Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 18.
Published in final edited form as: ACS Appl Bio Mater. 2019 Sep 27;2(11):4986–4997. doi: 10.1021/acsabm.9b00710

Near-Infrared Hybrid Rhodol Dyes with Spiropyran Switches for Sensitive Ratiometric Sensing of pH Changes in Mitochondria and Drosophila melanogaster First-Instar Larvae

Yibin Zhang †,#, Shuai Xia †,#, Logan Mikesell †, Nick Whisman †, Mingxi Fang †, Tessa E Steenwinkel ‡, Kai Chen §, Rudy L Luck †,*, Thomas Werner ‡,*, Haiying Liu †,*
PMCID: PMC6945768  NIHMSID: NIHMS1061290  PMID: 31912007

Abstract

Near-infrared hybrid rhodol dyes (probes A and B) for sensitive ratiometric visualization of pH changes were prepared by incorporating hemicyanine dyes into traditional rhodol dyes. This approach was based on π-conjugation changes involving a rhodol hydroxyl group as a spiropyran switch upon pH changes. Electronic spectra of probes A-2 and B-2 contain sharp absorption peaks at 535 nm and fluorescence peaks at 558 nm with similar π-conjugation and a closed spiropyran form at a basic pH of 10.2. However, acidic pH conditions break down the hemiaminal ether groups, leading to indolenium moieties and significantly extending the π-conjugation within the rhodol fluorophores, resulting in additional near-infrared emissions for probes A-1 and B-1. As a result, probes A and B exhibit gradual decreases of the absorption peaks at 535 nm and gradual increases in absorption peaks at 609 and 622 nm upon transition from basic to acidic pH, respectively. Both probes display ratiometric fluorescence sensing responses to pH downgrades from 10.2 to 3.6 with visible fluorescence decreases at 558 nm, as well as corresponding increases of the near-infrared fluorescence peaks at 688 and 698 nm, respectively. They exhibit fast, sensitive, and selective fluorescence responses with clearly defined ratiometric features to pH changes and show low cytotoxicity and excellent cell permeability. Our probes were successfully applied to ratiometrically detect pH changes in mitochondria, D. melanogaster first-instar larvae, and to visualize the mitophagy process caused by either cell nutrient starvation or drug treatment.

Keywords: near-infrared fluorescence, ratiometric imaging, pH, rhodol dye, spiropyran switch, mitochondria

Graphical Abstract

graphic file with name nihms-1061290-f0001.jpg

1. INTRODUCTION

Mitochondrial defects or dysfunctions are closely related to some cardiovascular and neurological diseases, such as Alzheimer’s disease.1–4 It is very important to precisely and quantitatively detect mitochondrial pH values because they are closely linked to the unique functions and biochemical processes of mitochondria.5–9 Ideal fluorescent probes for pH sensing in mitochondria should be able to specifically target mitochondria, possess ratiometric sensing capability with two well-defined visible and near-infrared emissions, and effectively track mitophagy. Many rhodol-based probes have been developed,2 and rhodol dyes hold exceptional photo-physical advantages, including high fluorescent quantum yields, excellent photostability, pH sensitivity, and high absorption coefficients.10,11 In order to prevent cellular and tissue damage due to excitation using wavelengths less than 600 nm,12–14 a variety of near-infrared rhodol dyes have been reported, where either the central oxygen atom has been substituted with N, C, or Si atoms or an additional amine has been incorporated for advantageous near-infrared imaging.15–20 However, most of these probes are single-wavelength-based fluorophores without ratiometric sensing features, and they often register systematic errors, such as excitation light fluctuations, heterogeneous samples, probe concentration deviations, and varied localization in organelles in live cells.15–18,21 Spiropyran molecular switches have been widely utilized in the design of optical probes because they respond reversibly to chemical stimuli through chemical structural changes.22–25 Most optical probes are based on absorbance changes involving conversion between the spiropyran and hemicyanine forms due to an external stimulus, and many often lack ratiometric fluorescence responses because of nonfluorescent spiropyran forms.22–27 Effective spiropyran switches with high fluorescence are not yet fully developed in order to ratiometrically detect mitochondrial pH changes.

Here, we present a simple but very effective way to prepare proton-activated ratiometric near-infrared rhodol hydride dyes (probes A-2 and B-2) for sensing pH changes by conjugating hemicyanine dyes into traditional rhodol dyes containing spiropyran molecular switches. The probes allow for ratiometric pH detection through π-conjugated modulation involving rhodol hydroxyl groups in the spiropyran switches under a pH stimulus. Acidic pH readily activates the breakdown of the hemiaminal ether moieties of the probes to produce indolenium moieties, extend rhodol π-conjugation, and lead to new fluorescence peaks with a considerable bathochromic shift to the near-infrared region for probes A-1 and B-1 (Chart 1). Moreover, under basic conditions, the free hydroxyl group of the rhodol fluorophore reacts as a nucleophile toward the indolenium moiety, forming a spiropyran switch, and leads to an emission at 563 nm for the rhodol fluorophores with significantly reduced π-conjugation (Chart 1). Gradual pH increases result in decreases of fluorescence peaks at 688 and 698 nm, as well as corresponding increases in the fluorescence peak at 563 nm for probes A and B, respectively. The probes undergo reversible structural transformations from hemicyanine configurations to spiropyran forms with significantly reduced π-conjugation upon pH increase. They demonstrate sensitive ratiometric sensing of pH variances in mitochondria and D. melanogaster first-instar larvae and can monitor mitochondrial delivery to lysosomes during mitophagy, induced by cell nutrient starvation and drug treatment.

Chart 1.

Chart 1.

Open in a new tab

Probes A and B Undergo Reversible Structural Changes between Closed Spiropyran and Opened Hemicyanine Forms upon pH Changes

2. EXPERIMENTAL SECTION

Instrumentation.

1H NMR and 13C NMR spectra of the fluorescent probes in CDCl3 solution were recorded using a Varian Unity Inova NMR spectrophotometer at 400 and 100 MHz. Absorption spectra were collected with a PerkinElmer Lambda 35 UV/vis spectrometer, while fluorescence spectra were performed on a Jobin Yvon Fluoromax-4 spectrofluorometer.

Reagents.

All solvents and chemical reagents were purchased from Fisher scientific or Sigma-Aldrich. The silica gel (200–300 mesh) from Sigma-Aldrich was used for column chromatographic purification, while silica gel plates from Sigma-Aldrich were employed to conduct thin-layer chromatography (TLC) analyses. The rhodol dye-bearing formyl group (3) was prepared and characterized according to a reported procedure.28

Synthesis of Probe A.

