Aldehyde N,N-dimethylhydrazone-based fluorescent substrate for peroxidase-mediated assays

Numerous assays based on peroxidase activity have been developed for the detection of analytes due to the various optical peroxidase substrates. However, most substrates are sensitive to light and pH and are over-oxidized in the presence of excess H2O2. In this study, 2-((6-methoxynaphthalen-2-yl)methylene)-1,1-dimethylhydrazine (MNDH), a fluorescent peroxidase substrate prepared from naphthalene-based aldehyde N,N-dimethylhydrazone, was developed. MNDH showed quantitative fluorescence changes with respect to the H2O2 concentration in the presence of horseradish peroxidase (HRP), and the MNDH/HRP assay showed no changes in fluorescence caused by over-oxidation in the presence of excess H2O2. Further, MNDH was thermo- and photostable. Additionally, the assay could be operated over a considerably wide pH range, from acidic to neutral. Moreover, MNDH can be used to detect glucose quantitatively in human serum samples by using an enzyme cascade assay system.


Introduction
Peroxidase activity, which involves the oxidation of organic substrates with the aid of H 2 O 2 , is extensively utilized in various bio-sensors, such as multi-enzyme cascade-based assays and enzyme-linked immunosorbent assays. [1][2][3] The widespread application of peroxidases in sensing systems can be attributed to (1) the large number of optical peroxidase substrates, (2) the very high turnover numbers of these catalyst systems, and (3) the ease of combining peroxidases with other enzymes that produce H 2 O 2 as a byproduct. 4 Horseradish peroxidase (HRP), a natural heme-containing enzyme, is one of the most commonly used enzymes in sensing systems owing to its high specicity, sensitivity, and stability for conjugation with antibodies. [4][5][6] HRP is oxidized to oxHRP by using hydrogen peroxide, which then catalyzes the oxidation of the optical substrate, while oxHRP is reduced to HRP. 7 The resulting changes in the optical properties of the substrate enable qualitative and quantitative analyses of the analyte.
As optical substrates undergo changes in color, uorescence, or chemiluminescence in the presence of peroxidases, these systems have been widely applied in sensing systems. Commonly used peroxidase substrates include colorimetric substrates such as 2,2 0 -azino-bis(3-ethylbenzothiazoline-6sulfonate) (ABTS) and 3,3,5,5-tetramethylbenzidine (TMB), uorescent substrates such as Amplex Red, and chemiluminescent substrates such as luminol. [8][9][10][11] Thus, a wide range of substrates having different optical properties exists, and suitable substrates can be selected in accordance with assay operating conditions such as pH. Colorimetric substrates are mainly used in acidic-to-neutral conditions, and the color changes are detectable by the naked eye. [12][13][14] However, in complex matrices such as biological samples, the colored radical products of ABTS and TMB formed through peroxidasemediated oxidation can react with other antioxidants, resulting in a loss of color. 15,16 In contrast, uorescent and chemiluminescent substrates have higher sensitivities, and thus, lower concentrations can be used. [17][18][19] However, uorescent substrates such as Amplex Red can only be used at pH 7-8 because oxidized Amplex Red, a uorescent resorun, is further oxidized to the non-uorescent resazurin under acidic conditions. 20 The chemiluminescence of luminol is also inhibited at an acidic pH. 21 In addition, in the case of Amplex Red, if a small amount of its product (resorun) is present, photo-oxidation occurs even in the absence of H 2 O 2 , thereby lowering the detection sensitivity. 22 Moreover, these common substrates are excessively oxidized by high concentrations of H 2 O 2 , resulting in a loss of color or uorescence. 23,24 Hence, new uorescent peroxidase substrates must be (1) universally applicable at an acidic-to-neutral pH, (2) photostable, and (3) not overly oxidized by excess H 2 O 2 .

N,N-
Dimethylhydrazone is a well-known protecting group for ketones but is not used to protect aldehydes because complete deprotection is difficult. 25

Materials and instrumentation
Chemical reagents were purchased from commercial sources (Sigma-Aldrich, Tokyo Chemical Industry, and Duksan Pure Chemical, Korea) and used without further purication. Human serum was purchased from Sigma-Aldrich. 1 H and 13 C nuclear magnetic resonance (NMR) spectra were recorded using a JEOL 400 MHz NMR spectrometer. High-resolution mass spectra (HRMS) were recorded using a Bruker Impact II quadruple time-of-ight (QToF) mass spectrometer with an electrospray ionization (ESI) source. Melting point analysis was performed using a Büchi M-560 melting point apparatus. The uorescence spectra were recorded using an Agilent Cary Eclipse uorescence spectrophotometer, and absorbance spectra were recorded on a JASCO V-630 UV-Vis spectrophotometer. Fourier-transform infrared (FTIR) spectra were recorded on a Thermo Scientic NICOLET iS10 spectrometer using a KBr disc (Thermo Scientic, 25 Â 4 mm).

