Abstract
The pH in organisms is strictly regulated within an appropriate range for homeostasis, and many pathological processes often result in pH dysregulation to distinct intervals that serves as targets for diagnosis and treatment. Thus, a mechanism of selective and gating response for different proton levels in an integrated nanodevice is needed to report acidic biological environments with specific pH dynamic ranges. Here we report the design of band-pass light-up nanokits (BLINK) with pH heterogating capacity in opposite directions to achieve selective reporting of the acidic intervals (minimum to 0.25 pH unit) of pathophysiological environments. The proton heterogated property of BLINK integrates ultra-fast (<15 ms) and cascade nanophase transition of transistor-like polymeric modules and tandem FRET relay between encoded fluorophores, permitting the selective transformation of acidic extracellular or intracellular signals to exponentially amplified fluorescence readout without introducing the signals outside the targeted pH ranges. This study offers a valuable tool for pH sensing, and paves the way for the development of interval-based optical imaging technology.
Introduction
Acid-base homeostasis and pH regulation are of great importance to the organism physiology1,2. In particular, many biological processes, such as enzymatic reaction3, antigen processing and presentation4,5, lysosome catabolism6, and glycolysis7,8, operate effectively within specific pH ranges. For example, cancer cells mainly rely on aerobic glycolysis (known as Warburg effect), leading to the extracellular mild acidic pH (pHe) in the range of 6.5–6.99. Thus, interval-based pH sensing technology will have significant implications for biochemical processes monitoring, medical imaging, and clinical diagnosis.
Many methods, such as electrochemical configurations10, and optical systems11, have been employed for pH sensing. Electrochemical pH sensors enable real-time monitoring of pH dynamics with high precision and the electrode tip can be miniaturized for intracellular pH sensing and label-free imaging12,13. Nevertheless, the invasiveness of microelectrode insertion may cause physical disruption to cells and thereby limiting their applicability in prolonged live-cell studies. In comparison, alternative imaging techniques including optical methods, magnetic resonance imaging (MRI)14,15,16 and radioactive tracers for positron-emission-tomography (PET)17,18 have been developed for pH sensing, offering high spatial-temporal resolution and compatibility with live-cell imaging. For optical imaging, diverse modalities have been employed ranging from absorptiometry19 and luminescence measurement20,21 to infrared/Raman spectroscopic sensing22,23,24,25,26. Especially, fluorescence-based pH indicators combined with cutting-edge nanotechnology27,28 are wildly applied for non-invasive and real-time bioimaging. In addition to the traditional fluorescent chemical dyes, genetically encoded pH biosensors29,30,31,32,33,34,35,36 have been developed to visualize the pH-related biological processes and events directly in situ. However, selective targeting and sensing of certain pH ranges remains a challenge. Although researches have reported OFF-ON-OFF fluorescent pH sensors37,38,39,40,41,42,43,44,45, these small molecular sensors fail to discriminate the narrow pH range in biological systems due to their broad pH responsiveness (∆pH~4–6 units). Moreover, these probes are also limited by restricted fluorescence emission and fixed pKa. Therefore, probes with sharp pH response as well as tunability for pH-intervals and fluorescence emission are urgently needed for biomedical imaging.
Biological systems have evolved to exploit cooperativity in macromolecular assembly for molecular recognition46,47,48, such as cooperative activation of ion channels49. Moreover, many synthetic self-assembled nanostructures with positive cooperativity are applied for chemical and biological sensing50, which exhibit fast temporal responses with enhanced sensitivity and specificity. Recently, we have demonstrated that hydrophobic micellization-driven cooperative nanophase separation renders transistor-like signal transformation of nanomaterials with distinct and sharp pH thresholds51,52,53,54,55,56. In this work, we developed a library of band-pass light-up nanokit (BLINK) with tunable pH heterogating capacity by integrating sharp and cascade phase transition with tandem fluorescence resonance energy transfer (FERT) mechanism (Fig. 1a). Based on the mechanism of cascade-heterogated nanophase separation, the low-pass (LP) transistor-like module of BLINK forms the high pH threshold (pHt1), while the high-pass (HP) transistor-like module forms the low pH threshold (pHt2), resulting in a pH band-pass (BP) feature of fluorescence emission. Utilizing BLINK with optimized pHt1 and pHt2 as endosome-trackers, we could specifically track and report the endocytic organelles with distinct lumenal pH intervals by excluding the more acidic organelles as endosome maturation (Fig. 1b). Moreover, we exploited BLINK to specifically transform the metabolic acidosis of extracellular tumor microenvironment to exponentially amplified fluorescence output by eliminating the off-target signals from the more acidic endocytic vesicles after nanoparticle cellular internalization (Fig. 1b). Thus, BLINK can serve as a toolkit for pH regulation study in cell biology and a precise targeting platform for the drug/gene delivery.
a scheme of BLINK nanotransistor with binary OFF-ON-OFF response as compared to traditional OFF-ON-OFF small molecular probes. Transition of BLINK from OFF1 to ON is triggered at pHt1, while the transition from ON to OFF2 occurs when pH value drops to pHt2. Thus, BLINK[t2, t1] could selectively target specific pH-interval between pHt1 and pHt2 with a narrow resolution of 0.25 pH unit. However, OFF-ON type pH probe would provide biased signals of pH lower than the targeted pH window. As for dual-pH-responsive molecular probes, they require a very wide range of 4–6 pH units to achieve OFF-ON-OFF transition. b based on the concept of cascade-heterogated nanotransistor, BLINKs with distinct pHt1 and pHt2 exhibit the selective amplification of acidic extracellular tumor microenvironmental signals (pHe ~6.5–6.9) rather than the pH signals from bloodstream and endocytic organelles upon nanoparticle internalization. For the imaging of acidic endocytic organelles, BLINKs with optimized pHt1 and pHt2 keep OFF at pH higher than pHt1 and lower than pHt2, while the nanoparticles are turned ON in the unique pH ranges of early endosome (pHi ~6.0–6.5), late endosome (pHi ~5.5–6.0), and lysosome (pHi ~4.5–5.5) without any crosstalk.
Results
Design and fabrication of BLINK
We use the ultra-pH-sensitive (UPS) poly(ethylene glycol)-b-poly(alkyl amino ethyl methacrylate) (mPEG5k-b-PR) copolymers with tunable and sharp pH transitions (pHt) to construct a library of BLINK with different pH intervals. BLINK has a micellar nanostructure that comprises of two pH-gated modular polymers with distinct pHt: a polymer with high pHt (pHt1) conjugated with a fluorescent quencher (mPEG5k-b-P(R1-r-Q)), and a polymer with low pHt (pHt2) encoded separately with a fluorescence donor (mPEG5k-b-P(R2-r-D)) and acceptor (mPEG5k-b-P(R2-r-A)). Mechanistically, mPEG5k-b-P(R2-r-Q) serves as the low-pass gate that functions as the quencher of the signal of mPEG5k-b-P(R2-r-A), while the mPEG5k-b-P(R2-r-D) and mPEG5k-b-P(R2-r-A) co-work as the high-pass gate that can achieve fluorescence resonance energy transfer (FRET) from donor to acceptor (Fig. 2a, Supplementary Figs. 1, 2). We hypothesize that at the pH higher than pHt1 (pH > pHt1), all the fluorescence signals are quenched (OFF1 state) in the micelle core due to the tandem FRET relay from donor to acceptor, and to quencher molecules upon the excitation of donor dye. Then, as the pH drops below pHt1 (pHt2 <pH <pHt1), the mPEG5k-b-P(R1-r-Q) module is rapidly dissociated from the BLINK due to the protonation of its tertiary amine, and the fluorescence signal of acceptor is exponentially amplified (ON state) from the residual nanoparticle. Finally, as the pH drops below pHt2 (pH <pHt2), the residual nanoparticle is completely disassembled into mPEG5k-b-P(R2-r-D) and mPEG5k-b-P(R2-r-A) unimers, and the fluorescence signal of acceptor turns OFF (OFF2 state) due to the abolishment of the FRET effect between the donor and acceptor. Thus, BLINK exhibits an OFF-ON-OFF pH band-pass (BP) performance in acceptor channel upon the excitation of donor dye, while presents OFF-ON pH low-pass (LP) property in acceptor channel upon the excitation of the dye itself.
a molecular mechanism of BLINK for heterogated pH responsiveness. Intermolecular FRET between the conjugated dyes changed as the proton-triggered sequential disassembly of the copolymers with different pHt, which leads to distinct luminescence at different stages upon the excitation of the donor or acceptor. b normalized absorption and emission spectra of the selected dyes (DEAC as donors, Cy3.5 as acceptors, BBQ650 as quenchers). c representative fluorescence spectra of BLINK at different stages with the excitation of donors (λex = 405 nm, OFF1: pH 7.4, ON: pH 6.3, OFF2: pH 5.4). d normalized fluorescence intensities (F.I.) as a function of pH for BLINK library with tunable pHt1 and pHt2, of which the ΔpH (pHt1‒pHt2) was narrowed to about 0.25. e fluorescent images of BLINK library showed in (d) at different pH values. Samples were placed in the 384-well plates. f normalized fluorescence intensities (F.I.) as a function of pH value for BLINK library with tunable pHt1 and fixed pHt2. g fluorescent images of BLINK library showed in (f) at different pH values. h fluorescent images of exemplary BLINK library with different FRET pairs and corresponding quenchers. Source data are provided as a Source Data file.
