Nutritional properties of selected superfood extracts and their potential health benefits

Materials

Chia (Organic Road, Australia), flax (Lotus, Australia), hulled sunflower seeds (Natural Road, Australia) and hulled hemp seeds (Hemp Foods Australia, Australia) were purchased commercially. A pre-ground and processed commercial hemp flour (Hemp Foods Australia) was also purchased, this flour was made from whole hemp seed with the oil pressed out and some of the protein extracted.

Flour preparation

Approximately one cup of seeds were ground using a blade mill-type Multi Grinder (Sunbeam, Australia) for approximately 5 min at room temperature to create a seed meal. The meal was sieved and larger pieces were ground again. This flour was sealed in a falcon tube and placed in a sealed container in the fridge with silica beads to prevent excess moisture.

Seed and flour nutritional profile

Where relevant, nutritional information such as protein, energy (kJ), fat, dietary fibre and carbohydrate contents were taken from the commercial packaging of the purchased seeds. Lipid content and fatty acid profiles of flours were determined by the Waite Lipid Analysis Service, at the South Australian Health and Medical Research Institute (Liu, Mühlhäusler & Gibson, 2014). Total lipid in whole seed flour was determined by modified (Folch, Lees & Sloane Stanley, 1957) and fatty acid profiles were determined by gas chromatography of transesterified lipids following (Liu, Mühlhäusler & Gibson, 2014).

Crude aqueous body temperature seed extracts

To generate a crude aqueous body temperature extract, 4 g of flour was added to 200 mL of RO (reverse osmosis) water at 37 °C in a borosilicate glass beaker and the mixture was left to agitate on a heated magnetic stirrer to maintain this temperature. After 3 h, the mixture was transferred into 50 mL polypropylene tubes and centrifuged at 1,000 rpm (200 × g) for 10 min to pellet any debris. The supernatant was removed and freeze dried to a constant weight using a Freezone six freeze dry system (Labconco, Kansas City, MO, USA). Yield of extract was determined as the percentage of freeze-dried extract mass to mass of flour (Figure 1). Freeze dried extracts were dissolved in sterile PBS (Life Technologies^™, Carlsbad, CA, USA) at a concentration of 5 mg/mL for cell culture experiments and sterile RO water for compositional analysis.

IEC-6 and Caco-2 cell culture

IEC-6 (CRL1592) and Caco-2 (HTB-37) cells originally purchased from the ATCC were provided by SEPB and cultured in Gibco^® Dulbecco’s Modified Eagle Medium (DMEM; Life Technologies^™, Carlsbad, CA, USA) with 10% Fetal Bovine Serum (FBS; Sigma, St. Louis, MO, USA), antibiotic/antimycotic solution (10,000 units penicillin, 10 mg streptomycin and 25 μg amphotericin B per mL) (Sigma, St. Louis, MO, USA). The cells were maintained in a 37 °C incubator with 5% CO₂ in 75 cm² flasks (Greiner bio-one, Kremsmünster, Austria) and the media was changed every 72 h. Cells were passaged at 70–80% confluency and split at a 1:3 ratio using Trypsin (Gibco^®, Amarillo, TX, USA) to dissociate the cells from the flask surface. Cells were stained with trypan blue (Gibco^®, Amarillo, TX, USA) and counted using a haemocytometer on an Olympus inverted microscope.

