Available online 12 January 2017
- a Laboratorio de Neurofarmacología Celular, Centro de Biofísica y Bioquímica
- b Unidad de Proteómica, Centro de Biología Estructural, Instituto Venezolano de Investigaciones Científicas (IVIC). Apartado 20632 Caracas, 1020-A. Venezuela
- Received 3 October 2016, Revised 26 December 2016, Accepted 9 January 2017, Available online 12 January 2017
- Species from Brownea genus are used in the traditional medicine for treatment of hemorrhages.
- Females with heavy menstrual blood loss (menorrhagia) drink Brownea grandiceps Jacq. flowers decoctions to reduce the bleeding.
- This is the first work about a condensed tannin (called Browplasmin) that present anti-plasmin effect with an activity as effective as reference compounds used for treatment of menorrhagia like tranexamic acid.
Following Venezuelan traditional medicine, females with heavy menstrual blood loss (menorrhagia) drink Brownea grandiceps Jacq. flowers (BG) decoctions to reduce the bleeding. In a previous study, we demonstrated that BG aqueous extract (E) possesses a potent anti-fibrinolytic activity capable of inhibiting plasmin, the main serine-protease that degrades fibrin. It is widely known that plasmin inhibitors are often used as anti-fibrinolytics to reduce bleeding during surgeries with high risk of blood loss such as cardiac, liver, vascular, tooth extraction and large orthopedic procedures, as well as for menorrhagia treatments. The aim of this work was to isolate and characterize from BGE the compound responsible for the reported anti-fibrinolytic activity.
Materials and methods
A decoction of BG was prepared; then it was homogenized, centrifuged and lyophilized to obtain BGE. Subsequently the extract was fractionated by gel filtration and reverse phase using HPLC and the active compound was characterized by MALDI-ToF MS. The kinetic parameters of anti-plasmin activity were evaluated by an amidolytic assay using a chromogenic substrate; also the anti-plasmin activity was estimated by fibrin plate method. Data were analyzed by nonparametric statistics.
The active compound was a condensed tannin denominated Browplasminin, which is capable of inhibiting the plasmin activity in a dose-dependent manner when measured in fibrin plates or by the amidolytic activity method; it also has a minor effect on the FXa activity. However, it does not affect the activity of other serine-proteases such as trypsin, t-PA or u-PA. Browplasminin consists predominately of heteroflavan-3-ols of catechin with B-type linkages, and extents up to heptadecamers (~ 5000 Da), with hexose residues attached to the polymer that presents a high degree of galloylation. Its IC50 for plasmin was 47.80 μg/mL and for FXa was 237.08 μg/mL, while the Ki were 0.76 and 61.61 μg/mL for plasmin and FXa, respectively.
The overall outcome of this study suggests that Browplasminin could be responsible for reducing heavy menstrual bleeding in women because its kinetic parameters showed that is a good plasmin inhibitor.
- BGE, Brownea grandiceps Jacq. flowers aqueous extract;
- HPLC, high-performance liquid chromatography;
- MALDI-ToF, matrix-assisted laser desorption/ionization - time of flight;
- MS, mass spectrometry;
- FXa, factor X activated;
- t-PA, tissue plasminogen activator;
- u-PA, urokinase plasminogen activator;
- TA, tranexamic acid;
- EACA, ε- aminocaproic acid;
- g, grams;
- g, gravity;
- TFA, trifluoroacetic acid;
- ACN, acetonitrile;
- DHB, 2,5-Dihydroxybenzoic acid;
- CI, confidence interval
Chemical compounds studied in this article
- Aprotinin (PubChem CID: 22833874);
- Benzamidine (PubChem CID: 80289);
- ε-aminocaproic acid (PubChem CID: 564);
- Tranexamic acid (PubChem CID: 5526);
- Acetonitrile (PubChem CID: 6342);
- Trifluoroacetic acid (PubChem CID: 6422);
- Ammonium acetate (PubChem CID: 517165);
- Sodium chloride (PubChem CID: 5234);
- Cesium chloride (PubChem CID: 24293);
- Calcium chloride (PubChem CID: 5284359);
- Trizma base (PubChem CID: 6503);
- Hydrochloride acid (PubChem CID: 313);
- Imidazole (PubChem CID: 795)
The menstrual bleeding of women in reproductive age is produced by an activation of local fibrinolysis in the endometrium, which depends on the balance between production and release of plasminogen activators and inhibitors. Endometrial fibrinolysis is excessive in women with heavy menstrual blood loss (menorrhagia), because there are elevated levels of endometrium derived plasmin and plasminogen activators that result in an increased local fibrinolysis (Philipp, 2011). In surgeries with high risk of blood loss such as cardiac, liver, tooth extraction as well as in menorrhagia treatments, plasmin (trypsin-like serine protease that degrades fibrin) is an important target to reduce bleeding (Lien and Milbrand, 2006, Schouten et al., 2009 and Philipp, 2011).
