Actinomycete strain, Streptomyces sp. strain NEAE-94, has been tested for its cholesterol oxidase activity with a plate-based method, the formation of the pink areas around the colonies indicated the presence of cholesterol oxidase activity (Fig. 1). The process of cholesterol-oxidation is the oxidation of cholesterol with the use of cholesterol oxidase to 4-cholesten-3-one and hydrogen peroxide. The hydrogen peroxide generated by cholesterol oxidation was then combined with 4-aminoantipyrine and phenol by peroxidase to produce quinoneimine coloration (Supplementary Figure 1). The promising strain was identified based on morphological, cultural, physiological and chemotaxonomic characteristics, in addition to 16S rRNA sequence.
Cultural characteristics of Streptomyces sp. strain NEAE-94
The isolate showed well growth on four different media, including oatmeal agar, inorganic salt-starch agar, tyrosine agar and yeast malt extract agar. Weak growth was observed on glycerol asparagines agar and peptone-yeast extract iron agar (Supplementary Table 1). Aerial mycelium color was whitish yellow on yeast extract-malt extract agar (Fig. 2A); yellow on oatmeal agar, tyrosine agar and inorganic salt-starch agar (Fig. 2B), while is faint yellow on peptone-yeast extract iron agar and glycerol asparagines agar. However, the substrate mycelium develop a yellow color on oatmeal agar, inorganic salt-starch agar, yeast extract -malt extract agar; brownish orange color on tyrosine agar media. Whereas, a faint orange color was developed on both glycerol asparagine agar and peptone-yeast extract iron agar media. A yellow diffusible pigment was produced in inorganic salt-starch agar, yeast extract-malt extract agar, tyrosine agar and glycerol asparagine agar; a faint yellow pigment was produced in oatmeal agar. The diffusible pigment was not pH indicator. No pigments were produced in peptone-yeast extract iron agar.
Physiological properties of Streptomyces sp. strain NEAE-94
The physiological properties of Streptomyces sp. strain NEAE-94 are listed in Table 1. It utilized trehalose, D(+)xylose, D(+)mannose, rhamnose, D(+)galactose, raffinose, D(-)fructose, D(+)glucose, L-arabinose, sucrose, maltose and cellulose, but could not utilized ribose as the sole carbon source. Streptomyces sp. strain NEAE-94 has reduced nitrate to nitrite. Coagulation and peptonization of milk were positive. Production of protease, α–amylase (Fig. 2C), cellulase, gelatinase and asparaginase was positive. On the other hand, production of uricase, lecithinase and chitosanase was negative. The optimal growth temperature was 30oC and the optimal pH was 7.0. Streptomyces sp. strain NEAE-94 grew in the presence of NaCl up to 5 % (w/v). Streptomyces sp. strain NEAE-94 showed positive antimicrobial activities against Staphylococcus aureus, E. coli, Bacillus subtilis and pseudomonas aeruginosa, but no activities were shown against Candida albicans, Rhizoctonia solani, Aspergillus niger, Fusarium oxysporum, Alternaria solani, Sacchromyces cerevisiae, Bipolaris oryzae or Klebsiella pneumonia. Streptomyces sp. strain NEAE-94 did not produce melanoid pigments in tyrosine agar, peptone-yeast extract iron agar or tryptone-yeast extract broth.
Morphological features of Streptomyces sp. strain NEAE-94
Morphological characteristics of Streptomyces sp. strain NEAE-94 was observed by scanning electron micrograph after incubation on medium of starch nitrate agar medium at 30°C for 14 days. Microscopic observation of Streptomyces sp. strain NEAE-94 showed rectiflexibiles spores chains (Fig. 3). In general, chains of mature spores are long. Spore shape is elongated (0.593– 0.754 x 0.995–1.341 μm), irregular and the spore surface is smooth (Fig. 3).