Compounds 3 (208 mg, 0.5 mmol) and 4 (150 mg, 0.5 mmol) were added to dry ethanol (10 mL) and were stirred at room temperature for 16 h (Scheme 1). The mixture was concentrated under reduced pressure, diluted with dichloromethane, washed by water and brine, dried with anhydrous Na2SO4, filtered, and concentrated. The resulting residue was purified by using flash column chromatography gradient elution with methanol to dichloromethane ratios increasing from 2% to 5%, affording probe A-3 (70 mg, 26%). Probe A-3: 1H NMR (300 MHz, CDCl3) δ: 8.04–7.96 (m, 1H), 7.71–7.56 (m, 3H), 7.11–7.27 (m, 2H), 7.11–7.03 (m, 1H), 6.91–6.80 (m, 1H), 6.63–6.48 (m, 1H), 6.41–6.35 (m, 2H), 6.31 (dd, J = 9.0, 2.6 Hz, 2H), 6.28–6.31 (m, 1H), 5.56 (dd, J = 10.2, 6.9 Hz, 1H), 3.33 (q, J = 7.1 Hz, 4H), 2.71 (d, J = 7.2 Hz, 3H), 1.28 (s, 3H), 1.13–1.20 (m, 6H), 1.12 (s, 3H). 13C NMR (75 MHz, CDCl3) δ: 169.77, 156.21, 153.01, 149.67, 148.22, 136.76, 134.87, 129.56, 128.70, 127.74, 126.16, 125.01, 124.27, 121.80, 119.40, 118.56, 115.71, 111.65, 108.49, 107.12, 106.95, 105.28, 104.81, 102.76, 97.90, 52.40, 51.88, 44.73, 29.23, 26.43, 20.42, 12.85. LCMS (ESI): calculated for C37H35N2O4 [M] 570.2, found [M + H] 571.3. HRMS (ESI): 571.25865. Probe A-1: 1HNMR (300 MHz, CDCl3 + CF3COOH) δ: 8.21–8.15 (m, 1H), 7.74–7.60 (m, 3H), 7.60–7.54 (m, 1H), 7.54–7.45 (m, 2H), 7.29 (s, 1H), 7.26–7.19 (m, 2H), 7.13–7.03 (m, 2H), 6.99 (dd, J = 9.8, 2.4 Hz, 1H), 6.91 (d, J = 2.4 Hz, 1H), 6.52 (d, J = 13.0 Hz, 1H), 3.70 (q, J = 7.3 Hz, 4H), 3.42 (d, J = 1.1 Hz, 3H), 1.45 (s, 3H), 1.29–1.42 (m, 6H), 1.24–1.35 (m, 3H). See Figures S1–S4.

Scheme 1.

Scheme 1.

Open in a new tab

The Probe Synthesis of Near-Infrared Rhodol Dyes (Probes A and B) and Their Structural Changes in Trifluoracetic Acid Treatment or under UV Radiation and Heat Treatment

Synthesis of Probe B.

Compounds 3 (208 mg, 0.5 mmol) and 7 (175 mg, 0.5 mmol) were added to dry ethanol (10 mL) and stirred at room temperature for 16 h (Scheme 1). The mixture was concentrated under reduced pressure and diluted with dichloromethane. The product was washed with brine solution, dried with anhydrous Na2SO4, filtered, and concentrated. The resulting residue was purified by using a flash column chromatography gradient elution with methanol to dichloromethane ratios increasing from 2% to 5%, yielding probe B-3 (86 mg, 28%). Probe B-3: 1H NMR (300 MHz, CDCl3) δ: 8.02 (d, J = 7.5 Hz, 1H), 7.94–7.91 (m, 1H), 7.82–7.78 (m, 1H), 7.77–7.73 (m, 1H), 7.67–7.59 (m, 2H), 7.41–7.37 (m, 1H), 7.24–7.18 (m, 2H), 6.97 (dd, J = 8.6, 4.6 Hz, 1H), 6.66 (d, J = 2.9 Hz, 1H), 6.55–6.48 (m, 2H), 6.42 (d, J = 4.1 Hz, 1H), 6.38 (d, J = 2.6 Hz, 1H), 6.33–6.30 (m, 1H), 5.66–5.62 (m, 1H), 3.32 (q, J = 7.1 Hz, 4H), 2.81 (d, J = 5.7 Hz, 3H), 1.63 (d, J = 1.6 Hz, 3H), 1.30 (s, 4H), 1.13 (t, J = 7.0 Hz, 6H). 13CNMR (75 MHz, CDCl3) δ: 169.79, 156.56, 149.68, 146.03, 134.89, 129.67, 128.94, 126.52, 126.44, 125.03, 124.34, 121.73, 118.54, 115.64, 108.50, 106.31, 105.93, 105.26, 102.56, 97.89, 53.99, 44.73, 24.55, 22.03, 12.80. LCMS (ESI): calculated for C41H37N2O4 [M] 620.2, found [M + H] 621.3. HRMS (ESI): 621.27428. Probe B-1: 1H NMR (300 MHz, CDCl3 + CF3COOH) δ: 8.18–8.12 (m, 2H), 8.05 (d, J = 8.1 Hz, 1H), 7.85–7.75 (m, 3H), 7.62–7.56 (m, 2H), 7.47 (d, J = 7.5 Hz, 1H), 7.34–7.28 (m, 2H), 7.14 (d, J = 7.5 Hz, 1H), 7.03 (d, J = 9.8 Hz, 1H), 6.98–6.93 (m, 2H), 6.89 (d, J = 2.2 Hz, 1H), 6.61 (d, J =13.0 Hz, 1H), 3.68 (d, J = 7.5 Hz, 4H), 3.53 (s, 3H), 1.63 (s, 3H),1.36 (s, 3H), 1.36–1.23 (m, 6H). See Figures S5–S8.

Cell Culture and Cytotoxicity Assay.

HeLa cells were nurtured in DMEM (modified Eagle’s medium, Gibco) supplemented with 10% FBS (fetal bovine serum, Fisher Scientific) at 3 °C with 5% CO2 humidified air and then subcultured every 2–3 days until an 80% confluence was attained. A cytotoxicity assay was carried out using a standard MTT assay against HeLa cells. The HeLa cells were seeded at a density of about 6000 cells per well in a 96-well plate and grown for 24 h, followed by adding fresh culture medium containing probe A or B with different concentrations from 0, 5, 10, 20, 50, to 75 μM for further incubation of 48 h. The cells were then incubated for 4 h with 500 μg/mL of tetrazolium salt (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide), which would be reduced to water-insoluble purple formazan by metabolically active cells. Dark purple crystals were dissolved with dimethylsulfoxide (DMSO), and the cell viability rate was determined by measuring the absorbance at 490 nm. The cell viability rate was calculated by Vrate = (A − AB)/(AC − AB) × 100%, where A is the absorbance of the experimental group, AC the absorbance of the control group (cell medium was used as control), and AB the absorbance of the blank group (no cells).

