2 O2, respectively, the k) NADH and l) TPZ absorption
spectra were statistically significant.

LP filter, gradually increased with increasing pH value, consis- Lum-Ir and TPZ were 17.2% and 4.05%, respectively. As shown
tent with luminol’s properties. in Figure 2b, the dynamic light scattering (DLS) and Zeta poten-
The therapeutic efficacy against deep hypoxic tumors was sig- tial monitoring revealed that LLT@Lip maintained colloidal sta-
nificantly enhanced by combining hypoxic-activated antitumor bility for 15 days.
prodrugs, such as TPZ, with EPDT and CPDT. This approach The UV–vis absorption of TPZ was used to study the drug
enables the design of a versatile nano-theranostic nanoplatform release profile of LLT@Lip in pH 7 buffer. As shown in Figure
capable of achieving effective antitumor therapeutic efficacy. Cell- S10 (Supporting Information), typical liposome-mediated drug
like liposomes were used to load PSs and prodrugs, water-soluble release characteristics were observed, showing 10.9% and 7.3%
TPZ was loaded in the internal water phase, and hydrophobic cumulative release of TPZ and Lum-Ir within 26 h, thereby sug-
Lum Ir-and LnNP-Ir were loaded in the hydrophobic layer.[18] The gesting that the designed LLT@Lip was an ideal candidate for
molecular structures of Lum-Ir were validated by nuclear mag- systemic drug delivery. Moreover, the chemiluminescence signal
netic resonance (1H NMR) spectroscopy (Figure S8, Supporting of LLT@Lip was observed using a CCD camera. In the presence
Information). As shown in Figure 2a, the TEM data revealed that of H2 O2 , the LLT@Lip luminescence decayed exponentially over
the size of LLT@Lip was ≈210 nm, and CSS nanoparticles in the a 35 min period (Figure S11, Supporting Information). Thus the
hydrophobic layer can be clearly observed. Meanwhile, UV–vis decrease in the LLT@Lip and H2 O2 concentrations was accom-
absorbance and PL spectra were utilized to verify the successful panied by the weakening of the chemiluminescence intensity of
encapsulation of LnNPs-Ir, Lum-Ir, and TPZ into the liposomes. LLT@Lip (Figure 2c,d), implying that luminol successfully sensi-
The UV–vis spectrum of LLT@Lip revealed the characteristic ab- tized Ir(III) complex moiety in the presence of H2 O2 under acidic
sorption peaks of Luminol, Ir(III) complex, and TPZ at 290, 368, conditions. Thus, these results indicate that LnNPs and luminol
and 460 nm, respectively (Figure S9, Supporting Information). are efficient energy donors for the as-designed Ir(III) complex PS
The absorbance analysis revealed that the loading efficiencies of in PDT.

Small 2025, 21, 2500553 2500553 (5 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

