An indigenous dye-decolourising bacterium Micrococcus endophyticus (ES37) was isolated from dye contaminated soil and identified by 16S rDNA sequencing. The bacterial strain ES37 exhibited 97.19% of dye removal capacity in Luria-Bertani broth composition within 48 h, while the culture containing yeast extract showed 53.4% decolourisation in 72 h. In the absence of carbon and nitrogen sources, the bacterial strain failed to decolourise the dye, even on extended incubation. The effect of environmental factors on decolourisation was investigated by Plackett–Burman design and the significant parameters were lactose, yeast extract and pH. Optimisation of these factors was done by response surface methodology with central composite design; the decolourisation ranged from 0.43 to 77.49%. The optimised levels of lactose, yeast extract and pH were found to be 0.85% (w/v), 0.71% (w/v) and 7.5%, respectively. Under the optimal conditions, decolourisation of remazol golden yellow by ES37 strain was 81.61%, which was in agreement with the predicted value of 79.99%. These findings revealed the interactions and importance of environmental factors on dye decolourisation using native bacteria and also their standard point for an effective dye removal process.

ABBREVIATIONS

  • ANOVA

    analysis of variance

  • BOD

    biological oxygen demand

  • CCD

    central composite design

  • COD

    chemical oxygen demand

  • DOE

    design of experiments

  • PBD

    Plackett–Burman design

  • PCR

    polymerase chain reaction

  • RSM

    response surface methodology

INTRODUCTION

Nowadays, management of water pollution is one of the major environmental challenges for ecologists due to rapid industrialisation. Enormous amounts of chemicals including dyes are manufactured and utilised in materials, such as textiles, printing, pharmaceuticals, foods, toys, papers, plastics and cosmetics (Moosvi et al. 2007). Among these, discharge of coloured effluents from dye manufacturing units and textile processing industries impose the majority of environmental problems. The effluents from these dyeing industries increase the biological oxygen demand (BOD) and chemical oxygen demand (COD) levels in water bodies. The quantity and composition of effluents depend upon the dyeing process and the type of dye being used. Among all dyes, the azo group of dyes comprises about 70% occurrence in effluents due to its wide range of application in textile industries (Khouni et al. 2012).

Prior to dyeing, grey fabrics are consigned to washing, scoring, desizing and bleaching practices (Aly et al. 2010). In textile industries, high concentrations of salt are utilised for fixation of fibre reactive dyes on fabric materials. Textile wastewater also contain heavy metals, which are combined with the azo dye complex in order to resist biological degradation; this causes serious health hazards when it comes in contact with domestic water (Cooper 1995). Draining of coloured effluents into an aquatic environment generates a foul smell, reduces oxygen levels, sunlight penetration and pH, which in turn affects the natural ecosystem (Asgher et al. 2009; Molina-Guijarro et al. 2009). Enormous quantities of textile wastewater also pollute groundwater and soil fertility (Pourbabaee et al. 2005).

The dyes that are released into the water bodies are carcinogenic and mutagenic in nature and are hoarded in living organisms through bioaccumulation (Lima et al. 2007). For instance, the presence of monoazo and anthoquinone groups in dyes are found to cause allergic reactions (Vander Zee et al. 2003). In the human body, the liver and kidneys metabolise azo dye, using the azo reductase enzyme, into aromatic amines; this is then excreted in the urine. Sometimes, it causes bladder cancer and induces tumour formation (Brown & Devito 1993).

Currently, various physical, chemical and biological treatment technologies have been implemented for removal of colour from the effluents of textile dyeing industries (Mahmoued 2010). The main drawbacks in the implementation of conventional physical and chemical methods are the high cost, low efficiency, inapplicability to a wide variety of dyes and also the generation of secondary wastes (Rauf & Salman Ashraf 2009). Microbial decolourisation is an eco-friendly and cost-effective method that produces less sludge compared to chemical decomposition processes (Verma & Madamwar 2003).

In the last few decades, much research has focussed on biodegradation or bioaccumulation of dyes using autochthonous micro-organisms, due to their ability to survive in higher pH, temperature, salt concentration and a nutrient deficient medium. Therefore, biological treatment of dye removal, involving various operational parameters, must be analysed and optimised prior to on-site execution. Conventional optimisation methods have resulted in a large number of unreliable experiments at an incomplete stage (Senthilkumar et al. 2012).

Statistical tools, such as design of experiments (DOEs), have assisted in predicting the optimum response value with a few experiments (Montgomery 2001). Plackett–Burman design (PBD) and response surface methodology (RSM) application are used to find combined and individual effects of the significant factors that are involved in the process (Plackett & Burman 1946; Chen 1994). Mostly, they are used to reduce the chemical usage and the time involved in the process by optimisation (Ayed et al. 2011). Optimisation of dye decolourisation has specifically focussed on the factors playing an important role in large-scale treatment, by a bio-statistical approach using bacteria that was not explored well in dye decolourisation.