Multiple pharmacological approaches have been proposed in an attempt to attenuate the undue fibrinolytic system activation and to decrease blood loss. Some synthetic or natural molecules such as trans-4-aminomethylcyclohexane-carboxylic acid or tranexamic acid (TA), 6-aminohexanoic acid or ε- aminocaproic acid (EACA) and aprotinin are used as therapeutic agents in plasmin inhibition (Mannucci, 1998, Munoz et al., 1999 and Tengborn, 2012). TA (157 Da) and EACA (131 Da) are plasmin synthetic inhibitors, while aprotinin (6.512 Da) is a natural inhibitor. The first two, are lysine analogous that act as reversible inhibitors by competitively blocking the lysine-binding site of plasminogen/plasmin, thus preventing its attachment to fibrin (Tengborn, 2012), while aprotinin is a polypeptide of 58 residues, isolated from bovine lung, that directly inhibits plasmin and other serine-proteases (Mannucci,1998). Aprotinin is not used to reduce heavy menstrual blood loss anymore; as TA is more potent than EACA to treat menorrhagia, it is the main therapeutic alternative to control this disorder now-a-days (Mannucci, 1998 and Tengborn, 2012).
TA can reduce menstrual bleeding by approximately 50% and in most studies it is administered from days 1 to 4 or 5 of menses, in a dose of 4 g/day, during 3 consecutive menstrual cycles. However, nausea, vomiting, diarrhea, and some impaired color vision has been reported after the administration of TA (Philipp, 2011).
In most developing countries, plant remedies are the most prevalent treatments, where recipes have been handed down from generation to generation and every culture has used decoctions or extracts of leaves, flowers, barks or roots to treat various medical problems as hemorrhages, wound healing, menstrual disorder between others (Agra et al., 2007, Díaz and Ortega, 2006, Klitgaard, 1991a, Klitgaard, 1991b and Otero et al., 2000).
Brownea genus, comprising evergreen small trees with capitate racemes axillary or terminal of red, yellow or white flowers; which are found from the south of Mexico to Peru, including Antillean islands, Jamaica, Trinidad and Tobago ( Agra et al., 2007, Bussmann and Sharon, 2006 and Klitgaard, 1991a). Species from this genus are used in the traditional medicine for treatment of hemorrhages ( Díaz and Ortega, 2006, Márquez et al., 2005 and Otero et al., 2000). In Colombia and Brazil, some plant species from this genus such as B. ariza Bentham and B. rosademonte, are mostly employed in folk medicine as cicatrizant and to neutralize hemorrhages caused by snake venom ( Otero et al., 2000). However, in Venezuela and Ecuador, flowers of B. grandiceps Jacq. (popularly known as mountain rose or rose of Venezuela) are consumed as decoction in traditional medicine to reduce the bleeding in females with heavy menstrual blood loss ( Díaz and Ortega, 2006, Klitgaard, 1991a, Klitgaard, 1991b and Pereira and Brazón, 2015).
Women with menorrhagia inform that decoction of Brownea grandiceps Jacq. flowers (BG) increases their quality of life because this clinical problem alters their emotional, social and working life (personal communication). In a previous study, we demonstrated that BG aqueous extract possesses a potent anti-fibrinolytic activity that inhibits plasmin and which is also a rich source of bioactive compounds with activity on other human proteases ( Pereira and Brazón, 2015). As until now, an anti-plasmin component had not been isolated from BG, the aim of this work was to purify and characterize the plasmin inhibitor from Brownea grandiceps flowers aqueous extract.
2. Materials and methods
Ammonium acetate (CH3COONH4), acetone, NaCl, CsCl, aprotinin, ε- aminocaproic acid (EACA), tranexamic acid (TA), benzamidine, imidazole, trypsin (ref. T1426-1G) and thrombin (ref. 605190-1000U) were obtained from Sigma-Aldrich® (St. Louis, MO, USA). trifluoroacetic acid (TFA) and acetonitrile (ACN) for HPLC were purchased from Merck (Massachusetts, USA). DHB and calibration solution for mass spectrometry were acquired from Bruker Daltonik GmbH (Bremen, Germany). Fibrinogen, chromogenic substrates: S-2251 (H-D-norleucyl-cyclohexylalanyl-lysine para-nitroanilide diacetate salt, ref. 251 L) for plasmin, S-2222 (Methoxycarbonyl-D-cyclohexylglycylglycyl-arginine-para-nitroanilide acetate, ref. 222 L) for factor X activated (FXa) and trypsin, S-2288 (Methylsulfonyl-D-cyclohexyltyrosil-glycyl-arginine paranitroaniline acetate, ref. 444 L) for tissue plasminogen activator (t-PA) and S-2444 (carbobenzoxy-L- γ-glutamyl(α-t-butoxy)-glycyl-arginine paranitroaniline monoacetate salt, ref. 244 L) for urokinase plasminogen activator (u-PA), as the enzymes: plasmin (ref. 411), FXa (ref. 526), t-PA (ref. ADG170) and u-PA (ref. 128) were purchased from Sekisui Diagnostics (Lexington, USA).
2.2. Collection of plant material and preparation of aqueous extract
Brownea grandiceps Jacq. flowers were collected from 3-m-tall wild trees, fertilizers free, in the areas surrounding the Posada Mocundo, Carabobo State, Venezuela (10° 14’ 14’’ N–68° 15’ 44’’ W). The identity of this plant, which was collected in March 2014, was confirmed by Reina Gonto and a voucher specimen (number 5699) was deposited in the herbarium of the Instituto Venezolano de Investigaciones Científicas (IVIC), Caracas, Venezuela.
The Brownea grandiceps flowers aqueous extract (BGE) was prepared from a decoction using 229 g of Brownea grandiceps fresh flowers (1 inflorescence) in 2.5 L of double distilled water at 100 °C for 40 min which was then homogenized in a blender during 5 min. The homogenate was centrifuged at 27.500 g for 30 min at 4 °C; the supernatant was lyophilized, weighed (23 g) and stored at 4 °C until use.