Cellular Fluorescence Imaging.

HeLa cells were seeded in 35 mm confocal glass bottom dishes (MatTek) with 1 × 105 cells per dish and cultured for 24 h before imaging. HeLa cells were incubated with 5 μM of probe A or B and either 10 μM Mito Tracker blue or 10 μM Lysosensor blue in normal medium in the presence of 1% DMSO containing fetal bovine serum or the medium containing 1% DMSO without fetal bovine serum for 30 min and 2 h, followed by washing the cells twice with PBS buffer before imaging. HeLa cells were also incubated with probe A or B with 5 μM cyanine dye (IR-780) in normal cell culture medium for 30 min and washed twice with PBS buffer prior to imaging. For visualization of intracellular pH changes, HeLa cells were incubated in different pH citric buffers, containing nigericin (5 μg/mL) for 30 min to equilibrate the intracellular and extracellular pH.29–34 The cells were then incubated with 10 μM probe A or B for 15 min and then rinsed with FBS buffer twice before imaging. A confocal fluorescence microscope (Olympus IX 81, Olympus America Inc.) was used to obtain cell images. The cells were imaged with a 100× objective lens for the colocalization imaging experiments and with a 60× objective lens for the other imaging experiments. The fluorescence of LysoSensor Blue or Mito Tracker Blue (blue channel) was measured from 425 to 475 nm under 405 nm excitation; the visible fluorescence (green channel) and near-infrared fluorescence (red channel) of the probes under 488 nm excitation were recorded from 525 to 575 nm and from 650 to 750 nm, respectively. The near-infrared fluorescence of the probes (magenta channel) under excitation of 559 nm was recorded at 650–750 nm. The images were further processed with an Olympus FV10-ASW 3.1 viewer using Image Pro6 software.

In Vivo Experiments with D. melanogaster First-Instar Larvae.

A nine-well glass viewing dish was used to conduct fluorescence imaging of D. melanogaster first-instar larvae with probe A. The larvae were divided into four groups with ten freshly hatched first-instar larvae. The larvae in the first group were submerged in 500 μL of PBS for 2 h for the blank control. The larvae in the second, third, and fourth groups were incubated with 10 μM probe A for 2 h in 500 μL of buffers with pH values 9.5, 7.4, and 5.0, respectively. Following incubation with probe A in the different buffer solutions, the larvae were washed three times with PBS buffer and transferred with water onto microscope slides. Confocal fluorescence images were taken with an Olympus IX 81 microscope before the water dried on the slides. The fluorescence imaging protocol used for the D. melanogaster larvae was identical to how we imaged the cells.

3. RESULTS AND DISCUSSION

Construction of Ratiometric Near-Infrared Hydride Rhodol Dyes for pH.

A formyl-functionalized rhodol dye (3) was reacted with 1,2,3,3-tetramethyl-3H-indolium iodide (4) and 1,2,3,3-tetramethyl-3H-benzo(e)indolium iodide (5) in ethanol at room temperature to form probes A-3 and B-3, respectively, in order to introduce spiropyran switches into traditional rhodol dyes (Scheme 1). The rhodol derivative (3) was prepared by condensing 2-(4-diethylamino-2-hydroxybenzoyl)benzoic acid (1) with 4-(diethylamino)-salicylaldehyde (2) in sulfuric acid at 100 °C for 2 h.28 The probes A-2 and B-2 under neutral and basic pH in aqueous solutions are composed of a rhodol fluorophore and N,O-disubstituted hemiaminal ether residue linked through a carbon–carbon double-bonded connection (Chart 1). The incorporation of a hydroxyl group allows for variance in the π-conjugation of the rhodol dyes upon pH changes, which achieves sensitive ratiometric determination of pH changes (Chart 1). Under neutral and basic conditions, the free hydroxyl group acting as a nucleophile attacks the indolenium moiety, forming an N,O-disubstituted hemiaminal ether group, and reduces π-conjugation. The rhodol fluorophore with a closed spiropyran configuration emits in the visible region (Chart 1). Under acidic pH conditions, breakdown of the hemiaminal ether residues occurs, generating indolenium moieties and considerably enhancing the π-conjugation in the rhodol fluorophores, resulting in a near-infrared emission with a bathochromic shift (Scheme 1).

Optical Study of Probes in Organic Solvents.

Probes A-3 and B-3 are colorless in common organic solvents, such as tetrahydrofuran, dichloromethane, dimethylformamide, and DMSO, as the probes have closed forms of both spiropyran and spirolactone rings. The addition of trifluoroacetic acid to probes A-3 and B-3 in DMSO solution results in ring opening of both spiropyran and spirolactone configurations, leading to new absorption peaks at 546 and 542 nm (Figures S55 and S57), two fluorescence peaks at 575 and 706 nm for probe A-1 and 585 and 712 nm for probe B-1 (Figures S56 and S58) (Scheme 1). Probe B-1 with more extended π-conjugation has both absorption and emission peaks occurring at longer wavelengths than probe A-1. All intermediates and probes were characterized by NMR and mass spectrometry (see Supporting Information).

Probes A-3 and B-3 are colorless without any significant absorption and emission peaks in DMSO solution (Figures S75 and S80). Under UV radiation of 254 nm, a new broad absorption peak at 460 nm (Figure S75) and a new fluorescence peak at 523 nm appear (Figure S76), increase with time, and become maximized within 90 min for probe A-3 (Figures S76–S78) because UV radiation transforms the closed spiropyran ring of the probe into the hemicyanine configuration as probe A-4 (Scheme 1). In addition, the probe with the opened hemicyanine configuration, i.e., A-4, can be reversibly converted into a probe with a closed spiropyran ring structure, i.e., A-3, if heated at 80 °C (Scheme 1) (Figures S78–S79). Similar photolysis and heat effects to probe A-3 were observed for probe B-3 (Figures S80–S84), which shows longer absorption and emission peaks at 507 and 530 nm under UV radiation, respectively (Figures S83–S84), due to a more extended benzo(e)indolium π-conjugation in the opened hemicyanine configuration (Scheme 1). Probe A-3 is almost colorless and shows extremely weak fluorescence at 582 nm in DMSO solution with a trace amount of water (Figure S85). However, addition of 10% water to DMSO solution of probe A-3 causes a significant fluorescence increase at 582 nm, indicating that water facilitates ring opening of the spironolactone configuration of probe A-3 and leads to π-conjugation enhancement of probe A-2. These structural changes triggered by water were further confirmed by 1H NMR spectra of probe A-3 and A-2 in DMSO-d6 solution in the absence and presence of D2O solution (Figures S86–S87).

Probe Optical Responses to pH Changes.

