Optimized decolorization of two poly azo dyes Sirius Red and Sirius Blue using laccase-mediator system

Abstract

Factors, namely pH, laccase-like activity, dyes concentration as well as 1-Hydroxybenzotriazole (HBT) concentration was examined. The results indicated that the maximum decolorization yield and rate reached 98.30 ± 0.10% and 5.84 ± 0.01%/min, respectively for Sirius Blue, and 99.34 ± 0.47% and 5.85 ± 0.12%/min, respectively for Sirius Red after 4 h. The presence of the redox mediator 1-hydroxybenzotriazole (HBT) greatly improved the decolorization levels. The optimum concentrations of HBT, dyes, and laccase were 0.62 mM, 50 mg/L, and 0.89 U/mL respectively at pH 4.58 for both dyes. Phytotoxicity tests using treated and untreated dyes proved that the applied treatment slightly decreased the toxicity of the by-products. However, the germination index (GI) increased from 14.6 to 36.08% and from 31.6 to 36.96% for Sirius Red and Sirius Blue, respectively. The present study focused on the treatment of two recalcitrant azo dyes, namely: Sirius Blue (Direct Blue 71) and Sirius Red (Direct Red 80). The decolorization was performed using cell-free supernatant from Coriolopsis gallica culture with high laccase activity. Response surface methodology (RSM) and Box-Behnken design were applied to optimize the decolorization of the two tested dyes. The effect of four.

Introduction

Population growth and industrialization have increased water consumption and deteriorated its quality. The impact of these disturbances has been noticed at different environmental levels, especially in humans, animals, and plants which are all reliant on clean water (Singh et al. 2020; Al-Tohamy et al. 2022). Domestic, agricultural, and industrial wastewater are the main causes of water quality degradation. These contaminated waters are poorly treated and managed, causing the release of recalcitrant and toxic molecules such as heavy metals, harmful solvents, dyes, and pesticides (Akpor and Mammo 2011; Kant 2012; Sarkar et al. 2017). Currently, water contamination due to the inability of textile industries to properly dispose of their wastewater is one of the major challenges affecting the whole world. Wastewater containing dyes is a major pollutant of the environment. Textile industries generate large amounts of highly colored wastewater containing a diverse range of persistent pollutants (Benkhaya et al. 2017; Sarkar et al. 2017; Rafiq et al. 2021). In general, wastewaters produced from textile industries contain dyes, chemical components (acetic acid, sulfides, etc.) (Yaseen et al. 2019). These affect negatively the balance of the ecosystem, especially the aquatic life by reducing the diffusion of light and gaseous exchanges in the discharge medium and increasing the biological and chemical oxygen demand (Kant 2012; Sarkar et al. 2017; Benkhaya et al. 2022). Around 7 × 10⁷ tons of synthetic dyes are generated annually worldwide. The textile industry uses over 10,000 tons of dyes, frequently classified into diverse groups based on their origin, structure, and application (Al-Tohamy et al. 2022; Chandanshive et al. 2020). Among the most commonly used synthetic dyes in the textile industry, we can cite Azo dyes which are the largest used in the textile industry class (above 70%) (Benkhaya et al. 2020; Thangaraj et al. 2021). Humans may come across azo dyes in via ingestion or direct skin which have a damaging influence on many tissues in the human body (Manickam and Vijay 2021). Furthermore, since the azo dyes are not biodegradable and rather resistant to conventional physicochemical degradation, they may result in the appearance of various chronic diseases namely hypertension, chronic cramps, sporadic diseases as well as sarcoma of the spleen, apart from genetic mutations (Kant 2012; Sarkar et al. 2017).

