6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (trolox), tocopherols (α-, β-, γ-, and δ-), phytosterols, squalene, fatty acid methyl esters (37 FAME mix), 2,2′-azinobis (3-ethylbenzothiazoline)-6-sulfonic acid (ABTS) and 1,1-diphenyl-2-picrylhydrazyl (DPPH) were obtained from Sigma-Aldrich (Bellefonte, PA, USA). All other solvents were of chromatography grade and ultrapure water (Milli-Q system, Millipore, Bedford, MA, USA) was used throughout the experiments.

Avocado fruits (P. americana Mill. cv. Hass, Pinkerton, and Reed) were harvested from Hainan, Chain in February 2020 and stored at room temperature until full maturity before manually separated into seeds, pulp and peel. The weight of the whole fruit, pits, peel and pulp of more than 50 avocado fruits were evaluated. The pulp was immersed into citric acid solution (1%, w/v) for 15 min to avoid possible enzymatic darkening (7).

Avocado pulp was pulped into puree in a beater with a creamy consistency and then either oven-dried or vacuum-dried or freeze-dried as detailed below. The puree was spread onto stainless steel trays (80 × 100 cm) and then either dried in an oven (Fanen 320 SE, Brazil) with forced air circulation at 60°C or in a vacuum drier (Memmert Inc. VO 500 model, Germany) at 40°C and 600 mbar for 24 h. The temperatures were chosen according to the results of dos Santos et al. (8). For freeze drying, pulp was lyophilized at-40°C for 24 h with a freeze-drier (Edwards Pirani 501, West Sussex, United Kingdom). All drying was performed in triplicate and the dried pulp was vacuum packaged with light-proof bags and stored at-50°C for further use. Oil extraction was performed by Soxhlet extraction with petroleum ether, as described by AOAC 920.39. Yield was calculated as the ratio of the weight of oil extract to the dry weight of puree.

The moisture content, dry weight, total sugars, crude fat, crude fiber, protein and ash contents were determined according to AOAC, 2000. Fat content was determined as the method described by Rodríguez-Carpena et al. (14). The slip melting point, iodine value, free fatty acids content, saponification value and unsaponifiable matter content were analyzed as described in AOCS, 1999.

The fatty acid composition was analyzed as the method stated by Tan et al. (15) with minor modifications. FAME solution was prepared by adding 0.1 g of oil into hexane solution (5 ml) and sodium methoxide reagent (250 μl). Then, 5 ml of sodium chloride solution (26.5%, w/v) was added. After vigorous shaking for 15 s, the solution was stood at room temperature for 10 min to allow stratification. The top layer (1 μl) was analyzed with a gas chromatograph (Thermo Fisher Scientific Inc., Waltham, MA, USA) coupled with a polar capillary column (30 m × 0.25 mm, 0.25 μm, Austin, Texas, USA) and a flame ionization detector. The initial column temperature was maintained at 100°C for 2 min and then increased to 230°C at a rate of 5°C/min and was held there for 10 min. The temperatures of the injector and the detector were fixed at 250°C throughout the analysis. FAME standards were employed to identify the peaks on the obtained chromatograms.

Solid fat content (SFC) was measured using a Bruker Minispec (Model Mq 20) pulse nuclear magnetic resonance (NMR) spectrometer (Karlsruhe, Germany). Samples in the NMR tubes were heated at 80°C for 60 min, and then held at the corresponding measuring temperatures. After 30 min, SFC measurements were taken at 5°C intervals over the range of 0–40°C.

The contents of tocopherols were determined by HPLC as previously described (16). Folin–Ciocalteu reagent was employed to determine the total phenolic content according to the method of Rubilar et al. (17) with some modification. Phytosterols and squalene were identified and quantified with GC–MS. (18)

Antioxidant activity was assessed via DPPH assay as the method described by Sun et al. (19) with some changes. Briefly, 50 ml of various concentrations of AO solution (25, 50, 100, 200 or 250 mg/ml, dissolved in methanol) was blended with 2 ml of DPPH solution (24 mg/ml) and then incubated in the dark at room temperature. After 30 min, absorbance was then measured at 517 nm using a UV–visible spectrophotometer (Varian Cary 50, Mulgrave, Australia). Tert-butyl hydroquinone was recruited as a reference. Radical scavenging activity was calculated as follows: radical scavenging activity = (A_DPPH − A_AO)/A_DPPH × 100%. A_DPPH and A_AO are the absorbance of the DPPH solution and the test solution, respectively.

All experiments were carried out in triplicate. Results are presented as mean ± standard deviations. Analysis of variance and multiple comparison was performed with Duncan’s tests. p values less than 0.05 were considered statistically significant.

