MOF-Based G−Quadruplex/Hemin DNAzymes for Cascade Reaction

Abstract

DNA-based biomimetic enzymes have attracted extensive attention due to their unique structure and stability compared to natural enzymes. Meanwhile, the specific sequences of DNA itself also have a catalytic effect. Herein, we first designed three guanine-rich DNA sequences numbered c−Myc3c, PG4TC, and TTT to construct g−quadruplex/hemin DNAzymes. Then, the g−quadruplex/hemin DNAzymes with the best activity were selected by a comprehensive examination of activity, degradation rate, and affinity. Subsequently, the stability and reusability of UiO66−DNAzymes were investigated using UiO66 as the carrier to immobilize DNAzymes. The results showed that UiO66−DNAzymes had excellent reusability and stability. Finally, UiO66−DNAzymes were successfully used for glucose detection by cascading with glucose oxidase (GO_x) with a detection limit of 0.62 μM. The constructed glucose sensor had a good specificity, which is of great significance for developing a novel, accurate, fast, and economical glucose detection sensor.

1. Introduction

Natural enzymes are efficient biocatalysts with mild reaction conditions and high selectivity specificity. However, the stability and activity of natural enzymes are difficult to reach the level of practical application [1,2]. Therefore, it is urgent to study biomimetic enzymes with good stability and low cost to meet the needs of practical production and application. Natural enzymes tend to be susceptible to inactivation in cascade reactions, while biomimetic enzymes have become ideal catalysts to replace natural enzymes [3,4,5]. Particularly, deoxyribozymes (DNAzymes) are single-stranded DNA molecules with enzymatic catalytic activity and structure recognition ability obtained by in vitro screening technology [6]. DNA molecules provide a rich reaction microenvironment due to their unique helical structure, hairpin structure, tetrad structure, and various hair non-canonical structures [7,8,9,10]. DNAzymes have numerous advantages compared to natural enzymes, such as easy access, strong hydrolytic stability, resistance to extreme environments, etc. Thereupon, DNA-based mimetic enzymes have attracted extensive interest [11]. Nevertheless, the catalytic activity of early developed DNAzymes was not satisfactory, so it is crucial to develop methods to improve the catalytic activity.

G−quadruplex/hemin DNAzymes are also known as peroxide-mimicking DNAzymes [12,13]. They are composed of a guanine-rich DNA sequence folded into a quadruplex structure and have peroxidase catalytic activity in the presence of hemin [14]. Different DNA sequences affect how g−quadruplexes bind to hemin, thereby affecting the activity of DNAzymes [15,16]. Therefore, the design of specific DNA sequences is one of the directions to optimize the activity of DNAzymes. Moreover, immobilized DNAzymes are another means to improve the activity of DNAzymes. Enzyme immobilization is one of the most attractive methods to improve enzyme performance due to the inherent fragility and difficult recovery of natural enzymes [17]. Metal-organic frameworks (MOFs) have the advantages of flexible pore structure, abundant coordination sites, and diverse functional groups [18,19]. The structural characteristics of MOFs make them the preferred materials for enzyme immobilization, which can be used in biosensors and other fields. Therefore, MOFs are an ideal choice to immobilize DNAzymes. Currently, there are many studies about embedding DNAzymes on MOFs to improve the activity of DNAzymes [20,21,22,23], but all of them were used in electrocatalysis and have few biological applications.

Peroxidase can oxidize chromogenic substrates such as ABTS and TMB in the presence of H₂O₂ [24]. Based on this, researchers have developed various biosensors for detection applications, including fluorescent g−quadruplex DNAzyme sensors, colorimetric g−quadruplex DNAzyme sensors, electrochemical g−quadruplex sensors, etc. [25]. However, there was no report on g−quadruplex sensors for glucose detection to our knowledge. Glucose, as an important intermediate involved in human metabolism, is closely related to various human diseases [26]. Therefore, the importance of developing an accurate, fast, facile, and economical sensor to detect glucose is self-evident.

