2.1 General experimental procedures

NMR spectra {¹H, ¹³C, HSQC, HMBC [71 ms (7 Hz)], COSY} were measured in CDCl₃, on a Bruker Avance 300 (300.13 MHz for ¹H and 75.47 MHz for ¹³C) spectrometer or Bruker Avance 500 with cryoprobe, (500.13 MHz for ¹H and 125.76 MHz for ¹³C) and using TMS as internal standard. Chemical shifts were reported in δ units (ppm) and coupling constants (J) in Hz. The MS spectra were obtained using ESI(+) with a Micromass Q-Tof 2™ mass spectrometer (Manchester, UK). The HRMS-ESI were obtained using a MicroTof spectrometer with Apollo II with ESI interface, using a voltage of 4500 V applied to the needle, and a counter voltage between 100 and 150 V applied to the capillary. Column chromatography [CC] was performed with silica gel 60 (Aldrich Chemistry, 63–200 μm). Preparative thin layer chromatography (prep. TLC) was performed on glass plates (20 × 20 cm) precoated with Merck silica gel 60 GF₂₅₄ (0.5 mm thickness) and activated at 100–110 °C for 12 h. Merck silica gel 60 GF₂₅₄ plates were used for TLC. Spots were detected in TLC under UV light (λ 254 and 366 nm). Carbon tetrachloride (CAS Number 56-23-5, 99.9% purity) was purchased from Sigma Aldrich. All other chemicals and solvents were purchased from Panreac and Acros Organics and were of analytical grade or bi-distilled commercial solvents.

GC–MS analysis of triacylglycerol was performed using a Trace gas chromatograph (2000 series) equipped with a Thermo Scientific DSQ II mass spectrometer (Waltham, MA). Separation of the TMS-derivatives resulting from the alkaline hydrolysis of triacylglycerol and silylation of correspondent fatty acid and glycerol, was carried out in a DB-1 J&W capillary column (30 m × 0.32 mm inner diameter, 0.25 μm film thickness) using helium as the carrier gas (35 cm s⁻¹). The chromatographic conditions were as follows: initial temperature, 80 °C for 5 min; temperature rate, 4 °C min⁻¹ up to 285 °C, which was maintained for 10 min.; injector temperature, 250 °C; transfer-line temperature, 290 °C; split ratio, 1:50. The mass spectrometer was operated in the electron impact (EI) mode with energy of 70 eV, and data were collected at a rate of 1 scan s⁻¹ over a range of m/z 33–750. The ion source was kept at 250 °C. Chromatographic peaks were identified by comparing their mass spectra with the equipment mass spectral library (NIST05 MS Library Database).

2.2 Plant collection and extract preparation

Fresh specimens of S. ramosissima J. Woods were collected in the vegetative state, from June to October 2011 at Troncalhada salt pan, Ria de Aveiro (40° 38′ 43.38″ N, 8° 39′ 44.59″ W). A voucher specimen was identified by Helena Silva and deposited in the Herbarium of the Department of Biology, University of Aveiro, Portugal (AVE), under the reference “AVE 6606”. After rinsed with distilled water, samples were stored at −20 °C until further use.

The fresh aerial parts (1.2 kg) of S. ramosissima were chopped into small pieces and extracted with ethanol pro-analysis grade (EtOH, 5 L, three times) at room temperature for 24 h using an overhead stirrer. The mixture was filtered and the ethanol was removed using a rotary vacuum evaporator at 50 °C. The resulting extract was then freeze-dried, yielding 34.7 g of ethanolic extract (SREE).