Study of the operating mechanism of the MNDH/HRP system
Combinations of MNDH (20 mM, ACN 5%), HRP (50 mU mL À1 ), and H 2 O 2 (50 mM) were added to an acetate buffer solution (20 mM) of pH 4.0, and the uorescence spectra of samples were recorded at 25 C for 5 min.
For large-scale sample preparation, MNDH (100 mM, ACN 5%), HRP (250 mU mL À1 ), and H 2 O 2 (250 mM) in a pH 4.0 buffer solution (acetate, 20 mM) were placed in a 500 mL volumetric ask. The mixture was extracted three times by using chloroform. The organic layer was then collected and concentrated under reduced pressure, and the oxMNDH was obtained by performing column chromatography using chloroform. 1  MNDH and H 2 O 2 titration using the MNDH/HRP system MNDH (0 to 30 mM, ACN 5%) was added to a buffer solution (acetate pH 4.0 or Tris-HCl pH 7.0, 20 mM) containing HRP (50 mU mL À1 ) and H 2 O 2 (50 mM), and the uorescence spectra were recorded at 25 C at 15 s intervals over 10 min for the pH 4.0 and 60 min for the pH 7.0. H 2 O 2 (0-30 or 20 mM) was added to buffer solution (acetate pH 4.0 or Tris-HCl pH 7.0, 20 mM) containing HRP (50 mU mL À1 ) and MNDH (20 mM, ACN 5%), and uorescence spectra were recorded at 25 C at 15 s intervals over 10 min for the pH 4.0 and 60 min for the pH 7.0.
The initial reaction velocity (V 0 ) was calculated from the initial linear change in the uorescence intensity from 0 to 60 s at pH 4.0 and from 0 to 5 min at pH 7.0. The Michaelis-Menten constant (K m ) and maximum reaction velocity (V max ) were obtained from the Lineweaver-Burk plot.

Thermo-and photostability of MNDH
A batch of MNDH solutions (400 mM, ACN 100%) was incubated for 6 h at 60 C on a hot plate. Another batch of MNDH solutions (400 mM, ACN 100%) was incubated for 24 h in a light-box with a 25 W light-emitting diode lamp (6000 K). Then, the MNDH (20 mM, ACN 5%) incubated on the hot plate or in a light-box was added to the buffer solution (acetate pH 4.0 or Tris-HCl pH 7.0, 20 mM) containing HRP (50 mU mL À1 ) and H 2 O 2 (50 mM), and uorescence spectra were recorded at 25 C for 5 min for the pH 4.0 and 30 min for the pH 7.0.
Serum glucose assay using the MNDH/HRP system Human serum samples were prepared by applying ultra-ltration and diluted by a factor of 37.5. The samples were then added to the buffer solution (Tris-HCl pH 7.0, 20 mM) containing HRP (0.1 U mL À1 ), GOx (0.5 U mL À1 ), and MNDH (20 mM, ACN 5%), and uorescence spectra were recorded at 25 C for 30 min at 30 s intervals.

Results and discussion
Synthesis and characterization of MNDH as the peroxidase substrate MNDH was synthesized in a single step, yielding a white crystalline solid, and the 1 H-and 13 C-NMR spectra, HRMS, and melting point data (Fig. S1-S3 †) conrmed that MNDH had been cleanly synthesized. Aer synthesis, we investigated the pH range over which MNDH can be used as a peroxidase substrate via reaction with H 2 O 2 . Under various pH conditions (pH 4.0 to 9.0), time-dependent changes in the uorescence of MNDH solutions containing HRP were observed in both the presence and absence of H 2 O 2 (Fig. S4 †). In the absence of H 2 O 2 , no change was observed in the uorescence of the solution over time, regardless of the pH. In the presence of H 2 O 2 , the use of a lower pH resulted in a rapid increase in the uorescence intensity of the solution over time. In particular, the uorescence signal was saturated within 3 min at pH 4.0. Although the change in the absolute uorescence intensity of the solution at pH 7.0 was slower and lower than that at pH 4.0, the difference in the uorescence intensity in the presence of H 2 O 2 was approximately 65-times greater aer 5 min than that in its absence (at pH 7.0). This is because the uorescence intensity at 375 nm for MNDH was almost zero at pH 7.0. Consequently, MNDH can be used as a uorescent substrate in acidic-to-neutral solutions.

