Glucose oxidase immobilized on ZIF-7-III: composite formation, optimization and integration in an electrochemical biosensor for selective glucose detection

Abstract

Coordination polymers (CPs) can be used as supporting materials for enzyme immobilization to overcome limitations arising from mass transfer barriers, active site blocking, and low enzyme loading, as previously demonstrated, employing metal–organic frameworks (MOFs). We report on a new composite containing glucose oxidase (GOD) and ZIF-7-III (Zn(bIm)₂), focusing on three aspects: formation, optimization of catalytic properties, and application. The immobilization of GOD on ZIF-7-III at room temperature was systematically studied by varying the amount of GOD and the reaction time. ZIF-7-III/GOD composites were obtained, as confirmed by Fourier transform infrared (FTIR) spectroscopy and thermogravimetric analysis (TGA). GOD incorporation significantly slows the crystallization of ZIF-7-III. The optimized composite exhibits relative catalytic activity of 98% at maximum enzyme loading, enhanced stability across a broad pH range (3 < pH < 9), and stability at elevated temperatures (up to 80 °C). Storage stability retains 66% activity after 60 days, which is important for possible applications. The use of the composite in an electrochemical biosensor for glucose detection was also demonstrated. Cyclic voltammetry and amperometric measurements were performed, demonstrating high repeatability (RSD <6%) and selectivity against common interferents. A linear response to glucose concentrations in the millimolar range was observed, with a detection limit of 0.01 mmol L⁻¹. Notably, the sensor effectively detected glucose in human plasma samples, indicating its potential for real-world glucose monitoring.

---

Introduction

Early detection of diabetes is becoming increasingly important due to the growing proportion of the population affected by this condition. Glucose sensors can be useful tools in both detecting the disease and detecting increases in glucose concentrations due to hyperglycemia in general. Thus, fast and precise glucose detection is mandatory in order to monitor irregular blood glucose concentrations.^1–5 One method for the detection of glucose is the use of enzyme-based electrochemical biosensors. This method was established in recent years due to its high sensitivity to low glucose concentrations combined with a high analyte selectivity as a result of the high specificity of the enzyme.^6–9 Glucose biosensors are based on the use of glucose oxidase (GOD) catalyzing the oxidation of glucose to gluconic acid and hydrogen peroxide.^10,11 Despite its widespread use, the use of GOD for glucose detection is limited by the low thermal and chemical stability of the enzyme.^12,13 To address these limitations in the utilization of GOD in biotechnological processes, various immobilization strategies have been developed to preserve the biological activity of the enzyme under various conditions. Immobilization methods not only increase the thermal and chemical stability of the enzyme but also allow the enzyme to be recovered.^14,15 Within the last few decades, various support materials for enzyme immobilization were developed, including organic and inorganic materials such as biopolymers, metal oxides, and carbon.^16–19 The selection of an immobilization technique must take different factors into account such as cost, reagent toxicity, and enzyme leaching.^16,20 Also, the activity and the stability of enzymatic biosensors can be influenced by the immobilization method. The same is true for the sensitivity, which may decrease due to enzyme denaturation and conformational alterations, especially at the active site of the enzyme induced by immobilization.²¹ Immobilization of the enzyme can be performed either in one step or by a two-step method. In the latter, the enzyme is adsorbed in or on a preformed porous support material or covalently attached to the support. Adsorption in pores is thus limited by the pore size of the support material, while the synthesis of support materials with suitable functional groups on the surface can make covalent attachment more challenging. The advantage of the two-step method is its versatility, allowing a wide variety of support and enzyme combinations, although adsorbed enzymes may leach more easily. In contrast, in the one-step method, the enzyme is entrapped or encapsulated during the formation of the support material, i.e. in situ, and a composite is formed. While leaching is a minor problem, this method faces immense synthetic challenges in finding synthesis conditions to form the support material while avoiding enzyme degradation, i.e. temperatures below 35 °C, short reaction times in the range of minutes to a few hours, and neutral pH which is highly dependent on the enzyme to be incorporated.^22,23 Additionally, reactants must be carefully selected to avoid denaturation of the enzymes. Metal–organic frameworks (MOFs), also known as porous coordination polymers, can be obtained under mild reaction conditions using starting materials that are compatible with enzyme stability. Immobilization by entrapment or encapsulation using these compounds is advantageous since they offer potentially high accessibilities and thus catalytic activities of the immobilized enzymes due to their high specific surface area and the presence of pores.²⁴ The accessibility of the enzyme can be further improved by structural defects, changes in particle morphology and size, or by using etching processes leading to an increase in porosity.^25,26 For example, Hu et al. synthesized enzyme-MOF composites using the zeolitic imidazolate framework ZIF-8 as the support. Enhanced substrate accessibility and enzyme activity were attributed to structural defects in the enzyme–MOF composites.²⁷ Due to the limited stability of enzymes and reaction conditions where MOFs are formed, most of the reported studies used ZIF-8 and MIL-100 as the support.²⁸ A ZIF system that has been well studied and forms different framework structures is ZIF-7. Three different phases are formed, which are known as ZIF-7-I, ZIF-7-II, and ZIF-7-III. All of the phases have the same network composition [Zn(bIm)₂], bIm⁻ = benzimidazolate, which contains Zn²⁺ ions that are tetrahedrally coordinated by four bIm⁻ ligands. Each bIm⁻ ligand is bridging two Zn²⁺ ions through a bis-monodentate linkage.²⁹ These tetrahedra are interconnected to a 3D framework in the metastable phases ZIF-7-I and -II. In water, the nonporous, thermodynamically stable phase ZIF-7-III is formed that contains layers of strongly distorted ZnN₄ tetrahedra. ZIF-7-III exhibits a layered crystal structure and is the densest of the three ZIF-7 phases.^29–32

In this article, we present the formation of ZIF-7-III/GOD composites using a one-step (in situ) synthesis, optimization of the enzyme loading, as well as the catalytic performance of the composites in the oxidation of glucose. The use of the composite as the active material in an electrochemical biosensor is also presented, demonstrating glucose detection with good sensitivity and high selectivity (Scheme 1).