2.3 Animals and experimental design

All animal experiments were conducted in compliance with national guidelines and the European Directive (2010/63/EU) on the care and use of laboratory animals. Five-week-old male CD-1 mice, supplied by Harlan laboratories (Barcelona, Spain) and weighing 34.61 ± 4.49 g, were housed in polycarbonate cages placed in an acclimatized chamber suitable for small rodents under standard conditions: constant temperature of 22 ± 2 °C, relative humidity of 40–60% and 12 h light/darkness cycle. Mice were fed with regular rodent chow (A04, SAFE diets, France) and water ad libitum, and allowed to acclimate for a week before the experimental study. Animals were randomly divided into four experimental groups of four mice each, as follows: Control mice (water ad libitum); SREE (received orally 50 mg kg⁻¹ b.w./day of SREE for 22 days); CCl₄ [a single subcutaneous injection (250 μL) from the mixture of CCl₄ (0.2 mL kg⁻¹ b.w.) in olive oil on the 22th day]; SREE+CCl₄ (received orally 50 mg kg⁻¹ b.w./day of SREE for 22 days before the CCl₄ administration as described above). The experiment lasted 23 days and the assigned doses of CCl₄ and SREE were based on the previous studies of Irie et al. (2010) and Rathod et al. (2011), respectively.

2.4 Phytochemical characterization

The ethanolic extract of S. ramosissima (SREE) (33.2 g dry weight) was suspended in 500 mL of water and successively partitioned with n-hexane, dichloromethane and ethyl acetate. The hexane fraction (2.2 g) was subjected to CC using a step gradient elution of n-hexane:ethyl acetate (100:0–80:20). The purification of fractions 1, 3 and 5, by prep. TLC [n-hexane:ethyl acetate, 96:4; chloroform/toluene/methanol, 8:1:0.2 and 8:1:0 respectively] yielded, the pure compounds 1 (ethyl linolenoate, 14.7 mg), 3 (octadecyl (E)-ferulate, 1.5 mg) and 2 (sitostanol, 3.1 mg), respectively.

The dichloromethane fraction (18.1 g) was eluted on CC with a gradient system [n-hexane:dichloromethane (1:9), dichloromethane:ethyl acetate (1:0 and 3:7) and ethyl acetate:ethanol (1:1 and 0:1)] yielding 10 fractions (A – J). Fraction A (1.39 g) was rechromatographed on CC eluted with n-hexane:chloroform (from 100:0 to 0:100%]) to afford the pure compound saliramoester (8) (92 mg) from subfraction A₅, while the ethyl (E)-2-hydroxycinnamate (5) (2 mg) was obtained from subfraction A₁₄ after prep. TLC purification [n-hexane:ethyl acetate (85:15)]. Fraction C (1.1 g) was chromatographed on CC eluted with n-hexane:ethyl acetate (100:0–50:50). Subfraction C₄ (315 mg) was further purified by prep. TLC [n-hexane:ethyl acetate (9:1) twice] to give octadecyl (E)-ferulate (3) (4 mg) and eicosanyl (E)-ferulates (4) (3 mg), while ethyl (E)-2-hydroxycinnamate (5) (4 mg) was also isolated from subfraction C₅ after purification by prep. TLC [n-hexane:ethyl acetate (9:1), twice]. Fraction F (6.4 g) was chromatographed on CC [n-hexane:ethyl acetate, (100:0–0:100)]. The subfractions F₁ and F₂ (207 mg) were purified by prep. TLC [n-hexane:ethyl acetate, (8:2)] to yield a triacylglycerol (7) (112 mg), while F₅ (552 mg) was rechromatographed on prep. TLC (dichloromethane, twice) to yield octadecyl (E)-ferulate (3) (22 mg) and eicosanyl (E)-ferulates (4) (18 mg). The scopoletin (6) (3 mg) was obtained from the fraction G after being rechromatographed on CC [dichloromethane:ethyl acetate, (100:0–70:30)], followed by prep. TLC [n-hexane:ethyl acetate (1:1)].

The ethyl acetate fraction (0.3 g) was chromatographed on prep. TLC [n-hexane:ethyl acetate (1:2)] to yield saliramophenone (9) (7 mg), and saliramopyrrole (10) (6 mg).

