# Amperometric method for the determination of cellulase activity and its optimization using response surface method – Journal of Analytical Science and Technology

Jun 24, 2022

### Optimization of the protocol for the amperometric detection of cellulase activity

The efficiency of the amperometric evaluation of the cellulase activity is influenced by several factors, such as the hydrolysis temperature and the pH value, the amount and concentration of CMC, the incubation time with CMC, but also by the amount of glucose oxidase used, the concentration of dissolved oxygen (DOC), rate of stirring and the depth of the glucose oxidation reaction (follow-up time of the reaction-related decrease in the oxygen concentration, further referred as assessment time). In the present study, the effect of three independent input variables such as incubation time, assessment time and glucose oxidase concentration, was optimized using surface response method with Box-Behnken design to achieve maximum normalized signal change ΔI of the oxygen sensor. This approach allowed the determination of optimal conditions from a minimal number of runs, helped to reveal the effect of independent variables, and provided information for estimating the results to design an assay. The other abovementioned factors were kept unchanged: temperature 25 °C, pH 5.0 (0.1 M acetate buffer), the concentration of CMC 1% and the stirring speed of the magnetic stirrer was set to a constant speed to ensure uniform stirring and avoid formation of bubbles and vortices, that affect the oxygen sensor signal. All used solutions were air-saturated: the DOC level at 100% saturation at 25 °C is 8.26 mg/l (Dissolved oxygen saturation calculator 2021). The temperature 25 °C was selected as optimal to make the experimental procedure as simple as possible, although the rate of enzymatic catalysis increases along with the increase of temperature. The pH value 5.0 was chosen based on the definition of the cellulase unit (Sigma-Aldrich 2021) as the most favourable in order to achieve the highest cellulase activity. In order to maximize the glucose produced during the CMC hydrolysis, which leads to larger changes in the measured signal and consequently a higher system sensitivity, a relatively high cellulase concentration (0.57 U/ml) was used in optimization studies.

The optimization results are shown in Table 1. Based on the summary statistics of model fitting, the best model to maximize the sensitivity of the assessment was the linear model. The final equation in terms of actual factors was as follows:

$$y = 0.114363 + 0.006417 times {text{X}}_{1} + 0.010625 times {text{X}}_{2} + 0.000083 times {text{X}}_{3 }$$

(2)

where X1, X2 and X3 are the incubation time, (GOD) concentration and assessment time, respectively. The fitting characteristics (p < 0.0001, F = 34.30) imply that the model was significant. The ANOVA test indicated that the incubation time and glucose oxidase (GOD) concentration were significant (p < 0.05) while the effect of the assessment time was not significant (p = 0.5321). The predicted by the model and experimental values of the normalized biosensor signal change were in good correspondence (Additional File 1: Fig. S1) confirmed by the small difference (~ 0.06) between the predicted R2 and adjusted R2 values (0.8020 and 0.8619, respectively). The signal to noise ratio (adequate precision) had a value of 17.59 indicating adequate signals in the entire design space.

Based on the statistical analysis and the purpose of this study to maximize the sensitivity of the system, we chose the maximum values of all input factors to build the calibration curves for the assessment of cellulase activity: the sample incubation time with CMC was 70 min, GOD concentration 12 U/ml, and the measurement time 600 s. The latter was chosen considering that by registering the biosensor output signal for a longer time, it was possible to characterize and analyse the system sensitivity (dependencies between the assessed cellulase activity and biosensor signal) at various selected time points.

### Amperometric determination of cellulase activity

For the determination of cellulase activity, we used CMC. CMC is a water-soluble derivative of cellulose, which major difference is the presence of some anionic carboxymethyl groups replacing hydrogen atoms from hydroxyl groups in the pristine cellulose molecule (Rahman et al. 2021). The properties of CMC depend on the degree and uniformity of substitution, but also the degree of polymerization (DP): the solubility increases with the decreased DP and increased substitution rate, while the viscosity increases with the increased DP (Ergun et al. 2016). However, CMC is widely used to characterize endoglucanase enzyme activity due to the presence of amorphous sites which are ideal for the assessment of endoglucanase action (Nagl et al. 2021).

