Manganese Phosphate Self-assembled Nanoparticle Surface and Its application for Superoxide Anion Detection

Quantitative analysis of superoxide anion (O2·−) has increasing importance considering its potential damages to organism. Herein, a novel Mn-superoxide dismutase (MnSOD) mimics, silica-manganous phosphate (SiO2-Mn3(PO4)2) nanoparticles, were designed and synthesized by surface self-assembly processes that occur on the surface of silica-phytic acid (SiO2-PA) nanoparticles. The composite nanoparticles were characterized by fourier transform infrared spectroscopy (FTIR), transmission electron microscopy (TEM), scanning electronic microscopy (SEM), electron diffraction pattern, energy dispersive spectroscopy (EDS) and elemental mapping. Then the electrochemical measurements of O2·− based on the incorporation of SiO2-Mn3(PO4)2 onto the surface of electrodes were performed, and some satisfactory results were obtained. This is the first report that manganous phosphate (Mn3(PO4)2) nanoparticles with shape-controlled, but not multilayer sheets, were utilized for O2·− detection. The surface self-assembly technology we proposed will offer the ideal material to construct more types biosensor and catalytic system for its nanosized effect.

. However, the intrinsic drawbacks of DNA, including high cost, instability, and storage difficulty, may limit their widely applications of electrochemical sensors. Dai also reported the high efficient catalysis of Mn 2 P 2 O 7 , which was used as a SOD mimic for O 2 ·− detection 26 . There is a serious problem in dealing with the preparation of these reported MnSOD mimics. It is that the conventional synthesized MnSOD mimics that reported in the previous literatures have multilayer sheet structure with uncontrolled shape, thickness and size. This approach will bring resources waste and low catalytic efficiency. We wonder how it is possible to utilize surface self-assembly technology and nanotechnology to construct a more efficient MnSOD mimic for promoting analytical properties.
In this paper, SiO 2 -Mn 3 (PO 4 ) 2 NPs were synthesized by surface self-assembly processes that occur on the surface of SiO 2 -phytic acid (SiO 2 -PA). To the best of our knowledge, there are no reports employing surface coating technique to immobilize Mn 3 (PO 4 ) 2 onto the surface of NPs for O 2 ·− detection. The SiO 2 -Mn 3 (PO 4 ) 2 NPs have many advantages, like controllable shape with nanoscale, high specificsurface area than that of nano-sheet structure, low cost, simple preparation process, non-toxic, and so on. This novel MnSOD mimic we prepared is utilized to fabricate biosensors, and the electrochemical measurements of O 2 ·− based on the incorporation of SiO 2 -Mn 3 (PO 4 ) 2 onto the electrodes surface are performed. Figure 1 showed the fourier transform infrared (FTIR) spectroscopy of SiO 2 NPs (a) and SiO 2 -PA NPs (b). For curve (a), the appearance of characteristic peak at 1106 cm −1 and 957 cm −1 were attributed to the O-Si-O bonds stretching vibration, indicating that SiO 2 NPs were successfully synthesized 27 . Compared with unmodified SiO 2 NPs, the SiO 2 -PA NPs illustrated three extra peaks at 2928, 1552 and 695 cm −1 , which should be attributed to -C-NH 2 stretching, symmetric -NH 2 stretching, and the bending vibrations of -NH in APTES, respectively 28 . The results indicated that APTES was successfully modified onto the surface of SiO 2 NPs 29 . More importantly, an adsorption peak at 1092 cm −1 was observed due to the overlap of the characteristic peak of phosphate group (PO 4 3− ) and the peak of asymmetric O-Si-O stretching 30 . The results confirmed that the SiO 2 NPs were successfully modified by APTES and PA.  As shown in Fig. 2a, the Zeta potential of SiO 2 surface was − 38.5 mV, which was attributed to many -OH and other oxygen-containing groups that present in the SiO 2 NPs surface. When modified with APTES, the Zeta potential of APTES-SiO 2 NPs increased to + 22.3 mV that due to the amine groups on the surface of the particles (Fig. 2b). However, the Zeta potential measurements for SiO 2 -PA NPs (Fig. 2c) showed a negative surface charge that owing to the six phosphate groups of PA. When Mn 2+ ions in solution were self-assembled onto the surface of SiO 2 -PA NPs, the zeta potential increased to − 14.1 mV. The change of Zeta potential indicated that SiO 2 -Mn 3 (PO 4 ) 2 NPs were successfully synthesized by self-assembly technology based on the electrostatic interaction that between Mn 2+ ions and the phosphate groups 31 .

