Expression optimization, purification, and functional characterization of cholesterol oxidase from Chromobacterium sp. DS1

Cholesterol oxidase is a bifunctional bacterial flavoenzyme which catalyzes oxidation and isomerization of cholesterol. This valuable enzyme has attracted a great deal of attention because of its wide application in the clinical laboratory, synthesis of steroid derived drugs, food industries, and its potentially insecticidal activity. Therefore, development of an efficient protocol for overproduction of cholesterol oxidase could be valuable and beneficial in this regard. The present study examined the role of various parameters (host strain, culture media, induction time, isopropyl ß-D-1-thiogalactopyranoside concentration, as well as post-induction incubation time and temperature) on over-expression of cholesterol oxidase from Chromobacterium sp. DS1. Applying the optimized protocol, the yield of recombinant cholesterol oxidase significantly increased from 92 U/L to 2115 U/L. Under the optimized conditions, the enzyme was produced on a large-scale, and overexpressed cholesterol oxidase was purified from cell lysate by column nickel affinity chromatography. Km and Vmax values of the purified enzyme for cholesterol were estimated using Lineweaver-Burk plot. Further, the optimum pH and optimum temperature for the enzyme activity were determined. This study reports a straightforward protocol for cholesterol oxidase production which can be performed in any laboratory.


Introduction
Cholesterol oxidase (EC 1.1.3.6) is a bacterial flavoenzyme belonging to the oxidoreductase enzyme family and catalyzing the oxidation and isomerization of cholesterol as the first step in the cholesterol catabolism. The enzyme structure includes two domains: the FAD binding domain and the substrate binding domain [1]. The enzyme is classified in two types according to the nature of the bond between FAD cofactor and apoenzyme: Type I, the FAD cofactor non-covalently linked to the protein; and Type II, the cofactor covalently bound to the apoenzyme [2]. Both types of enzymes have found wide applications as useful biotechnological tools. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 pre-culture medium and incubated under the same conditions. Once the cells reached the mid-exponential phase (OD 600 nm ' 0.6), induction of protein expression was performed with addition of 0.5 mM IPTG and incubated at 37˚C while shaking for 6 h. Next, the cells were harvested by centrifugation at 7000 x g, 4˚C, and 10 min, and then the cells were resuspended in 0.5 mL of PBS buffer at pH 7.0. The suspended cells were disrupted by sonication for 10 min at a frequency of 20 kHz, amplitude of 75%, and duty cycle of 0.7 s and the lysate was centrifuged for 20 min at 13000 x g, 4˚C. Enzyme activity assay was conducted based on the method of Allain et al. on the crude extract [27,28]. The assay mixture contained 100 mM potassium phosphate pH 7.0, 1 mM cholesterol, 21 mM phenol, 1.4 mM 4-aminoantipyrine and 5 U/mL peroxidase. The reaction was started by addition of 100 μL sample to 1 mL assay mixture and the appearance of the red chromophore was monitored continuously at 500 nm and the experimental results were reported as mean of three experiments ± SD. Blanks without enzyme or without cholesterol were routinely run in parallel. One unit of activity was defined as the formation of 1 μmol of hydrogen peroxide (0.5 μmol of quioneimine dye) per min at 25˚C.
Culture media optimization. To determine the effect of different culture media on the yield of active ChO, three types of media were evaluated (LB, TB, and SB). The overnight culture of selected Rosetta-gami2(DE3) harboring pET24-ChO plasmid was made in 3 mL of LB medium. Then, 10 mL of each medium was inoculated with the pre-culture bacteria with the ratio of 1:100. When OD 600 nm was reached at 0.6, the cultures were induced with 0.5 mM of IPTG and incubated at 37˚C, 160 rpm for 6 h. The cultures were harvested and the pellets were resuspended in 0.5 mL of PBS buffer. After sonication, the cell lysate was centrifuged at 13000 x g, 4˚C, and 20 min. The total activity of the recombinant enzyme was measured by performing enzyme assay in the crude extract for production determination.
Optimum induction time. Four shaking flasks containing 10 mL of TB/Antibiotics medium were inoculated by 100 μL of overnight cultured Rosetta-gami2(DE3)-pET24-ChO and incubated at 37˚C, 160 rpm. When OD 600 nm of the cultures reached 0.3, 0.6, 1.2, and 1.8, induction was made through 0.5 mM IPTG. The induced cultures were incubated for 6 h at 37˚C, 160 rpm. After disruption of harvested cells and centrifugation, quantification of active enzyme was performed by enzyme activity assay.
Optimum inducer concentration. Rosetta-gami2(DE3) cells containing pET24-ChO were grown overnight in LB/Antibiotics medium. Fresh cultures (5 flasks) containing 10 mL TB medium were inoculated (1:100) and incubated at 37˚C, 160 rpm. When the OD 600 nm of cultures reached 0.6, induction was made individually by IPTG concentrations of 0.05, 0.1, 0.25, 0.5, and 1 mM. After 6 h of incubation at 37˚C, 160 rpm, the cells were harvested and disrupted by sonication and then the enzyme activity was measured.
Induction temperature and post-induction incubation time. The effect of different incubation temperatures (25˚C, 30˚C, and 37˚C) as well as four post-induction incubation times (6,8,16, and 24 h) was evaluated on the ChO production. For this purpose, three flasks containing 20 mL of TB/Antibiotics medium were inoculated by 0.2 mL pre-cultured Rosetta-gami2(DE3) harboring ChO gene, where the induction was made at OD 600 nm ' 0.6 by IPTG at a final concentration of 0.1 mM. After induction, each flask was shaken at 160 rpm and 25˚C, 30˚C, and 37˚C. Appropriate post-induction incubation time was determined through withdrawing 2 mL of culture medium from each flask at different time intervals (6,8,16, and 24 h). The collected samples were centrifuged and pellets were resuspended in the buffer, and then the cells were disrupted by sonication. Enzyme activity assay performed for quantification of the expressed recombinant enzyme in the prepared crude extract.