Scheme 1 Outline of the publication which reports the green synthesis of the ZIF-7-III/GOD composite (top), its catalytic properties as a catalyst for converting glucose to gluconic acid and hydrogen peroxide (middle), and the integration in a biosensor for glucose detection (bottom).

Experimental

Materials

Acetic acid (Merck, ≥99%), aqueous ammonia solution (Merck, 32%), benzimidazole (HbIm) (Merck, >99.5%), Coomassie Brilliant Blue G-250 (Merck), D(+)-glucose-monohydrate (Merck), 3,5-dinitro salicylic acid (Sigma Aldrich, 98%), ethanol (Merck, >96%), glucose oxidase from Aspergillus niger (type II, Sigma Aldrich GmbH, 15 000–25 000 units per g solid), phosphoric acid (Merck, 85%), potassium sodium tartrate (Merck, 99%), sodium acetate (Merck, >99.5%), sodium hydroxide (Merck, >98%) and zinc nitrate hexahydrate (Merck, ≥99%) were commercially obtained and used without further purification.

Analytical techniques

Powder X-ray diffraction (PXRD) measurements were performed on a Bruker D8 Advance diffractometer with Cu Kα₁ radiation (λ = 1.5406 Å) in the range of 5° to 35°. The average d-spacing was calculated by using the Bragg equation.^31,33 Fourier transform infrared (FT-IR) spectra (4000–350 cm⁻¹) were recorded using a JASCO FTIR 6300 spectrometer. Thermogravimetric analysis (TGA) was carried out on a Linseis STA PT 1600 instrument under a dynamic air atmosphere (flow rate = 6 ml min⁻¹) with a heating rate of 8 °C min⁻¹ from 25–600/800 °C. The UV-visible absorbance was measured using a UV/Vis spectrophotometer (160 u Shimadzu). Field emission scanning electron microscopy (FE-SEM) images were recorded on a TSCAM MIRA3 instrument. Electrochemical measurements were performed using an IM6x electrochemical station (Zahner-Elektrik GmbH & Co. KG, Germany) and a computer-controlled potentiostat (Autolab electrochemical analyzer model PGSTAT30, Eco Chemie, The Netherlands). A conventional three-electrode system was used, including an Ag/AgCl (saturated KCl) electrode as the reference electrode, a platinum wire as the counter electrode, and a bare or modified glassy carbon electrode (GCE) as the working electrode.

Optimized synthesis of ZIF-7-III

ZIF-7-III was prepared according to the previously reported procedure with some modifications.³⁴ Briefly, 0.118 g (1 mmol) of HbIm was combined with 29 mL of deionized water and stirred for 5 minutes at room temperature. Subsequently, 0.148 g (0.5 mmol) of Zn(NO₃)₂·6H₂O was introduced to the suspension. Following 10 minutes of stirring, 0.1 mL of 32% ammonia solution (equivalent to 1.656 mmol NH₃) was added to the reaction mixture, resulting in the immediate formation of a milky suspension. Finally, after two hours, the white solid (ZIF-7-III) was separated by centrifugation (10 000 rpm, 5 min), washed three times with deionized water, and dried at room temperature.

Synthesis of the ZIF-7-III/GOD composite

The ZIF-7-III/GOD composite was obtained by including the aqueous GOD solution in the process of ZIF-7-III synthesis. Different amounts of GOD ranging from 5 to 30 mg were used for the composite formation. In the following procedure, only the synthesis of the composite that provided the highest catalytic activity is described.

In the optimized synthesis, the aqueous GOD solution (4 mL, 5 mg GOD mL⁻¹) was added to the HbIm suspension (0.118 g (1 mmol) suspended in 25 mL deionized water. Then, 0.148 g of Zn(NO₃)₂·6H₂O (0.5 mmol) was introduced into the mixture, followed by stirring for 10 minutes. After adding 0.1 mL of a 32% ammonia solution (1.653 mmol NH₃), the reaction mixture was stirred vigorously at room temperature for 2 h. Ultimately, the yellowish precipitate was collected through centrifugation and rinsed three times (each time with 8 mL of deionized water). The remaining precipitate was allowed to dry at ambient temperature while the supernatants were preserved to determine the enzyme loading. The amount of immobilized GOD was calculated using the Bradford method, and details are given in Section S1.1.³⁵ Before measuring the loading, the composite material was sonicated in an ice bath with a 50% duty cycle for 30 min to remove physically adsorbed enzymes.

Results and discussion

The synthesis, characterization and optimization of a catalytically active ZIF-7-III/GOD composite are shown and its stability under harsh conditions, i.e. changes in pH, temperature, or storage time, is demonstrated, which allows integration of the composite in a biosensor. Accordingly, the results and discussion are divided into three major sections.

In situ immobilization of GOD on ZIF-7-III was accomplished via a room temperature synthesis procedure in an aqueous solution using zinc nitrate and benzimidazole. The pH of the solution was adjusted by the addition of aqueous NH₃. In addition, the effect of synthesis time and amount of GOD in the reaction mixture on the GOD loading and crystallization were studied. The formation of the ZIF-7-III/GOD composites was confirmed by PXRD, FT-IR, TGA, and catalytic measurements. The Bradford method³⁵ was used to determine and optimize the GOD concentration in the composite. TGA was also employed to confirm the loading. The results are presented in the following sections.

Synthesis of ZIF-7-III and the ZIF-7-III/GOD composite

ZIF-7-III was synthesized using a slightly modified method, as reported by Ebrahimi et al.³⁴ by adding an aqueous NH₃ solution (32%) directly to the reaction mixture containing 29 mL H₂O, 1 mmol HbIm, and 0.5 mmol Zn²⁺ ions. The amount of NH₃ solution was varied between 0 and 10 mL (0–165.4 mmol) (Section S1.2). The addition of 1 or 10 mL of aqueous NH₃ led to the formation of ZIF-7-III with a high degree of crystallinity, whereas the absence of NH₃ resulted in a product with low crystallinity (Fig. S2). However, the pH of the reaction mixtures at this condition at the end of the reaction was 10, leading to denaturation of GOD. The use of 0.1 mL aqueous NH₃ resulted in pH 6, which is close to the optimal pH for GOD activity,¹⁰ as well as a crystalline product. Hence, these synthesis conditions were used for further investigations.

