Turn-on Fluorescent Probe for Dopamine Detection in Solutions and Live Cells Based on in Situ Formation of Aminosilane-Functionalized Carbon Dots

Abstract

Dopamine (DA) is a critical biomarker for a variety of neurological diseases. A simple and rapid DA detection method will provide aids for clinical diagnosis and treatments for those diseases. In this work, we developed a novel pretreatment-free method for dopamine detection using carbon dots as a turn-on fluorescent probe that was in situ synthesized. In a mild condensation reaction between N-[3-(Trimethoxysilyl)propyl]ethylenediamine (AEATMS) and dopamine, the aminosilane-functionalized carbon dots (SiCDs) were produced, which can be directly used as the dopamine detecting probe. SiCDs exhibited green fluorescence with excitation/emission maximum at 380/495 nm. By measuring the fluorescent intensity of SiCDs, DA concentrations in the reaction system were easily quantified. The linear range of the assay was between 0.1 to 100 μM with a limit of detection (LOD) of 56.2 nM. The probe is of good selectivity and the recoveries of the developed method were in the range of 101.77–119.91% with RSDs within 3.67% in human serum sample tests. We also found that these SiCDs can be synthesized within living MN9D cells under the 37 °C and generated bright fluorescence, which can serve as an indication for the distribution of DA in the cells. The described method exhibit potential in DA detection and live cell imaging for its feature of facility, inexpensiveness, and sensitivity.

Graphical abstract

Green fluorescence emitting carbon dots were in situ synthesized from silane and dopamine in the testing samples. The fluorescent intensities were correlated to dopamine concentrations in the samples. SiCDs can be synthesized inside live cells and indicate the distribution of dopamine.

1. Introduction

Dopamine (DA), a catecholamine, serves as neurotransmitter and hormone in human body and regulates renal, hormonal, and central nervous activities. Particularly in the brain, dopamine pathways participate in motor control and hormone release. These pathways and relevant cells form the neuromodulatory dopamine system[1]. Its malfunction is frequently associated with some of the most infamous nervous diseases such as schizophrenia, Parkinson’s, and Huntington’s diseases [2]. DA concentrations in blood, urine, cerebrospinal fluid provide crucial hints for prevention, diagnosis, and treatment of these diseases.

There are plenty of approaches for DA detection, such as high-performance liquid chromatography, [3], electrochemistry[4], colorimetry[5], capillary electrophoresis[6], eletrogenerated chemiluminescence[7], fluorescence, and enzymatic method[8]. Among those methods, fluorescence methods outstand others with their sensitivity, efficiency, and convenience in operation. To date, a variety of fluorescent probes have been synthesized, including boronic acid derivatives[9], carbon dots[10, 11], gold nanoclusters[12], polydopamine nanoparticles[13], ZnO quantum dots[14], et al. However, the syntheses of the fluorescent probes are usually energy and time consuming, while the sample pretreatments of those fluorescent probes based-methods are often complicated. Therefore, a simpler and straightforward method is needed in experimental and clinical practice.

Carbon dots (CDs) have been widely used in biochemical analyses as fluorescent probes[15, 16]. Various reagents including glucose, organosilane, and citric acid were used as precursors to synthesize CDs. Organosilane-functionalized CDs exhibit good biocompatibility, water dispersibility, cost-effectivity, and capability of surface functionalization, which especially suits aqueous samples testing[17–19]. Compared to complex reaction conditions required for other CDs[20], those syntheses for organosilane-functionalized carbon dots can be carried out under less demanding conditions in the reaction between polybasic acid and organosilane such as N-(3-triethoxysilylpropyl)ethylenediamine (AEATES), N-(2-aminoethyl)-3-aminopropylmethyldimethoxysilane (AEAPMS), (3-aminopropyl)triethoxysilane (APTES), N-(triethoxysilylmethyl)aniline (AMTES), (3-mercaptopropyl)trimethoxysilane (MPTES), 2-[3-(trimethoxysilyl)propoxy]oxirane (GMTMS)[21–24]. Zhang et al. developed a highly sensitive and selective DA detecting method using water-soluble organosilane-functionalized nanoparticles synthesized from trisodium citrate dehydrate and (3-Aminopropyl)trimethoxysilane (APTMS)[25]. Sensitive and selective as this method was, the synthesis of the nanoparticles was energy-consuming and involved complicated microwave-assisted procedures. Cai and co-workers prepared a nanoprobe by an in situ reaction between glucose in serum and APTES, the fluorescent intensity of which was in turn used to determine the glucose level in serum[26]. This work indicated that in situ detection based on organosilane-functionalized carbon dots has great potential in the analysis of other small biomolecules of interest that can be used as carbon dots precursors.

