Near-infrared fluorescent probe based on rhodamine derivative for detection of NADH in live cells

Abstract

A near-infrared fluorescent probe was prepared for selective detection of reduced nicotinamide adenine dinucleotide (NADH) in live cells. The probe turns off the fluorescence with a closed spironolactone switch. However, reduction of the probe by NADH turns on fluorescence at 740 nm. Theoretical calculations suggest a more planar arrangement between the rhodamine and quinoline moieties with increased π-delocalization resulting from reduction.

1. Introduction

Reduced nicotinamide adenine dinucleotide (NADH) and its phosphate ester (NADPH) serve as the most essential coenzymes in all living organisms and exert significant roles in antioxidation, energy metabolism, gene expression, immunological functions, mitochondrial functions, calcium homeostasis, biosynthesis, aging, cell death, and carcinogenesis [1–3]. Significant abnormal NAD(P)H concentration levels could lead to diseases such as diabetes, DNA damage, Cancer, Alzheimer’s, and Parkinson’s diseases [4–11]. Accurate detection of NAD(P)H distribution and dynamic variations will provide an insightful understanding of their biological functions and pathological characters [2,4–7,11]. Compared with traditional methods such as HPLC [12,13], capillary electrophoreses [14], electrochemistry [15,16] and enzymatic cycling assay [13], fluorescence detection possesses advantages, including tracking the dynamics of processes in live cells for both in vitro and in vivo imaging [2,3,11,17–30]. NAD(P)H emits fluorescence at 450 nm with low fluorescence quantum yield [7,8,10,11]. In order to avoid the fluorescence background from NAD(P)H [7,10], it is imperative to develop fluorescent probes for NAD(P)H detection in the deep red and near-infrared region [2,3,18].

In this communication, we report a near-infrared fluorescent probe for selective detection of NADH in live cells. The probe shows extremely weak fluorescence with a closed spirolactone configuration. However, the presence of NADH causes the reduction of the probe and significantly increases the probe fluorescence with considerably enhanced π-conjugation of rhodamine derivative (Scheme 1). The probe has been used to detect NADH concentration changes in HeLa cells caused by the addition of glucose, and different extracellular pyruvate/lactate ratio levels.

Scheme 1.

Chemical structure changes of fluorescent probe in response to NADH.

2. Materials and instruments

2.1. Materials

Unless specifically indicated, all reagents were obtained from commercial suppliers. Compound 3 was synthesized and characterized according to a reported procedure [31].

2.2. UV–Vis absorption and fluorescence spectroscopy

Absorption spectra were collected on a Cary 60 UV–Vis spectrometer, and fluorescence spectra were conducted using a Jobin Yvon Fluoromax-4 spectrofluorometer.

2.3. Cell culture and cellular toxicity testing

HeLa cells were cultured in DMEM (Dulbecco’s Modified Eagle’s Medium) containing 10 % FBS (Fetal Bovine Serum) under an atmosphere of 5 % CO₂ at 37 °C [32–36]. Cellular cytotoxicity of the probe was evaluated with MTT methods. Living cells were placed into 96 well plates with a density of 10,000 in each well for 24-h incubation. Then, the cells were replaced with fresh culture medium containing different probe concentrations ranging from 0 to 40 μM for another 24-h incubation. Next, 10 μL MTT (5 mg/mL) was added to each well for 4-h incubation. Lastly, 100 μL DMSO was added to each well to dissolve the formazan and measurements were conducted using a Microplate Reader at 570 nm.

2.4. Fluorescence imaging of NADH in live cells

A series of experiments were performed to explore the feasibility of probe A in detecting NADH in cultured HeLa cells. First, HeLa cells were incubated with 10 μM probe A at 37 °C for 120 min as a control experiment and imaged to detect endogenous NADH for an all substrate-dependent assay. In order to detect intracellular NADH changes caused by exogenous NADH through diffusion into live cells, we treated HeLa cells with 20 μM exogenous NADH for 120 min, washed the cells with a PBS buffer (pH 7.4), and further incubated with a cell medium containing 10 μM probe A for 120 min. For co-localization experiments to determine where the probe locates within live cells, HeLa cells pretreated with 20 μM exogenous NADH were further incubated with 5 μM rhodamine 123 (as a mitochondria-specific probe) and 10 μM probe A for 120 min. Two fluorescence channels were used to collect cellular fluorescence signals from 520 nm to 580 nm under excitation of 488 nm for rhodamine 123, and from 720 nm to 770 nm under 633 nm excitation for probe A. For each substrate-dependent assay, different substrates such as glucose, lactate, and pyruvate were incubated with HeLa cells in serum-free DMEM/HBSS (1:9) solutions. In glucose-dependent studies, HeLa cells were first treated with 20 mM glucose for 120 min and then further incubated with 10 μM probe A in serum-free DMEM/HBSS solution for 120 min. In lactate/pyruvate-dependent experiments, HeLa cells were incubated with 10 mM lactate and 5 mM pyruvate for 120 min, respectively, and then further incubated with 10 μM probe A for 120 min.Cellular fluorescence images were collected from 720 nm to 770 nm under 633 nm excitation. Confocal images were collected with Olympus FluoViewTM FV1000 using the FluoView software.