The composition of the ZIF-7-III/GOD composites with respect to GOD loading and the impact of GOD on the crystallization process was studied by adding different GOD amounts ranging from 0 to 30 mg to the reaction mixture (Section S1.3) and reactions were carried out by varying the reaction time between 5 min to 2 h. In the absence of GOD, reflections of ZIF-7-III are clearly observed after 15 min (see the PXRD patterns in Fig. S3) while in the presence of GOD (20 mg per batch), crystalline material is only found after 1.5 h (see the PXRD patterns in Fig. S4). Thus, GOD slows down (inhibits) the crystallization indicating its entrapment in the composite. The results of the optimization study of the GOD loading are shown in Table 1. As expected, increasing the GOD amount from 5 to 20 mg in the reaction mixture resulted in an increase of GOD in the composite. When 20, 25, and 30 mg of the enzyme were employed, the amount of immobilized enzyme was constant (∼16.5 mg), indicating that the support had reached its maximum loading capacity.³⁶ The catalytic activity of the ZIF-7-III/GOD was also investigated to confirm the accessibility of the enzymes to glucose molecules. Thus, the amount of immobilized enzyme was kept constant in each catalytic reaction, and the glucose oxidase activity was determined by the Miller method (Section S1.5). The results are also summarized in Table 1. The activity of ZIF-7-III/GOD increased gradually from 62% to 98% as the GOD amount in the reaction mixture increased from 5 to 20 mg. Samples obtained with larger amounts of GOD, i.e. 25 and 30 mg, were catalytically less active. The decrease has been previously attributed to aggregation or entrapment of enzymes or an impeded mass transfer of substrates to the active sites.^26,37 Also, the orientation of the enzyme can have a strong impact on its catalytic activity.^21,38 The highest activity of the ZIF-7-III/GOD composite, with a relative loading of 82% and an activity of 98%, was found when 20 mg of the enzyme was employed. This corresponds to a GOD loading of 11.7% in the composite, as 140 mg of the composite was obtained in the synthesis. This composite was used for all subsequent studies.

Table 1 Effect of GOD amount in the reaction mixture on GOD loading and catalytic activity of the ZIF-7-III/GOD composite^a

Entry	Amount of GOD in the synthesis (mg)	GOD immobilization (mg)/(%)	Relative catalytic activity^b (μmol min^−1 ml^−1)/(%)
a The synthesis of ZIF-7-III/GOD was carried out at room temperature for 2 h, varying the amount of GOD and using 1 mmol HbIm, 0.5 mmol Zn^2+, and 0.1 mL aqueous NH3 (32%) solution (Vtot = 29 mL). GOD loading was determined by the Bradford method (S1.1), and the relative catalytic activity of the immobilized enzyme was determined by the Miller method after a reaction time of 25 min at pH 6 and room temperature. A constant weight ratio of substrate to immobilized enzyme was used. b Relative catalytic activity in % relates to the activity of the ZIF-7-III/GOD composite in comparison to the free enzyme under identical catalytic reaction conditions.
1	5	4.5/90	0.037/62
2	10	8.6/86	0.042/70
3	15	12.75/85	0.049/82
4	20	16.4/82	0.059/98
5	25	16.7/68	0.054/90
6	30	16.7/55	0.051/85

---

Characterization of ZIF-7-III and ZIF-7-III/GOD

Characterization of ZIF-7-III and the ZIF-7-III/GOD composite, obtained using the optimized synthesis conditions, was carried out using PXRD (Fig. 1 (top)). Broad reflections in both PXRD patterns are observed, indicating the formation of small crystallites, which was also supported by electron microscopy. The presence of GOD in the reaction mixture also leads to minor differences in the positions of the 002 Bragg reflection corresponding to the distance between the layers of ZIF-7-III (Fig. 1 (top)). This slight change in the interlayer d-spacing (9.81 Å to 9.91 Å) cannot be explained by the intercalation of the large GOD molecules (6.0 × 5.2 × 7.7 nm³)¹⁰ and thus the enzyme must have been encapsulated in or entrapped on the surface of ZIF-7-III.^38,39

Fig. 1 Top: PXRD patterns of simulated ZIF-7-III (black), as well as ZIF-7-III (red) and ZIF-7-III/GOD (green) obtained using the optimized synthesis conditions, middle: FTIR spectra of free GOD (blue), ZIF-7-III/GOD (green), and ZIF-7-III (red). Bottom: SEM images of (a) ZIF-7-III, (b) ZIF-7-III/GOD, and (c) ZIF-7-III/GOD after its use as a catalyst in six consecutive reactions.

FT-IR spectroscopy clearly confirms the presence of GOD in the ZIF-7-III/GOD composite (Fig. 1 (middle)), and the bands are assigned in Table S1. The observed characteristic vibrational bands are in line with the reference spectra of the pure compounds (Table S2).^38,40,41 Thus, the sharp bands corresponding to the C–H stretching and bending vibrations, as well as those related to ring vibrations of ZIF-7-III, are clearly observed in the composite.⁴¹ The bands for GOD are much broader due to the different local environments of the different groups in the enzyme, but the characteristic bands related to the peptidic bonds at 1646 and 1577 cm⁻¹ are also observed for the composite. These are due to the C=O stretching vibration and the N–H bending with a contribution from the C–N stretching vibrations, respectively.^42,43

The morphology of the products was determined by SEM. The SEM micrographs (Fig. 1 (bottom)) of the samples show that in the presence and absence of GOD, block-shaped crystals are formed, but the shape is less regular and the surface much rougher in the composite, showing the presence of many structural defects. This observation is in line with the presence of GOD molecules entrapped on the surface of ZIF-7-III, since the enzymes will modify the crystal growth during the synthesis. The block-shape morphology and smooth surface observed for ZIF-7-III are in good agreement with the results from a previous study by Liu et al.²⁹ The SEM micrograph of the composite used in six consecutive catalytic reactions is also presented in Fig. 1 (bottom (c)). The characteristic block-shaped morphology and the rougher surface with many defects are still present.