Results and Discussion
The TEM and SEM images were also employed to further confirm the formation of SiO 2 -Mn 3 (PO 4 ) 2 NPs. Figure 3A revealed that the spherical SiO 2 NPs were obtained with the average particle size of 75 nm. After surface self-assembly of PA and Mn 2+ sequentially, the two sizes of SiO 2 -PA NPs and SiO 2 -Mn 3 (PO 4 ) 2 NPs showed a slight increase (Fig. 3B,C), respectively. Furthermore, the electron diffraction pattern displayed an amorphous diffraction pattern of Mn 3 (PO 4 ) 2 that deposited on the surface of silica (see the inset from Fig. 3C). And the corresponding elemental mapping of oxygen (O), silicon (Si), phosphorus (P), and manganese (Mn) from the SiO 2 -Mn 3 (PO 4 ) 2 NPs were indicated in Fig. 3D. The energy dispersive spectroscopy (EDS) of SiO 2 -Mn 3 (PO 4 ) 2 NPs showed that the different atomic percentages were 85.32% (O), 13.10% (Si), 1.47% (P), and 0.11% (Mn), respectively. It can be concluded that Mn 3 (PO 4 ) 2 was firmly coated onto the outer surface of the SiO 2 -PA NPs. Mn 3 (PO 4 ) 2 layer has little effect on the size growth of SiO 2 NPs because it was only monolayer of Mn 3 (PO 4 ) 2 molecular that self-assembled onto the outer surface of SiO 2 NPs based on the electrostatic interaction. Here, the stability of Mn 3 (PO 4 ) 2 supported on SiO 2 NPs was evaluated by Zeta potential measurement after three weeks of storage. As was shown in Figure S1, the Zeta potential of the SiO 2 -Mn 3 (PO 4 ) 2 NPs had almost no change after three weeks of storage, indicating the Mn 3 (PO 4 ) 2 NPs have long-term stability. SEM images were also used to investigate the surface texture change after Mn 3 (PO 4 ) 2 coating on SiO 2 NPs, and the particle size of Mn 3 (PO 4 ) 2. Figure S2 showed the SEM images of the SiO 2 NPs, SiO 2 -PA NPs and SiO 2 -Mn 3 (PO 4 ) 2 NPs, respectively. It can be observed from the SEM results that the samples with Mn 3 (PO 4 ) 2 coating can remain its original spherical morphology. Meantime, it can also be seen in this figure that particle size of the samples showed a slight increase after Mn 3 (PO 4 ) 2 coating on SiO 2 NPs, which was in consistence with the results obtained by TEM images as showed  Figure 4A illustrated the synthesis process of SiO 2 -Mn 3 (PO 4 ) 2 NPs. The formational mechanism of this biomimetic enzyme could be explained as follows: After dropping into MnSO 4 solution, PO 4 3− ions, derived from the surface of SiO 2 -PA NPs, were in combination with Mn 2+ ions by electrostatic interaction to form Mn 3 (PO 4 ) 2 . When the PO 4 3− ions were consumed, the monolayer of Mn 3 (PO 4 ) 2 molecular was self-assembly on the outer surface of SiO 2 -PA NPs with controllable morphology. In addition, only aggregated Mn 3 (PO 4 ) 2 particles were observed in the absence of SiO 2 NPs with the same reaction conditions (Supplementary Figure S3).
A schematic drawing of the stepwise construction process of modified glassy carbon electrode (GCE) was described in Fig. 4B. The electrochemical properties of the SiO 2 -Mn 3 (PO 4 ) 2 /Multi-walled carbon nanotubes (MWCNTs)/GCE were investigated by cyclic voltammetry (CV) and Electrochemical Impedance Spectroscopy (EIS) [32][33][34][35] (Supplementary Figure S4). Figure 5A displayed that all fabrication process of SiO 2 -Mn 3 (PO 4 ) 2 / MWCNTs/GCE were carried out by CV in nitrogen saturated phosphate buffered solution (PBS) at a scan rate of 100 mV·s −1 . In the working potential range of 0-0.9 V, there was no electrochemical signal can be observed at the bare GCE (curve a). In contrast, the SiO 2 -Mn 3 (PO 4 ) 2 /GCE exhibited a pair of weakly redox peaks (curve b). When the MWCNTs/GCE was modified with SiO 2 -Mn 3 (PO 4 ) 2 NPs, the oxidation-reduction peaks were more obviously observed (curve c) that due to the excellent electronic conductivity of MWCNTs (Supplementary Figure S5), and the sensitivity of this biosensor was largely improved 36 . Moreover, the peak currents of SiO 2 -Mn 3 (PO 4 ) 2 /MWCNTs/GCE (curve c) were much larger than that of Mn 3 (PO 4 ) 2 /MWCNTs/GCE (curve d).  Results demonstrated that the electro-catalytic effect of SiO 2 -Mn 3 (PO 4 ) 2 was much higher than that of Mn 3 (PO 4 ) 2 aggregated particles. It can be attributed to that the nanosized SiO 2 -Mn 3 (PO 4 ) 2 possessed high specific surface area. As a result, it will help improve the catalytic efficiency of O 2 ·− in the electrolyte. When the bare GCE was only modified with MWCNTs, the background current was more clearly observed (curve e).