An avocado fruit includes three anatomical parts: peel, pulp and seed. The morphometric measurement results of the whole fruit, peel, pulp and seed from the three avocado varieties are demonstrated in Table 1. In the light of all significantly different measurements among these avocado varieties, a clear differentiation among them was indicated. Reed showed the largest fruit and seed. Hass presented the lowest pulp percentage (60.59%), which agreed with the data reported by Costagli and Betti (20). Moisture content in the fresh pulp from Hass was 73.35 ± 0.45%, and its crude fat content was 16.53 ± 0.13%, which was significantly higher than that of Pinkerton (12.55 ± 1.43%) and Reed (14.25 ± 0.99%). This data was similar to a previous study that reported the lipids content in avocado pulp to range from 10 to 15%, showing its dependence on genotype as well as other factors (e.g., cultivar., season and growing conditions) (13, 21). Unlike oil extracted from other fruits such as palm kernel oil, AO is extracted from pulp rather than seeds as they contain very little oil (<2%) and have hepatoxic substances (22). Furthermore, compared to Reed and Pinkerton varieties, Hass variety is rich in edible oils (23). Therefore, the pulp of Hass would be a better candidate for obtaining AO.

AOs were extracted from the three avocado varieties that were subjected to different predry-treatments (freeze drying, oven drying and vacuum drying). The yields of the five oils are demonstrated in Figure 1A. It was found that drying methods had significant effect on AO yield with the highest yield obtained by freeze drying. Comparatively, vacuum drying pretreatment produced the lowest oil yield (51.13 ± 4.33%). This result was in accordance with other studies (2, 24). The changes of AO yield resulted from different drying treatments could result from drying temperature and drying method (9). Besides, the results also indicated that under the same kind of drying method (freeze drying) significant different AO yields could be observed from varieties (Figure 1A). The highest oil yield (65.30 ± 0.10%) was observed for Hass variety, which was 12.10 and 8.23% higher than those of Pinkerton and Reed varieties, respectively. Therefore, freeze drying treatment could increase oil yield and Hass showed the highest extraction efficiency among the three varieties investigated in the present study. This agreed well with some earlier works (7, 24).

Figure 1B shows the SFC profile of the AOs. The SFCs of freeze-dried Hass oil (FHO), freeze-dried Pinkerton oil (FPO) and freeze-dried Reed oil (FRO) at 0°C were 13.36, 7.30 and 7.25%, respectively, and decreased almost to 0% at 40°C, showing consistency to previous findings (25, 26). Notably, FHO stood out from the other two oils (FPO and FRO) due to its higher SFC in a wide temperature range. This is contrary to the results reported by Yanty et al. (27), who detected low SFC for Australian Hass oil. However, the Hass oil from West Malaysia displayed high SFC from 13.57 to 18.00% at 0°C. This difference could result from the variance in fatty acids and triglycerides composition (25). In terms of the influence of drying treatments, the SFC curves for oven-dried Hass oil (OHO), vacuum-dried Hass oil (VHO) and FHO intertwined with each other in a small range, suggesting no obvious difference was induced from drying treatments. Therefore, Hass can be a promising base material in making cream and lotion for skin nourishment. Reed and Pinkerton, on the other hand, could be used for making sauce, marinade and salad dressing. In general, these results illustrate that the variety but not predry-treatment affects the SFC in AO. With respect to the viscosity profile (Figure 1C), FHO is remarkably different from others, which may be due to its unique physicochemical properties (2, 28).

The two factors influencing the functionality of oil are physicochemical properties and bioactive constituents. Therefore, the appearance, microstructure, fatty acid composition, quality indices as well as the tocopherol, phytosterol, squalene and polyphenol contents of the AOs obtained from the three avocado varieties that were dried with different methods were examined systematically.

The antioxidant capacities of the AOs determined by DPPH and ABTS assay are presented in Figure 3. The DPPH scavenging capacity ranged from 81 to 97%, which indicated favorable radical scavenging activity of all the AOs (Figure 3D). In terms of the effect of predry-treatments, vacuum drying endowed AO with higher radical scavenging activity as DPPH scavenging activity of VHO was greater than that of OHO and FHO. In addition, Hass oil extracted from oven-dried and vacuum-dried avocado pulp exhibited significantly higher DPPH scavenging capability than that of Pinkerton and Reed ones. With respect to ABTS scavenging activity, the Reed oil exhibited the highest capacity, followed by Pinkerton oil and Hass oil (Figure 3E). The high ABTS scavenging activity of Pinkerton oil was probably associated with the content of tocopherols and phenolic (Table 3). Similar to the results obtained in the DPPH assays the Hass oil from vacuum-dried pulp showed higher radical scavenging activity than that of oils from other predry-treatments. Likewise, a great decrease in the antioxidant activity of phenolic compounds was observed after its subjection to oven-drying (14, 23, 36).