Fluorescent carbon dots in situ synthesized under mild experimental conditions could be a solution to DA imaging in cells. The present DA imaging methods include Falck-Hillarp [27] and immunohistochemistry method. But either of them can be performed on living cells since they all kill the cells before the observation[28]. There were attempts to directly observe DA in live cells utilizing its intrinsic mid-ultraviolet autofluorescence[29], but the excitation/emission wavelengths of 270/320 nm require pulsed femtosecond lasers to execute a two-photon or higher order excitation strategy, which needs expensive instrumentation. Bera et al. proposed a Falck-Hillarp method-like solution: using O-phthalaldehyde (OPA) to penetrate live cells and react with the native catecholamines inside cells to form fluorescent adducts that can be imaged by single-photon fluorescence microscope [30]. But the morphology of the cells would change in the intervening process, and cell damages are still likely to happen due to OPA’s cytotoxicity. In view of the disadvantages of the above methods, organosilane-functionalized CDs might be the ideal probe to imaging native DA in living cells for the low toxicity of aminosilane and easily-achievable synthesizing conditions.

Herein we report a method for DA sensing and quantitative analysis using in situ prepared aminosilane-functionalized carbon dots (SiCDs). SiCDs were prepared by condensation reaction between DA and aminosilane under mild conditions then characterized using UV-Vis spectroscopy, fluorescence spectroscopy, high-resolution transmission electron microscopy (HR-TEM), and flourier transform infrared spectra (FT-IR). The correlation of the fluorescent intensity of the product SiCDs in the final reaction solution to the original concentration of DA was investigated, and a novel pretreatment-free method was established to measure DA concentration in solutions based on that correlation. We also explored the in situ formation of SiCDs inside MN9D cells under cell cultural conditions to evaluate its potential for DA imaging.

2. Experimental Section

2.1. Materials

Uric acid (UA) and 95% N-[3-(Trimethoxysilyl)propyl]ethylenediamine (AEATMS) were purchased from Aladdin Company (Shanghai, China). Dopamine hydrochloride was purchased from J&K Scientific Ltd. (Beijing, China). Lactic acid (LA), citric acid (CA), imidazole, and lactose were purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (Tianjin, China). Glucose, L-ascorbic acid (AA), urea, sodium chloride (NaCl), calcium chloride (CaCl₂), magnesium chloride (MgCl₂), and potassium chloride (KCl) were purchased from Tianjin Bodi Ltd. (Tianjin, China). Glycine (Gly) was purchased from Solarbio Science & Technology Co., Ltd. (Beijing, China). Quinine sulfate was purchased from Chengdu Kelong Chemical Reagent Factory (Chengdu, China). Human serums were purchased from Chengdu Stemstar Biomedical Technology Ltd. (Chengdu, China). O-phthalaldehyde (OPA), adrenaline (Adr), noradrenaline (NA) were purchased from Chengdu Micxy Chemical Co., Ltd. (Chengdu, China). Cell counting kit-8 (CCK8), minimum Eagle’s medium (MEM), fetal bovine serum (FBS), penicillin, and streptomycin were purchased from Bimake Ltd. (Shanghai, China). MN9D cells were purchased from Fenghbio Ltd. (Changsha, China). All solutions were prepared with deionized water produced by a Milli-Q (18.2 MΩ) system (Millipore, Bedford, MA, USA). All above-mentioned chemicals and solvents were of analytical reagent grade if not otherwise noted and used without further purification.

