Moringa oleifera Lam. leaves and seeds were collected from a Cameroonian plantation and sampled by a ‘traditional healer’. Leaves (L) and seeds (S), dried in the sun, were collected from the same trees, and they were used in traditional African medicine preparations. Moreover, a non-conventional preparation was prepared using fresh seeds (FS).

M. oleifera extracts (MOE) were prepared according to African traditional methods: L, S and FS were ground with a mortar and pestle and boiled for 15 min in bidistilled water. Other types of extracts were obtained from the same MO plant parts, but they were powdered, resuspended in bidistilled water and frozen for 24 h at −80°C. All extracts were centrifuged, and the supernatants were recovered, filtered (0.22 µm) and stored at −20°C. The extracts obtained from the boiling and freezing procedures are, from now on, referred to as boiled and frozen extracts, respectively.

Human Jurkat E6-1 lymphoid and THP1 monocytoid cell lines (American Type Culture Collection, Manassas, VA, USA) were grown in a suspension culture at a density of 7×10⁵ cells/ml.

Human PBMCs were obtained from 17 healthy blood donors attending the local Blood Transfusion Unit of Policlinico ‘Tor Vergata’ in Rome. PBMCs were separated by density gradient from the buffy coat according to the Ficoll-Hypaque standard technique (Lonza, Morristown, NJ, USA) and were cultured at a density of 10⁶ cells/ml. Tumour cell lines, Jurkat and THP1, and PBMCs from healthy donors were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc.), 2 mM glutamine (HyClone; GE Healthcare Life Sciences, Logan, UT, USA), 50 U/ml penicillin and 50 U/ml streptomycin (HyClone; GE Healthcare Life Sciences). All cell lines and PBMCs were cultured at 37°C in a 5% CO₂ humidified atmosphere in the presence or absence of the various MO extracts (MOE) at different concentrations, which varied between 0.01 and 10 mg of fresh plant weight equivalent (FW) per ml of culture medium, for 72 h. The ethical approval for the collection and use of human samples was obtained in 2014, from ethical board of ‘Tor Vergata’ hospital, protocol number 15/14 (D.M.08.02.2013-D.G.R.146/2013; D.D.G.467 del 25.07.2013), all patients provided written informed consent.

Amounts of viable and dead cells were assessed by the trypan blue (Euroclone S.p.A., Milan, Italy) exclusion test, and a negative (water) proliferation control was used. After 72 h, viable and dead cells, treated and untreated, were calculated as a fold change with respect to untreated cells at Time 0.

Cumulative results from at least three different measurements were used to calculate the MOE concentration required to reduce cell proliferation by 50% (EC50) or to induce death in 50% of cells (LD50). EC50 and LD50 were evaluated in all cell lines and in PBMCs using sigmoidal dose-response regression curves, using GRAPH PAD PRISM software.

BCL2 and SIRT1 intracellular expression was evaluated by flow cytometry analysis. After 72 h, transfected cells were harvested, fixed and permeabilized with 70% ethanol and incubated with PE-conjugated anti-human BCL2 (BD Biosciences, Franklin Lakes, NJ, USA), rabbit anti-human SIRT1 (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and anti-rabbit IgG-PE antibodies (Calbiochem; Merck KGaA, Darmstadt, Germany). Stained cells were analysed using a CytoFLEX flow cytometer (Beckman Coulter, Inc., Brea, CA, USA) and CytExpert 1.2 software (Beckman Coulter, Inc.).

After 72 h of incubation with MOE, apoptosis was evaluated by flow cytometry (CytoFLEX; Beckman Coulter, Inc.) assessment of hypodiploid events. Cells were harvested, washed twice in PBS and incubated for 20 min in 70% EtOH at −20°C. After incubation, cells were washed and stained in propidium iodide 1.25 µg/ml. Events were gated on forward vs. orthogonal scatter in such a way that degraded DNA from cell debris or from doublets was excluded and nuclei from viable, apoptotic and necrotic cells were assayed. Data acquisition and analyses were performed for 150,000 events for each sample and analysed using CytExpert 2.0 (Beckman Coulter, Inc.).

The percentage of DNA synthesis in the cells was detected by the bromodeoxyuridine (BrdU) In Situ Cell Proliferation kit (FLUOS, Roche), according to the manufacturer's instructions, and analysed using CytoFLEX (Beckman Coulter, Inc.).