To test this hypothesis, we first synthesized a series of amphiphilic mPEG5k-b-PR block copolymers, where mPEG5k is hydrophilic poly(ethylene glycol) and PR is an ionizable segment. Each mPEG5k-b-PR copolymer was conjugated with different fluorescent dyes (Q, D, A dyes) via amide linkage. Then, BLINK was engineered through the co-assembly of mPEG5k-b-P(R1-r-Q), mPEG5k-b-P(R2-r-D), and mPEG5k-b-P(R2-r-A) copolymers at different molar fraction, where PR1 and PR2 copolymers have predesignated pHt values of pHt1 and pHt2 (pHt1 > pHt2). We screened the appropriate donor-acceptor pairs for BLINK. 7-diethylamino-3-carboxy coumarin (DEAC), which exhibited broad emission spectrum, was widely used as FRET donor57,58,59; Cy3.5 was chosen as the acceptor due to the substantial overlap between its excitation spectrum and the emission spectrum of DEAC (Fig. 2b). These dyes were efficiently conjugated to mPEG5k-b-PR copolymers. The conjugation number was measured by UV–Vis spectroscopy (Supplementary Table 1). To confirm the FRET effect between the chosen dyes, hybrid micelles of the DEAC-conjugated and Cy3.5-conjugated poly(ethylene glycol)-b-poly(dipropylaminoethyl methacrylate) copolymers (mPEG5k-b-nPDPA, pKa = 6.08) were fabricated at pH 7.4. When DEAC was excited at 405 nm, a strong emission of Cy3.5 centered at 610 nm was captured (Supplementary Fig. 3). When the hybrid micelles dissociated at pH 5.4, only the emission of DEAC at 470 nm was shown under the excitation of DEAC. Similar results were reproduced by adjusting the ratio (wt./wt.) of two components. These results demonstrated that the pH-gated module with high-pass feature was successfully fabricated by engineering the hybrid micelles of the DEAC-conjugated and Cy3.5-conjugated copolymers.
We next combined the high-pass (pHt2) and low-pass (pHt1) pH-gated modules to fabricate BLINK[t2, t1]. BBQ650, as a suitable FRET quencher of Cy3.5, was chosen and conjugated to poly(ethylene glycol)-b-poly(ethylpropylaminoethyl methacrylate) copolymer (mPEG5k-b-PEPA). BLINK was prepared via the co-assembly of the BBQ650-conjugated mPEG5k-b-PEPA, DEAC- and Cy3.5-conjugated mPEG5k-b-nPDPA copolymers (Fig. 2c). At neutral pH (OFF1), the emission of Cy3.5 upon the excitation of the DEAC was effectively quenched by BBQ650. When the BBQ650-conjugated mPEG5k-b-PEPA dissociated at pH 6.3, the emission of Cy3.5 was dramatically amplified (ON). Furthermore, at pH 5.4, the emission of Cy3.5 subsequently diminished due to the dissociation of the micelles (OFF2). The fluorescence ON/OFF1 and ON/OFF2 ratios reached over 10-fold. These results demonstrated that BLINK achieved the OFF-ON-OFF properties of the acceptor emission upon the excitation of the donor at different pH values.
BLINK is tunable for the pH-interval reporting and multi-color encoding
We next investigated the tunability of BLINK in different pH intervals. By using mPEG5k-b-PR copolymers with different pHt, we successfully engineered a small library of BLINKs with tunable pHt1 and pHt2 values in a broad pH range from 5.2 to 6.6 (Fig. 2d, Supplementary Fig. 4). All the BLINKs can achieve very narrow pH range resolution (ΔpH = 0.2–0.3) due to the sharp pH transition (ΔpH10–90% ~ 0.25) of LP and HP gates for each BLINK. As shown in Fig. 2e, the FRET images of BLINKs as a function of pH values presented a signal light-up in a pH band-pass manner, whereas the emission for acceptor itself appeared a pH low-pass characteristic with persistent fluorescent signal under the pH below pHt1 (Supplementary Fig. 5). To further verify the tunability of pH-gating capacity, we fabricated a series of BLINKs with fixed pHt2 and variable pHt1, with a pH span from 0.3 to 1.1 (Fig. 2f, g, Supplementary Figs. 6, 7). Collectively, we established a library of BLINK nanoparticles with tunable low-pass and high-pass gating capacity.
We also verified the universality of BLINK strategy in other D-A-Q FRET pairs. In the DEAC-Cy3.5-BBQ650 pairs, the acceptor dye Cy3.5 was successfully replaced by Cy3 and DQFRB dyes60 with maximum emission of 570 nm and 660 nm, respectively (Supplementary Figs. 8, 9). By altering the acceptors, BLINK showed a similar pH band-pass imaging readout, except for the corresponding changes in fluorescence emission (Fig. 2h, Supplementary Figs. 10, 11). We also employed TVP as a donor (λex/λem = 470/590 nm)61 (Supplementary Fig. 12a). Accordingly, Cy5.5 and Cy7 were selected as the acceptor and quencher, respectively (Supplementary Figs. 12, 13). In this case, BLINK composed of TVP-Cy5.5-Cy7 pairs also exhibited the band-pass fluorescence illumination with an excitation of 465 nm and emission of 720 nm (Fig. 2h, Supplementary Fig. 14). Altogether, we successfully engineered a library of multicolored BLINKs using different UPS copolymers encoding different FRET pairs and quenchers.
Proton-driven cascade nanophase transition and FRET relay enable band-pass performance of BLINK
To further understand the molecular mechanism of pH-gating capacity of BLINK, we performed several physicochemical experiments based on BLINK[6.1,6.5] that composed of mPEG5k-b-(nPDPA-r-DEAC), mPEG5k-b-(nPDPA-r-Cy3.5) and mPEG5k-b-(PEPA-r-BBQ650). First, we characterized the 1H-NMR spectra of the nanoprobes in D2O at different apparent pH values (pHa). As shown in Fig. 3a, only the proton resonance peak of the mPEG5k segment appeared at the OFF1 state, while the resonance of the PR segment was completely suppressed owing to the highly hydrophobic micelle cores. When the pHa dropped to about 6.96, resonance of the PEPA segment gradually appeared, implying the protonation of tertiary amine and the solvation of this segment (ON state). For the nPDPA segment with higher hydrophobicity, the resonance signals did not appear until the pHa further decreased to about 6.41 (OFF2 state). These results demonstrated that the pH-sensitive copolymers in BLINK were sequentially protonated at different pH environment.
a illustration of BLINK[6.1, 6.5], and 1H-NMR spectra (in D2O) of the nanoprobe at different apparent pH values (pHa), pKw (D2O, 25 °C) = 14.87, pH (pD, neutral D2O, 25 °C) = 7.44. b correlogram acquired by DLS and the TEM images of BLINK at different states. Scale bar = 100 nm. c fractogram of FFF-MALS for the nanoprobe at different states. The light-scattering intensity at 90° (left, dash lines) and the molar mass determined for each fraction (right, solid lines) were shown. The cross flow (0.8 mL min−1) in FFF separation system for OFF2 state was lower than that (3.5 mL min−1) for OFF1 and ON state to acquire approximate elution time. Fraction of the assembly rather than the disassembly at ON state was shown and determined. d visualization of the assembly and the disassembly fraction of the nanoprobe at different stages by ultrafiltration (MWCO = 100 kDa). e Stopped-flow fluorescence measurement of the nanoprobe during two state-transition processes (OFF1-to-ON and ON-to-OFF2 transitions). f time-resolved fluorescence energy transfer (λex = 405 nm, extracted at different time) of the nanoprobe at ON state. g schemes showing the molecular mechanism of BLINK, which based on pH-triggered sequential disassembly and FRET effect. Source data are provided as a Source Data file.