2.3. Purification of plasmin inhibitor
An anti-plasmin inhibitor was isolated from a solution of 6 mg BGE in 100 µL of 20 mM CH3COONH4 pH 4.7, using a Protein Pack 300TM column (7.8×30 cm, Waters, USA) coupled to a Waters HPLC 2695 with a quaternary pump and a photodiode array detector model 2996. Elution of BGE components was performed at a 0.5 mL/min flow rate under isocratic conditions with 20 mM CH3COONH4 pH 4.7 during 60 min at room temperature (20 °C) and detected at a wavelength of 280 nm. Fractions were manually collected and used to determine the inhibitory activity of plasmin. The active fraction (3.21 mg in 100 µL of HPLC water) was applied to an analytic C18 reverse phase column (4.6×250 mm, Vydac, USA) equilibrated with 0.12% TFA in water and eluted with a linear gradient of solvent A (0.12% TFA in water) to 60% of solvent B (0.1% TFA in ACN). The flow rate was 1 mL/min, and the process was monitored at a wavelength of 230 nm during 60 min. The anti-plasmin fraction (0.63 mg in 100 µL water) was re-purified through the same column with a linear gradient of 18 to 27% solvent B in 20 min. The active fraction was then called Browplasminin and was freezed-dried and stored at −20 °C until use.
2.4. MALDI-ToF analysis
The spectra were recorded on a Bruker MALDI-ToF/ToF mass spectrometer Autoflex III Smartbeam (Bruker Daltonics, Germany) provided with a smart beam laser (200 Hz). The mass spectra were acquired in both positive and negative ion mode, over a mass range of 400–5000 Da, with an acceleration voltage of 20 kV and a reflectron voltage of 23 kV; DHB was used as matrix. For analysis in negative mode, Browplasminin was resuspended in water (10 mg/mL). Browplasminin was mixed with a solution of DHB (20 mg/mL in 30% ACN, 0.01% TFA) at equal volumes. Then, 4 μL of the resulting mixture was placed on a MALDI steel target. Alternatively, to promote the formation of ion adducts [M+Na+] or [M+Cs+] in positive mode, Browplasminin was dissolved in acetone 30% (20 mg/mL) and it was mixed with DHB in the same solvent (20 mg/mL) at equal ratios. Two microliters of NaCl (1.5 mg/mL) or CsCl (1.520 mg/mL) were added to the sample/matrix solution and 4 μL of this mix were spotted.
2.5. Inhibition of plasmin amidolytic activity
The effect of BGE and its fractions on plasmin amidolytic activity was evaluated by the method of Guerrero and Arocha-Piñango (1992). The assay was carried out in 96-well polystyrene plates (NuncTM, USA), containing a mixture of 60 µL of recommended buffer for the chromogenic substrate (S-2251), 20 µL of plasmin (0.652 µM) pre-incubated at 37 oC for 30 min with BGE (100 µg/mL), fractions (100 µg/mL) or Browplasminin (6.25, 12.5, 25, 50 and 100 µg/mL) and then, 20 µL of S-2251 (0.8 mM) was added. After, the plate was incubated at 37 oC for 30 min and the absorbance at 405 nm was measured. The results were expressed as residual amidolytic activity percentage. The positive controls were aprotinin (100 µg/mL), EACA (100 µg/mL) and TA (100 µg/mL).
2.6. Inhibitory effect on other serine-proteases
To determine the specificity of the anti-plasmin component, the effect of Browplasminin was also evaluated on trypsin, FXa, t-PA and u-PA amidolytic activity, using a similar protocol as described above. The concentrations of FXa, t-PA, u-PA and trypsin, were 0.7 µM, 1.38 µM, 1 IU/µL and 0.08 µg/µL, respectively. The chromogenic substrates for the enzyme mentioned and the concentrations used were: S-2222 (0.8 mM), S-2288 (1.2 mM) and S-2444 (1 mM) respectively. The results were expressed as residual amidolytic activity percentage. Aprotinin (100 µg/mL), EACA (100 µg/mL), TA (100 µg/mL) and Benzamidine (10 mM) were used as controls.
2.7. Anti-fibrinolytic Activity
Anti-fibrinolytic activity was evaluated by the fibrin plate method (Marsh and Arocha-Piñango, 1972). The fibrin films on 3.5 cm plates were prepared, incubating at 25 °C for 30 min 1.5 mL of 0.1% fibrinogen (containing 10% of plasminogen as contaminant) in imidazole saline buffer and 75 µL of bovine thrombin (10 IU/mL, in 0.025 M CaCl2). Then, ten microliters of a mixture of plasmin (0.652 µM) with BGE, fractions or Browplasminin (100 µg/mL), pre-incubated at 37 °C for 30 min, were applied over the films. Such plates were then incubated at 37 °C for 24 h. The diameters of the lysed areas were measured and the residual fibrinolytic activity was expressed in mm2. Imidazole saline buffer, plasmin, aprotinin (100 µg/mL), EACA (100 µg/mL) and TA (100 µg/mL) were used as controls.
2.8. IC50 Calculation
To determine the concentration of Browplasminin required to inhibit 50% of the amidolytic activity (IC50) of plasmin or FXa, the IC50 was calculated from fitting the dose-response data using a logarithmic non-linear regression analysis performed using the LibreOffice Calc version 220.127.116.11 (The Document Foundation, Germany).