M. oleifera extracts were analysed by a Shimadzu HPLC (Shimadzu Corporation, Kyoto, Japan) equipped with a diode array detector (DAD) SPD-M20A (Shimadzu Corporation). Analytes were separated using a Kynetex C18 column 2.6 µm × 75×2.1 mm (Phenomenex, Torrance, CA, USA). Data acquisition and peak integration were performed with a LabSolutions software (Shimadzu Corporation). The mobile phases used were 1% trifluoroacetic acid at pH 2.5 (solvent A) and acetonitrile (solvent B). The system was run with the following gradient elution program: 0 min, 85% solvent A; 2.5 min 65% solvent A; 7 min, 25% solvent A; 9 min, 25% solvent A; 13.5 min, 85% solvent A; and 25 min, 85% solvent A. The flow rate was kept constant throughout the analysis at 1 ml/min⁻¹, and the injection volume was 20 µl. Chromatographic profiles were obtained at 254 nm.

The amount of total simple phenols and flavonoids in the plant samples was measured using a method that has been widely reported (27,28). Briefly, for total simple phenols quantitation, the equivalent of 400 µg of plant extract was mixed with 100 µl of Folin-Ciocalteu reagent and 80 µl of 0.7 M Na₂CO₃, incubated for 1 h in the dark and spectrophotometrically analysed at 760 nm. For the measurement of flavonoid levels, the equivalent of 400 µg of plant extract was adjusted with ddH₂O to reach 250 µl of final volume, which was added to 30 µl of pure methanol, 2 µl of 10% AlCl₃ and 2 µl of 1 M CH₃CO₂K. The sample was left in the dark for 15 min and absorbance was read at 415 nm. The concentration of total phenols and flavonoids was obtained by comparing the sample absorbance values with calibration curves properly obtained using adequate amounts of pure gallic acid (GA) and quercetin (Q) as standards. Consequently, the results were reported as µg of GA or Q equivalents per 100 mg of sample fresh weight (µg GAE or QE/100 mg SFW).

The total small RNA pool of MO (mol-small RNA pool) was extracted from 50 mg FW of MOE seeds using NucleoSpin miRNA in accordance with the manufacturer's instructions (NucleoSpin miRNA experiments protocols^®; MACHEREY-NAGEL GmbH & Co. KG, Düren, Germany).

The presence of microRNA purified, as reported above, from the aqueous extract of MOE-S (mol-miR) was evaluated by RT-qPCR, as has been widely reported (29,30). In brief, cDNA was synthesized using a specific reverse transcription kit for microRNA (miRCURY LNA Universal RT microRNA PCR, Synthesis kit II; EXIQON; Qiagen GmbH, Hilden, Germany), according to the manufacturer's guidelines. To verify the absence of nucleases in the reaction and evaluate the efficiency of retro-transcription and qPCR amplification, 10⁸ copies of a synthetic spike-in control miRNA (UniSp6; EXIQON; Qiagen GmbH) were added to each RNA sample before conversion to cDNA. qPCR was carried out in a 10 µl reaction volume that included 20 ng cDNA, 50% SYBR green (ExiLENT SYBR^® Green master mix; EXIQON; Qiagen GmbH) and 1 µl of a mixture containing pre-designed PCR primers specific for microRNA amplification (microRNA LNA PCR primer sets; EXIQON; Qiagen GmbH). The qPCR assay was performed using a Bio-Rad (IQ5) thermal cycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Amplification parameters were set as recommended by the instruction manual of the EXIQON pre-designed primers (Table I).

Quantification of mol-miRs was performed using the threshold cycle (Ct) comparative method according to the MIQE guidelines (29) comparing the mol-miRs expression respect to the lower expressed miR (mol-miR858b). 5S rRNA was used as a housekeeping gene.

All data were presented as the mean ± SD of at least three independent experiments performed on THP1 and Jurkat cells and PMBCs from 10 healthy donors. Data analyses were performed using the SPSS statistical software system (v.17.0; SPSS, Inc., Chicago, IL, USA).

Comparison between treated vs. untreated cells for trypan blue assay, apoptosis assay, BCL2 and SIRT1 intracellular protein expression were all conducted using t-test. A comparison of trypan blue apoptosis and BrdU assays results, in response to the different concentration of extracts, was carried out using an ANOVA and a Bonferroni significant difference test as a multiple comparison test. *P<0.05, **P<0.01 and ***P<0.001 were considered to indicate a statistically significant difference. For non-parametric correlations, a Pearson correlation coefficient was calculated.