In order to clarify the structure of triacylglycerol (7) the compound was subjected to GC–MS analysis after alkaline hydrolysis and silylation according to published methodology (Isca et al., 2014b). Briefly, about 20 mg of the compound 7 was dissolved in 10 mL of 1 mol dm⁻³ KOH in 10% aqueous methanol. The mixture was heated at 100 °C, under a nitrogen atmosphere, for 1 h. The reactional mixture was cooled, acidified with 1 mol dm⁻³ HCl (final pH lower than 3), and extracted three times with dichloromethane (3 × 10 mL). After dichloromethane evaporation, the hydrolysed compounds were dissolved in 250 μL of pyridine and the hydroxyl groups silylated by addition of 250 μL of BSTFA, 50 μL of TMSCl, and heating the mixture at 70 °C for 30 min.. Thereafter, the reactional mixture was injected into the GC–MS apparatus.

2.5 Spectroscopic data for compounds 1–10

3.1 Mice behaviour and biometrical parameters

The toxicity model of carbon tetrachloride (CCl₄) is widely used to assess the beneficial effects of plant extracts against xenobiotic-induced oxidative stress, which has been recently investigated in the male reproductive system (Khan, 2012; Yüce et al., 2014).

No significant changes in behaviour and physical appearance from control and SREE groups were observed. In contrast, all mice exposed to CCl₄ (CCl₄ and SREE+CCl₄ groups) presented, 12–24 h after the CCl₄ dose, lusterless fur and piloerection, less activity and weak response to handling. The results of the gonadal biometric measurements (Table 1) showed that, statistically, no significant effect (p > 0.05) were observed on testicular weight, volume and gonadosomatic index among groups treated either with CCl₄ or SREE. Although CCl₄ intoxication is able to impair testosterone levels (Khan, 2012), apparently on a macroscopic analysis, it was not sufficient to affect testicular growth in our experimental conditions.

3.2 Histology and morphometry

The results of the histopathological examination of the testis (Fig. 1) revealed that in the control and SREE groups the percentage of tubules affected was low (lesion score minimal or slight) and not statistically significant, as also shown in histological sections (Fig. 2-A and -B, respectively). The lesion score in CCl₄ group increased significantly to the marked level when compared with control and SREE groups (Fig. 1), since tubular degeneration, such as architectural disorder of seminiferous epithelium with germ cell dissociation and sloughing into the lumen, was evident (Fig. 2-C). These findings are in agreement with previous acute studies in the rat (Khan, 2012). The degree of the described lesions was significantly reduced (p < 0.05) in SREE+CCl₄ group when compared with CCl₄ group (Figs. 1 and 2-D), where SREE pre-treatment prevented approximately 45% and 21% of the CCl₄-induced epithelium disruption and germ cell exfoliation, respectively. Concerning the morphometric analysis (Table 2), we concluded that testicular histomorphology is not affected by SREE oral treatment or toxic exposure to CCl₄ since no statistical significant effect (p > 0.05) was observed in the volume, diameter, area and length of seminiferous tubules from SREE or CCl₄-treated animals, when comparing all the groups. In this regard, the protective action of SREE could only be evidenced at the histopathological level.

Close

Figure 1

Effects of S. ramosissima ethanolic extract (SREE) on histopathological lesions in CCl₄-induced acute testicular injury. Frequency of affected tubules is expressed as mean ± SEM. Significantly different (p < 0.05) when compared to: a = control group; b = SREE treated group; c = CCl₄ treated group. Lesions were rated as 1: minimal, 2: slight, 3: moderate, 4: marked and 5: severe (Lanning et al., 2002).

Export to PPT

Close

Figure 2

Testicular sections (400×) of mice treated with SREE and CCl₄. Control (A) and SREE (B) groups with normal histology. SE = seminiferous epithelium, I = interstitial space, L = lumen. CCl₄-treated mice (C) presented epithelium disruption (asterisk) and germ cell exfoliation (arrow). These findings were less noted in SREE+CCl₄ group (D). Bar = 50 μm.

Export to PPT

3.3 Structural characterization of secondary metabolites isolated from SREE

In order to identify the secondary metabolites potentially responsible for the near to normal histology observed in SREE-treated groups, the SREE was fractionated and each fraction was purified by preparative chromatographic techniques according to the experimental section, affording ten compounds (Fig. 3).