2.9. Kinetics of plasmin inhibition
The IC50 value was also determined directly by the simple method of Copeland et al., 1995, in which the dissociation constant (Ki) value of tight-binding inhibitors of enzymes can be calculated from graphical analysis of the dose-response curve. This alternative method, allowed the construction a dose-response plot as a function of different inhibitor concentrations (6.25, 12.5, 25, 50 and 100 µg/mL) using fixed concentrations of plasmin (0.652 µM) and S-2251 (0.8 mM) as substrate.
The data for the concentration dependent inhibition were analyzed with Eq. (1):
When [Ι] = ΙC50, the value of V / Vο is 0.5 and the value of Vο / V must be 2.0, then the Ki value of competitive tight-binding inhibitors may be expressed in terms of the Eq. (2):
2.10. Statistical analysis
Since the sample size (n=3) was small and the distribution of data unknown, the use of nonparametric statistical procedures was mandatory (Colquhoun, 1971). Data are presented as medians and their 90% confidence interval (CI) calculated with the procedure of Hodges and Lehmann (Hollander and Wolfe, 1973).
3. Results and discussion
3.1. Isolation of the plasmin inhibitor
The aqueous extract of Brownea grandiceps flowers (6 mg in 100 µL water) was fractionated by HPLC using the Protein Pak 300TM semi-preparative size exclusion column under isocratic conditions, four fractions were collected and denominated F1 – F4 as depicted in Fig. 1A. The last fraction (F4) showed the highest activity of all on plasmin. Fraction F4 (3.21 mg in 100 µL water) was further chromatographed by analytical reverse phase HPLC using a C18 column (Fig. 1B) and it was separated in 8 new fractions with retention times between 12.88–30.00 min, these fractions were denoted F4-a – F4-h. Anti-plasmin activity was detected in the F4-h fractions which eluted at 21.10 min, this fraction (0.63 mg in 100 µL water) was rechromatographed in the same column and the active fraction that eluted at 5.78 min was named Browplasminin (Fig. 1C).
Browplasminin was thus isolated employing three chromatographic steps: first, a size exclusion chromatography that allowed obtains the low molecular weight fraction (F4) with high anti-plasmin activity. Second, a separation based on polarity using a reverse phase column where the majority of the components require of 12 to 30 min (12–30% ACN gradient) to elute. This step allowed obtaining a low polarity fraction and with specificity for plasmin (F4-h). Similar results were obtained by Jin and Row (2007) who successfully separated five catechins: the (+) catechin, (-) epicatechin, (-) epigallocatechin gallate and (-) epigallocatechin, using a 10–30% ACN gradient. A third reverse phase step produced the purified Browplasminin.
3.2. MALDI-ToF analysis
Condensed tannins as well as other phenolic compounds can be studied using mass spectrometry techniques, such as MALDI-ToF which detects and allowed to characterize high molecular weight proanthocyanidins (Fulcrand et al., 2008). Browplasminin was analyzed by MALDI-ToF mass spectrometry and [M+Na]+ or [M+Cs]+ pseudo molecular ions were generated and detected separately in positive mode, while [M-H]- pseudo molecular ions were obtained and studied in negative mode. The results showed that Browplasminin consists of predominating heteroflavan-3-ols of catechin with B-type linkages, and extend up to heptadecamers with hexose residues attached to the polymer that presents a high degree of galloylation and a molecular mass of approximately 5000 Da (Table 1, Fig. 2 and Fig. 3).
Polymer Afzelechin Catechin Gallocatechin [M + Na]+ calculated [M + Na]+ detected [M + Cs]+ calculated [M + Cs]+ detected [M - H]- calculated [M - H]- detected Dimer 1 1 0 585 585.0726 695 - 561 561.4493 0 2 0 601 601.0977 711 - 577 577.4679 0 1 1 617 617.1465 727 727.3595 593 593.4865 Trimer 1 2 0 873 873.6940 983 983.6000 849 - 0 3 0 889 889.5397 999 999.5172 865 865.0340 0 2 1 905 905.7588 1015 - 881 881.2339 0 1 2 921 921.6294 1031 1031.4091 897 897.2843 Tetramer 1 0 3 1161 1161.7991 1271 1271.0157 1137 1137.4435 0 4 0 1177 1177.7108 1287 1287.3306 1153 1153.4581 0 3 1 1193 1193.8573 1303 1303.4537 1169 1169.3812 0 2 2 1209 1209.7640 1319 - 1185 1185.5648 Pentamer 1 4 0 1449 1449.0116 1559 1559.2095 1425 1425.4473 0 5 0 1465 1465.3619 1575 1575.2159 1441 1441.7540 0 4 1 1481 1481.3157 1591 1591.4082 1457 1457.7330 Hexamer 1 5 0 1737 1737.2510 1847 1847.4553 1713 1713.2187 0 6 0 1753 1753.2098 1863 1863.2400 1729 1729.1918 0 5 1 1769 1769.2264 1879 1879.1970 1745 1745.1649 Heptamer 1 6 0 2025 2025.2556 2135 2135.3275 2001 2001.1533 0 7 0 2041 2041.2424 2151 2151.3313 2017 2017.1539 0 6 1 2057 2057.3288 2167 2167.3481 2033 2033.1182 Octamer 1 7 0 2313 2313.2548 2423 2423.6880 2289 2289.4788 0 8 0 2329 2329.3163 2439 2439.3253 2305 2305.7157 0 7 0 2345 2345.3376 2455 2455.3824 