We further investigated the morphology of BLINK at different states. Dynamic light scattering (DLS) and transmission electron microscopy (TEM) results (Fig. 3b, Supplementary Fig. 15) showed that BLINK was a spherical particle with hydrodynamic diameter of 30.6 nm and polydispersity of 0.081 at OFF1 state. At ON state, the measured correlation curve decayed more rapidly, indicating that the particle size slightly decreased due to the dissociation of protonated mPEG5k-b-PEPA copolymers, with a diameter of 28.0 nm and polydispersity of 0.112. While at OFF2 state, the DLS results did not meet quality criteria because the nanoprobe dissociated completely as TEM images showed. To finely parse the phase transitions of BLINK at different stages, we used a field-flow fractionation coupled with multi-angle static light scattering (FFF-MALS) system to measure the molar weight (Mn) and the aggregation number (Nagg, determined as Mn/Munimer) of the nanoparticles (Fig. 3c). The Mn of the assembly entity was decreased from 2.3 × 106 to 1.5 × 106 at OFF1 and ON states, corresponding to the Nagg from 107 to 68. Moreover, this phase transition could be visualized by gel column chromatography (Supplementary Fig. 16a) and ultrafiltration method (MWCO = 100 kDa, Fig. 3d). The nanoprobe was stayed in the upper layer at pH 7.4, while during the phase transition from OFF1 to ON when pH drops to 6.3, the protonated mPEG5k-b-PEPA labeled with BBQ650 were separated from the residue nanoparticle labeled with Cy3.5 and DEAC. The isolated protonated polymers at ON state using ultrafiltration method were characterized by 1H-NMR as BBQ650-conjugated mPEG5k-b-PEPA copolymers (Supplementary Fig 16b). At the phase transition from ON to OFF2 states, the nanoprobe was completely dissociated into unimers as demonstrated by FFF-MALS analysis; the dissociated unimer was thoroughly collected in the lower layer at pH 5.4 as indicated by ultrafiltration method. These results demonstrated that BLINK exhibited a cascade micelle-to-unimer phase transition over pH changes.
To test the temporal response of BLINK upon pH changes, stopped-flow experiments were applied to monitor the fluorescence intensities during the phase transition processes. As shown in Fig. 3e, the kinetics of the phase transition were ultra-fast with a recovery half-life (t1/2 for OFF1-to-ON transition) of 14 ms and a decay half-life (t1/2 for ON-OFF2 transition) of 5 ms. Additionally, the time-resolved fluorescence emission spectra of the nanoprobe at ON state showed an efficient energy transfer process from DEAC to Cy3.5 (Fig. 3f, Supplementary Fig. 17). Therefore, the molecular mechanisms of band-pass pH response were clarified, which was based on pH-triggered cascade nanophase separation and the FRET relay effect as we hypothesized (Fig. 3g).
We further conducted a series of in vitro stability tests to evaluate the feasibility of utilizing the BLINK nanoprobe for biological applications. BLINK[6.1, 6.5] showed stable and reproducible pH responsiveness over 1-week storage (Supplementary Fig. 18a) and with biochemical factors and bioactive molecules including fetal bovine serum, potassium chloride, hydrogen peroxide, glutathione and sodium chloride (Supplementary Fig. 18b). Moreover, BLINK[6.1, 6.5] exhibited consistent pH-responsive behavior both in the presence and absence of FBS, demonstrating its stability for monitoring dynamic pH fluctuations in biological samples (Supplementary Fig. 18c–e). We also test the nanoprobe at different concentrations of 10 μg/mL and 500 μg/mL, and the results were consistent, indicating good reproducibility of BLINK nanoprobe (Supplementary Fig. 19).
BLINK specifically illuminates the acidic environment of distinct endocytic organelles
Endocytic pathway is linked with numerous biological6,62,63 and pathological64,65 processes, and plays an essential role in drug delivery66,67,68,69. The associated organelles maintain unique lumenal pH values at different stages70 for cellular homeostasis. BLINK can be taken up by different type of cells and the cellular signal increased in a dose-dependent manner (Supplementary Fig. 20a, b). The pH value of culture medium (>pHt1) has negligible influence on the nanoprobe uptake (Supplementary Fig. 20c). So, we next applied BLINK to monitor the lumenal pH of endocytic organelles. First, the Panc02 cells were treated with BLINK[6.1,6.5] for cellular internalization, and a proton-pump inhibitor (bafilomycin A1) was co-incubated to prevent the acidification of the endocytic organelles. Then, the cellular pH was clamped to 7.4, 6.3, or 5.4 with PBS buffers containing nigericin (10 μM) and valinomycin (10 μM), respectively. Using pH-sensitive dyes as reference, we verified that the cellular pH values can be uniformly clamped (Supplementary Fig. 21). Confocal images and flow cytometry results showed that BLINK[6.1,6.5] only fluoresced strongly at pH 6.3 in the band-pass channel, while exhibited strong fluorescence both at pH 6.3 and 5.4 in the low-pass channel (Fig. 4a Supplementary Fig. 20d). The cellular fluorescence intensity depends on nanoprobe concentration and instrument acquisition parameters. To normalize these variables, we employed the LP channel signal as an internal reference. As shown in Supplementary Figs. 20e and 22, the cellular BP/LP ratio at ON state was consistent over different nanoprobe concentrations under the same acquisition settings, indicating that the uneven distribution of nanoprobes within cells would not affect the detection results. This result validated the feasibilities of specific imaging of distinct endocytic organelles with different pH range by BLINK. So, we applied the pulse-chase assay to evaluate the maturation kinetics of endocytic organelles with BLINK[6.1, 6.5]. The fluorescence signals in band-pass channel gradually lighted up and then decayed during the chase imaging process, while the signal in low-pass mode remained consistent after its activation (Supplementary Fig. 23). Moreover, the pH detection results showed high consistency among different quantitative methods (Supplementary Fig. 24). These data revealed that BLINK[6.1,6.5] can specifically image the targeted endocytic organelles with lumenal pH from 6.1 to 6.5 during endosome maturation.
a representative fluorescent images of BLINK at different pH values in live Panc02 cells. The cells were treated with nigericin (10 μM) and valinomycin (10 μM) in the PBS buffers with different pH values. Red (BP), FRET-Cy3.5 excited by 405 nm; green (LP), Cy3.5 excited by 561 nm. b representative confocal and fast FLIM images of BLINK in live cells at single-organelle resolution. Panc02 cells were chased at 30 min after 10-min pulse with BLINK [6.1, 6.5]. Phasor analysis was applied to measure fluorescence lifetime of DEAC and the FRET efficiency between DEAC and Cy3.5. Blue, DEAC; red (BP), FRET-Cy3.5; c fast FLIM images of DEAC at different chase time for analysis of the endosome maturation. The fluorescence lifetime of DEAC increased as endosomes maturation. FLIM: λex = 440 nm, λem = 450-520 nm.
To corroborate the cascade phase transition of BLINK in endocytic organelles, we performed a FRET imaging analysis between the donor and acceptor in live cells treated with BLINK[6.1,6.5]. Results showed an efficient FRET effect from DEAC to Cy3.5 in the endocytic organelles with pH value clamped within interval of [6.1, 6.5] (Supplementary Fig. 25), while the FRET effect was abolished in organelles with pH value below pHt2. These data demonstrated the cascade micelle-to-unimer phase transition as the acidification of endocytic organelles during endosome maturation. We also used fluorescence lifetime imaging microscopy (FLIM) to probe the phase states of BLINK in single organelles resolution at different maturation time. Before FLIM imaging, confocal images were captured at 30 min after the 10-min pulse (Fig. 4b). All FRET-Cy3.5 signal dots were colocalized with a subset of DEAC signal dots. The colocalized dots showed shorter fluorescence lifetime of DEAC (τ = 1.045 ns) due to its energy transfer to Cy3.5 as compared to completely dissociated counterpart (τ = 1.702 ns). The phasor plot analysis also showed that the FRET efficiency reached about 57% at 30 min post-treatment. Moreover, the fluorescence lifetime gradually increased as endosome maturation (Fig. 4c and Supplementary Fig. 26), indicating the micelle-to-unimer transition from ON to OFF2 stage as the acidification of corresponding organelles. These results provided solid evidence for the work mechanism of BLINK in the cellular environment, and further validated the reliability of precise pH reporting of each endocytic organelle via BLINK imaging. We also verified the applicability of BLINK to different type of cell lines (Supplementary Fig. 27). Compared to the tumor cells and fibroblast cells, dendritic cells exhibited slower acidification kinetics (Supplementary Fig. 28), which can prevent proteolytic degradation and promote the antigen processing and presentation4,62.