Cell viability assay

A dose-response cell viability assay was performed as per Kumar, Nagarajan & Uchil (2018). IEC-6 or Caco-2 cells (Sambruy et al., 2001) were plated in 96 well plates at a concentration of 1 × 10⁵ cells/well in 50 µL of DMEM with 10% FBS, then left for 24 h to adhere. Extracts (5 mg/mL in PBS) were diluted in DMEM with 10% FBS to 1 mg/mL and then two-fold serial dilutions were prepared (resulting in concentrations of 1 mg/mL, 0.5 mg/mL, 0.25 mg/mL, 0.125 mg/mL, 0.0625 mg/mL and a vehicle control, 0 mg/mL). The extract treatments (50 µL) were applied to triplicate wells, and the plate was left to incubate for another 24 h. MTT solution (Invitrogen, Waltham, MA, USA) was diluted with PBS to 1 mg/mL, filter sterilized and 50 µL was added to every well using a multichannel pipette and the plate was incubated for 4 h at room temperature. A 100 µL volume of dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) was added to every well, the plate was covered in aluminium foil and agitated on a plate shaker for 15 min to extract the formazan product. Absorbance was read using a Thermo Multiskan Spectrum spectrophotometer at 570 nm. The results were converted into a percentage by comparing them to the vehicle control value:

$$Cellviability\left(\mathrm{\%}\right)={\displaystyle \frac{abs570\left(sample\right)}{abs570\left(control\right)}\ast 100}$$

Each cell viability assay was repeated to obtain 3–4 independent experiments per cell line.

Extract characterisation

Ferric reducing antioxidant power (FRAP) assay

A FRAP assay was performed in triplicate as per Benzie and Strain (Benzie & Strain, 1996). Each of the crude aqueous seed extracts were serially diluted using dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) at 1:2 dilution, eight times, with a starting concentration of 1 mg/mL. The ascorbic acid control was also diluted in the same manner. A ferrous sulphate solution was diluted using milli-Q water at concentrations of 0, 0.1, 0.2, 0.4, 0.5, 0.6, 0.8 and one mM. A 15 µL aliquot of each of the extract samples were added at multiple concentrations in a 96 well plate. A 15 µL volume of the FRAP reagent containing acetate buffer, TPTZ (2,4,6-Tris(2-pyridyl)-s-triazine) and FeCl₃ (iron (III) chloride) were added to every well and the results were read on a spectrophotometer at 570 nm.

Statistical analysis

Where applicable, statistical differences in extract composition were determined by ANOVA with a post-hoc Tukey’s Honestly Significant Difference (HSD) test in GenStat 15^th Edition.

For MTT assays, marginal models for viability were fitted using Generalized Estimating Equations using R version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria), accounting for repeated experiments assuming an independence working correlation structure. Cubic splines were used to explore the non-linear dose-response relationship of concentration on viability. Separate marginal models were fitted with categorical concentrations to allow for comparisons to the DMEM + FBS (vehicle) group.

Fatty acid and micronutrient analyses of ground flours

Prior to preparing extracts for cell treatment, the seeds were ground into a flour. Tests on the flour indicated that they all contain beneficial and essential micronutrients including magnesium, calcium, iron and zinc (Table 1). Flax seed flour had the highest content of omega-3 fatty acids in comparison to omega-6 at a ratio of 11:3, followed by chia flour at 3:1 (Table 1). Comparatively, hulled hemp seed flour, processed hemp flour and sunflower seed flour all had higher levels of omega-6 compared to omega-3; sunflower flour had the highest ratio (1:614 of omega-3:omega-6). Chia seeds showed a significantly higher amount of carbohydrates (37%) in comparison to the other seeds (Table 1). Hemp flour had the highest level of dietary fibre (67%), followed by chia (35%) then flax (29%). Sunflower had the highest total fat content (57%), followed by hemp seed (54%), flax (43%), and chia (29%), with hemp flour being the lowest (9%) (Table 1).

All seeds contained magnesium with hulled hemp seeds having the highest level (5,359 mg/kg), followed by flax (4,425 mg/kg), while chia, sunflower and hemp flour had slightly lower levels (3,296–3,525 mg/kg). Both the hulled hemp seed and hemp flour contained the highest levels of iron (196 mg/kg) compared to the other seeds (80–145 mg/kg). Hulled hemp seed had the highest level of zinc, followed by hulled sunflower, hemp flour and chia, with flax having the lowest level of zinc (Table 1). Only flax seed had detectable levels of sodium at 1,073 mg/kg compared to the other seeds. Chromium, cobalt, selenium and molybdenum were present in very low levels or not detectable.