16S rRNA gene sequence analysis and phylogenetic analysis
The obtained 16S rRNA sequence of Streptomyces sp. strain NEAE-94 was determined which gave an almost complete sequence with 1536 bp and further subjected to the BLAST search [37] of the GenBank database and the results showed homologies with other relevant sequences of many species belonging to the Streptomyces genus. The phylogenetic tree (Fig. 4) was constructed by using the tree-making neighbor-joining algorithm method of Saitou and Nei [43] using MEGA 3.0 software [38]. Streptomyces sp. strain NEAE-94 shared gene similarity of 99.38% to that of Streptomyces anulatus strain BZ10-24, query cover 94% (GenBank accession no. KC493992.1); 99.59% to that of Streptomyces parvus strain 3151, query cover 94% (GenBank accession no. EF063462.1); 99.38% to that of Streptomyces flavofuscus strain NRRL B-2594, query cover 94% (GenBank accession no. EF178690.1) and 99.19% to that of Streptomyces fimicarius strain BWL-H1, query cover 95% (GenBank accession no. MG197994.1).
Taxonomic conclusions
The whole morphological and physiological properties of Streptomyces sp. strain NEAE-94 and its closest phylogenetic neighbors of the genus Streptomyces which showed significant similarities are shown in Table 1. Streptomyces sp. strain NEAE-94 mainly have the same characteristics as Streptomyces anulatus, Streptomyces flavofuscus, Streptomyces parvus and Streptomyces fimicarius in that it produced rectiflexibiles spore chains and did not produce melanin pigments. However, Streptomyces sp. strain NEAE-94 mainly differed from Streptomyces fimicarius in that it produces yellow aerial mycelium, yellow substrate mycelium, and yellow diffusible pigment on ISP medium 2. On the other hand, Streptomyces fimicarius lost the ability to produce a diffusible pigment and not produce distinctive substrate mycelium pigment. Also Streptomyces sp. strain NEAE-94 differed from Streptomyces fimicarius and Streptomyces parvus in the pattern of utilization of carbon sources (Table 1). Moreover, Streptomyces sp. strain NEAE-94 mainly differed from Streptomyces flavofuscus in the diffusible pigment and color of substrate mycelium.
The comparative study between the related species of the genus Streptomyces and Streptomyces sp. strain NEAE-94 (Table 1) indicated that it mostly related to Streptomyces anulatus [44]. Accordingly, Streptomyces sp. strain NEAE-94 was identified as Streptomyces anulatus strain NEAE-94 (accession number is KC354803). The strain (Streptomyces anulatus) has been deposited in the Culture Collection Ain Shams University (CCASU). The culture collection accession number CCASU 20202 is assigned to the deposited strain.
Screening of significant factors for production of cholesterol oxidase using Plackett–Burman design
The design matrix used for screening of the significant factors for production of cholesterol oxidase and the appropriate responses are shown in Table 2. The mycelial growth of Streptomyces anulatus strain NEAE-94 during cholesterol oxidase production in shake flask in submerged fermentation is shown in Fig. 2D. The results showed broad variability of the cholesterol oxidase activity (0.87 to 11.03 U/mL) reflecting the significance of the medium optimization for enhanced cholesterol oxidase production. The highest production of cholesterol oxidase (11.03 U/mL) was achieved in the run no. 18 using 50 mL medium/250 mL conical flask consists of (g/L): Glucose 5; starch 10; cholesterol 3; yeast extract 4; peptone 4; (NH4)2SO4 4; FeSO4.7H2O 0.01; MgSO4.7H2O 0.5; NaCl 0.5; K2HPO4 1 and pH 7; inoculum size was 4 % (v/v) and incubated for 5 day at 37°C using an agitation speed of 150 rpm. In the run no. 4, the lowest production of cholesterol oxidase was obtained (0.87 U/mL) using 50 mL medium/250 mL conical flask consists of (g/L): Glucose 5; starch 10; cholesterol 1; yeast extract 1; peptone 1; (NH4)2SO4 4; FeSO4.7H2O 0.05; MgSO4.7H2O 0.1; NaCl 1; K2HPO4 1 and pH 9; inoculum size was 4 % (v/v) and incubated for 7 day at 37°C using an agitation speed of 100 rpm.
