Electrocatalytic miRNA Detection Using Cobalt Porphyrin-Modified Reduced Graphene Oxide

Metalated porphyrins have been described to bind nucleic acids. Additionally, cobalt porphyrins present catalytic properties towards oxygen reduction. In this work, a carboxylic acid-functionalized cobalt porphyrin was physisorbed on reduced graphene oxide, then immobilized on glassy carbon electrodes. The carboxylic groups were used to covalently graft amino-terminated oligonucleotide probes which are complementary to a short microRNA target. It was shown that the catalytic oxygen electroreduction on cobalt porphyrin increases upon hybridization of miRNA strand (“signal-on” response). Current changes are amplified compared to non-catalytic amperometric system. Apart from oxygen, no added reagent is necessary. A limit of detection in the sub-nanomolar range was reached. This approach has never been described in the literature.


Synthesis of Metalated Cobalt Tetra(4-carboxyphenyl)Porphyrin (TCPP).
In a glass tube adapted to microwave, cobalt acetate (36 mg, 3 eq) was added to a solution of TCPP (25 mg, 1 eq) in DMF (5 mL). The reaction medium was microwaved during 20 min at 120 °C (800 W), then the solvent was evaporated under vacuum. The solid was recrystallized in chlorhydric acid. The metalated porphyrin was filtered and washed three times in water, then dried under vacuum.
Probe Grafting and miRNA Hybridization. NH 2 -modified DNA probes (pDNA-29b-1) were covalently grafted on CoTCPP/RGO/GCE in 0.1 M MES buffer containing 150 mM EDC + 300 mM NHS. The reaction was carried out overnight at 37 °C. Electrodes were then washed with water and PBS and incubated in PBS at 37 °C for 1 h to release physisorbed DNA probes.

Methods and Apparatus.
For electrochemical experiments, a conventional one-compartment, three-electrode cell was used with a VMP3 potentiostat (Bio-Logic, Claix, France), the data being collected by EC-lab ® software from Bio-Logic. To prepare the working electrodes, their surface were polished with 0.3 µm alumina slurry on microfiber pads for 1 min. The residual alumina particles were removed by sonication in distilled water then acetonitrile for 30 s, respectively. The electrodes were air-dried before use. The counter electrode was a platinum grid and the reference electrode a commercial saturated calomel electrode-SCE (Metrohm, Villebon, Swisserland). The electrolytic solution was argon-saturated PBS, or aerated PBS for oxygen reduction experiments. Square wave voltammetry (SWV) scans were repeated until complete stabilization of the electrochemical signal (i.e., no difference observed between two successive SWV scans). The metalation reaction was performed with a microwave synthesis reactor Anton Paar Monowave 300 (Graz, Austria).

RGO-Modified Electrodes
Reduced graphene oxide (RGO) was used in order to immobilize CoTCPP through strong π-π interactions [27][28][29][30], and to increase the specific surface area of GC electrodes. First of all, various quantities of RGO were drop-casted on electrodes, and the electrochemical capacitance measured by cyclic voltammetry (Figure 2A).  Figure 2B shows the capacitive charge as a function of the RGO quantity immobilized on the electrodes. The areal capacitance is approximately 300 F g −1 cm −2 .

RGO: CoTCPP-Modified Electrodes
RGO was first modified with CoTCPP by mixing RGO and CoTCPP in a 1:2 water-acetonitrile (vol/vol) mixture under magnetic stirring during 72 h, then isolated by centrifugation and washed first with ethanol then with distillated water. The best result was obtained with 200 µg mL −1 RGO and 1.8 × 10 −4 M CoTCPP, for which the resulting CoTCPP-modified RGO can form a stable suspension in water up to 1.17 mg mL −1 . These conditions were used for the following experiments.
A suspension of 200 µg mL −1 CoTCPP/RGO in ultrapure water (10 µL) was drop-casted on GC electrodes, then let to dry 24 h before use. Figure 3 shows cyclic voltammograms of CoTCPP/RGO-modified, CoTCPP-modified and RGO-modified GC electrodes, under oxygenated or deoxygenated conditions. As previously shown in Figure 1, on bare GC electrode, oxygen electroreduction occurs at low potentials (curve a). On a RGO-modified electrode, reduction starts at potentials slightly higher (Figure 3, curve c); however, the peak maximum occurs below -0.8 V. On a CoTCPP-modified electrode (Figure 3, curve b), the electroreduction peak occurs at -0.26 V. On a CoTCPP/RGO-modified electrode (Figure 3, curve a), the electroreduction wave is shifted 70 mV more positive than without RGO, with a peak current slightly higher. In these experimental conditions, the peak current is limited by oxygen diffusion.