Because of the high number of textile manufactures and the vast amounts of dye-containing wastewater, adequate and efficient management solutions are required to prevent the ecosystem contamination and promote sustainability. Based on the wastewater properties, textile dye wastewater can be treated physically, chemically, biologically, or even through a combination of these methods (Praveen et al. 2022; Chau et al. 2023). Various physical methods such as sedimentation, adsorption, membrane filtration, and ion exchange are generally used to treat dye-containing wastewater as they are economical and easy to apply (Kant 2012; Benkhaya et al. 2022). Chemical treatments such as coagulation-flocculation, advanced oxidation processes and electro-chemical technology are known to be expensive and limited methods because of the high electrical energy requirements, large amounts of used chemicals, and the need for proper equipment. Moreover, the use of chemical approaches for dye removal releases toxic metabolites and by-products generated during the treatment process (Crini and Lichtfouse 2019; Wang et al. 2021). On the other hand, biological approaches to dye-containing wastewater treatment are economically feasible, environmentally safe, and the generated byproducts during the microbial metabolic activity are non-toxic, compared to other approaches (Jain et al. 2022). This type of treatment can be applied using one or several highly-biodegradable microorganisms where dye-containing wastewater, is dealt with under aerobic or anaerobic conditions (Liu et al. 2018). The two primary techniques utilized for dye decolorization treatment of textile wastewater are adsorption and degradation (Mahmood et al. 2016; Sarkar et al. 2017). Bacteria (Senelisile et al. 2022), algae (Fekry et al. 2018), yeast (Dammak et al. 2022), and fungi (Ben Ayed et al. 2022a, b), as well as enzyme-based systems (Bello-Gil et al. 2018), represent feasible biological possibilities for converting dye molecules into non-toxic compounds in textile dye wastewater treatment. Fungi have been proven to be effective in the bioremediation of dyes (Ghariani et al. 2019; Ngo and Tischler 2022). The main advantage of using fungi for dye-containing wastewater treatment is their ability to accelerate their metabolism in order to achieve optimal environmental conditions (Zafiu et al. 2021). Enzymatic degradation is carried out by laccases and peroxidases (MnP, LiP and Dye-P) (Saratale et al. 2011; Ngo and Tischler 2022). Among these enzymes, laccase was the most popular due to its capacity to oxidize a broad range of substrates including lignin, dyes and antibiotics. Such capacities enabled its use in many fields such as pulp and paper, food, pharmaceutical and textile industries (Zhang et al. 2021; Ben Ayed et al. 2022a, b; Zofair et al. 2022; Kyomuhimbo and Brink 2023). To enhance its efficiency, laccase-mediated systems have been investigated to treat recalcitrant micro-pollutants and detoxify industrial sewage (Fillat et al. 2017; Parra Guardado et al. 2019; Chen et al. 2021; Zofair et al. 2022). In fact, the redox potential of white-rot fungi laccases can reach 800 mV, which is lower than that of nonphenolic compounds (up to 1500 mV). Thus, using a mediator can improve the electrons transfer between the substrate and laccase. The mediator can be first oxidized by the laccase into instable free radical which can oxidize nonphenolic substrates (Hilgers et al. 2018), so expanding the substrate range of laccase reactions. Previous studies reported the capacity of such a system to remove textile dyes and bisphenol A from water (Benzina et al. 2013; Daâssi et al. 2014, 2016).

The main objective of this study was the optimization of the degradation of two azo dyes; Sirius Blue (Direct Blue 71) and Sirius Red (Direct Red 80) using laccase-like active cell-free supernatant from Coriolopsis gallica. The optimization study was performed using Box-Behnken experimental design with three levels and four factors, i.e., pH, HBT concentration, laccase activity, and dye concentration. Two responses were measured: the decolorization yield and rate. The toxicity of the treated dyes was also assessed by measuring the germination index.

Materials and methods

Fungal strain and culture conditions

The Coriolopsis gallica, strain BS 9 (OR234862) was used for laccase production. For the short-term conservation, the strain was maintained in a PDA medium. The liquid pre-culture was made up of 10 mL of M7 (pH 5.5) which contains (g/L): glucose 10, peptone 5, yeast extract 1, ammonium tartrate 2, KH2PO4 1, MgSO4 0.7H2O 0.5, KCl 0.5, and 1 mL of trace element solution which encloses (in g/L): B4O₇Na₂ · 10H₂O 0.1, CuSO₄ · 5H₂O 0.01, FeSO₄ · 7H₂O 0.05, MnSO₄ · 7H₂O 0.01, ZnSO₄ · 7H₂O 0.07, (NH4)6Mo7O24 · 4H₂O 0.01 (Berrio et al. 2007). For the preculture inoculation, 4 agar plugs containing the strain were cut from the plate stock and then the 100 mL flasks were incubated at 30 °C and 160 rpm for 4 days. The mycelium was then partially fragmented using glass beads (0.6 mm). The obtained suspension was later used to inoculate 250 mL (1 L flask) of M7 medium containing 300 µM CuSO₄. The incubation was performed over a period of 8–10 days in the same conditions as the precultures (at 30 °C and 160 rpm).

Dyes

The dyes used in this study were obtained from a textile company based in “Ksar Helal”, Tunisia. Their chemical properties are displayed in Table 1.

Table 1.

Chemical properties of Sirius Blue and Sirius Red

Properties	Sirius blue (direct blue 71)	Sirius red (direct red 80)
CAS number	4399-55-7	2610-10-8
Molecular weight (g/mol)	1029.87	1373.07
EC number	224-531-4	220-027-3
Color index	34,140	35,780
Chemical formula	C40H23N7O13S4 · 4Na	C45H26N10Na6O21S6
λmax (nm)	594	528
Molecular structure

Cell-free extract preparation and laccase-like activity assays

The fungal culture was filtered through Whatman filter paper. The resulting filtrate was concentrated with a dialysis membrane cut-off 5 KDa using PEG 12000 and then stored at − 20 °C. The enzymatic activity was determined based on the oxidation of 10 mM 2,6-dimethoxyphenol (DMP). The reaction was carried out at 469 nm, pH 5.5 in 100 mM tartrate buffer for 1 min. 50 µL of the laccase preparation was added later (Ben Ayed et al. 2022a, b).