Herein, we first designed three DNA sequences and screened out the g−quadruplex/hemin DNAzymes system with high activity through a series of optimization. Then, the DNAzymes were immobilized on UiO66 to improve their stability and catalytic ability. Subsequently, UiO66−DNAzymes were cascaded with GO_x to construct an efficient, accurate, and cost-effective biosensor platform for glucose detection. This work constructed a cascade catalytic platform between DNA-mimetic enzymes and natural enzymes, which provided an inspiring general strategy for glucose detection.

2. Results and Discussion

2.1. Design and Characterization of G−Quadruplex/Hemin DNAzymes

The preparation process of g−quadruplex/hemin DNAzymes is shown in Figure 1a. In the presence of K⁺, ssDNA molecules fold into a parallel g−quadruplex structure. After adding hemin, the g−quadruplex will form peroxidase-activated DNAzymes with hemin. Peroxidase enzyme can oxidize ABTS to generate chromogenic ABTS⁺ in the presence of H₂O₂. To verify the peroxidase activity of DNAzymes, ABTS and H₂O₂ were added to DNAzyme solution, g−quadruplex DNA solution, or hemin solution, respectively, and then the color change of the solution was observed. As shown in Figure 1b, the DNAzyme solutions appeared blue in the presence of H₂O₂, but g−quadruplex DNA solutions or hemin solutions were colorless. This result proved that DNAzymes did have peroxidase activity, while g−quadruplex DNA or hemin had no peroxidase activity. At the same time, it also verified the successful formation of g−quadruplex/hemin DNAzymes.

Then, three DNA sequences were designed to optimize the activity of DNAzymes (Table S1). By studying the initial reaction rate and enzyme kinetics of the three sequences, the g−quadruplex/hemin DNAzymes with higher activity were screened out. As shown in Figure 1c, the initial reaction rate of c−Myc3c was 259.72 nM/s, which was much higher than that of PG4TG and TTT. The enzymatic kinetic parameters of the three DNA sequences were further studied (Figure S1). The V₀ and H₂O₂ concentration curves were fitted by using the Michaelis–Menten model, and the specific enzyme kinetic parameters are summarized in

Table 1. The kinetic parameters of the three DNA sequences.

Sequences Name	Km (mM)	Vmax (nM s−1)	kcat (s−1)
c-Myc3c	0.4367	463.7	1.016
PG4TC	0.5711	394.2	0.6902
TTT	0.5097	253.0	0.4964

. According to the data in Table 1, the K_m of the c−Myc3c sequence was the smallest among the three sequences, indicating that the c−Myc3c had the highest affinity for the substrate. Meanwhile, the V_max and k_cat of the c−Myc3c sequence were much higher than the other two sequences, showing that the c−Myc3c had the most excellent activity among the three sequences.

The stability of the enzyme is significant for practical applications, so the initial degradation rate of DNAzymes synthesized from three DNA sequences in the presence of 0.5 mM H₂O₂ was investigated. To obtain the specific degradation rate, the extinction coefficient of DNAzymes should be acquired first. DNAzymes have a characteristic absorption peak at 404 nm, while hemin also has an absorption peak at the same wavelength. To avoid interference with DNAzymes by hemin, hemin should be fully saturated with the g-quadruplex sequences. Therefore, the DNAzyme was constructed with a hemin concentration of 2 μM and a DNA concentration of 5 μM. As shown in Figure 1d, the calculated extinction coefficients of c−Myc3c, PG4TC, and TTT were 1.265 × 10⁵ M⁻¹cm⁻¹, 1.005 × 10⁵ M⁻¹cm⁻¹, and 1.270 × 10⁵ M⁻¹cm⁻¹, respectively. According to Figure 1e, the DNAzymes of the three sequences started to degrade in the presence of 0.5 mM H₂O₂. The slowest degradation rate was TTT among the three sequences, with a degradation rate of 1.030 nM s⁻¹. The initial degradation rates of c-Myc3c and PG4TC were 1.212 nM s⁻¹ and 1.318 nM s⁻¹, respectively. The specific data were listed in

Table 2. The degradation and dissociation constant of the three DNA sequences.

Sequences Name	εcomplex (M−1cm−1)	Vd (nM/s)	Kd (μM)
c−Myc3c	1.265×105	1.212	3.683
PG4TC	1.005×105	1.318	5.820
TTT	1.270×105	1.030	9.239

.