Table 3 shows the statistical analysis of the results of the Plackett–Burman design.
Fig. 5A shows the main effect of the individual independent factors on the cholesterol oxidase production. Fig. 5A revealed that, temperature, agitation speed, pH, starch, cholesterol, peptone, yeast extract, ammonium sulphate and K2HPO4 positively affect cholesterol oxidase production, whereas the remaining factors named incubation time, glucose, inoculum size, MgSO4, NaCl and FeSO4 negatively affect cholesterol oxidase production. The data revealed that, starch (G), peptone (K) and ammonium sulphate (L) with higher P-values (0.9271, 0.9573, 0.9370; respectively), lower effects (0.09, 0.05 and 0.07; respectively) and lower contribution % (0.04, 0.02 and 0.04; respectively) are insignificant factors. The Pareto chart shows absolute effects values and illustrates the significance order of the factors that influence cholesterol oxidase production. The Pareto chart shows a reference line, any absolute effect value extending past this reference line is highly essential (Fig. 5B).
The values of the determination coefficient (R2 = 0.9996) and the adjusted determination coefficient (Adj. R2 = 0.9978) are very high and suggests a strong model significance [45]. The smaller P-value of the factor reveals that the factor is more essential for cholesterol oxidase production. Cholesterol, agitation speed with a P-value of <0.0001 was determined to be the most significant factors, followed by the concentration of yeast extract, incubation time (0.0001) then glucose (0.0002) (Table 3). The P-value < 0.05 (0.0001) implies that the model terms are significant. The F value of the model (573.21) means that it is significant. The first order polynomial equation representing the production of cholesterol oxidase in relation to the independent factors was obtained by neglecting the insignificant factors: (see Equation 3 in the Supplementary Files)
Where Y is the production of cholesterol oxidase and A,B,C,D,E,F,H,K,M,N,O,P are temperature, the time of incubation, size of inoculum, speed of agitation, pH, concentration of glucose, concentration of cholesterol, concentration of yeast extract, concentration of K2HPO4, NaCl, MgSO4 and FeSO4; respectively.
The residuals' normal probability plot is a valuable tool for detecting and explaining the systemic deviations from the normality [46]. Fig. 5C displays a normal probability plot of the residuals. The residuals have been drawn against a theoretical normal model distribution in such a manner that the points for cholesterol oxidase production should form an approximately straight line. Departures from this straight line show deviations from normality. The normal probability plot of the residuals shows points close a diagonal line; so that residuals seem to be distributed nearly normal. This means the model was well designed to the findings of the experiments. Fig. 5D shows the plot of the predicted cholesterol oxidase production versus actual values, while the dots collected around a diagonal line reveals the model's excellent fit.
For assessment of Plackett-Burman design precision, the following production medium was used for cholesterol oxidase production (g/L): FeSO4.7H2O 0.01; MgSO4.7H2O 0.5; NaCl 0.5; K2HPO4 1; yeast extract 4; cholesterol 3 and pH 7. The production medium was inoculated with inoculum size of 4% (v/v) and incubated at a temperature of 37°C in a shaker incubator at 150 rpm for 5 days. The production of cholesterol oxidase using the previous medium was 11.03 U/mL, which was increased 2.45 times compared to the enzyme activity obtained before application of Plackett-Burman design (4.51 U/mL).
Optimization of the selected significant variables by Box–Behnken design
Depending on the statistical analysis of Plackett–Burman design results, the factors with positive impact on production of cholesterol oxidase (temperature, agitation speed, pH, starch, cholesterol, peptone, yeast extract, ammonium sulphate and K2HPO4) have been maintained at their highest levels (+1). Whereas, the factors with negative impact on cholesterol oxidase production (incubation time, inoculum size, MgSO4, NaCl and FeSO4) were fixed at their low (-1) levels. On the other hand, the insignificant factors (starch, peptone and (NH4)2SO4) have been omitted in the subsequent experiments.
