Available online at www.sciencedirect.

com

Bioresource Technology 99 (2008) 75657572

Factorial design for the optimization of enzymatic detection of cadmium in aqueous solution using immobilized urease from vegetable waste
Om Prakash a, Mahe Talat a,*, S.H. Hasan b, Rajesh K. Pandey a
b a Department of Biochemistry, Faculty of Science, Banaras Hindu University, Varanasi 221005, India Department of Applied Chemistry, Institute of Technology, Banaras Hindu University, Varanasi 221005, India

Received 13 November 2007; received in revised form 8 February 2008; accepted 10 February 2008 Available online 18 April 2008

Abstract Free as well as alginate immobilized urease was utilized for detection and quantitation of cadmium (Cd2+) in aqueous samples. Urease from the seeds of pumpkin (Cucumis melo), being a vegetable waste, was extracted and puried to apparent homogeneity (Sp. Activity 353 U/mg protein; A280/A260 = 1.12) by heat treatment at 48 0.1 C and gel ltration through Sephadex G-200. The homogeneous enzyme preparation was immobilized in 3.5% alginate leading to 86% immobilization and no leaching of the enzyme was found over a period of 15 days at 4 C. Urease catalyzed urea hydrolysis by both soluble and immobilized enzyme revealed a clear dependence on the concentration of Cd2+. The inhibition caused by Cd2+ was non-competitive (Ki = 1.41 105 M). The time dependent inhibition both in the presence and in absence of Cd2+ ion revealed a biphasic inhibition in the activity. A Response Surface Methodology (RSM) for the parametric optimization of this process was performed using two-level-two-full factorial (22), central composite design (CCD). The regression coecient, regression equation and analysis of variance (ANOVA) was obtained using MINITAB 15 software. The predicted values thus obtained were closed to the experimental value indicating suitability of the model. In addition to this 3D response surface plot and isoresponse contour plot were helpful to predict the results by performing only limited set of experiments. 2008 Elsevier Ltd. All rights reserved.
Keywords: Urease; Cd2+; Immobilization; Response surface methodology (RSM); Full factorial design

1. Introduction Extensive research has been done to investigate the possibilities oered by enzymes in biotechnological and envin ronmental applications (Husain and Jan, 1999; Dura and Esposito, 2000). Enzymatic reactions have proved to be very promising tools to identify major pollutants such as heavy metals (Brack et al., 2000; Jung et al., 1995). However, use of free enzymes show some major drawbacks such as thermal instability, susceptibility to attack by proteases, activity inhibition, high sensitivity to several denaturing agents, the impossibility of separating and reusing free catCorresponding author. Tel.: +91 0542 2307323; fax: +91 0542 2368174. E-mail address: mahetalat04@gmail.com (M. Talat). 0960-8524/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.02.008
*

alyst at the end of the reaction, etc. The use of immobilized enzymes has proved to be more advantageous than the free n et al., 2002). However, high cost and limenzymes (Dura ited availability of immobilized enzyme preparations are two important limitations in the wider applications of enzymes for routine detection of heavy metal ions (Tischer and Kasche, 1999; Gupta and Mattiasson, 1992). Among the techniques used for immobilization, entrapment in natural biopolymers is favored for various reasons; e.g., nontoxicity of the matrix, possibility of the variation in the bead size and high percentage of immobilization and cost eectiveness (Kierstan and Bucke, 2000; Smidsrd and Skjaok-Br, 1990). Considering these, calcium alginate mediated entrapment has attracted much attention. Apart from the matrix, enzymes availability and behaviour towards the pollutant are also important issues of consideration.

7566

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572

Enzymes are often specic to inhibitor and in many cases the inhibitory eect of investigated pollutant is related to its biological toxicity (Krawczyk et al., 2000). Owing to its pronounced sensitivity, urease (urea amidohydrolase, EC 3.5.1.5) has been considered as a model enzyme for application as a probe for heavy metal ions. Interestingly, dierent metals exhibit quite dierent behaviour in their ability to act as urease inhibitor (Kuswandi, 2003; Prakash and Vishwakarma, 2001; Krajewska et al., 2004). Most of the studies have utilized urease obtained from jack bean which being expensive. Therefore, there still exits a need to have urease from a non-conventional, un-utilized and cheaper source for versatile applications. Employing very simple steps, urease was puried to apparent homogeneity from an agricultural waste, i.e., the dehusked seeds of pumpkin (Cucumis melo) and entrapped into alginate beads. In the present communication, we describe the interaction of thus puried soluble as well as calcium alginate immobilized urease with cadmium ion. In addition to this, the eect of the variables i.e. Cd2+ ion concentration and time of interaction, which are aecting the activity of immobilized urease enzyme, were evaluated by using Response Surface Methodology (RSM) a statistical and graphical technique. Dierent factorial designs are available in RSM techniques (Khuri and Cornell, 1987; Mason et al., 1989). Here two-leveltwo-factor full factorial central composite design (CCD) model was used (Heck et al., 2005). The predicted result by the response surface central composite design (CCD) model was then compared with the experimental results. 1.1. Design of experiments