To obtain information about MO's effect on cell viability, Jurkat cells, THP1 cells and PBMCs were treated with extracts obtained from the leaves (L), seeds (S) and fresh seeds (FS) of MO, after being frozen (MOE-f) or boiled in water (MOE-b).

THP1 and Jurkat cells and PBMCs from healthy donors were treated with frozen or boiled MOE in a concentration ranging from 0 to 10 mg/ml.

Seventy-two hours after treatment, cell viability was analysed using a trypan blue assay.

For each type of preparation, the MOE concentration that was able to reduce cell proliferation by 50% (EC50) or induce death in 50% of cells (LD50), with respect to the number of cells at time 0, was estimated.

After 72 h, the treatment of THP1 (Fig. 1A, D and G) and Jurkat (Fig. 1B, E and H) cells with boiled extracts of the three different parts of the MO plant induced a significant reduction of proliferation, with respect to untreated cells, in a dose dependent manner, starting from 0.01 mg/ml (P<0001 for each concentration vs. untreated cells; Fig. 1).

Jurkat and THP1 cells showed a different susceptibility to MOE treatments: Although MOE-b reduced cell growth in both tumour cells at a low concentration, it showed a lower toxicity in THP1 cells with respect to Jurkat cells. Indeed, THP1 cells demonstrated a significant increase in trypan blue positive cells at a concentration of 1 mg/ml for leaves (Fig. 1A), 2.5 mg/ml for fresh seeds (Fig. 1D) and 5 mg/ml for seeds (Fig. 1G), whereas a cytotoxic effect was evident in Jurkat cells starting from 0.25 mg/ml of leaves (Fig. 1B), 0.01 mg/ml of fresh seeds (Fig. 1E), and 0.1 mg/ml of seeds (Fig. 1H).

MOE treatments at a low concentration preserved viability in PBMCs from healthy donors; a decrease of cell proliferation associated with a significant increase of trypan blue positive cells was detectable at the concentration of 0.25 mg/ml for leaves (Fig. 1C) and 5 mg/ml for all seed preparations (Fig. 1F-I).

Based on these results, EC50 at 72 h of MOE treatment was calculated compared to control cells.

As shown in Table II, THP1 cells were the most sensitive to treatments with the different boiled parts of M. oleifera. In fact, for all boiled preparations, a concentration between 0.014 and 0.020 mg/ml was enough to reduce proliferation of THP1 cells by 50%, whereas for Jurkat cells, concentrations of 0.129, 0.121 and 0.035 mg/ml for leaves, seeds and fresh seeds, respectively, were necessary to reduce proliferation by 50%. PBMCs appeared to be less sensitive to the treatments: Concentrations of 1.94 mg/ml of leaves and of more than 10 mg/ml of all seed extracts were required to reach the EC50 value.

The different MO tissue boiled extracts induced the death of 50% of cells (LD50) more effectively in Jurkat cells with respect to THP1 cells and PBMCs. In detail, LD50 in THP1 cells and PBMCs was obtained for an MOE concentration of higher than 10 mg/ml, while a concentration between 2–4 mg/ml was sufficient in Jurkat cells (Table III).

The treatment of THP1 and Jurkat cells with frozen extracts of the three different parts of the MO plant induced a significant reduction of proliferation, in a dose dependent manner, starting from 0.01 mg/ml (P<0001 for each concentration vs. untreated cells; Fig. 2). The treatment became cytotoxic with a different intensity in THP1 cells with respect to Jurkat cells: For THP1 cells starting from 5 mg/ml for leaves, 2.5 mg/ml for seed samples (Fig. 2A and G), and from 1 mg/ml for fresh seed preparations (Fig. 2D) and in Jurkat cells starting at 0.1 mg/ml for all preparations (Fig. 2B, E, H).

A reduction of viable PBMCs was observed when they were exposed to 2.5–10 mg/ml of leaves and fresh seeds (Fig. 2C and F) and 0.25–10 mg/ml of the seed preparation (Fig. 2I). In PBMCs, the cytotoxic effect of frozen MOE appeared at high concentration (5–10 mg/ml) for the treatment with L and FS (Fig. 2C and F), while S extract promoted the development of trypan blue positive cells starting from 0.1 mg/ml (Fig. 2I).

EC50 values were also measured for the treatments using frozen preparations of leaves, fresh seeds and seeds of MOE (Table IV).