3. Methods

3.1. Synthetic route

In order to prepare a near-infrared rhodamine-based fluorescent probe for the detection of NADH, we conducted a condensation reaction of 9-(2-carboxyphenyl)-6-(diethylamino)-1,2,3,4-tetrahydroxanthylium perchlorate (3) with quinoline-3-carbaldehyde in acetic acid solution under reflux conditions, affording the near-infrared rhodamine dye 5 (Scheme 2). Compound 3 was prepared and characterized according to the reported procedure [31]. Fluorescent probe A for NADH was prepared through methylation of rhodamine dye 5 with methyl trifluoromethanesulfonate in dichloromethane solution at room temperature (Scheme 2). The probe was characterized by ¹H, ¹³C NMR, and mass spectroscopy.

Scheme 2.

Synthetic route to prepare the fluorescent probe.

3.2. Synthetic of the probe

When compound 3 [31] (475 mg, 1 mmol) and quinoline-3-carbaldehyde (157 mg, 1 mmol) were added to 10 mL CH₃COOH, the mixture was stirred for 5 h at 80 “C. After the reaction mixture cooled to room temperature, the solution was extracted with 50 mL dichloromethane, washed by water and brine, respectively (Scheme 2). The organic layer was collected, dried over anhydrous Na₂SO₄ and filtered. After the filtrate was concentrated, the residue was purified to get compound 5 by using flash column chromatography through gradient elution with methanol to dichloromethane ratio from 5 % to 10 % (252 mg, yield = 49%). ¹HNMR (400 MHz, Methanol-d₄) δ 9.08 (d, J = 2.2 Hz, 1H), 8.62 (s, 1H), 8.30 (d, J = 6.1 Hz, 2H), 8.05 (d, J = 7.9 Hz, 2H), 7.83 (t, J = 7.5 Hz, 2H), 7.75 (t, J = 7.4 Hz, 1H), 7.70 – 7.65 (m, 1H), 7.34 (dd, J = 7.5, 1.3 Hz, 1H), 7.30 – 7.23 (m, 2H), 7.12 (d, J = 9.5 Hz, 1H), 3.73 (q, J = 7.1 Hz, 4H), 3.09 (s, 2H), 2.43 (s, 2H), 1.87 (d, J = 7.8 Hz, 2H), 1.33 (t, J = 7.1 Hz, 6H). 13C NMR (101 MHz, Methanol-d₄) δ 11.78, 21.51, 25.74, 26.99, 95.51, 127.66, 128.59, 128.77, 130.10, 131.1, 137.69, 146.69, 151.73. MS/Z = 515.21 (calcd. 515.23292 for C₃₄H₃₁N₂O₃⁺).

When compound 5 (256 mg, 1 mmol) and MeOTf (163 mg, 1 mmol) were added to 10 mL dichloromethane solution, the mixture was stirred at room temperature for 5 h (Scheme 2). 50 mL H₂O was added to the above solution, and extracted with 40 mL dichloromethane. The organic layer was collected, dried with anhydrous MgSO₄ and filtered. The filtrate was collected and the solvent was evaporated under reduced pressure. The residue was purified by using flash column chromatography through gradient elution with methanol to dichloromethane ratio increasing from 5% to 10% in order to elute the target probe (98 mg, yield = 37%) MS/Z = 530.26 (calcd. 530.25 for C₃₅H₃₄N₂O₃⁺). ¹HNMR (400 MHz, Methanol-d₄) δ 9.58 (s, 1H), 9.29 (s, 1H), 8.49 (dd, J = 18.5, 8.6 Hz, 2H), 8.31 – 8.27 (m, 1H), 8.16 (d, J = 7.7 Hz, 1H), 8.10 – 8.00 (m, 2H), 7.78 (d, J = 7.5 Hz, 1H), 7.71 (t, J = 7.4 Hz, 1H), 7.33 – 7.27 (m, 1H), 7.06 – 6.91 (m, 3H), 4.74 (d, J = 1.4 Hz, 3H), 3.63 (q, J = 7.2 Hz, 4H), 3.03 (d, J = 8.5 Hz, 2H), 2.32 (s, 2H), 1.84 (s, 2H), 1.36 – 1.12 (m, 6H). 13C NMR (101 MHz, Methanol-d₄) δ 145.82, 135.97, 133.25, 130.82, 130.53, 130.05, 118.52, 95.84, 45.54, 45.25, 26.67, 21.73, 11.76.