2.2. Synthesis of SiCDs and Detection of DA in Aqueous Solution

SiCDs were prepared following the method described in a work developed by Cai and co-workers with certain modifications[26]. Briefly, 300 μL of AEATMS was added into 2.7 mL of DA solutions of various concentrations (0.1–160 μM). Solutions were mixed by vortex then put in 60°C water bath for 60 min. After cooled to room temperature, fluorescent spectra of the resultant solutions were recorded with an excitation wavelength of 380 nm. The intensities and DA concentrations were calculated to create a regression equation that describes the correlation between them.

2.3. Characterization of SiCDs

Ultraviolent spectra (UV), excitation and emission fluorescent spectra of the above-mentioned solutions were recorded by a Shimadzu UV-1800 UV spectrophotometer and a Shimadzu RF-5301PC spectrofluorophotometer, respectively. Solutions were transferred into a 1000 molecular weight cut-off (MWCO) dialysis bag and dialyzed for 8 hours and then freeze-dried to obtain SiCDs powder to proceed morphology observation and other optical feature analyses. Fluorescent lifetime analysis was performed on an FLS980 fluorescence spectrometer. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) images were obtained by a Tecnai G2 F20 transmission electron microscope. Energy-dispersive X-ray spectroscopy (EDS) pattern was obtained by an Oxford Ultim Extreme EDS equipped on TEM. Fourier transform infrared spectrum (FT-IR) was recorded in KBr by a Perkin-Elmer FT-IR spectrometer. Size distribution of SiCDs was obtained by a Malvern Zetasizer Nano. X-ray photoelectron spectroscopy (XPS) images were obtained by a Thermo ESCALAB 250XI multifunction imaging electron spectrometer.

2.4. Selectivity of SiCDs for DA Detection

An aliquot of 300 μL AEATMS was added into 2.7 mL solutions of 50 μM DA, NaCl, KCl, CaCl₂, MgCl₂, imidazole, urea, Gly, glucose, CA, UA, LA, lactose, AA, Adr, and NA respectively. Each solution was mixed and put into 60 °C water bath for 60 min. After cooled to room temperature, the fluorescent intensity of each solution at the wavelength of 495 nm was recorded under the excitation wavelength of 380 nm.

2.5. DA Detection in Human Serum Samples

After diluted for 10 folds, human serum samples were spiked with different amounts of DA (10, 20, and 30 μM). Then, 300 μL of AEATMS was added into 2.7 mL of as-prepared human serum samples, respectively. A group of DA solutions with different concentrations (5, 10, 15, 20, 25 μM) were also prepared and mixed with AEATMS, as the standard solution group. Solutions were then mixed and put into 60 °C water bath for 60 min. After cooled to room temperature, the fluorescent intensity at the wavelength of 495 nm of each solution was recorded with an excitation wavelength of 380 nm. Intensities and DA concentrations of the standard solution group were calculated to establish a standard curve. Using the equation of the curve, DA concentrations in serum sample group solutions were calculated.

2.6. Cell culture and CCK8 Assay

MN9D cells were used to evaluate the cytotoxicity of AEATMS and OPA. Briefly, the cells were cultured in MEM containing non-essential amino acids (NEAA) and supplemented with 10% fetal bovine, penicillin (100 U mL⁻¹), and streptomycin (100 U mL⁻¹) at 37°C and 5% CO₂. Then, MN9D cells at the density of 10⁴ per well were seeded in a 96-well plate in the cell medium. After overnight incubation at 37°C and 5% CO₂, the culture medium was replaced with 100 μL medium that contained different concentrations of AEATMS (0.2, 0.5, 1, 2, 5, and 10 mM), or 0.1 mM OPA for another 24 h incubation. Then 10 μL WST-8 solution was added into each well and the absorbances at 450 nm were recorded following the CCK8 manufacturer’s instruction. Cells cultured with MEM were used as blank control.

2.7. DA Distribution Observation in Live Cells

MN9D cells were incubated with 0.1 mM OPA for 30 min as described in Bera et al.’s work[30] or 1 mM AEATMS for 24 hours, respectively. After the incubation, the cells were rinsed three times with PBS and observed under a Leica TCS SP8 laser scanning confocal fluorescent microscope.