The presence of GOD in the ZIF-7-III/GOD composite material was also confirmed by TGA. The TG curves of ZIF-7-III, free GOD, and ZIF-7-III/GOD are shown in Fig. 2. The TG curve of ZIF-7-III exhibits a well-defined plateau, and decomposition of the framework takes place between 450 and 600 °C.²⁹ The TG curve of the ZIF-7-III/GOD composite shows a TG curve similar to ZIF-7-III but with an additional decomposition step between 25 and 480 °C corresponding to the decomposition of GOD followed by decomposition of ZIF-7-III. The residue obtained at 600 °C was identified as ZnO (Fig. S6) and this information was used to quantify the amount of immobilized GOD in the ZIF-7-III/GOD composite (Table S4). Thus, a weight loss of 11.1% was observed which corresponds very well to the GOD loading determined by the Bradford method (11.7%).

Fig. 2 TGA curves of free GOD (blue), ZIF-7-III/GOD (green), and ZIF-7-III (red).

Catalytic experiments

After optimization of the synthesis conditions and composition (GOD loading), the properties of the ZIF-7-III/GOD composite were determined. The immobilization of enzymes can improve their stability compared to free ones, however, bioactivity and mass transport to the active sites must be maintained.^23,44,45 The reaction conditions for the activity assay, i.e., the conversion of glucose to gluconic acid and H₂O₂, were studied using different reaction times (5 min–2 h), pH values (3–9), and reaction temperatures (30–80 °C). In addition, the storage stability and the reusability of the composite material were investigated in six consecutive catalytic reactions. Each catalytic reaction was carried out three times, and the error bars represent the standard deviation.

To study the accessibility of the glucose to the catalytic center of the enzyme in the ZIF-7-III/GOD composite, the amount of converted glucose after 5, 10, 15, 20, 25, and 30 min was determined and compared to the amount obtained using free GOD (Fig. 3). Details are given in Section S2.2.1 and Table S5. After a reaction time of 25 min or higher, almost full conversion of glucose to gluconic acid was observed, and the relative catalytic activity of the composite was 98%. The activity of the composite is much smaller at shorter reaction times, which can be explained by hindered accessibility of the active center of the immobilized enzymes. The relative catalytic activity of solutions containing GOD, GOD/HbIM, GOD/Zn²⁺, and GOD/NH₃ are also shown in Fig. 3 for comparison (Section S2.2.2 and Table S5).

Fig. 3 The relative catalytic activity of the immobilized enzyme as a function of reaction time for the conversion of glucose to gluconic acid and H₂O₂ using the optimized ZIF-7-III/GOD composite (blue squares). The relative catalytic activity of the immobilized enzyme was determined by the Miller method after different reaction times (5, 10, 15, 20, 25 min) at pH 6 and room temperature. A constant weight ratio of substrate to immobilized enzyme was used. In addition, experiments were carried out to determine the catalytic activity of GOD in solution and in the presence of the starting materials; HbIm (orange square), Zn(NO₃)₂ (green square) and NH₃ (pink square). Precursors have no obvious effect on GOD activity. Enzyme activity levels exceeding 100%, as indicated by the observed values, are within the normal range.⁵⁰

After fixing the reaction time to 25 min, the influence of pH (3–9) on the catalytic activity of free and immobilized GOD was investigated (Fig. 4(a) and Section S2.2.3). The optimal pH values for maximum activity of the free and immobilized GOD were 6 and 7, respectively. It is noteworthy that the immobilized GOD showed higher tolerance under alkaline and acid conditions compared to the free GOD. The relative catalytic activity of the free GOD is only 28 and 21% at pH 3 and 9, respectively, while under the same conditions, the immobilized GOD retained 61 and 58% of its initial activity. Thus, improved stability and activity in a wider pH range is observed for the ZIF-7-III/GOD composite.

Fig. 4 (a) Effect of pH on the relative catalytic activity of free GOD and ZIF-7-III/GOD, (b) effect of temperature on the relative catalytic activity of free GOD and ZIF-7-III/GOD, (c) comparison of storage stability of the ZIF-7-III/GOD composite and free GOD, and (d) relative catalytic activity of the ZIF-7-III/GOD sample used in six consecutive catalytic reactions.

In addition to the chemical stability, the thermal stability of the immobilized enzyme, relative to the free enzyme, is an essential factor for possible industrial applications. Therefore, the relative catalytic activity of the free and immobilized GOD was investigated at different reaction temperatures (Fig. 4(b) and Section S2.2.4). At low reaction temperatures (30–40 °C), only minor differences in activities were found, while at higher temperatures (>40 °C), the immobilized GOD exhibits a significantly higher stability compared to the free GOD. Thus, at a temperature of 70 (80)°C, the relative catalytic activities are 31 (11) and 62 (46)% for free and immobilized GOD, respectively. Hence, the temperature range of the enzyme activity could be extended by immobilizing GOD.

The storage stability and reusability of biocatalytic systems are critical factors in potential industrial applications from the practical and economic points of view. The results of the storage stability of the free enzyme and ZIF-7-III/GOD in buffer solutions at 4 °C are shown in Fig. 4(c). Details are provided in Section S2.2.5. The activity of the biocatalyst was about 66% of its initial activity after 60 days, whereas the activity of free GOD was less than 25% of the initial activity under the same conditions, demonstrating the increase in storage stability due to the immobilization of the enzyme. To determine the reusability of ZIF-7-III/GOD, the composite was recovered from the reaction mixture by centrifugation, washed with phosphate buffer solution (PBS), dried at room temperature, and reused in another reaction with fresh substrates (S2.2.6). This experiment was repeated for six cycles (Fig. 4(d)). A decrease in enzyme activity of the immobilized GOD was observed after six cycles, maintaining 77.7% of its initial activity. The characterization results of the recovered composite after 6 cycles are presented in Fig. S8.

