Both aqueous and methanolic extracts from C. coccineum were evaluated for their content of phenolics and flavonoids, and for their total antioxidant capacity. The selected ET-based methods were the DPPH assay, ABTS-based assay in order to measure the TEAC (Trolox Equivalents Antioxidant Capacity), and the FRAP method. The selected HAT-based method was the ORAC assay based on pyrogallol red. The full results are summarized in Table 1. Both extracts showed significant antioxidant capacities, with a few differences between them. ORAC-PYR assay gave in both cases the highest antioxidant capacity value, in accordance with the fact that this method is the only HAT-based, being thus able to evaluate overall antioxidant capacity, while FRAP, TEAC, and DPPH only evaluate the presence of reducing compounds (they are in fact electron-transfer based methods). This is in agreement with previously reported studies about the risk that ET-based assays may underestimate total antioxidant capacity [11]. Rached et al. (2010) have measured the antioxidant potential of specimens of C. coccineum grown in Algeria, using two complementary assays, namely inhibition of DPPH radical and β-carotene bleaching [12]. Other authors have studied the antioxidant activity in specimens of C. songaricum, an Asian species of the genus Cynomorium, also widely used in traditional Chinese medicine and known as “suoyang” [13,14].

When data reported in these studies are compared to what we found for the Maltese mushroom grown in Sardinia the results are not similar. In fact, the DPPH-based IC₅₀ values of our extracts are significantly higher, showing a lower antioxidant capacity than Algerian C. coccineum and C. songaricum (4.1 μg/mL and 13.5 μg/mL for Algerian C. coccineum ethanolic and water extracts, and about 35 μg/mL for C. songaricum) [12,13]. On the contrary, the antioxidant capacity measured with FRAP assay appeared to be higher for our specimens (0.051 and 0.167 mmol Fe^II/g for C. songaricum methanolic and water extracts respectively [14]). These conflicting data can be explained by the significant chemical and methodological differences between the two assays. Accordingly, for the full evaluation of the antioxidant capacity of any extracts, the use of a variety of assays is usually recommended [10,15]. The total content of phenolics was similar when comparing methanolic and aqueous extracts, and lower if compared with the content reported in other Cynomorium studies [12,14]. Among the total phenolics, the content of flavonoids represents a significant part, and this value for both extracts showed no significant differences if compared with the total flavonoids content observed in C. coccineum specimens picked in Algeria [12].

HPLC investigation of the fresh Maltese mushroom extracts revealed that the SFE extract had a chromatographic profile similar to that obtained for the methanol-water extraction (not shown). The chromatograms of both extracts showed a peak, identified as gallic acid, and two zones at about 32–33 and 37–39 min retention time of unidentified peaks. The UV-Vis spectra (200–600 nm) of these peaks revealed similarities with procyanidins and gallic acid derivatives, suggesting that they can be procyanidin oligomers. In the methanol-water extract of whole C. coccineum extract, the anthocyan cyanidin 3-O-glucoside was also detected. This is in agreement with that reported by Harborne et al. [16] who found the presence of cyanidin 3-O-glucoside in the red-brown inflorescences of C. coccineum where, like other anthocyans, it plays the role of attracting insect pollinators. Such investigation [16] reported that this is the only anthocyan of this plant, but in our extracts another five anthocyans were detected and quantified (Table 2). Interestingly, the external layer of C. coccineum is the richest in anthocyans, which are responsible for the intense plant color (Figure 1). The presence of considerable amounts of gallic acid and cyanidin 3-O-glucoside in the Maltese mushroom, when compared with other sources [17], is worthy of note. In fact, these are plant polyphenolic compounds naturally present in the human diet. Their use in dietary supplementation is well documented [18] as it can improve the antioxidative potential and nutritional and functional qualities of several foods [19], and is known to reduce the risk of disease [20,21].

Taken together, these data demonstrate that extracts of C. coccineum show a significant total antioxidant potential, and that flavonoids reasonably give a contribution to the overall antioxidant capacity. This is quite interesting, since flavonoids seem to be involved in the anti-carcinogenic and anti-mutagenic activities of plant extracts [22].