The importance of combining EPDT and CPDT to overcome the combination of EPDT and CPDT, enabling effective therapy
the limitations of deep tumor PDT was further demonstrated for deep hypoxic tumors.
by investigating the relationship between 980 nm NIR-triggered In addition to generating photosensitization-mediated ROS,
ROS production of LLT@Lip and tissue depth. To achieve this, the Ir(III) complex PS can also deplete nicotinamide adenine din-
a 1% intralipid medium was used to mimic biological tissues, ucleotide (NADH), a key mitochondrial oxidant, through elec-
and ROS production at different thicknesses was measured us- tron transfer.[19] The depletion of NADH leads to disruption of
ing a DPBF probe. The results revealed that there was a threefold the NADH/NAD+ balance leading to oxidative stress in the can-
decrease in EPDT efficacy after passing through a 2 mm tissue cer cell, thereby enhancing the therapeutic effect (Figure 2j). As
barrier (Figure 2e), highlighting the need for integrating multiple shown in Figure S15 (Supporting Information), and Figure 2k,
PDT-activated methods to enhance treatment efficacy. the Ir(III) complex in LnNP-Ir and Lum-Ir treated groups was
The ROS generation profiles of LLP@Lip were evaluated un- sensitized to oxidize NADH to NAD• under 980 nm irradiation
der various treatment conditions. Ir(III), as a Type-I/II PS, can and H2 O2 , respectively. This was evident by the gradual decrease
sensitize O2 to form 1 O2 through energy transfer, or generate of the 350 nm peak. Notably, reductive Ir(II) (formed from Ir(III)
ROS, such as O2 •− through electron transfer to the substrate. By after electron loss) reduced TPZ to activate its cytotoxicity. After
utilizing the Type I/II photochemical processes of the Ir(III) com- photosensitization in the NADH environment, the characteristic
plex, the ROS production ability of LLT@Lip (including EPDT absorption peak of TPZ was gradually reduced by LnNP-Ir and
and CPDT) was thoroughly assessed. For Type II photodynamic Lum-Ir (Figure S16, Supporting Information; Figure 2l), suggest-
effects, a singlet oxygen sensor green dye (SOSG, peak emission ing that the Ir(III) complex has a photocatalytic effect on TPZ-
at 525 nm) was used as a chemical trapping agent for 1 O2 detec- induced toxicity.
tion. First, the Type-II PDT potential of LnNPs-Ir and Lum-Ir was Encouraged by the results discussed above, the antitumor ef-
confirmed using SOSG (Figure S12, Supporting Information; fects of LLT@Lip were investigated in vitro. Figure 3a shows the
Figure 2f). The conversion of SOSG into its highly fluorescent antitumor principle of LLT@Lip. The external 980 nm light acti-
SOSG endoperoxide suggested that both LnNPs-Ir under 980 nm vates EPDT, while endogenous overexpressed H2 O2 in TME trig-
irradiation and Lum-Ir with H2 O2 oxidation demonstrated robust gers CPDT. These two PDT processes simultaneously undergo
Type-II photodynamic activity. These results highlight the dual Type I/II PDT, consuming O2 in the TME, which in turn activates
capacity for Type-II PDT activation using both 980 nm excitation TPZ to initiate chemotherapy. Importantly, EPDT and CPDT still
and H2 O2 , effectively overcoming the limitation of depth pene- maintain the low oxygen dependence of Type-I PDT in hypoxic
tration of conventional PDT. environments, allowing LLT@Lip to achieve efficient tumor ther-
Since Type-II PDT (1 O2 generation) is more common in Ir(III) apy across both temporal and spatial dimensions. First, the cellu-
complexes, we focused on providing robust evidence for the less lar uptake and cellular distribution of LLT@Lip were determined.
common Type-I PDT mechanism. Electron paramagnetic reso- The blue UCL of LnNPs and red fluorescence of Ir(III) complexes
nance (EPR) and fluorescence probe methods were conducted confirmed the cellular uptake of LLT@Lip after 2 h of incubation.
to investigate Type-I photodynamic effects. Typically, EPR is the As shown in Figure 3b and its corresponding line-profile plot, the
most reliable technique for detecting short-lived ROS, employing graph quantifies the fluorescence intensity distribution of each
5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as a trapping agent for channel (G, B, R) along the white line segment in the merged
the detection of O2 •− . As shown in Figure 2g, a distinct six-line panel. The fluorescence signals of the Ir(III) complexes matched
EPR signal of equal intensity was observed for the DMPO/O2 •− well with the green signal of the Mito-tracker, which was indica-
adduct, confirming the presence of O2 •− . Next, the ability of tive of their mitochondria-targeting properties, which arise from
LLT@Lip to generate O2 •− was demonstrated using the dihy- the lipophilic moiety of Ir(III) complexes. This was further val-
drorhodamine 123 (DHR123) probe. DHR123 is non-fluorescent idated by the corresponding fluorescence intensity distribution
but emits green fluorescence upon reacting with O2 •− . As shown curve of Mito-G and Ir(III) complexes. Furthermore, the cellular
in Figure S13 (Supporting Information) and Figure 2h, strong uptake rate of LLT@Lip by 4T1 cells was quantified using DSL be-
green fluorescence signals were observed upon 980 nm irradia- fore and after incubation (Figure S17, Supporting Information).
tion and H2 O2 stimulation, respectively, thereby indicating that The uptake rates of LLT@Lip were determined 7% and 14% for
both Lum-Ir-based CPDT and LnNPs-Ir-based EPDT are capa- 2 and 6 h of incubation, respectively.
ble of generating O2 •− . Furthermore, LLT@Lip demonstrates a The effect of EPDT, CPDT, and chemotherapy on the cytotoxi-
higher rate of O2 •− production when both H2 O2 and 980 nm were city of 4T1 cells was evaluated by MTT assay. In a normoxic envi-
applied simultaneously. These results confirmed that the Ir(III) ronment, the cell survival rate was >90% at TPZ concentrations
complex demonstrated a significant Type I photodynamic pro- <2.5 μg mL−1 (Figure S18a, Supporting Information). However,
cess, enabling LLT@Lip to still have superior PDT efficacy even the cell survival rate dropped sharply by 50% when the O2 con-
in the later stages of treatment when hypoxia is more severe. centration decreased to 2%, indicating that TPZ was effectively
The total ROS level was evaluated using 9,10-anthracenediyl- activated under these conditions. Thus, the TPZ (2.5 μg mL−1 )
bis (methylene) and dimalonic acid (ABDA, absorption peak at was selected for further experiments in the study. As shown
380 nm). The data suggested that the LLT@Lip group maintained in Figure 3c, the survival rate in hypoxic conditions was 30%
the peak intensity, while groups exposed to 980 nm light or H2 O2 when the concentration of LnNPs-Ir was 200 μg mL−1 , com-
environment showed ABDA consumption of 29% and 24%, re- pared to the control and 808 nm irradiation groups. This was
spectively (Figure S14, Supporting Information; Figure 2i). No- because LnNP does not produce UCL under 808 nm excitation.
tably, when both stimuli were applied concurrently, 49% of ABDA Meanwhile, significant cell-killing effects were also observed for
was consumed. This enhanced ROS production resulted from the CPDT of Lum-Ir@Lip (Figure 3d). As a chemiluminescence