The present study aimed to explore the potentiality of native bacteria (which were isolated from dye contaminated soil) in the dye decolourisation process, and also to employ them in decolourisation of a reactive azo dye, remazol golden yellow.

MATERIALS AND METHODS

Isolation and screening of dye-decolourising bacteria

Dye contaminated soil was collected in sterile polyethylene bags from a textile industry located in Erode, Tamil Nadu, India. For bacterial isolation, 1.0 g of soil sample was subjected to dilution and cultured using a pour plate technique in nutrient agar plates containing 5 g/L peptone, 5 g/L sodium chloride, 3 g/L yeast extract, 20 g/L agar (pH 7.0 ± 0.2). Inoculated plates were incubated at 35 ± 2 °C and 45 ± 2 °C for 24–48 h in order to isolate mesophilic and thermophilic bacteria, respectively. All steps in the culture isolation and inoculation procedure were carried out in a sterile cabinet under laminar flow conditions. The distinct bacterial colonies were purified by repeated streaking on nutrient agar; their dye decolourisation ability was then tested. The dye decolourisation proficiency of refined bacterial isolates was tested by dot plate assay in Luria-Bertani agar plates containing 10 g/L casein enzymic hydrolysate, 5 g/L yeast extract, 10 g/L sodium chloride and 15 g/L agar amended with various concentrations (50, 100, 150, 200 and 250 mg/L) of remazol golden yellow dye (RNL). The inoculated plates were incubated at 37 °C for 4 days and the decolourising zones around the bacterial colonies were observed.

Identification of dye-decolourising bacteria

The selected feasible dye-decolourising bacterial strain ES37 was identified by the 16S rDNA sequencing method. The genomic DNA isolation, extraction, polymerase chain reaction amplification and 16S rDNA sequencing of the amplified gene were carried out at Xcelris Labs Ltd, Ahmedabad, India.

Phylogenic tree analysis and Genbank accession

The generated 16S rDNA consensus sequences were initially analysed in an NCBI server (www.ncbi.nlm.nih.gov/BLAST) using the BLASTN tool. Closely related sequences were obtained and aligned with the multiple alignment software program Clustal W. A phylogenetic tree was constructed by the neighbour joining (NJ) method in MEGA version 5.0 based on bootstrap analysis (1,000 replicates). The nucleotide sequences were deposited in the NCBI GenBank with accession number JX042488.

Decolourisation efficiency in various media

The decolourisation competence of selective bacterial strains was determined with various aqueous media containing a 100 mg/L concentration of remazol golden yellow dye, namely: (i) Luria-Bertani broth: 10 g/L casein enzymic hydrolysate, 5 g/L yeast extract, 10 g/L sodium chloride; (ii) yeast extract broth: 5 g/L yeast extract and 5 g/L sodium chloride; (iii) Bushnell and Hass broth: 0.2 g/L magnesium sulphate, 1 g/L di-potassium hydrogen phosphate, 0.02 g/L calcium chloride, 0.05 g/L ferric chloride, 1 g/L ammonium nitrate. About 1% (v/v) of 24 h culture was inoculated and incubated at 37 °C for 3 days, under static condition. Every 24 h, the samples were withdrawn and centrifuged at 5,590 × g for 20 min. Absorbance (OD) of the cell free supernatant was measured using UV spectrophotometer (Cyberlab UV-100, USA) at 412 nm. Experiments were conducted in triplicate; the mean values were used to determine the decolourisation percentage using Equation (1) as follows: 
formula
1

Optimisation of dye decolourisation via DOEs

DOEs is a bio-statistical tool, which reveals the individual and interactive effects of multiple factors involved in a process, by screening up to ‘n–1’ variables in just ‘n’ number of trials. Estimation of regression coefficients and prediction of the fittest model were carried using the statistical software Minitab Version 15 (Tripathi & Srivastava 2012).

Plackett–Burman design

The 2 k-factorial PBD trials were carried out in order to select the important factors that influence decolourisation of remazol golden yellow by M. endophyticus ES37 and also to evaluate the relative importance of the significant factors that are involved in the process. Scrutinised factors and their actual values are shown in Table 1. Plackett–Burman experimental design, depicted in Table 2, is based on the first-order model Equation (2) as follows: 
formula
2
where Y is the response, Xi is factor levels, i is the factor number, β0 is the model intercepts term, βi is the linear effect, βii is the squared effect, and βij is the interaction effect between Xi and Xj on the dye decolourisation process.
Table 1

Actual values of factors for PBD

Test variable Lactose % (w/v) Yeast extract % (w/v) pH Temperature (°C) Inoculums size % (v/v) Dye concentration (mg/L) Incubation period (h) 
Low level (−) 0.1 0.1 30 100 24 
High level (+) 1.0 1.0 45 10 300 72 
Test variable Lactose % (w/v) Yeast extract % (w/v) pH Temperature (°C) Inoculums size % (v/v) Dye concentration (mg/L) Incubation period (h) 
Low level (−) 0.1 0.1 30 100 24 
High level (+) 1.0 1.0 45 10 300 72 
Table 2