In summary, the characterization of the composite and the results of the catalytic studies confirm the successful immobilization of GOD, its improved stability at elevated temperatures and a wide pH range, as well as the catalytic activity. Although ZIF-7-III itself is nonporous, the high catalytic activity observed after GOD immobilization clearly suggests effective enzyme incorporation. In principle, these results can be explained by two possible mechanisms for enzyme immobilization (Fig. 5), i.e. entrapment of enzymes either on ZIF-7-III particle surfaces (1) or between ZIF-7-III particles (2), which leads to the formation of agglomerates. Whilst PXRD cannot be used to distinguish between the two mechanisms, there are a number of results that strongly support entrapment on the particle surface. The slower crystallization of ZIF-7-III/GOD indicates an interaction between the ZIF-7-III precursors, especially the Zn²⁺ ions, and GOD in the reaction mixture, which could be purely electrostatic or by coordination of –COO⁻ and histidine groups of the enzyme. This modified crystallization process results in the formation of ZIF-7-III particles with rougher surfaces as seen in the SEM micrographs. In addition, the high immobilization efficiency of up to 82% and the large amount of immobilized GOD in the ZIF-7-III/GOD composite, confirmed by TG measurements and the Bradford method, indicate the existence of intimate interactions between GOD and ZIF-7-III. For comparison, adsorption of GOD on pre-synthesized ZIF-7-III was also investigated (two-step method), but no immobilization was observed, confirming the advantage of the one-step (in situ) synthesis procedure (Fig. S9 and Section S2.2.7). The entrapped enzymes are catalytically active, but the relative catalytic activity of the composite in comparison to the free GOD (Table 1) is decreased, which can be due to hindered accessibility of glucose to the enzyme or a lower catalytic activity of the entrapped GOD. Mechanism 1 is also supported by the results of dynamic light scattering (DLS) measurements, the reusability study, as well as the GOD loading investigation on pre-synthesized ZIF-7-III. DLS measurements (Fig. S10) show the presence of particles with an average diameter of 400 nm, which demonstrates the presence of individual particles and no agglomeration in the PBS. In addition, slight leaching of enzymes after each catalytic reaction is observed, which is in line with intimate interactions between GOD and ZIF-7-III. The attempt to immobilize the enzyme with ZIF-7-III (by two-step method) resulted in a solid material containing no GOD, as demonstrated by the Bradford method.

Fig. 5 Scheme showing the two possible mechanisms for enzyme immobilization, i.e. entrapment of enzymes on/in ZIF-7-III particle surfaces (top) or between ZIF-7-III particles, which leads to the formation of agglomerates (bottom).

Biosensor fabrication and testing (using ZIF-7-III/GOD)

Electrochemical response of GCE/ZIF-7-III/GOD. The characterization results and the catalytic properties of the optimized ZIF-7-III/GOD composite have demonstrated its high relative catalytic activity and its improved chemical and storage stability. Thus, the composite was also used as an electrochemical biosensor. It should be mentioned that direct immobilization of GOD on an electrode is not as efficient and often leads to denaturation of the enzyme.^46,47 The fabrication of the ZIF-7-III/GOD biosensor was accomplished by drop casting ZIF-7-III/GOD onto a GCE. Details are provided in Section S2.3.1. The following sections describe the electrochemical characteristics as determined by cyclic voltammetry (CV) and chronoamperometry (CA). In addition, sensor characteristics such as stability, repeatability, reusability, and selectivity are reported, as well as the first results on the determination of glucose in a real sample in O₂-saturated PBS at pH = 7.4.

The cyclic voltammograms of the GCE/ZIF-7-III and GCE/ZIF-7-III/GOD in the presence of O₂-saturated PBS are shown in Fig. S11 and Fig. 6(a), respectively. No redox peaks are found for GCE/ZIF-7-III, while for GCE/ZIF-7-III/GOD, two peaks are observed, the anodic and cathodic peaks, at E_pa = −340 mV and E_pc = −420 mV. In contrast, the response of the composite in N₂-saturated PBS resulted only in two low-intensity redox peaks at the same potentials (Fig. 6(b)). The peak potential separation (ΔE_p) is about 80 mV, indicating a quasi-reversible redox-behavior, which can be explained by the redox couple FAD/FADH₂, the redox centers of GOD, suggesting a good direct electron transfer between the electrode and GOD (eqn (1) and Fig. 6(d)),^48,49

GOD-FAD + 2H⁺ + 2e⁻ ⇌ GOD-FADH₂ (1)

Fig. 6 Cyclic voltammetry responses at a glassy carbon electrode coated with ZIF-7-III/GOD with E_pa = −340 mV and E_pc = −420 mV in (a) O₂-saturated PBS solution, (b) N₂-saturated PBS solution, and (c) O₂-saturated PBS solution containing different concentrations of glucose: 0, 0.3, 0.5, 0.7, 0.8, 0.9 and 1 mmol L⁻¹. The PBS concentration used in all measurements is 0.1 mol L⁻¹ and the pH is 7.4. All CV curves were recorded against Ag/AgCl (saturated KCl) (scan rate: 50 mV s⁻¹). (d) Reaction scheme of the glucose sensor by the ZIF-7-III/GOD modified electrode.

The cathodic peak current at E_pc = −420 mV in the O₂-saturated PBS is significantly higher compared to the measurement in N₂-saturated PBS, which is due to the fast re-oxidation of the reduced form of GOD (GOD-FADH₂) to its oxidized form (GOD-FAD) by the dissolved O₂, and H₂O₂ is formed (eqn (2)).

GOD-FADH₂ + O₂ → GOD-FAD + H₂O₂ (2)

The electrochemical biosensor was also used to measure indirectly the amount of glucose in the solution. The CV response curves of the ZIF-7-III/GOD biosensor at different glucose concentrations ranging from 0 to 1 mmol L⁻¹ are shown in Fig. 6(c). A decrease in the GOD reduction peak current at approximately −420 mV and an increase in the GOD oxidation peak current at approximately −340 mV are observed with increasing glucose concentration.^10,50 In the presence of glucose, GOD-FAD is chemically reduced to GOD-FADH₂ (eqn (3)) and gluconic acid is formed, leading to a decreased reduction peak current according to eqn (1).^49,51

Glucose + GOD-FAD → GOD-FADH₂ + Gluconic acid (3)

Chronoamperometric measurements record the current at a fixed voltage as a function of time, which can be used to determine the concentration of an analyte and the sensitivity of a sensor. The CA curves of solutions containing different amounts of glucose ranging from 0.05 to 3 mmol L⁻¹ were measured in O₂-saturated PBS (pH = 7.4), and the response curves for all glucose concentrations are shown in Fig. 7(a). As previously observed in the CV curves, the current depends on the glucose concentration, a decrease is observed with increasing glucose concentrations, and a stable current is observed within approximately 10 seconds. A linear response is found for glucose concentrations ranging from 0.05 to 1.1 mmol L⁻¹, with a detection limit of 10.2 μmol L⁻¹ (see Fig. 7(b)).