Aqueous seed extract composition

To produce material that could be used for treating the cells, aqueous extracts FSE, CSE, HSE, HHE and HFE were generated from flax seed, chia seed, hulled sunflower, hulled hemp and hemp flour, respectively (Fig. 1). The same method was used on each seed yet varying amounts of extractable material were produced. Yield was highest for FSE at 39.46%, followed by HHE at 32.34%, HSE at 25.06%, CSE at 16.28% and lastly HFE at 8.06%. These numbers were calculated by comparing the initial weight (4 g) with the freeze-dried weight of the aqueous extract.

MTT cell viability assays

Both cell lines showed a similar response to each of the seed extracts (Figs. 2, 3, S2 and S3). Caco-2 and IEC-6 cell viability showed little change when treated with HHE or HFE when compared to the vehicle control (Figs. 2 and 3). FSE resulted in the largest increase in cell viability in both Caco-2 and IEC-6 cells (at the highest dose, 1 mg/mL, 163% and 195% of the vehicle control respectively) (Figs. 2 and 3).

Protein and monosaccharide analyses

Analyses were performed to determine the protein and monosaccharide content of each crude aqueous seed flour extract. The protein analysis showed FSE had the highest levels of protein, followed by CSE when measured in the body temperature water extracts (Fig. 4). These levels differed slightly from the commercial label, which was based on the whole seed prior to extraction (Table 1). Flax, chia and hulled sunflower seed all originally had the same amount of protein per gram, however post-extraction, different levels were observed, with FSE having the highest protein content of 20.62 ± 1.28 (%w/w) and HSE having the lowest content of 7.60 ± 0.15 (%w/w) (Fig. 4). A monosaccharide analysis was performed which showed the presence of key polysaccharide building block monosaccharides, in varying concentrations between extracts (Fig. 4). CSE and FSE had the characteristic monosaccharides found in the mucilage fraction of each seed type. HSE, HFE and HHE extracts showed high levels of glucose but limited levels of the other monosaccharides quantified (Fig. 4).

Mineral analysis of the aqueous extracts

Essential minerals were tested both pre- and post-extraction as this allowed for comparison of the availability of each mineral. The results showed high variability for each extract for the minerals measured (Fig. 5). For minerals such as potassium, chromium, selenium, iodine and copper, there was a slight increase in concentration from the flour to the extract from every seed. Sodium was only present in flax seed, with a three-fold increase in concentration in the extract. Hulled hemp and chia seed showed a similar level of magnesium in both the flour and corresponding extract, with the concentration increasing in the extract. The levels of magnesium were relatively similar in flax seed, hemp flour and hulled sunflower and did not change significantly in the extracts. Flax seed showed consistent iron levels in both the flour and the extract, hulled sunflower showed a slight increase in the extract and chia seed, hulled hemp and hemp flour showed a slight decrease (Fig. 5). Chia seed had higher levels of calcium in the flour sample, but there was a decrease in the extract. Manganese showed a substantial increase in concentration in the extract for hemp flour only. Cobalt was enriched in the extracts from flax seed and chia seed compared to the initial flour sample. Nickel showed an initial high concentration in hulled hemp but was almost entirely absent in the extract.

Extract composition

The protein, monosaccharide and mineral analyses were combined and compared for the total extracted material (Fig. 4). This figure shows that even with the combined protein, carbohydrates and minerals content quantified, there was still over 50% of the aqueous extract which was not categorised. The extract composition of FSE and HSE is predominantly protein, while for HFE it is carbohydrates. For CSE and HHE, more than 75% of the extract is currently uncharacterised.

Antioxidant capacity of body temperature extracts

The antioxidant capacity of each of the seed extracts was determined using a ferric reducing antioxidant power (FRAP) assay (Fig. S1). HFE and HHE extracts had particularly low antioxidant capacity, while FSE and CSE showed a slight but significantly higher antioxidant capacity (P < 0.05). HSE showed the highest antioxidant capacity of all the extracts.