A. Apparatus and materials

Carbon fiber (Φ7 μm, Toray Industries Co., Ltd. Japan), glass capillary (Φ1 mm) and cooper wire (Φ0.5 mm) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Graphite powder conductive paint (Ted Pella, USA), DA, AA, HAuCl₄, Tris(2-carboxyethyl) phosphine (TCEP) and 6-Mercaptohexanol (MCH) were purchased from Sigma-Aldrich (St. Louis, MO). Cathodic electrophoresis paint (CEP) was obtained from Aomei Chemical Co., Ltd. (Shanghai, China). The DNA sequence (P1) in the experiment is 5’-SH-(CH₂)₆-GTCTCTGTGTGCGCCAGAGACACTGGGGCAGATATG GGCCAGCACAGAATGAGGCCC-MB-3’ (Sangon Biotech Co., Ltd., Shanghai, China). All solutions were prepared with ultrapure water.

B. Preparation of carbon fiber microelectrode

The carbon fiber microelectrode (CFME) was prepared following the protocol developed by Huang et al. [24]. The carbon fiber was cleaned by sonication for 5 min successively in acetone, alcohol and ultrapure water, and dried at 37 ℃. It was stuck onto a 10-cm cooper wire with graphite powder conductive paint and inserted into the glass capillary tip with an inner diameter of 20 μm, with an exposed tip of about 0.5 mm. The glass capillary was sealed using an alcohol lamp. Finally, the tip of carbon fiber was etched to ∼ 200 μm length and ∼ 500 nm diameter by flame.

C. Fabrication of nano-flower gold microelectrode

The prepared CFME was encapsulated by CEP. The CEP solution was prepared by diluting CEP with water (1:5, by mass). The CFME tip was cleaned by sonication gently in ultrapure water and fully immersed in the CEP solution. In the deposition process of CEP, the CFME was used as the working electrode, and a platinum electrode was used as the reference and counter electrode. For uniform deposition, the platinum electrode was bent into circular shape (∼ Φ1.5 cm) and the CFME was located at the circle center. The deposition was carried out under the "i-t" mode, with an initial potential of -8 V. After 100 s, the encapsulated CFME was washed with ultrapure water, heated to 80 ℃ and kept for 5 min. The CEP would be solidified and shrunk. A small area of CFME would be exposed. The apparent area of the exposed CFME was calculated from the steady current detected in 10 mmol L^-1 K₃[Fe(CN)₆] (containing 0.5 mol/L KCl) using Eq. (1),

where i₀ is the steady state current detected in 10 mmol L^-1 K₃[Fe(CN)₆], r₀ is the effective radius of the exposed tip of CFME, n is the electron transfer number in the oxidation-reduction reaction, D_c is the diffuse coefficient of the agent (7.2×10^-6 cm²/s for K₃[Fe(CN)₆]), and F is Faraday constant.

The encapsulated CFME of a similar size was selected in the experiment (∼ 250 nm in radius) and immersed in a HAuCl₄ solution containing 20 mmol L^-1 HAuCl₄ and 0.5 mol/L HCl. Au deposition was carried out at -8 V for 100 s using DC potential amperometry in a three-electrode system with Ag/AgCl serving as a reference electrode.

D. Electrochemical detection of DA of NGFME

The prepared NGFME was allowed to incubate with 25 nmol/L P1 loading buffer (containing 1 mol/L NaCl, 20 mmol L^-1 MgCl₂, 20 mmol L^-1 PB and 3 mM TCEP) overnight in the dark. After that, the microelectrode was treated with 2 mmol L^-1 MCH to tune the loading density of P1. The DA was detected in a solution containing 100 mM NaCl, 2 mM MgCl₂ and 20 mM PB, using a three-electrode system with "AC Voltammetry (ACV)" mode on the electrochemical station. Note that the solution of DA and AA should be prepared fresh at the time of use to prevent oxidation of the reagents.

E. Electrochemical detection of DA and AA of CFME

DA and AA detection of CFME was conducted in the detection solution containing 100 mM NaCl, 2 mM MgCl₂ and 20 mM PB (pH 7.4) using a three-electrode system through ACV mode on the electrochemical station.

A. Fabrication and characterization of NGFME

The process of NGFME fabrication is illustrated in Fig. 1(a). Scanning electron microscope (SEM) images of the microelectrode tip are shown in Figs. 1(b)–1(e). Figure 1(b) is a general view of the CFME tip, revealing its smooth needle-like shape. After encapsulated by CEP, the tip surface became rugged (Fig. 1(c)). Figure 1(d) shows the NGFME tip, with flower-like structure after Au deposition. Cyclic voltammetry (CV) curves tested in 0.05 mol/L H₂SO₄ demonstrate the component of the flower-like structure is Au, with the special reductive peak of Au at 0.863 V (Fig. 1(f)). The structure is sized at about 100 μm. Figure 1(e) shows that surface of the flower-like structure is spiny with thorns in nanometer level, with its surface area being comparable to a macroscopic gold electrode of 2-mm diameter.

