Determination of FVIIa-sTF Inhibitors in Toxic Microcystis Cyanobacteria by LC-MS Technique

The blood coagulation cascade involves the human coagulation factors thrombin and an activated factor VII (fVIIa). Thrombin and fVIIa are vitamin-K-dependent clotting factors associated with bleeding, bleeding complications and disorders. Thrombin and fVIIa cause excessive bleeding when treated with vitamin-K antagonists. In this research, we explored different strains of toxic Microcystis aeruginosa and cyanobacteria blooms for the probable fVIIa-soluble Tissue Factor (fVIIa-sTF) inhibitors. The algal cells were subjected to acidification, and reverse phase (ODS) chromatography-solid phase extraction eluted by water to 100% MeOH with 20%-MeOH increments except for M. aeruginosa NIES-89, from the National Institute for Environmental Studies (NIES), which was eluted with 5%-MeOH increments as an isolation procedure to separate aeruginosins 89A and B from co-eluting microcystins. The 40%–80% MeOH fractions of the cyanobacterial extract are active against fVIIa-sTF. The fVIIa-sTF active fractions from cultured cyanobacteria and cyanobacteria blooms were subjected to liquid chromatography-mass spectrometry (LC-MS). The 60% MeOH fraction of M. aeruginosa K139 exhibited an m/z 603 [M + H]+ attributed to aeruginosin K139, and the 40% MeOH fraction of M. aeruginosa NIES-89 displayed ions with m/z 617 [M  ́ SO3 + H]+ and m/z [M + H]+ 717, which attributed to aeruginosin 89. Aeruginosins 102A/B and 298A/B were also observed from other toxic strains of M. aeruginosa with positive fVIIa-sTF inhibitory activity. The active fractions contained cyanobacterial peptides of the aeruginosin class as fVIIa-sTF inhibitors detected by LC-MS.