Mechanistic study of MNDH
To determine the mechanism of the MNDH/HRP system, the changes in the uorescence of MNDH in a pH 4.0 buffer solution containing a combination of HRP and H 2 O 2 were observed. The uorescence spectra of the MNDH solutions containing either HRP or H 2 O 2 were consistent with those of the MNDH solutions without both HRP and H 2 O 2 (Fig. 1). In contrast, for the MNDH solution containing both HRP and H 2 O 2 , the uorescence intensity at 375 nm increased signicantly. Thus, the uorescence of MNDH was affected only in the presence of both HRP and H 2 O 2 . This indicates that oxHRP, which is formed in the presence of H 2 O 2 , catalyzes the oxidation of MNDH, suggesting that MNDH can be used as a new uorescent substrate for peroxidases. To conrm the structure of the oxidized form of MNDH (oxMNDH) formed in the presence of H 2 O 2 and HRP, a large-scale sample was prepared, and 1 H-and 13 C-NMR and FTIR measurements were performed. As shown in the 1 H-NMR spectra ( Fig. S2 and S5 †), the peaks corresponding to the dimethyl and N]CH moieties of the dimethylhydrazone group of MNDH disappear aer oxidation to yield oxMNDH. In addition, in the 13 C-NMR spectra (Fig. S6 †), no peaks corresponding to the dimethyl group are observed. Furthermore, the obtained NMR spectra of oxMNDH are found to be consistent with the NMR spectra of 6-methoxy-2-naphthonitrile ( Fig. S7 and S8 †), 27,28 and the IR spectrum of oxMNDH contains a new peak at 2222 cm À1 , which can be attributed to a C^N group (Fig. S9 †). These results reveal that oxMNDH has the structure of 6-methoxy-2-naphthonitrile (C 12 H 9 NO). Therefore, we conclude that the N,N-dimethylhydrazone moiety in MNDH was oxidized to a nitrile group in the presence of HRP and H 2 O 2 via the Cope elimination reaction per our original design.
Kinetic study of the enzymatic reaction and H 2 O 2 detection in the MNDH/HRP system Next, a kinetic study of the peroxidase reaction was conducted using MNDH as the peroxidase substrate. The samples were prepared by xing the concentration of either MNDH or H 2 O 2 while changing the concentration of the other; these experiments were performed in a pH 4.0 or 7.0 buffer solution containing HRP. Concentration-dependent uorescence of the assay was observed for both MNDH and H 2 O 2 . In particular, as the concentration of MNDH or H 2 O 2 increased, the uorescence intensity of the solution and the initial rate (V 0 ) of the enzymatic reaction increased (Fig. 2, S10 and S11 †). At pH 4.0, MNDH was rapidly oxidized by HRP and H 2 O 2 to the extent that the uorescence intensity was saturated within approximately 3 min, and the uorescence increased signicantly when the MNDH concentration was 20-30 mM (Fig. 2a and S10 †). In addition, for H 2 O 2 concentrations of 0-15 mM, the concentration-dependent change in the uorescence intensity was linear, and the limit of detection (LOD) reached as low as 0.03 mM (Fig. 2b and S12a †). At pH 7.0, the uorescence intensity increased relatively slowly with time (Fig. S11 †). The increase was linear between H 2 O 2 concentrations of 0 and 7.5 mM, and the LOD was 0.06 mM (Fig. 2b and S12b †). As shown by these LOD values, low concentrations of H 2 O 2 can be detected by the MNDH/HRP system. Next, we used the Michaelis-Menten equation to obtain the kinetic parameters. When the reciprocal of V 0 was plotted against the reciprocal of MNDH or H 2 O 2 concentration, linear plots having excellent correlation coefficients (R 2 ; 0.956-0.996) were obtained. As shown in Table 1, K m and V max are 18.14 mM and 0.38 DF s À1 , respectively, for H 2 O 2 , and 0.66 mM and 7.11 DF s À1 , respectively, for MNDH, at pH 4.0. These K m values are much lower than those of other widely used colorimetric substrates (for example,   S10a and b †). Hence, these results indicate that MNDH is a promising new uorescent peroxidase substrate.