To study the catalysis effect of the SiO 2 -Mn 3 (PO 4 ) 2 NPs, the biosensor in PBS and PBS of containing 1.0 μmol L −1 O 2 ·− were measured by CV, respectively. As shown in Fig. 5B, in the PBS containing of 1.0 μ mol L −1 O 2 ·− (curve a), both anodic and cathodic peak currents that corresponding to the redox reaction of in PBS (curve b) clearly increased that can be attributed to the oxidation and reduction of O 2 ·− , respectively 37 . According to the previous reports 38 To further prove this proposed mechanism/reaction, X-ray photoelectron spectroscopy (XPS) analysis was carried out to analyze the composition and chemical configuration of the SiO 2 -Mn 3 (PO 4 ) 2 NPs before and after electrocatalysis process. More details about the XPS spectra of Mn 2p were presented in Figure S6.
The  Figure S8). In addition, the response of SiO 2 -Mn 3 (PO 4 ) 2 / MWCNTs/GCE toward O 2 ·− generated by XAN/XOD was investigated by amperometric measurements 40 . As shown in Figure S9, with successive additions of XAN to the solution, a stepwise increase of the current response was observed.
To evaluate the anti-interference performance of detecting O 2 ·− , the biosensor was examined by successive additions of O 2 ·− and interfering substances into a 0.1 M PBS at 0.484 V. Figure 7A  , respectively (Fig. 7B). Figure S10  ·− . To verify the stability, the biosensor was monitored after being stored for three weeks in a refrigerator. Figure 7C indicated that the current response was no apparent decrease, which was much longer than those obtained for enzyme-based O 2 ·− biosensors [41][42][43][44] . In order to investigate the binding firmness of SiO 2 -Mn 3 (PO 4 ) 2 NPs decorated onto the MWCNTs/GCE electrode, the reuse ability of the SiO 2 -Mn 3 (PO 4 ) 2 /MWCNTs/GCE electrode was tested. Figure S11 displayed the CV curves of the biosensor for 20 cycles, which showed almost overlap curves, indicating the biosensor we prepared had a good cycle stability that can attributed to the good binding state of the SiO 2 -Mn 3 (PO 4 ) 2 NPs and the MWCNTs/GCE electrode. Real-time detection performance of the biomimetic enzyme sensor has also been monitored by detecting O 2 ·− released from the HeLa cells. The amperometric responses of the biosensor were obtained at applied potentials of 0.484 V versus Ag/AgCl in 2 mL 0.1 M PBS (pH 7.4) containing 0.5 × 10 5 cells·mL −1 . After the injection of 4 μ g mL −1 phorbol 12-myristate 13-acetate (PMA), which was reported to generate O 2 ·− from live cells 45 , the current gradually increased at SiO 2 -Mn 3 (PO 4 ) 2 /MWCNTs modified electrode. In this work, PMA was used as a stimulant for the cell to exude O 2 ·− 46,47 . Figure 7D indicated that the strong current signal (0.01457 μ A, curve a) was caused by O 2 ·− released from the HeLa cells, considering that SiO 2 -Mn 3 (PO 4 ) 2 could selectively decomposes O 2 ·− . According to the above linear relationship, the O 2 ·− concentration of 0.0765 μ M was calculated. Thence, the amount of O 2 ·− releasing from per 10 5 cells was calculated to be 0.153 nmol. Meanwhile, in the absence of the HeLa cells and the presence of the treatment of PMA, no obvious current response can be seen on the screen (curve b). To test the effectiveness of this technology, the biomimetic enzyme sensor has also been used to detect the concentration of O 2 ·− in plasma (Supplementary Figure S12).