Close

Figure 3

Chemical structures of compounds 1-10 isolated from S. ramosissima aerial parts.

Export to PPT

Compounds 1–7 proposed structures were identified by exhaustive analysis of their 1D and 2D NMR spectra (¹H, ¹³C, DEPT, COSY, NOESY, HSQC e HMBC) and MS data (see spectroscopic data in Material and Methods section) as well as comparing with literature data.

Compound 1 was identified as ethyl linolenoate (ω3 PUFA) mainly based on: (a) the ¹³C NMR signal at δ_C 173.9 ppm and at δ_C 127.1–131.9 ppm characteristic, respectively, of the carbon on the ester group and double bonds; (b) the ¹H NMR signals at δ_H 1.25 (t, J 7.2 Hz) and δ_H 4.12 ppm (q, J 7.2 Hz) indicating the presence of an ethyl group bound to an electronegative atom like oxygen; (c) the ESI-MS signal at m/z 307 corresponding to the [M+H]⁺ ion, showing a C-18 fatty acid. The structure of compound 1 was unequivocally confirmed by the analysis of 2D NMR spectra (COSY, HSQC and HMBC). The ethyl group cannot be an artefact since the SREE was obtained at room temperature. On the other hand, both ethyl and methyl ester forms were reported to occur in the nature (Herrera-Valencia et al., 2012; Huh et al., 2010). Linolenic acid is widely distributed in nature and was previously identified on Salicornia sp. (Lu et al., 2010; Isca et al., 2014b); however, its ethyl derivative was here for the first time identified in Salicornia species.

In the ¹³C NMR spectrum of compound 2, all the signals (with one exception) appear below 56.4 ppm, highlighting its strong aliphatic character. The exception is the signal at δ_C 71.4 ppm suggesting the presence of one CHOH group. The DEPT NMR spectrum allowed inferring the existence of six CH₃, twelve CH₂, nine CH groups and two quaternary carbons. The signal in the MS spectrum at m/z 399, corresponds to the [M-18+H]⁺ ion, suggesting a compound with the molecular formula C₂₉H₅₂O. These data and the analysis of the 2D NMR spectra allowed the confirmation that compound 2 is sitostanol also known as stigmastanol. Compound 2 is recognized for its serum LDL-cholesterol lowering effects (Ramjiganesh et al., 2001) and was previously identified on Salicornia species (Isca et al., 2014b; Salt and Adler, 1985).

In the compound 3 a (E)-ferulic nucleus was clearly identified by: (a) the signals at δ_H 6.30 and 7.61 ppm (d, J 15.9 Hz) characteristic of (E)-vinylic system; (b) the signals at δ_H 6.92 (d, J 8.1 Hz), 7.04 (d, J 1.8 Hz) and at 7.07 ppm (dd, J 1.8, 8.1 Hz) typical of a tri-substituted aromatic ring; (c) signal at δ_H 3.93 ppm (s, δ_C 55.9 ppm) with connectivity with the signal at δ_c 146.7 ppm (HMBC spectrum) showing an aromatic methoxy group; (d) signal at δ_H 5.85 ppm without proton-carbon correlation on HSQC spectrum, typical of non-label hydroxy proton. The presence of aliphatic chain esterified with the ferulic acid can be deduced from: (a) ¹³C NMR signal at δ_C 167.4 ppm characteristic of the carbon on an ester group; (b) the intense methylene signals at δ_H 1.25–1.68 ppm besides the triplet at δ_H 0.87 ppm (J 6.8 Hz) characteristic of aliphatic methylic protons. The ESI-MS signal at m/z 447 attributed to [M+H]⁺ ion suggesting a molecular formula of C₂₈H₄₆O₄. The complete analysis of the 2D NMR spectra confirmed the structure of compound 3 as octadecyl ferulate.