in a kitchen blender for 2 min, ltered through muslin cloth and centrifuged for 15 min at 04 C at 21,500g. The supernatant was ltered through a thick layer of prewashed glass wool to remove excess fat. 2.3. Purication of urease The following steps were taken to purify the crude enzyme preparation. 2.3.1. Heat treatment Crude enzyme preparation (3.0 ml) was heated at 48 0.1 C in a water bath for 10 min and was immediately chilled in crushed ice. This was centrifuged at 21,000g for 15 min at 04 C. The supernatant was collected. 2.3.2. Gel ltration Heat treated enzyme (1.5 ml, 2.42.6 mg protein) was loaded on a Sephadex G-200 column (1.5 40 cm, preequilibrated with degassed 25 mM Trisacetate buer, pH 7.5). The elution was carried out at 46 C at a ow rate of 20 2 ml h1 with degassed extraction buer. Various 2.0 ml fractions containing urease activity were pooled and concentrated against solid sucrose. The enzyme preparation was kept frozen (at 20 C) in small aliquots until further use. The enzyme preparation (sp act 353 12 U mg1 protein) showing a single enzyme and protein band on native 7.5% PAGE (at pH 8.3), was employed for the study. 2.4. Calcium alginate beads preparation

A two-level-two-factor (22) full factorial experiment was designed to observe the eect of the parameters inuencing activity of immobilized enzyme. The two factors considered are (i) concentration of Cd2+ (low 0.05 mM and high 0.1 mM) and (ii) interaction time of Cd2+ with enzyme (550 min). 2. Methods 2.1. Materials Pumpkin seeds were procured from the local market and dehusked just before soaking. Tris was obtained from Boehringer Mannheim Gmbh, Germany. Bovine serum albumin was obtained from Sigma Chemical Co., USA. Sephadex G-200 was from Pharmacia Fine Chemicals, Uppsala, Sweden. Urea (enzyme grade), cadmium acetate, Nesslers and Folin-Ciocalteau reagents were from Qualigens Fine Chemicals, Mumbai. All other reagents were analytical grade chemicals either from BDH or E. Merck, India. 2.2. Isolation of urease Dehusked pumpkin seeds (6 g) were soaked in 25 mM Trisacetate buer (pH 7.5) for 8 h at 46 C, homogenized

Solution (3.5%) (w/v) of sodium alginate was prepared in 25 mM Trisacetate buer (pH 7.5) by stirring for 2 h at room temperature and was stored at 4 C. Suitably diluted urease solution (0.7 mg protein ml1) was mixed in chilled sodium alginate solution and dropped into 100 ml of chilled and continuously stirring 400 mM CaCl2 solution with the help of a micropipette. The beads formed were allowed to stir for 90 min for complete calcium alginate formation. Beads were collected, washed thoroughly with 25 mM Trisacetate buer (pH 7.5) to remove excess Ca2+ and stored in the same buer at 4 C. During the course of immobilization, the left over CaCl2 solution was analyzed for protein and enzyme activity to ascertain leaching, if any. 2.5. Urease activity assay Enzyme activity was assayed in 50 mM Trisacetate buffer (pH 8.0). An aliquot (0.8 ml) of buer and 1.0 ml of 250 mM urea in the same buer were brought to 30 C. The reaction was started by adding 0.2 ml of suitably diluted enzyme. After 10 min, 1.0 ml of 10% trichloroacetic acid was added to stop the reaction. The total reaction mixture was transferred to a measuring ask (50 ml) and the