Box–Behnken design [42] was used to obtain the optimal levels of the most significant factors influencing cholesterol oxidase production by Streptomyces anulatus strain NEAE-94 and to study the interaction effects among these factors. In the current study, fifteen experiments with various combinations of agitation speed, cholesterol concentration and yeast extract concentration were performed and the experimental and predicted cholesterol oxidase production and residuals for the fifteen trials are provided in Table 4.
Based on the variations in the agitation speed, cholesterol concentration and yeast extract concentration, the results showed variations in the cholesterol oxidase production. Cholesterol oxidase production ranged from 5.64-27.31 U/mL. The lowest cholesterol oxidase production by Streptomyces anulatus strain NEAE-94 (5.64 U/mL) was achieved in the 13th run when the agitation speed was 100 rpm, cholesterol concentration was 4 g/L and yeast extract concentration was 3 g/L. The maximum value of cholesterol oxidase production was achieved in the 12th run with value of 27.31 U/mL, when agitation speed was 150 rpm, cholesterol concentration was 4 g/L and yeast extract concentration was 5 g/L.
The maximum cholesterol oxidase production obtained from this study (27.31 U/mL) by Streptomyces anulatus strain NEAE-94 is superior and greater than most of the other reported values such as the cholesterol oxidase production by Rhodococcus equi no. 23 (0.24 U/mL) [47], Micrococcus sp. (3.68 U/mL) [48], Streptomyces A (2.44 U/mL) [49], Brevibacterium sp. (1.483 U/mL) [50], Streptomyces lavendulae (2.21 U/mL) [51], Streptomyces sp. (6.2 U/mL) [52], Bacillus cereus (1.67 U/mL) [53] and Streptomyces fradiae (0·03 U/mL) [21].
The analysis of variance (ANOVA) for multiple regression analysis
Table 5 contains multiple regression analysis and ANOVA for the results of the Box–Behnken design. The ANOVA of the multiple regression analysis show the model to be highly significant, as can be seen from the low probability value (<0.0001) and the value of Fisher’s F-test (113.82) (Table 5). The current R2 and adjusted R2 values were 0.9951 and 0.9864; respectively, implying that the model is appropriate to represent the real relationship between cholesterol oxidase production and the selected factors. The largest R2 value showed that experimental and expected cholesterol oxidase production values are in excellent agreement [54]. Predicted R2 value of 0.9427 shows that the model is sufficient to predict the value of the production of cholesterol oxidase in the range of factors used. Accuracy and reliability of the model can be seen in the small percentage of the coefficient of variation value (CV=5.44%), mean value (15.16), adequate precision value (31.469), PRESS value (40.03) and standard deviation value (0.82) (Table 5).
The significance of each coefficient was defined in terms of both P and F values listed in Table 5. It can be seen from the P-values and F-values that the linear coefficients of cholesterol concentration, interaction between the agitation speed and cholesterol concentration; agitation speed and yeast extract concentration; cholesterol concentration and yeast extract concentration and quadratic effects of agitation speed, cholesterol concentration and yeast extract concentration are significant as it is evident from the F-values of 119.05, 7.80, 62.24, 7.80, 321.59, 79.45, 526.73; respectively, and P-values of 0.0001, 0.0383, 0.0005, 0.0383, <0.0001, 0.0003, <0.0001; respectively. On the other hand, P-values of the linear coefficients of agitation speed (X1) and yeast extract concentration (X3) indicate that they had nonsignificant effects on cholesterol oxidase production by the strain under study.
The quadratic model of Box–Behnken design used for cholesterol oxidase production by Streptomyces anulatus strain NEAE-94, with a non-significant lack of fit (F-value 1.51 and P-value = 0.4219) and a very low P-value< 0.0001 was shown in the fit summary results (Supplementary Table 2). The largest adjusted and predicted R2 of 0.9864 and 0.9427 and the lowest standard deviation (0.82) was reported in the summary statistics of the quadratic model.
The non- significant lack-of-fit, the high value of adjusted and predicted R-squared, low PRESS value, high F-value, low standard deviation and high adequate precision indicating the validity and high degree of accuracy of the model prediction for the production of cholesterol oxidase by Streptomyces anulatus strain NEAE-94.
