Fig.1 Photophysical properties of the designed PSAs and their underlying TICT mechanism. UV–vis absorption and PL spectra of (A) 5TZP and (B) 2TZP in DMSO solution ([c] = 1 × 10–5 M); (C) normalized PL spectra of 5TZP and 2TZP in solids; (D) absolute fluorescence quantum yields (PLQYs) of 5TZP and 2TZP in DMSO solution ([c] = 1 × 10–5 M) and in solids. (E) Plots of I/I0 for 5TZP and 2TZP in DMSO/water mixtures with different water fractions ([c] = 1 × 10–5 M). (F) Calculated molecular planarity of 5TZP and 2TZP; (G) RMSD of 5TZP (top) and 2TZP (down) (ground state: blue, excited state: red); plots of reorganization energy versus normal-mode wavenumber of (H) 5TZP and (I) 2TZP (inset: proportions of bond length, bond angle, and dihedral angle contributing to reorganization energy); PL spectra of (J) 5TZP and (K) 2TZP in different solvents; (L) plot of the emission wavelength of 5TZP and 2TZP versus the water fraction from 10% to 90%; schematic illustration of S1 energy-level changes of (M) 5TZP and (N) 2TZP in THF and CHCl3; (O) the frontier molecular orbitals of HOMO and LUMO, and electron–hole analysis at S0 and S1 based on optimized geometries of 5TZP and 2TZP; calculated potential energy surfaces of (P) 5TZP and (Q) 2TZP of φ1 and φ3 in S1 states over different dihedral angles.#

Scheme.1 Schematic Illustration of the “Nanocrystal-to-Nanocrystal (NCNC) Transformation” in TME for On-Demand Photo-STING Therapy#

本研究提出了“分子构象扭曲协同质子捕获”的分子工程路线，首次设计合成了兼具酸敏二甲氨基（DMA）给体、三苯胺（TPA）给体与带正电荷噻唑受体的异构光敏剂2TZP（以5TZP为对比异构体）。理论计算与晶体学分析表明，2TZP中正电荷噻唑与邻近苯环发生光诱导结构平面化，增强了扭曲分子内电荷转移（TICT）与聚集诱导发光（AIE）效应，显著减小单线态-三线态能隙（$\Delta E_{\text{ST}} = \mathbf{0.129\text{ eV}}$）。在弱酸刺激下，质子化驱动二聚体沿b轴发生约90°旋转与晶格滑移，利用晶体内部高达45.76%的自由体积与高晶格相似性，在保持单晶结构的同时实现从六棱柱到长宽比为4的长方体棱柱的完全晶型转变，并暴露正电荷位点以特异性结合富电子的RNA，形成“电子库-泵”效应加速电子转移，使系间窜越（ISC）和三线态产率提高至10.98%。

实验全面评估了2TZP纳米晶体的理化性质、细胞行为及双侧小鼠肿瘤模型的免疫激活效果。2TZP纳米晶体表现出6.70的酸解离常数（$\text{p}K_{\text{a}}$）与仅0.70个pH单位的超窄响应区间；在pH 6.5条件下与RNA结合后，荧光强度恢复21.09倍，总ROS生成增强25.80倍（显著优于无定形纳米颗粒的4.48倍）。细胞实验显示，弱酸环境促使转变后的纳米晶体靶向核仁（与RNA共定位率达80.03%），在光照下诱导核DNA大量泄漏并激活cGAS-STING通路，使4T1细胞中磷酸化STING（p-STING）表达上调8.87倍，光细胞毒性指数达24.83。在双侧4T1乳腺癌小鼠模型中，光照仅作用于原发肿瘤，不仅实现2/5小鼠原发肿瘤完全消退，更通过诱导树突状细胞成熟（28.90%）、浸润性CD8⁺ T细胞增加（6.01倍）及效应记忆T细胞（$\text{T}_{\text{em}}$）增加（3.52倍），实现了对未照射远端肿瘤83%的生长抑制率，40天生存率达到100%，并显著抑制了肺血行转移。

全流程贯穿了“分子合成与量化计算 $\rightarrow$ 单晶结构解析与纳米晶相变表征 $\rightarrow$ 飞秒/纳秒瞬态吸收动力学解析 $\rightarrow$ 活细胞核仁靶向与免疫通路验证 $\rightarrow$ 活体近红外荧光成像与双侧肿瘤免疫治疗”。实验同时证实，2TZP纳米晶体在水、PBS与细胞培养基中存放7天仍能保持晶体形貌与粒径稳定，且酸碱诱导的纳米晶相变具备高度的可逆循环性；荷瘤小鼠在治疗期间无显著体重下降或主要器官损伤，表现出优异的生物相容性与靶向安全性。

