Optimization of L-asparaginase activity of Actinobacteria isolated from Guaviare river sediments in Colombia

Purpose: To optimize the L-asparaginase activity of Actinobacteria isolated from Guaviare river sediments in Colombia. Methods: Actinobacterial strains were evaluated for their L-asparaginase activity using phenol red plates and Nessler’s assays. Strains with L-asparaginase activity were identified based on 16S ribosomal rRNA sequencing, and a central composite design was used to study nutritional and growth factors that could improve L-asparaginase activity. L-asparaginase protein was detected using western blotting and the cytotoxicity of L-asparaginase preparations was evaluated against MDA-MB231 and L929 cell lines. Results: Kitasatospora atroaurantiaca, Streptomyces griseoluteus, and Streptomyces panaciradicis were cultured in medium with lactose as a carbon source and a combination of asparagine and malt extract as nitrogen sources. These strains showed L-asparaginase activities of 29.4, 114.06, and 34.08 U/mg, respectively, and half-maximal inhibitory concentration (IC50) values of 25.61 ± 2.15, 8.18 ± 1.61, and 165.29 ± 1.06 ppm, respectively, against MDA-MB 231 cells. Western blotting analysis revealed the presence of an L-asparaginase monomer with a molecular weight of 37 kDa. Conclusion: Kitasatospora atroaurantiaca, Streptomyces griseoluteus, and Streptomyces Panaciradicis produce L-asparaginases with low L-glutaminase activity and promising cytotoxic activity and thus may be useful for the management of acute lymphoblastic leukemia.


INTRODUCTION
L-asparaginase (L-asparagine amidohydrolase, EC 3.5.1.1) is a potential anti-cancer enzyme. This enzyme decreases the L-asparagine concentration by catalyzing the deamination of the amino acid into L-aspartic acid and ammonium, leading to cell death [1]. Patients with acute lymphoblastic leukemia (ALL) are treated with Escherichia coli L-asparaginase; however, they can suffer an adverse reaction because of the L-glutaminase activity together with the short half-life of the enzyme [3][4].
Therefore, there is a high demand for oncolytic enzymes because cancer cells are more sensitive to them [2]. L-asparaginases produced by microorganisms and plants have been studied, and several attempts have been made to increase their activity [9][10][11][12] using different experimental designs, such as the Box-Behnken [13] and Plackett-Burman [12] methods. Likewise, attempts have been made to produce enzyme recombinantly, employing techniques such as directed evolution and epitope engineering, to introduce the gene into E. coli [8]; however, the recovery of the enzyme was low, it showed poor biochemical properties [6], and was impure [7] Thus, the present study focused on searching for novel sources of L-asparaginase with low Lglutaminase activity, and the effects of Lasparaginase from Actinobacteria isolated from the Guaviare river (Colombia).

EXPERIMENTAL Biological materials
The strains analyzed belonged to the Actinobacteria biobank of La Sabana University and were isolated from sediments of the Guaviare river [18].

Selection of Actinobacteria with L-Asparaginase or L-Glutaminase activity
The morphologies of the strains were evaluated on ISP-3 medium (also known as oatmeal agar) using Gram staining and scanning electron microscopy (SEM) (Phenom pro, Thermo Fisher Scientific, Netherlands) [14]. Strains selection was performed using a phenol red plate assay [15,16]. Strains were cultured in ISP-5-agar (glycerol-L-asparagine) with 0.009 % (v/v) phenol red and incubated at 37 °C for 7 days [15].

Determination of the enzymatic activities
The enzymatic activities were quantified using nesslerization [16]. The strains were grown in ISP-5 medium at 30 °C, with shaking at 150 rpm (Innova 42R, New Brunswick™, USA) for seven days. After centrifugation at 5000 g (Mikro 22R, Hettich, Germany) for 30 min, the biomass was filtered through a 0.22-µm pore size hydrophilic polyvinylidene fluoride (PVDF) membrane, washed and dried at 80 °C for 24 h. The biomass content was measured and the supernatant (5 mL) was subjected to nesslerization [16]. The enzymatic activity is expressed in enzyme units (U), which were defined as the amount of enzyme used to release 1 μmol of ammonium per unit time (U = μmol/min) [14]. Protein quantification was performed using a bicinchoninic acid (BCA) assay with bovine serum albumin (BSA) at 2 mg/mL as a standard [17].