We further compared the BLINK with the traditional probes and biomarkers for endocytic organelles. LysoTracker, a widely used fluorophore of acidic organelles in live cells, was co-incubated with BLINK[6.1,6.5] (Fig. 5a and Supplementary Fig. 29). pH band-pass signals indicated the organelles with pH values between 6.1 and 6.5, while the pH low-pass signals represented the organelles with pH values below 6.5. The colocalization results showed that LysoTracker only stained a small population of endocytic organelles with a pH value between 6.1 and 6.5. To further corroborate whether BLINK can be activated in different endocytic organelles, we transfected A549 cells with green fluorescent protein (GFP)-fused Rab5a, Rab7a, and LAMP1 for the tracking of early endosomes, late endosomes, and lysosomes, respectively. For BLINK[6.1,6.5], 86% of the band-pass signals were colocalized with Rab5a-positive organelles, whereas only 19% and 6% of the band-pass signals were colocalized with Rab7a-positive and LAMP1-positive organelles, respectively (Fig. 5b, c and Supplementary Fig. 30a–c). As for BLINK[5.3,5.7], 82% and 33% of the band-pass signals were colocalized with Rab7a-positive and LAMP1-positive organelles, respectively (Fig. 5d, e and Supplementary Fig. 30d–f), whereas only 4.5% of signals were located in Rab5a-positive organelles. Altogether, these data demonstrated that BLINK with tunable pHt1 and pHt2 thresholds can accurately discriminate and report the pH regulation of the endocytic organelles during endosome maturation at single organelle resolution.
a representative confocal images of live cells co-incubated with BLINK[6.1, 6.5] (100 μg mL−1) and LysoTracker deep red (10 nM). Colocalization analysis at single-organelle resolution showed the threshold pH of LysoTracker deep red for acidic organelles imaging. Red (BP), FRET-Cy3.5 excited by 405 nm; green (LP), Cy3.5 excited by 561 nm; blue, LysoTracker deep red excited by 633 nm. b representative confocal images showed colocalization of BLINK[6.1, 6.5] (red) with different endocytic organelles, including early endosomes (Rab5a-GFP, green), late endosomes (Rab7a-GFP, green) and lysosomes (LAMP1-GFP, green). c frequency of colocalization between BLINK[6.1, 6.5] with biomarkers from (b). d representative confocal images showed colocalization of BLINK[5.3, 5.7] (red) with different endocytic organelles (green). e frequency of colocalization between BLINK[5.3, 5.7] with biomarkers from (d). Source data are provided as a Source Data file.
BLINK specifically images the acidotic tumor pHe in tumor-bearing mice
For in vivo imaging studies, we first engineered BLINK[6.5,6.9] to target the acidic extracellular pH environment in tumors. BLINK nanoprobe could passively accumulate in the tumor tissues via enhanced permeability and retention (EPR) effect. BLINK[6.5,6.9] (50 mg kg−1) was intravenously injected into mice bearing subcutaneous 4T1 breast tumors, and the isolated organs were imaged through the pH band-pass and low-pass channels at 24 h post-injection, respectively (Fig. 6a). Significant contrast between tumor and other tissues was observed in pH band-pass channel. Especially, the tumor/liver (T/L) ratio for pH band-pass channel was 3.29 ± 0.86, while this parameter was only 1.56 ± 0.38 at low-pass channel (p = 0.0128, Supplementary Fig. 31). These results demonstrated that BLINK can provide more accurate imaging of the acidic tumor microenvironment through excluding the more acidic endocytic signals after cellular endocytosis as compared to low-pass pH-gated (without pHt2) nanoprobes (Fig. 6b).
a representative fluorescent images of the isolated organs at 24 h post-injection of BLINK[6.5, 6.9]. For pH band-pass imaging, λex = 430 nm, λem = 620 nm; for pH low-pass imaging, λex = 570 nm, λem = 620 nm. b illustration of the optimized specificity and sensitivity for tumor pHe imaging by tuning the thresholds of pHt1 and pHt2. Compared to the traditional one-gated (without pHt2) pH-responsive probes, BLINK eliminated the signals in the more acidic endocytic compartments. c heat map shows the T/L (tumor/liver) ratio of nanoprobes with different pHt1 and pHt2 (n = 3 biological independent mice). d representative pH band-pass images of the isolated organs in different groups at 24 h post-injection of nanoprobes. Bay-876 and AZM were use as inhibitors of GLUT1 and carbonic anhydrase IX to neutralize the tumor pHe. e T/L ratio of different groups in (d). Data are presented as mean ± s.d. (n = 3 biological independent mice). One-way analysis of variance (ANOVA). f bioimaging of orthotopic 4T1-tumor-bearing mice with luciferase reporters. g confocal images of the tissue sections for pH band-pass imaging channel. λex = 405 nm, λem = 590-650 nm. h–l in vivo imaging of different cancer types in different tumor models. h orthotopic Panc02 models (pancreatic cancer); i orthotopic CT26 models (colorectal cancer cancer); j mesentery metastases models (4T1). k orthotopic Hepa1-6 models (hepatocarcinoma). l orthotopic HepG2 models (hepatocellular carcinoma). Images were captured at 24 h post-administration of nanoprobes. Source data are provided as a Source Data file.
By tuning the threshold of pHt1 and pHt2, we prepared a library of BLINK with 12 nanoparticles. The pHe imaging efficacy of BLINKs was screened in 4T1 tumor-bearing mice to optimize the sensitivity and specificity. As shown in Fig. 6c, BLINK exhibited high specificity for tumor pHe imaging with high T/L ratio (>2) in a variety of combinations of pHt1 and pHt2 modules. BLINK[6.3,6.9] with the highest T/L ratio of 3.78 was exploited for further imaging studies.
To verify the specificity of tumor pHe imaging by BLINK[6.3,6.9], we next evaluated the T/L ratio of control groups treated with inhibitors of tumor acidosis or OFF/ON type UPS nanoprobe71 (UPS6.9 with pHt of 6.9, Fig. 6d, e). Bay-876 (10 mg kg−1 every 12 h), an orally active glucose transporter 1 (GLUT1) inhibitor72, and acetazolamide sodium (AZM, 100 mg kg−1 every 12 h), a carbonic anhydrase (CA) IX inhibitor72, were administrated 12 h before BLINK[6.3,6.9] injection and repeated thrice, respectively. Treatment with these metabolic inhibitors resulted in a significant decrease of T/L ratio (1.32 ± 0.54 for Bay-876, p = 0.0032, and 0.59 ± 0.26 for AZM, p = 0.0003). Moreover, UPS6.9 as traditional pH-responsive nanoprobes with high accumulation and clearance in the liver, also showed lower T/L ratio as compared to BLINK (0.35 ± 0.12, p = 0.0001). For in vivo tumor pHe imaging, BLINK[6.3,6.9] also exhibited high tumor-to-normal tissues contrast in orthotopic 4T1 breast tumor models (Fig. 6f). The fluorescence imaging of organ sections showed that the BLINK[6.3,6.9] was specifically activated in tumors rather than the normal tissues such as muscles (Fig. 6g).
Finally, we test the universality of BLINK[6.3,6.9] for the imaging of acidotic tumor pHe in several tumor-bearing mice models, including orthotopic Panc02 pancreatic cancer, CT26 colorectal cancer, and 4T1 mesenteric metastases models. The imaging results demonstrated that BLINK[6.3,6.9] can efficiently amplify the acidic tumor pHe signals with high contrast over the surrounding normal tissues (Fig. 6h–j). Encouraged by the high T/L contrast of BLINK[6.3,6.9], we also challenged the imaging of orthotopic liver cancer. As shown in Fig. 6k, l, BLINK[6.3,6.9] distinguished the Hepa 1–6 and HepG2 xenografts from normal liver tissues with high contrast. Collectively, these results highlighted that BLINK[6.3,6.9] can serves as a specific and powerful nanoprobe for tumor pHe imaging of various solid tumors.