Fig. 7 (a) Chronoamperometric curves for the GCE/ZIF-7-III/GOD biosensor at a constant voltage of −450 mV versus Ag/AgCl (saturated KCl) using glucose solutions with concentrations ranging from 0.05 to 3 mmol L⁻¹. (b) The calibration curve of the GCE/ZIF-7-III/GOD shows the concentration dependence of the current after 10 s in the range from 0.05 to 1.1 mmol L⁻¹.

CV measurements were also carried out in the presence of H₂O₂ and the absence of glucose. A high cathodic peak current at −420 mV is observed, which increases with the amount of added H₂O₂ (Fig. S12). The increase is due to the electrochemical oxidation of H₂O₂ to O₂ at the electrode–electrolyte interface (eqn (4)).⁵² The O₂ is reduced by GOD-FADH₂, resulting in the formation of GOD-FAD and H₂O₂ according to eqn (2), which in turn leads to an increase in the cathodic current (eqn (1)).

H₂O₂ → O₂ + 2H⁺ + 2e⁻ (4)

Stability, repeatability, reusability, and selectivity studies. Storage stability assessments of the GCE/ZIF-7-III/GOD biosensor, conducted through daily response current measurements of 0.5 mmol L⁻¹ glucose, demonstrated a 90% initial current retention after two weeks, indicating robust storage stability. Furthermore, the biosensor exhibits good repeatability (10 repeated measurements) and reproducibility (10 single measurements) with a relative standard deviation of less than 6% and high selectivity against interfering substances such as ascorbic acid, uric acid, mannose, galactose, and fructose, even at concentrations ten times higher than glucose (Fig. 8).

Fig. 8 The electrochemical response of the GCE/ZIF-7-III/GOD biosensor in 0.1 mol L⁻¹ O₂-saturated PBS (pH  =  7.4) containing 0.2 mmol L⁻¹ glucose and 2.0 mmol L⁻¹ of various interferences (scan rate: 50 mV s⁻¹).

The performance of the electrochemical biosensor presented in this work is compared with some of the published glucose sensors (Table 2). The linear range of the GCE/ZIF-7-III/GOD biosensor is similar to those reported in prior studies.^49,52–57 However, unlike some of these studies that employed complex composites for electrode modification, this work utilizes a simple procedure for in situ enzyme immobilization, eliminating the need for additional substances such as Nafion. While g-C₃N₄ (Table 2, entry 4) serves as a straightforward modifier, the linear range of the resulting sensor is only found for higher concentrations from 2 to 50 mmol L⁻¹.⁵⁴ The lower limit of detection (LOD) of the GCE/ZIF-7-III/GOD biosensor is close to the range found for other sensors.

Table 2 A comparison of the analytical characteristics of this sensor and other sensors used to detect glucose

Materials	Linear range (mmol L^−1)	LOD (mmol L^−1)	Ref.
FTO: fluorine-doped tin oxide, CNTs: carbon nanotubes, PEI: polyethylenimine, GOx: glucose oxidase, MoS2: molybdenum disulfide, NRs: nanorods, GCE: glassy carbon electrode, RGO: reduced graphene oxide, g-C3N4: 2-dimensional graphitic carbon nitride, NSs: nanosheets, BiOI: bismuth oxyiodide, GOD: glucose oxidase, AMP: aminopropyl-magnesium phyllosilicate, MWCNT: multiwalled carbon nanotubes.
FTO-CNTs/PEI/GOx	0.07–0.7	0.07	49
GOx/MoS2 NR/GCE	3.0–7.0	3.0	52
GOx/Fe3O4/RGO/GCE	0.5–12	0.05	53
GOx/g-C3N4/GCE	2–50	0.005	54
TiO2 NSs/BiOI NSs/GOx	0.01–1	0.01	55
Nafion-GOD/AMP-GCE	0.1–1.1	0.035	56
ZIF-8-GOx/Pd/MWCNT	1–10	0.05	57
GCE/ZIF-7-III/GOD	0.05–1.1	0.01	This work

---

Determination of glucose in a real sample. A human plasma sample was used to investigate the performance of the biosensor for practical applications. The concentration of glucose in the plasma was determined by a blood glucose meter. The GCE/ZIF-7-III/GOD biosensor was used for the determination of glucose in this plasma sample. Thus, the plasma was diluted with a phosphate buffer solution (pH 7.4) to yield a concentration of 0.24 mmol L⁻¹ and then spiked with pre-defined amounts of glucose to yield glucose concentrations ranging from 0.29 to 0.74 mmol L⁻¹. The chronoamperometric response was recorded and the concentration of glucose in the samples was determined using the calibration curve (Fig. 7(b)). The analytical results are summarized in Table 3, with relative standard deviation (RSD) calculated to be less than 6%, indicating that the GCE/ZIF-7-III/GOD biosensor is feasible for practical analysis.

Table 3 Glucose detection in a plasma sample using the GCE/ZIF-7-III/GOD biosensor^a

c (glucose) (mmol L^−1)	Δc (glucose) (mmol L^−1)	c calc. (glucose) (mmol L^−1)	c obs. (glucose) (mmol L^−1)	Recovery (%)	RSD (%)
a Each experiment was carried out three times to determine the relative standard deviation. c (glucose) = glucose concentration in PBS after the addition of a plasma sample; Δc = increase of glucose concentration in PBS by the addition of a glucose standard solution; ccalc. (glucose) = calculated concentration due to the presence of plasma and glucose standard solution; cobs. (glucose) = concentration determined from the chronoamperometric response using the calibration curve; recovery = 100 × observed concentration/(original concentration + added concentration).
	0.05	0.29	0.33	113	5.25
0.24	0.1	0.34	0.36	105	3.91
	0.5	0.74	0.71	96.0	2.50