2321 23212445 Polymer Afzelechin Catechin Gallocatechin [M + Na]+ calculated [M + Na]+ detected [M + Cs]+ calculated [M + Cs]+ detected [M - H]- calculated [M - H]- detected Nonamer 1 8 0 2601 2601.7129 2711 2711.3730 2577 2577.8876 0 9 0 2617 2617.3587 2727 2727.2340 2593 2593.9589 0 8 1 2633 2633.3423 2743 2743.5238 2609 2609.8579 Decamer 1 9 0 2889 - 2999 - 2865 2865.5669 0 10 0 2905 - 3015 3015.8587 2881 2881.3149 0 9 1 2921 - 3031 3031.3059 2897 2897.2576 Undecamer 1 10 0 3177 - 3287 3287.0259 3153 3153.3707 0 11 0 3193 - 3303 3303.1196 3169 3169.7681 0 10 1 3209 - 3319 - 3186 3186.7018 Dodecamer 0 12 0 3481 - 3591 3591.7866 3457 3457.5681 Tridecamer 0 13 0 3769 - 3879 - 3745 3745.6782 Tetradecamer 0 14 0 4057 - 4167 - 4033 4033.3955 Pentadecamrer 0 15 0 4345 - 4455 - 4321 4321.7355 Hexadecamer 0 16 0 4633 - 4743 - 4609 4609.3887 Heptadecamer 0 17 0 4921 - 5031 - 4897 4897.4825
The spectra, summarized in Fig. 2, were characterized by repetitive patterns of peaks with a distance of 288 Da between each set that corresponded to catechin/epicatechin monomeric units (Khanbabaee and van Ree, 2001), the addition of which produced polymeric molecules. Thereby, an equation was developed to predict the mass distribution of polyflavans using the terms: 290+288(a) + c, where 290 represents the molecular weight of the terminal catechin/epicatechin unit, 288 is the molecular weight of the extended catechin/epicatechin unit, the alphabetic character a, represents the degree of polymerization, while alphabetic character c, is the weight of the adduct ion employed (23 for sodium, 133 for Cesium and −1 for hydrogen). So, spectra analysis provided detection of oligomeric sodium adducts that extended up to nonamers (m/z=2617.3587) and of cesium adducts up to dodecamers (m/z=3591.7866) (Table 1). Negative-ion mode allowed more resolution and the detection of masses that correspond to an oligomeric series of catechin/epicatechins up to the heptadecamers (m/z=4897.4825) (Table 1).
Fig. 3A shows that each periodic peak series presented signals 16 Da higher that might be related to the presence of an additional hydroxyl group at the position 5' of the B-ring. Furthermore, it was possible to detect signals 16 Da lower that might the indicated monomers that contain only one hydroxyl group attached to B aromatic moiety. Thus, it was suggested that Browplasminin contains flavan-3-ols type catechin/epicatechins, gallocatechin/epigallocatechins and afzelechin/epiafzelechins, meaning, procyanidins, prodelphinidins and propelargonidins, respectively (Khanbabaee and van Ree, 2001).
The mass spectra also showed that the signals of catechin/epicatechin oligomers are more intense with respect to the gallocatechin/epigallocatechin and afzelechin/epiafzelechin peaks series, indicating that procyanidins were the main constituents of Browplasminin. The masses of sodium adduct ions [M+Na]+ obtained in this work had the same values as in Krueger et al. (2000), who attributed them to the presence of procyanidins, prodelphinidins and properlagonidin polymers in tannins from grape seeds (Krueger et al., 2000 and Reed et al., 2005), leaves and needles from Salix alba, Picea abies, Fagus sylvatica and Tilia cordata ( Behrens et al., 2003) and oligomeric procyanidin fractions from grape seeds and condensed tannin fractions from apple (Mané et al., 2007). Whereas the masses of cesium adducts ions [M+Cs]+ detected here agree with the masses observed for tannins from Canarium album ( Zhang and Lin, 2008), Kandelia candel and Rhizosphora mangle leaves ( Zhang et al., 2010a), Casuarina equisetifolia stem bark and fine roots ( Zhang et al., 2010b), and procyanidins and prodelphinidins from Grevillea robusta leaves ( Wei et al., 2012).
Moreover, the spectra contained masses that are 2 Da lower than those already described (Fig. 3B) that could be attributed to A-type interflavanic linkages that occur between flavan-3-ol subunits (Wei et al., 2011 and Flamini, 2013). These peaks showed lower intensity with respect to the corresponding compounds with B-type bonds, thus implying that, in Browplasminin there are links of B-type and A-type bonds, with the B-type linkages predominating. MALDI-ToF MS has revealed the presence of A-type interflavan bonds in condensed tannins isolated from cranberry (Foo et al., 2000), Tilia cordata ( Behrens et al., 2003), and Machilus pauhoi ( Wei et al., 2011).
In all spectra, the periodical occurrence of sets of peaks with mass differences of 152 Da was detected (Fig. 2, Fig. 3A). These signals correspond to one galloyl ester residue at the heterocyclic C-ring (Flamini, 2013). The negative mode is the best analytical condition for the detection of galloylated structures, followed by spectra performed in positive mode using CsCl, whereas poor signals are expected in analysis carried out with NaCl.