Molecular identification of Actinobacteria strains
The 16S rRNA gene was amplified using polymerase chain reaction (PCR) in a thermocycler (BIO-RAD iCycler, USA) with extracted DNA as the template [18]. The PCR reaction was confirmed by electrophoresis in 1× Tris-Borate-EDTA (TBE) buffer and compared with a ladder of molecular markers (HyperLadder™ IV, Bioline, London). The amplicons were sequenced by BIOS-SIB (Colombia) and compared with entries in the GenBank databases and evaluated according to their percent identity.

Western blotting
Protein extracts (30 μg) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes using a Semi-Dry Blotter (EU-4000, C.B.S. Scientific Company Inc, USA) for 30 minutes at 100 V. The membranes were blocked with 5% non-fat dry milk-TBST (Tris-buffered saline-Tween 20) buffer for 1 h and incubated overnight at 4 °C with a 1:1000 dilution of anti-L-asparaginase antibodies conjugated with horseradish peroxidase (HRP) (GTX 40848, GeneTex) in 1% non-fat dry milk-TBST buffer. Bands were detected using myECL Imager (Thermo Scientific, USA). A commercial Lasparaginase (Medac GmbH, Hamburg, Germany) was used as the positive control.

Evaluation of the effect of carbon and nitrogen sources on the L-asparaginase activity
The best three carbon sources were determined among soluble starch, lactose, glycerol, glucose and sucrose [2,10,15,19] at 1 % (w/v) each, using the phenol red plate assay [15,20]. The best carbon source was then confirmed using nesslerization. The same tests were performed to determine the best nitrogen source among meat extract, yeast extract, potassium nitrate, peptone, malt extract, L-asparagine, and Lglutamine [6,20] at 1% (w/v), and two combinations of L-asparagine with malt extract or potassium nitrate (both at 0.05%).

Optimization of nutritional and growth factors
The influence of different nutritional and fermentation conditions on the L-asparaginase activity was determined using a small central composite design (CCD) with five levels of each factor ( Table 2). The experimental results of CCD were fitted by the response surface regression procedure using a second order polynomial equation: where is the predicted response, 0 is the regression coefficient, is the linear coefficient, is the quadratic coefficient, is the interaction coefficient, and is the coded level of the independent variables [12].

Statistical analysis
Analysis of variance (ANOVA) was applied to the established regression and was carried out for the experimental designs and the models, with statistical significance defined as p < 0.05, in Design Experts® v7 software. The surface plots were constructed using STATISTICA software.

Actinobacteria with L-asparaginase and/or Lglutaminase activity
Gram-positive bacteria with the typical morphology of the Actinobacteria (Figure 1 A-D) were identified using Gram staining and SEM. Actinobacteria showed white aerial and gray vegetative mycelia ( Figure  1E), with well-developed and non-fragmented hyphae, with ramifications accompanied by fragments stick or coconut form, the presence of spores in spiral chain on the aerial mycelium ( Figure 1F), and the presence of branched thin filaments and spiral structures ( Figure 1G); Among 375 strains tested, 20 presented L-asparaginase and/or L-glutaminase activity, as determined by the phenol red assay [1,10] (  Figure 1H).  (5C-486), using Chromas ® [13] and Bioedit ® to obtain the consensus sequences, which were aligned using BLAST ® [22]. Each strain presented identities greater than 99% with the sequences deposited in GenBank.

Western blotting
Bands with a molecular weight of 37 kDa were detected for the five strains (as an example Figure 3 show western blotting for S. panaciradicis). The presumed L-asparaginase bands migrated similarly to the positive control ( Figure 3, lane 2); therefore, we hypothesized that the L-asparaginase enzymes from the isolated Actinobacteria had a similar molecular mass to the L-asparaginase from E. coli. In addition, the intensity of the band increased after each optimization step.

Effect of carbon and nitrogen sources on Lasparaginase activity
The nesslerization experiment showed that the best carbon source to maximize the Lasparaginase activity was lactose Kitasatospora atroaurantiaca, Streptomyces griseoluteus, and Streptomyces panaciradicis were selected for further study.