3. Results and Discussion

3.1. Preparation and Characterization of SiCDs

As shown in Fig. 1a, SiCD solution had maximum UV absorbance at 270 and 310 nm that distinguished SiCDs’ from DA and AEATMS. To evaluate SiCDs’ fluorescent emission property under different excitation, emission spectra were recorded with the excitation wavelengths from 310 to 450 nm. SiCDs demonstrated wavelength-independent fluorescent emission under excitation of 310–420 nm while with the excitation wavelengths from 430 to 450 nm the emission peak shifted from 495 to 525 nm (Fig. S1). Fig. 1b shows the fluorescent excitation and emission spectra of SiCDs that peaked at 380 nm and 495 nm, respectively.

Fig. 1.

(a) UV spectra of DA, AEATMS, and SiCDs. (b) Fluorescent excitation (380 nm) and emission (495 nm) spectra of SiCDs. Inset: photographs of SiCD solution under daylight (left) and 365 nm UV light (right). (c) HR-TEM image of SiCDs. (d) EDS pattern of SiCDs. (e) FT-IR spectrum of SiCDs. (f) size distribution of SiCDs.

After SiCDs were dialyzed and freeze-dried into powder as described above, TEM image, EDS pattern and FT-IR spectra were recorded. In the HR-TEM image (Fig. 1c), amorphous SiCDs’ diameters ranged from 1.2 to 3.7 nm and their averaged diameter was 2.0 nm (±0.3 nm). EDS pattern is presented in Fig. 1d and it indicates that SiCDs consisted of N, O, Si elements and the weight ratios are 4.40% for N, 6.09% for O, and 1.14% for Si, respectively. The Cu can be ascribed to the carbon-coated copper grid used in the measurement. Corresponding to the EDS pattern, absorbance peaks of FT-IR spectrum (Fig. 1e) reveals the following chemical bonds’ existing in SiCDs: N-H (stretching vibration at 3362 cm⁻¹, bending vibration at 1592 cm⁻¹), O-H (stretching vibration at 3285 cm⁻¹), C-H (stretching vibration at 2930 cm⁻¹, bending vibration at 1465 cm⁻¹), C-N (stretching vibration at 1315 cm⁻¹), Si-O-Si (stretching vibration at 1125 cm⁻¹). Fig. 1f presents the size distribution of SiCDs obtained from dynamic light scattering (DLS). The SiCDs diameter had a very narrow distribution range of 2.0 to 4.2 nm, which confirmed the TEM observation results. The element composition, size distribution, and morphology of SiCDs fitted the description of organosilane-functionalized carbon dots[25].

Consistent with the FT-IR spectrum, XPS spectra (Fig. 2) revealed SiCDs’ elements composite and surface chemical structure. The full range spectrum (Fig. 2a) shows five major peaks at 101.9, 153.0, 283.9, 398.0, and 531.0 eV that can be assigned to Si 2p, Si 2s, C 1s, N 1s, O 1s, respectively[24]. High-resolution spectra of each signal were displayed as well. C 1s spectrum in Fig. 2b can be deconvoluted into four components that peaked at 284.1, 284.8, 285.5, and 287.7 eV and assigned to C-Si, C-C/C=C, C-N, and C-O, respectively[18, 26, 31]. N 1s spectrum in Fig. 2c can be deconvoluted into two components that peaked at 398.5 and 399.1 eV and assigned to C-N-C and N-(C)₃[18, 26, 31]. O 1s spectrum in Fig. 2d can be deconvoluted into three components that peaked at 531.0, 532.0, and 533.2 eV and assigned to C=O, C-OH, and Si-O, respectively[18, 26, 31]. Si 2p spectrum in Fig. 2e can be deconvoluted into two components that peaked at 101.8 and 102.5 eV and assigned to Si-C and Si-O[18, 26, 31].

Fig. 2.

XPS spectra of SiCDs. (a) Full range, (b) C 1s, (c) N 1s, (d) O 1s, (e) Si 2p, respectively.