The optimum levels of agitation speed, cholesterol concentration and yeast extract concentration giving the maximum cholesterol oxidase production was evaluated by a second-order polynomial equation. Cholesterol oxidase production can be predicted by applying the following second-order regression equation in terms of the independent variables: (see Equation 4 in the Supplementary Files)
Where Y is the cholesterol oxidase production, X1 is the coded value of agitation speed, X2 is the coded value of cholesterol concentration and X3 is the coded value of yeast extract concentration.
Three dimensional (3D) surface and Contour plots
To understand the interaction among the three factors (X1 - X3) and the optimum level of each factor required for the maximum cholesterol oxidase production, the 3D curves and its corresponding contour plots were generated by plotting the cholesterol oxidase production on the Z axis versus two factors are allowed to vary and the third variable is fixed at its zero level (shown in Figs. 6A–C). Fig. 6A represents the cholesterol oxidase production as the simultaneous effect of agitation speed (X1), cholesterol concentration (X2) while yeast extract was kept at the central point (5 g/L). The cholesterol oxidase activity increases gradually by increasing cholesterol concentration and agitation speed till reach its optimum, but further increase in both cholesterol concentration and agitation speed leads to decrease in cholesterol oxidase activity. By solving the equation (4), the highest cholesterol oxidase production of 27.21 U/mL could be reached using 5 g/L yeast extract at the optimal predicted levels of agitation speed and cholesterol concentration of 150 rpm and 4.8 g/L; respectively.
Enhanced production of cholesterol oxidase was recorded with the use of different compounds as cholesterol, yeast extract [21]; malt extract, potato starch and peptone [55] as substrates. Yehia et al. [56] noted that the growth and breakdown of cholesterol by the tested bacterial isolates were largely and severely affected by cholesterol concentration in the culture medium. The largest cholesterol breakdown by the Enterococcus hirae was achieved using 1 g/L cholesterol. Sojo et al. [57] and Yazdi et al. [21] reported that the largest cholesterol breakdown and maximum cholesterol oxidase production by the Rhodococcus erythropolis and Streptomyces fradiae was achieved using 2 g/L cholesterol.
There are different and opposite effects of nitrogen sources on the production of cholesterol oxidase in the literature. Voelker and Altaba [58] recorded an increase in the production of cholesterol oxidase by organic nitrogen much greater than inorganic nitrogen. The reason is that organic nitrogen may contain the most types of growth factors and amino acids required for the microbial growth and could be immediately metabolized by cells, thereby supporting the cholesterol oxidase production [51]. Maximum cholesterol oxidase production by Rhodococcus equi 2C and Rhodococcus equi no. 23 was recorded using 0.3% [21] and 0.4–0.5%, w/v yeast extract [47]; respectively. On the other hand, Moradpour et al. [59] and Ahmad [60] reported that the best sources of nitrogen for cholesterol decomposition by Pseudonocardia compacta S-39 were ammonium nitrate, ammonium sulphate and sodium nitrate [61]. Liu et al. [62] reported that the best sources of nitrogen for maximum cholesterol oxidase production by Arthrobacter simplex were ammonium salts.
Appropriate oxygen must be given in the fermentation media using shaken cultures to satisfy the organism's growing demands and to produce the desired end product. The suitable agitation speed ensures that the dissolved oxygen in the medium is adequately supplied and can become vital for microbial biosynthesis of certain end products.
The three-dimensional surface and contour plots in Fig. 6B illustrates cholesterol oxidase production as a function of agitation speed (X1) and yeast extract concentration (X3) while cholesterol concentration (X2) was fixed at the central point (4 g/L). Fig. 6B indicates that low agitation speed (X1) results in lower cholesterol oxidase production and with increasing the agitation speed, the cholesterol oxidase production increases beyond 150 rpm after which cholesterol oxidase production was reduced. Lower and higher concentrations of yeast extract (X3) results in lower cholesterol oxidase production, and the maximum cholesterol oxidase production, obviously obtained at the central level of the yeast extract concentration. By analysis of Fig. 6B and solving the equation (4), the maximum predicted cholesterol oxidase production of 26.55 U/mL could be reached using 4 g/L cholesterol at the optimal predicted levels of agitation speed (150 rpm) and yeast extract concentration (5 g/L).