Small 2025, 21, 2500553 2500553 (6 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Figure 3. a) Schematic diagram of the antitumor therapy of LLT@Lip. b) Cellular uptake and distribution of LLT@Lip in 4T1 cells. The adjacent graph
shows the corresponding line-profile plot. Scale bar: 20 μm. c–e) The cell viabilities of the 4T1 cells co-incubated with different concentrations of LnNP-
Ir@Lip, Lum-Ir@Lip, and LLT@Lip. f–h) ROS detection in 4T1 cells under normoxia and hypoxia conditions using the DCFH-DA, SOSG, and DHR123
as probes. Scale bar: 50 μm. i) Detection of mitochondrial activity using JC-1 in both normoxic and hypoxic environments. j) Flow cytometry results
demonstrating cell apoptosis under different incubation conditions. Error bars represent SD of duplicate experiments for each sample (n = 3). *P
< 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Small 2025, 21, 2500553 2500553 (7 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

substrate, 100 μM H2 O2 had no significant effect on cell survival chemotherapy was 90.3%, consistent with the MTT results.
rate (Figure S18b, Supporting Information). Thus, this H2 O2 These results demonstrated that both CPDT and EPDT operated
concentration was used in this study as a pre-requisite for eval- through Type I/II photochemical reaction processes mediated
uating the CPDT efficacy. At a Lum-Ir@Lip concentration of by the Ir(III) complex PS, effectively addressing the limitation
200 μg mL−1 , the survival rate of the cells was 47% in normoxic of low ROS efficiency of traditional Type II PDT due to tumor
O2 concentration, whereas it increased to 55% in 2% O2 . This hypoxia. Meanwhile, when combined with hypoxia-activated
observation was attributed to the weakening of the Type-II PDT prodrugs, they achieved improved tumor inhibition efficiency.
effect under hypoxia while the Type-I PDT effect was sustained. The advantages of NIR-IIc imaging on PDT guidance were
Additionally, under hypoxic conditions, LLT@Lip at the same further demonstrated by employing LiErF4 @LiYF4 -based probes
concentration induced a higher tumor cell inhibition rate (33%) (Er-probe) that emitted 1530 nm NIR-IIb luminescence, with sur-
(Figure 3e). This could be attributed to the accelerated activation face modification, same as that of LiTmF4 @LiYF4 -based (Tm-
of TPZ, which compensates for the decreased efficacy of Type-II probe). As shown in Figure S20 (Supporting Information), the
PDT caused by O2 consumption. two probes were mixed to achieve the same integral lumi-
The ROS-generating capability of the LLT@Lip was also evalu- nescence intensity in NIR-IIb and NIR-IIc subwindows upon
ated in tumor cells via Type-I/II photodynamic performance test- 808 nm excitation. The mixed probes were then sealed in cap-
ing. First, O2 •− and 1 O2 were detected in 4T1 cells using 2,7- illary tubes, and their NIR-II imaging properties were observed
dichlorodihydrofluorescein diacetate (DCFH-DA). As shown in by immersing in 1% intralipid solutions at different depths (1
Figure 3f, the fluorescence imaging results and the quantita- to 3 mm) (Figure 4a). NIR-IIb and NIR-IIc regions were distin-
tive statistics (Figure S20a, Supporting Information) revealed that guished by using different filters – 1500 nm LP + 1700 nm SP
compared to the PBS control group, the green fluorescence sig- for NIR-IIb and 1700 nm LP for NIR-IIc. At the same depth, the
nals of DCFH were significantly enhanced in the Lum-Ir+ H2 O2 , DSL intensity of NIR-IIc was higher than NIR-IIb (Figure 4c),
LnNP-Ir + 980 nm laser, and LLT@Lip+ H2 O2 + 980 nm laser demonstrating that the NIR-IIc probe offered sharper imaging
groups. The LLT@Lip + H2 O2 + 980 nm laser group induced and deeper penetration depth compared to its NIR-IIb counter-
the strongest fluorescence signals, indicating that the combined part. Further, the mixed probes formed a cross-shaped pattern
EPDT and CPDT effectively elevated the ROS levels. Additionally, at different depths in 1% lipid solution (Figure 4b). The trans-
the fluorescence signal from normoxic conditions was stronger verse capillary depth was 1 mm, while the longitudinal depth was
than in hypoxic conditions, consistent with the results of the solu- 3 mm. Figure 4d shows that the SBR (signal-to-background ratio)
tion experiments. Furthermore, when SOSG was used as the spe- for the 1–3 mm depth in NIR-IIc imaging was 2.87, which was
cific probe for 1 O2 (Figure 3g), the green fluorescence appeared lower than the SBR of 3.34 observed in NIR-IIb imaging. This in-
brighter in the normoxic environment (Figure S19b, Support- dicated that the NIR-IIc photons had lower attenuation, thereby
ing Information). In contrast, almost no fluorescence signal was indicating a deeper penetration depth for the NIR-IIc lumines-
observed in the hypoxia (2% O2 ) environment, suggesting that cence. The in vivo imaging performances of NIR-IIb and NIR-
while Type-II photodynamic activity was suppressed in a hypoxic IIb c were compared by intravenously injecting tumor-bearing
environment, it exhibited the expected activity level in a normoxic mice with a mixed aqueous solution of CSS3-based NIR-IIc and
environment. Simultaneously, when DHR123 was used to detect Er-based NIR-IIb probes, both adjusted to the same luminous
O2 •− , a stable green fluorescence signal was observed under all intensity upon 808 nm excitation. Figure 4e shows fast Fourier
cultivation conditions (Figure 3h; Figure S19c, Supporting In- transform (FFT) images that illustrate the spatial frequency dis-
formation), indicating that the Type-I photodynamic process re- tribution across tumor regions, thereby highlighting the spec-
mained stable. These findings are crucial for understanding the tral characteristics of the images. NIR-IIc exhibited higher lev-
mechanisms of the LLT@Lip nanoplatform in complex physio- els of both intermediate-frequency and high-frequency signals
logical microenvironments and for optimizing future therapeutic than NIR-IIb, suggesting that NIR-IIc imaging offered improved
strategies. spatial resolution over the NIR-IIb subwindow. This enhance-
The PDT effect on mitochondria was also examined by using ment could be attributed to reduced tissue scattering and more
the 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-imidacarbocyanine balanced photon absorption at longer wavelengths in the NIR-
iodide (JC-1) probe to measure mitochondrial membrane poten- II region. As shown in Figure 4g, the SBR of NIR-IIc and NIR-
tial (Figure 3i). The green fluorescence of JC-1 was more intense IIb images of the blood vessels inside the tumor was 1.55 and
under normoxic conditions due to the effective progression of 1.04, respectively. This data further indicated that SBR correlated
Type-II PDT, indicating severe mitochondrial damage (Figure with wavelength, further confirming the advantages of NIR-IIc
S19d, Supporting Information). Meanwhile, the mitochondria imaging.
incubated with LLT@Lip, H2 O2, and 980 nm laser under hy- Furthermore, LLT@Lip also functioned as an excellent ratio-
poxic conditions were also seriously damaged, likely due to metric luminescence probe capable of measuring the imaging
the reduction of TPZ to cytotoxic chemotherapy drugs. The depth by leveraging its dual-wavelength excitation and NIR-IIc
antitumor mechanism of LLT@Lip was further elucidated by emission properties. Theoretically, the luminescence intensity ra-
using Annexin V-FITC/PI flow cytometry to assess its ability tio (LIR) of LLT@Lip is only related to the power density of the
to induce cell apoptosis (Figure 3j). The flow cytometry results excitation light and tissue penetration depth due to the differen-
indicated a substantial increase in cells undergoing apoptosis tial absorption and scattering of 980 and 808 nm photons in tis-
following various treatments. The results revealed that the sues. Furthermore, in vitro, mimic experiments were conducted
apoptosis rates of CPDT and EPDT were 58.97% and 71.4%, to gather in-depth information. As shown in Figure 4h, a capil-
respectively, while the apoptosis rate of PDT combined with lary filled with LLT@Lip was immersed in 1% intralipid solutions