To assess the affinity magnitude of the three g−quadruplex sequences to hemin, the dissociation constant K_d of the DNAzymes was determined. The absorbance change of DNAzymes at 404 nm was recorded by titrating DNA solutions with different concentrations (0–10 μM) into 5 μM hemin, and K_d was reckoned by fitting a curve of α versus DNA concentration. Figure 2 showed the changes in the titration spectra of three g−quadruplex sequences, and the specific K_d statistics were shown in Table 2. The affinity of the three g−quadruplex sequences to hemin was c−Myc3c, PG4TC, and TTT in descending order. Combined with the previous activity experiments, the affinity between the g−quadruplex sequences and hemin was positively correlated with the formation of DNAzymes. Therefore, considering the catalytic activity, stability, and affinity, the c−Myc3c sequence was finally selected for subsequent immobilization studies.

2.2. UiO66−DNAzymes Immobilization

Subsequently, the DNAzymes with a c−Myc3c sequence were immobilized on UiO66 (Figure 3a) and the properties of the immobilized DNAzymes were studied. To demonstrate the successful immobilization of UiO66−DNAzymes, TG was used to characterize UiO66 and UiO66−DNAzymes (Figure 3b). The UiO66−DNAzymes showed a slight weight loss below 100 °C, which was not seen in the UiO66 curve. When heated to 800 °C, UiO66 retained 43% of its initial weight, which was slightly higher than UiO66−DNAzymes. The breakdown of DNAzymes was responsible for the excess loss in UiO66−DNAzymes, indicating that DNAzymes were successfully immobilized on UiO66. FTIR was further performed to investigate the relationship between immobilized UiO66−DNAzymes. As illustrated in Figure 3c, the peak of UiO66−DNAzymes changed slightly but not noticeably, which could be attributed to the small amount of immobilized DNAzymes. To better observe the morphology, particle size, and dispersion of UiO66 and UiO66−DNAzymes, the UiO66 and UiO66−DNAzymes were characterized by SEM and TEM. As shown in Figure 3d, the morphology of UiO66 was an ortho-octahedron with about 500 nm. Likewise, the morphology, dispersibility, and particle size of UiO66−DNAzymes were unchanged (Figure 3e), indicating that DNAzymes had no effect on UiO66. Subsequently, the optimal immobilized ratio was attained by varying the amount of UiO66 (0–4 mg). As shown in Figure 3f, the absorbance of the supernatant at 404 nm decreased significantly with the addition of UiO66, indicating that most of the DNAzymes were immobilized on UiO66. The immobilization yield was up to 92% after adding 2 mg of UiO66. However, the immobilization rate of the DNAzymes increased slowly with the further amount of UiO66 (Figure 3g). Considering the economy, 2 mg UiO66 was finally selected for immobilization.

Subsequently, the storage stability and reusability of the immobilized DNAzymes were investigated. As shown in Figure 4a, the free DNAzymes were inactivated after 4 days of storage, while the UiO66−DNAzymes retained 46.4% activity at this time. Remarkably, UiO66−DNAzymes still maintained 29.25% activity after 10 days. From Figure 4b, the UiO66−DNAzymes preserved approximately 50% activity after repeated use 7 times. Furthermore, UiO66−DNAzymes still retained high activity in different temperatures and pH (Figure 4c,d), displaying the robust stability of the UiO66−DNAzymes. These results all indicated that UiO66 had a protective effect on the DNAzymes structure, preventing DNAzyme decomposition.

Then, the enzymatic kinetics of UiO66−DNAzymes were studied, and the specific enzyme kinetic parameters were shown in Table S2. The K_m of UiO66−DNAzymes was higher than that of free DNAzymes, indicating that immobilization reduced the affinity of DNAzymes to a substrate. DNAzymes were adsorbed on the UiO66 surface, making them more difficult to contact the substrate with greater resistance. At the same time, the V_max of immobilized DNAzymes was smaller than that of free DNAzymes, which also proved the above conclusion.