Plackett–Burman experimental design for screening of significant factors on dye decolourisation

Run order Lactose % (w/v) Yeast extract % (w/v) pH Temperature (°C) Inoculum size % (v/v) Dye concentration (mg/L) Incubation period (h) DV-1* DV-2* 
1.0 0.1 30 100 72 
1.0 1.0 45 100 24 
0.1 1.0 30 10 100 24 −1 
1.0 0.1 45 300 24 −1 −1 
1.0 1.0 45 10 100 72 −1 −1 
1.0 1.0 30 10 300 24 −1 
0.1 1.0 45 300 72 −1 
0.1 0.1 45 10 100 72 −1 
0.1 0.1 45 10 300 24 
10 1.0 0.1 30 10 300 72 −1 
11 0.1 1.0 30 300 72 −1 
12 0.1 0.1 30 100 24 −1 −1 
Run order Lactose % (w/v) Yeast extract % (w/v) pH Temperature (°C) Inoculum size % (v/v) Dye concentration (mg/L) Incubation period (h) DV-1* DV-2* 
1.0 0.1 30 100 72 
1.0 1.0 45 100 24 
0.1 1.0 30 10 100 24 −1 
1.0 0.1 45 300 24 −1 −1 
1.0 1.0 45 10 100 72 −1 −1 
1.0 1.0 30 10 300 24 −1 
0.1 1.0 45 300 72 −1 
0.1 0.1 45 10 100 72 −1 
0.1 0.1 45 10 300 24 
10 1.0 0.1 30 10 300 72 −1 
11 0.1 1.0 30 300 72 −1 
12 0.1 0.1 30 100 24 −1 −1 

*DV 1 and 2, dummy variable; +1, high concentration; −1, low concentration.

A Plackett–Burman experimental run was performed in a basal broth medium with abiotic controls in static incubation. Response percentage dye decolourisation was assayed. A positive or negative coefficient value of factors specifies their relative contribution to dye decolourisation (Du et al. 2012). A larger magnitude of Student t-test value with smaller P-value of factors indicates higher significance on dye decolourisation (Karthikeyan et al. 2010). Only the factors which had a P-value (probability > F) of less than 0.10 were considered for further experiments (Khelifi et al. 2012).

RSM

The significant factors involved in decolourisation were optimised using RSM and central composite design (CCD). RSM and CCD involve a group of trial systems employed to assess the correlation between experimental factors and the response. The selected factors, namely yeast extract, lactose and pH, were considered at five coded levels, −α, −1, 0, +1 and +α (Table 3); α = 2n/3, ‘n’ is the number of parameters and ‘0’ is the central point. The actual values of experimental factors were calculated by using Equation (3) (Paul et al. 1992): 
formula
3
Table 3

Actual values of the significant factors

  Five levels of variables
 
Variables Unit − α (−1.68179) − 1 + α (+1.68179) 
Yeast extract % (w/v) − 0.20681 0.1 0.55 1.306807 
Lactose % (w/v) − 0.20681 0.1 0.55 1.306807 
pH – 3.636414 10.36359 
  Five levels of variables
 
Variables Unit − α (−1.68179) − 1 + α (+1.68179) 
Yeast extract % (w/v) − 0.20681 0.1 0.55 1.306807 
Lactose % (w/v) − 0.20681 0.1 0.55 1.306807 
pH – 3.636414 10.36359 
The experimental model of CCD is shown in Table 4. All the other factors used in this study were constant: 0.5% (w/v) sodium chloride, 200 mg/L dye concentration, 18 h culture (OD 610 nm = 0.5) of 7.5% (v/v) inocula, 37 °C temperature and 48 h incubation period. The correlation and relationship of these factors and their measured response in CCD can be described by linear quadratic model as in Equation (4): 
formula
4
where Y is the percentage of decolourisation, β0 is the intercept term, β1, β2 and β3 are linear coefficients of each significant factor, β11, β22, β33 are quadratic coefficients, β12, β13, β23 are interaction coefficients and X1, X2, X3, X4 are coded factors. The best experimental model was determined by coefficient R2. A regression model equation was used to find the predicted optimal values of the selected factor and to subsequently reveal the interaction of these factors within a stipulated level (Zhao et al. 2010). The maximum decolourisation potential of the bacteria was verified under predicted optimal condition, obtained in RSM. Experiments were conducted in triplicate and the decolourisation percentage was compared to the predicted result from RSM.
Table 4