Time-scan spectrum of fluorescent intensity of SiCDs under 380 nm excitation (Fig. S2) exhibits SiCDs’ stability under UV irradiation. With quinine sulfate (quantum yield 54%) as a standard reference, the quantum yield of SiCDs was measured to be 9.72%. According to the fluorescent decay curve of SiCDs (Fig. S3), it fits a second-order exponential function with the fluorescent lifetime of 0.61 ns (τ₁) and 4.22 ns (τ₂).

As described by Zhong et al.[32], the formation mechanism of hydrothermal synthesized quantum dots using organosilane is that large silicon nanoparticles are firstly produced through the hydrolysis of silane in the presence of reducing reagent in aqueous solution, then large particles divide into small silicon crystal nuclei under microwave irradiation or heating, and finally, those nuclei grow into silicon quantum dots through an Ostwald ripening process. However, Oliinyk et al. pointed out that the nanoparticles formed in this process were more likely to be carbon nanoparticles produced by condensation reactions instead of crystalline silicon nanoparticles since their TEM image, XRD pattern, and fluorescent properties have more resemblance to silane-synthesized carbon dots rather than plasma synthesized Si nanoparticles[33]. This work provided more evidence for Oliinyk’s assumption. AEATMS and other aminosilanes are often used as surface modifiers or cross-linking agents for their properties to hydrolyze into silanol in aqueous environments. The unstable silanol tends to condensate through dehydration with hydroxyl groups, which were evidenced by UV absorbance spectra of pure AEATMS, heated AEATMS aqueous solution, and SiCD solution (Fig. S4) in this work. UV absorbance spectrum of the heated AEATMS aqueous solution was different from pure AEATMS but highly resembled that of SiCDs solution, while the absorbance increased significantly. The Si-O-Si signal in FT-IR and XPS spectra also confirmed the occurrence of condensation. Moreover, the fluorescent properties of these solutions are dopamine-dependent, that is, when dopamine was absent, there was no fluorescent emission. These facts indicated that silanol hydrolyzed from AEATMS in aqueous solutions might condensate and aggregate, and dopamine will condense with the silanol through dehydration and further aggregate to form the fluorescent carbon dots. To evaluate the stability of SiCDs, the fluorescent intensity of SiCD solution at 495 nm was measured 7 days in a row. As shown in Fig. S6, the fluorescent intensity grew continuously days after the synthesis reaction. Based on the assumption of AEATMS-dopamine aggregation and condensation, this can be explained as the result of aggregation under room temperature. SiCDs also exhibited stable fluorescence in different pH phosphate buffer (Fig. S7).

3.2. Sensitivity of SiCDs for DA Detection

To quantify the positive correlation between fluorescent intensities of SiCDs and DA concentrations, fluorescent spectra of a series of SiCD solutions were recorded and are presented in Fig. 3a. With concentrations of DA ranging from 0.1 to 140 μM, the fluorescent intensity at 495 nm increased gradually. Especially, between 0.1 μM and 100 μM, fluorescent intensities and DA concentrations exhibited excellent linear correlation (R² = 0.998) that can be described by the following equation:

$$\text{y}=2.1107+6.59227\text{x}$$

Fig. 3.

(a) Fluorescent emission spectra of SiCD solutions with DA concentration varying from 0.1 μM to 140 μM (excitation: 380 nm). Inset: enlarged graph of spectra of 0.1 μM, 0.2 μM, 0.5 μM, 1 μM, 2 μM DA concentration solutions. (b) Fluorescent intensity at 495 nm response against different DA concentration and its correlation with concentration exhibited good linearity. Inset: enlarged graph of part of the calibration curve.

Since the fluorescent probe was prepared consuming DA in the reaction mixture, the regression curve was calculated using fluorescent intensities rather than relative intensities. The calibration curve of concentration-FL intensity correlation is presented in Fig. 3b. Using the regression equation of the curve, limit of detection (LOD) was calculated to be 56.2 nM (3 times the signal-to-noise ratio) which is satisfactory compared with similar fluorescent DA detecting methods. The linear range of the proposed method was also wide among those methods. Table 1 summarizes and compares some of the recently reported dopamine detecting methods. The advantage of this preparation-free method made it a practical and pervasive choice under various conditions. Particularly with material as facile as AEATMS, SiCDs could be a promising probe for DA detection in varies samples.