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572

7567

volume was made to 50 ml with distilled water after adding 1.0 ml of Nesslers reagent. The amount of ammonia liberated was measured at 405 nm in a Spectronic 21UVD spectrophotometer. For assay of immobilized enzyme, the beads were incubated at 30 C for 10 min in standard assay medium. Following incubation, an aliquot of 1.0 ml was withdrawn from the reaction mixture and assayed as described. The beads were recovered from the reaction mixture and washed thoroughly with the buer and stored at 4 C. The percentage of immobilization is dened as the (total activity in immobilized beads/total activity of the soluble enzyme loaded) 100. A unit of enzyme activity was dened as the amount of enzyme required to liberate 1 lM of ammonia in 1 min under the test conditions dened above (30 C, 50 mM Trisacetate buer, pH 8.0, 250 mM urea). Protein was estimated by Folin-Ciocalteau reagent calibrated with crystalline bovine serum albumin (Lowry et al., 1951). 2.6. Eect of Cd2+ on the activity of the soluble and immobilized urease Stock solution of cadmium acetate was made in 50 mM Trisacetate buer (pH 8.0) and diluted with the same buffer as required. The activity of suitably diluted enzyme was determined in the presence of varying concentration of Cd2+ added in the standard assay mixture. For the direct eect, the enzyme alone was incubated with the desired concentration of Cd2+ for 10 min at 30 C and the treated enzyme was assayed for the activity. The rate of inactivation was studied at two concentrations of Cd2+. Timedependent interaction was studied by incubating the enzyme with 0.05 mM and 0.08 mM of Cd2+ at 30 C. Aliquots were withdrawn at specic time intervals and assayed for the activity. The results reported are mean of 56 replicate experiments carried out with fresh batch of puried enzyme. 2.7. Response surface methodology design of experiment Response surface methodology (RSM) is an empirical statistical technique employed for multiple regression analysis by using quantitative data. It solves multivariable data which is obtained from properly designed experiments to solve multivariable equation simultaneously (Tan et al. (2008)). The graphical representation of their functions is called Response Surface, which was used to describe the individual and cumulative eect of the test variables and their subsequent eect on the response. Easy way to estimate Response Surface, Factorial designs is the most useful scheme for the optimization of variables with a limited number of experiments. A variety of factorial designs are available to accomplish this task (Azaroghar and Dalai, 2005). The most successful and best among them is the central composite design (CCD) which is accomplished by

adding two experimental points along each coordinate axis at opposite sides of the origin and at a distance equal to the semi diagonal of the hyper cube of the factorial design and new extreme values (low and high) for each factor added in this design (Kumar et al., 2008). If the factorial is a full factorial then, a 2k 1=4 1

Since in this study two factors such as concentration of cadmium and interaction time were considered thus k = 2. So a 1:414 Furthermore, the total number of experimental point (N) in a CCD can be calculated by following equation: N 2 k 2k x 0 2

where N is the number of experimental run, k is the number of variables and xo is the number of central points. Thus for this design total number of experimental runs will be 13 (k = 2; x0 = 5). Data is thus obtained from the central composite design was subjected to a second order multiple regression analysis to explain the behaviour of the system using the least square regression methodology. X X X Y b0 bi X i bii X 2 bij X i X j  3 i where, Y is the predicted response; Xi, X 2 i , Xj are independent variables in coded values; b0 is the constant; bi is linear eect; bii is squared eect and bij is interaction eect. The analysis of results was performed with statistical and graphical analysis software (Minitab Release 15, 2006). This software was used for regression analysis of the data obtained and to estimate the coecient of regression equation. ANOVA (analysis of variance) which is statistical testing of the model in the form of linear term, squared term and interaction term was also utilized to test the signicance of each term in the equation and goodness of t of the regression model obtained (Huiping et al., 2007). This response surface model was also used to predict the result by isoresponse contour plots and three dimensional surface plots. Contour plot is the projection of the response surface as a two dimensional plane where as 3D surface plots is the projection of the response surface in a three dimensional plane (Box and Hunter, 1957). 3. Results and discussions 3.1. Extraction of enzyme The procedure for extraction of urease from dehusked seeds of pumpkin was optimized by varying the pH and molarity of the extraction buer. The seeds were soaked overnight at 46 C in 25 mM Trisacetate buer of pH range varied from 6.0 to 8.5. The soaked seeds along with the buer were crushed in a kitchen blender giving high strokes for 2 min. The supernatant obtained after centrifugation revealed an increase in the activity with increasing