CCD matrix for optimisation of dye decolourisation

Run order Pt type Blocks Yeast extract % (w/v) Lactose % (w/v) pH 
0.1 0.1 
1.0 0.1 
0.1 1.0 
1.0 1.0 
0.1 0.1 
1.0 0.1 
0.1 1.0 
1.0 1.0 
−1 −0.20 0.55 
10 −1 1.31 0.55 
11 −1 0.55 −0.20 
12 −1 0.55 1.31 
13 −1 0.55 0.55 3.63 
14 −1 0.55 0.55 10.36 
15 0.55 0.55 
16 0.55 0.55 
17 0.55 0.55 
18 0.55 0.55 
19 0.55 0.55 
20 0.55 0.55 
Run order Pt type Blocks Yeast extract % (w/v) Lactose % (w/v) pH 
0.1 0.1 
1.0 0.1 
0.1 1.0 
1.0 1.0 
0.1 0.1 
1.0 0.1 
0.1 1.0 
1.0 1.0 
−1 −0.20 0.55 
10 −1 1.31 0.55 
11 −1 0.55 −0.20 
12 −1 0.55 1.31 
13 −1 0.55 0.55 3.63 
14 −1 0.55 0.55 10.36 
15 0.55 0.55 
16 0.55 0.55 
17 0.55 0.55 
18 0.55 0.55 
19 0.55 0.55 
20 0.55 0.55 

RESULTS AND DISCUSSION

Enumeration and screening of dye-decolourising bacteria

The total heterotrophic bacterial population of the textile dye contaminated soil was found to be 62 × 106 CFU/g. The presence of a high bacterial population in the dye-contaminated soil has drawn attention to the utilisation of native azo dye resistant bacteria in dye degradation (Rajee & Patterson 2011). Dye decolourisation efficiency of distinct bacterial colonies using plate assay at different concentrations, ranging from 50 to 250 mg/L of remazol golden yellow dye in a Luria-Bertani medium, was determined. Twenty bacterial colonies were observed to possess a positive zone of clearance at 100 mg/L concentration. The ratio of removal of dye by bacteria was directly propotional to the concentration of the dye (Jadhav et al. 2008). M. endophyticus ES37 produced clear zones in all the concentrations of dye and was scrutinised for further investigations. Elisangela et al. (2009) have emphasised that dye-decolourising micro-organisms from contaminated samples of the textile industry have maximum dye removal capacity when compared to the strains that are isolated from other sources.

Identification of dye-decolourising bacteria

The identification of the dye-decolourising bacterial strain was done by 16S rDNA sequencing method. The evolutionary history was analysed using the NJ method. The phylogenetic tree of the sequenced strain is shown in Figure 1. The tree was constructed from nucleotide sequences of ES37, which demonstrated 99% similarity with M. endophyticus YIM 56238 (accession number EU005372). Hence, the strain was identified as Micrococcus endophyticus ES37 (accession number JX042488). A number of Micrococcus sp. has been reported for decolourisation of textile dyes in the previous literature; however, no reports are available on decolourisation of dyes using M. endophyticus. Similar to M. endophyticus, the strains Micrococcus luteus (accession number AJ409096) and Micrococcus sp. (accession number AJ313024) also have the ability to decolourise the dye present in contaminated soil (Moosvi et al. 2007).
Figure 1

Phylogenetic tree of Micrococcus endophyticus (ES37) showing the relationship between selected bacterial strains. The per cent numbers at the nodes indicate the levels of bootstrap support based on NJ analyses of 1,000 replicates. Brackets represents sequence accession numbers.

Figure 1

Phylogenetic tree of Micrococcus endophyticus (ES37) showing the relationship between selected bacterial strains. The per cent numbers at the nodes indicate the levels of bootstrap support based on NJ analyses of 1,000 replicates. Brackets represents sequence accession numbers.

Decolourisation of remazol golden yellow in various compositions

Growth of micro-organisms mainly depends upon the supplementation of nutrients in media. Nutrients are necessary for microbial growth and cell metabolic activities (Moosvi et al. 2007). The decolourisation ability of M. endophyticus ES37 was observed using different media compositions. The strain showed 53.59% dye removal within 24 h. After additional incubation for 48 h, the bacterial cells achieved 97.19% decolourisation in Luria-Bertani broth composition. Saratale et al. (2010) reported that Micrococcus glutamicus NCIM-2168 required 42 h in order to achieve maximum decolourisation of dyes, which may be due to the higher molecular weight and presence of inhibitory groups in the dyes. The maximum decolourisation ratio in LB broth was enhanced by amendment of casein enzymic hydrolysate. In yeast extract broth, the culture exhibited 6.52% and 15.23% decolourisation after the incubation period of 24 h and 48 h, respectively. Finally, the strain achieved 53.4% decolourisation of remazol golden yellow dye in 72 h. Similar results were observed in the decolourisation of Orange MR dye at 900 ppm by Micrococcus sp. DBS 2; decolourisation of up to 36.34% has been reported (Rajee & Patterson 2011). The organic nitrogen source, contributed by the yeast extract in the media, is used to reduce azo bonds by regeneration of NADH as an electron donor (Chen et al. 2003). However, the extended incubation period of M. endophyticus ES37 in Bushnell Haas broth composition failed to decolourise remazol golden yellow dye after using it as the sole source of carbon and nitrogen. The absence of supplementary carbon and nitrogen sources in the bacterial culture exemplifies the essentiality of a co-substrate to support bacterial growth and decolourisation (Nigam et al. 1996).