Discussion
In this study, we developed cascade-heterogated proton nanotransistors for unambiguously discriminating and reporting distinct pH ranges in biological environments. The BLINK technology leverages pH-driven cascade three-phase transition of multi-modular nanoprobes to modulate the FRET relay between encoded dye molecules, thereby achieving pH band-pass performance. The BLINK integrates two transistor-like polymeric modules, each of which can digitize analog proton signals with high fidelity without crosstalk. We observed the elegant and sharp OFF1-ON-OFF2 three-phase transitions in vitro and in biological systems, which is the prerequisite for specifically distinguishing unique and narrow pH ranges. Building on the mechanism of cascade-heterogated proton nanotransisitor, we successfully fabricated a library of BLINKs for specific imaging of pathophysiological acidic environments, including glycolytic acidotic extracellular milieu (pHe = 6.5–6.9), early endosomes (pHi = 6.0-6.5), and late endosomes (pHi = 5.5–6.0) with specific pH ranges.
So far, conventional pH-sensitive probes (e.g., LysoTracker) specifically target the highly acidic lysosomes rather than low acidic early endosomes and late endosomes. Leveraging our technology, we have achieved high-resolution imaging of acidic early compartments of the endocytic pathway without introducing signal crosstalk from off-target ranges. More importantly, we achieved the specific imaging of glycolytic acidotic pHe, a ubiquitous hallmark of cancer, which displays a narrow and fluctuating pH ranges across different types of cancers. Previously, we reported the development of low-pass UPS nanoprobes (pHt = 6.9) to image solid tumors by the non-linear amplification of pHe signals. However, the internalization-triggered activation of nanoprobes by normal cells adjacent to tumor tissues may lead to blurred contrast between tumor and normal tissues, especially the orthotopic carcinoma in liver (a mononuclear macrophage system for nanoprobe uptake and activation). By targeting the specific range of pHe, BLINK with an optimized pH range has shown broad tumor specificity for a variety of cancer types (breast, pancreatic, colorectal, liver and peritoneal metastasis) in the orthotopic xenograft mouse models with higher tumor-to-normal tissue contrast as compared with low-pass UPS nanoprobes.
In summary, we developed the transistor-like band-pass proton nanoprobes based on the mechanism of cascade-heterogated nanophase separation and tandem FRET relay. The heterogated proton nanotransistors encoded with distinct and tunable pH thresholds demonstrate the selective imaging of multiplex acidic intracellular and extracellular compartments with high contrast. This study strives for further improvement on the design of pH-sensing techniques, and offers insights into the interval-based biomedical imaging.
Methods
Ethical statement
This research complies with all relevant ethical regulations. All animal studies were conducted in accordance with the National Institute Guide for the Care and Use of Laboratory Animals. The experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Peking University (Accreditation number: LA 2019039).
Materials
All chemicals were purchased from J&K Scientific unless otherwise stated. 7-Diethylaminocoumarin-3-carboxylic acid (DEAC) was purchased from Bide pharm Technology Co., Ltd. Dyes including DQFRB, BBQ650, and TVP were prepared following published literatures (see Supplementary Information). Cy3, Cy3.5, Cy5.5 and Cy7 were purchased from Lumiprobe Corporation. Monomers, 2-aminoethyl methacrylate hydrochloride (AMA-MA) and 2-(diethylamino) ethyl methacrylate (DEA-MA) were purchased from Polyscience Company. Macroinitiator mPEG5k-Br and other monomers, such as 2-(dibutylamino) ethyl methacrylate (DBA-MA), 2-(dipropylamino) ethyl methacrylate (nDPA-MA), and 2-(N-ethyl-N-propyl amino) ethyl methacrylate (EPA-MA), were prepared in our lab33. Copper(I) bromide and N, N, N’, N″, N’-pentamethyldiethylenetriamine (PMDETA) were purchased from Sigma-Aldrich. Nigericin sodium salt, valinomycin, acetazolamide sodium, and bafilomycin A1 were purchased from MedChemExpres. pHrodoTM Red Dextran, Alexa Fluor 647 Dextran, LysoTrackerTM Deep Red and LysoTrackerTM Green DND-26 were purchased from Invitrogen. BAY-876 was purchased from Shanghai Yuanye Bio-Technology Co., Ltd.
Synthesis of mPEG5k-b-PR and mPEG5k-b-P(R-r-D/A/Q) dye-conjugated block copolymers
Dye-free block copolymers mPEG5k-b-PR were synthesized by atom transfer radical polymerization (ATRP) method32. Taking mPEG5k-b-PEPA as an example, the typical procedure was as follows. First, mPEG5k-Br (0.04 mmol, 200 mg), EPA-MA (3.2 mmol, 636.8 mg), AMA-MA (0.12 mmol, 20 mg), PMDETA (20 μL) were dissolved with 1 mL of dimethylformamide (DMF) and 1 mL of 2-propanol in a polymerization tube. The catalyst CuBr (6 mg) was added to the mixture under a nitrogen atmosphere after three cycles of freeze-pump-thaw. Then, polymerization was performed at 40 °C for 12 h. Afterwards, the reaction mixture was successively dialyzed against ethanol with 0.1 % (v/v) PMDETA and against Milli-Q water. Finally, a white powder was obtained by lyophilization. The purified block copolymers were characterized by 1H-NMR (AVANCE III 400, Bruker, USA).
For dye conjugation of copolymers, 20 mg of mPEG5k-b-PR was dissolved in 500 μL DMF, and each dye (D for donor, A for acceptor, and Q for quencher) dissolved in DMF (10 mg mL−1) was added at a required feeding ratio. Then, O-benzotriazol-1-yl-tetramethyluronium (HBTU, 1.2 eq. to dye) was added into the mixture. The reaction mixture was maintained at room temperature overnight in the dark. Subsequently, the crude product was purified by dialysis (MWCO = 10 kDa) against methanol. Finally, the dye-conjugated copolymers were dispersed in Milli-Q water, lyophilized, and kept at −20 °C for storage. The number of conjugated dyes in each copolymer was determined by a UV–Vis spectrophotometer (UH5300, Hitachi, Japan).
Fabrication of BLINK nanoprobe
BLINK nanoprobe was formulated by a pH-induced self-assembly procedure. Firstly, three dye-labeled polymers mPEG5k-b-P(R-r-D), mPEG5k-b-P(R-r-A), and mPEG5k-b-P(R-r-Q) were separately dissolved in 10 mM HCl to a final concentration of 5 mg mL−1. Then, the three solutions were mixed at a 1:1:1 ratio by volume. The mixture was further diluted with Milli-Q water to a total copolymer concentration of 1 mg mL−1. To fabricate BLINK by co-assembly, the pH value of the mixed solution was adjusted to around 8 with 1 M NaOH solution. Subsequently, the inorganic salts were removed through micro-ultrafiltration system (MWCO = 100 kDa), and BLINK solution was adjusted to 10 mg mL−1 as stock solution for further studies.
pH-dependent fluorescence activation of BLINK nanoprobe
The BLINK stock solution was diluted to a final polymer concentration of 100 μg mL−1 in phosphate buffer saline (10 mL, pH 7.4). Then, the working solution of BLINK was titrated to predesignated pH with 10 or 100 mM HCl and a pH-meter was used to monitor the pH value. Meanwhile, fluorescence emission spectra of BLINK at different pH values were obtained on a fluorescence spectrophotometer (F-7000, Hitachi, Japan). The excitation wavelength (λex) of different donors and the emission wavelength (λem) of different acceptors are as follows: DEAC (405 nm), TVP (460 nm); Cy3 (430-700 nm), Cy3.5 (430–650 nm), DQFRB (430–750 nm), Cy5.5 (500–800 nm). The intensities at maximum emission wavelength of acceptors were used to quantify the OFF-ON-OFF properties of BLINK. To evaluate the stability of BLINK nanoprobe in vitro, biochemical factors and bioactive molecules including fetal bovine serum, potassium chloride, hydrogen peroxide, glutathione and sodium chloride were separately added to the solution. Time-resolved emission spectra at ON state were acquired on a transient fluorescence spectrometer (LifeSpec-II, Edinburgh Instruments, UK).