7568

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572

pH of the extraction buer and attained its maximum at pH 8.5.The specic activity, on the other hand, increased with increase in pH up to 7.5. Further increase in pH notably reduced the specic activity. Therefore, the optimum pH for extraction of urease in Trisacetate buer appeared to be 7.5. 3.2. Purication of enzyme 3.2.1. Heat fractionation Crude extract was heated at 48 0.1 oC for 10 min, chilled immediately and centrifuged. This step puried the enzyme by about 1.4-fold with 76% recovery. 3.2.2. Gel ltration Heat fractionated enzyme (158 20 U ml1, 1.61.8 mg protein ml1) was further puried by gel ltration through a Sephadex G-200 column. The enzyme was eluted in a single peak. Various 2 ml fractions containing urease activity were pooled and concentrated against solid sucrose. This step further puried urease by about 5.2-fold with 48% recovery (Table 1). Native 7.5% PAGE revealed single protein band under coomassie brilliant blue staining, silver staining as well as under urease-specic staining (Blattler et al., 1967). The puried enzyme showed a typical protein spectrum with maximum absorption at 278 nm and A280/A260 of 1.12 suggesting the preparation to be free of nucleotides. 3.3. Inhibitory eect of cadmium ions on urease Eect of Cd2+ on the activity of pumpkin urease (both soluble and alginate-immobilized) was studied in the concentration range of 0.010.1 mM. The desired concentration of it was added into the standard assay mixture and the activity was assayed. The results revealed a concentration dependent inhibition in the activity of both soluble and immobilized enzyme. The inhibition in the activity was more pronounced in the case of soluble enzyme than the alginate entrapped urease at all the concentrations of Cd2+ studied. Thus, 0.05 mM of it inhibited soluble enzyme by 46% and immobilized enzyme by 24% only. The observed dierence in the degree of inhibition might be due to the protection provided to the enzyme by alginate entrapment. Also, the hindered accessibility of the inhibitor to the active site of the enzyme in the immobilized state can not be ruled out.
Table 1 Purication of urease from dehusked seeds of pumpkin Steps Enzyme Specic activity Fold % (U/ml) (U/mg protein) purication Recovery 67.9 93.0 353.0 1.4 5.2 100 76 48

The nature of the inhibition of urease was studied by employing two concentrations (0.05 and 0.08 mM) of cadmium in such a manner that the inhibition was measurable. For the study, urea concentration in the assay mixture was varied from 2 to 125 mM. The results, when expressed by LineweaverBurk double reciprocal plot of substrate concentration vs. velocity (absorbance at 405 nm) revealed a non-competitive inhibition. The Ki was found to be 1.41 105 M for the urease. The interaction of the metal ion with the enzyme protein was so strong that the inhibition could not be reversed by dialysis (50 mM Trisacetate buer, pH 8.0, 46 C, 24 h). Similar irreversible interaction has also been shown for jack bean and pumpkin urease towards Hg2+ ion (Prakash et al., 2008). Time-dependent hydrolysis of urea by pumpkin urease was studied by incubating enzyme in the presence (0.05 and 0.08 mM) and absence of Cd2+ in the standard assay system. The urea hydrolysis in absence of Cd2+ progressed with time. In the presence of Cd2+ however, reaction progressed initially and then attained a steady state. The immobilized enzyme, on the other hand, exhibited improved stability towards the inhibitor. Thus, at a concentration of 0.05 mM, approximately 52.5% activity was retained even after 50 min and about 41.4% at 0.08 mM. Similarly, immobilization of Citrullus vulgaris urease on cyanuric chloride-DEAE-cellulose-ether has also been shown to exhibit improved stability towards inhibitory eect of heavy metal ions (Fahmy et al., 1998). Krajewska reported that the stability of jack bean urease against metal ion inactivation was considerably improved upon immobilization on glutaraldehyde-pretreated chitosan membrane (Krajewska, 1991). A signicant improved stability of the alginate immobilized watermelon urease towards water miscible organic solvents has recently been reported (Prakash and Upadhyay, 2006). 3.4. Response surface factorial design for the optimization of the process The variables that predominantly aecting the enzymatic detection of Cd2+ by immobilized urease were (i) concentration of Cd2+ solution and (ii) interaction time, and each parameter was having one lower value and one higher value. Thus for the two-level-two-full factorial central composite design was selected for which 13 experimental values (Table 2a) were required (Das and Giri, 1986). Experiments were performed according to the experimental plan (Table 2b) where the concentration of metal ion was changed from 0.05 to 0.1 mM and the interaction time from 5-50 min. The responses thus obtained for each combination of the variables are given in Table 2c. Signicant changes in enzyme activity were observed for all the combinations, implying that both these variables signicantly aecting the enzyme activity. 3.4.1. Interpretation of regression analysis The regression results obtained from central composite design (CCD) model are given in Table 3a where T values