Optimisation of dye decolourisation

Screening of significant factors by PBD

The influence of obligatory factors under static conditions, during decolourisation of remazol golden yellow by M. endophyticus ES37, was analysed using PBD. There was a wide variation in efficiency of remazol golden yellow decolourisation, ranging from 0.63 to 22.57% (Table 5). The statistical analyses of regression coefficients and estimated effects of the factors are given in Table 6. The coefficient values of yeast extract, pH and incubation period indicates that the increase in their concentration has a positive influence on the decolourisation process. Conversely, lower temperature, lactose concentration, inocula size and dye concentration are required to assist maximum dye removal ratio. Du et al. (2012) reported that yeast extract and pH had a positive effect on decolourisation of malachite green by Micrococcus sp. strain BD15, whereas a higher concentration of dye and higher temperature had negative effects on decolourisation; this supports the present observation. Anjaneya et al. (2011) pointed out that higher temperature leads to the loss of microbial metabolic activities during dye decolourisation. The necessity of using a lower concentration of dye and lactose in the culture medium may be related to the toxic effect on the bacterial cells, inhibition of azo reductase active sites and other enzyme systems involved in decolourisation (Tony et al. 2009; Jadhav et al. 2010). In the present study, factors possessing confidence levels greater than 90% were considered significant. A large t-value coupled with a low P-value specifies high significance of factors and model term.

Table 5

PBD of experimental and predicted decolourisation (%)

Run Lactose % Yeast extract %  Temperature Inoculum size % Dye concentration Incubation Percentage decolourisation
 
order (w/v) (w/v) pH (°C) (v/v) (mg/L) period (h) Experimental Predicted 
1.0 0.1 30 100 72 6.91 7.023 
1.0 1.0 45 100 24 7.28 7.89 
0.1 1.0 30 10 100 24 21.55 18.08 
1.0 0.1 45 300 24 4.88 3.7083 
1.0 1.0 45 10 100 72 8.17 7.56 
1.0 1.0 30 10 300 24 7.65 11.11 
0.1 1.0 45 300 72 22.57 20.56 
0.1 0.1 45 10 100 72 7.38 10.34 
0.1 0.1 45 10 300 24 1.64 1.79 
10 1.0 0.1 30 10 300 72 0.63 −1.85 
11 0.1 1.0 30 300 72 13.34 15.32 
12 0.1 0.1 30 100 24 5.08 5.43 
Run Lactose % Yeast extract %  Temperature Inoculum size % Dye concentration Incubation Percentage decolourisation
 
order (w/v) (w/v) pH (°C) (v/v) (mg/L) period (h) Experimental Predicted 
1.0 0.1 30 100 72 6.91 7.023 
1.0 1.0 45 100 24 7.28 7.89 
0.1 1.0 30 10 100 24 21.55 18.08 
1.0 0.1 45 300 24 4.88 3.7083 
1.0 1.0 45 10 100 72 8.17 7.56 
1.0 1.0 30 10 300 24 7.65 11.11 
0.1 1.0 45 300 72 22.57 20.56 
0.1 0.1 45 10 100 72 7.38 10.34 
0.1 0.1 45 10 300 24 1.64 1.79 
10 1.0 0.1 30 10 300 72 0.63 −1.85 
11 0.1 1.0 30 300 72 13.34 15.32 
12 0.1 0.1 30 100 24 5.08 5.43 
Table 6

Estimated effects of PBD

S. no. Variable Effect Coef. SE coef. T P 
Constant – 8.916 1.011 8.82 0.001* 
Lactose −6.018 −3.009 1.011 −2.98 0.041* 
Yeast extract 9.015 4.507 1.011 4.46 0.011* 
pH 5.782 2.891 1.011 2.86 0.046* 
Temperature −0.545 −0.272 1.011 −0.27 0.801 
Inoculum size −2.152 −1.076 1.011 −1.06 0.347 
Dye concentration −0.948 −0.474 1.011 −0.47 0.663 
Incubation period 1.822 0.911 1.011 0.90 0.418 
S. no. Variable Effect Coef. SE coef. T P 
Constant – 8.916 1.011 8.82 0.001* 
Lactose −6.018 −3.009 1.011 −2.98 0.041* 
Yeast extract 9.015 4.507 1.011 4.46 0.011* 
pH 5.782 2.891 1.011 2.86 0.046* 
Temperature −0.545 −0.272 1.011 −0.27 0.801 
Inoculum size −2.152 −1.076 1.011 −1.06 0.347 
Dye concentration −0.948 −0.474 1.011 −0.47 0.663 
Incubation period 1.822 0.911 1.011 0.90 0.418 

*Significant.

R2 = 90.74% R2(adj) = 74.53%.