EC50 of frozen MOE in THP1 and Jurkat cells is within a concentration range of 0.001–0.019 mg/ml. As well as for boiled extracts, PBMCs were less sensitive to the treatments. Indeed, for these cells, higher concentrations of extracts, with respect to tumour cell lines, were necessary to reduce cell proliferation by 50% (in detail, 5.195 mg/ml for L, 2.895 mg/ml for FS and 1.065 mg/ml for S).

Frozen MOE also showed a higher LD50 value in Jurkat cells than in THP1 cells and PBMCs (Table V).

In THP1 cells and PBMCs, a concentration higher than 10 mg/ml was necessary to reach LD50, while in Jurkat cells, a dose between 0.1–0.8 mg/ml was sufficient to induce death in 50% of cells.

Taken together, these data suggest that Jurkat and THP1 cells are more sensitive than PBMCs to treatment with MO aqueous extracts obtained by boiling, as well as freezing, in terms of cell proliferation.

Moreover, Jurkat cells were the most sensitive in terms of cytotoxic effect, compared to THP1 cells and PBMCs. In particular, in Jurkat cells, MOE-f was more toxic than MOE-b; indeed, a 10-fold lower concentration of all types of plant parts, in MOE-f with respect to MOE-b, was sufficient to reach the LD50 value.

To evaluate the effects of M. oleifera aqueous extracts on apoptosis, PBMCs from healthy donors and THP1 and Jurkat cells were treated with boiled and frozen aqueous preparations of different plant parts in a concentration ranging from 0 to 10 mg/ml. After 72 h, the apoptotic level was measured by assessing the percentage of hypodiploid nuclei, through propidium iodide (PI) staining and flow cytometry analysis.

The results obtained after cells were treated with MO boiled extracts are shown in Fig. 3. Among MOE-b, treatment with the L preparation induced a high level of apoptosis in both cell lines and PBMCs. The treatment increased the hypodiploid nuclei in THP1 cells by approximately 20% starting at a concentration of 0.5 mg/ml, while in Jurkat cells, a concentration twice as high (1 mg/ml) was necessary to have the same apoptotic level.

The PBMCs were less sensitive to MOE-b; 2.5 mg/ml of extracts were required to induce apoptosis of approximately 25% in treated cells (Fig. 3A).

The fresh seed preparation induced significant apoptotic levels only in THP1 cells, while it did not affect Jurkat cells or PBMCs, in which only higher concentrations induced a significant effect (Fig. 3B).

The effects of treatment with S boiled extracts were highly variable, depending on cell type. In THP1 cells, a significant increase in apoptotic level (20% of hypodiploid nuclei) from a concentration of 0.01 mg/ml was observed, while Jurkat cells needed 1 mg/ml of Fresh Weight (FW) to be affected by this treatment.

Interestingly, the boiled S preparation showed a specific pro-apoptotic effect on tumour cells, compared to the PBMCs of healthy donors, which were particularly resistant to the treatment. In fact, in PBMCs, the apoptotic level never exceeded 10%, even at the highest concentration used (10 mg/ml) (Fig. 3C).

Taken together, these results emphasized the non-significant (P>0,05) pro-apoptotic effect of the boiled mature seed preparation on lymphocytes from healthy donors, in contrast to the significant effect observed in the two tumour cell lines analysed.

In particular, we have also demonstrated that THP1 cells were more sensitive to the treatments in a dose dependent manner, while Jurkat cells appeared to be more resistant.

The evaluation of apoptosis after 72 h of treatment with MOE-f preparations underlined the high toxic effect of all these extracts on both cell lines and lymphocytes from healthy donors (Fig. 4).

Treatment with the frozen preparation of leaves induced a high level of apoptosis in all analysed cells. In particular, it increased the hypodiploid nuclei by approximately 25% with respect to the untreated cells; in THP1 cells, this increase occurred starting at a concentration of 2.5 mg/ml and in Jurkat cells, 5 mg/ml of FW induced approximately 50% of apoptosis.

The PBMCs were less sensitive; 5 mg/ml of MO was required to reach a 20% level of apoptosis (Fig. 4A).

The fresh seed preparation demonstrated a significant increase in apoptosis in THP1 and Jurkat cells at a low concentration. In detail, in THP1 cells, a significant increase of 30% in the apoptosis level, starting at a concentration of 0.25 mg/ml, was observed, while in Jurkat cells, we observed an 80% increase in apoptosis at a concentration of 0.5 mg/ml.

Interestingly, PBMCs from healthy donors were particularly resistant to the treatment.

In fact, the apoptosis level in these cells never exceeded 10%, even at the highest concentration (5–10 mg/ml; Fig. 4B).

