Cloning and Expression of Genes for Biodegrading Nodularin by Sphingopyxis sp. USTB-05

Biodegradation is efficient for removing cyanobacterial toxins, such as microcystins (MCs) and nodularin (NOD). However, not all the microbial strains with the microcystin-biodegrading enzymes MlrA and MlrC could biodegrade NOD. Studies on genes and enzymes for biodegrading NOD can reveal the function and the biodegradation pathway of NOD. Based on successful cloning and expression of the USTB-05-A and USTB-05-C genes from Sphingopyxis sp. USTB-05, which are responsible for the biodegradation of MCs, the pathway for biodegrading NOD by these two enzymes was investigated in this study. The findings showed that the enzyme USTB-05-A converted cyclic NOD (m/z 825.4516) into its linear type as the first product by hydrolyzing the arginine and Adda peptide bond, and that USTB-05-C cut off the Adda and glutamic acid peptide bond of linearized NOD (m/z 843.4616) and produced dimeric Adda (m/z 663.4377) as the second product. Further, based on the homology modeling of enzyme USTB-05-A, site-directed mutants of USTB-05-A were constructed and seven crucial sites for enzyme USTB-05-A activity were found. A complete enzymatic mechanism for NOD biodegradation by USTB-05-A in the first step was proposed: glutamic acid 172 and histidine 205 activate a water molecule facilitating a nucleophilic attack on the arginine and Adda peptide bond of NOD; tryptophan 176 and tryptophan 201 contact the carboxylate side chain of glutamic acid 172 and accelerate the reaction rates; and histidine 260 and asparagine 264 function as an oxyanion hole to stabilize the transition states.

In Figure 1a,b, the retention time of NOD was 13.46 min. The new peak of product A at the retention time of 7.26 min was highly increased at 12 h with a decrease of NOD. At 12 h, the CE of recombinant pET30a(+)/USTB-05-C/ BL21(DE3) containing 165 mg L −1 of protein was added. A new peak of product B at the retention time of 19.03 min appeared at 13 h and increased continuously until 24 h, while the peak of product A correspondingly decreased and disappeared (Figure 1c,d). The ultraviolet (UV) spectra of peak A and B were similar to that of NOD in the wavelength range of 200 nm to 370 nm (Figure 2), suggesting that Adda remained in product A and B. For the control group of treatment AC with CE of recombinant pET30a(+)/BL21(DE3) without protein USTB-05-C, the concentration of product A was shown to be constant ( Figure S1).    The UV spectra profiles of NOD and its two products.

LC-MS Analysis of NOD and Its Biodegradation Products
The m/z ratios of NOD and its biodegradation products were measured by LC-MS/MS ( Figure  3). The MS analysis of NOD in Figure 3a showed that the major ion (m/z 825.4516) was the protonated molecular ion of NOD [M+H] + . The molecular ion at m/z 843.4616 which was m/z 18 more than NOD indicated that product A was hydrolyzed NOD [M+H2O+H] + (Figure 3b). In the fragmentation spectrum of the m/z 843.4616 ion [M+H2O+H] + (Figure 3c) 16) and the phenylethymethoxy group PhCH2CHOCH3 (m/z 135.0805) via random fragmentation. However, all bonds and ions could be found through major peak analyses, except for the Adda and Arg bond. This analysis showed that enzyme USTB-05-A hydrolyzed the Adda-Arg bond in the first step of NOD biodegradation, converting cyclic NOD to the linear type. The protonated product B was detected at m/z 663.4377 (Figure 3d), coinciding with the dimeric ion of Adda (MW: 331). In the MS/MS spectrum for product B (Figure 3e)  In the control group for treatment A with CE of recombinant pGEX-4T-1/BL21(DE3) without protein USTB-05-A, the concentration of NOD was shown to be constant ( Figure S2). In treatment A with CE containing crude protein USTB-05-A, the results were similar to treatment AC during the first 12 h, but the peaks of product A and NOD remained constant and no other peaks appeared until 24 h. In the control group for treatment C with CE of the recombinant pET30a(+)/BL21(DE3) without protein USTB-05-C and treatment C group with CE containing crude protein USTB-05-C, the concentrations of NOD were shown to be constant ( Figure S3). These results indicated that recombinant enzyme USTB-05-A is involved in the first step of NOD biodegradation, and recombinant enzyme USTB-05-C is involved in the second step.

