Freezone 2.5 l

The pre-treated swim bladders were digested in 0.5 M acetic acid containing 2% pepsin (w/w) at a 1:40 (w/v) tissue/solution ratio. The mixture was continuously stirred at 4 °C for 24 h. After digestion, the viscous extract was filtered with two layers of cheesecloth and precipitated by adding NaCl to a final concentration of 1.2 M. The precipitate was collected by centrifuging at 16,000× g at 4 °C for 20 min using a Megafuge 16R centrifuge (Thermo Scientific Co., Waltham, MA, USA), and the resulting pellets were dissolved in 0.5 M acetic acid. This solution was purified using a dialysis membrane (14 kDa molecular weight cut-off) against distilled water for 72 h with a change of solution every 4 h. The resulting collagen was lyophilized (Free Zone 2.5 L, Labconco Corp., Kansas, MO, USA) and stored at −20 °C until further analysis.
The collagen yield was calculated based on the wet and dry weights of the raw material before and after processing, using Equation (1).
$$\mathrm{Yield}\left(\%\right)=\left[\frac{\mathrm{Wc}\left(\mathrm{g}\right)}{\mathrm{Wd}\left(\mathrm{g}\right)}\right]\times 100$$
where Wc is the weight of the lyophilized collagen and Wd is the dry weight of the initial swim bladder prior to pre-treatments.

In this study, two bacteria strains were used: A. vinelandii Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSM) 2289 and B. subtilis DSM 10. A. vinelandii was grown in Burk’s medium (BM) (Newton et al., 1953 (link)), whereas B. subtilis was grown in a minimal mineral medium (MM).
Bacterial filtrates were produced as described elsewhere (Ferreira et al., 2019b (link)). Shortly, bacteria were grown in Fe-depleted media until siderophore concentration was stable (72 and 48 h for A. vinelandii and B subtilis, respectively). Bacterial biomass was pelleted by centrifugation, and the supernatant was filtered through a 0.45-µm pore size filter. The resulting filtrate was mixed with corn starch (5 and 15 g L⁻¹ for B. subtilis and A. vinelandii, respectively), which was used as an anticaking agent. Then, the mixture was freeze dried (Labconco FreeZone 2.5 L coupled with a VacuuBrand RC 6 pump, USA). The resulting powder was homogenized and stored at 4°C until use.

Thornless blackberries (Rubus fruticosus ‘Ouachita’ PP17162) were handpicked at Jacob W. Paulk Farms, Inc. (Wray, GA, USA) over two summers. This cultivar features an erect-type morphology and was released by the University of Arkansas breeding program. Six lots of blackberries (~1200 g per lot) were collected each summer, vacuum packaged (Henkelman 600, Henkelman BV, The Netherlands) and stored at −40 °C. Representative samples (~150 g × 3) were randomly taken from all lots, the berries lyophilized using a FreeZone 2.5 L freeze dryer (Labconco Corporation, Kansas City, MO, USA), and then stored in a −80 °C freezer until analysis.

Frozen fish scales were mixed with distilled water with a ratio of 2–12% and then hydrolyzed hydrothermally at 85–135 °C for 15–120 min. The insoluble fish scale residues were removed by centrifugation at 8,000 g for 10 min and the supernatants were collected. After being freeze-dried with a lyophilizer (Freezone 2.5L, LABCONCO, Kansas City, MO, USA), tilapia scale protein powder was obtained and stored at −18 °C. Frozen fish scales were also treated by the conventional method as described by Wang et al. [23 (link)] with some modification. The scales were immersed in 0.1 M HCl for 2 h for decalcification and then washed with tap water until neutral to obtain the decalcified scales. The decalcified scales were mixed with distilled water (1:20, w/v) and heated at 60 °C for 2 h. The gelatin solution was obtained after centrifugation at 8000× g for 10 min. In addition, frozen fish scales were also dried at 50 °C for 12 h and then ground with an IKA A11 basic analytical mill (IKA, Staufen, Germany) to get the ground scales. As a comparison, the obtained ground scales and decalcified scales were further treated at 135 °C for 90 min to verify whether decalcification or pulverization is necessary for fish scales during hydrothermal treatment.

Fish bone proteins (FBP) were prepared through steam explosion-assisted extraction (SE) and hot-pressure extraction (HPE) processes, respectively. The steam explosion pretreatment was performed on a steam explosion device (QBS-200B, Gentle Science & Technology Co. Ltd., Hebi, China). About 300 g of fish backbones were put into a 5 L material vessel and exposed to the saturated steam at 159 ± 0.5 °C for 2 min, and finally, were terminated by explosive decompression. The exploded materials were collected and ground with an IKA A11 basic analytical mill (IKA, Staufen, Germany) to pass through a 0.38-mm-diameter mesh, and then extracted in distilled water (1:3, w/v) at 60 °C for 2 h in a water bath. In another set of experiments, fish backbones were further cut into segments of about 0.5 cm, mixed with three volumes of distilled water (w/v) and hanged in an extraction pot at 121 ± 0.5 °C for 70 min through HPE process according to Tan et al. [20 (link)]. The resulting mixtures were both filtered using gauze to remove the backbone residues. The filtrates were collected, freeze-dried with lyophilizer (Freezone 2.5 L, Labconco, Kansas City, MO, USA) and then stored at −20 °C until further analysis. The fish bone protein powder prepared by SE and HPE pretreatment was designated as SFBP and HFBP, respectively.