To predict the mass distribution of galloylated polyflavonoids, the equation used was the one described by Da Silva et al. (1992): 290 + 288a+ 152b + c, where 290 is the molecular weight of the terminal catechin/epicatechin unit, 288 is the molecular weight of the extended catechin/epicatechin unit, a represents the degree of polymerization, 152 represents the galloyl group, b is the degree of galloylation, c symbolizes the weight of the adduct ion employed (23 for sodium,133 for Cesium and −1 for hydrogen). The occurrence of galloyl residues on some of the repeating catechin/epicatechin units is one of the most common modifications in proanthocyanidins and has been reported for tannins isolated from oligomeric procyanidin fractions from grape seeds and condensed tannin fractions from apple (Krueger et al., 2000 and Mané et al., 2007), Canarium album plants ( Zhang and Lin, 2008) and Zyzygium cumini fruit ( Zhang and Lin, 2009).
Polyflavonoid peaks were followed by signals separated by 162 Da (Fig. 3C) which can be related to the presence of one glycoside group at the heterocyclic C ring, suggesting that Browplasminin is glycosylated (Zhang et al., 2010a). The glycosylated forms were detected in negative mode and spectra with [M+Cs]+ adducts. Fig. 3B displays an amplification of the spectrum with Cs+with details of the peaks; [M+H]+ adducts were observed with high intensity until a region of m/z=1500. Glycosylation is another important modification on flavonoids and has been detected in condensed tannins from Kandelia candel and Rhizosphora mangle ( Zhang et al., 2010a).
In brief, using three methods of spectrum acquisition for MALDI-ToF MS analysis it was possible to determine the main characteristics and components of the condensed tannin. The best analytical condition was the negative-ion mode which is consistent with the weak acidic nature of the procyanidins, where the proton dissociation in negative mode is much more feasible than protonation in positive-ion mode (Flamini and Traldi, 2010).
3.3. Plasmin inhibition
The results demonstrated that BGE, F4, F4-h, Browplasminin, aprotinin and TA at 100 µg/mL of final concentration, abolished the plasmin activity in fibrin plates, while EACA had a lesser activity (Table 2). Besides, when assayed, the effect of the extract, fractions and Browplasminin at a 100 µg/mL of final concentration on the amidolytic activity of plasmin was observed that Browplasminin inhibited the plasmin activity by 94.97%, whereas aprotinin, TA and EACA diminished the activity by 99.01, 81.70, and 20.72%, respectively (Table 2). These results corroborate those reported by us when we previously studied the BGE ( Pereira and Brazón, 2015). On the other hand, using the amidolytic method, it was observed that Browplasminin increases the inhibitory activity on plasmin in a dose-dependent manner (Fig. 4).
Sample Fibrin lysis (mm2) Inhibitory activity (%) Amidolytic activity (UA/min/µg) Inhibitory activity (%) Plasmin (Pln) 340.5 0 357.67 (230.67–430.67) 0 Pln + BGE 0 100 40.77 (20.88–69.88) 88.99 Pln + F4 0 100 47.89 (43.44–52.44) 86.61 Pln + F4-h 0 100 33.00 (29.89–40.00) 90.77 Pln + Browplasminin 0 100 18.00 (17.67 − 23.33) 94.97 Pln + Aprotinin 0 100 3.56 (3.33–4.00) 99.01 Pln + EACA 250 26.58 283.56 (283.33–288.67) 20.72 Pln + TA 0 100 65.44 (65.00–74.22) 81.70
The inhibitory effect of Browplasminin, a condensed tannin of approximately 5 kDa, has showed to be higher than the reference compounds EACA (131 Da) and TA (157 Da) but similar in molecular mass and inhibitory effect to aprotinin (Table 3). This find could help explain how the (anti-plasmin) anti-fibrinolytic activity of Browplasminin could be related to the reduction of blood loss reported by women that practice the Venezuelan traditional medicine (Díaz and Ortega, 2006 and Pereira and Brazón, 2015).
Sample Amidolytic activity (UA/min/µg) Inhibitory activity (%) Factor Xa (FXa) 339.00 (302.33 − 388.55) 0 FXa + BGE 90.89 (88.44 − 175.33) 73.19 FXa + F4 89.55 (88.00 − 114.22) 73.58 FXa + F4-h 88.21 (68.89–90.11) 73.98 FXa + Browplasminin 87.22 (69.66 − 99.87) 74.27 FXa + Benzamidine 8.44 (7.89 − 10.22) 97.51
This work is the first study that shows anti-plasmin activity in condensed tannins. The inhibitory activity of these compounds has been mostly characterized on digestive enzymes (Tamir and Alumot, 1969; Gonçalves et al., 2007; 2011), protein kinases (Wang et al., 1996), DNA topoisomerase II (Kashiwada et al., 1993), HIV-1 protease (Ma et al., 2000); fatty acid synthase (Zhang et al., 2008) and tyrosinases (Chai et al., 2012 and Chen et al., 2014). Nevertheless, plasmin inhibition has been described in flavonoids isolated from plants like the Blumea balsamifera ( Osaki et al., 2005, Mozzicafreddo et al., 2008 and Cuccioloni et al., 2009).
3.4. Inhibitory activity of Browplasminin on others serine-proteases
In order to determine the specificity of Browplasminin, its inhibitory activity was screened against serine-proteases in the coagulation and fibrinolytic systems using their respective chromogenic substrates. The results showed that Browplasminin had no significant effect on amidolytic activities of trypsin, t-PA or u-PA (data not shown). However, Browplasminin diminished FXa amidolytic activity in a dose-dependent manner, yet its effect was more efficient on plasmin (Fig. 4). In 2015 we reported similar findings when studying BGE activity on t-PA, u-PA, FXa and plasmin at concentrations ≤100 µg/mL (Pereira and Brazón, 2015).