LC-MS Analysis of NOD and Its Biodegradation Products
The m/z ratios of NOD and its biodegradation products were measured by LC-MS/MS ( Figure 3). The MS analysis of NOD in Figure 3a showed that the major ion (m/z 825.4516) was the protonated molecular ion of NOD [M+H] + . The molecular ion at m/z 843.4616 which was m/z 18 more than NOD indicated that product A was hydrolyzed NOD [M+H 2 O+H] + (Figure 3b

Recombinant USTB-05-A and its Mutants
As shown by the MC-LR biodegradation kinetics in Figure 4, the USTB-05-A enzyme can degrade about 75% of 18.7 mg/L NOD within 6 h and 85% within 24 h. However, the activities of the mutants against NOD were all completely abolished ( Figure 4), which confirmed that these seven sites were crucial for the USTB-05-A enzyme to biodegrade NOD.

Recombinant USTB-05-A and Its Mutants
As shown by the NOD biodegradation kinetics in Figure 4, the USTB-05-A enzyme can degrade about 75% of 18.7 mg/L NOD within 6 h and 85% within 24 h. However, the activities of the mutants against NOD were all completely abolished ( Figure 4), which confirmed that these seven sites were crucial for the USTB-05-A enzyme to biodegrade NOD.

Recombinant USTB-05-A and its Mutants
As shown by the MC-LR biodegradation kinetics in Figure 4, the USTB-05-A enzyme can degrade about 75% of 18.7 mg/L NOD within 6 h and 85% within 24 h. However, the activities of the mutants against NOD were all completely abolished (Figure 4), which confirmed that these seven sites were crucial for the USTB-05-A enzyme to biodegrade NOD.

Discussion
Biodegradation is essential for the reduction of NOD and MCs in water. Bourne et al. have detected the gene cluster for MC biodegradation, which includes mlrA, mlrB, mlrC, and mlrD from Sphingomonas sp. strain MJ-PV [27,35]. The enzyme MlrA encoded by the biodegradation gene mlrA