Crude extract 173 Heat treatment at 158 48 0.1 C Gel ltration 149 (Sephadex G-200)

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572 Table 2a Number of experimental runs required and their details for two-level factorial: full factorial Central composite design Factors 2 Replicates 1 Base runs 13 Base blocks 1 Total blocks 1 Total runs 13 a 1.41421

7569

Two-level factorial: full factorial Cube Center points Axial points in cube points 4 5 4

Center points in axial 0

Table 2b Experimental data showing the % residual activity of urease at dierent Cd(II) conc. (mM) and interaction time (min) Cd(II) conc. (mM) 0.10 0.08 0.05 0.05 0.05 0.08 0.08 0.08 0.08 0.05 0.08 0.05 0.08 Time (min) 15 40 22 18 25 25 22 40 18 50 50 30 5 % Residual activity 41.0 57.0 74.0 67.0 76.8 48.8 46.0 57.5 40.0 86.9 61.8 78.0 4.0

Table 2c Predicted % residual activity of urease after processing experimental data by MINITAB software Std order 8 10 4 9 11 13 7 12 3 6 1 2 5 Run order 1 2 3 4 5 6 7 8 9 10 11 12 13 Pt type 1 0 1 0 0 0 1 0 1 1 1 1 1 Blocks Cd(II) conc. (mM) 0.10 0.08 0.05 0.05 0.05 0.08 0.08 0.08 0.08 0.05 0.08 0.05 0.08 Time (min) 15 40 22 18 25 25 22 40 18 50 50 30 5 % Residual activity Experimental 41.0 57.0 74.0 67.0 76.8 48.8 46.0 57.5 40.0 86.9 61.8 78.0 4.0 Predicted 40.9999 60.1892 72.4829 66.0262 76.5748 48.0334 43.6720 60.1892 36.8559 85.6525 59.3527 81.9649 6.8041

1 1 1 1 1 1 1 1 1 1 1 1 1

The value of constant was found to be 52.192 which also do not depend on any factor and interaction of the factors. The eect of the linear factor i.e. Cd2+ ion concentration and interaction time was found to be highly signicant (P = 0.003 and 0.000, respectively) on the residual activity of enzyme. All the square terms such as (concentration)2 and (time)2 were also found to be signicant (P = 0.000). Since the squared terms were signicant that means there was a curved line relationship between residual activity and square factors. Whereas the interaction term concentration time (P = 0.610) was not found to be signicant. A positive sign of the coecient represents a synergistic eect, while a negative sign indicates an antagonistic eect. The linear variable concentration and square term time had a negative relationship with the residual activity. So with the increase of these factors there will be a decrease in the residual activity. Whereas the linear term (time), square term (concentration) and interaction term (concentration and time) had a positive eect on the residual activity which indicates that with an increase of these factors there will be an increase in the residual activity. Furthermore, high values of R2 (98.81%) and R2(adjusted) (97.96%) indicates a high dependence and correlation between the observed and the predicted values of response. This also indicates that 98.81% of result of the total variation can be explained by this model. The summary of ANOVA is shown in Table 3b. The ANOVA demonstrates that the regression model was highly signicant, as is evident from the calculated Fishers F value (116.08) and a probability (P) value of 0.000. The large value of F indicates that most of the variation in the response can be explained by the regression equation. The associated P value is used to estimate whether F is large enough to indicate statistical signicance. If P value is lower than 0.05, then it indicates that the model is statistically signicant (Zulkali et al., 2006). It was observed from this table that the coecients for the linear (P = 0.000) and the square terms (P = 0.000) were highly signicant whereas interaction (P = 0.610) eect was not signicant and thus conrm the applicability of the predicted model. The ANOVA table also shows
Table 3a Estimated regression coecients for % residual activity of urease versus Cd(II) conc. (mM), time (min) in coded units Term Constant Cd(II) conc. (mM) Time (min) Cd(II) conc. (mM) Cd(II) conc. (mM) Time (min) time (min) Cd(II) conc. (mM) time (min) R2 = 98.81% Coef. 52.192 8.781 25.939 18.521 SE coef. 1.800 1.958 1.655 2.843 T 28.991 4.486 15.675 6.514 P 0.000 0.003 0.000 0.000