Fluorescent images of nanoprobe solutions (100 μg mL−1) at different pH values were captured on an IVIS in vivo imaging system (IVIS SPECTRUM, PerkinElmer, USA). For pH band-pass imaging, the excitation filters were based on FRET donors, i.e., 430 nm for DEAC and 465 nm for TVP, respectively. While for pH low-pass imaging, the excitation filters were based on FRET acceptors, i.e., 530 nm for Cy3, 570 nm for Cy3.5, 570 nm for DQFRB, and 675 nm for Cy5.5. For two kind of imaging modalities, the emission filters were chosen for each acceptor, i.e., 580 nm for Cy3, 620 nm for Cy3.5, 660 nm for DQFRB, and 720 nm for Cy5.5.
pH-triggered sequential disassembly analysis
1H-NMR spectroscopy was used to confirm the core-shell structure of nanoprobes at different apparent pH (pHa) values. The samples were prepared as mentioned above, except that the solvent, acid, and buffer system were replaced by deuterated reagents.
The morphology of BLINK nanoprobes at different states was confirmed by transmission electron microscopy (JEM-1400, JEOL, Japan). The particles size and distribution were measured by dynamic light scattering analysis (Zetasizer Nano ZSP, Malvern, UK).
The molar mass and aggregation number (Nagg) of BLINK nanoprobes at different states were obtained and calculated by a field-flow fractionation coupled with multi-angle static light scattering (FFF-MALS) system. First, the samples were separated based on molecular weight and size by a FFF separation system (Eclipse, Wyatt technology, US) using PBS buffer with different pH values, which favored the analysis of assembled and unassembled components. The separation system was sequentially connected to a light scattering detector (DANW, Wyatt technology, US) and a refractive index detector (Optilab, Wyatt technology, US). The number-averaged molar masses (Mn) were reported by FFF-MALS system and the Nagg of nanoprobe was determined using the relation of Nagg = Mn/Munimer, where Munimer is the number-averaged molecular weight of block copolymers calculated by NMR.
The state-transition kinetics of BLINK nanoprobes was performed with a stopped flow spectrophotometer (SX-20, Applied Photophysics, UK). For the turn-on process, the nanoprobe solution (1 mg mL−1) was mixed with HCl at an equal volume from two syringes. Meanwhile, the fluorescence intensity was recorded using an excitation wavelength of 430 nm (D = DEAC) and a long-pass emission filter centered at 575 nm (A = Cy3.5). For the turn-off process, the protocols were similar, except that the initial state of the nanoprobes was adjusted to the ON state. To ensure the state-transition occurred exactly upon the mix, the concentrations of HCl were optimized in advance.
Cell culture
4T1 breast cancer, CT26 colorectal carcinoma, Panc02 pancreatic cancer, A549 lung carcinoma, Hepa1-6 hepatocarcinoma, HepG2 hepatocellular carcinoma cells and 3T3/NIH fibroblast cells were purchased from National Infrastructure of Cell Line Resource (Beijing, China). Murine DC2.4 cells were purchased from China Center for Type Culture Collection. 4T1 and CT26 cells were cultured in RPMI 1640 medium with 10% fetal bovine serum (FBS) and antibiotics (penicillin 100 U mL−1 and streptomycin 100 μg mL−1). Panc02, A549, Hepa1-6, 3T3/NIH, DC2.4 and HepG2 cells were cultured in DMEM medium with 10% FBS and antibiotics. All cells were cultured in a humidified environment with 5% CO2 at 37 °C.
Confocal imaging for cell trafficking
The pulse-chase method was applied to evaluate the cellular activation of BLINK nanoprobes as endosome-trackers. In a typical procedure, the cells were seeded in glass-bottom dishes (Corning) and incubated overnight at 37 °C in a 5% CO2 atmosphere for adhesion. The cells in complete medium were precooled at 4 °C for 10 min and then 100 μg mL−1 of BLINK nanoprobe was added. After another 10-min incubation at 4 °C, the medium was removed and the cells were washed three times with PBS buffer. Following the pulse, the cells were chased by confocal imaging (LSM880, Zeiss, Germany) at 37 °C. The 405 nm and 561 nm lasers were used for excitation of DEAC and Cy3.5, respectively, and the emission range was set to 590–650 nm. The ratiometric images were generated by ImageJ software (NIH).
For confocal imaging of BLINK at predesignated cellular pH values, the cells were pre-incubated with 0.1 μM bafilomycin A1 for 8 h. And the samples were treated according to the procedure described above, except that the cells were kept in the medium containing bafilomycin A1 (0.1 μM) to inhibit the acidification of endocytic organelles. Finally, the cells were kept in PBS buffers with different pH values containing 10 μM nigericin and 10 μM valinomycin, and the fluorescent images were captured with the same acquisition settings as previously mentioned.
For lysosome imaging, LysoTrackerTM Deep Red or LysoTrackerTM Green DND-26 (10 nM) was added and kept at 37 °C. LysoTracker Deep Red was excited with 633 nm laser and Cy5 emission filter was used for imaging, while 488 nm laser and GFP filter were used for LysoTracker Green.
For colocalization studies of BLINK with endocytic organelles, A549 cells stably transfected with Rab5a-GFP, Rab7a-GFP, or LAMP1-GFP were used, respectively. Fluorescent images were captured by confocal laser scanning microscopy (A1R-Storm, Nikon, Japan). BLINK (DEAC as donors) were excited by 405 nm laser, and TRITC filter were used for imaging capture. Meanwhile, GFP-labeled organelles were imaged by 488 nm laser excitation with GFP filter. The colocalization between BLINK and organelles were analyzed with ImageJ software (NIH).
Cellular uptake and pH-induced activation analysis
The quantitative analysis of cellular uptake and pH-induced activation were also performed on a flow cytometer (Cytoflex LX, Beckman Coulter, US). Live cells were seeded in 24-well plates (8 × 104 cells per well) and incubated overnight. For cellular uptake analysis, the cells were treated with BLINK nanoprobe at 4 °C for 10 min, and then incubated at 37 °C for 3 h allowing endocytic organelle maturation. Cells were subsequently harvested and cellular fluorescence (fully activated LP signal) was measured via Y610 channel on Cytoflex LX. For pH-induced cellular fluorescence activation analysis, cells were preincubated with bafilomycin A1 to inhibit the acidification of endocytic organelles during maturation, followed by modulating intracellular pH values with buffers containing nigericin and valinomycin as mentioned before. The BP signal was measured via V610 channel on Cytoflex LX. Cell populations were gated for a live population using FSC and SSC plot of cell only sample. The gate was set to remove cell debris (small FSC v SSC) and large aggregates of cells (large FSC or SSC) and used across all samples.
Fluorescence lifetime imaging microscopy (FLIM)
BLINK nanoprobe was loaded into live cells as described above. The FLIM images were acquired with a STELLARIS 8 confocal microscope platform (Leica, Germany). DEAC was excited by a 440-nm modulated diode laser working at a 20 MHz repetition rate, and images were acquired upon detection in the 450–520 nm range by a Leica HyD X detector. Confocal images based on fluorescence intensities were acquired before fluorescence lifetime imaging, with a 405 nm laser excitation and two HyD R detectors ranged from 430–520 nm (DEAC) and 590–650 nm (Cy3.5), respectively. Phasor plot analysis (average fluorescence lifetime and FRET efficiency) was performed with LASX FLIM software (Leica, Germany).
Animal models
Female BALB/c mice (6–8 weeks, 18–20 g), female C57BL/6J mice (6–8 weeks, 18–20 g) and female NU/NU mice (6–8 weeks, 18–20 g) were purchased from Vital River Laboratory Animal Center (Beijing, China). Animals were housed in groups of 3–5 mice per cage under specific pathogen free conditions and allowed to acclimatize for one week before experimentations. The animals were housed in a humidity-controlled environment at 25 °C with free access to food (standard chow diet, catalog number HD8013, Huanyu Biotechnology Co., Ltd.) and water in a 12 h dark/light cycle. Orthotopic 4T1 breast cancer model was established by inoculating 4T1 cells (1 × 106 cells/mouse) into the mammary fat pads of BALB/c mice. Tumor-bearing mice were applied for imaging studies when the tumors reached ~200 mm3. For intra-abdominal metastatic model, BALB/c mice were intraperitoneally injected with 4T1 cells (1 × 106 cells/mouse). To demonstrate the broad application for acidic tumor pHe imaging, other orthotopic tumor models, including Panc02 pancreatic cancer, CT26 colorectal carcinoma, Hepa1-6 hepatocarcinoma, and HepG2 hepatocellular carcinoma were established. For tumor-imaging studies, the maximal tumor size of 1500 mm3 was permitted by IACUC of Peking University. We confirm that the maximal tumor size was not exceeded in any of the experiments.