The ¹H and ¹³C NMR data of compounds 4 and 3 are identical. However, the ESI-MS spectra of compound 4 showed a peak at m/z 475 [M+H]⁺, corresponding to more 28 units (two CH₂) than compound 3. Thus, compound 4 was identified as eicosanyl (E)-ferulate. Although compounds 3 and 4 were previously found in nature (Serra et al., 2010; Kosma et al., 2012), they are described here for the first time in Salicornia genus, even if other alkyl ferulate and its antioxidant activity were recently reported in Salicornia herbacea (Wang et al., 2013).

Compound 5 was obtained as a yellow oil and the combination of ¹H and ¹³C NMR data showed the presence of an ethyl group esterified with an ortho-substituted cinnamic acid: (a) the presence of CO₂Et group was deduce as described above to compound 1; (b) an ortho-substituted aromatic ring deduced from ¹H NMR signals at δ_H 7.29, 7.31, 7.42 and 7.62 ppm, exhibiting HSQC correlations with δ_C 127.1, 131.0, 130.2 and 127.6 ppm, respectively; (c) a trans-vinylic system (δ_H 6.43 and 8.09 ppm, d, J 16.0 Hz; δ_C 120.9 and 140.4 ppm); (d) the peak at m/z 215 on ESI-MS spectrum corresponding to [M+Na]⁺ ion and showing a molecular mass of 192 u.m.a.. These data allowed to identify the compound 5 as ethyl (E)-2-hydroxycinnamate, a cinnamic acid derivative well-known for its high anticancer potential (De et al., 2011) and not previously identified in Salicornia species.

The exhaustive analysis of all spectroscopic data from compound 6 and its comparison with literature data (Siddiqui et al., 2007; Darmawan et al., 2012) allowed identify it as scopoletin, a well-known coumarin previously identified in Salicornia herbacea (Wang et al., 2013) with proved in vitro and in vivo biological properties (Gnonlonfin et al., 2012), including antioxidant activity (Wang et al., 2013). The assignment of all proton and carbon resonances (see Section 2.5) and their agreement with published data (Siddiqui et al., 2007; Darmawan et al., 2012) showed that several assignments made by Wang et al. (2013) must be interchanged (e.g. C-8, C-4, C-3).

The ¹H, ¹³C and DEPT NMR spectra of compound 7 exhibit the typical signals of the glycerol moiety: (a) signals at δ_H 4.15 and 4.31 ppm (dd, J 3.8; 11.8 Hz), with proton-carbon correlation on HSQC spectra, with δ_C 62.1 ppm assigned to OCH₂ groups; (b) a signal at δ_H 5.26 (multiplet) coupling with the previous two proton signals and with proton-carbon correlation on HSQC spectra with δ_C 68.9 ppm (OCH group). The presence of saturated fatty acids esterified was deduced from the spectroscopic evidences: (a) signals at δ_C 172.9 and 173.3 ppm whose shifts are typical of the carbon of ester groups; (b) an intense signal at δ_H 1.25 nearly as singlet, and a triplet at δ_H 0.88 ppm assigned to the aliphatic chain. Tetracosanoic acid and glycerol TMS derivatives were identified as the products of compound 7 hydrolysis (see Section 2.4). All these data corroborated the identification of compound 7 as triacylglycerol of the tetracosanoic acid. Several triacylglycerides were identified in seed oil of hybrid species Salicornia SOS-7 (El-Mallah et al., 1994) but none of them is the triacylglycerol of the tetracosanoic acid.

Additionally, three new compounds were isolated and structurally characterized: saliramoester, a long chain triester (8), saliramophenone, a propiophenone derivative (9) and saliramopyrrole a pyrrole-3-carbaldehyde derivative (10).

