Collection, plant identification and obtainment of the active fraction

The plant sample (inner bark, 5 kg) was collected at morning from a native population, located in the Caípe Velho Village (11º0’49”S, 37º13’21”W), São Cristóvão, SE, Brazil. The botanical identification was performed by Dr. Ana Paula do Nascimento Prata from the Biology Department of the Federal University of Sergipe (UFS) and a specimen was deposited in the UFS Herbarium under the voucher 014639. This sample was dried in a stove with air circulation at 40°C and after complete dehydration, it was milled for the extraction of the active fraction (hydrometanic fraction - HMF), rich in phenolic compounds (Dias et al., 2013) and whose studies have proved their effectiveness in the wound healing process (Viera et al., 2008; Silva et al., 2010).Thus, the HMF was obtained by liquid-liquid extraction from the ethanolic extract resuspended with 40% (v/v) methanol and successive washes with hexane, chloroform and ethyl acetate, resulting in 40.6% yield in relation to the dry matter. After its preparation, HMF was used for chemical and biological analyzes and for the preparation of the healing formulation (microemulsion with 10% hydrometanic fraction –10% HMFM) with a patent registered under the registration BR1020150073810.

Isolation and identification of the bioactive constituent of the active fraction

HMF (70.2 mg) was initially subjected to flash chromatography using silica gel (70-230 mesh) and AcOet:MeOH (0 to 50%) in polarity gradient. 26 subfractions were obtained which were monitored in TLC using silica 60H and mobile phase buOH: acetic acid: water (65:15:20); being revealed with ceric sulfate and UV light and collected according to the similarity of the chromatographic profile, forming 4 groups. Among them, phenolic profile G-I (4g) was eluted with MeOH in column chromatography (CC) using Sephadex (LD-20), yielding 56 fractions which were concentrated in a rotary evaporator under reduced pressure. These fractions were monitored by TLC following the same procedure described previously and assembled into 4 subgroups (sub-I, sub-II, sub-III and sub-IV) analyzed on HPLC/DAD-UV-vis from 190 to 680 nm. Elution was on analytical (5 μm, 25.0 x 0.46 cm) Shim-pack® PREP-ODS C₁₈ columns, by gradient (20 μL, water: 0-50% MeOH) for 30 min, with 1 mL.min^-1 flow. For elution, the subgroups were vacuum filtered using nylon membranes (2.5 cm x 0.45 cm d.i) and degassed in Ultracleaner 1400. Among them, the sub-I (67 mg) was eluted again on preparative C₁₈ column (5 μm, 25.0 x 2 cm), but in isocratic condition (1000 µL, water: acetonitrile 0.5% acetic acid, 85:15), DAD-UV-vis detection at 280 nm, flow of 5 mL.min^-1with the duration of 30 min, isolating the major compound AC-HMF-1. This isolated compound was concentrated in N₂ gas and analyzed by ¹H and ¹³C NMR in a Bruker 400 Ultrashield apparatus operating at 400 MHz for the ¹H nucleus and at 100 MHz for ¹³C.

Cell Viability of Macrophages J774

The viability of J774 cells (2 x 10⁴ cells) was assessed in triplicate after 24 h of continuous exposure to HMF and measured by the colorimetric assay of tetrazolium salt 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT) reduction to formazan, according to Mosmann (1983). For this, after cell adhesion in a 96-well ELISA reader plate, the culture medium was replaced with 200 μL of culture medium containing the FHM diluted in 0.5% DMSO at respective concentrations of 10, 50 and 100 μg.mL^-1. The plates were then incubated for 24 h. After this time, the culture medium was replaced with 200 μL MTT 0.5 mg.mL^-1, pre-filtered on a 0.22 mm milipore membrane and the plates were incubated for 3 h. After, the supernatant was carefully aspirated and 200 μL of DMSO were added to each well for the solubilization of formazan. The entire contents were transferred to a new ELISA plate and read at 570 nm, whose absorbances were normalized according to equation:

% VC = [DO (treated cells) − DO (blank)/ DO (control) − DO (blank)] × 100.