Decolorization assays

To evaluate the capacity of Coriolopsis gallica laccase to decolorize Sirius Blue and Sirius Red, a preliminary decolorization test was carried out at pH 5.5, 50 mg/L of dye, 1 U/mL of laccase-like activity, and at room temperature (approximately 30 °C). The reaction mixture and conditions were then optimized by adding HBT at different concentrations (0.2–0.6–1 mM).

Response-surface methodology by box-Behnken design

In order to collect most of the information about the decolorization process of the two tested dyes, taking into account cost and time, the Response-Surface Methodology (RSM) was applied using the Box–Behnken design. This methodology allows the optimum determination for all the tested responses simultaneously using the determined multivariable model that has already been verified and proven to be statistically valid. In this case, the optimum is obtained as a function of the most interesting levels of all the tested variables for each dye according to the main aim of the current study.

The chosen responses for both dyes in this study were the decolorization yield (Y_{S Blue} and Y_{S Red}) and the decolorization rate (R_{S Blue} and R_{S Red}) determined experimentally through Eqs. (1), and (2), respectively.

$$\mathbf{Decolorization}\mathbf{yield}\left(\%\right)=\frac{{\mathbf{A}}_{{\mathbf{t}}_0}-{\mathbf{A}}_{{\mathbf{t}}_{4\mathbf{h}}}}{{\mathbf{A}}_{{\mathbf{t}}_0}}\times 100$$ 1

where A_t0 and A_t4h are absorbances of the dye before and after 4h of incubation with the enzyme. The absorbances were determined at the maximum wavelengths 528 and 594 nm for Sirius Red and Sirius Blue, respectively. All the experiments were performed in a total volume of 1.5 mL reaction mixture in 100 mM tartrate buffer for 15 min. The initial decolorization percentage was determined based on Eq. (1) replacing A_{t 4h} with A_{t 15 min}.

$$\mathbf{Decolorization}\mathbf{rate}(\%/\mathbf{min})=\frac{\mathbf{initial}\mathbf{decolorization}(\%)}{\mathbf{time}}$$ 2

Four factors were tested three different times changing the quantity of the HBT, enzyme, and dye concentrations as well as pH as follows: HBT concentration (0.2, 0.6, and 1 mM); enzyme concentration (0.2, 0.6, and 1 U/mL); dye concentration (50, 100, 150 mg/L); and pH (3, 4.5, and 6). Each of these experiments was repeated three times achieving a total of 27 tested runs were determined by all possible combinations via the Box–Behnken design as shown in Table 2.

Table 2.

Box-Behnken experiment design applied to the decolorization process

Run	HBT (mM)	pH	Dye (mg/L)	Enzyme (U/mL)
1	0.2	3	100	0.6
2	1	3	100	0.6
3	0.2	6	100	0.6
4	1	6	100	0.6
5	0.6	4.5	50	0.2
6	0.6	4.5	150	0.2
7	0.6	4.5	50	1
8	0.6	4.5	150	1
9	0.2	4.5	100	0.2
10	1	4.5	100	0.2
11	0.2	4.5	100	1
12	1	4.5	100	1
13	0.6	3	50	0.6
14	0.6	6	50	0.6
15	0.6	3	150	0.6
16	0.6	6	150	0.6
17	0.2	4.5	50	0.6
18	1	4.5	50	0.6
19	0.2	4.5	150	0.6
20	1	4.5	150	0.6
21	0.6	3	100	0.2
22	0.6	6	100	0.2
23	0.6	3	100	1
24	0.6	6	100	1
25	0.6	4.5	100	0.6
26	0.6	4.5	100	0.6
27	0.6	4.5	100	0.6

All the responses were adjusted using a second-order polynomial equation, and multiple regressions were performed on the data to obtain an empirical template associated with the factors (Eq. 3).

$${\widehat{\mathbf{y}}}_{\mathbf{k}}={\beta }_0+\sum _{\mathbf{i}}{\beta }_{\mathbf{i}}·{\mathbf{x}}_{\mathbf{i}}+\sum _{\mathbf{ii}}{\beta }_{\mathbf{ii}}·{\mathbf{x}}_{\mathbf{i}}^2+\sum _{\mathbf{i}}\sum _{\mathbf{j}>\mathbf{i}}{\beta }_{\mathbf{ij}}·{\mathbf{x}}_{\mathbf{i}}·{\mathbf{x}}_{\mathbf{j}}$$ 3

where ŷ_k is the calculated responses with the determined model; β₀, β_i, β_ij and β_ii are the model’s intercepts, linear, interactions, and quadratic coefficients, respectively; x_i is the level‐coded factor variable, and the subscript 1 ≤ i ≤ 4 corresponds to the tested factors.