To better assess the antioxidant activity observed with in vitro chemical assays, the methanolic extract obtained from C. coccineum was subsequently tested for its protective effect in a biochemical assay of oxidative stress, namely the thermal (140 °C), solvent-free degradation of cholesterol for 1 h or 2 h. The methanolic extract was chosen as the richest in total phenolics (Table 1), and also because, in general, methanolic extracts are more active when studying lipid models. Our model of lipid oxidation has been widely used to assess the antioxidant properties of extracts and natural compounds [23,24]. The consumption of cholesterol and its transformation in its oxidized products (7-keto and 7β-OH) were measured as markers of the oxidative process. At 140 °C, cholesterol was an oil, and more than 80% of the initial compound disappeared within 1 h of heating, with a significant related increase of the oxysterols 7β-OH and 7-keto, as previously observed [23,24].

Figure 2 shows the antioxidant activity obtained in the presence of different amounts (1–50 μg) of extract during cholesterol oxidation for 1 h and 2 h. Antioxidant activity is reported as percentage of cholesterol protection (Figure 2A), calculated considering the percentage of sterol consumption in the presence of the antioxidant with respect to total cholesterol consumption without antioxidant (100% of consumption or 0% of protection). The extract showed a significant inhibitory activity of cholesterol degradation from 2.5 μg at the 1 h time point (40% protection) to 5 μg at the 2 h time point (70% protection, data not shown).

Figure 2B shows the values (μg) of the oxysterols 7-keto and 7β-OH formed during cholesterol oxidation at 1 h in the absence (0) or in the presence of methanolic extract. The extract (5 μg) efficiently prevented the formation of the oxysterols at 1 h.

The scavenging ability of the methanolic extract observed in simple in vitro chemical assays was thus confirmed in this model of lipid oxidation. C. coccineum extract protected sterol against free radical attack and inhibited oxysterol formation, showing scavenging ability against peroxyl radicals ROO• [21,22]. Cholesterol, present in free and in esterified forms in biological membranes, lipoproteins, and food of animal origin, may undergo oxidation when exposed to oxidative stress, and cholesterol oxidation products, oxysterols, exhibit a host of biological activities of relevance for biomedical research (cytotoxicity, angiotoxicity, and mutagenicity) [23,25].

The protective effect of the methanolic extract against the liposome oxidative injury was also evaluated. As an index of the Cu²⁺-induced lipid peroxidation, the variation of the levels of unsaturated fatty acids was analyzed, together with the rise in the level of the oxidation product MDA. Figure 3 shows the values (expressed as μg/mg liposomes) of the main unsaturated fatty acids (18:1, 20:4 n-6, 22:4 n-6, 22:5 n-6, and 22:6 n-3) measured in the control (Ctrl) and following liposome oxidation at 37 °C for 24 h with 5 μM CuSO₄ in the absence (0) and in the presence of different amounts (10, 25, 50 μg/mL) of the extract.

The Cu²⁺-mediated degradation of liposomes resulted in a significant decrease of the unsaturated fatty acids 22:6 n-3, 20:4 n-6, and 22:5 n-6 in the oxidized control (0) (Figure 3A). The consumption of polyunsaturated fatty acids was accompanied by an increase of MDA level in the system (MDA values: 0.46 ± 0.17 and 3.15 ± 0.60 μM in the Ctrl and oxidized control samples, respectively). The addition of the methanolic extract to the system exerted a significant protection, with respect to the oxidized control, against the oxidation of polyunsaturated fatty acids at all the tested concentrations (Figure 3A), and also significantly reduced the formation of the oxidative product MDA (Figure 3B).

From our observations, the methanolic extract significantly preserved liposome fatty acids from oxidative damage, maybe protecting the lipid particles by scavenging oxygen radicals and chelating coppers ions present at the aqueous phase [24]. Liposomes are considered an important membrane model for the study of lipid peroxidation. Much research has been carried out to understand in depth the effects of lipid oxidation in phospholipid membranes to get tools to prevent, modulate and treat the oxidative damage [26]. The oxidative degradation of unsaturated fatty acids, essential components of biological membranes and LDL, is well known to play a role in the development of tissue damage and in a wide range of pathological events [23,25,26].