In aqueous solutions, spirolactone rings of the probes open as the bridging O atom is easily protonated (Chart 1). Upon excitation at 480 nm in pH 10.2 buffer, probes A and B produce similar sharp absorption peaks at 535 nm and fluoresce at the same wavelength of 558 nm, which indicates that both probes possess the same π-conjugation systems with closed spiropyran forms, i.e., probes A-2 and B-2, at pH 10.2. A gradual decrease in the pH leads to gradual decreases of the absorption peaks at 535 nm and gradual increases in new absorption peaks at 609 and 622 nm (Figures 1 and 2) for probes A-1 and B-1, respectively (Chart 1), because acid converts the spiropyran rings into hemicyanine structures. Probe B-1 displays a longer absorption wavelength at 622 nm due to better benzo(e)-indolium π-conjugation with the opened hemicyanine configuration than probe A-1 (Figure 2). Gradual pH decreases in solutions containing either probe A or B result in significant decreases of the fluorescence peaks at 558 nm and in considerable increases of a new near-infrared fluorescence peak at 688 or 698 nm under 480 nm excitation. This is because acidic pH activates breakdown of the N,O-disubstituted hemiaminal ether residues, leading to the opening of the spiropyran rings and converting the closed-ring structures into a hemicyanine configuration with significantly enhanced π-conjugation (Chart 1). Under excitation at 480 nm, probe A-1 has fluorescence quantum yields of 38.2% and 13.6%, corresponding to the emission peak at 558 nm at pH 10.2 and the near-infrared fluorescence peak at 688 nm at pH 3.6, respectively (see Section 3 of the Supporting Information). Probe B displays similar ratiometric fluorescence sensing responses to pH decreases from 10.2 to 3.6 with decreases of visible fluorescence at 558 nm, as well as fluorescence increases of the near-infrared fluorescence peak at 698 nm (Figure 2). Probe B-1 has a similar fluorescence peak at 558 nm with the closed spiropyran configuration at pH 10.2 to that for probe A-1 but displays a slightly longer near-infrared fluorescence at 698 nm with the opened hemicyanine configuration, i.e., B-2, at pH 3.6. Probe B has a fluorescence quantum yield of 34.6% for the fluorescence peak at 558 nm in pH 10.2 buffer and exhibits a fluorescence quantum yield of 10.1% in pH 3.6 buffer for the peak at 698 nm under 480 nm excitation. Near-infrared fluorescence of both probes increases with pH decreases under 600 nm excitation (Figure 3). Additionally, probes A and B reversibly respond to pH variations related to opening of the probe spiropyran rings with average pKa values of 8.26 and 7.10, respectively (Figures S49–S54). The lower pKa value of probe B is due to the delocalization of a pair of unshared electrons from the nitrogen atom in the spiropyran ring into the naphthalene moiety, which requires a stronger acidic condition to protonate this nitrogen atom and open the ring.

Figure 1.

Figure 1.

Open in a new tab

Absorption (left) and emission (right) spectra of 5 μM of probe A in different pH buffers containing 30% ethanol under 480 nm excitation.

Figure 2.

Figure 2.

Open in a new tab

Absorption (left) and emission (right) spectra of 5 μM of probe B in different pH buffers containing 30% ethanol under 480 nm excitation.

Figure 3.

Figure 3.

Open in a new tab

Fluorescence spectra of 5 μM of probes A (left) and B (right) in different pH buffers containing 30% ethanol under 600 nm excitation.

Computational Study of the Probes.

The aforementioned electronic transitions were investigated via theoretical calculations, where the structures of each molecule, i.e., A-1–4 and B-1–4, depicted in Chart 1 and Scheme 1 were optimized (see Supporting Information Section 2 which includes procedural details). Absorptions were calculated in the UV range at 310 and 330 nm for A-3 and B-3, respectively (Figures S20 and S39), whereas those for A-4 and B-4 were calculated to be at 450 and 473 nm (Figures S27 and S46). The calculations were also useful in tentatively assigning the 1H NMR spectra for A-3 (Figure S21) and B-3 (Figure S40) and in assigning the resonance for the carbon atom in the lactone ring, as indicated in Figures S23 and S42, respectively. Of primary interest was the nature of the absorptions under the different pH environments. We found that the calculated absorptions were much lower than those measured experimentally, as listed in Table S25 for A-1–2 and B-1–2. These calculated values are outside of the expected range35 (i.e., <0.25 eV), which may be ascribed to a underestimation of transition energies with the seven different functionals utilized.36 However, the conformational and orbital information is useful because the trend in the absorptions is maintained. Structural optimizations reveal a major conformational change in the molecules, ranging from almost planar forms in A-1 (Figure S9) and B-1 (Figure S28) to bent forms in A-2 (Figure S15) and B-2 (Figure S34), with respect to the xanthene and hemicyanine sections. While the benzoic acid moieties in the rhodol section of the molecule are almost perpendicular to the xanthene plane at A-1 (75°), A-2 (83°), B-1 (65°), and B-2 (82°), the hemicyanine moieties in A-1 and B-1 are attached to the rhodol section via C to C atom single bonds, which have torsion angles of −19° and −30°, respectively. In A-2 and B-2, this linkage is in the form of the spiropyran ring, which results in angles between the xanthene and hemicyainine planes of 79° and 83°, respectively (Figure 4).

Figure 4.

Figure 4.

Open in a new tab

Current density difference illustrations as iso-surfaces of probes A-1 (top left), A-2 (top right), B-1 (bottom left), and B-2 (bottom right). Red areas represent values for the various densities of −1.500e−5, and blue are for 1.500e−5; see scale on top of illustration.

These different geometries result in the observable transitions in A-1 and B-1 consisting of HOMO–1 → LUMO orbitals, whereas those for A-2 and B-2 consist of HOMO–1 → LUMO orbitals (Figures S47 and S48). The drawings in Figure 2 illustrate that the HOMOs in probes A-1 and B-1 involve π-orbitals from the xanthene and hemicyanine sections of both molecules, judging from the movement of electron density from the red sections of the molecule to the blue. Notably, in A-1, the charge transfer occurs from the left end of the xanthene ring (i.e., the part involving the diethylammonium moiety) to the right. In B-1, the origin is at the right end, i.e., the hemicyanine section. In A-2 and B-2, the charge transfer is equivalent (HOMO–1 → LUMO) and involves primarily π-orbitals on the xanthene section of the molecule (Figure 4), probably because of the conformation noted previously. Here, the HOMOs consists of orbitals localized on the hemicyanine section of the molecules (see Figure S47 for A-2 and Figure S48 for B-2).

Probe-Selective and -Reversible Responses to pH, Photostability, and Cytotoxicity.