Antiproliferative activity

The antiproliferative potential of HMF was verified by cytotoxicity analysis using the MTT method (Mosmann, 1983) against the following tumor cell lines: human promyelocytic leukemia (HL-60), human ovary (OVCAR-8), colon carcinoma (HCT-116) and glioblastoma (SF295), provided by the National Cancer Institute (NCI). In a 96-well plate, the cell lines were plated at the concentration of 0.1 × 10⁶ cells.mL^-1and incubated for 24 h in a stove with 5% CO₂ at 37°C. Later, 50 μg.mL^-1 of HMF was added to each well and incubated for 72 h in a 5% CO₂ stove at 37°C. Then, the plate was centrifuged (15 xg/15 min) at 4°C and the supernatants removed. Then 150 μL of the MTT solution (0.5 mg.mL^-1) was added and again incubated for 3 h. The plate was centrifuged (30 g/10 min) at 4°C, the supernatants were discarded and the pellets resuspended in 150 μL of pure sterile DMSO. For the formazan quantification by viable cells, the absorbance was read in an ELISA at 595 nm. Doxorubicin (Glenmark Laboratories) was used as a positive control (0.003 to 0.25 μg.mL^-1). HMF was found to cause more than 75% inhibition of growth in at least one of the cell lines tested and, therefore it was tested with increasing concentrations ranging from 0.024 to 50 μg.mL^-1 to determine its IC₅₀ (minimum inhibitory concentration capable of causing 50% of maximal effect) (Ribeiro et al., 2012). The analyzes were performed in triplicate and the IC₅₀ and their respective confidence intervals (95% CI) were performed by non-linear regression.

DPPH radical-scavenging activity

The determination of the 2,2-diphenyl-1-picrylhydrazyl (DPPH) sequestering ability was done with 10% MFHM and the 10% gallic acid Microemulsion (10% GAM) according to Brand-Wiliams et al. (1995) and the consumption of this free radical was monitored by decreasing the absorbances of test solutions in different concentrations. The samples concentrations used were: 10% HMFM at 15, 20, 40, 50 and 60 μg.mL^-1; 10% GAM at 30, 60, 80 and 100 μg.mL^-1. The absorbance measurements of the mixtures (0.3 mL of sample solution or positive control and 2.7 mL of DPPH stock solution at 40 μg.mL^-1 were made at 515 nm in 1, 5 and 10 min and every 10 min until 1 h. The mixture of methanol (2.7 mL) and methanol solution of the extract (0.3 mL) was used as blank.

From the equation of the calibration curve (y = 110.547 – 0.02804a, 0.998) and the absorbance values in 30 min for each concentration tested, the percentages of remaining DPPH (%DPPH_REM) were determined. The expression of the results was made by the concentration required to decrease the initial concentration of DPPH by 50% (EC₅₀), by Inhibition Percentage (IP) and by the antioxidant activity index (AAI). EC₅₀ was calculated by plotting the %DPPH_REM at 60 min opposed to the sample concentrations being expressed in μg.mL^-1 ± standard deviation. The IP was calculated with de 60 µg.mL^-1 at 30 min. The AAI was calculated according to Scherer and Godoy (2009) in which AAI = %DPPH_REM (end)/EC₅₀ being considered unsatisfactory AAI < 0.5, moderate with 0.5 < AAI < 1.0, strong with 1.0 < AAI < 2.0 and very strong with AAI > 2.0.

Healing activity

The method was performed according to Martins et al. (2006) and Carvalho et al. (2009) with modifications and the tests described were approved by the Animal Research Ethics Committee of the Federal University of Sergipe (CEPA/UFS) under protocol 57/2012. The animals, Rattus norvegicus Wistar line (3.5 months, 200-300 g) were randomly selected, housed in cages at 22 ± 3ºC, with 12 h/12 h cycle, free access to food (Labina ®) and water ad libitum. They were divided into 3 groups: (i) No treatment Group (NTG, n = 15); (ii) Vehicle Group (VG, n = 15) treated with 100 μL of the formulation in the absence of MHF, and; (iii) 10% hydrometanic fraction microemulsion group (10% HMFMG, n = 15) treated with 100 μL of the microemulsion produced from the 10% hydrometanic fraction. Each group was subdivided into 3 subgroups (n = 15) according to the date of the sacrifice period: 7, 14 and 21 days after surgery. The animals were anesthetized intramuscularly with thiopental at a dose of 50 mg.kg^-1 and wound formation was performed by means of a 0.8 mm diameter punch after the application of povidine 1% for local asepsis. The measurement of the area of the wound was done through a digital caliper and the formulation corresponding to each tested group was applied except NTG. At the end of the surgical procedure, the animals were housed in their cages and monitored for anesthetic recovery, as well as normalization of the respiratory rate and the beginning of the active search for water and food. The formulations were applied daily and the animals were examined for their mobility in the open field, presence or absence of secretions, crusts or necroses. On days 7, 14 and 21, animals were weighed, anesthetized with ether, euthanized by decapitation and the wounds measured by digital caliper. The wounds were measured on days 0 (immediately after surgery), 7, 14 and 21 postoperative (immediately after euthanasia) and the area of the wounds was calculated by the equation of Prata et al. (1998): A= π.R.r, where A is the area (cm²); "R", the major radius and "r", the minor radius. The degree of contraction, expressed as percentage, was measured by the equation proposed by Ramsey et al. (1995): % of wound contraction= W0-Wi / W0 x 100, in which W0 = initial wound area on day 0 and Wi= area of the wound on the day of its excision and the results of area and contraction of the wounds were expressed as mean ± standard deviation, submitted to analysis of variance and tested by Tukey (p < 5). The animals were placed in plastic bags for infecting material, frozen in an appropriate freezer and collected through the collection of biological waste by the institution.