Unless otherwise specified, all the reagents, standards and solvents were of HPLC graded, purchased from Sigma-Aldrich (Milan, Italy) and used without further purification. CuSO₄·5H₂O was supplied by Carlo Erba (Milan, Italy). Standard of cyanidin 3-O-glucoside was purchased from Extrasynthese (Genay, France). Ultrapure water (18 mΩ) was obtained with a Milli-Q Advantage A10 System apparatus (Millipore, Milan, Italy).

C. coccineum was collected in south western Sardinia (Italy), in April 2012. Reference material for stems (AR-CC-2012/4/1) was deposited in the collection of the Department of Biomedical Sciences, University of Cagliari. The specimens were kept cool and transferred to the laboratory within one hour from harvest. In a typical collection round, the stems (500 g) were gently cleaned to remove earthy residue and then dried with a blotter. The specimens were cut into slices having a thickness of about 0.5 cm and hung up to dry in air flow at 45 °C for 12 h in a food dehydrator (Rommelsbacher ElektroHausgeräte GmbH, Dinkelsbühl, Germany). After drying, the weight of the dried material was about 185 g.

Two steps of extraction were performed on the same plant material. In the first one, dried and ground C. coccineum material was extracted with supercritical CO₂ (SFE) at 40 °C and 250 bars for 4 h. This extraction step, performed in a supercritical laboratory apparatus, produced an apolar fraction (fixed oil). The second extraction step involved the use of a highly polar solvent on the same matrix in order to obtain a polar fraction. To this end, the exhausted C. coccineum powder was subjected to maceration by using methanol or water at ambient temperature in a stoppered container with frequent agitation for a period of 72 h at 4 °C. The extracts were filtered and the supernatants obtained were dried in a rotary vacuum evaporator. For HPLC-DAD analysis, an alternative extraction procedure of C. coccineum was carried out. Fresh material was extracted with a methanol-water mixture (9:1, v/v). Extraction was performed on whole plant, plant without the external colored layer and on the separated colored layer alone (Figure 1b). For the extraction, 3 g of material were put in a plastic vial with 6 mL of methanol-water mixture, finely ground with an Ultra-Turrax^® mixer (International PBI SpA., Milan, Italy) and extracted in an ultrasonic bath for 30 min at 30 °C. The suspension was centrifuged at 6000 rpm, the supernatant was separated and the residue resuspended in 4 mL of methanol and extracted in a sonicator for 30 min. After centrifugation, the supernatant was separated and added to the previous one. The solution was dried in vacuo and the residue was kept at −20 °C in the dark until analysis.

Folin-Ciocalteu reagent was used to determine total soluble phenolics content according to a previously described method [27]. Gallic acid was used as the standard (linearity range 0.05–0.6 mM), and results were calculated as gallic acid equivalents per gram of dry material (mM GAE/g) using a standard curve.

Total flavonoids content was measured by using a previously described method [10]. Catechin was used as the standard (linearity range 0.1–0.6 mM) and results were expressed as Catechin Equivalent per gram of dry material (mM CE/g).

Detection and quantitative analyses of phenolic compounds were carried out using an LC-DAD method slightly modified from Tuberoso et al. [28]. An HPLC Varian system ProStar was employed, fitted with a pump module 230, an autosampler module 410 with a 10 μL loop, and a ThermoSeparation diode array detector SpectroSystem UV 6000lp (ThermoSeparation, San Jose, CA, USA), set at 280, 360 and 520 nm. Separation was obtained with a Gemini C18 column (150 × 4.60 mm, 3 μm, Phenomenex, Casalecchio di Reno, BO, Italy) using 0.2 M phosphoric acid (solvent A), and acetonitrile (solvent B) at a constant flow rate of 0.8 mL/ min. The gradient (v/v) was generated decreasing from 90% of solvent A to 80% in 10 min; to 60% in 20 min; to 40% in 40 min; to 10% in 50 min. Chromatograms and spectra were elaborated with a ChromQuest V. 2.51 data system (ThermoQuest, Rodano, Milan, Italy). Anthocyanins were detected and dosed at 520 nm, flavonols at 360 nm, and all the other compounds at 280 nm. Gallic acid and cyanidin 3-O-glucoside standard solutions were prepared in methanol, and the working standard solutions in ultrapure water. Calibration curves were built with the method of external standard, correlating the area of the peaks with the concentration. The correlation values were 0.9991–0.9999 in the range of 0.2–20 mg/L. Two hundred milligrams of dry extracts were dissolved in 5 mL of the methanol-water mixture (9:1, v/v), diluted with water (1:10–1:50, v/v), and injected in HPLC without any further purification.