---

Conclusions

The utilization of the coordination polymer ZIF-7-III as the support for enzyme immobilization has been demonstrated. Through a one-step synthesis of ZIF-7-III in the presence of GOD, a ZIF-7-III/GOD composite with a loading of ca. 11 wt% and an efficiency of 82% was obtained. ZIF-7-III is known to be nonporous; therefore, the immobilized enzymes have to be incorporated on the external surfaces of the particles. Immobilization leads to a decrease in enzyme accessibility and only after a reaction time of 25 min almost full conversion of glucose to gluconic acid is observed which corresponds to a relative catalytic activity of 98% for the composite. At the same time, immobilization leads to a higher temperature and storage stability, as well as pH tolerance for GOD. The ZIF-7-III/GOD composite was used in six consecutive reactions and only a decrease of 23% in relative catalytic activity was observed. Deposition of the composite onto a GCE was accomplished and the setup was used as a biosensor for glucose. High selectivity and detection in the range of 0.05 to 1.1 mmol L⁻¹ range is demonstrated and the sensor has also been successfully tested using a plasma sample. The study demonstrates that coordination polymers that can be obtained in water using mild reaction conditions are suitable candidates for enzyme immobilization. Further studies will focus on the immobilization of other enzymes and using other coordination polymers. In brief, in this work we effectively harnessed the synergistic effects of structure and material and enhanced the application efficiency of biosensors.

Author contributions

Sahar Aghayani: methodology, investigation, analysis, writing; Shahram Tangestaninejad: project administrator, supervision, editing; Norbert Stock: data validity, editing; Bastian Achenbach: methodology, editing; Mehrnaz Bahadori: validation, editing; Majid Moghadam: supervision, validation; Seyedeh Fatemeh Nami-Ana: validation; Maryam Sharifi: validation; Valiollah Mirkhani: supervision; Iraj Mohammadpoor-Baltork: supervision.

Conflicts of interest

There are no conflicts to declare.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. All data generated or analyzed during this study are included in this published article.

Supplementary information is available. Some experimental parts, characterization of ZIF-7-III, and ZIF-7-III/GOD, catalytic studies, and biosensor fabrication and testing have been presented in supplementary information. See DOI: https://doi.org/10.1039/d5tb01119a

Acknowledgements

We gratefully acknowledge the financial support provided by the Ministry of Science, Research and Technology of Iran and the Research Council of the University of Isfahan. We also extend our sincere thanks to the University of Kiel for their support, with special appreciation given to the team in the Department of Inorganic Chemistry for their invaluable assistance with various measurements.