The effect of significant factors on decolourisation is explained by regression Equation (5), which is used to find the predicted decolourisation (%) of any desired combination of factors in the experimental range. The fittest model was confirmed by the determination of correlation coefficient (R2 = 0.9074), which is nearer to 1; it is denoted for good statistical replication between the experimental and predicted responses. 
formula
5
Analysis of variance (ANOVA) of the linear model is presented in Table 7. This explains the presence of variations in the net result caused by each factor that is relative to the total variation. According to a Pareto chart (Figure 2) the screened factors, in the order yeast extract, lactose and pH, were most effective for remazol golden yellow dye decolourisation.
Table 7

ANOVA for PBD

S. No. Source DF Seq SS Adj SS Adj MS F P 
Main effects 480.188 480.188 68.598 5.60 0.058* 
Lactose 108.661 108.661 108.661 8.87 0.041* 
Yeast extract 243.811 243.811 243.811 19.90 0.011* 
pH 100.283 100.283 100.283 8.18 0.046* 
Temperature 0.891 0.891 0.891 0.07 0.801 
Inoculum size 13.889 13.889 13.889 1.13 0.347 
Dye concentration 2.698 2.698 2.698 0.22 0.663 
Incubation period 9.955 9.955 9.955 0.81 0.418 
Residual error 49.015 49.015 12.254 – – 
 Total 11 529.204     
S. No. Source DF Seq SS Adj SS Adj MS F P 
Main effects 480.188 480.188 68.598 5.60 0.058* 
Lactose 108.661 108.661 108.661 8.87 0.041* 
Yeast extract 243.811 243.811 243.811 19.90 0.011* 
pH 100.283 100.283 100.283 8.18 0.046* 
Temperature 0.891 0.891 0.891 0.07 0.801 
Inoculum size 13.889 13.889 13.889 1.13 0.347 
Dye concentration 2.698 2.698 2.698 0.22 0.663 
Incubation period 9.955 9.955 9.955 0.81 0.418 
Residual error 49.015 49.015 12.254 – – 
 Total 11 529.204     

*Significant.

Figure 2

Effects of influencing factors on dye decolourisation in a Pareto chart.

Figure 2

Effects of influencing factors on dye decolourisation in a Pareto chart.

RSM approach using CCD

The significant factors (yeast extract, lactose and pH) and their combined effect were studied by RSM for optimisation of remazol golden yellow decolourisation using M. endophyticus ES37. As per CCD, the dye removal rate varied from 0.43 to 77.49% (Table 8). Second-order polynomial model Equation (6) and CCD results were linked in order to elucidate the dependence of the dye removal ratio by multiple regression analysis. The estimated regression coefficient for the model is given in Table 9. The magnitude of coefficient had high significance (P = 0) for all factors during the decolourisation process. In addition, the correlation coefficient (R2) was determined by regression analysis and was found to be 83.43%, thus a higher value of R2 indicated a good correlation between the factors and CCD dye decolourisation process. 
formula
6
where Y is predicted dye decolourisation (%), and X1, X2 and X3 were the coded values of lactose, yeast extract and pH, respectively.
Table 8

CCD mediated decolourisation (%) with predicted values

    Percentage decolourisation
 
 
Trial Lactose (X1Yeast extract (X2pH (X3Experimental Predicted Residual 
0.1 0.1 22.09 6.12 15.96 
0.1 56.68 50.03 6.64 
0.1 31.55 21.54 10.00 
60.2 56.82 3.37 
0.1 0.1 27.76 30.10 −2.34 
0.1 51.21 60.18 −8.97 
0.1 40.2 45.81 −5.61 
52.33 67.26 −14.93 
−0.20 0.55 3.32 13.53 −10.21 
10 1.31 0.55 77.25 68.49 8.75 
11 0.55 −0.20 39.42 45.63 −6.21 
12 0.55 1.31 69.31 64.55 4.75 
13 0.55 0.55 3.64 0.43 21.33 −20.90 
14 0.55 0.55 10.36 69.74 50.28 19.45 
15 0.55 0.55 74.81 72.59 2.21 
16 0.55 0.55 77.49 72.59 4.89 
17 0.55 0.55 71.3 72.59 −1.29 
18 0.55 0.55 70.99 72.59 −1.60 
19 0.55 0.55 70.5 72.59 −2.09 
20 0.55 0.55 70.71 72.59 −1.88 
    Percentage decolourisation
 
 
Trial Lactose (X1Yeast extract (X2pH (X3Experimental Predicted Residual 
0.1 0.1 22.09 6.12 15.96 
0.1 56.68 50.03 6.64 
0.1 31.55 21.54 10.00 
60.2 56.82 3.37 
0.1 0.1 27.76 30.10 −2.34 
0.1 51.21 60.18 −8.97 
0.1 40.2 45.81 −5.61 
52.33 67.26 −14.93 
−0.20 0.55 3.32 13.53 −10.21 
10 1.31 0.55 77.25 68.49 8.75 
11 0.55 −0.20 39.42 45.63 −6.21 
12 0.55 1.31 69.31 64.55 4.75 
13 0.55 0.55 3.64 0.43 21.33 −20.90 
14 0.55 0.55 10.36 69.74 50.28 19.45 
15 0.55 0.55 74.81 72.59 2.21 
16 0.55 0.55 77.49 72.59 4.89 
17 0.55 0.55 71.3 72.59 −1.29 
18 0.55 0.55 70.99 72.59 −1.60 
19 0.55 0.55 70.5 72.59 −2.09 
20 0.55 0.55 70.71 72.59 −1.88 
Table 9