Large-scale production of recombinant ChO
Large-scale production of ChO was performed under optimized conditions. Pre-culture was made by 5 mL of LB [kanamycin (50μg/mL) and chloramphenicol (25μg/mL)] inoculated by 50 μL of Rosetta-gami2(DE3)-pET24-ChO glycerol stock. Subsequently, 500 mL of TB/Antibiotics was inoculated by 5 mL of pre-cultured media. When OD 600 nm reached 0.6, IPTG (0.1 mM) was added and induction of ChO gene expression was done at 30˚C, 160 rpm for 16 h. The harvested bacterial pellet was resuspended in 10 mL of resuspension buffer (PBS, NaCl 0.3 M, Imidazole 5 mM, pH 7) and disrupted by sonication. The cell lysate was centrifuged at 13000 x g, 4˚C, for 20 min with the supernatant used for ChO purification via affinity chromatography.

Purification of recombinant ChO
Nickel-chelating affinity chromatography (Ni-CAM HC Resin-Sigma) was used for purification of recombinant ChO. For this purpose, initially the column (2 mL) was equilibrated with 30 mL of equilibration buffer (PBS, NaCl 0.3 M, Imidazole 5 mM; pH 7.0) at 1 mL/min. Then, 10 mL of the enzyme crude extract containing 210 mg protein was loaded onto the column, where the column was washed with equilibrium buffer until the absorbance at 280 nm reached the basal level. To elute proteins, elution buffer (PBS, NaCl 0.3 M, and Imidazole 150 mM; pH 7.0) was used and the released proteins were fractionated. The purity of fractionated samples was evaluated by SDS-PAGE 10%. The pure fractions were pooled together and dialyzed against 50 mM sodium phosphate buffer at 4˚C, pH 7.0 for 16 h. To determine the purification yield and specific activity of the recombinant ChO, the protein concentration and enzyme activity of the samples were measured by Bradford [29] and enzyme assay, respectively.