Fig 6C shows cholesterol oxidase production as influenced by cholesterol concentration (X2) and the concentration of yeast extract (X3) by maintaining the agitation speed at the central point (150 rpm). With an increased concentrations of both cholesterol and yeast extract, cholesterol oxidase production by the selected strain (Streptomyces anulatus strain NEAE-94) was improved and the maximum cholesterol oxidase production was obtained at the middle levels of two factors, and further increase of cholesterol concentration or yeast extract concentration decreases cholesterol oxidase activity. By analysis of Fig. 6C and solving the equation (4), the maximal predicted cholesterol oxidase production of 27.21 U/mL could be reached using agitation speed (150 rpm) and the optimal predicted levels of 4.8 g/L cholesterol and yeast extract concentration (5 g/L).
In this study, the maximum cholesterol oxidase production (27.31 U/mL) is obtained using the following medium formula (g/L): yeast extract 5, cholesterol 4, FeSO4.7H2O 0.01, MgSO4.7H2O 0.5, NaCl 0.5, K2HPO4 1, pH 7, inoculum size 4 % (v/v), temperature 37°C, agitation speed 150 rpm, medium volume 50 mL and incubation time 5 days.
El-Naggar et al. [14] used the 2-level Plackett–Burman experimental design in 20 experimental run to evaluate the importance of fifteen process variables of medium components and operating conditions for the production of cholesterol oxidase by Streptomyces cavourensis strain NEAE-42. The most important factors that significantly influenced cholesterol oxidase production were initial pH, cholesterol and (NH4)2SO4 concentrations. El-Naggar et al. [14] reported that the optimal levels of the three selected process factors for maximum cholesterol oxidase production (20.521 U/mL) as obtained from the central composite design were pH 8; cholesterol concentration 3 g/L; (NH4)2SO4 8 g/L. However, Ahmad and Goswami [63] optimized the medium for production of cholesterol oxidase by Rhodococcus sp. NCIM 2891 using the classical and statistical methods. They reported that the maximum cholesterol oxidase production (3.25 U/mL) was obtained in the statistically optimized medium under the optimal levels of the process factors that were 2.5g/L (NH4)2HPO4, 9 g/L yeast extract and 3.5 g/L cholesterol in approximately 60 hours of cultivation at 30°C and pH 7.0. On the other hand, Srivastava et al. [64] applied various statistical optimization techniques to enhance the cholesterol oxidase production by Streptomyces rimosus MTCC 10792. They reported that, out of the examined factors, yeast extract, dextrose, starch and ammonium carbonate were the most significant factors and the maximum cholesterol oxidase production was 5.41 U/mL in the optimized medium using the optimum concentrations of the four variables that were (g/100 mL medium): 0.05 ammonium carbonate, 0.1 starch, 0.8 dextrose and 0.99 yeast extract.
Moradpour et al. [59] screen various process variables that had a major impact on cholesterol oxidase production by Streptomyces badius using Plackett-Burman design and optimized these variables using Box-Behnken design. They reported that, yeast extract, pH, Tween 20 and temperature were the most significant factors and the maximum cholesterol oxidase production by Streptomyces badius was 2.05 U/mL in the optimized medium using the optimum levels of the four variables that were determined to be: yeast extract, 0.45 %; pH, 6.5; Tween 20, 0.05 % and 30°C. Moreover, ElBaz et al. [65] used a two-step statistical approach to optimize the production of cholesterol oxidase from Bacillus pumilus. The maximum cholesterol oxidase production (90 U/mL) was obtained after 6 days of fermentation at pH 8 with medium/flask ratio of 0.35 and the concentrations of cholesterol, NH4NO3, yeast extract and Tween 80 were 0.2, 0.3, 1 and 0.2%; respectively.