2.3. Glucose Detection

With the increase in glucose-related diseases, the development of methods to detect glucose has been a research hotspot. Therefore, an efficient, accurate, and economical sensor for glucose detection was constructed by cascading UiO66−DNAzymes and GO_x. The response of the glucose sensor was investigated in HEPES buffer (pH = 7.5) using different glucose concentrations (0–500 μM). As shown in Figure 5a, the detection limit of the constructed sensor was 0.62 μM. Then, to verify the selection specificity of the sensor, 500 μM glucose was replaced with equal concentrations of other sugars (xylose, fructose, mannose, lactose) and reacted with BSA. The selection specificity of the sensor was assessed by the absorbance change of the supernatant at 420 nm. It could be seen that the absorbance of other interfering substances was extremely low (Figure 5b), showing that the constructed glucose sensor had excellent specificity.

3. Experimental

3.1. Materials

Zirconium (IV) tetrachloride (ZrCl₄, 99%), terephthalic acid (H₂BDC, 98%), N, N-dimethylformamide (DMF, ≥99.9%), and glucose were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Hemin (AR), dimethyl sulfoxide (DMSO), GO_x, and H₂O₂ were purchased from Shanghai yuanye Bio-Technology Co., Ltd. Hydrochloric acid (HCl, 36.0–38.0 wt%) was obtained from Damao (Tianjin, China).

3.2. Characterization

Transmission electron microscopy (TEM, Hitachi H-800) and scanning electron microscopy (SEM, JSM-5600LV) were used to observe the morphologies and structures of UiO66 and UiO66−DNAzymes. Fourier transform infrared spectroscopy (FTIR, Nicolet model 205 spectrometer) was performed to characterize the functional groups of UiO66 and UiO66−DNAzymes. Thermogravimetric (TG) analysis was undertaken on a STA7300 Thermal gravimetric analyzer (Hitachi HTIACHI) heating from 25 °C to 800 °C in air at a rate of 10 °C/min.

3.3. Preparation Methods

3.3.1. Preparation of G−Quadruplex/Hemin DNAzymes

The three DNA sequences were designed according to the literature [15,16,27] and are shown in Table S1. DNA stock solution was prepared in a HEPES buffer (50 mM, pH 7.5) and stored at 4 °C. The diluted DNA solution was added to a HEPES buffer containing 100 mM KCl and the g−quadruplex structure was obtained after a 5 min reaction at 95 °C. Then, DNA and hemin were dissolved in DMSO at a molar concentration of 1:2, and the g−quadruplex/hemin DNAzymes were produced after incubation for 1 h at room temperature.

3.3.2. Preparation of UiO66

UiO66 was prepared by the conventional hydrothermal method [18]: ZrCl₄ (104 mg) and H₂BDC (72.5 mg) were dissolved in 100 mL DMF. Then, a little hydrochloric acid was added to ultrasonicate for 10 min. The solution was placed in a Teflon-lined autoclave, followed by a hydrothermal reaction at 130 °C for 10 h. The products were centrifuged and washed three times with ethanol and water to remove the unreacted substrate.

3.3.3. Preparation of UiO66−DNAzymes

The above obtained UiO66 (2 mg) was added to the prepared DNAzyme solution and immobilized at room temperature for 1 h at 800 rpm. The precipitate was collected by centrifugation and washed three times with HEPES buffer to remove the non-immobilized DNAzymes. The amount of UiO66 was changed (0–4 mg) to optimize the immobilization ratio.

3.4. Enzymatic Properties Determination

3.4.1. Enzyme Activity

Enzyme activity was calculated by measuring the ability of the g−quadruplex/hemin DNAzymes to oxidize ABTS. A total of 0.5 mM H₂O₂ was added to the prepared DNAzyme solution to oxidize ABTS (0.5 mM), and then the absorbance change of the solution was measured at 420 nm. The molar absorption coefficients were 36,000 M⁻¹ cm⁻¹ for ABTS⁺ [14]. The standard deviations were calculated from three parallel experiments.

The enzyme kinetic constant was determined by changing the H₂O₂ concentration (0–5 mM). The enzyme kinetic curve of V₀ and H₂O₂ was nonlinearly fitted to obtain the Michaelis constant K_m and the maximum velocity V_max. k_cat was calculated using the formula k_cat = V_max/[catalyst concentration].