Small 2025, 21, 2500553 2500553 (8 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Figure 4. a) NIR-II luminescence imaging of a capillary tube filled with PEGlation of Tm-based and Er-based probe solutions, immersed at various depths
in 1% intralipid solution. NIR-IIb and NIR-IIc signals were recorded through coupling with the filters upon 808 nm excitation (50 mW cm−2 ). b) Er-based
and Tm-based probes were mixed at the same luminescence intensity upon 808 nm excitation and placed in cross-shape capillary tubes. The NIR-IIb
and the NIR-IIc subwindows were collected by a 1500 nm LP + 1700 nm SP filter and a 1700 nm LP filter, respectively. c) The luminescence intensity of
Tm and Er probes at different imaging depths of (a). d) SBR statistics of Er-based and Tm-based nanoparticles at 1 and 2 mm depths. e) Luminescence
imaging and the FFT images of the tumor by intravenous injection of a mixed aqueous solution of CSS3 and Er-based nanoparticles. f,g) The spatial
frequency distribution and the SBR statistics of blood vessels of (e). h) LIR images from luminescence imaging of LLT@Lip in 1% intralipid solution at
different depths captured under 808 and 980 nm irradiation. i) The depth-dependent LIR decay curves.

Small 2025, 21, 2500553 2500553 (9 of 13) © 2025 Wiley-VCH GmbH

 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

at different depths and irradiated with fixed power densities of 2 min at an interval of 2 min) at 4 h postinjection. A steady in-
980 and 808 nm light. Figure 4i demonstrates that the LIR under crease was observed in the body weight of mice during the treat-
different excitation wavelengths exhibited an exponential decay ment course, which indicated the preferable biocompatibility of
trend with increasing depth below 2.5 mm, indicating that the LLT@Lip (Figure 5f). After 15 days of monitoring, the tumors
LIR method can be effectively used for in vivo tissue depth detec- were isolated, and images were captured. The results showed ex-
tion. LIR probes have advantages of sensitivity, environmental ro- cellent antitumor therapeutic effects exhibited by the combina-
bustness, deep-tissue imaging, real-time monitoring, and quan- tion of EPDT and CPDT with chemotherapy (Figure 5g,h). How-
titative analysis, making them a superior tool for tumor depth ever, the LnNP-Ir@Lip + 808 nm group had a relatively higher tu-
measurement.c mor volume, confirming that 808 nm excitation could not sensi-
NIR-IIc imaging-guided PDT was done to evaluate the supe- tize Ir(III) complexes to perform EPDT. Also, the LLT@Lip group
rior luminescence imaging performance in the NIR-IIc subwin- had a reduced level of tumor volume compared with the Lum-
dow. To evaluate the biological safety of LLT@Lip, we assessed Ir@Lip group, demonstrating the advantages of combination
key biochemical markers (ALT, AST, ALP) and blood routine pa- therapy in deep hypoxic tumors. This property was attributed to
rameters in both control and LLT@Lip-treated mice. As shown in the fact that both EPDT and CPDT could consume O2 , aggravat-
Figure S21 (Supporting Information), there were no significant ing the hypoxic state of TME, activating TPZ for chemotherapy,
differences in these indices between the control group and the and synergizing PDT to inhibit tumor growth. Furthermore, his-
LLT@Lip group. This indicates that LLT@Lip exhibits good bio- tological analysis after H&E staining showed that both the PBS
compatibility and does not cause adverse effects on liver function control group and the 980 nm group exhibited densely packed
or blood profiles. LLT@Lip (200 μL of 10 mg mL−1 ) was intra- tumor tissues. On the contrary, tumor tissues in the LLT@Lip
venously injected, followed by the monitoring of the circulation + 980 nm group extensively damaged the tumor cells, as char-
and enrichment in vivo under 808 nm laser irradiation (0.5 mW acterized by their condensed nuclei, indicating noticeable apop-
cm−2 ) (Figure 5a). The whole blood vascular system was easily totic activity (Figure 5i). The TUNEL immunofluorescence re-