Discussion
Biodegradation is essential for the reduction of NOD and MCs in water. Bourne et al. have detected the gene cluster for MC biodegradation, which includes mlrA, mlrB, mlrC, and mlrD from Sphingomonas sp. strain MJ-PV [27,35]. The enzyme MlrA encoded by the biodegradation gene mlrA is crucial for opening the cyclic structure of MCs by cleaving the Adda-Arg peptide bond. After, linearized MCs are biodegraded by the enzyme MlrB and Mlr C. Studies about biodegradation genes or pathways of MC degredation can be found, but less information is available on NOD, which is also a typical type of cyanobacterial toxin in natural water bodies [6,46]. Most of the literature emphasized the toxicity, toxin biosynthesis, detection methods of NOD and bacterial strains for NOD biodegradation. It has been shown that not only microbial populations [33,47,48], but also single strains [27][28][29] are able to biodegrade MCs, NOD or both. The bacterial strains in the Sphingomonas genus are the most reported strains to biodegrade MCs. The promising strain Sphingopyxis sp. USTB-05 has a strong ability to biodegrade MCs, and has at least three enzymes encoded by biodegradation genes [31,49]. Sphingopyxis sp. USTB-05 was also found to have the ability to biodegrade NOD and two biodegradation products were observed by HPLC at a wavelength of 238 nm. The pathway for NOD biodegradation by Sphingopyxis sp. USTB-05 was also speculated [44] but not entirely clear as biodegradation by the whole strain of Sphingopyxis sp. USTB-05 was a consecutive process, not step by step.
With the cloned USTB-05-A and USTB-05-C we previously studied on MCs [30], the functions of the USTB-05-A and USTB-05-C enzymes were evaluated for NOD biodegradation (Figures 1 and 2). The results showed that the USTB-05-A enzyme was active in the first step of NOD biodegradation, and the USTB-05-C enzyme participated in further biodegradation from the second step. The USTB-05-A enzyme hydrolyzed cyclic NOD (m/z 825.4516) (Figure 3a) into product A (m/z 843.4616) (Figure 3b). Based on the analysis of the MS/MS spectrum for product A (Figure 3c), it could be inferred that the USTB-05-A enzyme broke the Adda-Arg bond in the ring of NOD through the process of adding one hydrogen on the NH 2 group of Adda and linking one hydroxyl on the carboxyl group of arginine. The linearized NOD was the product in the first step of NOD biodegradation ( Figure 5). The result was similar to that reported by Kato et al.: nodularin cleaved at the Arg-Adda bond by strain B-9 [50]. The Arg-Adda bond break also was the first step for MC biodegradation [27]. Product B (m/z 663.4377) in the second step was observed as the dimeric ion of Adda (MW: 331), which was due to the catalysis of the linearized NOD by the USTB-05-C enzyme (Figure 3d). A previous report [51] showed the same result for the dimeric ion of Adda (m/z 663). However, for the formation of dimeric Adda and the rest of the molecule, there were no explanations in the literature. With the MS/MS spectrum for the second product (Figure 3e), it could be inferred that the recombinant enzyme USTB-05-C cut off the Adda-Glu band of linearized NOD, producing Adda as the product in the second step of biodegradation ( Figure 5).
Based on our previous work on the cloning of USTB-05-A (also known as mlrA) in Sphingopyxis sp. USTB-05 [30] and homology modeling of the USTB-05-A enzyme, seven mutants were constructed. Briefly speaking, the structure for the USTB-05-A enzyme was predicted with eight transmembrane α-helices (TM1-TM8). Following the template of MmRce1 with an active site of H 260 AIH 263 NE 265 , USTB-05-A was assumed to bear an abortive infectivity (ABI) domain composed of four transmembrane helices TM4-TM7. The ABI domain was surrounded by the other four helices, which together form a large cavity accessible to substrates such as NOD. The critical catalytic residues (E140, H173, H227 and N231) of MmRce1 were found to be well conserved in the USTB-05-A enzyme (E172, H205, H260, and N264). Two conserved Trp residues of W144 and W169 in MmRec1 (W176 and W201 in USTB-05-A) were found to face the carboxylate side chain of E140 (E172 in USTB-05-A) for a better activation of catalytic water. Then, these residues were verified by following docking-and site-directed mutation experiments [45]. By comparing the enzyme activities of recombinant USTB-05-A and its mutants, seven crucial sites (E172, W176, W201, H205, H260, N264, and E265) for the activity of the USTB-05-A enzyme were found. For MC-LR biodegradation kinetics in our previous study, the point mutants E172A, H205A, and H260A resulted in a complete loss of enzymatic activity. W176A and W201A mutants led to a decrease in the reaction rate to some extent and N264A results in a significant decrease in MlrA activity [45]. However, for the NOD biodegradation kinetics in this study, all seven mutants were completely inactivated. The results showed that these seven sites were crucial for MC-LR biodegradation, and even more important for NOD biodegradation.

Conclusions
The enzymatic pathway for biodegrading nodularin by Sphingopyxis sp. USTB-05 was clarified in this study. The enzyme USTB-05-A converted cyclic NOD into its linear type as the first product by hydrolyzing the Arg-Adda peptide bond, and USTB-05-C cut off the Adda-Glu peptide bond of linearized NOD and produced Adda as the second product. A complete enzymatic mechanism for NOD biodegradation by USTB-05-A was also proposed. The active site Glu172 and His205 activated a water molecule, facilitating a nucleophilic attack on the Adda-Arg peptide bond of NOD. Trp176 and Trp201 contacted the carboxylate side chain of Glu172 and accelerated the reaction rate. His260 and Asn264 functioned as an oxyanion hole to stabilize the transition states. The findings provide good foundations for investigating the biodegradation mechanism of both MCs and NOD in the future.