Introduction
The blood coagulation cascade [1][2][3][4][5] is composed of intrinsic, extrinsic and common pathways involving human coagulation factors. It is initiated by vascular injury and tissue factor (TF) exposure, which triggers the extrinsic pathway [2]. The extrinsic pathway involves activated factor VII-tissue factor (fVIIa-TF) complex activated by Ca 2+ , cephalin or phospholipid [6]. The activation of the fVIIa-TF complex triggers activation of factor X (fX) to activated factor X (fXa) leading to activation of activated factor II (fIIa) or thrombin generation [2]. Thrombin generation needs fVIIa-TF complex, which initiates coagulation and has become the target of therapeutic studies [7]. The activated factor
Analysis of the fVIIa-sTF active extracts of M. aeruginosa K139 and NIES-89 by LC-MS [25] gave a good lead for the active compounds present as fVIIa-sTF inhibitors ( Figure 2, Table 2  We have isolated aeruginosin K139 (30) but unfortunately, complete chemical shift assignments were not determined [26]. The paper by Nishizawa et al. [24] published aeruginosin K139 (30) chemical structure by MS elucidation. However, the stereochemistry of the compound was not deduced. Aeruginosin K139 (30) will be elucidated completely in our next paper. Moreover, aeruginosin K139 (30) has a chemical structure similar to aeruginosin 602 (31) reported by Welker et al. [27]. Aeruginosins K139 (30) and 602 (31) have identical fragmentation pattern reported by Nishizawa et al. [24] and Welker et al. [27]. Both compounds were also elucidated using the LC-MS technique. However, for consistency, this paper will refer aeruginosin with m/z 603 [M + H] + as aeruginosin K139 (30) al. [27]. Aeruginosins K139 (30) and 602 (31) have identical fragmentation pattern reported by Nishizawa et al. [24] and Welker et al. [27]. Both compounds were also elucidated using the LC-MS technique. However, for consistency, this paper will refer aeruginosin with m/z 603 [M + H] + as aeruginosin K139 (30)  We tested other toxic Microcystis strains for the presence of aeruginosins. Aeruginosins could also be found in some other strains of toxic Microcystis, with the presence of aeruginopeptins and microcystins. Indeed, the M. aeruginosa M228 strain was positive against fVIIa-sTF assay. The aeruginopeptins or microcystin-YR (20), with tR 14.9-18.4 min, co-existed with the active compounds. However, testing of the pure compounds of aeruginopeptins and microcystins (Table 1)  We tested other toxic Microcystis strains for the presence of aeruginosins. Aeruginosins could also be found in some other strains of toxic Microcystis, with the presence of aeruginopeptins and microcystins. Indeed, the M. aeruginosa M228 strain was positive against fVIIa-sTF assay. The aeruginopeptins or microcystin-YR (20), with t R 14.9-18.4 min, co-existed with the active compounds. However, testing of the pure compounds of aeruginopeptins and microcystins (Table 1)    The EC 50 s, calculated by Biodatafit [28], of the 40% MeOH fraction of M. aeruginosa NIES-89 containing aeruginosin 89A/B (26/27) were 0.010 µg/mL and 7.123 µg/mL for thrombin and fVIIa, respectively. Thus, the 40% MeOH fraction of M. aeruginosa NIES-89 had computed 0.001 thrombin/fVIIa ratio. The dual inhibitory activity of aeruginosins 89A/B (26/27), and also K139 (30), against thrombin and fVIIa enzymes, make aeruginosins good candidates for fVIIa-sTF inhibitors.
Mar . Drugs 2016, 14, 0000   9 The EC50s, calculated by Biodatafit [28], of the 40% MeOH fraction of M. aeruginosa NIES-89 containing aeruginosin 89A/B (26/27) were 0.010 μg/mL and 7.123 μg/mL for thrombin and fVIIa, respectively. Thus, the 40% MeOH fraction of M. aeruginosa NIES-89 had computed 0.001 thrombin/fVIIa ratio. The dual inhibitory activity of aeruginosins 89A/B (26/27), and also K139 (30), against thrombin and fVIIa enzymes, make aeruginosins good candidates for fVIIa-sTF inhibitors. We have detected aeruginosins 98A (36) and B (37) from M. aeruginosa NIES-98. The MeOH fractions from the aforementioned cyanobacteria are not active in the fVIIa-sTF assay. Thus, from our readings, we compare the fVIIa-sTF inhibitory activity of aeruginosins to phenylamidine. Kadono [29] has denoted the importance of phenylamidine P1 moiety in fVIIa inhibition, which has an inhibitory activity against fVIIa-sTF. The presence of the cyclic amino alcohol moiety in aeruginosins may contribute to efficient binding against fVIIa. However, this hypothesis needs to be established by a structure-activity relationship and subject to another paper. Based on Kadono's paper [29], inhibitors "1-5" with linear structure and containing three peptide bonds exhibit both thrombin and fVIIa inhibitory activities. The number of peptide bonds contributes to the fVIIa inhibitory activity of the compounds and lessens its thrombin inhibition. The addition of one more peptide bond gives promising fVIIa-TF inhibitory activities. This additional peptide bond has been noted in inhibitors "2" to "5" [29] and aeruginosins. The presence of P3 moiety in aeruginosins has certain effects on inhibition of fVIIa and thrombin. The fVIIa and thrombin have the same catalytic triad Ser195-His58-Asp102, S1 pocket, and activation site Arg-Ile [30,31].
Aeruginosins from toxic Microcystis cyanobacteria is a class of fVIIa-sTF inhibitors with thrombin-inhibiting activity. The aeruginosins could be developed into a specific fVIIa-sTF inhibitor that may avoid bleeding and bleeding complications. Some common fVIIa scaffolds from our review [19] have been identified, and we have correlated to the scaffolds of the cyanobacteria origin. The We have detected aeruginosins 98A (36) and B (37) from M. aeruginosa NIES-98. The MeOH fractions from the aforementioned cyanobacteria are not active in the fVIIa-sTF assay. Thus, from our readings, we compare the fVIIa-sTF inhibitory activity of aeruginosins to phenylamidine. Kadono [29] has denoted the importance of phenylamidine P1 moiety in fVIIa inhibition, which has an inhibitory activity against fVIIa-sTF. The presence of the cyclic amino alcohol moiety in aeruginosins may contribute to efficient binding against fVIIa. However, this hypothesis needs to be established by a structure-activity relationship and subject to another paper. Based on Kadono's paper [29], inhibitors "1-5" with linear structure and containing three peptide bonds exhibit both thrombin and fVIIa inhibitory activities. The number of peptide bonds contributes to the fVIIa inhibitory activity of the compounds and lessens its thrombin inhibition. The addition of one more peptide bond gives promising fVIIa-TF inhibitory activities. This additional peptide bond has been noted in inhibitors "2" to "5" [29] and aeruginosins. The presence of P3 moiety in aeruginosins has certain effects on inhibition of fVIIa and thrombin. The fVIIa and thrombin have the same catalytic triad Ser195-His58-Asp102, S1 pocket, and activation site Arg-Ile [30,31].
Aeruginosins from toxic Microcystis cyanobacteria is a class of fVIIa-sTF inhibitors with thrombin-inhibiting activity. The aeruginosins could be developed into a specific fVIIa-sTF inhibitor that may avoid bleeding and bleeding complications. Some common fVIIa scaffolds from our review [19] have been identified, and we have correlated to the scaffolds of the cyanobacteria origin. The arginine and its arginine-derivatives (argininal and argininol) are essential for its fVIIa-sTF inhibition. In addition, structure-activity relationship (SAR) studies will be done in order to deduce the most active scaffold in aeruginosin. We hope to establish a particular SAR study between basic P1 arginine of aeruginosins and fVIIa enzyme. We will also consider the fVIIa enzyme and P3 moiety interaction as proposed in the study. Furthermore, synthesis and modifications have been deemed to make it specific for fVIIa. Assays involving a combination of co-factor(s) and enzymes (TF-fVIIa-fXa-fIIa, etc.) will be performed for a better diagnostic test for the specificity of aeruginosins.