Probe cytotoxicity and selectivity in the presence of potential interfering biological molecules and ions were investigated. We measured fluorescence spectra of probe A or B in the presence of 100 μM metal ions (Al3+, Ca2+, Co2+ Cu2+, Fe3+, K+, Mg2+, Mn2+, Ni2+, and Zn2+), 100 μM anions (CN−, S2O32−, HSO3−, SO32−, ClO−, HCO3−, CO32−, S2−, NO3−, PO43−, SO42−, and Cl−), 100 μM various amino acids, and biothiols (100 μM cysteine, 100 μM homocysteine, and 10 mM GSH) in pH 3.8, 7.5, or 9.9 buffers (Figures S59–S70). These ions and molecules do not significantly change the fluorescence ratios, indicating that the probes can detect the intracellular pH without interference caused by intracellular biological molecules and ions. Furthermore, the probes show selective responses to pH values in the absence and presence of reactive oxygen and nitrogen species, such as 50 μM nitric oxide (NO), peroxynitrite (ONOO−), sodium peroxide, and 100 μM hydrogen peroxide (Figures S71–S72). The probes also show good photostability throughout 2 h of excitation at 480 and 550 nm (Figures S88–S91). In addition, probes A and B exhibit reversible florescence responses to pH changes between 4.0 and 10.0 (Figures S73–S74). The cytotoxicity of the probes was investigated with an MTT cell viability assay. The results in Figure S92 show that the probes are nontoxic and allow high cell-proliferation rates even under high probe concentrations.

Cellular Fluorescence Imaging in Mitochondria.

Being positively charged, the probes should be able to selectively target mitochondria with negative electric potentials across their inner membranes through electrostatic interactions. To test this hypothesis, we conducted intracellular localization tests of the probes via colocalization experiments with organelle trackers, such as Lysosensor blue, mitochondria-targeting cyanine dye (IR-780), and Mito Tracker blue (a cationic fluorescent probe that accumulates specifically in mitochondria), using HeLa cells in normal culture medium. As expected, cellular fluorescent intensity increases with the probe concentration increases in probes A and B (Figures 5 and S93). The observed high Pearson correlation coefficients (at more than 0.95) of probes A and B with cyanine IR-780 dye confirm that the probes selectively stain mitochondria in live cells. The Pearson correlation coefficients of probes A and B (channel I) with Mito Tracker blue in HeLa cells during 30 min of incubation in normal medium containing FBS are high at 0.95 and 0.97, respectively (top of Figure 6, Figures S95, S96 (probe A) and S111, S112 (probe B)). In contrast, lower Pearson correlation coefficients of 0.67 and 0.35 are observed for probes A and B (channel I) with Lysosensor blue in HeLa cells in 30 min cell incubation in normal medium containing FBS, respectively (Bottom of Figure 6, Figures S97, S98 (probe A) and S113, S114 (probe B)). Upon an increase of the incubation time to 2 h, the average Pearson correlation coefficients of probes A and B (channel I) with Mito Tracker blue are 0.95 and 0.94, respectively (Figures S99, S100 (probe A) and S115, S116 (probe B)), while the average Pearson correlation coefficients of probes A and B (channel I) with Lysosensor blue increase to 0.69 and 0.51, respectively (Figures S101, S102 (probe A) and S117, S118 (probe B)). These high Pearson correlation coefficients of probes A and B with Mito Tracker blue convincingly demonstrate that the probes stay in mitochondria in live cells in normal medium.

Figure 5.

Figure 5.

Open in a new tab

Fluorescence imaging of HeLa cells with probe A under excitation at 488 and 559 nm, 5 μM of cyanine dye (IR-780) under excitation at 630 nm in normal medium in the presence of FBS with 30 min of incubation. Scale bar: 20 μm.

Figure 6.

Figure 6.

Open in a new tab

Fluorescence imaging of HeLa cells with 5 μM of probe A, 10 μM Mito Tracker blue (upper), or 10 μM Lysosensor blue (bottom) under excitation at 405 nm in normal medium in the presence of FBS at 30 min incubation. Scale bar: 20 μm.

In order to further confirm that the probes selectively accumulate in mitochondria in live cells, we further treated HeLa cells with an uncoupler of mitochondrial oxidative phosphorylation, i.e., carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), to disrupt the mitochondrial membrane potential, resulting in acidification of mitochondria.5 HeLa cells were incubated with either probe A or B for 30 min in a normal cell culture medium and then subjected to an additional 20 min incubation with FCCP dissolved in PBS buffer. We observed a visible fluorescence decrease and near-infrared fluorescence increase with dramatic color changes in the ratio images of the near-infrared channel II divided by the visible channel I from blue to green for probe A (Figures 7 and S94). These results indicate that FCCP treatment leads to apparent acidification of mitochondria because of the disruption of the mitochondrial membrane potential caused by FCCP treatment, further confirming that the probes stay in mitochondria before the FCCP treatment.

Figure 7.

Figure 7.

Open in a new tab

Fluorescence imaging of HeLa cells with 3.0 μM of probe A in normal medium with 20 min incubation under excitation at 488 and 559 nm as control experiment after FCCP treatment. The FCCP treatment was further conducted as follows: Cells were incubated with probe A for 20 min in normal medium and further incubated with 20 μM of FCCP for 20 additional minutes. The cells were washed before the confocal imaging was conducted. Scale bar: 50 μm.

Mitophagy Caused by Cell Nutrient Starvation.

Starvation is well-known to induce mitophagy,37 and we then determined if the probes can illuminate the mitophagy process that occurs from nutrient starvation. During mitophagy, lysosomes infuse into defective mitochondria, producing acidic autolysosomes, resulting in a decrease in the pH value inside mitochondria.37 Probe A was incubated with HeLa cells in serum-free medium at different incubation times. Our cellular imaging results during nutrient starvation show that incubation times from 10 min to 3 h cause a gradual decrease in the visible fluorescence in the green channel and a gradual increase of the near-infrared fluorescence in the red channel under 488 nm excitation (Figure 8). A significant color change in the merged images of visible and near-infrared fluorescence from deep green to brown color is also observed (Figure 8), indicating that the probe fluorescence changes in the visible and near-infrared channels are related to mitophagy induced by nutrient starvation. Probes A and B remain in mitochondria within 30 min of incubation of HeLa cells in serum-free medium. This is evident as probes A and B (channel I) with Mito tracker blue in HeLa cells show high Pearson correlation coefficients of 0.92 and 0.91, respectively (Figure 9 upper, S103, S104 (probe A) and S119, S120 (probe B)), while they have lower Pearson correlation coefficients of 0.44 and 0.46 with Lysosensor blue, respectively (Figure 9, lower, S105, S106 (probe A) and S121 and S122 (probe B)). During a 2 h incubation in serum-free medium, the fluorescence images of probes A and B (channel I) with Lysosensor blue in HeLa cells show high Pearson correlation coefficients of 0.93 and 0.93, respectively (bottom of Figure 10, Figures S109, S110 (probe A) and S125, S126 (probe B)). In contrast, fluorescence images of probes A and B (channel I) with Mito Tracker blue have lower Pearson correlation coefficients of 0.49 (top of Figure 10, Figures S107, S108 for probe A) and 0.58 (Figures S123, S124 for probe B) under 2 h incubation during cell starvation. These results indicate that the probes were localized to acidic autolysosomes during mitophagy caused by cell starvation after the 2 h incubation period. These results demonstrate that the probes can be used to track the mitophagy process induced by nutrient starvation (Figures 8–10 and S103–S109, S119–S126).