4. Optical properties of the probe

The intermediate compound 5, shows a broad absorption peak at 550 nm and a fluorescence peak at 660 nm in pH 7.4 PBS buffer containing 10% DMSO (Figs. S8–S9). In contrast, the end product probe A shows a prominent absorption peak at 550 nm due to the closed spironolactone structure in the absence of NADH (Fig. 1). After the addition of 50 μM NADH reduces probe A, it triggers a new near-infrared absorption peak at 710 nm with significantly enhanced π-conjugation of rhodamine dye (Scheme 1), concomitant with a decrease in the absorption peak at 550 nm (Fig. 1). Additionally, gradual increases in NADH concentrations cause gradual increases in absorption at 710 nm and gradual decreases in absorption at 550 nm. The nature of this change demonstrates that this probe is very sensitive to NADH. Computing and plotting the ratio of these changes results in a linear relationship to increases in NADH concentrations up to stoichiometric equivalence of the probe (Fig. S10).

Fig. 1.

The absorption spectra of 10 μM of probe A (left and right) in the absence and presence of different concentrations of NADH in pH 7.4 PBS buffer containing 5% DMSO solution at room temperature with a 120-min incubation time of the probe with NADH and their linear responses. The arrows indicate the direction of decreasing and increasing NADH concentrations in both illustrations which result in gradual decreases in absorption at 550 nm and gradual increases in absorption at 710 nm.

Probe A displays no fluorescence in the absence of NADH and contains a closed spironolactone ring configuration in the rhodamine dye (Scheme 1 and Fig. 2). Gradual increases of NADH concentration in pH 7.4 PBS buffer containing 10 μM probe A causes a new fluorescence peak at 740 nm and results in gradual increases in this fluorescence peak (Fig. 2). Reduction of probe A by NADH, which results in an opened ring spironolactone configuration, presumably significantly enhances the π-conjugation of the rhodamine fluorophore, leading to the new fluorescence peak at 740 nm (Scheme 1). The probe shows nice linear fluorescence responses to gradual increases in NADH concentration (Fig. S11). A peak at M/Z = 531.33 (calcd. 531.67 for C₃₅H₃₅N₂O₃⁺, Fig. S6) obtained using an electrospray mass spectrometer provides evidence of the formation of the reduction product of probe A, i.e., AH, by NADH. We also obtained similar agreement between the experimentally determined mass of the non-reduced compound, i.e., probe A, and that calculated (found MS/Z = 530.26, calcd. 530.67 for C₃₅H₃₄N₂O₃²⁺, Fig. S7).

Fig. 2.

The fluorescence spectra of 10 μM probe A (left and right) in the absence and presence of different concentrations of NADH in pH 7.4 PBS buffer under excitation of 670 nm at room temperature with a 120-min incubation time of the probe with NADH. The arrows indicate the result on the fluorescence with gradual increases in NADH concentration.

In practical applications, the selectivity of the probe is crucial. After a series of external substances interference tests, it can be concluded that the probe displays high selectivity to NADH over reactive oxygen, nitrogen, and sulfur species such as peroxynitrite, singlet oxygen, hydroxyl radicals, hydrogen peroxide, sodium hypochlorite, sodium hypobromite, cysteine, homocysteine, glutathione, sodium hydrosulfide, and hydrogen sulfide (Fig. S12). Good selectivity means this probe can effectively measure NADH concentrations in the presence of other bioactive molecules in living systems.

5. Theoretical results

To better understand the response mechanism, we conducted theoretical calculations on the cations in probes A and AH to determine the nature of the transformation. In the optimized geometries, an inter-planar angle of 42.2° between the rhodamine and quinoline moieties in probe A pertains as a consequence of the spironolactone ring, which reduces to 21.5° upon ring-opening in probe AH, see Figs. S13 and S16. This results in more π-orbital delocalization between the rhodamine and quinoline moieties resulting in a shift in the absorbance to lower frequencies. The extent of the delocalization can be seen in the current density diagram for the HOMO to LUMO transition in probe AH (Fig. 3). The calculated absorbance for the probe A and reduced AH were at 561 and 615 nm are within the range suggested for such calculations of 0.3 eV (i.e., 0.04 and 0.27 eV respectively) [37].

Fig. 3.

Drawings of the current density difference for the HOMO → LUMO transition as isosurfaces for probe AH in water. The numerical range values of the color scale displayed at the top of the figure are ± 5.000 e⁻⁵. The nature of the specific transitions is indicated, and illustrations of the LCAOs are presented in the supporting information.