3.4.2. DNAzymes Degradation Rate

H₂O₂ (0.5 mM) was added to the prepared DNAzyme solution, and the change of absorbance was observed at 404 nm within 30 min. The initial degradation rate V_d was calculated by Equation (1).

$$V_d =\frac{\left|{\mathit{dC}}_{\mathit{complex}}\right|}{\mathit{dt}}=\frac{\left|\mathit{dA}\right|}{dt\times b\times {\epsilon }_{\mathit{complex}}}$$

(1)

where complex is DNAzymes, dt is the change of time, dA is the change of absorbance at 404 nm, b is the optical path, and ε is the extinction coefficient of the DNAzyme at 404 nm.

3.4.3. DNAzymes Dissociation Constant (K_d)

DNAzymes were prepared by titrating different concentrations of DNA sequences (0–10 μM) to 5 μM hemin solution, and the change of absorbance between 300–700 nm was recorded by UV-vis. The fraction of hemin (α) was calculated by Equation (2). K_d was obtained by fitting a curve of α versus DNA concentration.

$$\alpha =\frac{A_f-A}{A_f-A_b}$$

(2)

where A_f is the absorbance of hemin alone at 404 nm, A_b is the absorbance of DNAzymes saturated with hemin at 404 nm, and A is the absorbance when DNA was added.

3.5. Enzyme Stability

Storage stability: free and immobilized DNAzymes were stored at 4 °C for 10 days, and the changes in enzyme activity were recorded at 2, 4, 6, 8, and 10 days, respectively. Maximum enzyme activity was set to 100%. The three parallel experiments were performed.

Reusability: the reacted UiO66−DNAzymes were centrifuged to collect the precipitate, washed three times with a HEPES buffer, and then the substrate ABTS and H₂O₂ were added to react again. The enzyme activity was taken to be 100% the first time.

Temperature stability: free and immobilized DNAzymes were incubated at a different temperature (20–70 °C) for 1 h, and other conditions remained unchanged. The enzyme activity was measured according to method 3.4.1. The maximum enzyme activity was set to 100%. The three parallel experiments were performed.

pH stability: free and immobilized DNAzymes were incubated in different pH (3–10) for 1 h, and other conditions remained unchanged. The enzyme activity was measured according to method 3.4.1. The maximum enzyme activity was set to 100%. The three parallel experiments were performed.

3.6. Glucose Detection Application

The total reaction system was 1 mL. The prepared UiO66−DNAzymes were added to a HEPES buffer containing 100 μg/mL GO_x, 1.5 mM ABTS, and 0–500 μM glucose for 30 min. Then, the solution was centrifuged and the absorbance of the supernatant at 420 nm was measured by a UV-vis spectrophotometer. The three parallel experiments were performed.

Selection specificity: the total reaction system was 1 mL. The prepared UiO66−DNAzymes were added to a HEPES buffer containing 100 μg/mL GO_x, 1.5 mM ABTS, and 500 μM of other sugars (xylose, fructose, mannose, lactose) for 30 min. Then, the solution was centrifuged and the absorbance of the supernatant at 420 nm was measured by a UV-vis spectrophotometer.

4. Conclusions

In summary, DNAzymes were constructed with hemin based on three guanine-rich DNA sequences, c−Myc3c, PG4TC, and TTT. Comprehensively examining the activity, degradation rate, and affinity of the three sequences, the results showed that the peroxidase activity and affinity with the substrate of the DNAzymes by the c−Myc3c sequence were much higher than those of the other two sequences. Then, the DNAzymes numbered c−Myc3c were immobilized on UiO66, and the stability and reusability of UiO66−DNAzymes were improved. Finally, an efficient, accurate, and economical sensor for glucose detection was successfully constructed by cascading UiO66−DNAzymes and GO_x. Using different glucose concentrations (0–500 μM), the response of the glucose sensor in a HEPES buffer (pH = 7.5) was studied. The detection limit of the constructed sensor was 0.62 μM, and the glucose sensor also had excellent specificity. Our study explored a new direction for DNAzyme catalysis and offered a facile and versatile biosensor platform for glucose detection. Considering that the base sequences affect the catalytic activity of enzymes, exploring the design of specific DNA sequences seems limitless.