miRNA Detection
To graft pDNA-29b-1 probes, CoTCPP/RGO-modified GC electrodes were dipped into 500 μL of a solution containing 150 mM EDC + 300 mM NHS, at 37 °C for 2 h. After that, the electrodes were washed with ultrapure water and immersed in 500 μL of H 2 O + 10 −7 M pDNA-29b-1 for 2 h at 37 °C, then washed and rinsed with PBS during 45 min at 37 °C under stirring, then with ultrapure water. Hybridization solutions containing target miRNA in PBS (from 10 −11 M up to 10 −9 M) were prepared and heated above the melting temperature of the duplex for 5 min. pDNA-29b-1/CoTCPP/RGO electrodes were dipped into this solution and then kept at the hybridization temperature for 2 h, then washed with PBS at 50 °C. Immediately, SWV was used to characterize hybridization. The main oxygen reduction peak current (situated slightly above -0.2 V vs. SCE) was used as the transduction signal.
As shown on Figure 4, curve b, pDNA-29b-1 probe grafting leads to a relative current decrease of -30% compared to curve a (no grafted probe), while hybridization with the complementary miR-29b-1 strand (10 −9 M) leads to a +20% increase (curve c) compared to curve b. Under the same experimental conditions, but for incubation without any miR target (curve f), the current change is not significant, while one can observe a slight current decrease (-10%) for incubation with a non-complementary target (miR-141, curve i). These experiments, replicated at least three times, gave the following averaged current changes, for a concentration of miR target (when present) of 10 −9 M: grafting (-25 ± 8)%, hybridization (+18 ± 8)%, blank (0 ± 2)%, non-complementary (-10 ± 3)%. One can make the hypothesis that the conformational change [33] induced by the transition from single strand pDNA-29b-1 probe to double-stranded pDNA-29b-1. miR29b-1 hybrid influences the reduction current. This could be due to changes in weak bond interactions between the DNA strands and the porphyrin [20][21][22], or changes in the diffusion kinetic of oxygen at the graphene interface.
Various target concentrations were investigated, from 10 −11 M to 10 −9 M. The calibration curve was constructed from the relative current change, expressed as a percentage (%I peak /I peak ), before and after hybridization, using the following Equation (1): (1) where I Peak , Probe and I Peak , Hyb are the currents corresponding to the SWV reduction peak at -0.2 V before and after hybridization, respectively ( Figure 5).

Conclusions
We have elaborated CoTCPP/RGO-modified electrodes by reducing graphene oxide then functionalizing it by CoTCPP through π-π interactions. These electrodes combine the high electrical conductivity and specific area of RGO with the electrocatalytic properties of cobalt porphyrin towards oxygen electroreduction, in neutral saline solution (PBS). CoTCPP carries carboxylic groups which were used to covalently graft oligonucleotide probes through peptide bonds. In addition, CoTCPP may interact with nucleic acids through weak bonds. Hybridization of this probe with short miRNA targets leads to a change in conformation of these DNA strands, from a random coil structure to a well-organized double-strand. This molecular reorganization may influence the electrocatalytic behavior of CoTCPP towards oxygen reduction through two different ways. First, weak interactions between CoTCPP and the nucleic acids may be broken upon hybridization; secondly, the diffusion of oxygen, which was lowered due to the randomly coiled probes, may be restored upon hybridization. These two phenomena may participate to the current increase which was recorded. To the best of our knowledge, it is the first reported attempt to transduce miRNA. DNA hybridization through an electrocatalytic system based on porphyrin and oxygen reduction. This catalytic approach, conjugated to the use of reduced graphene oxide, allows high current densities despite very low quantity of electroactive material. Detection of miRNA was taken as a possible application example because they are promising disease response biomarkers. For application in concrete cases, optimizations will be necessary (decrease of the detection limit, analysis in complex media, sensitivity to single-nucleotide mismatch).