Design and statistical analysis

The Minitab^® 19.2020.1 Statistical Software (64-bit) (© 2020 Minitab, LLC All rights reserved) was used for the experimental design, statistical analysis, figures drawing, and the determination of the optimums. The coefficients of the factors were determined using the least‐squares method. Student’s t‐test was used to identify the level of significance of both the fitted model and the factors influences and their interactions. A probability level that is less than 0.05 (p < 0.05, corresponding to 95%confidence level) indicates that the model and the factors influences were significant. The quality of the model was analyzed using the coefficient of determination (R²), the adjusted coefficient of determination (R²_adj), and the root-mean squared error (RMSE).

Phytotoxicity assay

The phytotoxicity assay was carried out according to Benzina et al. (2013) with slight modifications. The germination test was performed with radish seeds (Raphanus sativus). A Whatman filter paper was soaked with 2 mL of sterile distilled water in a Petri dish. 5 mL of the solution (treated and untreated dyes) was finally added and 10 seeds were spread on the paper. Incubation took place in the dark at 22 °C for 7 days. The Germination Index GI was calculated as follows:

$$\mathbf{GI}\left(\%\right)=\frac{(\mathbf{number}\mathbf{of}\mathbf{germinated}\mathbf{seeds}\mathbf{in}\mathbf{sample})\times (\mathbf{average}\mathbf{roots}\mathbf{length}\mathbf{of}\mathbf{treated}\mathbf{seedlings})}{(\mathbf{number}\mathbf{of}\mathbf{germinated}\mathbf{seeds}\mathbf{in}\mathbf{control})\times (\mathbf{average}\mathbf{roots}\mathbf{length}\mathbf{of}\mathbf{control}\mathbf{seedlings})}\times 100$$ 4

Results and discussion

Decolorization without a mediator

In order to determine the capacity of the cell-free extract to decolorize the azo dyes, the decolorization was performed without HBT first. The decolorization was performed at room temperature (30 °C), pH 5.5, 50 mg/L of dye, and with 1 U/mL of laccase-like activity. Sirius Red and Sirius Blue were decolorized respectively to 72.72 ± 0.24% and 81.22 ± 1.02%. These results seemed to be impressive compared to other azo dyes (Forootanfar et al. 2016). This effectiveness could be explained by the high redox potential of the used laccase. In fact, it is known that basidiomycete white-rot fungi have a high redox potential which enables them to efficiently oxidize the substrates with high redox potential. Indeed, the efficiency as well as the rate of the oxidation reaction are governed by the electron transfer between the substrate and the T1Cu site, which is the binding site within the laccase's active site (Mateljak et al. 2019; Takur et al. 2022).

Optimization of the decolorization of Sirius Blue and Sirius Red

The Box-Behnken design was used to optimize the biotreatment of Sirius Blue and Sirius Red in the presence of HBT by laccase like activity of Coriolopsis gallica.

Table 3 displays the obtained results of Sirius Blue and Sirius Red decolorization yields by mediated laccase enzyme (with already very low standard deviation values for all the experiments 0.06–1.83% for Sirius Blue and 0.02–2.71 For Sirius Red). The obtained results after repeating the experiments were almost similar, with slight negligible differences, confirming their stability. Compared to other enzymes used for the removal of Sirius Blue, (Mehrabian et al. 2018; Yanto et al. 2019), the enzyme used in our experiment seems to have reached valuable results, between 81.43 ± 0.26 and 97.33 ± 0.16%. Ben Ayed et al. (2022a, b) also reached a removal rate of 82% of the Reactive Black 5 azo dye treated with the same enzyme. On the other hand, Sirius Red decolorization yields by the same studied enzyme went from 26.65 ± 0.02 to 95.94 ± 0.21.

Table 3.

Decolorization yields (Y in%) and rates (R in %/min) of Sirius Blue and Sirius Red