Bacterial Strains, Culture Conditions and Reagents
Recombinant pGEX-4T-1/USTB-05-A/BL21(DE3) and pET30a(+)/USTB-05-C/BL21(DE3) were previously studied on MCs and used in this study for NOD biodegradation [30,49]. E. coli TOP10 and E. coli BL21(DE3) were purchased from Sangon Biotech (Shanghai, China). The recombinant and host strains were cultured in Luria-Bertani (LB) medium [41] on a shaker at 200 rpm at an appropriate temperature. The vectors pGEX-4T-1 and pET30a(+), restriction enzymes BamHI, XhoI, SacI and NotI, the plasmid mini-prep kit and the polymerase chain reaction (PCR) kit were obtained from Sangon Biotech (Shanghai, China). Standard NOD was purchased from Enzo Science Inc., Farmingdale, NY, USA. All other chemicals were analytical grade. USTB-05-A is likely a glutamate protease belonging to type II CAAX prenyl endopeptidases. Combined with the biodegradation of NOD by USTB-05-A and its mutants, a complete enzymatic mechanism for NOD biodegradation by USTB-05-A is proposed: Glu172 and His205 activate a water molecule facilitating a nucleophilic attack on the Adda-Arg peptide bond of NOD; Trp176 and Trp201 contact the carboxylate side chain of Glu172 and accelerate the reaction rates; and His260 and Asn264 function as an oxyanion hole to stabilize the transition states.

Conclusions
The enzymatic pathway for biodegrading nodularin by Sphingopyxis sp. USTB-05 was clarified in this study. The enzyme USTB-05-A converted cyclic NOD into its linear type as the first product by hydrolyzing the Arg-Adda peptide bond, and USTB-05-C cut off the Adda-Glu peptide bond of linearized NOD and produced Adda as the second product. A complete enzymatic mechanism for NOD biodegradation by USTB-05-A was also proposed. The active site Glu172 and His205 activated a water molecule, facilitating a nucleophilic attack on the Adda-Arg peptide bond of NOD. Trp176 and Trp201 contacted the carboxylate side chain of Glu172 and accelerated the reaction rate. His260 and Asn264 functioned as an oxyanion hole to stabilize the transition states. The findings provide good foundations for investigating the biodegradation mechanism of both MCs and NOD in the future.

Expression of Gene USTB-05-A and USTB-05-C
The expression of recombinants was performed as previously described [30,49]. The confirmed clones of pGEX-4T-1/USTB-05-A/ BL21(DE3) and pET30a(+)/USTB-05-C/ BL21(DE3) were grown in LB medium with 100 µg mL −1 ampicillin or 50 µg mL −1 kanamycin for 3 h, respectively. The inoculated culture was grown at 37 • C and 200 rpm to an optical density at 600 nm (OD 600nm ) of 0.6, induced with isopropyl-β-D-thiogalactoside (IPTG) at a final concentration of 0.1 mmol L −1 and incubated at 30 • C for 3 h. The flask culture was grown in the same conditions above with an incubation amount of 10% and culture volume of 50 mL per 500 mL shake flask. One liter of recombinant cell culture was harvested by centrifugation (12,000 g, 4 • C, 25 min), washed thrice with PBS (50 mM, pH 7.3) and resuspended in 20 mL PBS to a final concentration of 20-30 g dry cell weight per liter. The cells were disrupted by sonication (400 W, 20 min) in an ice bath, and centrifuged (15,000 rpm, 4 • C, 30 min). The supernatant was collected as CE. Protein content was determined according to Bradford [52].