Optimized nutritional and growth factors
Experiments were randomized using the design matrix and the experimental responses of the Lasparaginase specific activity for the different trials were assessed ( Table 2). The regression equations were obtained ANOVA analysis (Table 1).  All models showed statistical significance (p < 0.05) and the lack of fit was not significant, showing correct correlation between the different studied variables and responses for strain Streptomyces panaciradicis ( Table 3).  The CCD results identified the optimal values for carbon and nitrogen source concentrations, temperature, and pH ( ). The maximum theoretical L-asparaginase activity for Kitasatospora atroaurantiaca was 38.7 U/mg    Cytotoxic activity MDA cells exposed to diverse concentrations of protein broths with L-asparaginase activity showed a dose-dependent decrease in viabilitycompared with untreated cells after 48 h of treatment. The protein broths from the strains presented a half maximal inhibitory concentration (IC 50 ) between 1 and 200 ppm on the MDA cell line ( Table 5). The IC 50 value of doxorubicin was less than 10 ppm against the L929 cell line. The IC 50 values for each enzymatic broth were greater than 200 ppm against L929 cells.

DISCUSSION
When used in chemotherapy, L-asparaginase caused problems related to its side effects associated with its L-glutaminase activity.
Therefore, in the present study, we identified five Actinobacterial strains with high L-asparaginase and no L-glutaminase activity.
Different carbon sources have different effects on L-asparaginase production from Streptomyces. S. longsporusflavus and S. albidoflavus produced the maximum amount of Lasparaginase when grown on soluble starch [13,23] whereas S. phaeochromogenes and S. tendae produced the maximum activity on glycerol and sucrose, respectively [24]. However, few studies used lactose for Actinobacterial Lasparaginase production. Furthermore, the nitrogen sources used during fermentation has an impact on L-asparaginase production [1,2,19]. El-naggar et al [25] suggested that Lasparaginase activity increases with the concentration of L-asparagine in the medium. In the present study, we found that L-asparagine and malt extract stimulated L-asparaginase activity more than using one nitrogen source alone. However, for all the strains evaluated, Lasparagine was not essential to obtain Lasparaginase activity, as seen when malt extract or potassium nitrate alone were used as nitrogen sources.
For Kitasatospora atroaurantiaca and S. panaciradicis, temperature showed a quadratic relationship with the L-asparaginase activity, whereby there was a maximum value beyond which an increase in the factor did not correspond to an increase in activity. The concentrations of carbon and nitrogen sources showed an inverse correlation with the L-asparaginase activity in Streptomyces panaciradicis.
The L-asparaginase activities obtained from Streptomyces panaciradicis (114.06 U/mg), Streptomyces griseoluteus (34.08 U/mg), and Kitasatospora atroaurantiaca (29.4 U/mg) were much greater than those attained from other microorganisms, such as Aspergillus terreus (10.97 U/mg) [10], Streptomyces thermoluteus (4.6 U/mg), and Streptomyces avermitilis (5.6 U/mg) [26]. Kumari et al [17] reported the production of L-asparaginase from Streptomyces griseoluteus isolated form marine sediments, with a enzymatic activity of 16.88 U/mg before purification. Therefore, the three strains identified in the present study produced L-asparaginase with high activity and low levels of L-glutaminase activity.
The Streptomyces panaciradicis, Streptomyces griseoluteus, and Kitasatospora atroaurantiaca Lasparaginases displayed similar cytotoxic activities against MDA cells (0.190, 3.421, and 0.786 mU/mL, respectively) compared with other bacterial L-asparaginases. L-asparaginase purified from Enterobacter cloacae showed an IC 50 value of approximately 11.8 U/mL [27] and Pokrovskaya et al [28] reported that the recombinant L-asparaginase produced by Yersinia pseudotuberculosis showed an IC 50 value of 10 U/mL.

Streptomyces
griseoluteus, Kitasatospora atroaurantiaca, and Streptomyces Panaciradicis (the last two reported for the first time as Lasparaginase producers) are potential sources of L-asparaginase with low L-glutaminase activity. Optimizing the nutritional and growth factors of Kitasatospora atroaurantiaca, Streptomyces griseoluteus, and Streptomyces panaciradicis increases L-asparaginase activity between 9 and 90 times. L-asparaginases from these strains possess good cytotoxicity against MDA-MB-231 cells in vitro and low performance on the untransformed cell line L929, suggesting they are good candidates for the treatment of acute lymphoblastic leukemia.