Culture Condition
Five-liter to ten-liter cyanobacterial cultures of 50 strains M. aeruginosa and Anabaena strains were grown in M. aeruginosa (MA) and C medium with N-Tris(hydroxymethyl) methyl-3-aminopropanesulfonic acid (TAPS) rather than Tris (hydroxymethyl) aminomethane (CT) media [39] for fVIIa-sTF and thrombin inhibitory assays. The M. aeruginosa K139 strain was grown in C medium with Bicine in preference for Tris (hydroxymethyl) aminomethane (CB medium) [24]. The M. aeruginosa strains were obtained from Microbial Culture Collection, National Institute for Environmental Studies (NIES), Japan unless otherwise indicated. The cultures were grown in a 5-L glass bottle by aeration at 20˝C for 2-4 weeks with continuous light except M. aeruginosa NIES-89 under 12L:12D cycle. The algal cells were centrifuged using Kubota 7000 centrifuge at 9000 rpm before lyophilization. The lyophilized cells were stored at´30˝C until micro-extraction.

Extraction
The freeze lyophilized algal cells (100 mg) were extracted with 3 mL (ˆ3) 5% acetic acid, homogenized for 30 min, and centrifuged using Kubota 5920 at 4000 rpm. The resulting supernate was evaporated in vacuo at 40˝C. The supernate was eluted by solid phase extraction (SPE) using Sep-Pak Vac 6 mL (1 g) C18/tC18 cartridge (Waters brand). Increasing concentrations of MeOH from water to 100% MeOH with 20% increments was used to elute the supernate. For M. aeruginosa NIES-89, a 5%-increment MeOH was used to separate aeruginosins from microcystins. The cyanobacterial extracts and pure peptides from Microcystis were subjected for in vitro assays. Standard microcystins were bought from Wako Pure Chemical Industries, Ltd., Osaka, Japan. The thrombin assay was performed following the procedure by Anas et al. [22,40,41], in parallel with fVIIa, fVIIa-sTF assays. The crude MeOH fractions active against fVIIa-sTF were subjected to LC-MS experiment to determine the active compounds present.

Serine Protease Inhibitory Assays
All assay experiments were done in a cold condition at 4˝C using an ice bucket until pre-incubation and reaction at 37˝C.

