Close this search box.

Optimized spray-dried conditions’ impact on fatty acid profiles and estimation of in vitro digestion of spray-dried chia/fish oil microcapsules – Scientific Reports

Procurement of raw materials

CS (Gazala’s Pantry, Faisalabad, Punjab, Pakistan) and skin-on fish fillets (Labeo rohita, rohu, a member of the carp family) were purchased from an SB Department Store, Faisalabad, Punjab, Pakistan). All Sigma–Aldrich® chemicals (St. Louis, MO, USA) were procured from the scientific stores in Punjab, Pakistan, and were at least research grade. The seeds were cleaned by washing with tap water and then dried at room temperature (27 ± 2 °C) to remove any dirt and other unnecessary materials. Fillets used for oil extraction were ~ 1 m long and ~ 2 kg. A total of ~ 50 kg of fish fillets were used.

Chemical composition of the raw materials

Moisture contents (MC)

The MC of the samples was estimated using the official AOAC method 930.1534. Samples (5 g) were dried in a hot-air-oven (Memmert, Äußere Rittersbacher, Germany) at 105 °C for 1 h in petri dishes (Fudau Cell Culture, Luolong, Henan, China). A digital scale (SF–400A, The Stationers, London, UK) was used for all weighing. After heating, the samples were cooled in a desiccator containing phosphorus pentoxide. The MC was calculated:

$$MC left( % right) = frac{{{text{W}}2 – {text{ W}}3}}{{{text{W}}2 – {text{W}}1}} times 100$$


where: W1 = Weight of petri dish, W2 = Weight of petri dish with sample, W3 = Weight of petri dish with sample after drying, W2–W3 = Loss of moisture, W2–W1 = Weight of fresh sample.

Ash content

The ash content was estimated using the AACC35 basic method 08–01. A 5 g sample was poured into a dry, pre–weigh crucible. Samples were placed in an electric muffle furnace (FHX–12, Daihan Scientific, Largo, FL, USA) at ~ 500–550 °C for 6 h. The crucibles were cooled in a desiccator and weighed immediately after room temperature was attained. The ash content was:

$$Ash contents left( % right) = frac{{{text{Wr}}}}{{{text{Ws}}}} times 100$$


where; Wr = Weight of residue, Ws = Weight of sample.

Crude protein

The crude protein was estimated using the oldest AOAC34 Kjeldahl method 64–50. A 2 g sample in a 10 mL test tube (Taian Youlyy, Xintai, Shandong, China) had a digestion tablet added (Merck, St. Louis, MO, USA) and 20 mL of 0.1 M H2SO4 for digestion. After 3–4 h digestion, a bright yellow color was obtained. The solution was cooled at room temperature. After cooling, distilled water was added to 50 mL. H2SO4 to trap the ammonia that was then volatilized during the distillation process. The ammonia was collected into a flask with 4% boric acid and a methyl indicator. The mixture was back titrated with 0.1 N H2SO4 to determine total nitrogen (Eq. 3). A conversion factor was used to convert nitrogen to crude protein. Crude protein (%) = Nitrogen % × 6.25, which was the calibration factor selected.

$$Nitrogen left( % right) = frac{{left( {{text{A}} – {text{B}}} right) times {text{N}} times 1.4007}}{{{text{W }}left( {text{g}} right)}}$$


where: A = mL of alkali of blank, B = mL of alkali of sample, N = Normality of alkali, W = Weight of sample.

Crude fat

The fat was measured with minor modifications using the AACC36 official method 30–10. A 2 g sample of raw material was wrapped in filter paper (grade 40: 8 μm). Then, 2 mL of 95% ethanol was gently added. After that, 10 mL of HCl was added and gently stirred. The beaker was placed in a water bath at 75 °C and mixed for 0.5 h until the solution was completely hydrolyzed. Then, 25 mL of ether was added to the solution and stirred for 1 min. The rinsed contents of the beaker were added into an extraction tube with 25 mL of redistilled petroleum ether (Sigma Aldrich, St. Louis, MO, USA) and again agitated for 60 s. The final solution was kept at room temperature with no disturbance until a layer of fat appeared on the fluid surface. The fat content was then alienated from the solution using centrifugation (Megafuge 8R Small Benchtop Centrifuge, Thermo Fisher Scientific, Dreieich, Germany) at 30,300 × g (600 rpm with a fixed angel rotor) for 15–20 min. After centrifugation, the fat content was filtered through Whatman No. 1 filter paper (Cytiva, Marlborough, MA, USA). The solvents were volatilized in a hot air oven at 100 °C for half hour.