Estimated regression coefficients for CCD

S. no. Variable Coef. SE coef. 
Constant 72.5917 5.548 13.085 0.000* 
Lactose 16.3401 3.681 4.439 0.001* 
Yeast extract 5.6242 3.681 1.528 0.158 
pH 8.6070 3.681 2.338 0.041* 
Lactose*Lactose −11.1646 3.583 −3.116 0.011* 
Yeast extract*Yeast extract −6.1866 3.583 −1.727 0.115 
pH*pH −13.0031 3.583 −3.629 0.005* 
Lactose*Yeast extract −2.1575 4.809 −0.449 0.663 
Lactose*pH −3.4575 4.809 −0.719 0.489 
10 Yeast extract*pH 0.0725 4.809 0.015 0.988 
S. no. Variable Coef. SE coef. 
Constant 72.5917 5.548 13.085 0.000* 
Lactose 16.3401 3.681 4.439 0.001* 
Yeast extract 5.6242 3.681 1.528 0.158 
pH 8.6070 3.681 2.338 0.041* 
Lactose*Lactose −11.1646 3.583 −3.116 0.011* 
Yeast extract*Yeast extract −6.1866 3.583 −1.727 0.115 
pH*pH −13.0031 3.583 −3.629 0.005* 
Lactose*Yeast extract −2.1575 4.809 −0.449 0.663 
Lactose*pH −3.4575 4.809 −0.719 0.489 
10 Yeast extract*pH 0.0725 4.809 0.015 0.988 

*Significant.

R2 = 83.43% R2(adj) = 68.51%.

ANOVA was used to evaluate the implication and competence of the fittest second-order polynomial model on maximum dye decolourisation. Linear (P = 0.003) and quadratic (P = 0.007) effects of the factors had a greater influence on the dye decolourisation process. Nevertheless, the interaction terms (P = 0.867) were not statistically significant. This indicates that all the factors had a specific effect on decolourisation. Counter plots of a range of responses with the desired combination of significant factors in the experimental range are illustrated in Figure 3. These figures signify the relationship and interaction effect of any two factors with varying concentrations tested on dye decolourisation, while the third factor was maintained in the mid-value.
Figure 3

Response contour plots between significant factors on decolourisation (%): (a) effect of yeast extract and lactose concentration on dye decolourisation (%); (b) effect of pH and lactose concentration on dye decolourisation (%); (c) effect of pH and yeast extract concentration on dye decolourisation (%).

Figure 3

Response contour plots between significant factors on decolourisation (%): (a) effect of yeast extract and lactose concentration on dye decolourisation (%); (b) effect of pH and lactose concentration on dye decolourisation (%); (c) effect of pH and yeast extract concentration on dye decolourisation (%).

In the validation study, a response optimiser was applied to establish optimal concentrations of important factors: lactose, yeast extract and pH were found to be 0.85% (w/v), 0.71% (w/v), and 7.5%, respectively. Under these predicted optimal conditions, about 81.61% of decolourisation was attained in accordance with a predicted value of 79.99%. Based on the results observed by Saratale et al. (2009), added nutrients are substrates needed for the functioning of the azo reductase enzyme system. Rajee & Patterson (2011) observed that optimal pH 6 is required for decolourisation of azo dye by an autochthonous bacterium: Micrococcus sp. DBS; pH contributes to the transportation of dye molecules across the cell membrane. Similar to the present study, Du et al. (2012) reported that under optimised conditions, decolourisation of malachite green by Micrococcus sp. strain BD15 was achieved with an efficiency of 96.9%, which was within the predicted level (92.04%). M. glutamicus NCIM 2168 completely decolourised an initial concentration of 50 mg/L sulphonated diazo dye Reactive Green 19A within 42 h at 37 °C and pH 8 (Saratale et al. 2009). Under optimal conditions, decolourisation of textile wastewater increased significantly; this may be due to the influence of variations in abiotic factors that are involved in the decolourisation process (Chaudhari et al. 2013).

CONCLUSION

The present study clearly emphasises that the indigenous bacterial strain M. endophyticus ES37 has the ability for bioremediation of azo dye. The interactive effect of significant parameters and their optimal conditions were determined by statistical applications. The favourable conditions of environmental parameters most suitable for dye decolourisation resembled normal conditions. However, the toxicity of the degraded substances must be analysed prior to industrial applications.

ACKNOWLEDGEMENTS

The authors sincerely thank the Vice Chancellor and Registrar of Periyar University, Salem for the financial support of University Research Fellowship (URF) to accomplish this research work. The authors are also thankful to the Head, Department of Microbiology, Periyar University, for providing the facilities.