Thrombin Inhibitory Assay
Thrombin inhibitory assays were performed following the procedure of Anas et al. [22,40,41] using 1 mg/mL and 100 µg/mL concentrations with H 2 O, 50% EtOH or 100% EtOH as solvents. The final concentration in each assay was 100 µg/mL and 10 µg/mL, respectively. Leupeptin was used as a positive control from Peptide Institute, Osaka, Japan. The Bz-Phe-Val-Arg¨pNA HCl was purchased from Bachem AG (Bubendorf, Switzerland) and used as a substrate. Solvents H 2 O, 50% EtOH, and 100% EtOH were used as negative controls. Pure compounds were tested at a final concentration of 1 µg/mL unless otherwise indicated.

FVIIa and FVIIa-sTF Assays
Preparation of L-α-Cephalin or 3-sn-Phosphatidylethanolamine Buffer The fVIIa and fVIIa-sTF assays used L-α-cephalin buffer solution. The fVIIa-sTF assay was performed following the procedure by Nakagura et al. [42] with modification. The L-α-cephalin as buffer solution was prepared as follows: Buffer (A): Five hundred milliliters (500 mL) of water was added to 6.057 g of Tris (hydroxymethyl)aminomethane (Nacalai Tesque, Kyoto, Japan) to make 100 mM Tris-HCl solution; 4.383 g NaCl (Nacalai Tesque) was added to the resulting solution to make 100 mM NaCl, and 500 mg bovine serum albumin (BSA) (Sigma, A7284, St. Louis, MO, USA) was added. The pH was adjusted to 7.40; Buffer (B): A 200 mL of Buffer A was added to 0.3329 g of CaCl 2 (Nacalai Tesque). The resulting solution (Buffer B) was adjusted to pH 7.48 before it was stored at 4˝C in preparation for the next day experiment. A 30 µg/mL 3-sn-phosphatidylethanolamine from the bovine brain (Sigma, USA) or L-α-cephalin was added to Buffer B on the day of the experiment.

FVIIa Assay
The 80 µL 3-sn-phosphatidylethanolamine buffer, 50 µL of 100 mM fVIIa enzyme in a buffer, and 20 µL of sample solution were dispensed in each well of a 96-well plate (Iwaki: 3881-096, Tokyo, Japan). The 96-well plate with the solution was pre-incubated at 37˝C for 5 min separately together with 1 mM of Chromozyme t-PA (N-Methylsulfonyl-D-Phe-Gly-Arg-4-nitranilide acetate), from Roche Diagnostics (Mannheim, Germany), dissolved in water as a substrate. The 50 µL of the substrate was added, and the mixture was agitated to start the reaction. The absorbance was noted at 405 nm using Thermo Scientific Multiskan FC microplate photometer until favorable binding was observed.

FVIIa-sTF Assay
The same buffer preparation for fVIIa assay was used for the fVIIa-sTF inhibitory assay. The fVIIa: sTF ratio was 0.30 µg/mL: 0.39 µg/mL, and was prepared in Section 3.3.2.

Preparation of FVIIa Enzyme
The human factor VIIa (HFVIIa) enzyme, purchased from Enzyme Research Laboratories, South Bend, IN, USA, was added and adjusted with 20 mM Tris-HCl/0.1 M NaCl/pH 7.4. The final enzyme concentration should be 95.06 µg/mL. The 100 µL enzyme solutions were stored in plastic cryogenic vials (Iwaki: 2712-002, Tokyo, Japan) at´80˝C until use. The fVIIa enzyme, 95.06 µg/mL, and 100 µL volume solution was added to 7.822 mL of 3-sn-phosphatidylethanolamine buffer on the assay preparation.

Preparation of Soluble Tissue Factor (sTF or F3-28H)
The sTF or Recombinant Human Soluble Tissue Factor (F3-28H) or Human F3 was purchased from Creative Biomart, Shirley, NY, USA. The sTF was added with 10 mM PBS, pH 7.4, to make 1 mM (25.624 µg/mL), and transferred in 300 µL volumes in plastic cryogenic vials (Iwaki: 2712-002, Tokyo, Japan), stored at´80˝C until use. The sTF solution (25.624 µg/mL, 300 µL) was added to 4.7 mL of the 3-sn-phosphatidylethanolamine buffer in an amber bottle before use.