Fig.2 Acid-triggered single-crystal transformation by protonation and structural relaxation. (A) 1H NMR spectra of 2TZP in CDCl3 before and after addition of CF3COOD; (B) unit-cell diagrams, (C) molecular structures, (D) dimer arrangements, and (E) molecular packing viewed along the b-axis of 2TZP under different conditions (pH= 7.4 and 6.5); (F) illustration of conformational rotation and structural relaxation in 2TZP dimers during the acid stimulation process; (G) calculated energy landscapes for the proposed acid-triggered transformation in 2TZP. The visualized isosurfaces of (H) IGM analysis and 2D plots of δgIntra (blue) and δgInter (red) for the dimer structure in (I) 2TZP (pH= 7.4) and (J) 2TZP (pH= 6.5).#

Fig.3 NCNC transformation and RNA-binding study of 2TZP. high resolution transmission electron microscopy (HR-TEM) and SAED images of (A) 2TZP NCs (pH= 7.4) and (B) 2TZP NCs (pH= 6.5); (C) the predicted crystal growth of 2TZP (pH= 7.4) and 2TZP (pH= 6.5) by BFDH prediction; (D) the calculated attachment energies and total face area in different crystal faces for 2TZP (pH= 7.4) and 2TZP (pH= 6.5); (E) FL spectra and (F) corresponding pH-dependent emission intensity of 2TZP NCs in the indicated pH buffers; (G) TEM and SAED images of 2TZP NCs (pH= 6.5) after adding RNA; (H) DLS analysis of the size distribution and (I) zeta potential of the 2TZP NCs under different conditions; (J) linear relationship between the fluorescence intensity of 2TZP NCs and RNA or DNA; (K) fluorescence images of 2TZP NCs with DNA or RNA in different pH buffers; (L) FL responses of 2TZP NCs to RNA and other interferences (50 μM) (1–12: Cu+, Cu2+, Ag+, Zn2+, ClO–, HSO3–, GSH, BSA, PBS, DMEM, DNA, and RNA); molecular docking study showing the interaction position and type of (M) 2TZP (pH= 7.4) and (N) 2TZP (pH= 6.5) binding to RNA.#

Fig.4 Mechanism of RNA-Enhanced Intersystem crossing to promoting ROS generation. (A) total ROS generation of 2TZP NCs under light irradiation; (B) ·OH, (C) O2•–, and (D) 1O2 generation of 2TZP NCs detected by APF, DHR123, and SOSG under light irradiation, respectively; the femtosecond transient absorption (fs-TAS) of (E) 2TZP NCs (pH= 7.4), (I) 2TZP NCs (pH= 6.5), and (M) 2TZP NCs (pH 6.5+RNA); extracted kinetic curves of (F) 2TZP NCs (pH= 7.4), (J) 2TZP NCs (pH= 6.5), and (N) 2TZP NCs (pH= 6.5+RNA); the differences in the species-associated decay (SAD) spectra and population-versus-time curves of (G, H) 2TZP NCs (pH= 7.4), (K, L) 2TZP NCs (pH= 6.5), and (O, P) 2TZP NCs (pH 6.5+RNA), respectively; (Q) time-resolved PL (TRPL) signals at 600 nm of 2TZP NCs (pH= 7.4), 2TZP NCs (pH= 6.5), and 2TZP NCs (pH 6.5+RNA); the kinetics of nanosecond transient absorption (ns-TAS) for (R) 2TZP NCs (pH= 7.4), (S) 2TZP NCs (pH= 6.5), and (T) 2TZP NCs (pH 6.5+RNA).#