For the amperometric assay, we followed the decrease of DOC corresponding to the normalized output signal change ΔI (calculated as 1−I/I0) [35] at different cellulase concentrations (Fig. 3). As indicated above, the sensitivity of the system is determined by the difference in the normalized output signal at different CMC concentrations. Considering that, the DOC curves decrease for different cellulase concentrations (in Fig. 3) at different times (marked with grey dashed lines); we can see that the difference in ΔI increases along with the increase in the measurement period. Analysis of the slopes of the calibration plots based on the data collected at different times between 50 and 600 s revealed a hyperbolic correlation (Fig. 4). The correlation clearly indicated that the sensitivity of the assessment system does not change if we increase the assessment time over 300 s, when the sensor signal has reached 97% of the steady-state and the signal is only controlled by diffusion (Rinken and Tenno 2001).

It is also important to mention that while the shorter assessment times lead to a lower sensitivity of the system, the use of a shorter assessment time enables the detection range of the cellulase activity in the samples to be extended.

Based on these considerations, the normalized output signal change value ΔI300 was used to evaluate the cellulase activity The ΔI300 value was linearly dependent on the cellulase activity of the cellulase (between 0.5 and 15 U) with the slope s = 0.057 ± 0.002 1/U (Fig. 5). This value, showing the sensitivity of the assay, was similar to the slope of the dependence of the reaction depth (calculated with the biosensor dynamic model (Rinken et al. 1996)) on cellulase activity (Additional file 1: Table S1), which was 0.055 ± 0.001 1/U, giving an additional indication that using the optimal protocol, the maximum sensitivity of the assay can be achieved in a limited time period using the transient phase data of the process. The reproducibility of the assay was very good, as reflected by the R2 value of 0.9928, and the relative standard deviation of the y-intercept σ was 0.013.

The limit of detection (LOD) and limit of quantification (LOQ) of the amperometric assessment were calculated as following (ICH 2005):

$${text{LOD}} = frac{3.3*sigma }{s}$$

(3)

$${text{LOQ}} = frac{10*sigma }{s}$$

(4)

where σ is the standard deviation of the y-intercept and s is the slope of the calibration curve. The LOD and LOQ values were 1.21 ± 0.06 U and 3.66 ± 0.17 U, respectively.

For testing, the results of the amperometric assessment were compared with those of the DNS method and viscosimetry. A commercial cellulase preparation from TCI Chemicals with a declared cellulase activity 29,000 cellulase U/g was used as a model sample for this test. The apparent cellulase activity of this preparation, evaluated in triplicates with different measurement methods and expressed in units of cellulase activity per gram of solid preparation, is shown in Table 2.

Apparent cellulase activity as determined by various methods was substantially different. The amperometric assay, which measured the total glucose produced by complete hydrolysis of CMC, gave an apparent cellulase activity of 4053 ± 7 U/g, which is ~ 2 and 3 times lower than the DNS and viscosimetric methods. This lower apparent activity is likely to be caused by the low β-glucosidase activity compared to other cellulase enzymes in the commercial preparation examined. It has also been shown that an inappropriate ratio of endoglucanase, exoglucanase and β-glucosidase can lead to an accumulation of cellobiose, resulting in an inhibition of the cellulase activity (Saritha Mohanram 2015). As expected, the apparent activity obtained by the DNS method (Additional File 1: Fig. S2) was higher than the activity measured by the amperometric method, which measures the levels of reducing sugars, including both cellobiose and glucose (Gusakov et al. 2011). The apparent cellulase activity obtained with the DNS method is also explained by the fact that cellobiose, one of the main products of cellulases, is broken down by the DNS itself and measured as approx. 1.5 glucose molecules rather than one reducing end group (Gusakov et al. 2011). The viscosimetry, which mainly focuses on the assessment of endocellulase activity, showed the highest apparent activity of 12,236 ± 350 U/g (Additional File 1: Fig. S3), however, a synergistic effect between the components is required for the complete and effective hydrolysis of cellulose.

The results obtained clearly show the need to identify the method used to evaluate of cellulase activity, since commercial cellulase preparations differ extensively in the amounts of the various enzymes, which leads to fluctuations in the rate and extent of hydrolysis of cellulose substrates.

Compared to other currently available methods for the assessment of cellulase activity, the amperometric method allows a quick and specific detection of glucose, which is the final product of the hydrolysis of celluloses, in the presence of the hydrolysis intermediate cellobiose. The proposed method is simple, does not require the quenching of the glycolytic reaction, and depending on the enzymatic activity of the sample and aim of analysis, it allows the modification of the sensitivity and detection range of the assay.