Thermo-and photostability of MNDH
The stability of the peroxidase substrate in the presence of light and heat during storage and use is crucial because, on receiving energy from these sources, unstable substrates can undergo a range of reactions that can diminish their uorescence properties. These problems reduce the accuracy and precision of analyte detection, thereby affecting the reliability of the assay. For example, Amplex Red, which is a widely used substrate, has low photostability and requires protection from light during use, making it an inconvenient reagent. To verify the thermostability of our system, the MNDH solution was heated on a hot plate at 60 C for 6 h and then cooled to room temperature. The change in the uorescence of the MNDH solution aer adding it to a combination of H 2 O 2 and HRP in a pH 4.0 buffer solution was then measured at 25 C (Fig. 3). The uorescence signal of the sample containing preheated MNDH was similar to that of unheated MNDH, indicating thermostability. Next, the photostability of MNDH was tested. Aer exposure to 25 W, 6000 K light for 24 h in a light-box, the uorescence of the MNDH solution was measured using the same method as that used for the thermostability experiments; no changes were observed, indicating that the MNDH solution is photostable. Further, in the absence of HRP and H 2 O 2 , the light-irradiated MNDH showed no changes in uorescence, indicating that the MNDH had not been photo-oxidized. These results were also observed at pH 7.0 ( Fig. S13 †). Although MNDH had a lower emission wavelength than previously reported uorescent peroxidase substrates such as Amplex Red, it exhibited comparatively higher thermo-and photostability. [31][32][33][34] Therefore, as a result of its easier storage, MNDH can be used as a substitute for Amplex Red.

Application of MNDH for glucose detection
Finally, a model study based on glucose detection was conducted to conrm if MNDH can be applied as a peroxidase substrate for multi-enzyme cascade assays. Aer glucose was added at various concentrations to a pH 7.0 buffer solution containing GOx, HRP, and MNDH, the uorescence intensity was measured. As the glucose concentration increased, the uorescence intensity increased and, nally, became saturated ( Fig. 4a and S14a †). The plot of uorescence intensity versus glucose concentration showed excellent linearity (R 2 ¼ 0.987) between glucose concentrations of 0 and 20 mM, having a low LOD of 0.31 mM (Fig. S15 †). In addition, the selectivity of the MNDH/GOx/HRP assay for glucose in the presence of other saccharides, including Gal, Fru, Mal, Lac, and Suc, was tested. The concentrations of the other saccharides were maintained at 10 times the concentration of glucose. Aer the samples were prepared in the same manner as the glucose titration, the uorescence intensities of the samples were measured. As shown in Fig. 4b and S14b, † unlike glucose, the other saccharides do not cause any signicant increase in uorescence, indicating that the MNDH/GOx/HRP assay system is highly glucose-specic. The applicability of the MNDH/GOx/HRP assay system to human serum samples was also tested. Human serum samples were purchased from Sigma-Aldrich and prepared by applying ultraltration to remove proteins. The MNDH/GOx/ HRP assay system was used to determine the concentration of  glucose in human serum to be 5.04 AE 0.01 mM, which was nearly identical to that (5.19 AE 0.03 mM) measured by using a glucose meter. Thus, the proposed system can accurately measure the concentration of glucose in human serum, indicating its potential for application in the diagnosis of diseases for which glucose is a biomarker.

Conclusions
MNDH, a new uorescent peroxidase substrate based on the aldehyde N,N-dimethylhydrazone, was developed. The synthesis of MNDH is simple in a single step and requires commercially available reagents. MNDH was oxidized to uorescent oxMNDH containing a cyano group by using HRP and H 2 O 2 , and the MNDH/HRP assay system can be operated at an acidic-toneutral pH. Further, the detection of H 2 O 2 at micromolar levels is possible. MNDH showed considerable thermo-and photostability and was not over-oxidized in the presence of excess H 2 O 2 . Moreover, MNDH can be applied to an enzyme cascade assay system for the detection of glucose in human serum. Thus, MNDH is expected to replace existing peroxidase substrates in various sensing elds.

Conflicts of interest
There are no conicts to declare.