The ¹H, ¹³C and DEPT NMR spectra of compound 8 showed its strong aliphatic character with the presence of large number of CH₂ groups and several CH₃ groups (δ_H 2.30–0.84 ppm; δ_C 39.8–16.4 ppm) with two OCH₂ and one OCH groups (δ_H 4.89–4.05 ppm; δ_C 70.7–61.2 ppm), three ester groups (δ_C 173.59, 173.95 and 174.03 ppm) and a trisubstituted vinylic system (δ_H 5.33 ppm, δ_C 118.1 and 142.6 ppm) with a vicinal methyl group [δ_H 1.69 ppm (3H, s); δ_C 16.4 ppm] and a vicinal CH₂O group [δ_H 4.59 ppm (d, J 7.1 Hz); δc 61.2 ppm]. The relative position of these substituents was confirmed by the connectivities showed on Fig 4a and b. The NOESY correlations (Fig 4c) between methylic protons at δ_H 1.69 ppm and the oxymethylenic protons at δ_H 4.59 ppm, besides the NOE effect between the vinylic proton and the methylenic protons at δ_H 2.00 ppm showed a vinylic system with (E)-configuration. The signal at δ_H 4.89 ppm (1H, m) exhibits on HSQC spectrum correlation with the oxymethynic carbon at δ_C 70.7 ppm, COSY coupling with the methylic protons at δ_H 1.19 ppm (d, J 6.6 Hz) and connectivity on HMBC spectrum with the carbonyl carbon on a ester group at δ_C 173.6 ppm. These data and the connectivities showed on Fig. 4d, allowed to elucidating other part of structure 8 (Fig. 4). Another portion of the molecule includes the second oxymethylenic group (δ_C 64.4 ppm) and the third ester group (δ_C 174.03 ppm) bonded to CH₂ groups as proved by the connectivities showed on Fig. 4e. Moreover, the signals at δ_H 0.84 and 0.85 ppm (d, J 6.6 Hz) are correlated with two equivalent methylic carbons at δ_C 19.7 ppm, and both exhibit connectivities with methynic and methylenic carbons at δ_C 32.8 and 37.4 ppm respectively (Fig. 4f) showing the presence of a terminal i-propyl group. Identical spectroscopic evidences (Fig. 4g) showed the presence of a second terminal i-propyl group. The most functionalized structural elements are thus identified, and these are connected to each other by the chains of CH₂ groups, the length was established by MALDI TOF MS showing the peak at m/z 622 relative to [M]⁺ and by MALDI TOF MS/MS (+) showing the main peaks at m/z 172, 200 and 340 and assigned to the fragments [C₁₀H₂₀O₂]⁺, [C₁₂H₂₄O₂]⁺ and [C₂₀H₃₆O₄]⁺, respectively, resulting of the fragmentation pattern point out on Fig. 5. Thus, the molecular formula of compound 8 was established as C₃₈H₇₀O₆. All the described spectroscopic data allowed to establish the structure of compound 8 as (E)-6-{[4-methyl-6-((4-methylpentanoyl)oxy)hex-4-enoyl]oxy}heptyl6-methylheptanoate, a long chain triester which we have named saliramoester.

Close

Figure 4

Main HMBC connectivities (→) (a, b, d–g), and NOESY (↔) correlations (c) observed for compound saliramoester (8).

Export to PPT

Close

Figure 5

Ions observed in TOF MS MALDI spectrum of saliramoester (8).

Export to PPT

The ¹H NMR spectrum of compound 9 showed two signals at δ_H 6.89 and 7.90 ppm (d, J 8.5 Hz) with proportional integrals of two protons each, coupling with each other and exhibiting correlations with signals at δ_C 115.5 and 130.7 ppm, which are characteristic of protonated carbons on a para-substituted aromatic ring. The triplets at δ_H 4.02 and 3.17 ppm coupling each other (J 5.2 Hz) (Fig. 6 double arrow), exhibit HSQC correlations with the carbons at δ_C 58.0 and 39.0 ppm, respectively, are assigned to two vicinal CH₂ groups. The ¹³C NMR spectrum exhibits signals at δ_C 198.9, 160.5 and 58.0 ppm assigned to ketone carbon, non-protonated aromatic carbon and a methylenic carbon, being the last two bonded to electronegative atoms like oxygen. The connectivities shown in Fig. 6 (single arrow) confirm the relative position of these structural elements and showing a propiophenone skeleton which [M+H]⁺ should be around 189. The absence of other signals in the ¹H NMR spectrum and the peaks at m/z 337 and 353 on ESI-MS spectrum ([M+Na]⁺ and [M+K]⁺ respectively) suggests that compound 9 is not a single propiophenone moiety but rather two propiophenone skeletons bonded by an oxygen atom at C-3 or C-4′ resulting a compound with symmetry (thus only half of the protons and carbons were required to be assigned based on the NMR analysis). The connectivity between the protons at δ_H 7.90 ppm and the carbon at δ_C 160.5 ppm observed on HMBC spectrum optimized for long range J_H/C of 4 Hz, showed that compound 9 is a propiophenone dimer possessing a 4″-O- $4^{‴}$ ether bond identified as 1,1′-[oxybis(4,1-phenylene)]bis(3-hydroxypropan-1-one, named saliramophenone.