Fig. 1.Full-screen View

(Color online) Schematics of a NGFME (a); Scanning electron microscope (SEM) images of the tip of CFME (b); CEP-encapsulated CFME (c); NGFME (d) and structure of the NGFME (e); and cyclic voltammetry curve (4 segments) of the NGFME measured in 0.05 mol/L H₂SO₄ (f).

B. The fabrication of micro E-DNA aptasensor

A novel E-DNA micro aptasensor was fabricated based on the NGFME for DA detection. As shown in Fig. 2(a), DA aptamer was modified with thiol motif at 5’ and Methylene (MB) molecule at 3’ end. The aptamer was assembled on surface of the NGFME through Au-S bonding. While DA was absent in the detection system, the presence of extended single-strand DNA kept MB away from the NGFME and hindered the electron transfer between MB and the gold surface. Therefore, ACV signal of the specific reducing peak of MB at -0.28 V was low (black curve in Fig. 2(b)). While DA was present in the system, DA would bind to its aptamer and cause the formation of secondary structure (Fig. 2(a)), hence the enhanced electron transfer between MB and gold and the increase of -0.28 V ACV signal (red curve in Fig. 2(b)). Therefore, DA detection can be achieved by this micro E-DNA sensor.

Fig. 2.Full-screen View

(Color online) Schematic illustration (a) of the fabrication of micro E-DNA sensor based on NGFME; ACV signals (b) of the micro E-DNA sensor before and after reaction with 1 mM DA; and the "Signal Increase %" response (c) of the micro E-DNA sensor to different concentration of DA aptamer containing in loading buffer.

C. The optimization of loading concentration of DA aptamer

Sensitivity of the E-DNA sensor is related directly with loading density of the probe on the sensor surface [25]. High loading density shall hinder the aptamer folding and increase the background signal. With a low loading density of aptamer, the signal increase caused by the aptamer folding shall not be obvious. Thus, the concentration of loading concentration of P1 was optimized. We defined the signal increase of current at -0.28 V as "signal increase%" = [(It-I₀)/I₀]×100%, where It and I₀ were the current at -0.28 V in presence and absence of DA, respectively. The DA aptamer concentration in loading buffer were 5–1000 nmol/L. From Fig. 2(c), in presence of 1 mM DA, the signal increase is the highest at DA aptamer concentration of 25 nM, which was chosen as the loading concentration of DA apatemer.

D. DA detection of the CFME and micro E-DNA sensor in the interference of AA

Microelectrodes with carbon tip were always employed for monitoring neurotransmitter release from cell [11, 13, 14]. The carbon microelectrode is sensitive and fast responding to DA, but not selective. We tested the response of 1 mM DA in the interference of 1 mM AA of CFME. As shown in Fig. 3(a), both the DA- and AA-generated oxidation curves on CFME and the peaks overlapped with each other. The peak current detected from the mixture of DA and AA was about 2 folds higher than that from DA only. Therefore, selective detection of DA cannot be achieved at the coexistence of AA by CFME. After modifying the CFME with Au deposition and DA aptamer, the electrode responded selectively to DA. As shown in Fig. 3(b), no obvious signal increase response was observed at high concentration of AA (10 mM). What is more, the coexistence of AA with DA did not disturb the detection of DA, either. The result showed great potential for the detection of DA in vivo. The high selectivity of the micro E-DNA sensor was contributed to the high affinity between DA and its aptamer [9, 26, 27] (with the dissociation constant K_d = 0.7 μM in solution [26]).

Fig. 3.Full-screen View

(Color online) Electrochemical behavior (a) of DA and AA at CFME without modification; and comparison (b) of "signal increase%" of the micro E-DNA sensor by incubating with DA and AA. The concentration of DA and AA were 1 mM and 10 mM, respectively.

E. Quantitative detection of DA and the response time

To test the quantitative detection capability of the sensor in the interference of AA, calibration curve of the assay was obtained at different DA concentrations from 25 μm to 5 mM in the coexistence of 10 mM AA. As shown in Fig. 4(a). The signal increased with the DA concentration. A linear response was obtained in 25–250 μM ranges and the linear equation was "signal increase%" = 0.05092c_DA (μM) + 0.1677 with a correlation coefficient of 0.9575. The detection limit of DA was 25 μm. The result illustrated that quantitative and selective detection of DA can be realized even the concentration of AA is several hundred folds higher than that of DA.

Fig. 4.Full-screen View

Signal increase of the micro E-DNA sensor for target DA in the coexistence of 10 mM AA, as function of (a) DA content and (b) reaction time. The inset in (a) is the amplification of liner range from 25 to 250 μm. The data in (b) were taken every 30 s.

The response time of this micro E-DNA sensor to DA was tested through ACV measurement every 30 s. As shown in Fig. 4(b), this micro sensor responded rapidly to DA and exhibited near saturated signaling within 30 s of the addition of DA. The fast response of the sensor showed the enormous potential in real-time application.