Run	Sirius blue		Sirius red
	YS Blue (%)	RS Blue (%/min)	YS Red (%)	YS Red (%/min)
1	84.52 ± 0.16	5.43 ± 0.01	43.79 ± 2.50	1.64 ± 0.06
2	88.09 ± 0.27	5.56 ± 0.01	71.73 ± 2.65	2.29 ± 0.12
3	87.64 ± 0.12	5.25 ± 0.02	77.22 ± 2.71	1.93 ± 0.06
4	91.91 ± 0.19	5.69 ± 0.03	90.90 ± 0.76	5.09 ± 0.13
5	96.44 ± 0.08	6.15 ± 0.02	92.65 ± 1.76	5.75 ± 0.08
6	94.33 ± 0.08	5.62 ± 0.00	94.87 ± 0.27	5.18 ± 0.08
7	97.09 ± 0.16	6.32 ± 0.02	94.61 ± 0.12	5.92 ± 0.01
8	96.17 ± 0.11	6.13 ± 0.01	95.94 ± 0.21	5.66 ± 0.02
9	94.20 ± 0.06	5.60 ± 0.01	93.46 ± 0.47	3.30 ± 0.26
10	96.72 ± 0.06	6.05 ± 0.01	94.77 ± 0.24	5.60 ± 0.05
11	96.21 ± 0.09	6.09 ± 0.01	95.14 ± 0.18	5.50 ± 0.00
12	97.33 ± 0.16	6.28 ± 0.01	94.47 ± 0.29	5.79 ± 0.04
13	90.18 ± 0.33	5.66 ± 0.02	84.74 ± 0.06	3.05 ± 0.05
14	91.45 ± 0.57	5.78 ± 0.01	83.06 ± 1.44	4.04 ± 0.43
15	81.43 ± 0.26	5.30 ± 0.01	40.46 ± 0.71	1.59 ± 0.01
16	88.86 ± 0.13	5.34 ± 0.00	84.04 ± 1.51	2.69 ± 0.20
17	96.54 ± 0.28	6.07 ± 0.02	95.44 ± 0.18	5.58 ± 0.13
18	96.67 ± 0.41	6.27 ± 0.03	89.74 ± 0.76	6.01 ± 0.05
19	93.94 ± 0.14	5.73 ± 0.04	93.17 ± 0.11	3.64 ± 0.11
20	96.08 ± 0.49	6.02 ± 0.08	95.38 ± 0.05	5.56 ± 0.03
21	82.23 ± 0.19	5.32 ± 0.01	26.65 ± 0.02	0.73 ± 0.03
22	87.17 ± 0.44	5.03 ± 0.06	79.31 ± 2.25	1.31 ± 0.14
23	86.70 ± 1.83	5.47 ± 0.15	73.36 ± 0.82	2.90 ± 0.02
24	91.12 ± 0.12	5.86 ± 0.04	92.58 ± 0.25	5.07 ± 0.02
25	96.53 ± 0.23	6.17 ± 0.00	95.12 ± 0.48	5.63 ± 0.06
26	96.67 ± 0.24	6.15 ± 0.01	95.28 ± 0.32	5.63 ± 0.03
27	96.86 ± 0.08	6.18 ± 0.01	95.68 ± 0.32	5.71 ± 0.03

Moreover, the results of Sirius Blue decolorization rates achieved very high values: 5.03 ± 0.06–6.32 ± 0.02%/min (with very low standard deviation values: 0–0.15%/min). This proves that the enzymatic decolorization of Sirius Blue with the studied enzyme was very fast compared to other dyes. For example, the removal rates of Direct Black 166 and Direct Yellow 107 were both 1.34%/min (Forootanfar et al. 2016). As for Sirius Red, was more recalcitrant since its decolorization yield was lower. In addition, the decolorization rate in the case of Sirius Red was slower since only 15 among 27 experiments reached a higher rate than 5.03%/min (the minimum rate of decolorization in the case of Sirius Blue). Additionally, the standard deviation of this parameter is also very low (0.00–0.43%/min), indicating some stability of the decolorization process kinetics for the second dye (Sirius Red) regardless of the incubation conditions. These results can be explained by the differences in the molecular structure and weight (Singh et al. 2020). In fact, the molecular weight of Sirius Red, (a tetra azo dye), 1373,07 g/mol is higher than that of Sirius Blue, (a triazo dye), 1029.87 g/mol, (Table 1).

Compared to other azo dyes, especially mono and diazo ones, the obtained results were quite satisfactory. For example, the decolorization percentage of Acid Red 18, a mono-azo dye which treatment should be easier, reached 97% in 15 min (6.47%/min) (Ashrafi et al. 2013).

To test the model quality, R² and R²_adj of decolorization yields and rates of both dyes were determined (Table 4).

Table 4.

Model summary

	Sirius blue		Sirius red
	YS Blue (%)	RS Blue (%/min)	YS Red (%)	RS Red (%/min)
R2 (%)	95.55	98.17	84.20	93.18
R2adj (%)	94.60	97.18	80.85	91.74
RMSE	1.13 (unit of YS Blue)	0.055 (unit of RS Blue)	1.13 (unit of YS Res)	0.055 13 (unit of RS Red)

R² and R²_adj obtained for both responses and for the two dyes were quite high. The values of model regression parameters indicate proper fitness of the model. In fact, R² and R²_adj are statistical parameters that determine the goodness of fit of the used model. R² measures how well the regression model captures the patterns in the data. The closer its value to 100%, the better the model translates the relationship between the variables. While R²_adj is an adjusted version of R² that highlights the issue of overfitting. Unlike R², it only adds new predictors if it improves the model’s predicting power.

However, the root-mean squared error (RMSE) of all responses is very low. This means that the identified models have minimum errors between the experimental results (real results) and the calculated ones for all the responses and for all of the applied conditions.

All the achieved results using the chosen statistical coefficients (R²_, R²_adj and RMSE) confirm the suitability of all the identified models applied in this work for all of the responses (Y_{S Blue}, R_{S Blue}, Y_{S Red} and R_{S Red}).

Decolorization yield and rate modeling

Table 5 provides the determined coefficient of Eq. 3 of all of the studied responses using the statistical Student t-test: decolorization yields (Ŷ_{S Blue} and Ŷ_{S Red}) and rates (Ȓ_{S Blue} and Ȓ_{S Red}) for Sirius Blue and Sirius Red dyes. The equations show the coefficients of the main effects, interactions between variables and quadratic effects.