Kinetic characterization of purified ChO
The recombinant ChO activity was assayed at different temperatures (30˚C-80˚C) in order to determine the recombinant enzyme optimum thermal activity. The optimum pH of the recombinant enzyme activity was also determined by the enzyme activity assay at 25˚C under various pH (3)(4)(5)(6)(7)(8)(9)(10)(11) conditions. The buffer systems were prepared according to Doukyu et al. [27]. The K m and V max values for cholesterol was estimated based on Lineweaver-Burk plots of the data obtained from the assay solution containing 0 to 1 mM of cholesterol substrate.

Optimization of recombinant ChO expression
Optimal host strain for ChO expression. The ability of BL21(DE3), BL21(DE3)pLysS, and Rosetta-gami2(DE3) for producing recombinant ChO was evaluated using shake-flask culture. Accordingly, for all of the three cultures, induction were made by 0.5 mM IPTG when OD 600 nm reached 0.6 and protein expression continued for 6 h at 37˚C, 160 rpm. After sonication, the cell lysates were centrifuged and the supernatants were recovered as crude lysates to determine units of enzyme activity produced per liter of culture medium. The Rosetta-gami2 (DE3) cells yielded the highest level of active recombinant ChO with 218 U/L activity (Fig 1). In addition, the total magnitude of active enzyme obtained from BL21(DE3)pLysS was higher compared to the value obtained from BL21(DE3).
Optimal culture medium for ChO expression. Three different culture media (LB, TB, and SB) were evaluated to achieve the optimum production of soluble ChO by the Rosetta-gami2(DE3) cells. To compare the effect of different culture media, overnight culture of Rosetta-gami2(DE3) containing pET24-ChO was made in LB medium at 37˚C. Freshly prepared LB, TB, and SB media were inoculated by pre-culture inoculum (1%) and incubated at 37˚C until OD 600 nm was reached 0.6. Subsequently, the cultures were induced with 0.5 mM IPTG and were grown for another 6 h at 37˚C, 160 rpm. Cell biomass and total enzyme activity obtained by different media types are revealed in Fig 2. The highest cell biomass (OD 600 nm = 4.1) was achieved by TB medium. Also, enzyme assay indicated that TB medium supports high production of recombinant ChO in comparison to the other media.
Pre-induction growth optimization. To determine the best Rosetta-gami2(DE3)-pET24-ChO growth phase for induction of ChO production, four shake flasks were examined in parallel, and each culture was induced at different growth phases. ChO production was induced with 0.5 mM IPTG in early exponential (0.3), mid-exponential (0.6), late exponential (1.2), and stationary (1.8) growth phases. The results, represented in Fig 3, indicate that the highest yield of active enzyme (550 U/L) was obtained when induction occurred in the mid-exponential (0.6) growth phase. Furthermore, higher cell biomass (OD 600 nm = 5.1) accumulation was observed when induction was made in the stationary (1. Optimal induction temperature and post-induction incubation time. Under the best conditions achieved so far [Rosetta-gami2(DE3), TB medium, induction at OD 600 nm = 0.6, and 0.1 mM IPTG], the effect of various induction temperatures as well as post-induction incubation times was investigated on ChO production. To this end, three flasks containing TB medium were cultivated under optimized conditions. After addition of IPTG (0.1 mM), the flasks were incubated at 25˚C, 30˚C, and 37˚C, separately, with 2 mL samples withdrawn from each flask at different time intervals (6,8,16, and 24 h). ChO activity assay showed that the yield of recombinant active enzyme markedly increased when the induced culture medium was incubated at 30˚C for 16 h. As represented in Fig 5,

Large-scale enzyme production
Putting all the optimal conditions together, we made 500 mL culture of Rosetta-gami2(DE3)-pET24-ChO at 30˚C for 16 h through induction by 0.1 mM IPTG in the mid-exponential growth Expression optimization, purification and characterization of cholesterol oxidase phase (OD 600 nm ' 0.6). After the enzyme crude extract preparation, the total enzyme activity and total protein concentration were measured. As reported in Table 1, the total enzyme activity and total protein content were 1160 U and 380 mg for 500 mL of the culture, respectively.