visualized using a 1700 nm LP filter; tiny blood vessels (diame- sults confirmed the combination therapy of Type I/II PDT and
ter: 458 μm) fitted by Gaussian-fitting of the cross-sectional inten- chemotherapy-induced tumor cell apoptosis (Figure 5j). Also, the
sity were clearly distinguishable at 1 h post-injection (Figure 5b). vital organs (heart, liver, spleen, lung, kidney) in LLT@Lip +
An optimum imaging outcome was obtained at 40 min postin- 980 nm group did not exhibit any significant damage (Figure S22,
jection based on the full width at half maximum (FWHM) and Supporting Information), supporting the excellent biocompati-
SBR statistics for the cross-section of the blood vessels (yellow bility of LLT@Lip.
straight line) on the back of mice (Figure 5b). Figure 5a,d shows
the angiography at different time points within 8 h. 1 h after 3. Conclusion
intravenous injection, LLT@Lip accumulates in the liver, which
becomes clearly visible. By 8 h post-injection, the liver is sig- This study describes the development of an 1800 nm NIR-
nificantly illuminated. This was attributed to the metabolism of IIc-emitting phototheranostic nanoplatform with an organic-
LLT@Lip in the microvessels. The half-time of LLT@Lip was cal- inorganic hybrid nanoPS of LnNP-Ir and Lum-Ir conjugates,
culated as 168 min based on the luminescence signal intensity and TPZ, a hypoxic-activated chemotherapeutic prodrug, within
of the same site. It was found to be significantly greater than the liposomal delivery system, which was specifically designed
that of the ICG used in clinical settings. It benefitted the ves- for deep hypoxic tumor therapy. The incorporation of Ir(III)
sel imaging and enriched the tumor site (Figure 5c). The results complex-based Type I and II PSs of LnNP-Ir and Lum-Ir was
also showed that LLT@Lip reached its maximum accumulation shown to facilitate EPDT and CPDT via the FRET mechanism,
at the tumor site at 4 h post-administration (Figure 5d) when a which was activated by an external 980 nm laser and endoge-
photoswitch to 980 nm irradiation could initiate EPDT to obtain nous H2 O2 , respectively. The dual-modality PDT strategy en-
the maximum therapeutic benefits. Additionally, LLT@Lip, as an sured highly oxidative apoptotic ability, even in deep hypoxic tu-
LIR probe, could provide in-depth details on blood vessels and mors. The synergistic effects of EPDT and CPDT were found
their surrounding tumor state. The tumor tissue was found to to have accelerated the activation of hypoxia-activated TPZ pro-
be clearly illuminated at 980 and 808 nm excitations, and the drugs, enhancing the overall therapeutic outcomes. The study
depth information of the tumor was preliminarily analyzed via results showed that the combination of high spatiotemporal reso-
LIR maps. lution real-time NIR-IIc luminescence imaging-guided PDT with
Next, the in vivo antitumor performance of LLT@Lip was eval- a hypoxia-activated chemotherapy prodrug TPZ could overcome
uated in mice. 4T1 tumor-bearing Balb/c mice were randomly the dual challenges of limited efficacy in deep hypoxic tumors.
categorized into seven groups (n = 5 each), namely Control, This innovative cancer therapy presents a combination of inter-
980 nm, LnNP-Ir@Lip + 808 nm, Lum-Ir@Lip, LnNP-Ir@Lip nal and external excitation modalities for PDT. It offers a safe,
+ 980 nm, LLT@Lip, LLT@Lip + 980 nm. Figure 5e presents the efficient, and accurate antitumor strategy for clinical deep hy-
treatment plan. Each group used different treatment techniques poxic tumor therapy, signifying a significant step forward in can-
to administer the intravenous injections. The antitumor therapy cer treatment.
with NIR-IIc guidance was conducted due to its excellent imag-
ing performance. First, a nontherapeutic 808 nm laser was used
for real-time NIR-IIc imaging until the tumor site was identi- 4. Experimental Section
fied. The tumor site was exposed to 980 nm light under imaging Animal Experiments: All animal experiments were handled following
guidance to trigger EPDT (0.5 W cm−2 for 10 min, irradiation for the care and use of research animals by the animal experiment ethics