Table 5.

Student t-test results and coefficients of the regression equation

Response term	ŶS Blue (%/min)	ȒS Blue (%/min)	ŶS Red (%/min)	ȒS Red (%/min)
Constant	96.69***	6.165***	95.36***	5.658***
Linear
HBT concentration (x1)	1.15***	0.143***	3.23*	0.730***
pH (x2)	2.08***	0.018	13.87***	0.661***
Dye concentration (x3)	− 1.46***	− 0.177***	− 3.03*	− 0.502***
Enzyme concentration (x4)	1.13***	0.199***	5.37***	0.748***
Square
x1 · x1	− 0.15	− 0.077***	− 0.67	− 0.330**
x2 · x2	− 8.72***	− 0.623***	− 24.10***	− 2.780***
x3 · x3	− 0.24	− 0.034*	0.44	0.014
x4 · x4	− 0.67*	− 0.094***	− 1.60	− 0.234
Interaction
x1 · x2	0.18	0.078***	− 3.56	0.628***
x1 · x3	0.50	0.021	1.97	0.370*
x1 · x4	− 0.35	− 0.064***	− 0.49	-0.502**
x2 · x3	1.54***	− 0.021	11.31***	0.028
x2 · x4	− 0.13	0.168***	− 8.36**	0.397**
x3 · x4	0.30	0.087***	− 0.22	0.078

(·): p > 0.05: No significant effect; (*): p < 0.05 Significant effect; (**): p < 0.01 Very significant effect; (***): p < 0.001 Very high significant effect

For Sirius Blue, the four studied factors were equally significant, for both responses, with p-value less than 0.001 corresponding to a very high level of statistical significance, except for pH for the decolorization rate (p-value more than 0.05). For the quadric effects, only two were significant for the dye decolorization (in %), where the p-values were less than 0.05 and 0.001 (initial enzyme concentration and pH, respectively). For the decolorization rate, however, the four factors were effective (p-value less than 0.001 for all the factors other than the initial dye concentration). Concerning the interactions between the variables, few interactions were highly significant for both responses (p-value < 0.001), for example pH × initial dye concentration for Ŷ_{S Blue}, and pH × initial enzyme concentration for Ȓ_{S Blue}.

As for Sirius Red, the four factors were proven to be important, too with a p-value less than 0.001 for both responses Ŷ_{S Red} and Ȓ_{S Red}, except for the initial HBT and dye concentrations which had p-values < 0.005 for Ŷ_{S Red} (Table 5).

The most effective factor for the decolorization yields of both dyes was the pH; the initial enzyme concentration was also effective but to a lesser extent. On the other hand, the impact of the four factors on the decolorization rate of both dyes was roughly similar but negative for the dye concentration.

Effects of the individual factors

In order to determine the importance and effect of the studied factors on the decolorization yields and rates, the main effects plots (Fig. 1) were studied and analyzed. The variation of the pH affected both responses for the two dyes. Y_{S Blue}, Y_{S Red} (decolorization percentages of Sirius Blue and Sirius Red, respectively), R_{S Blue}, and R_{S Red} (decolorization rates of Sirius Blue and Sirius Red, respectively) reached their maximum slightly after level 0 (pH 4.5), approximately at pH 5. These results are in agreement with previous studies that reported that fungal laccase-mediated decolorization is generally more effective at slightly acidic conditions (Forootanfar et al. 2016; Singh et al. 2020). However, at an alkaline pH, the hydroxide anion can bind to the T2/T3 coppers of the enzyme which interrupt the electron transfer from T1 to T2/T3 centers (Forootanfar et al. 2016; Singh et al. 2020). In addition, the laccase of Coriolopsis gallica, as well as that of Trametes trogii (which are phylogenetically related), were demonstrated to be active and stable at pH between 4 and 7 (Ben Ayed et al. 2022a, b). As for HBT, increasing its initial concentration improved the decolorization yield and rate for both dyes (Fig. 1). However, when tested at level 1 (1 mM HBT), the decolorization rate of Sirius Blue and Sirius Red become more stable, which can be caused by the product saturation together with the presence of less product. In fact, the concentration of radicals generated from the oxidation of the mediator by the enzyme might be high, which might limit the access of the substrate to the enzyme (Brooks et al. 2012). Some studies have proved also that at high concentration, the mediator can affect the decolorization process negatively (Benzina et al. 2013; Forootanfar et al. 2016; Hirai et al. 2006). Guardado et al. (2019) explained that an excess of the mediator can produce more radicals which may attack the laccase active sites (competitive inhibition between the radicals and the substrate).

Fig. 1.