In vivo and ex vivo fluorescence imaging
BLINK nanoprobes (50 mg kg−1, DEAC as donor, Cy3.5 as acceptor, and BBQ650 as quencher) with predesignated pHt1 and pHt2 were administrated intravenously into the tumor-bearing mice. Fluorescent images were captured on an IVIS in vivo imaging system at 24 h post-administration. For pH band-pass imaging, 430 nm/620 nm filter pair was applied. While 570 nm/620 nm filter pair was chosen for pH low-pass imaging. For ex vivo fluorescence imaging, the mice were sacrificed, and tumor together with major organs were dissected. The isolated tumor and organs were imaged by IVIS system. As controls, acetazolamide sodium (100 mg kg−1 every 12 h, thrice, i.v.) or BAY-876 (10 mg kg−1 every 12 h, thrice, i.g.) was administrated 12 h before the intravenous injection of BLINK nanoprobe. The fluorescent images were captured at 24 h post-injection of BLINK. The fluorescence intensities were quantified using Living Image 4.3.1 (PerkinElmer, USA) with region of interest (ROI) analysis. Tumor/Liver (T/L) ratios were determined by comparing the average radiant efficiency in the tumor and the liver.
Statistics and reproducibility
Data were presented as mean ± s.d. Statistical analysis were performed using Origin 2019. Significant differences were assessed using paired t-test or one-way analysis of variance (ANOVA) test. P < 0.05 was considered statistically significant. pH-responsive experiments were performed at least three times with good reproducibility. TEM images and DLS in Fig. 3b were repeated thrice independently with similar results. Confocal and fast FLIM imaging of live cells incubated with BLINK nanoprobe were repeated at least three times with similar results, and a series of representative images from each group were shown, such as Figs. 4a–c and 5a, b, d. For in vivo and ex vivo imaging, the experiments were repeated independently at least three times using biological independent replicates that showed similar results and representative images were demonstrated. For fluorescence imaging of the tissue sections in Fig. 6g, the experiments were repeated independently three times using biological independent mice that showed similar results and representative images were shown.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information files. Source data are provided with this paper.
References
Hamm, L. L., Nakhoul, N. & Hering-Smith, K. S. Acid-base homeostasis. Clin. J. Am. Soc. Nephrol. 10, 2232–2242 (2015).
Aoi, W. & Marunaka, Y. Importance of pH homeostasis in metabolic health and diseases: crucial role of membrane proton transport. Biomed. Res. Int. 2014, 598986 (2014).
Rodrigues, R. C., Ortiz, C., Berenguer-Murcia, A., Torres, R. & Fernandez-Lafuente, R. Modifying enzyme activity and selectivity by immobilization. Chem. Soc. Rev. 42, 6290–6307 (2013).
Delamarre, L., Pack, M., Chang, H., Mellman, I. & Trombetta, E. S. Differential lysosomal proteolysis in antigen-presenting cells determines antigen fate. Science 307, 1630–1634 (2005).
Joffre, O. P., Segura, E., Savina, A. & Amigorena, S. Cross-presentation by dendritic cells. Nat. Rev. Immunol. 12, 557–569 (2012).
Okreglak, V. et al. Cell cycle-linked vacuolar pH dynamics regulate amino acid homeostasis and cell growth. Nat. Metab. 5, 1803–1819 (2023).
Gatenby, R. A. & Gillies, R. J. Why do cancers have high aerobic glycolysis?. Nat. Rev. Cancer 4, 891–899 (2004).
Webb, B. A., Chimenti, M., Jacobson, M. P. & Barber, D. L. Dysregulated pH: a perfect storm for cancer progression. Nat. Rev. Cancer 11, 671–677 (2011).
Estrella, V. et al. Acidity generated by the tumour microenvironment drives local invasion. Cancer Res. 73, 1524–1535 (2013).
Ghoneim, M. T. et al. Recent progress in electrochemical pH-sensing materials and configurations for biomedical applications. Chem. Rev. 119, 5248–5297 (2019).
Steinegger, A., Wolfbeis, O. S. & Borisov, S. M. Optical sensing and imaging of pH values: spectroscopies, materials, and applications. Chem. Rev. 120, 12357–12489 (2020).
Aref, M. et al. Potentiometric pH nanosensor for intracellular measurements: real-time and continuous assessment of local gradients. Anal. Chem. 93, 15744–15751 (2021).
Zhang, Y. et al. High-resolution label-free 3D mapping of extracellular pH of single living cells. Nat. Commun. 10, 5610 (2019).
Grashei, M. et al. Simultaneous magnetic resonance imaging of pH, perfusion and renal filtration using hyperpolarized (13)C-labelled Z-OMPD. Nat. Commun. 14, 5060 (2023).
Duwel, S. et al. Imaging of pH in vivo using hyperpolarized (13)C-labelled zymonic acid. Nat. Commun. 8, 15126 (2017).
Gallagher, F. A. et al. Magnetic resonance imaging of pH in vivo using hyperpolarized 13C-labelled bicarbonate. Nature 453, 940–943 (2008).
Sun, A., Tang, X., Nie, D., Fan, Y. & Tang, G. Positron emission tomography imaging of lesions pH using (11)C-labeled bicarbonate. Cancer Biother. Radiopharm. 33, 285–294 (2018).
Frullano, L., Catana, C., Benner, T., Sherry, A. D. & Caravan, P. Bimodal MR-PET agent for quantitative pH imaging. Angew. Chem. Int. Ed. Engl. 49, 2382–2384 (2010).
Hou, H. et al. Single-cell pH imaging and detection for pH profiling and label-free rapid identification of cancer-cells. Sci. Rep. 7, 1759 (2017).
Han, J. & Burgess, K. Fluorescent indicators for intracellular pH. Chem. Rev. 110, 2709–2728 (2010).
Yin, J., Hu, Y. & Yoon, J. Fluorescent probes and bioimaging: alkali metals, alkaline earth metals and pH. Chem. Soc. Rev. 44, 4619–4644 (2015).
Fang, L. et al. Surface-roughened SERS-active single silver nanowire for simultaneous detection of intracellular and extracellular pHs. ACS Appl. Mater. Interfaces 15, 20677–20685 (2023).
Bi, Y. et al. Reliable quantification of pH variation in live cells using prussian blue-caged surface-enhanced Raman scattering probes. Anal. Chem. 92, 9574–9582 (2020).
Zhao, X. et al. Branched Au nanoparticles on nanofibers for surface-enhanced raman scattering sensing of intracellular pH and extracellular pH gradients. ACS Sens. 5, 2155–2167 (2020).
Wilson, L. T. et al. Mitokyne: a ratiometric Raman probe for mitochondrial pH. Anal. Chem. 93, 12786–12792 (2021).
Zhang, L. et al. Recent progress of SERS nanoprobe for pH detecting and its application in biological imaging. Biosensors 11, https://doi.org/10.3390/bios11080282 (2021).
Modi, S. et al. A DNA nanomachine that maps spatial and temporal pH changes inside living cells. Nat. Nanotechnol. 4, 325–330 (2009).
Medintz, I. L. et al. Quantum-dot/dopamine bioconjugates function as redox coupled assemblies for in vitro and intracellular pH sensing. Nat. Mater. 9, 676–684 (2010).
Liu, A. et al. pHmScarlet is a pH-sensitive red fluorescent protein to monitor exocytosis docking and fusion steps. Nat. Commun. 12, 1413 (2021).
Li, Y. & Tsien, R. W. pHTomato, a red, genetically encoded indicator that enables multiplex interrogation of synaptic activity. Nat. Neurosci. 15, 1047–1053 (2012).
Shen, Y., Rosendale, M., Campbell, R. E. & Perrais, D. pHuji, a pH-sensitive red fluorescent protein for imaging of exo- and endocytosis. J. Cell Biol. 207, 419–432 (2014).
Burgstaller, S. et al. pH-lemon, a fluorescent protein-based pH reporter for acidic compartments. ACS Sens. 4, 883–891 (2019).
Chin, M. Y. et al. Genetically encoded, pH-sensitive mTFP1 biosensor for probing lysosomal pH. ACS Sens. 6, 2168–2180 (2021).
Johnson, D. E. et al. Red fluorescent protein pH biosensor to detect concentrative nucleoside transport. J. Biol. Chem. 284, 20499–20511 (2009).