Crude fiber

Crude fiber was obtained using AOAC34 Method No. 978.10. Sample (3 g) was digested using boiling H2SO4 (1.25%). Distilled and filtered through Whatman No. 1, water was used to wash the sample from the beaker. Samples were re–digested using NaOH (1.25%) and the beaker re–washed as previously. Then, the filtrate sample was collected in a dried crucible and placed on a hot plate to remove the excess water. The crucible was placed in a hot air oven for 2 h at 230 °C. The residues of the sample in the crucible were ashed in the muffle furnace for 3–5 h at 550–650 °C. The amount of crude fiber was estimated according to Eq. (4).

$$Crude fiber left( % right) = frac{{{text{W}}1 – {text{W}}2}}{{{text{Ws}}}} times 100$$


Where: W1 = Weight of crucible with fiber, W2 = Weight of crucible with ash, Ws = Weight of sample.

Nitrogen free extracts (NFE)

The NFE was:

$$NFE left( % right) = 100 – Mositure left( % right) + Crude fat left( % right) + Crude fiber left( % right) + Ash left( % right)$$


Oil extraction

CSO and FO were extracted using a cold press extraction and solvent extraction, respectively, according to the procedure of Rahim et al.21. Briefly, the CSO was extracted using a mini oil press model 6YL-550. The undesirable materials were removed from the extracted oil by sedimentation at room temperature for 24 h. The extracted oil was filtered through Whatman No. 1 and stored in the dark in screw-capped clear plastic bottles for a maximum of 1 wk. For the FO, solvent mixtures were prepared using methanol (Sigma Aldrich, 100 mL) and chloroform (Sigma Aldrich, 50 mL) at 27 ± 2 °C. Each fish sample (100 g) was soaked for 12 h in the already prepared solvent mixtures. A rotary evaporator was further used to evaporate the solvent mixture at 50 °C. The FO was saved in the dark in the plastic bottles for a maximum of 1 week.

Emulsion preparation

A blend of CSO and FO (50:50%) was produced at room temperature. All emulsions were prepared in the same manner as described in the recent research work of Rahim et al.14. An emulsion was progressively formed by combining 97 mL of distilled water with 3 mL of toluene. A 15 mL blend of CSO and FO and 85 mL of the above solution were gently mixed in another beaker. Then, soy lecithin (1%, w/w) was utilized as a natural emulsifier and stabilizer. Moreover, GA and MD (1:1 mass ratio, 15 g) were added as a wall material (WM) with a magnetic stirrer for 15 min. A homogenizer was used to mix at 110,00 rpm for 10 min.


SDM were prepared in a lab–scale mini spray–drier (Model number TPS–15, Toption Co., Shanghai, China)37. The spray–drier operating conditions for optimization were IAT of 125, 140, 155, 170, and 185 °C; WM of 5, 10, 15, 20, and 25%; PS of 3, 4, 5, 6, and 7 mL/min; and NS of 3, 5, 7, 9, and 11 s. The spray–dryer was preheated to the required temperature before feeding. A hydrostatic pump was used to pump the homogenized samples into the atomizer at the required flow rate. Then, the atomizer sprayed the raw materials into the drying chamber with hot air. Powder and gas entered into a cyclone separator, where they were separated in a glass collection tube. The SDM were inserted in a polythene zip bags (Local market, Faisalabad, Punjab, Pakistan) and kept at 27 ± 2 °C and humidity 40 ± 45% for a maximum of three weeks.

Characteristics of SDM

Fatty acid composition

Using the process of Rahim et al.21, methyl esters of fatty acids were evaluated using a gas chromatograph (GC, Model 7890–B, Agilent Technologies, Santa Clara, CA, USA). The esters of the samples were prepared using the standard method Ce 1f.–96 with some changes as explained by AOCS38. Briefly, 2 g of samples were inserted into a glass test tube with 2 mL of 1 M of caustic soda in methanol (100% concentration). Then, test tubes were placed into a boiling water bath at 95 °C until the mixture of samples became colorless. Then 3 mL of boron trifluoride was gently mixed-in and re-heated at 95 °C for 10 ± 3 min. After heating, hexane (3 mL) was mixed vigorously until two phases appeared. The supernatant was poured into GC vials and the peak areas were calculated using the software with the instrument and assumed to represent percent concentration assuming an equal response to each peak on a weight basis. The gas flow velocity for He, H2, and O2 (ratio 2:4:40) was adjusted from 20 to 25 mL/min at 185 °C to estimate the fatty acid profile using a SP–2560 capillary column 100 m × 0.25 mm id (Agilent Technologies, model 7890 B). FAME 37 (Supelco, St. Louis, MO, USA, a C37 alkane) and SLB–IL111 (Merck, St. Louis, MO, USA) internal standards were used for the fatty acid profile and their isomers.