Figure 8.

Figure 8.

Open in a new tab

Fluorescence imaging of HeLa cells incubated with 10 μM of probe A in serum-free medium under excitation at 488 and 559 nm with a scale bar of 50 μm.

Figure 9.

Figure 9.

Open in a new tab

Fluorescence imaging of HeLa cells with 5 μM of probe A, 10 μM Mito Tracker blue (top, Ex 405 nm), and 10 μM Lysosensor blue (bottom, Ex 405 nm) in serum-free medium for 30 min incubation. Scale bar: 20 μm.

Figure 10.

Figure 10.

Open in a new tab

Fluorescence imaging of HeLa cells with 5 μM of probe A, 10 μM Mito Tracker blue (top, Ex 405 nm), or 10 μM Lysosensor blue (bottom, Ex 405 nm) in serum-free medium for 2 h incubation. Scale bar: 20 μm.

We also used rapamycin to induce mitophagy in normal culture medium and determined if probes A and B can be used to track mitophagy induced by this drug.5 Fluorescence images display significant increases in the cellular near-infrared fluorescence and decreases in the cellular visible fluorescence with substantial color changes from blue to green. These results indicate that probe A is located within the autolysosomes formed by fusing damaged mitochondria with lysosomes during mitophagy, due to the rapamycin treatment (Figures 11 and S127). As a consequence of this localization, the probe effectively detects mitochondrial acidification during mitophagy caused by either nutrient starvation (Figures S8–10, S103–S109, and S119–S126) or drug treatment with rapamycin (Figures 11 and S127).

Figure 11.

Figure 11.

Open in a new tab

Fluorescence images of HeLa cells incubated with 3.5 μM of probe A and rapamycin (100 nM) in normal culture medium after a 1 h incubation time with probe A under excitation of 488 and 559 nm with scale bars of 50 μm. Images from top to bottom: probe A only (top), probe A and rapamycin (middle), cells incubated with rapamycin for 24 h and then further incubated with probe A for 1 h (bottom).

Visualization of Intracellular pH Changes.

We further tested if the probes can detect intracellular pH changes ratiometrically in mitochondria by incubating HeLa cells with probe A or B, using buffers with different pH values in the presence of 5 μM K+/H+ ionophore nigericin, which is used to achieve an equilibration between the pH of the external buffer and the intracellular pH value.29–34 A gradual decrease of the intracellular pH from 10.0 to 3.0 leads to a visible fluorescence decrease and a corresponding near-infrared fluorescence increase at 488 nm excitation for probe A (Figures 12 and S128–S132). The merged cellular images of the visible and near-infrared fluorescence of probe A in the third column undergo significant changes from a deep green to brown and then finally a deep red upon cellular pH changes from 10.0 to 3.0. Similar fluorescence ratiometric responses of probe B under these conditions were observed (Figures S131 and S132). We further conducted colocalization experiments of probes A and B with Mito Tracker blue or Lysosensor blue with intracellular pH 5.0 and 6.0 (Figures 13 and S133–S139). The Pearson correlation coefficients of probe B (channels I, II, and III) with Mito Tracker blue (channel IV) in HeLa cells are high at 0.97, 0.96, and 0.98 after a 2 h incubation of probe B and Mito Tracker blue with HeLa cells in pH 6.0 normal medium, containing FBS and 5 μM nigericin, respectively (Figure 13). This is clearly depicted in the overlapped cellular images and the scatter plots, indicating that probe B stays in mitochondria in live cells after a 2 h probe incubation with HeLa cells in pH 6.0 normal medium containing 5 μM nigericin. Similar high Pearson correlation coefficients of probe A with Mito Tracker blue were also obtained under pH 5.0 or 6.0 (Figures S133 and S135). These results demonstrate that probes A and B can effectively detect pH changes in mitochondria ratiometrically (Figures 12, 13, and S129–S139) and that they can be utilized for the visualization of mitophagy processes induced by cell starvation (Figures 8–10, S103–S109, and S119–S129) and drug treatment (Figures 11 and S127).

Figure 12.

Figure 12.

Open in a new tab

Fluorescence images of HeLa cells incubated with 10 μM of probe A in different pH buffers containing 5 μM nigericin. Scale bar: 50 μm.

Figure 13.

Figure 13.

Open in a new tab

Fluorescence images of HeLa cells incubated with 5 μM of probe B in pH 6.0 buffer containing 5 μM nigericin and 10 μM Mito Tracker blue (Ex 405 nm). Scale bar: 10 μm. Pearson correlation coefficient between channel I and channel IV: 0.97. Pearson correlation coefficient between channel II and channel IV: 0.96. Pearson correlation coefficient between channel III and channel IV: 0.98.

Reactive oxygen species (ROS) from mitochondria inside most mammalian cells are produced due to mitochondrial damage in a variety of pathologies and play a very important role in redox signaling from mitochondria to the rest of the cell. We also used probe A to detect pH fluctuations in live cells under oxidative stress. Hydrogen peroxide treatment of HeLa cells results in a cellular fluorescence decrease in the near-infrared channel and a fluorescence increase in the visible channel over time, indicating that pH increases over time in mitochondria due to the H2O2 treatment (Figure S140), which is consistent with what was reported in the literature.9

Additionally, we conducted photobleaching studies in HeLa cells by confocal fluorescence microscopy. Cellular visible and near-infrared fluorescence intensities decrease slightly during a 240 s photobleaching experiment, as detected by confocal fluorescence microscopy (Figures S141–S144), indicating that the probes show photostability in live cells, and this confirms our in vitro results (Figures S88–S91).

Fluorescence Imaging of D. melanogaster First-Instar Larvae.