Enzymatic Activity of Recombinant USTB-05-A and USTB-05-C
The experiments were carried out with five treatments as following:

•
Control groups for treatment A and treatment C: the CEs of recombinant bacteria pGEX-4T-1/BL21(DE3) without protein USTB-05-A and recombinant bacteria pET30a(+)/BL21(DE3) without protein USTB-05-C were added into two tubes with PBS containing NOD, respectively. Treatment AC: at 0 h, the CE containing crude protein USTB-05-A was added into PBS containing NOD. At 12 h, the CE containing crude protein USTB-05-C was added to the above solution.

•
Control group for treatment A and treatment C: at 0 h, the CE containing crude protein USTB-05-A was added into the test tubes with PBS containing NOD. At 12 h, the CE of pET30a(+)/BL21(DE3) without protein USTB-05-C was added to the above solution. Table 1 shows these experiment conditions in brief. In each reaction system of 4 mL, the initial concentration of NOD was 13.7 mg L −1 . The protein concentrations of CEs containing USTB-05-A and USTB-05-C were 158 mg L −1 and 165 mg L −1 , respectively. All test tubes were placed in the incubator at 30 • C with shaking at 200 rpm. Samples of 200 µL were taken at 0 h, 3 h, 12 h, 13 h, 18 h and 24 h, respectively. Concentrated hydrochloric acid (2 µL) was added immediately to stop the reaction. Samples were centrifuged (15,000 rpm, 2 min) and the undiluted supernatants were taken for the detection by HPLC.
For product identification, samples at several reaction time points were taken and purified through the C 18 solid-phase extraction cartridge (Waters, OASIS™ HLB, USA). Methanol (40 µL) was added at a rate of 1 mL min −1 . The treated sample was collected and analyzed by liquid chromatography tandem-mass spectrometry (Agilent 6530 Accurate-Mass Q-TOF LC/MS, Agilent Technologies Inc., Wilmington, DE, USA) to determine the m/z ratios of NOD and the biodegradation products. Table 1. Experiment conditions of the treatments in brief.

Treatment Experiment Condition
Control

Site-Directed Mutagenesis and Enzymatic Activity Comparison
Based on our previous work involving the cloning of USTB-05-A (also known as mlrA) in Sphingopyxis sp. USTB-05 [30] and homology modeling of enzyme USTB-05-A, site-directed mutagenesis for the construction of seven mutants including mlrAE172A, mlrAW176A, mlrAW201A, mlrAH205A, mlrAH260A, mlrAN264A, and mlrAE265A was performed by overlap extension PCR as previously reported [45]. The expression of recombinant USTB-05-A mutants and preparation of CEs were in accordance to the steps for recombinant USTB-05-A mentioned above. For enzymatic activity comparison of recombinant USTB-05-A and mutants, the initial concentration of NOD was 18.7 mg L −1 and the protein concentration of CEs was 350 mg L −1 . Samples of 200 µL were taken at 0 h, 6 h, 9 h, 12 h and 24 h, respectively.

Analysis Methods
The concentration of NOD was detected by high performance liquid chromatography (HPLC) (LC10ATVP, Shimadzu Co., Ltd., Tokyo, Japan) with a UV diode array detector at 238 nm. An Agilent TC-C 18 column (4.6 × 250 mm) (Agilent, 1200 series, Wilmington, DE, USA) was used. Chromatographic grade acetonitrile and water containing 0.05% (v/v) trifluoroacetic acid (35:65, v/v) was used as the mobile phase. The flow rate was 1.0 mL min -1 , and the injection amount was 20 µL. The spectra of NOD and its biodegradation products were detected at the time of HPLC analysis.
For MS detection, an analytical column OOD-4622-ECLC (100 mm × 4.6 mm i.d., 2.6 µm, Phenomenex Inc., Torrance, CA, USA) was used at a maintained temperature of 30 • C. Precursor ions for samples and internal standards were determined from mass spectrum obtained during infusion into the Agilent 6530 Accurate-Mass Q-TOF LC/MS. An electrospray ionization (ESI) source was used. The MS detection conditions were: capillary voltage of 4000 V, nebulizer gas (N 2 ) pressure of 35 psi, desolvation gas flow rate of 10 L min −1 and desolvation gas temperature of 300 • C.