Close

Figure 6

Coupling deduced by COSY spectrum (↔) and main HMBC (→) connectivities observed for compound saliramophenone (9).

Export to PPT

The ¹H and ¹³C NMR data suggest that compound 10 is a small secondary metabolite. In fact, only four non-equivalent protons and six carbon NMR signals were observed in the respective spectra. The proton signal at δ_H 9.61 ppm as singlet exhibits HSQC correlation with the carbon signal at δ_C 177.9 ppm indicating the presence of an aldehyde group, while the two doublets at δ_H 7.22 and 6.53 ppm (J 3.5 Hz, 1H each) exhibit HSQC correlation with the signals at δ_C 122.7 and 110.0 ppm respectively, suggesting the presence of two tri-substituted vinylic systems. The most shielded proton appears at δ_H 4.73 ppm also as singlet (the integral is proportional to 2H). It exhibits proton-carbon correlation with the signal at δ_C 57.8 ppm suggesting the presence of oxymethylenic group. All these spectroscopic evidences among the connectivities observed on the HMBC spectra (Fig. 7) allowed to establishing the relative position of structural elements discussed above and conclude that the compound 10 is a pyrrole-3-carbaldehyde derivative. The ESI-MS spectrum showed a peak at m/z 141 corresponding to [M]⁺ and confirmed the proposed structure of compound 10 identified as 5-hydroxy-1-(hydroxymethyl)-1H-pyrrole-3-carbaldehyde named saliramopyrrole.

Close

Figure 7

Main HMBC connectivities observed for compound saliramopyrrole (10).

Export to PPT

To our best knowledge, the compounds 8–10 are here reported for the first time in nature.

The structural diversity observed on the isolated compounds 1–10 in addition to the completely new skeleton described to the saliramophenol (Isca et al., 2015) suggests that S. ramosissima can be exceptionally prolific in producing structurally diverse secondary metabolites, some of them being strong candidates responsible for the testicular preservation observed in the current study. For instance, α-linolenic acid, the core of compound 1 available in vivo after enzymatic hydrolysis, is the substrate for the production of docosahexaenoic acid, which is important to maintain the spermatogenesis function within the testicular environment through incorporation on germ cell and spermatozoa membranes (Roqueta-Rivera et al., 2010). Octadecyl (3) and eicosanyl (4) (E)-ferulates are alkyl ester derivatives of ferulic acid, a phenolic compound with recognized antioxidant properties through free radical scavenging, including the inhibition of both the acute and subchronic toxicity of secondary free radicals generated by CCl₄ (Kim et al., 2011; Srinivasan et al., 2005), and approved in some countries as an additive to prevent oxidation in foods (Graf, 1992). Alkyl ferulates demonstrated to exert higher antioxidant activity than the ferulic acid (Anselmi et al., 2004; Kikuzaki et al., 2002) and can also be enhanced when combined with different phytochemicals (Wang et al., 2013).

Although little is known about the effects of several of the isolated compounds upon the male reproductive system and further studies will be needed to elucidate such potential, the antioxidant potential of other isolated compounds is well known and may contribute to the prevention of ROS-mediated stress, enhancing the SREE antioxidant effect in vivo.