DPPH assay was performed as already described [15]. Trolox was used for the calibration curve (linearity range 5–50 μM), and results were calculated as Trolox Equivalents per gram of dry material (mM TE/g), and as IC₅₀ (the extract concentration necessary to decrease DPPH radical concentration by 50%).

FRAP was measured as described [15]. Results were calculated as Trolox Equivalents per gram of dry material (mM TE/g) and mmol of Fe(II) per gram of dry material (mmol Fe^II/g).

The TEAC assay was performed according to Re and coworkers [29]. Trolox was used for the calibration curve (linearity range 0.1–0.8 mM) and the results were expressed as Trolox Equivalents per gram of dry material (mM TE/g), and as IC₅₀.

ORAC-PYR was determined using the pyrogallol red method where APH was the radical releaser [30]. Trolox was used for the calibration curve (linearity range 0.1–0.8 mM), and results were expressed as Trolox Equivalents per gram of dry material (mM TE/g).

The cholesterol oxidation assay was conducted in dry state [23]. Aliquots of 0.5 mL (2586 nmol) of cholesterol solution (2 mg/mL methanol) were dried in a round-bottom test tube in vacuo and then incubated in a bath at 140 °C for 1 or 2 h (oxidized controls) under artificial light exposure. Controls (non-oxidized cholesterol) were kept at 0 °C in the dark. In a different set of experiments, aliquots of methanol extract of C. coccineum (1–50 μg) in methanol solution (1 mg/mL, 100 μg mL) were added to 0.5 mL of cholesterol solution (2 mg/mL methanol), the cholesterol/extract mixtures were dried in vacuo and then incubated in dry state in a bath at 140 °C for 1 or 2 h. During the same time points, non-oxidized control samples of cholesterol were kept at 0 °C in the dark. The oxidation was stopped by cooling and adding 1 mL of methanol. Analyses of cholesterol, 7-ketocholesterol (7-keto), and 7β-hydroxycholesterol (7β-OH) were carried out with a liquid chromatograph (Agilent Technologies 1100, Palo Alto, CA, USA) equipped with a diode array detector (DAD) as previously described [23].

Liposomes were prepared according to the method of Bangham with slight modifications [24]. Aliquots (300 μg) of liposomes in 1 mL of saline solution were incubated for 24 h in the presence of 5 μM CuSO₄ at 37 °C in a thermostatic water bath, exposed to air and artificial light. Controls were kept at 0 °C in the dark. Different amounts of methanolic extract (10, 25, 50 μg), redissolved in methanol solution (10 mg/mL, 5 mg/mL and 2 mg/mL), were added to 300 μg of liposomes in 1 mL of saline solution and the mixtures were incubated for 24 h at 37 °C. The oxidation was stopped by cooling the samples in ice/water. Aliquots of liposome samples (900 μL) were immediately dissolved in ethanol and subjected to mild saponification and unsaturated fatty acid analysis by HPLC was conducted as previously described [24]. Aliquots (100 μL) of liposome samples were used for malondialdehyde (MDA) quantification. MDA levels were measured using the method described by Templar and coworkers [31]. MDA-TBA adduct quantification was conducted by HPLC-DAD analysis as previously reported [24].

GraFit 7 (Erithacus Software, London, UK) [32], and R 2.5.1 software (R Foundation for Statistical Computing, Vienna, Austria) [33], Graph Pad INSTAT software (GraphPad software, San Diego, CA, USA) were used to calculate the means and standard deviations. Evaluation of statistical significance of differences was performed by analysis of variance (One-way ANOVA), using Bonferroni Multiple Comparisons Test.