Main effects plot for Sirius Blue decolorization (a) and its rate (b), as well as Sirius Red decolorization (c) and its rate (d) after 4 h as a function of the coded value of factors

Concerning the enzyme effect, the higher its initial concentration was; the higher the decolorization percentages/rates were, Y_{S Blue}, Y_{S Red}, R_{S Blue,} and R_{S Red}. This particular result was evident since there were more active sites available and the dye could be rapidly broken. This result is consistent with data from (Ben Ayed et al. 2022a, b). The dye concentration affected the responses negatively for both dyes: the higher the initial concentration was the lower the decolorization rates and yields were for Sirius Blue and Sirius Red. In fact, the quantity of the molecules competing for the restricted number of enzyme active sites increased as the substrate concentration increased. Furthermore, steric hindrance or substrate inhibition may occur which reduces the enzyme's capacity to bind to the dye (Kokkonen et al. 2021).

Effect of the interactions between factors

To study the interactions between the factors and their effect, the response surfaces were prepared (Figs. 2, 3, 4, and 5). The effects of interaction between pH and other factors of Sirius Blue and Sirius Red are illustrated in Figs. 2a, c, e, 3a, c, e, 4a, c, e, 5a, c, and e.

Fig. 2.

Interaction effects in the decolorization yield of Sirius Blue: a HBT × pH, b Dye × HBT, c pH × Dye, d Dye × Enzyme, e pH × Enzyme, and f HBT × Enzyme

Fig. 3.

Interaction effects in the decolorization rate of Sirius Red: a HBT × pH, b Dye × HBT, c pH × Dye, d Dye × Enzyme, e pH × Enzyme, and f HBT × Enzyme

Fig. 4.

Interaction effects in the decolorization yield of Sirius Red: a HBT × pH, b HBT × Dye, c pH × Dye, d Dye × Enzyme, e pH × Enzyme, and f HBT × Enzyme

Fig. 5.

Interaction effects in the decolorization rate of Sirius Red: a HBT × pH, b Dye × HBT, c Dye × Enzyme, d pH × Dye, e pH × Enzyme, and f HBT × Enzyme

It was shown that increasing the pH to approximately 5 reduced the decolorization and its rate for both dyes, whatever the HBT concentration was. This was also the case with the interactions between pH × Dye (Figs. 2c, 3c, 4c, and 5 c) or pH × Enzyme (Figs. 2e, 3e, 4e, and 5 e). In fact, pH can affect HBT, laccase, and dyes similarly. Forootanfar et al. (2016) had explained the enzyme activity decrease by the possibility of binding hydroxide anions at higher pH, which interrupts the electron transfer. Besides, at higher pH, the dye structure might be affected. For example, acid azo dyes may become deprotonated, which decreases their solubility in water. On the other hand, basic azo dyes become positively charged, affecting their interactions with laccase and the HBT (Aksu and Tezer 2005). This may be the case with the mediator, which redox properties can be affected by changes in pH, leading to a decrease in its effectiveness. For instance, at higher pH values, the solution becomes depleted of protons, which can cause compounds like HBT to become deprotonated as a means to stabilize the system. Figures 2b, 3b, 4b, and 5b show the effect of HBT and dye concentration on both responses for both dyes. Achieving over 95% and 6%/min for decolorization yield and rate, respectively, was possible by increasing the HBT concentration to 1 mM (level 1) and reducing the dye concentration to 50 mg/L (level − 1). In fact, by reducing the dye concentration, the remaining substrate molecules become easier for the laccase to target and oxidize. The electron transport may be improved by combining this with a greater mediator concentration. This result was expected since it is consistent with fundamental ideas in enzyme kinetics and mediator-assisted processes. Figures 2d, 3d, 4d, and 5d illustrate the effect of the dye when combined with the enzyme. The highest decolorization percentage was reached at the highest enzyme level and at the lowest dye level at fixed pH and HBT concentrations. In fact, this is also expected since there is enough enzyme for the substrate, in this case. These results are in accordance with those found by (Yadav et al. 2021). The HBT concentration effect and that of the enzyme are illustrated in Figs. 2f, 3f, 4f, and 5f. Maximizing these two variables at a fixed pH and dye concentrations (4.5 and 100 mg/L respectively) increased the responses to 100% and 98%, respectively for Sirius Red and Sirius Blue decolorization, and to ~ 6 and 6.25%/min decolorization rates for Sirius Blue and Sirius Red. Increasing the mediator concentration can improve the electron transfer through the amelioration of the efficiency of electron exchange between the dye and the laccase active sites. Thus, more substrate can be oxidized in a shorter time (Gu et al. 2021). Similarly, increasing the enzyme concentration improves the substrate cleavage. In fact, the number of active sites will grow which will increase the number of dye molecules to treat and so the decolorization yield and rate. Moreover, the combination can have a synergistic impact on the decolorization process: the mediator functions as an electron shuttle efficiently carrying electrons, while the enzyme provides the catalytic power to oxidation (Mehrabian et al. 2018). But this growth is generally limited, as at some high concentrations of both factors (mediator and laccase), the decolorization may decrease mainly because of the substrate limitation or active site saturation (Benzina et al. 2012; Çifçi et al. 2019).