Conclusion
In this case, the SiO 2 -Mn 3 (PO 4 ) 2 NPs, synthesized via self-assembly technique and nanotechnology, were applied in a biomimetic enzyme biosensor for the detection of O 2 ·− . Results revealed that SiO 2 -Mn 3 (PO 4 ) 2 /MWCNTs/ GCE showed high electrocatalytic activity toward O 2 ·− , lower detection limit and wide detection range. Furthermore, the biosensor that assembled under optimal conditions exhibited high selectivity of O 2 ·− in the presence of related interference, such as H 2 O 2 , UA, AA, DA and Cys. Meanwhile, the long-term stability and good reproducibility of this biomimetic enzyme biosensor were proved. Compared with the Mn 3 (PO 4 ) 2 multilayer sheets, the modified GCE of SiO 2 -Mn 3 (PO 4 ) 2 with high specific surface area exhibited more excellent analytical performance. Consequently, the biomimetic enzyme-free sensor was successfully applied to detecting O 2 ·− that released from live cells, which holds a great promising platform for the reliable monitoring of major diseases in future.  were received from Aladdin Chemistry Co. Ltd (Shanghai, China). Potassium phosphate tribasic trihydrate (K 3 PO 4 ·3H 2 O), manganese sulfate monohydrate (MnSO 4 ·H 2 O) and dimethyl sulfoxide (DMSO) were obtained from Sinopharm Chemical Reagent Co. Ltd. Potassium hyperoxide (KO 2 ) was purchased from Alfa Aesar. Multiwalled carbon nanotubes (MWCNTs) was purchased from Shenzhen Nanotech Port Co. Ltd. Triton X-100, nafion (5 wt% solution in lower aliphatic alcohol), 18-crown-6, phorbol 12-myristate 13-acetate (PMA), dopamine (DA), cysteine (Cys), ascorbic acid (AA) and uric acid (UA) were acquired from Aladdin Sigma-Aldrich Co. (USA). Hydrogen peroxide (H 2 O 2 , 30%) was received from Beijing Chemical Works (China). Phosphate buffer solution (PBS) was obtained by dissolving 8.0 g NaCl, 0.2 g KCl, 1.44 g NaH 2 PO 4 and 0.24 g KH 2 PO 4 in 1000 mL double-distilled water.

Materials
Apparatus. The morphologies of the samples were recorded by transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HITACHI H-7650, Japan). Scanning electron microscope (SEM) images were obtained by a Scanning electron microscope (JSM-6300, Japan). XPS measurements were performed on a Thermo ESCALAB 250 using a monochromic Al X-ray source (1486.6 eV). All the electrochemicals data were measured by CHI 760D electrochemical workstation (Shanghai Chenhua, China). Fourier transform infrared (FTIR) spectra of the SiO 2 NPs and SiO 2 -PA NPs were obtained from a VARIAN Cary 5000 Fourier transform infrared spectrophotometer (VARIAN, USA). Surface potential of the samples was performed by Zeta potential analyzer (Malvern Instruments ZS90). All experiments were carried out using a three-electrode cell equipped, which consisted of a platinum electrode, saturated calomel electrode (SCE) and working electrode. Preparation of SiO 2 -Mn 3 (PO 4 ) 2 NPs. Firstly, SiO 2 NPs were synthesized by the reverse microemulsion method as reported previously by Bagwe 48 . In a typical synthesis, triton X-100 (10.62 g), hexanol (9.6 mL) and cyclohexane (45 mL) were mixed in a 100 mL round-bottomed flask under stirring for 10 min, and then water (2.88 mL) was added to the mixture at room temperature. After being stirred for 0.5 h, NH 3 ·H 2 O (600 μ L) and TEOS (1200 μ L) were dropped into the above clear solution, respectively. Next, the mixture was allowed to stir for a further 24 h at room temperature. The resulting NPs were collected by centrifugation and dried at 60 °C under vacuum condition for 24 h. Then, the modified process of SiO 2 was briefly described as follows: SiO 2 NPs (0.05 g) was dissolved in double-distilled water (20 mL), and TEOS (100 μ L) was added to the SiO 2 suspension with continuously stirring for 30 min at room temperature 49 . Then, the SiO 2 -NH 2 NPs were obtained by feeding appropriate amount of APTES. After stirring for 30 min, 120 μ L of PA/PA sodium salt hydrate buffer solution (pH = 7) was injected into the above solution with continues stirring for 24 h. The resulting NPs were washed with alcohol and double-distilled water. Finally, the SiO 2 -PA NPs were redispersed in water (10 mL), and then MnSO 4 aqueous solution (10 mL, 12 mM) was injected to the round-bottomed flask containing the SiO 2 -PA NPs under constant stirring for 1 h at 60 °C. After completion of the reaction, the obtained products were collected by centrifugation, washed with double-distilled water, and dried in a vacuum oven at 60 °C for 24 h.
Generation of superoxide anion. A stable O 2 ·− solution was prepared by dispersing KO 2 to DMSO (containing 18-crown-6). In accordance with the molar absorptivity of O 2 ·− in DMSO, the concentration of O 2 ·− was monitored by recording the absorbance of ferricytochrome c spectrophotometrically at 550 nm 50 . In particular, spectrophotometric measurement of the amount of ferricytochrome c that reduced by O 2 ·− referred to the following reaction: cytochrome c (Fe III ) + O 2 ·− = ferrocytochrome c (Fe II ) + O 2 , in which ferrocytochrome c exhibits a strong absorbance at 550 nm 51 . The linear relationship of the absorbance vs the ferrocytochrome c concentration was depicted in Figure S13. The linear equations were A 550 = 20.6c − 0.0044, R = 0.9986. The concentration of O 2 ·− can be calculated by the concentration of the formed ferrocytochrome c according to the above reaction formula 52 .