Fig.5 Intracellular localization and in vitro evaluation of phototherapeutic effect by 2TZP NCs. (A) colocalization study of 2TZP NCs with other commercial organelle-staining probes in HeLa cells; (B) fluorescence images of Hoechst 33342, SYTO green and 2TZP NCs in fixed HeLa cells treated with RNase or DNase; (C) detection of total ROS and O2–• in 4T1 cells by using DCFH-DA and DHE as intracellular fluorescence indicators; the corresponding fluorescence intensities of (D) DCFH-DA and (E) DHE (n = 3); dose–response cell viability curves and their corresponding IC50 values for 2TZP NCs against (F) HeLa, (G) MCF-7, and (H) 4T1 cells (n = 3); (I) IC50 and the corresponding PI (phototherapeutic index) values of 2TZP NCs against HeLa, 4T1, and MCF-7 cells (pH= 6.5); (J) flow cytometry profiles of apoptosis for 4T1 cells incubated with 2TZP NCs; (K) migration assay and (L) the corresponding semiquantitative analysis for 4T1 cells with various treatments after 24 h; (M) invasion assay for 4T1 cells with various treatments after 48 h. Data are expressed as the mean ± SD; p-values are denoted as follows: ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.#

Fig.6 In vitro cellular immune response. (A) γ-H2AX staining to visualize DNA damage and PicoGreen staining to visualize DNA fragmentation in 4T1 cells after different treatments; (B) the quantitative fluorescence intensity analysis of γ-H2AX and PicoGreen in the cytoplasm (n = 3); (C) expression of cGAS-STING proteins in 4T1 cells after various treatments; GAPDH was used as the internal reference protein. Cytokine detection of IFN-β (D) and IL-18 (E) in culture supernatants of 4T1 cells (n = 3); (F) CLSM images showing the expression of CRT (green channel) and HMGB1 (red channel) in 4T1 tumor cells after different treatments. The quantitative fluorescence intensity analysis of (G) CRT and (H) HMGB1 (n = 3); (I) volcano plot illustrating DEGs between control and 2TZP NCs+L at pH 6.5 group, highlighted in red (upregulated) and blue (downregulated); (J) bubble maps of significant enrichment pathways by KEGG enrichment analysis; (K) GSEA reveals pathway enrichments of various genes that changed in 2TZP NCs+L at pH 6.5 versus the control group ; (L) circular visualization of the results of functional enrichment analysis; (M) PPI network of DEGs involved in IL-17 signaling, apoptosis and DNA damage. Data are expressed as the mean ± SD; p-values are denoted as follows: ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.#

Fig.7 Assessment of the therapeutic efficiency of 2TZP NCs in vivo. (A) In vivo real-time FL images of tumor-bearing BALB/c nude mice after the injection of 2TZP NCs or PBS in tumor and normal tissue; (N: normal tissue; T: tumor tissue); (B) the quantitative fluorescence intensity analysis of tumor and normal tissues (n = 3); (C) FL images of 4T1 tumor-bearing mice after intravenous administration of 2TZP NCs; (D) the quantitative fluorescence intensity analysis of tumor (n = 3); (E) FL intensity of tumor and major organs after injection of 2TZP NCs for 24 h; (F) illustration of the therapeutic scheme for the bilateral tumor-bearing BALB/c mice with 2TZP NCs; plots of (G) primary tumor volume and (H) tumor weight under different treatments (n = 5); plots of (I) distant tumor volume and (J) tumor weight under different treatments (n = 5); (K) individual tumor growth curves for primary tumor and distant tumor; (L) survival curves of mice with different treatments during the therapy period (n = 5); (M) body weight of mice with different treatments (n = 5); Representative (N) H&E and (O) immunofluorescence images of TUNEL and Ki-67 staining of tumor tissues under different treatments (blue: nucleus, green: TUNEL, red: Ki-67). Data are expressed as the mean ± SD; p-values are denoted as follows: ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.#

Fig.8 Assessment of the antitumor immune responses of 2TZP NCs in vivo. (A) immunofluorescence images of CRT and HMGB1, CD4+ and CD8+ staining of tumor tissues under different treatments (top: green, CRT; red, HMGB1; down: green, CD4+; red, CD8+); (B) representative H&E staining of lungs from mice under different treatments; (C) analysis of mature DCs in lymph nodes and (D) the corresponding quantitative analysis under different treatments by flow cytometry (n = 3); (E) quantification of populations of CD8+ and CD3+ T cells in tumor tissues and (F) the corresponding quantitative analysis under different treatments by flow cytometry (n = 3); (G) representative scatter plots, (H) the corresponding quantification and (I) donut charts showing the proportions of CD3+ CD8+CD44+CD62L– Tem cells (n = 3); (J) the expression levels of IL-18, IL-1β, IFN-β, and IL-6 in serum investigated by ELISA (n = 5). Data are expressed as the mean ± SD; p-values are denoted as follows: ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.#