6. Cytotoxicity and cellular imaging of the probe

The probe possesses low cytotoxicity as an MTT assay shows that cell viability of 80% was observed at 40 μM of the probe concertation for 24-h incubation, indicating that the probe can be used for imaging application with good biocompatibility (Fig. 4) [32–36, 38–41]. We further tested whether the probe can respond to changes in NADH concentration in live cells. There is moderate fluorescence of probe A in HeLa cells, further incubation of HeLa cells stained with probe A with exogenous 20 μM NADH results in strong fluorescence in HeLa cells (Fig. 5), which indicates that NADH diffuses into the cells, further reduces the probe and triggers fluorescence increases with more significantly enhanced π-conjugation in the near-infrared rhodamine dye, which was consistent with what was observed in a test tube experiment (Fig. 2).

Fig. 4.

Cytotoxity and cell proliferation of probe A tested by MTT assay. The HeLa cells were incubated with 10, 20, 20 and 40 μM probe A for 24 h, and cell viability was measured by adding MTT reagent and collecting absorbance at 475 nm. The cell viability was directly proportional to the absorbance measured at 475 nm and normalized to control cells in the absence of the probe.

Fig. 5.

Fluorescence images of HeLa cells incubated with 10 μM probe A in the absence and presence of 20 μM NADH under 2-hour incubation. Untreated cells (without probe A) were used as a control (in the first column). Scale bar: 10 μm.

We hypothesized that the probe would specifically stain mitochondria through electrostatic interactions between the positively charged probe and the negative membrane potential in the inner mitochondrial membrane. In order to prove our hypothesis, we conducted colocalization experiments by incubating mitochondria-specific rhodamine 123 and 10 μM probe A in the presence of exogenous 10 μM NADH for 2 h (Fig. 6). Colocalization analysis shows that the Pearson correlation coefficient between probe A and rhodamine 123 was 0.96, indicating that probe A selectively stays in mitochondria (Fig. 6).

Fig. 6.

Fluorescence images of HeLa cells incubated with 10 μM and 5 μM rhodamine 123 in the presence of 10 μM NADH for 2-hour incubation. The Pearson correlation coefficient is 0.96. Cellular fluorescence images were collected from 720 nm to 770 nm under 633 nm excitation. Scale bar: 10 μm.

In order to demonstrate whether the probe can be used to detect endogenous NADH changes in live cells, we stained HeLa cells with probe A for 120 min and further incubated the cells with 20 mM glucose for 120 min (Fig. 7). We also incubated HeLa cells with 20 mM glucose for 120 min as a control experiment to assess potential fluorescence background from HeLa cells in the presence of glucose. There was no fluorescence observed in HeLa cells in the presence of glucose without probe A in the control experiment (Fig. 7). Incubation of HeLa cells with 10 μM of probe A shows moderate intracellular fluorescence due to reduction of the probe by endogenous NADH inside HeLa cancer cells. However, further incubation of the probe-stained HeLa cells with 20 mM glucose for 120-minutes results in stronger near-infrared fluorescence in HeLa cells (Fig. 7). In cancer cells, the intracellular NADH level significantly depends on glycolysis. Glycolysis consumes glucose and produces NADH. As a result, incubation of HeLa cells with probe A and glucose results in stronger fluorescence through further reduction of the probe by glucose-triggered NADH in the HeLa cells. These results show that the probe can visualize endogenous NADH changes generated by a glucose stimulus (Fig. 7).

Fig. 7.

Fluorescence images of HeLa cells incubated with 20 mM glucose (a control experiment), and incubated with 10 μM probe A in the absence and presence of 20 mM glucose for 120 min. Scale bar: 10 μm.

The extracellular pyruvate/lactate ratio has served as a marker of intracellular NADH/NAD⁺ standing as lactate dehydrogenase highly efficiently catalyses the oxidation of NADH by transporting two electrons from NADH to pyruvate, generating lactate [42]. We also investigated the effect of pyruvate/lactate imbalance on NADH levels in live cells. The cellular fluorescence intensity is significantly reduced under incubation of HeLa cells with exogenous pyruvate, while lactate treatment of HeLa cells results in stronger intracellular fluorescence of the fluorescent probe (Fig. 8). These results suggest that the probe can be employed to monitor endogenous NADH concentration changes caused by an imbalance in the pyruvate/lactate ratio in live cells.

Fig. 8.

Fluorescence images of HeLa cells incubated with 10 μM probe A in the absence and presence of 10 mM lactate and 6 mM pyruvate ratio for 120 min, respectively. Scale bar: 10 μm.

7. Conclusions

We have prepared a near-infrared rhodamine-based fluorescent probe to detect intracellular NADH changes in HeLa cells caused by the addition of glucose. The probe can also monitor different extracellular pyruvate/lactate ratio levels.