Microscopic wounds analysis

The surgical specimens obtained from NTG, VG and 10% GMFHM were submitted to 10 μm thick sections in cryostat, and then placed on glass  slides and stained by the hematoxylin-eosin method (H-E)

Cytotoxicity in J774 Macrophages

HMF significantly favored the viability of J774 macrophages at all concentrations tested (10, 50 and 100 μg.mL^-1) (ρ <0.05) when compared to untreated cells (negative control) showing concentration-dependent behavior (Figure 4). According to Ribeiro et al. (2012), even showing a reduction was not equal to or greater than 75%, from which the plant extract can be considered cytotoxic. On the other hand, for Neri-Numa et al. (2014), an extract that inhibits more than 50% of cell growth and exhibits concentration-dependent behavior is anti-proliferative. The literature warns that this plant species has toxicity, however, such studies used very high concentrations and/or doses in comparison with the data from this study. In this context, Oliveira et al. (2013) demonstrated that A. cochliacarpa bark extracts caused hepatotoxicity in mice at  concentrations  of  125  to  1000 mg.mL^-1, higher than in this study. However, Lima et al. (2014) concluded that the ethanolic extract of Pithecellobium cochliocarpum, synonym of A. cochliacarpa, was a toxic agent in the in vitro tests against Artemiasalina, as well as in pre-clinical tests (acute) by intraperitoneal route. In this route, doses ranged from 160 (maximum dose free of mortality) to 414.72 mg.Kg^-1 (minimum dose capable of leading to 100% death of the animals). However, when orally administered, it showed low toxicity; the doses used ranged from 1 to 5 g.kg^-1. Thus, exposure conditions, administered or absorbed dose, time and frequency of exposure and routes of administration are variables that should be considered.

 

 

Antiproliferative activity

The antiproliferative potential of HMF against four human tumor cell lines, was not satisfactory for all tested strains without acceptable and compatible results with what is established by the National Cancer Institute of the United States (NCI, USA). According to this institute, the IC₅₀ limit value for extracts considered active as or for anticancer agent should be ≤ 30 μg.mL^-1 (Buriol et al., 2009; Ribeiro et al., 2012). Such concentrations, from a pharmacological point of view, are more feasible for potential clinical application (Rosa et al., 2014). In addition, HMF was not able to inhibit 75% survival of the cancer cell lines used, so their IC₅₀ were not determined.

Antioxidant activity

The antioxidant effect of 10% MHMF, produced for wound treatment, was more effective than that used as a positive control (10% GAM) differing significantly (ρ < 0.05) (Table 1).However, when comparing the antioxidant action of this formulation with the antioxidant effect of unincorporated HMF, there is a significant difference when comparing them both, 4.06 ± 1.09 (Dias et al., 2013) which allows to extrapolate that in 6 times for this formulation to reach the same EC_50. However, this result does not eliminate the effect of the 10% MFHM, since the AAI shows that the formulation had strong antioxidant activity, which was not observed with  10%  GAM,  whose potential was moderate.

Still, compared to the others EC₅₀ observed in the literature for extracts of plants considered good antioxidants, it is lower, for example the one observed in the extract of Acacia caesia (L.), whose EC₅₀ was approximately 109 μg.mL^-1, 4.5 times greater than 10% MFHM (Thambiraj and Paulsamy, 2012). The same behavior was observed with methanolic extracts from the leaves and bark of Goniothalamus velutinus (Airy Shaw), with EC₅₀ of 155.32 and 203.96 μg.mL^-1, respectively (Iqbal et al., 2015). It is worth mentioning that, researches carried out in different databases, showed only one study with nanostructured formulation and the same biological purpose, but based on silicone, to which trans-resveratrol was incorporated and whose antioxidant potential was unsatisfactory, inhibiting 19.85% of the DPPH^•. In this sense, both 10% MHMF and 10% GAM acted in the fight against the DPPH free radical action, presenting percentage of inhibition of 93.43 and 61.37%, respectively (Table 1).