and P values along with the constant and coecients are mentioned. The T value is used to determine the signicance of the regression coecients of the parameters and the P value is dened as the smallest level of signicance leading to rejection of null hypothesis. In general, the larger the magnitude of T and smaller the value of P, the more signicant is the corresponding coecient term (Ravikumar et al., 2007).

18.096 1.684

2.823 3.152

6.410 0.000 0.534 0.610

R2(pred.) = 78.07% R2(adj.) = 97.96%

7570

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572

Table 3b Analysis of variance for % residual activity of urease versus Cd(II) conc. (mM), time (min) in coded units Source Regression Linear Square Interaction Residual error Lack-of-t Pure error Total DF 5 2 2 1 7 6 1 12 Seq. SS 5613.50 4500.84 1109.90 2.76 67.70 67.58 0.12 5681.21 Adj. SS 5613.50 3363.60 1112.58 2.76 67.70 67.58 0.12 Adj. MS 1122.70 1681.80 556.29 2.76 9.67 11.26 0.12 F 116.08 173.88 57.52 0.29 90.11 P 0.000 0.000 0.000 0.610 0.080

y 192:660 4878:54 concentration of Cd2 2:89428 interaction time 29633:1 concentration of Cd2 0:0357451 interaction time 2:99421 concentration of Cd2 interaction time 4
2 2

The predicted values of residual activity of enzyme obtained using Eq. (4) are closed to the experimental values proving that the model is fully applicable. 3.4.2. Interpretation of residual graph The normality of the data can be checked by plotting the normal probability plot (NPP) of the residuals. The normal probability plot is a graphical technique for assessing whether or not a data set is approximately normally distributed (Box and Hunter, 1957). The residual is the dierence between the observed and the predicted value (or the tted value) from the regression. If the points on the plot fall fairly close to the straight line then the data are normally distributed. Fig. 1(i) shows normal probability plot of residual values. It could be seen that the experimental points were reasonably aligned suggesting normal distribution. The results can be shown in Fig. 1(ii) with the help of a histogram. A histogram of the residuals shows the distribution of the residuals for all observations. The gure shows an almost symmetrical histogram (bell shaped, i.e. the
(ii)

Table 3c Estimated regression coecients for % residual activity of urease versus Cd(II) conc. (mM), time (min) using data in uncoded units Term Constant Cd(II) conc. (mM) Time (min) Cd(II) conc. (mM) Cd(II) conc. (mM) Time (min) time (min) Cd(II) conc. (mM) time (min) Coef. 192.660 4878.54 2.89428 29633.1 0.0357451 2.99421

no residual error, which means the variation in the response data can be very well explained by the model. Multiregression analysis was performed to obtain a quadratic response surface model (Table 3c) and equation thus obtained was
(i)
99 90

4 2

Percent

Residual
-5.0 -2.5 0.0 2.5 5.0

50

0 -2 -4 0 20 40 60 80

10 1

Residual

Fitted Value

(iii)
3

(iv)
4 2 2

Frequency

Residual
-4 -3 -2 -1 0 1 2 3

0 -2 -4 1 2 3 4 5 6 7 8 9 10 11 12 13

Residual

Observation Order

Fig. 1. (i) Normal probability plot of the dierence between the observed and the predicted value. (ii) Histogram of the dierence between the observed and the predicted value. (iii) Dierence between the observed and the predicted value versus the tted value when the variance is constant. (iv) Dierence between the observed and the predicted value versus the order of the data due to the cumulative eect of interaction time and Cd2+ concentration.