Small 2025, 21, 2500553 2500553 (10 of 13) © 2025 Wiley-VCH GmbH
 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Figure 5. a) NIR-IIc imaging of blood vessels and tumor in mice post-intravenous injection of LLT@Lip (10 mg mL−1 ) under 808 nm excitation (50 mW
cm−2 ) at different time points. b) Measurement of FWHM of blood vessels in (a). c) Intravascular circulation time statistics of LLT@Lip. d) LIR imaging
of tumor and ratio maps were obtained under 980 and 808 nm excitation 4 h post-injection of LLT@Lip. e) NIR-IIc imaging-guided PDT therapeutic
schedule. f) Body weights of the mice after different therapies. g) Average tumor volume from each group after the whole therapy. h) Tumor images for
each treatment group. i,j) H&E and TUNEL staining of tumor sections from different treatment groups after 15 days of treatment. Error bars represent
SD of duplicate experiments for each sample. Error bars represent SD of duplicate experiments for each sample (n = 5). *P < 0.05, **P < 0.01, ***P
< 0.001, ****P < 0.0001.

Small 2025, 21, 2500553 2500553 (11 of 13) © 2025 Wiley-VCH GmbH
 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

committee. Female Balb/c mice (6–8 weeks, weight: ≈20 g) were pur- [5] a) R. A., H. Wang, C. Nie, Z. Han, M. Zhou, O. O. Atinuke, K. Wang,
chased from Beijing Vital River Company. The ethical approval statement X. Wang, S. Liu, J. Zhao, W. Qiao, X. Sun, L. Wu, X. Sun, Nat. Com-
was approved by the Animal Protection and Utilization Committee of Jilin mun. 2023, 14, 6644; b) H. Guo, Z. Cao, J. Li, Z. Fu, S. Lin, L. Wang,
University (No. 20240314–15). J. Liu, ACS Nano 2023, 17, 5059; c) D. Kang, H. S. Kim, S. Han, Y.
Statistical Analysis: All data were shown as mean ± standard devia- Lee, Y. P. Kim, D. Y. Lee, J. Lee, Nat. Commun. 2023, 14, 2755; d) S.
tion (SD), and the significance of differences among groups was evaluated Zeng, J. Chen, R. Gao, R. Chen, Q. Xue, Y. Ren, L. Liu, C. Tang, H.
with Student’s t-test (*p < 0.05, **p < 0.01, ***p < 0.001). Data signifi- Hu, N. Zeng, S. Wen, H. Zhang, C. Liu, C. Fang, Adv. Mater. 2023, 36,
cances were as follows: *p < 0.05, **p < 0.01, and ***p< 0.001. The error 2308780.
bars in Figures 2b and 3c,e represent the standard deviation (n = 3). The [6] a) Y. Yu, L. Xiang, X. Zhang, L. Zhang, Z. Ni, Z. H. Zhu, Y. Liu, J. Lan,
error bars in Figure 5f,g represent the standard deviation (n = 5). XRD
W. Liu, G. Xie, G. Feng, B. Z. Tang, Adv. Sci. 2023, 10, 2302395; b) Y.
patterns in Figure 1c were normalized. All data analysis and plotting were
Xu, R. Liu, H. Yang, S. Qu, L. Qian, Z. Dai, ACS Appl. Mater. Interfaces
performed with OriginPro 9 software (OriginLab Co., Northampton, MA)
2022, 14, 25197; c) R. Zhao, X. Liang, B. Zhao, M. Chen, R. Liu, S.
and GraphPad Prism (version 6.0, CA, USA).
Sun, X. Yue, S. Wang, Biomaterials 2018, 173, 58; d) Y. Q. Zhang, W.
L. Liu, X. J. Luo, J. P. Shi, Y. Z. Zeng, W. L. Chen, W. H. Huang, Y. Y.
Zhu, W. L. Gao, R. H. Li, Z. H. Ming, L. X. Zhang, R. Q. Yang, J. Z.
Supporting Information Wang, G. J. Zhang, Adv. Sci. 2023, 10, 2205294; e) X. Zhang, C. Li, Y.
Zhang, X. Guan, L. Mei, H. Feng, J. Li, L. Tu, G. Feng, G. Deng, Y. Sun,
Supporting Information is available from the Wiley Online Library or from Adv. Funct. Mater. 2022, 32, 2207259.
the author. [7] a) N. Zhang, K. P. Francis, A. Prakash, D. Ansaldi, Nat. Med. 2013,
19, 500; b) D. Mao, W. Wu, S. Ji, C. Chen, F. Hu, D. Kong, D. Ding, B.
Liu, Chem 2017, 3, 991; c) C. Wu, M. Cui, L. Cai, C. Chen, X. Zhu, Y.
Acknowledgements Wu, J. Liu, H. Wang, Y. Zhang, ACS Appl. Mater. Interfaces 2022, 14,
13094; d) D. Wang, B. Xue, T. Y. Ohulchanskyy, Y. Liu, A. Yakovliev, R.
The authors gratefully acknowledge the financial support provided by Ziniuk, M. Xu, J. Song, J. Qu, Z. Yuan, Biomaterials 2020, 251, 120088;