 

 

Healing activity

Index of Clinical Wound Retraction (CWR) and area of granulation

According to Figure 5,both 10% HMFMG and VG showed improvement in wound healing at 7 days of treatment, being  equal  to  each other (ρ <0.05) and differing only in relation to NTG. Wound contraction is closely related to myofibroblastic differentiation which is a phenotypic alteration suffered by fibroblasts that acquire contractile phenotype in response to TGF-β, and cytokine released by monuclear leukocytes (lymphocytes and macrophages/histiocytes) in the injured area (Thannickal et al., 2003). In this sense, the CWR and 10% HMFMG were the same, probably due to the presence of oleic acid in the base formulation, since fatty acids stimulate the synthesis of TGF-β and induce the expression of contractile proteins, such as α-actin of smooth muscle, in fibroblasts, differentiating them in myofibroblasts (Mishra and Simonson, 2008; Santos et al., 2014). In fact, several studies have highlighted that unsaturated fatty acids, including the oleic used in this formulation, act in this process by aiding tissue repair (Cardoso et al., 2011). Therefore, the similar behavior may suggest the low effectiveness of HMF because the compounds present did not act synergistically with oleic acid for this purpose, although there is a tendency. However, when the analysis of the granulation area is observed, it can be seen that at 7 days, the lowest area obtained was in 10% HMFMG, while the largest was VG. Namely, the granulation area or tissue is the point of reference for tissue repair. Its histological characteristic consists of the presence of new and small blood vessels, as well as in the proliferation of fibroblasts. Its size depends on the defect in the tissue created by the wound and the intensity of the inflammation (Santos et al., 2014). Thus, it is suggested that HMF compounds acted to reduce the inflammatory process, resulting in a smaller area of granulation. It is noteworthy that the 10% MHMF showed a strong antioxidant behavior, with a percentage of inhibition above 90% (Table 1), which reaffirms the possibility of being active in the reduction of inflammation since antioxidants have a positive correlation with the anti-inflammatory process (Oliveira et al., 2013). Furthermore, previous research by Da Silva et al. (2011) and Saturnino-Oliveira et al. (2014) showed the effectiveness of A. cochliacarpa in the anti-inflammatory action. At 14 days, it was observed that the three groups showed a similar pattern of CWR and granulation area (p <0.05) (Figure 5). However,  regarding  the  observations, at 7 days of treatment, there was an increase in the percentage of retraction of the wounds and an accentuated decrease in the areas of granulation. Such behavior in this period is expected since there is a reduction in the population of myofibroblasts, certainly induced by apoptosis (Ribeiro et al., 2009). However, at 21 days, it was observed that 10% HMFMG presented 100% retraction of the wound area and lower area of granulation differing significantly (ρ <0.05) from VG and NTG, which obtained equal behavior. This event is probably associated with some of the chemical constituents of A. cochliacarpa HMF or their synergistic action, which may have increased the local production of TGF-β, thus increasing myofibroblastic differentiation. Vieira et al. (2008) demonstrated the positive action of flavonoids in the healing process and, later, Dias et al. (2014) studying the chemical profile of A. cochliacarpa have claimed it to be a species rich in flavonoids. In fact, the major constituent of 10% HMFMG, the + (-) catechin is already cited as a flavonoid with anti-inflammatory potential. However, it has already been cited as a TGF-β inhibitor, one of the growth factors closely related to the process of wound contraction, an effect not observed in this experiment. It is worth noting that at 17 days of this experiment, 10% HMFMG animals stopped receiving the topical application since their wounds were already fully contracted. Similar results were obtained in animals treated with the nanoemulsion eucalyptus oil, oil-in-water type, similar to the present study, at 16 days (Sugumar et al., 2014).

 

 

Microscopic wounds analysis

As for collagen deposition (Figure 6), the pattern observed was quite similar between groups over the experimental times, with predominance of collagen III in 7 days, depositing reticulated greenish fibers, characteristic of this type of collagen. On the 14th day, there was both the deposition of type I collagen, with golden and reddish fibers, well interlaced, as well as type III. In 21 days, it was possible to notice the resumption of collagen type III deposition  in  the  three  groups,  mainly NTG and VG, in detriment of type I, as observed on the 7th day. This may be associated with the remodeling process of type I collagen and the synthesis and deposition of type III collagen. This resumption of collagen III deposition at the wound site is probably due to the papillary dermis which also synthesizes it (Isaac et al., 2010).  Among  the  three groups analyzed, in 21 days, 10% HMFMG showed lower deposition of collagen I than the others. This fact is associated with a more accentuated repair process that should have been favored by the constituents of A. cochliacarpa HMF, which, as previously mentioned, caused an increase in the differentiation of myofibroblasts and accelerated the remodeling process with its successive stages of production, digestion and orientation of collagen fibrils (Hatanaka and Curi, 2007).

 

 

Determination of total sulfhydriles (SH)

According to Table 2, the concentration of SH on the 7th day presented a significant difference (ρ<0.05) between 10% HMFMG and NTG being higher in the first, and may be associated with a higher antioxidant effect. However, on the 14th and 21st days, there was no difference between any of the groups analyzed.