Notes and references

1. R. Yuan, B. Yan, C. Lai, X. Wang, Y. Cao, J. Tu, Y. Li and Q. Wu, ACS Omega, 2023, 8, 22099–22107.
2. M. J. Jara Fornerod, A. Alvarez-Fernandez, M. Michalska, I. Papakonstantinou and S. Guldin, Chem. Mater., 2023, 35, 7577–7587.
3. K. L. Wolkowicz, E. M. Aiello, E. Vargas, H. Teymourian, F. Tehrani, J. Wang, J. E. Pinsker, F. J. Doyle, M. Patti, L. M. Laffel and E. Dassau, Bioeng. Transl. Med., 2021, 6, e10201.
4. N. O. Gomes, R. T. Paschoalin, S. Bilatto, A. R. Sorigotti, C. S. Farinas, L. H. C. Mattoso, S. A. S. Machado, O. N. Oliveira and P. A. Raymundo-Pereira, ACS Sustainable Chem. Eng., 2023, 11, 2209–2218.
5. G. Zhang, Q. Li, J. Li, M. Pan, Y. Wang, R. Su, W. Qi and W. Zhang, Langmuir, 2023, 39, 8779–8786.
6. A. M. Ashrafi, M. Sýs, E. Sedláčková, A. S. Farag, V. Adam, J. Přibyl and L. Richtera, Sensors, 2019, 19, 2939.
7. R. Cai, Z. Zhang, H. Chen, Y. Tian and N. Zhou, Sens. Actuators, B, 2021, 326, 128842.
8. M. Z. H. Khan, M. R. Hasan, S. I. Hossain, M. S. Ahommed and M. Daizy, Biosens. Bioelectron., 2020, 166, 112431.
9. M.-J. Lee, J.-H. Choi, J.-H. Shin, J. Yun, T. Kim, Y.-J. Kim and B.-K. Oh, ACS Appl. Nano Mater., 2023, 6, 12567–12577.
10. S. B. Bankar, M. V. Bule, R. S. Singhal and L. Ananthanarayan, Biotechnol. Adv., 2009, 27, 489–501.
11. J. Kim, A. S. Campbell, B. E.-F. De Ávila and J. Wang, Nat. Biotechnol., 2019, 37, 389–406.
12. J. Lu, M. Nie, Y. Li, H. Zhu and G. Shi, Colloids Surf., B, 2022, 217, 112602.
13. X. Lian, Y.-P. Chen, T.-F. Liu and H.-C. Zhou, Chem. Sci., 2016, 7, 6969–6973.
14. S. Rafiei, S. Tangestaninejad, P. Horcajada, M. Moghadam, V. Mirkhani, I. Mohammadpoor-Baltork, R. Kardanpour and F. Zadehahmadi, Chem. Eng. J., 2018, 334, 1233–1241.
15. V. Asadi, R. Kardanpour, S. Tangestaninejad, M. Moghadam, V. Mirkhani and I. Mohammadpoor-Baltork, RSC Adv., 2019, 9, 28460–28469.
16. N. R. Mohamad, N. H. C. Marzuki, N. A. Buang, F. Huyop and R. A. Wahab, Biotechnol. Biotechnol. Equip., 2015, 29, 205–220.
17. A. R. Ismail and K.-H. Baek, Int. J. Biol. Macromol., 2020, 163, 1624–1639.
18. J. Zdarta, A. Meyer, T. Jesionowski and M. Pinelo, Catalysts, 2018, 8, 92.
19. Y. R. Maghraby, R. M. El-Shabasy, A. H. Ibrahim and H. M. E.-S. Azzazy, ACS Omega, 2023, 8, 5184–5196.
20. C. Mateo, J. M. Palomo, G. Fernandez-Lorente, J. M. Guisan and R. Fernandez-Lafuente, Enzyme Microb. Technol., 2007, 40, 1451–1463.
21. A. Sassolas, L. J. Blum and B. D. Leca-Bouvier, Biotechnol. Adv., 2012, 30, 489–511.
22. C. Sicard, Angew. Chem., Int. Ed., 2023, 135, e202213405.
23. J. Boudrant, J. M. Woodley and R. Fernandez-Lafuente, Process Biochem., 2020, 90, 66–80.
24. S. S. Nadar, L. Vaidya and V. K. Rathod, Int. J. Biol. Macromol., 2020, 149, 861–876.
25. P.-H. Hsu, C.-C. Chang, T.-H. Wang, P. K. Lam, M.-Y. Wei, C.-T. Chen, C.-Y. Chen, L.-Y. Chou and F.-K. Shieh, ACS Appl. Mater. Interfaces, 2021, 13, 52014–52022.
26. J. Yang, W. Huang, W. Zhang, K. Wei, B. Pan and S. Zhang, Langmuir, 2023, 39, 2312–2321.
27. C. Hu, Y. Bai, M. Hou, Y. Wang, L. Wang, X. Cao, C.-W. Chan, H. Sun, W. Li, J. Ge and K. Ren, Sci. Adv., 2020, 6, eaax5785.
28. R. Banerjee, A. Phan, B. Wang, C. Knobler, H. Furukawa, M. O’Keeffe and O. M. Yaghi, Science, 2008, 319, 939–943.
29. Y. Peng, Y. Li, Y. Ban, H. Jin, W. Jiao, X. Liu and W. Yang, Science, 2014, 346, 1356–1359.
30. P. Zhao, G. I. Lampronti, G. O. Lloyd, M. T. Wharmby, S. Facq, A. K. Cheetham and S. A. T. Redfern, Chem. Mater., 2014, 26, 1767–1769.
31. D. A. Sherman, M. Gutiérrez, I. Griffiths, S. Mollick, N. Amin, A. Douhal and J. Tan, Adv. Funct. Mater., 2023, 33, 2214307.
32. Q.-F. Yang, X.-B. Cui, J.-H. Yu, J. Lu, X.-Y. Yu, X. Zhang, J.-Q. Xu, Q. Hou and T.-G. Wang, CrystEngComm, 2008, 10, 1534.
33. S. Xia, M. Ni, T. Zhu, Y. Zhao and N. Li, Desalination, 2015, 371, 78–87.
34. M. Ebrahimi and M. Mansournia, Mater. Lett., 2017, 189, 243–247.
35. M. M. Bradford, Anal. Biochem., 1976, 72, 248–254.
36. N. S. Rios, S. Arana-Peña, C. Mendez-Sanchez, Y. Lokha, V. Cortes-Corberan, L. R. B. Gonçalves and R. Fernandez-Lafuente, Catalysts, 2019, 9, 576.
37. J.-P. Chen and W.-S. Lin, Enzyme Microb. Technol., 2003, 32, 801–811.
38. Y. Pan, H. Li, J. Farmakes, F. Xiao, B. Chen, S. Ma and Z. Yang, J. Am. Chem. Soc., 2018, 140, 16032–16036.
39. D. A. Sherman, M. Gutiérrez, I. Griffiths, S. Mollick, N. Amin, A. Douhal and J. Tan, Adv. Funct. Mater., 2023, 33, 2214307.
40. Y. Sun, Y. Li and J.-C. Tan, ACS Appl. Mater. Interfaces, 2018, 10, 41831–41838.
41. N. Buğday, S. Altın and S. Yaşar, Int. J. Energy Res., 2022, 46, 795–809.
42. M. Portaccio, B. Della Ventura, D. G. Mita, N. Manolova, O. Stoilova, I. Rashkov and M. Lepore, J. Sol-Gel Sci. Technol., 2011, 57, 204–211.
43. F. Gutierrez, M. D. Rubianes and G. A. Rivas, Sens. Actuators, B, 2012, 161, 191–197.
44. E. T. Hwang and S. Lee, ACS Catal., 2019, 9, 4402–4425.
45. D. Sillu and S. Agnihotri, ACS Sustainable Chem. Eng., 2020, 8, 900–913.
46. K.-Y. Hwa and B. Subramani, J. Med. Bioeng., 2015, 4, 297–301.
47. E. Lebègue, H. Smida, T. Flinois, V. Vié, C. Lagrost and F. Barrière, J. Electroanal. Chem., 2018, 808, 286–292.
48. Z. Dai, G. Shao, J. Hong, J. Bao and J. Shen, Biosens. Bioelectron., 2009, 24, 1286–1291.
49. M.-H. Lin, S. Gupta, C. Chang, C.-Y. Lee and N.-H. Tai, Microchem. J., 2022, 180, 107547.
50. G. Wang, N. M. Thai and S.-T. Yau, Biosens. Bioelectron., 2007, 22, 2158–2164.
51. Y. Zhu, Y. Qi, M. Xu and J. Luo, Colloids Surf., A, 2023, 661, 130908.
52. D. Van Tuan, D. T. T. Ngan, N. Thi Thuy, H. Lan, N. Thi Nguyet, V. T. P. Thuy, N. D. Dien, V. V. Thu, P. H. Vuong and P. D. Tam, VNU J. Sci. Math. Phys., 2023, 39, 1.
53. H. Teymourian, A. Salimi and S. Firoozi, Electroanal., 2014, 26, 129–138.
54. K. Tian, H. Liu, Y. Dong, X. Chu and S. Wang, Colloids Surf., A, 2019, 581, 123808.
55. B. Wu, Z. Cheng, Y. Hou, Q. Chen, X. Wang, B. Qiao, D. Chen and J. Tu, RSC Adv., 2022, 12, 19495–19504.
56. L.-M. Liu, F.-Y. Liu, X.-X. Liu, C.-F. Huang, Z.-J. Wang and D.-Y. He, Int. J. Electrochem. Sci., 2020, 15, 6001–6011.
57. R. Singh, M. Musameh, Y. Gao, B. Ozcelik, X. Mulet and C. M. Doherty, J. Mater. Chem. C, 2021, 9, 7677–7688.

---