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572

7571

errors are normally distributed with mean zero) (Minitab Release 15, 2006). Fig. 1(iii) plots the residuals versus the tted values (predicted response). The residuals are scattered randomly about zero i.e. the errors have a constant variance. This last plot of Fig. 1(iv) is the residual value and the order of the corresponding observations. The plot is useful when the order of the observations may inuence the results which can occurs when data are collected in a line sequence. This plot can be helpful to a designed experiment in which the runs are not randomized. For residual activity data, the residuals appear to be randomly scattered about zero. No evidence exists that the regression terms are correlated with one another. 3.4.3. Interpretation of response 3D surface plots and contour plots The 3D response surface, which is a three dimensional graphic representation was used to determine the individual and cumulative eect of the variable and the mutual interaction between the variable and the dependent variable. The response surface analyzes the geometric nature of the surface, the maxima and minima of the response and the signicance of the coecients of the canonical equation. The polynomial response surface model obtained may be maximized or minimized to obtain the optimum points. Whereas a contour plot is a graphical technique for representing a three dimensional surface by plotting constant z-slices called contours, on a two dimensional format. That is, given a value for z, lines are drawn for connecting the (x, y) coordinates where that z value occurs (Ravikumar et al., 2007). To investigate the individual and interactive eect of these two factors on the residual activity of enzyme, three dimensional and contour plots were drawn with the help of Minitab Release-15 software and the inferences thus obtained are discussed below. 3.4.3.1. Response 3D surface plots. The surface plot (Fig. 2(i)) where residual activity of enzyme was represented by varying simultaneously Cd2+ concentration from 0.05 to 0.1 mM and time from 5 to 50 min. From this response surface plot this is also clear that residual enzyme activity to 50% concentration of Cd2+ should be 0.05 mM and exposure time should be nearly 20 min. The surface plot also describing individual and cumulative eect of these two test variable and test their subsequent eect on the response. 3.4.3.2. Contour plot. The isoresponse contour plots between the variables such as Cd2+ concentration and interaction time is given in Fig. 2(ii). The lines of contour plots (centre of the circle) predicting the values of residual enzyme activity for dierent Cd2+ ions concentration at dierent interaction time. These values are more or less same to the experimental values.

(i)

%Residual Activity

75 50 25 0 0.050 0.075 Cd(II) Conc (mM) 0.100 0 15 30

m
60 80 40 60

45
in
20

(ii)
50

40

Time (mins)

30

20

10

0.05

0.06

0.07

0.08

0.09

Ti

(m

s)

0.10

Cd(II) Conc (mM)

Fig. 2. (i) 3D surface plot of the cumulative eect of concentration of Cd2+ (mM) and interaction time (min) on residual activity of urease. (ii) Contour plot of the cumulative eect of concentration of Cd2+ (mM) and interaction time (min) on residual activity of urease.

4. Conclusion The present study was aimed to work out an inexpensive and simple procedure for immobilization of urease obtained from a rather cheap and non-conventional source, which could be utilized for detection and quantitation of Cd2+ present in polluted wastewater/industrial euents. The study conducted and results reported above suggest the versatile application of urease obtained from an agricultural waste, i.e. the discarded seeds of pumpkin and revealed its suitability for the detection of Cd2+ in water, industrial euents, soil etc. Furthermore, with the help of experimental data a model was predicted by statistical and graphical technique response surface methodology (RSM) where two-level-two factor (22) was used. The