FVIIa-sTF Assay Procedure
The 30 µL buffer, 100 µL fVIIa-sTF, and 20 µL sample solutions were added to a well in a 96-well plate. The solution was pre-incubated at 37˝C for 5 min, together with 1 mM Chromozym t-PA in water as a substrate. A 50-µL substrate was added to start the reaction, agitated, and the absorbance was monitored at 405 nm using Thermo Scientific Multiskan FC microplate photometer. The initial and final readings were noted for 40 min.

LC-MS Preparation of Samples and Determination of fVIIa-sTF Active Compounds
Acetonitrile (99.8% purity) was purchased from Necalai Tesque, Ultrapure Water (LC/MS grade), and Formic Acid (abt. 99%, LC/MS grade) was purchased from Wako Pure Chemical Industries, Ltd., Osaka, Japan. The reversed-phase C18 (ODS) methanol fractions, which were positive for fVIIa-sTF assays, were subjected to LC-MS and dereplicated to know the active compounds present.
One hundred microliters (100 µL) of 100 µg/mL from an EtOH solution of positive ODS fractions was transferred to a small vial. The EtOH solution was evaporated in vacuo at 40˝C before adding 100 µL of 10% MeCN to make up 100 µg/mL solution for LC-MS analysis.
The LC-MS analysis was performed using Thermo Finnigan LCQ deca XP Plus LCMS analytical instrument with Agilent 1100 Series capillary liquid chromatography system. The samples were analyzed using a solvent gradient from 10% MeCN with 0.1% HCOOH to 100% MeCN with 0.1% HCOOH over 60 min. The analysis was done using reversed phase super ODS (TSK-gel, TOSOH Bioscience, Tokyo, Japan) 50ˆ2 mm column, with flow rate 0.2 mL/min, 30˝C column oven, with 200˝C capillary temperature, and UV detection at 220 nm. Solvent optimization of M. aeruginosa NIES-89 40% MeOH fraction used gradient elution from 10% MeCN with 0.1% HCOOH to 15% MeCN with 0.1% HCOOH over 60 min using the aforementioned conditions and parameters. The LC-MS data were processed in Xcalibur Qual Browser ver. 1.2-1.3. The total ion chromatogram (TIC) and extracted ion chromatogram (EIC) were treated, and peaks were identified for the probable compounds present.

Conclusions
This research paves a new avenue for toxic Microcystis study on its role in medical research. We deduce the importance of serine protease inhibitory peptides aeruginosins from toxic Microcystis strains and relate it to the blood coagulation cascade using the LC-MS technique. Argal-containing aeruginosins are potent fVIIa-sTF inhibitors, which could be found in 40% to 80% MeOH ODS fractions in the study. Aeruginosins are potent fVIIa-sTF inhibitors, and we have detected six aeruginosins by LC-MS. The 40% MeOH fraction of M. aeruginosa NIES-89 containing a mixture of aeruginosins 89 A (26) and B (27) displays an EC 50 value of 7.123 µg/mL for fVIIa inhibitory assay and a thrombin inhibitory activity of 0.010 µg/mL. The aeruginosin 89 A (26)/B (27) has a dual inhibitory activity against thrombin and fVIIa with 0.001 thrombin/fVIIa inhibition ratio. We need to develop or increase the thrombin/fVIIa ratio for aeruginosin by subjecting it to a structure-activity relationship (SAR) study in the future. Increasing the thrombin/fVIIa ratio could make aeruginosin more specific to fVIIa, which could be done by peptide modification. Future directions of this research aim to establish the structure-activity relationship (SAR) study of different aeruginosins present in this paper. This research is our preliminary study for aeruginosins as probable fVIIa-sTF inhibitors of the blood coagulation cascade. We aim at establishing the concrete fVIIa-sTF scaffolds, which will result in less bleeding and bleeding complications from cyanobacteria, specifically Microcystis, as our future research. We need to develop a new drug that could inhibit fVIIa with less bleeding and bleeding complications in the future.