Sorvall lynx

Banana starch was extracted according to the previous procedure [18 (link)]. After green bananas were washed, their peels were removed. Then, the banana flesh was cut into 2 mm slices and immediately placed in the color-protection solution (20 g/L citric acid and 2 g/L ascorbic acid) for 15 min, followed by homogenization in a blender (Zhejiang Supor Co., Ltd., Zhejiang, China) with ultrapure water (1:2, w/v) for 2 min. The homogenized mixture was centrifuged (Sorvall LYNX, Thermo Fisher Technology Co., Ltd., Shanghai, China) at 5000 rpm for 10 min. The sediments were isolated and dried in an oven (CS101-ABN, Yongsheng Test Equipment Co., Ltd., Chongqing, China) at 45 °C for 6 h. The dried banana starch was ground into powder by a mortar (Beijing Zhongke Aobo Technology Co., Ltd., Beijing, China) and screened through a 100 mesh sieve (Shanghai Merxi Scientific Instrument Co., Ltd., Shanghai, China). The starch was stored in a brown reagent bottle and put in the freezer (Haier Co., Ltd., Qingdao, China) at 4 °C for further use. The prepared banana starch was native starch and named NS.

Guo et al. [20 (link)] was referenced for the solubility (S) and swelling power (SP) of starch. In a centrifuge tube, the sample was mixed with ultrapure water to prepare the suspensions (2.0%, w/v). The suspensions were heated at 50, 60, 70, 80, and 90 °C in a water-bath shaker (SWB-2000, Tianjin Aoxenes Instrument Co., Ltd., Tianjin, China) for 30 min. The mixture was centrifuged (Sorvall LYNX, Thermo Fisher Technology Co., Ltd., Shanghai, China) at 4000 rpm for 10 min after natural cooling. The sediments were dried (CS101-ABN, Yongsheng Test Equipment Co., Ltd., Chongqing, China) at 105 °C to a constant weight. The S and SP were calculated as Equations (2) and (3), respectively: $$\mathrm{S} (\%){=\mathrm{W}}_1 / {\mathrm{W}}_0\times 100$$
$$\mathrm{SP} \left(\mathrm{g} / \mathrm{g}\right){=\mathrm{W}}_2 / {\mathrm{W}}_0$$
where W₁ is the constant weight after drying, W₀ is the initial weight of the starch sample, and W₂ is the weight of the swollen starch.

Banana-starch–resveratrol complexes were prepared referring to Gao et al. [14 (link)] with some differences. Briefly, banana starch treated by ultrasound or untreated (10 g) and resveratrol (1.0 g, 10% based on starch weight) were suspended in 200 Ml of 30% ethanol solution with a stirring machine (HJ-6A, Jiangsu Jinyi Instrument Technology Co., Ltd., Jiangsu, China) at 150 rpm and 70 °C for 1 h. The suspensions were subsequently cooled and centrifuged (Sorvall LYNX, Thermo Fisher Technology Co., Ltd., Shanghai, China) at 5000 rpm for 10 min. The sediments were washed with a 50% ethanol solution to remove uncombined resveratrol, then centrifuged at 3000 rpm for 15 min. The resin step was repeated 3 times. Finally, the sediments were freeze dried (SCIENTZ-12N/A, Ningbo Xinzhi Biotechnology Co., Ltd., Ningbo, China) for 12 h and ground. The final samples were named Res-NS, Res-U₄₀S, Res-U₆₀S, Res-U₈₀S, and Res-U₁₀₀S.

Individual A. tumefaciens colonies were propagated in lysogeny broth at 28 °C, 220 rpm for 2 days, and harvested by centrifugation at 4629 g for 6 min (Sorvall Lynx, ThermoFisher Scientific, Waltham, MA). Pellets from each culture were resuspended in infiltration buffer (10 mm MES, 10 mm MgCl₂, pH 5.6, 100 μm acetosyringone) and adjusted to an OD₆₀₀ of 0.3 for the expression of the individual proteins, or an OD₆₀₀ of 0.9 and mixed together with 1 : 1 : 1 ratio (final OD₆₀₀ of each Agrobacterium suspension is 0.3) for the co‐infiltration of all three constructs. N. benthamiana plants were grown on custom‐mixed soil comprising of peat, 2.5 kg/m³ dolomite limestone, 1.3 kg/m³ base fertilizer, 2.7 kg/m³ Osmocote^® (applied every 3–4 months), 0.3 kg/m³ Exemptor^®, and 0.25 kg/m³ wetter in a controlled environment of 16‐h photoperiod generated by 400 W sodium lamps, 24 °C and 70% relative humidity (Pang et al., 2019 (link)). The first three mature leaves of plants grown for 3 weeks after pricking out were infiltrated using a needless syringe (Thuenemann et al., 2013a (link), 2013b (link)) and maintained at 23–25 °C with 16 h lighting.

The freeze-thaw stability of the samples was analyzed based on the report of Zheng et al. [22 (link)]. Briefly, starch (1.0 g) and ultrapure water (16 mL) were blended in a weighted centrifuge tube. Then, the suspension was heated (C22-RT22E01, Midea Group Co. Ltd., China) in a boiling-water bath for 30 min. Starch gels were frozen at −20 °C for 24 h in a freezer (Haier Co., Ltd., Qingdao, China). Then, the samples were thawed at room temperature for 6 h and centrifuged (Sorvall LYNX, Thermo Fisher Technology Co., Ltd., Shanghai, China) at 4500 rpm for 15 min. The supernatant was abandoned and the remaining precipitate was weighed. The freeze-thaw cycle was repeated three times. Syneresis reflects the stability of the freeze-thaw (Equation (5)).
$$\mathrm{Syneresis} (\%)=\left(\mathrm{W} / \mathrm{M}\right) \times 100$$
where W (g) is the weight of supernatant, and M (g) is the weight of original dry starch.