Purification of recombinant ChO
Nickel column affinity chromatography was used for purification of the recombinant ChO which contained 6×His -Tag in its N-terminal. The result obtained from SDS-PAGE analysis of eluted fractions indicated that pure ChO was mainly eluted by 150 mM imidazole. As displayed in Fig 6, according to lanes 2-9, ChO was highly purified. Fractions containing pure ChO were pooled and dialyzed against 50 mM sodium phosphate buffer at pH 7. Table 1 summarizes the data of purification steps. The overall yield of 82.5% and the overall purification growth of 3.1 times were achieved through Ni-CAM affinity chromatography.

Properties of the purified cholesterol oxidase
The activity of the recombinant ChO was determined at different temperatures and pH. The optimum temperature for ChO activity was determined to be 70˚C (Fig 7A). Nevertheless, the enzyme retained more than 60% of its activity at the temperatures between 50˚C to 80˚C Expression optimization, purification and characterization of cholesterol oxidase under the tested conditions. Under various pH conditions, maximal activity of the enzyme was observed at pH 6 ( Fig 7B). To calculate the K m and V max values of purified ChO, the activity of the enzyme was assayed with a range of cholesterol concentrations (0-1mM) at 25˚C, 0.1 M potassium phosphate buffer pH 7. For K m and V max estimation, 1/V was plotted against 1/ [S] in a Lineweaver-Burk plot, with K m and V max values obtained as 31 μM and 29.23 μmol/ min/ mL (Fig 8).