Tantama, M., Hung, Y. P. & Yellen, G. Imaging intracellular pH in live cells with a genetically encoded red fluorescent protein sensor. J. Am. Chem. Soc. 133, 10034–10037 (2011).
Rennick, J. J., Nowell, C. J., Pouton, C. W. & Johnston, A. P. R. Resolving subcellular pH with a quantitative fluorescent lifetime biosensor. Nat. Commun. 13, 6023 (2022).
Silva, A. P. & NimaláGunaratne, H. Direct visual indication of pH windows:‘off–on–off’fluorescent PET (photoinduced electron transfer) sensors/switches. Chem. Commun. 21, 2399–2400 (1996).
de Silva, S. A. et al. A fluorescent photoinduced electron transfer sensor for cations with an off-on-off proton switch. Tetrahedron Lett. 38, 2237–2240 (1997).
Gunnlaugsson, T., Leonard, J. P., Senechal, K. & Harte, A. J. pH responsive Eu(III)-phenanthroline supramolecular conjugate: novel “off-on-off” luminescent signaling in the physiological pH range. J. Am. Chem. Soc. 125, 12062–12063 (2003).
Diaz-Fernandez, Y. et al. Micelles for the self-assembly of “off-on-off” fluorescent sensors for pH windows. Chemistry 12, 921–930 (2006).
Pallavicini, P., Diaz-Fernandez, Y. A. & Pasotti, L. Smoothly shifting fluorescent windows: a tunable “off-on-off” micellar sensor for pH. Analyst 134, 2147–2152 (2009).
Hirano, T., Noji, Y., Shiraishi, T., Ishigami-Yuasa, M. & Kagechika, H. Development of an ‘OFF-ON-OFF’ fluorescent pH sensor suitable for the study of intracellular pH. Tetrahedron 72, 4925–4930 (2016).
Cidlina, A. et al. OFF-ON-OFF red-emitting fluorescent indicators for a narrow pH window. Chemistry 23, 1795–1804 (2017).
Ogasawara, H., Tanaka, Y., Taki, M. & Yamaguchi, S. Late-stage functionalisation of alkyne-modified phospha-xanthene dyes: lysosomal imaging using an off-on-off type of pH probe. Chem. Sci. 12, 7902–7907 (2021).
Gond, S., Yadav, P., Chauhan, B. S., Srikrishna, S. & Singh, V. P. Development of an ‘OFF-ON-OFF’ colorimetric and fluorometric pH sensor for the study of physiological pH and its bioimaging application. J. Mol. Struct. 1252, 132147 (2022).
Williamson, J. R. Cooperativity in macromolecular assembly. Nat. Chem. Biol. 4, 458–465 (2008).
Mahadevi, A. S. & Sastry, G. N. Cooperativity in noncovalent interactions. Chem. Rev. 116, 2775–2825 (2016).
Whitesides, G. M. & Grzybowski, B. Self-assembly at all scales. Science 295, 2418–2421 (2002).
Dixon, R. E., Navedo, M. F., Binder, M. D. & Santana, L. F. Mechanisms and physiological implications of cooperative gating of clustered ion channels. Physiol. Rev. 102, 1159–1210 (2022).
Rosi, N. L. & Mirkin, C. A. Nanostructures in biodiagnostics. Chem. Rev. 105, 1547–1562 (2005).
Li, Y., Wang, Y., Huang, G. & Gao, J. Cooperativity principles in self-assembled nanomedicine. Chem. Rev. 118, 5359–5391 (2018).
Feng, Q., Wilhelm, J. & Gao, J. Transistor-like ultra-pH-sensitive polymeric nanoparticles. Acc. Chem. Res. 52, 1485–1495 (2019).
Li, Y. et al. Molecular basis of cooperativity in pH-triggered supramolecular self-assembly. Nat. Commun. 7, 13214 (2016).
Zhao, T. et al. A transistor-like pH nanoprobe for tumour detection and image-guided surgery. Nat. Biomed. Eng. 1, https://doi.org/10.1038/s41551-016-0006 (2016).
Zhou, K. J. et al. Tunable, ultrasensitive pH-responsive nanoparticles targeting specific endocytic organelles in living cells. Angew. Chem. Int Ed. 50, 6109–6114 (2011).
Ma, X. P. et al. Ultra-pH-sensitive nanoprobe library with broad pH tunability and fluorescence emissions. J. Am. Chem. Soc. 136, 11085–11092 (2014).
Huang, C. B. et al. Real-Time monitoring the dynamics of coordination-driven self-assembly by fluorescence-resonance energy transfer. J. Am. Chem. Soc. 139, 9459–9462 (2017).
Cheng, D. et al. Selective visualization of the endogenous peroxynitrite in an inflamed mouse model by a mitochondria-targetable two-photon ratiometric fluorescent probe. J. Am. Chem. Soc. 139, 285–292 (2017).
Zhong, Y. et al. Enzyme-instructed self-assembly enabled monomer-excimer transition to construct higher ordered luminescent supramolecular assembly for activity-based bioimaging. Angew. Chem. Int. Ed. Engl. 60, 8121–8129 (2021).
Ren, T. B. et al. A general method to increase stokes shift by introducing alternating vibronic structures. J. Am. Chem. Soc. 140, 7716–7722 (2018).
He, X. et al. Phage-guided targeting, discriminative imaging, and synergistic killing of bacteria by AIE bioconjugates. J. Am. Chem. Soc. 142, 3959–3969 (2020).
Savina, A. et al. NOX2 controls phagosomal pH to regulate antigen processing during crosspresentation by dendritic cells. Cell 126, 205–218 (2006).
Scita, G. & Di Fiore, P. P. The endocytic matrix. Nature 463, 464–473 (2010).
Banushi, B., Joseph, S. R., Lum, B., Lee, J. J. & Simpson, F. Endocytosis in cancer and cancer therapy. Nat. Rev. Cancer 23, 450–473 (2023).
Gilleron, J. & Zeigerer, A. Endosomal trafficking in metabolic homeostasis and diseases. Nat. Rev. Endocrinol. 19, 28–45 (2023).
Gilleron, J. et al. Image-based analysis of lipid nanoparticle-mediated siRNA delivery, intracellular trafficking and endosomal escape. Nat. Biotechnol. 31, 638–646 (2013).
Wittrup, A. et al. Visualizing lipid-formulated siRNA release from endosomes and target gene knockdown. Nat. Biotechnol. 33, 870–876 (2015).
Madigan, V., Zhang, F. & Dahlman, J. E. Drug delivery systems for CRISPR-based genome editors. Nat. Rev. Drug Discov. 22, 875–894 (2023).
Rennick, J. J., Johnston, A. P. R. & Parton, R. G. Key principles and methods for studying the endocytosis of biological and nanoparticle therapeutics. Nat. Nanotechnol. 16, 266–276 (2021).
Casey, J. R., Grinstein, S. & Orlowski, J. Sensors and regulators of intracellular pH. Nat. Rev. Mol. Cell Biol. 11, 50–61 (2010).
Wang, Y. et al. A nanoparticle-based strategy for the imaging of a broad range of tumours by nonlinear amplification of microenvironment signals. Nat. Mater. 13, 204–212 (2014).
Neri, D. & Supuran, C. T. Interfering with pH regulation in tumours as a therapeutic strategy. Nat. Rev. Drug Discov. 10, 767–777 (2011).
Acknowledgements
This work was supported by the National Natural Science Foundation of China (NSFC) grants (82225044 to Y.W.) and the National Key Research and Development Programme of China (2021YFA1201200 to Y.W.). We would like to thank the State Key Laboratory of Natural and Biomimetic Drugs, Peking University Biological Imaging Core Facilities for confocal, animal, and tissue imaging services.
Author information
Authors and Affiliations
Contributions
Y.Y. and Y.W. are responsible for all phases of the research. B.C. assisted with the experimental design. F.W. performed GFP-fused cell lines. C.F., M.C. and M.P. helped with the establishment of tumor models and in vivo imaging. B.M. and R.Z. helped with syntheses of copolymers and dyes. Y.Y. and Y.W. wrote the manuscript. Q.Z. and Y.W. provided the conceptual advice and supervised the study. All authors discussed and commented on the results.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Chunrong Liu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Yang, Y., Chen, B., Wan, F. et al. Cascade-heterogated proton nanotransistors for multiplex pH-interval imaging. Nat Commun 16, 5770 (2025). https://doi.org/10.1038/s41467-025-61255-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-025-61255-6