It has been reported that some flavonoids are capable of inhibiting serine proteases related to haemostasis such as epigallocatechin gallate and tannic acid that repress u-PA, plasminogen activator inhibitor and elastase (Jedinak and Maliar, 2005, Cuccioloni et al., 2009 and Cale et al., 2010), quercetin which has inhibitory effects on trypsin, thrombin and u-PA (Jedinak et al., 2006 and Danihelová et al., 2013); delphinidin, cyanidin, pelargonidin and fisetinidin that abolishes the activating protease of Factor VII (Yamamoto et al., 2011).
3.5. Kinetic characteristics of Browplasminin
The IC50 for plasmin and FXa were estimated through a non-linear regression analysis (Fig. 4). Browplasminin inhibited 50% of plasmin activity at concentration of 47.80 μg/mL, while the IC50 for FXa is 237.08 μg/mL. The inhibition constants (Ki) deduced from Eq. (2) were 0.76 and 61.61 μg/mL for plasmin and FXa, respectively. The results indicated that the Browplasminin inhibition is more specific for plasmin. Condensed tannins with similar structural characteristics to Browplasminin exhibited IC50 values between 74 and 128 μg/mL (Mozzicafreddo et al., 2008 and Chen et al., 2014); this evidence suggests that Browplasminin could be considered a good plasmin inhibitor.
A relevant structural detail of Browplasminin is its high degree of galloylation (more hydroxyl groups in its structure), which allow more interaction of these hydroxyl groups through H-bonds with the active site of the enzyme (Mozzicafreddo et al., 2006; Cuccioloni et al., 2009; Chen et al., 2014), this suggests that Browplasminin could have a facilitated access to the catalytic pocket of the plasmin molecule in order to produce the inhibitory effect. In fact, Osaki et al. (2005) demonstrated that flavonoids from Blumea balsamifera with hydroxyl groups on 3' and 4' positions had a potent anti-plasmin effect. Thus, it is possible that a high content of galloyl residues in Browplasminin is related to its inhibitory efficiency. Nevertheless, more studies are needed to determine the type of inhibitor-enzyme interactions and if the galloylation of Browplasminin is involved in the mechanism of plasmin inhibition.
The BGE is a rich source of inhibitory compounds, principally phenolic molecules of a broad range of molecular weights with different efficiencies to inhibit plasmin and FXa. A condensed tannin isolated from BGE, which has been called Browplasminin, mainly showed anti-plasmin activity. Through MALDI-ToF mass spectrometry analysis it was determined that Browplasminin contains predominantly catechins of B-type which form a tannin polymer about 5000 Da that is highly galloylated and presents some residues of hexose. The inhibitory activity on plasmin (IC50=47.80 μg/mL, Ki = 0.76 μg/mL) of Browplasminin could be related to its high degree of galloylation, however, further research is required to determine its plasmin inhibition mechanism.
Statement of authors' contributions to manuscript
B.P (email@example.com) and J.B (firstname.lastname@example.org) conducted research, analyzed data and wrote the paper. J.B designed research. M.R (email@example.com) and E.V (firstname.lastname@example.org) conducted the assay with MALDI-ToF SM. All authors read and approved the final manuscript. The authors declare no conflict of interests.
The authors are indebted to the family Baasch Castrillo of Posada Mocundo, Carabobo State, Venezuela for the supply of Brownea grandiceps flowers. We greatly appreciated the technical assistance of Moises Sandoval, Felix Castillo and Milagros Mendez. This research was supported by grants from the Fondo Nacional de Ciencias, Técnología e Innovación-Programa Estímulo a la Innovación e Investigación (FONACIT-PEII) N° 2012001444 and to the Instituto Venezolano de Investigaciones Científicas (IVIC).