Kuppusamy and Kumar [53] used traditional one variable at a time method to find the key nutritional components such as different carbon sources, nitrogen sources, metal ions and different physical parameters like incubation time, temperature and pH to enhance cholesterol oxidase production by Bacillus cereus strain KAVK4. The highest production of cholesterol oxidase was achieved under flask conditions using the optimal levels of fructose (as a carbon source) in the production medium at a concentration of 2%, ammonium nitrate (as nitrogen source) at a concentration of 0.2% and magnesium sulphate as metal ion source at a concentration of 0.03%. The maximum production of cholesterol oxidase by Bacillus cereus strain KAVK4 of 1.67 U/mL was achieved at the optimum process variable values (incubation time was at 32 hrs, pH 7.5 at room temperature).
Yang and Zhang [50] used a three level central composite design to investigate the correlation between three independent variables (cholesterol, Tween-80 and treatment time) and cholesterol oxidase production by Brevibacterium sp. They found that, the optimal values of the three independent variables resulted in highest cholesterol oxidase production by Brevibacterium sp. (1.483 U/mL) were determined to be: 22.361 (min) treatment time, 0.2932% (v/v) Tween-80, 4.076 g/L cholesterol. In addition, El-Naggar et al. [22] used the Plackett-Burman design to evaluate the influence of nutritional and environmental variables for cholesterol oxidase production by Streptomyces aegyptia strain NEAE-102. They found that, out of fifteen variables screened by Plackett-Burman design experiments, pH, incubation time and cholesterol concentration were the most significant variables for cholesterol oxidase production. El-Naggar et al. [22] used a face centered central composite design to optimize the levels and analyze the combined effects of pH, incubation time and cholesterol concentration. The optimum levels of these variables for the high cholesterol oxidase production (15.631 U/mL) were determined to be: pH 6, 5 days of incubation time and 3 g/L cholesterol.
Varma and Nene [55] studied cholesterol oxidase production by Streptomyces lavendulae NCIM 2421. A peak of cholesterol oxidase activity of 1.8 U/mL was detected at 72 h. They found that, Streptomyces lavendulae NCIM 2421 is a constitutive producer of cholesterol oxidase where the addition of cholesterol to the medium did not enhance cholesterol oxidase activity. Whereas, Chauhan et al. [51] used orthogonal array method and response surface methodology to optimize medium for cholesterol oxidase production by Streptomyces lavendulae NCIM 2499. They reported that the model predicted maximum cholesterol oxidase production (2.21 U/mL) could be achieved after 72 h of incubation using the medium of the following composition (g/L): sodium chloride 0.7, MgSO4 2, K2HPO4 0.6, soyabean meal 20, malt extract 20, and glycerol 10 mL/L. An initial pH of 7.5 supported the maximum production of cholesterol oxidase. Moreover, Srivastava et al. [64] standardized the process of cholesterol oxidase production by studying a different range of various parameters at shake flask level. They found that the maximum cholesterol oxidase production by Streptomyces rimosus was achieved using the optimal levels of process variables which found to be inoculum size (3%, v/v), pH (7), incubation temperature (30°C), incubation time (48 h) and agitation speed (200 rpm). Fazaeli et al. [66] achieved maximum cholesterol oxidase production by Escherichia coli when the induced culture production medium was incubated for 24 hours at 15°C. Also, Niwas et al. [52] studied the effect of process variables on cholesterol oxidase production by Streptomyces sp. at shake flask level. The results indicated that the maximum cholesterol oxidase production (6.2 U/mL) was achieved using 0.05%, w/v cholesterol, pH 7 and 35°C, 200 rpm.
Verification of the model
Using the optimal levels of the process variables as obtained using Box–Behnken design, the experimental cholesterol oxidase production was verified and compared with the predicted value of cholesterol oxidase production (26.55 U/mL). The maximum experimental cholesterol oxidase production by Streptomyces anulatus strain NEAE-94 was 27.31 U/mL. The verification revealed a high degree of model accuracy (97.21%).