Discussion
Cholesterol oxidase, as an efficient biotechnological tool, is widely used across various industrial and clinical fields [30]. This enzyme is produced from a wide range of bacterial microorganisms with different enzymatic characteristics [5]. Since the thermal stability of the industrial and clinical enzymes is of great importance, selection of an enzyme with a high thermal stability and its successful over-expression could be very valuable. In the current study, we optimized the production of cholesterol oxidase from Chromobacterium sp. DS1 which is more stable at high temperatures than other commercially available ones. Selecting the suitable E. coli host strain is an essential step for the overexpression of a recombinant protein. This can be achieved by considering the expression plasmid in combination with the nature of heterologous protein. For this purpose, ChO-encoding gene was expressed in various E. coli hosts using the pET expression system. Rosetta-gami2(DE3) produced high levels of active enzyme in comparison to the other host strains. pET expression system based on T7 promoter was used for efficient expression of our desired gene. The DE3 strains of E. coli hosts are preferred for expression of recombinant proteins in plasmids containing T7 promoters. The reason is that this strains contain a copy of T7 polymerase gene on their chromosome which produces RNA polymerase for this promoter [31]. On the other hand, the nature of the desired gene has a significant effect on the yield of recombinant protein in different hosts. On the other hand, one of the most common problems encountered in heterologous protein expression is rare codons. To address this problem, two strategies are utilized [32]: one approach is replacing the rare codons with common ones in a new synthetic gene. This strategy in some cases can lead to decreased mRNA stability and protein yield [33]. The second approach is the expression of heterologous protein in engineered strains containing rare tRNAs [34]. An example of such engineered strains is the E. coli Rosetta strain which carries pRARE plasmid providing tRNA genes for the rare codons.
Furthermore, the culture medium composition affects the yield of the recombinant protein expression by altering the bacterial host metabolism [35]. Therefore, culture medium should be accurately selected to achieve the maximum possible value of recombinant protein. In this regard, the effect of three different media (LB, SB, and TB) in the production of the recombinant ChO was evaluated. The results indicated that TB was the best option for the production Expression optimization, purification and characterization of cholesterol oxidase of our desired protein. Superior buffering capacity, high concentration of yeast extract, and the use of glycerol as the carbon source supplement enable high biomass accumulation and high ChO production [36]. The appropriate bacterial growth phase at the time of expression induction and the optimal inducer concentration for maximal ChO production were determined individually. The results indicated that the yield of the enzyme was maximized when induction was made in the midexponential phase. On the other hand, evaluation of the biomass production under induction Expression optimization, purification and characterization of cholesterol oxidase at different growth phases revealed that while the IPTG was added in the early exponential phase, biomass accumulation was lower than that of the stationary growth phase. Indeed, biomass production gradually rose when induction was made at the higher OD600. When induction was made in the early exponential growth phase, bacterial metabolic resources were channeled to produce recombinant protein comprising up to 50% of the total cellular protein [37,38]. Based on this reasoning, low cellular growth rate would be expected following early exponential phase induction. It was also observed that 0.1 mM IPTG concentration was large enough to induce ChO protein expression.
To investigate the effect of different post-induction temperatures and the post-induction incubation time on the yield of recombinant ChO, three incubation temperatures (25˚C, 30˚C, and 37˚C) were evaluated at different incubation times (6, 8, 16, and 24 h). Several studies have suggested that post-induction temperature as well as incubation time can affect the yield of recombinant protein production [39][40][41][42]. In addition, it was reported that different expression temperatures led to different enzymatic activities, although equal amounts of enzymes were synthesized at each temperature [43]. According to our results, reduction of the temperature down to 30˚C along with incubation time of 16 h enhanced the enzyme yield by approximately three times relative to the same condition at 37˚C. Generally, metabolic burden is Expression optimization, purification and characterization of cholesterol oxidase usually observed in bacteria-producing heterologous proteins [44]. High-rate produced recombinant proteins may accumulate in insoluble aggregates (inclusion body) as a direct consequence of overwhelming the host folding machinery [45]. In addition, hydrophobic interactions which constitute a key factor in the formation of inclusion bodies diminish if temperature is reduced [46][47][48].
In the current study, optimization of the recombinant ChO has resulted in 22.98-fold increase (2.115 U/mL) in comparison with the un-optimized condition (0.092 U/mL). El-Naggar et al. optimized cholesterol oxidase production from two Streptomyces strains in their original producer microorganisms. They reported an overall 4.97-fold increase (15.631 U/mL) in Streptomyces aegyptia COX production as compared with the un-optimized condition (3.1 U/ mL) [49]. In the second study, El-Naggar et al. found that the COX production by Streptomyces cavourensis after optimization process was 20.521 which is 6.19-fold higher than result obtained from un-optimized condition (3.31 U/mL) [50].
One step purification procedure was performed for recombinant ChO purification. In the end, the enzyme was purified 3.1 fold and overall yield of 82.5% was achieved. Dokyou et al. performed two step purification procedure for ChO isolation from E.coli Rosetta [51] and Chromobacterium sp.DS1 [52]. They used DEAE-cellulose DE52 column and achieved overall yield of 13% and 40% from E.coli Rosetta and Chromobacterium sp.DS1 respectively.
Characterization of the purified recombinant ChO indicated that the optimum temperature for the enzyme activity was 70˚C at pH 7 and the enzyme retained more than 60% of its activity between 50˚-80˚C. In comparison, optimum temperature for Brevibacterium sterolicum and Burkholderia cepacia cholesterol oxidase were reported to be 50˚C and 60˚C respectively [51]. Our result is in agreement with the result of Doukyu et al. (2009), suggesting that the cholesterol oxidase from Chromobacterium sp. DS1 is an active enzyme across a wide range of temperatures. Moreover, such a high optimum temperature (70ºC) have been reported for COXs from Pseudomonas aeruginosa [53] and Streptomyces fradiae [51]. Furthermore, activity assay at different pH revealed that the optimum pH for enzyme activity was 6. Various studies reported optimum pH of 6 for different cholesterol oxidases for example the COX from Brevibacterium sterolicum [54], Rhodococcus sp. [3], and Streptomyces sp. [12]. The K m value for the cholesterol was calculated to be 31 μM. This value is 5 μM larger than the value calculated by Doukyu et al. for cholesterol oxidase from Chromobacterium sp. DS1. On the other hand, the calculated K m was lower than that of enzymes from Streptomyces sp. SA-COO, B. cepacia ST-200, and P. fluorescens, but not that from Nocardia species [27].
In conclusion, our study indicated that the yield of recombinant ChO was markedly increased when ChO was produced under optimized protocol. In addition, purification and characterization of recombinant ChO verified again that cholesterol oxidase from Chromobacterium sp. DS1 is a thermo-stable enzyme with a broad range of thermal activity.