In THP1 cells, treatment with seed preparation induced a significant increase in apoptosis (25% of hypodiploid nuclei) starting at a concentration of 0.01 mg/ml; in Jurkat cells and PBMCs, 1 mg/ml of FW was necessary to reach this level of apoptosis (Fig. 4C).

These results confirm the results obtained on cell proliferation and cytotoxicity: MOE-f preparations were more toxic than MOE-b preparations in all cells analysed.

Considering the inclusion of MO seeds in the List of Plant and Vegetable Integrators, in respect to the European Pharmaceutical Plant Legislation, and their low cytotoxic effects (Tables II–III), the anti-proliferative and pro-apoptotic activities of MO seeds on Jurkat cells were characterized in detail.

After 72 h of treatment, a significant reduction in viability and an increase in apoptosis level were observed in Jurkat cells (Fig. 5A). Moreover, MOE-S-b induces a significant reduction in DNA synthesis compared to untreated cells, as demonstrated by the BrdU assay (Fig. 5B).

Pearson's correlation analysis of all these data show that the reduction of DNA synthesis correlated with the decreased number of viable cells (Fig. 5C) and the increase of apoptotic cells (Fig. 5D) treated with MOE-S.

The pro-apoptotic effect, induced by 1 mg/ml FW of MOE-S in Jurkat cell lines, was further investigated, evaluating the expression of SIRT1 and BCL2 proteins through flow cytometry analysis. The treatment of Jurkat cells with MOE-S-b induced a significant decrease of alive cells that express BCL2 and SIRT1 anti apoptotic proteins. This significant reduction was observed analysing the median intensity fluorescence (MIF) and the percentage of BCL2 (Fig. 5E-G) and SIRT1 protein expression (Fig. 5H-J).

To better understand and explain the previously observed biological activity of MOE-S on human cells, we analysed the HPLC-DAD chemical profiles of MOE-S and measured their total concentrations of simple phenols and flavonoids by spectrophotometric assays.

HPLC-DAD analysis demonstrated that L (green line), FS (pink line) and S (black line) samples presented very different chemical profiles (Fig. 6). In particular, L extracts appeared richer in secondary metabolites than FS and S preparations, which only showed little chromatographic peaks. In detail, all L extracts showed several secondary metabolites with retention times of between 9 and 12.5 min, which were not present in MO seed extracts. Likely, these peaks could be flavonoids, due to their higher affinity with the apolar B solvent used during the analysis. Generally, all profiles were richer in the initial part of the chromatogram with respect to the second part, indicating that polar compounds, including sugars, were abundant in all extracts. Surprisingly, we also observed that the metabolic profiles of each sample (L, FS and S) varied greatly according to the extraction method (boiled or frozen; Fig. 6A and B, respectively), in qualitative and quantitative terms. The boiling method appeared, on average, to better preserve the MO secondary metabolites in the extracts than the freezing procedure.

Spectrophotometric analysis measured the total amount of phenols and flavonoids in MOE (Fig. 6C). L preparations had more phenols and flavonoids than FS and S ones, both in boiled and in frozen extracts, confirming the previous HPLC-DAD observation. In particular, the amount of flavonoids was higher than simple phenols in L samples, supporting the existence of peaks at a high retention time in the chromatographic analysis of these extracts. In S and FS samples, simple phenols were more concentrated than flavonoids. However, S samples were generally richer in both of these two classes of plant compounds than the FS extract.

Results also demonstrated that the frozen preparation protocol degraded flavonoids but not simple phenols in L extracts, with respect to the boiling procedure, while it did not influence the concentration of secondary metabolites in FS and S.

A recent publication characterised the natural Olea europea small RNAs profile in olive drupes highlighting the presence of the most conserved plant miRNAs. The study demonstrated the effects of this small RNAs pool on apoptosis and proliferation in tumors cells (12).

Based on this result and on our recent paper reporting the miRNome of MO seeds (31,32), the mol-small RNA pool was extracted from MOE-b seeds and the presence of plant mol-miR was evaluated.

In this extract the qRT-PCR analysis revealed the presence of plant miRNAs (Fig. 7) belonging to 40 conserved families of plant microRNAs (32) involved in different important plant cellular pathways such as the response to abiotic stress, growth, differentiation and cell proliferation (32–34) (http://www.mirbase.org/search). This result suggests their possible involvement in apoptosis and proliferation, previously observed with Olea europea small RNAs.