- Agra et al., 2007
- Synopsis of the plants known as medicinal and poisonous in Northeast of Brazil
- Revista Brasilera de Farmacognosia, 17 (2007), pp. 114–140
- Bussmann and Sharon, 2006
- Traditional medicinal plant use in Northern Peru: tracking two thousand years of healing culture
- Journal of Ethnobiology and Ethnomedicine, 2 (2006), pp. 1–18
- Cale et al., 2010
- Characterization of a novel class of polyphenolic inhibitors of plasminogen activator inhibitor-1
- The Journal of Biological Chemistry., 285 (2010), pp. 7892–7902
- Chai et al., 2012
- NMR, HPLC-ESI-MS, and MALDI-ToF MS analysis of condensed tannins from Delonix regia (Bojer ex Hook) Raf. and their bioactivities
- Journal of Agricultural and Food Chemistry., 60 (2012), pp. 5013–5022
- Chen et al., 2014
- Condensed tannins from Ficus virens as tyrosinase inhibitors: structure, inhibitory activity and molecular mechanism
- PloS One., 9 (2014), pp. 1–12
- Colquhoun, 1971
- Lectures on Biostatistics: An Introduction to Statistics with Applications in Biology and Medicine
- Clarendon Press, Oxford (1971)
- Cuccioloni et al., 2009
- Natural occurring polyphenols as template for drug design. Focus on serine proteases
- Chemical Biology and Drug Design., 74 (2009), pp. 1–15
- Danihelová et al., 2013
- Inhibition of pathophysiological proteases with novel quercetin derivatives
- Acta Chimica Slovenica., 6 (2013), pp. 115–122
- Da Silva et al., 1992
- Procyanidin dimers and trimers from grape seeds
- Phytochemistry., 30 (1992), pp. 1259–1264
- Flamini and Traldi, 2010
- Mass spectrometry in grape and wine chemistry
- John Wiley & Sons, Inc, Hoboken, USA (2010)
- Flamini, 2013
- Recent applications of mass spectrometry in the study of grape and wine polyphenols
- IRSN Spectroscopy., 2013 (2013), pp. 1–45
- Gonçalves et al., 2007
- Inhibition of trypsin by condensed tannins and wine
- Journal of Agricultural and Food Chemistry., 55 (2007), pp. 7596–7760
- Hollander and Wolfe, 1973
- Non Parametric Statistical Procedures
- Wiley, New York (1973), p. 503
- Jedinak and Maliar, 2005
- Inhibitors of proteases as anticancer drugs
- Neoplasma., 52 (2005), pp. 185–192
- Jedinak et al., 2006
- Inhibition activities of natural products on serine proteases
- Phytotherapy Research., 20 (2006), pp. 214–217
- Jin and Row, 2007
- Separation of catechin compounds by retention theory in RP-HPLC
- Journal of Liquid Chromatography Related Technologies., 30 (2007), pp. 335–349
- Khanbabaee and van Ree, 2001
- Tannins: classification and definition
- Natural Product Reports., 18 (2001), pp. 641–649
- Klitgaard, 1991a
- Ecuadorian Brownea and Browneopsis (Leguminosae-Caesalpinioideae): Taxonomy, palynology, and morphology
- Nordic Journal of Botany, 11 (1991), pp. 433–449
- Krueger et al., 2000
- Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry of polygalloyl polyflavan-3-ols in grape seed extract
- Journal of Agricultural and Food Chemistry., 48 (2000), pp. 1663–1667
- Lien and Milbrand, 2006
- A disheartening story: aprotinin in cardiac surgery
- Critical Care., 10 (2006), p. 317
- Ma et al., 2000
- Inhibitory effects on HIV-1 protease of constituents from the wood of Xanthoceras sorbifolia
- Journal of Natural Products., 63 (2000), pp. 238–242
- Mané et al., 2007
- Assessment of the molecular weight distribution of tannin fractions through MALDI-TOF MS analysis of protein – tannin complexes
- Analytical Chemistry., 79 (2007), pp. 2239–2248
- Mannucci, 1998
- Hemostatic drugs
- The New England Journal of Medicine., 339 (1998), pp. 245–253
- Márquez et al., 2005
- Evaluación de la actividad coagulante “in vitro” del extracto total en etanol y fracciones de la corteza de Brownea ariza Bentham (Caesalpiniaceae)
- Revista Dugandia, Ciencias Básicas, Uniatlántico, 1 (2005), pp. 111–118
- Marsh and Arocha-Piñango, 1972
- Evaluation of the plate method for estimating plasminogen activators
- Thrombosis et Diathesis Haemorrhagica., 28 (1972), pp. 75–88
- Munoz et al., 1999
- Is ε-aminocaproic acid as effective as aprotinin in reducing bleeding with cardiac surgery
- A meta-analysis. Circulation, 99 (1999), pp. 81–89
- Osaki et al., 2005
- Sesquiterpenoids and plasmin-inhibitory flavonoids from Blumea balsamifera
- Journal of Natural Products., 68 (2005), pp. 447–449
- Schouten et al., 2009
- The effect of aprotinin, tranexamic acid, and aminocaproic acid on blood loss and use of blood products in major pediatric surgery: a meta-analysis
- Pediatric Critical Care Medicine, 10 (2009), pp. 182–190
- Tamir and Alumot, 1969
- Inhibition of digestive enzymes by condensed tannins from green and ripe carobs
- Journal of the Science Food and Agriculture., 20 (1969), pp. 199–202
- Tengborn, 2012
- Fibrinolytic inhibitors in the management of bleeding disorders
- Treatment of Hemophilia., 42 (2012), pp. 1–14
- Wei et al., 2011
- Antioxidant condensed tannins from Machilus pauhoi
- Journal of Medicinal Plants Research., 5 (2011), pp. 796–804
- Wei et al., 2012
- Structural composition and free radical scavenging activity of proanthocyanidins extracted from Grevillea robusta
- Records of Natural Product., 6 (2012), pp. 218–229
- Yamamoto et al., 2011
- Inhibitors of autoactivation of plasma hyaluronan-binding protein (factor VII activating protease)
- Biological &Pharmaceutical Bulletin., 34 (2011), pp. 462–470
- Zhang and Lin, 2008
- Tannins from Canarium album with potent antioxidant activity
- Journal of Zhejiang University Science. B., 9 (2008), pp. 407–415
- Zhang and Lin, 2009
- Antioxidant tannins from Syzygium cumini fruit
- African Journal of Biotechnology., 8 (2009), pp. 2301–2309
- Zhang et al., 2010a
- Condensed tannins from mangrove species Kandelia candel and Rhizophora mangle and their antioxidant activity
- Molecules., 15 (2010), pp. 420–431
- Zhang et al., 2010b
- Antioxidant tannins from stem bark and fine root of Casuarina equisetifolia
- Molecules., 15 (2010), pp. 5658–5670
© 2017 Elsevier B.V. All rights reserved.