For the first time, we demonstrated the feasibility of imaging pH changes in the model organism D. melanogaster. We incubated first-instar larvae of D. melanogaster with probe A in buffer solutions of three different pH values for 2 h, after which we conducted fluorescent confocal microscopy (Figure 14). The larvae incubated at pH 9.5 containing probe A showed strong visible fluorescence in the green channel and extremely weak near-infrared fluorescence in channels II and III under excitation at 488 and 559 nm. A decrease of the pH to 7.4 and then to 5.0 led to a strong decrease of the visible fluorescence of the gut tissue in channel I and a considerable near-infrared fluorescence increase in channels II and III under the same excitation conditions (Figure 14), which is consistent with the fluorescence responses of probe A to pH changes in buffer solutions (Figure 1).

Figure 14.

Figure 14.

Open in a new tab

Fluorescence images of D. melanogaster first-instar larvae incubated with 10 μM probe A in different pH buffers for 2 h. Scale bar: 200 μm.

4. CONCLUSION

Near-infrared ratiometric rhodol-based fluorescent probes have been developed by incorporating molecular spiropyran switches into traditional rhodol dyes for ratiometric sensing of pH changes in mitochondria and visualization of mitophagy processes induced by cell nutrient starvation and drug treatment. Probe A was also successfully used to image pH changes in D. melanogaster first-instar larvae for the first time based on a literature survey. The probes display quick fluorescence sensing responses to pH changes in near-infrared and visible channels and possess good selectivity, high sensitivity, low cytotoxicity, excellent cell permeability, and very useful ratiometric features to overcome systematic errors of intensity-based probes.

Supplementary Material

Supplemental informaton

ACKNOWLEDGMENTS

This research work was funded by the National Institute of General Medical Sciences of the National Institutes of Health under Award Numbers R15GM114751 and 2R15GM114751-02 to H.Y. Liu. A high-performance computing infrastructure at Michigan Technological University was used for the calculations.

Footnotes

Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsabm.9b00710.

Synthetic procedures, 1H and 13C NMR spectra of intermediates, probes A and B, high-resolution ESI-MS spectra of probes A and B, theoretical results, pKa determination, probe selectivity test, reversibility test, photostability test, and near-infrared fluorescence probes A and B in HeLa cells in buffers with different pH values (PDF)

The authors declare no competing financial interest.