Table 1.

An Overview on Recently Reported Dopamine Detection Methods

Materials used	Method	Require probe preparation	Range	LOD	Reference
Homogenate samples of mouse striatum	HPLC	Yes	0–2.51 μg/mL	0.031 μg/mL	[3]
Multi-wall carbon nanotubes and magnetic nanoparticles	Electrochemical	Yes	5–180 μM	0.43 μM	[4]
Gold nanoparticles	Colorimetric	Yes	5–600 nM	2 nM	[5]
Ag2Se quantum dots	Electrogenerated Chemiluminescence	Yes	0.5–19 μM	0.1 μM	[7]
Immobilized peroxidase	Enzymatic	Yes	33–1300 μM	2 μM	[8]
Carbon nanoparticles, Fe3+	Fluorometric	Yes	0–20 μM	0.32 μM	[10]
DA aptamer, carbon dots, nano-graphite	Fluorometric	Yes	0.1–5 nM	0.055 nM	[36]
Gold nanoclusters	Fluorometric	Yes	0.01–1 μM	10 nM	[12]
Graphene quantum dots	Fluorometric	Yes	0.25–50 μM	0.09 μM	[37]
Graphene oxide	Fluorometric	Yes	0.25–20 μM	94 nM	[38]
Polydopamine	Fluorometric	No	0.5–20 μM	40 nM	[13]
DNA/single-wall carbon nanotubes	Fluorometric	Yes	0.01–10 μM	11 nM	[39]
ZnO quantum dots	Fluorometric	Yes	0.05–10 μM	12 nM	[14]
CdSe/ZnS quantum dots	Fluorometric	Yes	0.1–20 μM	29.3 nM	[40]
Cu nanoclusters	Fluorometric	Yes	0.1–0.6 μM	0.1637 pM	[41]
AEATMS	Fluorometric	No	0.1–100 μM	56.2 nM	This work

3.3. Selectivity of SiCDs for DA Detection

DA is usually accompanied in actual bio-samples by other molecules such as AA, making it necessary to evaluate the selectivity of a new DA detection method. To examine the selectivity of SiCDs, common molecules in human body fluid (NaCl, KCl, CaCl₂, MgCl₂, imidazole, urea, Gly, glucose, CA, UA, LA, lactose, AA, Adr, NA, 50 μM) were chosen and respectively added into the reaction system as mentioned above. As shown in Fig. 4, most of them exhibited no significant FL intensities. It indicates that these molecules have no obvious interference to DA detection.

Fig. 4.

Fluorescent intensity at 495 nm of solutions of DA and other common molecules in biological environments. Concentrations of all substances were 50 μM: 1, Na⁺; 2, K⁺; 3, Ca²⁺; 4, Mg²⁺; 5, Imidazole; 6, Urea; 7, Gly; 8, Glucose; 9, CA; 10, UA; 11, LA; 12, Lactose; 13, AA; 14, Adr; 15,NA; 16, DA.

The similarity between the chemical structures of DA and AA might be the reason why resulting solutions in the AA group also exhibited weak fluorescence. AA and AEATMS in the solution might react in the solution and produce carbon dots. In fact, there are already several reports about synthesizing fluorescent nanoparticles from organosilane and AA[31, 34]. However, those reported methods are more demanding in reaction conditions, requiring microwave-heating, high pressure, or longer time. Meanwhile the amount of AA used as carbon dots precursor is usually of gram level. To produce highly fluorescent carbon dots while using sub-mircomolar level precursor existing in the detecting sample as describe in this work is rare. Therefore, with the mild condition applied in this work, few AA was transferred into fluorescent carbon dots and influence from AA is negligible as shown in Fig. 4.