the National Key Research and Development Program of China (D.L., e) Q. Wang, Y. Dai, J. Xu, J. Cai, X. Niu, L. Zhang, R. Chen, Q. Shen, W.
2021YFA0715603) and the National Natural Science Foundation of China Huang, Q. Fan, Adv. Funct. Mater. 2019, 29, 1901480; f) H. Chen, F.
(Y.C., 62075217, 62475262, H.C., 62305329, B.L., 22172154, and Q.L., Wu, X. Xie, W. Wang, Q. Li, L. Tu, B. Li, X. Kong, Y. Chang, ACS Nano
12404496), the Project of Science and Technology Agency, Jilin Province 2021, 15, 20643.
(Y.C., 20230508104RC and 20230402039GH), the Changchun Science and [8] a) Y. Chang, H. Chen, X. Xie, Y. Wan, Q. Li, F. Wu, R. Yang, W. Wang, X.
Technology Bureau (Y.C., 23GZZ06) and the China Postdoctoral Science
Kong, Nat. Commun. 2023, 14, 1079; b) Z. Feng, Y. Li, S. Chen, J. Li, T.
Foundation (H.C., 2023M733432 and Q.L., 2023M741350) and the au-
Wu, Y. Ying, J. Zheng, Y. Zhang, J. Zhang, X. Fan, X. Yu, D. Zhang, B.
tonomous project of SKLA (Y.C., SKLA-Z-2023-03).
Z. Tang, J. Qian, Nat. Commun. 2023, 14, 5017; c) T. Wu, Y. Wang, H.
Lin, J. Qian, Laser & Photonics Rev. 2024, 19, 202400628; d) Y. Zhong,
Z. Ma, F. Wang, X. Wang, Y. Yang, Y. Liu, X. Zhao, J. Li, H. Du, M.
Conflict of Interest Zhang, Q. Cui, S. Zhu, Q. Sun, H. Wan, Y. Tian, Q. Liu, W. Wang, K.
C. Garcia, H. Dai, Nat. Biotechnol. 2019, 37, 1322; e) Z. Feng, T. Tang,
The authors declare no conflict of interest. T. Wu, X. Yu, Y. Zhang, M. Wang, J. Zheng, Y. Ying, S. Chen, J. Zhou,
X. Fan, D. Zhang, S. Li, M. Zhang, J. Qian, Light Sci. Appl. 2021, 10,
197.
Data Availability Statement [9] a) Y. Chen, S. Wang, F. Zhang, Nat. Biomed. Eng. 2023, 1, 60; b) Z.
Chen, B. Yun, Y. Hou, X. Wang, X. Wang, J. Xu, L. Jiang, T. Han, H.
The data that support the findings of this study are available from the cor- Zhang, F. Zhang, Angew. Chem., Int. Ed. 2024, 137, 202416893; c) M.
responding author upon reasonable request. B. Liisberg, S. Lahtinen, A. B. Sloth, T. Soukka, T. Vosch, J. Am. Chem.
Soc. 2021, 143, 19399.
[10] Z. H. Chen, X. Wang, M. Yang, J. Ming, B. Yun, L. Zhang, X. Wang,
P. Yu, J. Xu, H. Zhang, F. Zhang, Angew. Chem., Int. Ed. 2023, 62,
Keywords 202311883.
deep tumors, hypoxia-activated chemotherapy, lanthanide-doped [11] K. Lv, H. Wang, X. Fu, S. Chen, R. Zhang, Y. Zhou, J. Feng, H. Zhang,
nanoparticles, near-infrared-II imaging, photodynamic therapy Small 2024, 20, 202404007.
[12] a) Y. C. Chen, Y. J. Liu, C. L. Lee, K. Y. Pham, D. Manoharan, S.
Received: February 14, 2025 Thangudu, C. H. Su, C. S. Yeh, Adv. Healthcare Mater. 2022, 11,
Revised: March 24, 2025 2201613; b) M. M. Chen, C. H. Xu, W. Zhao, H. Y. Chen, J. J. Xu,
Published online: April 8, 2025 Angew. Chem., Int. Ed. 2022, 61, 202117401; c) H. An, C. Guo, D.
Li, R. Liu, X. Xu, J. Guo, J. Ding, J. Li, W. Chen, J. Zhang, ACS
Appl. Mater. Interfaces 2020, 12, 17230; d) Y. Yang, S. Wang, L. Lu,
Q. Zhang, P. Yu, Y. Fan, F. Zhang, Angew. Chem., Int. Ed. 2020, 59,
18380.
[1] a) S. Chen, B. Li, Y. Yue, Z. Li, L. Qiao, G. Qi, Y. Ping, B. Liu, Adv.
[13] a) J. Ding, G. Lu, W. Nie, L. L. Huang, Y. Zhang, W. Fan, G. Wu, H. Liu,
Mater. 2024, 36, 2404296; b) X. Kang, Y. Zhang, J. Song, L. Wang, W.
H. Y. Xie, Adv. Mater. 2021, 33, 2005562; b) C. Liu, X. Wang, J. Liu, Q.
Li, J. Qi, B. Z. Tang, Nat. Commun. 2023, 14, 5216.
Yue, S. Chen, J. W. Y. Lam, L. Luo, B. Z. Tang, Adv. Mater. 2020, 32,
[2] C. Chen, C. Wu, J. Yu, X. Zhu, Y. Wu, J. Liu, Y. Zhang, Coordin. Chem.
2004685.
Rev. 2022, 461, 214495.
[14] a) R. Ding, X. Liu, W. Zhang, X. Chen, S. Chen, X. Yu, Z. Zhao, K.
[3] H. Abrahamse, M. R. Hamblin, Biochem. J. 2016, 473, 347.
Li, Adv. Funct. Mater. 2024, 34, 405499; b) J. Bonelli, E. Ortega-Forte,
[4] D. Chen, Q. Xu, W. Wang, J. Shao, W. Huang, X. Dong, Small 2021,
G. Vigueras, J. Follana-Berná, P. Ashoo, D. Abad-Montero, N. Isidro,
17, 2006742.
M. López-Corrales, A. Hernández, J. Ortiz, E. Izquierdo-García, M.