Responses optimization

In order to determine the most appropriate experimental conditions maximizing both decolorization yields and rates for the two dyes at the same time, the optimizer tool of Minitab 19 Software was used. All determined relationships between tested factors and the different responses (in precedent paragraph 2) were drawn graphically in Fig. 6. Maximizing all the responses in terms of decolorization yields and rates yielded the following conditions: initial HBT concentration: 0.62 mM; pH: 4.5–4.6; initial dye concentration: 50 mg/L and initial enzyme concentration: ~ 1 U/mL. Using these conditions, we can obtain via the equation models of all the responses: 98.18% and 100% of decolorization yields for Sirius Blue and Sirius Red, respectively, and 6.35%/min and 6.57%/min of decolorization rates for Sirius Blue and Sirius Red, respectively.

Fig. 6.

Optimized condition-set for Sirius Blue and Sirius Red decolorization

The obtained results were checked experimentally by applying the determined factors’ level and carrying out the experimental responses (the decolorization yield and rate for both dyes) results in triplicates. The achieved results are provided in Table 6.

Table 6.

Experimental and estimated responses for decolorization yields (Y_{S Blue} and Y_{S Red}) and rate (R_{S Blue} and R_{S Red}) for Sirius Blue and Sirius Red

	YS Blue (%)	RS Blue (%/min)	YS Red (%)	RS Red (%/min)
Experimental results	98.30 ± 0.10	5.84 ± 0.01	99.34 ± 0.47	5.85 ± 0.12
Estimated results	98.18	6.35	100	6.57

The decolorization yields for both dyes were approximately the same as the estimated ones. However, the decolorization rates were slightly different. This can be explained by the variation of the temperature as the experiments were performed at room temperature. In fact, temperature was shown to be an effective factor in many studies (Benzina et al. 2013; Forootanfar et al. 2016; Singh et al. 2020).

Phytotoxicity

The phytotoxicity analysis was carried out using germination test. Radish seeds were incubated with treated and untreated dyes. The test results are displayed in Fig. 7.

Fig. 7.

Toxic effects of Sirius Red, Sirius Blue and biodegraded products on the germination index of Raphanus sativus (T: treated dye; NT: untreated dye)

Figure 7 illustrates that the non-treated Sirius Red was more phytotoxic than the non-treated Sirius Blue, with GI of 14.6% and 31.60%, respectively. This difference can be explained by the chemical structure of the dyes, with Sirius Red being a tetra azo dye and Sirius Blue being a tri azo dye. In fact, Gooch et al. (2017) proved that the number of nitro groups is one of the factors that affect the toxicity of nitroaromatic molecules. Besides, according to (Hashemi and Kaykhaii 2022), the presence of nitro groups contributes to the production of aromatic amines and toxic molecules. On the other hand, the higher molecular weight of the first dye compared to the second one can potentially explain the observed differences (Singh et al. 2020).

Figure 7 also shows that, compared to the control, laccase-treated dyes were less toxic than the untreated ones. In fact, the cleavage of the parent molecules might produce less phytotoxic by-products. These findings are in line with the results achieved by (Rawat et al. 2018).

Conclusion

The main purpose of the current study was to optimize the decolorization of two recalcitrant dyes, Sirius Blue and Sirius Red, and maximize their decolorization yields and rates at the same time. The factors to optimize were the pH and the initial concentrations of HBT, dyes and enzyme. Results showed that the four studied factors had some limitations in the treatment of the two dyes, and that the optimum conditions were 4.5–4.6, 0.62 mM, 50 mg/L, 0.89 U/mL for pH, HBT concentration, dye concentration and enzyme concentration, respectively. The model was proved significant: For both dyes and responses, the determination coefficient (R²) values ranged between 84.20 and 98.17%. The adjusted R-squared (R²adj) values ranged between 80.85 and 97.18%. Additionally, the root mean square error (RMSE) ranged between 0.055 and 1.13. The germination test showed that the treatment did not totally eliminate the phytotoxicity. However, the germination index increased from 31.6 to 36.96% and from 14.6 to 36.08% for Sirius Blue and Sirius Blue, respectively. The mediator HBT enhanced the removal of the dyes and the generated metabolites seemed to be toxic towards the tested seeds. Further studies need to be carried out targeting such issues as toxicity removal, the toxicity of the generated products against other organisms using, for instance, cytoxicity and microtoxicity tests. These studies may also investigate other potential less costly and less toxic redox mediators than HBT such as Syringaldehyde, Vanillic acid, and/or Sinapic acid.

Author contributions

JB: Investigation, Writing original draft. IBA and BG: Revision of the manuscript. TM: Methodology, Conceptualization, Revision of the manuscript. BH: Conceptualization, Methodology, Formal analysis, Visualization. HZH: Supervision, Investigation. All authors read and commented on the manuscript.

Funding

This research was funded by the projects PHC-Utique CMCU 22/G0814 and PHC-Maghreb 23MAG31.

Data availability

All data are included within manuscript. Any other information about data will be made available on request.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest in the publication.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Ethical approval

Not applicable.