7572

O. Prakash et al. / Bioresource Technology 99 (2008) 75657572 Krajewska, B., 1991. Urease immobilized on chitosan membrane. Inactivation by heavy metal ions. J. Chem. Technol. Biotechnol. 52, 157 162. Krajewska, B., Zaborska, W., Chudy, M., 2004. Multi-step analysis of Hg2+ ion inhibition of jack bean urease. J. Inorg. Biochem. 98, 1160 1168. ska, M., Trojanowicz, M., 2000. Inhibitive Krawczyk, T.K., Moszczyn determination of mercury and other metal ions by potentiometric urea biosensor. Biosens. Bioelectron. 15, 681691. Kumar, A., Prasad, B., Mishra, I.M., 2008. Optimization of process parameters for acrylonitrile removal by a low-cost adsorbent using BoxBehnken design. J. Hazard. Mater. 150 (1), 174182. Kuswandi, B., 2003. Simple optical bre biosensor based on immobilised enzyme for monitoring of trace heavy metal ions. Anal. Bioanal. Chem. 376, 11041110. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folinphenol reagent. J. Biol. Chem. 193, 265 275. Mason, R.I., Funst, R.F., Hess, J.L., 1989. Statistical Design and Analysis of Experiments. Wiley, New York. Minitab Release 15, 2006. Statistical Software for Windows, Minitab Inc., USA. Prakash, O., Upadhyay, L.S.B., 2006. Immobilization imparts stability to watermelon urease to work in water miscible organic media. Biotechnol. Bioprocess Eng. 11, 140145. Prakash, O., Vishwakarma, D.K., 2001. Inhibition of urease from seeds of water melon (Citrullus vulgaris) by heavy metal ions. J. Plant Biochem. Biotechnol. 10, 147149. Prakash, O., Talat, M., Hasan, S.H., Pandey, R.K., 2008. Enzymatic detection of mercuric ions in ground-water from vegetable wastes by immobilizing pumpkin (Cucumis melo) urease in calcium alginate beads. Bioresource Technol. 99, 45244528. Ravikumar, K., Krishnan, S., Ramalingam, S., Balu, K., 2007. Optimization of process variables by the application of response surface methodology for dye removal using a novel adsorbent. Dyes Pigments 72, 6674. Smidsrd, O., Skjaok-Br, G., 1990. Alginate as immobilization matrix for cells. Trends Biotechnol. 8, 7178. Tan, I.A.W., Ahmad, A.L., Hameed, B.H., 2008. Optimization of preparation conditions for activated carbons from coconut husk using response surface methodology. Chem. Eng. J. 137, 462470. Tischer, W., Kasche, V., 1999. Immobilized enzymes: crystals or carriers? Trends Biotechnol. 17, 326335. Zulkali, M.M.D., Ahmad, A.L., Norulakmal, N.H., 2006. Oryza sativa L. husk as heavy metal adsorbent: optimization with lead as model solution. Bioresour. Technol. 97, 2125.

predicted value thus obtained using MINITAB software has been found close to the experimental value indicating the applicability of the model. References
Azaroghar, R., Dalai, A.K., 2005. Production of activated carbon from Luscar char: experimental and modeling studies. Micropor. Mesopor. Mater. 85, 219225. Blattler, D.P., Contaxis, C.C., Reithel, F.J., 1967. Dissociation of urease by glycol and glycerol. Nature 216, 274275. Box, G.E.P., Hunter, J.S., 1957. Multifactor experimental design for exploring response surfaces. Ann. Math. Stat. 28, 195241. Brack, W., Paschke, A., Segner, H., Wennrich, R., Schu u rmann, G., 2000. Urease inhibition: a tool for toxicity identication in sediment elutriates. Chemosphere 40, 829834. Das, M.N., Giri, N.C., 1986. Design and Analysis of Experiments. Wiley Eastern Ltd. n, N., Esposito, E., 2000. Potential applications of oxidative enzymes Dura and phenoloxidase-like compounds in wastewater and soil treatment: a review. Appl. Catal. B: Environ. 28, 8399. n, N., Rosa, M.A., DAnnibale, A., Gianfreda, L., 2002. ApplicaDura tions of laccases and tyrosinases (phenoloxidases) immobilized on dierent supports: a review. Enzyme Microbiol. Technol. 31, 907931. Fahmy, A.S., Bagos, V.B., Mohammed, T.M., 1998. Immobilization of Citrullus vulgaris urease on cyanuric chloride DEAE-cellulose ether: preparation and properties. Bioresour. Technol. 64, 121129. Gupta, M.N., Mattiasson, B., 1992. Unique applications of immobilized proteins in bioanalytical systems. Meth. Biochem. Anal. 36, 114. Heck, F.J., Floris, S.H., Herty, P.F., Ayub, M.A.Z., 2005. Optimization of cellulose-free xylanose activity produced by Bacillus Coagulans BL69 in solid-state cultivation. Process Biochem. 40, 107112. Huiping, L., Guoqun, Z., Shanting, N., Yiguo, L., 2007. Technologic parameter optimization of gas quenching process using response surface method. Comput. Mater. Sci. 38, 561570. Husain, Q., Jan, U., 1999. Detoxication of phenols and aromatic amines from polluted wastewater by using phenol oxidases. J. Sci. Ind. Res. 59, 286293. Jung, K., Bitton, G., Koopman, B., 1995. Assessment of urease inhibition assays for measuring toxicity of environmental samples. Water Res. 29 (8), 19291933. Khuri, A.I., Cornell, S.A., 1987. Response Surface Design and Analysis of Experiment. Marcel Dekker Inc., New York. Kierstan, M., Bucke, C., 2000. The immobilization of microbial cells, subcellular organelles and enzymes in calcium alginate gels. Biotechnol. Bioeng. 67, 726736.