Due to the structural similarity between catecholamines, carbon dots could be produced with the proposed method using other catecholamines such as adrenaline and noradrenaline. Fortunately, those carbon dots synthesized from adrenaline and noradrenaline appeared to have different photoluminescent properties from SiCDs prepared with AEAMTS in this work. The maximum excitation and emission wavelengths of them were both longer than the proposed SiCDs. As seen in the Fig. 4, when under the 380 nm excitation, adrenaline and noradrenaline groups’ emissions at 495 nm could be neglected.

3.4. DA Detection in Human Serum Samples

To test its ability of detection in a real complex matrix, the proposed SiCDs method was applied to measure DA concentrations in human serum samples spiked with different amounts of DA. The results are summarized in Table 2. Recoveries were calculated and ranged from 101.77 to 119.91%. RSDs were less than 4%.

Table 2.

Determination of DA in Human Serum Samples

Sample	DA added (μM)	DA found (μM)	Recovery (%), n=3	RSD (%), n=3
1	10	11.99	119.91	3.67
2	20	20.94	104.72	1.77
3	30	30.53	101.77	2.91

Recoveries in all groups are over 100%. It might indicate that there is still interference from other molecules in the sample. And to determinate sub-micromolar level DA in complex biological samples, SiCD method still needs improvement.

3.5. DA Detection in Live Cells

MN9D cells, one type of dopaminergic neurons, are the fusion of rostral mesencephalic tegmentum (RMT) cells from 14-day-old C57BL/6J mouse embryos and hypoxanthine phosphoribosyl-transferase (HPRT)-deficient N18TG2 neuroblastoma cells. Its immortality and sensitivity towards commonly used neurotoxin N-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) make MN9D cell line an ideal model for neurotransmitter interaction and traffic[35]. Comparing with other neurons such as PC-12 cells, MN9D cells have larger vesicles and quantal size and were therefore chosen for DA imaging observation. The CCK8 assay results presented in Fig. 5 showed that when exposed to 2 mM and lower concentrations of AEATMS, the cell viability of MN9D cells was at the same level as the control group. The low cytotoxicity of AEATMS made it possible for AEATMS to be incubated with cells without reducing cell activity or causing any change in cell morphology.

Fig. 5.

Cell viability tested by CCK8 assay. MN9D cells were incubated with 0, 0.2, 0.5, 1, 2, 5, and 10 mM AEATMS, and 0.1 mM OPA.

OPA can permeate cells and form fluorescent adducts with DA in live cells [30]. Therefore, cells incubated with OPA can be used as a positive control to verify the viability of SiCDs as DA sensing probes in live cells. Comparing Fig. 6b and Fig. 6d, the fluorescent microscope images of cells incubated with AEATMS and OPA exhibited the same fluorescence scattering pattern inside the cells. It showed that fluorescent nanoparticles were formed in the cells and can indicate the position of DA. The bright round-shaped spots inside the cells in the fluorescent images were likely to be DA containing vesicles or vesicular clusters. And judging from the bright field images of control and AEATMS incubated cells, the shapes of cells can maintain the same with the presence of SiCDs. Thus, in situ synthesizing SiCDs in living cells for DA sensing and imaging proved to be a feasible solution.

Fig. 6.

Laser scanning confocal fluorescent microscope images of MN9D cells. Panel (a) (c) (e) are bright field images, panel (b) (d) (f) are fluorescent images. (a) (b) control cells. (c) (d) cells incubated with 0.1 mM OPA. (e) (f) cells incubated with 1 mM AEATMS. The excitation wavelength was 405 nm.

4. Conclusions

Through the in situ synthesis of organosilane-sourced carbon dots, an efficient fluorescent dopamine detection method was established. It is the first report about carbon dots synthesis using dopamine existing in samples and in turn detecting DA level. With excellent photoluminescent properties, SiCDs can quantify DA in a wide range of 0.1 to 100 μM. The proposed method achieved a low LOD as well as good selectivity, and its performance in human serum samples was satisfactory. Furthermore, we proved that SiCDs can be formed inside living cells and capable of imaging DA distribution in cells. Simple, low-cost, and efficient, SiCDs have the potential to be utilized as a turn-on fluorescent sensing probe for DA detection in biological samples and live cells.