Fourier transform infrared (FTIR) and X-Ray diffraction analysis (XRD)

FTIR was operated from ~ 650 to 3500 cm−1 using an Agilent Technologies Cary 630 to assess the quantitative and qualitative analysis of the samples. The oil blend sample (S1), wall material sample (S2), and SDM sample (S3) were placed onto flat glass plates. Further, a very clean another plate was placed direct on top to get a clean film. The plate was kept in the special sample holder and the spectral wavenumber resolution was run at 4 cm−1 at ambient temperature until the CO2 peaks were minimized. Agilent MicroLab software was used to evaluate the signals and identify the peaks39. The crystallinity of samples was performed using an X-ray diffractometer (Model D8, Bruker, Berlin, Germany). The samples were placed in a slot and then pressed with frosted glass to get a good texture. XRD patterns were recorded at room temperature with a wavelength of 1.54 Å, angle 2θ ranging from 25 to 80º, and the instrument operated at 30 kV.

Impact of SDM in vitro with a simulated gastro–intestinal environment

Preparation of simulated gastric fluid (SGF)

A SGF was used for the in vitro release behavior of SDM as described by Minekus et al.40 with minor changes. NaCl (2 g) and 7 mL HCl (36%) were mixed with 900 mL deionized water (Sigma Aldrich) using a magnetic stirrer for 5 min and 3.2 g pepsin (Sigma Aldrich) to obtain a pH of 2.0 and the solution brought to 1000 mL with deionized water. The solution was incubated at 37 °C for 120 min with mixing at 100 rpm in an incubator shaker and kept at 4 ± 0.5 °C in a refrigerator for a maximum of 3 days.

Preparation of simulated intestinal fluid (SIF)

A SIF was prepared according to Goyal et al.41. A stock solution was prepared by dissolving 6.8 g of KH2PO4 in 850 mL of distilled water. The 0.2 M NaOH and 100 g of 1 × USP pancreatin (Sigma Aldrich) were brought to pH 1.2 with the hydrochloric acid solution, 0.1 M. The solution was incubated in the dark with mixing at 4 °C for 6 h and brought to 1000 mL with distilled water. It was refrigerated for a maximum of 3 days.

In vitro release behavior of SDM with SGF

The samples were prepared in test tubes by adding 5 g SDM containing 50% oil (w/w basis) to 900 mL of SGF. These were kept in the dark at 38 °C for 2 h. After incubation, 30 mL of petroleum ether and diethyl ether at 1:1 were added with magnetic stirring. A separatory funnel was used to separate the oil from the solution at room temperature. The separated oil was heated at 80 °C until the solvent evaporated. The oil was dried in the hot air oven for 30 min at 100 ± 5 °C42. The amount of oil in the capsule was calculated using the amount of starting oil in the SDM (loading capacity) and the percentage of oil released in each synthetic digestive fluid:

$$Oil release left( % right) = frac{{{text{Ao}}}}{{{text{As}}}} times 100$$


where, Ao = Amount of oil release, As = Amount of sample.

In vitro release behavior of SDM with SGF and SIF

To determine the oil release from SDM, 60 mL of both SGF and SIF were put in separate 500 mL beakers and mixed with 6 mL SDM. The SGF was incubated at 37 °C for 2, 3, and 4 h while shaking at 110 rpm. Subsequently, it was transferred into SIF and placed in the dark at 37 °C for 2 h and analyzed at 230 nm using a spectrophotometer (Analytik Jena AG–Specord 200 Plus, Jena, Germany). The quantity of released oil was calculated by evaporating the petroleum ether in a hot air oven at 80 °C for 30 min43.

Model fitting and statistical analysis

A central composite design (CCD) of the response surface methodology (RSM) with a total of 30 runs was used to optimize the four independent factors, while the response variables were: ALA, EPA, and DHA. All spray–drying runs were mixed to reduce repeatable errors in the model. The optimal values were evaluated for their level of significance with 5% (p ≤ 0.05) being significant using the Stat–Ease® software (version, Minneapolis, MN, USA). A statistical technique analysis of variance (ANOVA) was assessed to calculate the significant differences between independent variables and the validity of the experimental design. The validity or adequacy of the CCD was clarified using an estimation of the regression coefficient, R2–adj, and the lack of fit. The operating conditions of the spray–drier were presented in coded and actual levels that were predicted using the CCD, the maximum level was + 2, while the minimum level was − 2. Six central points (C1–C6) were evaluated to determine the response error. The linearity of the CCD for the independent factors were evaluated for the different dependent factors to estimate the quadratic effect of the spray–drier operating variables. The inter–day repeatability was calculated by analysis of the same emulsion for spray–drying one time/day for four days44. The dependent variables obtained according to the CCD were fitted to a second–order polynomial model and regression coefficients calculated. Statistix (version 8.1, analytical software, Tallahassee, Florida, USA) was used for the in vitro analysis.