REFERENCES

  • (1).Zielonka J; Joseph J; Sikora A; Hardy M; Ouari O; Vasquez-Vivar J; Cheng G; Lopez M; Kalyanaraman B Mitochondria-Targeted Triphenylphosphonium-Based Compounds: Syntheses, Mechanisms of Action, and Therapeutic and Diagnostic Applications. Chem. Rev 2017, 117 (15), 10043–10120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (2).Lin MT; Beal MF Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature 2006, 443 (7113), 787–795. [DOI] [PubMed] [Google Scholar]
  • (3).Vasquez-Trincado C; Garcia-Carvajal I; Pennanen C; Parra V; Hill JA; Rothermel BA; Lavandero S Mitochondrial dynamics, mitophagy and cardiovascular disease. J. Physiol 2016, 594 (3), 509–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (4).Cadonic C; Sabbir MG; Albensi BC Mechanisms of Mitochondrial Dysfunction in Alzheimer’s Disease. Mol. Neurobiol 2016, 53 (9), 6078–6090. [DOI] [PubMed] [Google Scholar]
  • (5).Liu Y; Zhou J; Wang LL; Hu XX; Liu XJ; Liu MR; Cao ZH; Shangguan DH; Tan WH A Cyanine Dye to Probe Mitophagy: Simultaneous Detection of Mitochondria and Autolysosomes in Live Cells. J. Am. Chem. Soc 2016, 138 (38), 12368–12374. [DOI] [PubMed] [Google Scholar]
  • (6).Xu W; Zeng ZB; Jiang JH; Chang YT; Yuan L Discerning the Chemistry in Individual Organelles with Small-Molecule Fluorescent Probes. Angew. Chem., Int. Ed 2016, 55 (44), 13658–13699. [DOI] [PubMed] [Google Scholar]
  • (7).Chen YC; Zhu CC; Cen JJ; Bai Y; He WJ; Guo ZJ Ratiometric detection of pH fluctuation in mitochondria with a new fluorescein/cyanine hybrid sensor. Chem. Sci 2015, 6 (5), 3187–3194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (8).Sarkar AR; Heo CH; Xu L; Lee HW; Si HY; Byun JW; Kim HM A ratiometric two-photon probe for quantitative imaging of mitochondrial pH values. Chem. Sci 2016, 7 (1), 766–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (9).Cao LX; Zhao ZS; Zhang T; Guo XD; Wang SQ; Li SY; Li Y; Yang GQ In vivo observation of the pH alternation in mitochondria for various external stimuli. Chem. Commun 2015, 51(97), 17324–17327. [DOI] [PubMed] [Google Scholar]
  • (10).Lee MH; Sessler JL; Kim JS Disulfide-Based Multifunctional Conjugates for Targeted Theranostic Drug Delivery. Acc. Chem. Res 2015, 48 (11), 2935–2946. [DOI] [PubMed] [Google Scholar]
  • (11).Han JY; Burgess K Fluorescent Indicators for Intracellular pH. Chem. Rev 2010, 110 (5), 2709–2728. [DOI] [PubMed] [Google Scholar]
  • (12).Fu W; Yan CX; Guo ZQ; Zhang JJ; Zhang HY; Tian H; Zhu WH Rational Design of Near-Infrared Aggregation-Induced-Emission Active Probes: In Situ Mapping of Annyloid-beta Plaques with Ultrasensitivity and High-Fidelity. J. Am. Chem. Soc 2019, 141 (7), 3171–3177. [DOI] [PubMed] [Google Scholar]
  • (13).Yan CX; Guo ZQ; Shen YY; Chen Y; Tian H; Zhu WH Molecularly precise self-assembly of theranostic nanoprobes within a single-molecular framework for in vivo tracking of tumor-specific chemotherapy. Chem. Sci 2018, 9 (22), 4959–4969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (14).Yan CX; Guo ZQ; Liu YJ; Shi P; Tian H; Zhu WH A sequence-activated AND logic dual-channel fluorescent probe for tracking programmable drug release. Chem. Sci 2018, 9 (29), 6176–6182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (15).Roth A; Li H; Anorma C; Chan J A Reaction-Based Fluorescent Probe for Imaging of Formaldehyde in Living Cells. J. Am. Chem. Soc 2015, 137 (34), 10890–10893. [DOI] [PubMed] [Google Scholar]
  • (16).Chen W; Xu S; Day JJ; Wang DF; Xian M A General Strategy for Development of Near-Infrared Fluorescent Probes for Bioimaging. Angew. Chem., Int. Ed 2017, 56 (52), 16611–16615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (17).Zhang SW; Adhikari R; Fang MX; Dorh N; Li C; Jaishi M; Zhang JT; Tiwari A; Pati R; Luo FT; Liu HY Near-Infrared Fluorescent Probes with Large Stokes Shifts for Sensing Zn(II) Ions in Living Cells. ACS Sens 2016, 1 (12), 1408–1415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (18).Grzybowski M; Taki M; Yamaguchi S Selective Conversion of P = O-Bridged Rhodamines into P = O-Rhodols: Solvatochromic Near-Infrared Fluorophores. Chem. - Eur. J 2017, 23 (53), 13028–13032. [DOI] [PubMed] [Google Scholar]
  • (19).Yuan L; Lin WY; Zhao S; Gao WS; Chen B; He LW; Zhu SS A Unique Approach to Development of Near-Infrared Fluorescent Sensors for in Vivo Imaging. J. Am. Chem. Soc 2012, 134 (32), 13510–13523. [DOI] [PubMed] [Google Scholar]
  • (20).Ren TB; Xu W; Jin FP; Cheng D; Zhang LL; Yuan L; Zhang XB Rational Engineering of Bioinspired Anthocyanidin Fluorophores with Excellent Two-Photon Properties for Sensing and Imaging. Anal. Chem 2017, 89 (21), 11427–11434. [DOI] [PubMed] [Google Scholar]
  • (21).Li XH; Gao XH; Shi W; Ma HM Design Strategies for Water-Soluble Small Molecular Chromogenic and Fluorogenic Probes. Chem. Rev 2014, 114 (1), 590–659. [DOI] [PubMed] [Google Scholar]
  • (22).Shao N; Jin JY; Wang H; Zheng J; Yang RH; Chan WH; Abliz Z Design of Bis-spiropyran Ligands as Dipolar Molecule Receptors and Application to in Vivo Glutathione Fluorescent Probes.J. Am. Chem. Soc 2010, 132 (2), 725–736. [DOI] [PubMed] [Google Scholar]
  • (23).Castet F; Rodriguez V; Pozzo JL; Ducasse L; Plaquet A; Champagne B Design and Characterization of Molecular Nonlinear Optical Switches. Acc. Chem. Res 2013, 46 (11), 2656–2665. [DOI] [PubMed] [Google Scholar]
  • (24).Yildiz I; Deniz E; Raymo FM Fluorescence modulation with photochromic switches in nanostructured constructs. Chem. Soc. Rev 2009, 38 (7), 1859–1867. [DOI] [PubMed] [Google Scholar]
  • (25).Barman S; Das J; Biswas S; Maiti TK; Singh NDP A spiropyran-coumarin platform: an environment sensitive photo-responsive drug delivery system for efficient cancer therapy. J. Mater. Chem. B 2017, 5 (21), 3940–3944. [DOI] [PubMed] [Google Scholar]
  • (26).Shiraishi Y; Sumiya S; Hirai T Highly sensitive cyanide anion detection with a coumarin-spiropyran conjugate as a fluorescent receptor. Chem. Commun 2011, 47 (17), 4953–4955. [DOI] [PubMed] [Google Scholar]
  • (27).Su T; Cheng FR; Cao J; Yan JQ; Peng XY; He B Dynamic intracellular tracking nanoparticles via pH-evoked “off-on” fluorescence. J. Mater. Chem. B 2017, 5 (17), 3107–3110. [DOI] [PubMed] [Google Scholar]
  • (28).Zhang Y; Ma L; Tang C; Pan S; Shi D; Wang S; Li M; Guo Y A highly sensitive and rapidly responding fluorescent probe based on a rhodol fluorophore for imaging endogenous hypochlorite in living mice. J. Mater. Chem. B 2018, 6 (5), 725–731. [DOI] [PubMed] [Google Scholar]
  • (29).Zhang JT; Yang M; Mazi WF; Adhikari K; Fang MX; Xie F; Valenzano L; Tiwari A; Luo FT; Liu HY Unusual Fluorescent Responses of Morpholine-Functionalized Fluorescent Probes to pH via Manipulation of BODIPY’s HOMO and LUMO Energy Orbitals for Intracellular pH Detection. ACS Sens 2016, 1 (2), 158–165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (30).Zhang SW; Chen TH; Lee HM; Bi JH; Ghosh A; Fang MX; Qian ZC; Xie F; Ainsley J; Christov C; Luo FT; Zhao F; Liu HY Luminescent Probes for Sensitive Detection of pH Changes in Live Cells through Two Near-Infrared Luminescence Channels. ACS Sens 2017, 2 (7), 924–931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (31).Fang MX; Xia S; Bi JH; Wigstrom TP; Valenzano L; Wang JB; Mazi W; Tanasova M; Luo FT; Liu HY A cyanine-based fluorescent cassette with aggregation-induced emission for sensitive detection of pH changes in live cells. Chem. Commun 2018, 54 (9), 1133–1136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (32).Wang JB; Xia S; Bi JH; Fang MX; Mazi WF; Zhang YB; Conner N; Luo FT; Lu HP; Liu HY Ratiometric Near-Infrared Fluorescent Probes Based On Through Bond Energy Transfer and pi-Conjugation Modulation between Tetraphenylethene and Hemicyanine Moieties for Sensitive Detection of pH Changes in Live Cells. Bioconjugate Chem 2018, 29 (4), 1406–1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (33).Zhang YB; Xia S; Fang MX; Mazi W; Zeng YB; Johnston T; Pap A; Luck RL; Liu HY New near-infrared rhodamine dyes with large Stokes shifts for sensitive sensing of intracellular pH changes and fluctuations. Chem. Commun 2018, 54 (55), 7625–7628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (34).Chen TH; Zhang SW; Jaishi M; Adhikari R; Bi JH; Fang MX; Xia S; Luck RL; Pati R; Lee HM; Luo FT; Tiwari A; Liu HY New Near-Infrared Fluorescent Probes with Single-Photon Anti-Stokes-Shift Fluorescence for Sensitive Determination of pH Variances in Lysosomes with a Double-Checked Capability. ACS Appl. Bio Mater 2018, 1 (3), 549–560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (35).Adamo C; Jacquemin D The calculations of excited-state properties with Time-Dependent Density Functional Theory. Chem. Soc. Rev 2013, 42 (3), 845–856. [DOI] [PubMed] [Google Scholar]
  • (36).Jacquemin D; Wathelet V; Perpète EA; Adamo C Extensive TD-DFT Benchmark: Singlet-Excited States of Organic Molecules. J. Chem. Theory Comput 2009, 5 (9), 2420–2435. [DOI] [PubMed] [Google Scholar]
  • (37).Eiyama A; Kondo-Okamoto N; Okamoto K Mitochondrial degradation during starvation is selective and temporally distinct from bulk autophagy in yeast. FEBS Lett 2013, 587 (12), 1787–1792. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental informaton
NIHMS1061290-supplement-Supplemental_informaton.pdf (11.2MB, pdf)

RESOURCES