Small 2025, 21, 2500553 2500553 (12 of 13) © 2025 Wiley-VCH GmbH
 16136829, 2025, 21, Downloaded from [Link] by National Tsing Hua University, Wiley Online Library on [12/08/2025]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] [Link]

Bosch, J. Rocas, Á. Sastre-Santos, J. Ruiz, V. Marchán, ACS Appl. [18] a) J. Wu, W. Liu, S. Tang, S. Wei, H. He, M. Ma, Y. Shi, Y. Zhu, S. Chen,
Mater. Interfaces 2024, 16, 38916. X. Wang, ACS Appl. Mater. Interfaces 2024, 16, 56850; b) E. Lavagna,
[15] S. Liu, H. Chen, Q. Wu, Y. Sun, Y. Pei, Z. Wang, D. Z. P. Güven, D. Bochicchio, F. Olgiati, F. Stellacci, G. Rossi, Nanoscale
Zhu, G. Li, M. R. Bryce, Y. Chang, Inorg. Chem. 2024, 63, 2021, 13, 16879; c) Y. Huang, H. Liu, Y. Zhao, H. Chen, Q. Li, X. Li, S.
16404. Hua, D. Cao, Y. Chang, RSC Adv. 2024, 14, 20152.
[16] L. Xia, X. Kong, X. Liu, L. Tu, Y. Zhang, Y. Chang, K. [19] a) Z. Liu, R. J. Deeth, J. S. Butler, A. Habtemariam, M. E. Newton, P.
Liu, D. Shen, H. Zhao, H. Zhang, Biomaterials 2014, 35, J. Sadler, Angew. Chem., Int. Ed. 2013, 52, 4194; b) Z. Liu, I. Romero-
4146. Canelón, B. Qamar, J. M. Hearn, A. Habtemariam, N. P. E. Barry, A.
[17] M.-K. So, A. M. Loening, S. S. Gambhir, J. Rao, Nat. Protoc. 2006, 1, M. Pizarro, G. J. Clarkson, P. J. Sadler, Angew. Chem., Int. Ed. 2014,
1160. 53, 3941.

Small 2025, 21, 2500553 2500553 (13 of 13) © 2025 Wiley-VCH GmbH