REFERENCES

REFERENCES
Brown
M. A.
Devito
S. C.
1993
Predicting azo dye toxicity
.
Crit. Rev. Env. Sci. Tec.
23
,
249
324
.
Chaudhari
A. U.
Savita
R. T.
Vijaykumar
L. M.
Kisan
M. K.
2013
Simultaneous decolourization of reactive Orange M2R dye and reduction of chromate by Lysinibacillus sp
.
KMK-A. J. Hazard. Mater.
262
,
580
588
.
Chen
K. C.
Wu
J. Y.
Liou
D. J.
Hwang
S. C. J.
2003
Decolourization of the textile azo dyes by newly isolated bacterial strains
.
J. Biotechnol.
101
,
57
68
.
Cooper
P.
1995
Colour in Dye House Effluent
,
1st edn
.
Society of Dyers and Colourists
,
Oxford
.
Du
L. N.
Zhao
M.
Li
G.
Zhao
X. P.
Zhao
Y. H.
2012
Highly efficient decolourization of malachite green by a novel Micrococcus sp. strain BD15
.
Environ. Sci. Pollut. Res.
19
,
2898
2907
.
Elisangela
F.
Zille
A.
Dias
G. F.
Ragagnin
M. C.
Durrant
L. R.
Cavaco
O. A.
2009
Biodegradation of textile azo dyes by a facultative Staphylococcus arlettae strain VN-11 using a sequential microaerophilic/aerobic process
.
Int. Biodeterior. Biodegrad.
63
,
280
288
.
Jadhav
U. U.
Dawkar
V. V.
Ghodake
G. S.
Govindwar
S. P.
2008
Biodegradation of direct red 5B, a textile dye by newly isolated Comamonas sp. UVS
.
J. Hazard. Mater.
158
,
507
516
.
Karthikeyan
K.
Nanthakumar
K.
Shanthi
K.
Lakshmanaperumalsamy
P.
2010
Response surface methodology for optimization of culture conditions for dye decolourization by a fungus, Aspergillus niger HM11 isolated from dye affected soil
.
Iran J. Microbiol.
2
,
213
222
.
Lima
R. O. A.
Bazo
A. P.
Salvadori
D. M. F.
Rech
C. M.
Oliveira
D. P.
Umbuzeiro
G. A.
2007
Mutagenic and carcinogenic potential of a textile azo dye processing plant effluent that impacts a drinking water source
.
Mutat. Res.
626
,
53
60
.
Molina-Guijarro
J. M.
Perez
J.
Munoz-Dorado
J.
Guillen
F.
Moya
R.
Hernandez
M.
Arias
M. A.
2009
Detoxification of azo dyes by a novel pH-versatile, salt-resistant laccase from Streptomyces ipomoea
.
Int. Microbiol.
12
,
13
21
.
Montgomery
D. C.
2001
Design and Analysis of Experiments
.
3rd edn
.
John Wiley
,
Hoboken, NJ
.
Nigam
P.
Banat
I. M.
Marchant
R.
1996
Decolourization of effluent from the textile industry by a microbial consortium
.
Biotechnol. Lett.
18
,
117
120
.
Paul
G. C.
Kent
C. A.
Thomas
C. R.
1992
Quantitative characterization of vacuolization in Penicillium chrysogenum using automatic image analysis
.
Trans. Chem. E.
70
,
13
20
.
Plackett
R. l.
Burman
J. P.
1946
The design of optimum multifactorial experiments
.
Biometrika
33
,
305
325
.
Pourbabaee
A. A.
Malekzadeh
F.
Sarbolouki
M. N.
Najafi
N.
2005
Aerobic decolourization and detoxification of a disperse dye in textile effluent by a new isolate of Bacillus sp
.
Biotechnol. Bioeng.
93
,
631
635
.
Rauf
M. A.
Salman Ashraf
S.
2009
Radiation induced degradation of dyes-an overview
.
J. Hazard. Mater.
166
,
6
16
.
Saratale
R. G.
Saratale
G. D.
Chang
J. S.
Govindwar
S. P.
2009
Ecofriendly degradation of sulfonated diazo dye CI reactive green 19A using M. glutamicus NCIM-2168
.
Bioresour. Technol.
100
,
3894
3905
.
Saratale
R. G.
Saratale
G. D.
Chang
J. S.
Govindwar
S. P.
2010
Decolourization and biodegradation of reactive dyes and dye wastewater by a developed bacterial consortium
.
Biodegrad.
21
,
999
1015
.
Tony
B. D.
Goyal
D.
Khanna
S.
2009
Decolourization of textile azo dyes by aerobic bacterial consortium
.
Int. Biodeterior. Biodegrad.
63
,
462
469
.
Vander Zee
F. P.
Bisschops
I. A. E.
Blanchard
V. G.
Bouwman
R. H. M.
Lettinga
G.
Field
J. A.
2003
The contribution of biotic and abiotic processes during azo dye reduction in anaerobic sludge
.
Water Res.
37
,
3098
3109
.
Verma
P.
Madamwar
D.
2003
Decolourization of synthetic dyes by a newly isolated strain of Serratia marcescens
.
World J. Microbiol. Biotechnol.
19
,
615
618
.