Increasing fish production in recirculating aquaculture system by integrating a biofloc-worm reactor for protein recovery

Highlights • Aquaculture waste was transformed into Tubificidae biomass for direct feeding.• Biofloc-worm reactor increased fish production by 17.1 % through protein recovery.• Worm predation enhanced aerobic denitrifier Deinococcus by 11.7 times.• Aerobic N cycling processes were increased, resulting in over 70 % gaseous N loss.


Text S1. Calculated methods
The carbon to nitrogen (C/N) ratio (RC/N) of the commercial feed and tapioca added to the B_RAS and BW_RAS was calculated using the following equation: RC/N = (F × PC × 16% × 8.8 + S × 82.8% × 44.4%)/(F × 35% × 16%) (1) Where F is the feeding rate (g day -1 ); PC is protein content in feed; 16% is the average nitrogen content of protein; 8.8 is the C/N ratio in the feeding commercial feed measured by Multi N/C 3100 TOC analyzer (Analytik Jena AG, Jena, Germany); S is the weight of daily tapioca addition; 82.8% is the carbohydrate in the commercial tapioca; and 44.4% is carbon content of tapioca.
The N content of commercial feed into the B_RAS and BW_RAS was calculated according to the following equation: (3) Where Nfeed is total N mass in the fish feed (g); Ftotal is total feeding amount of fish feed, 35% is the average protein content of fish feed, 16% is the average nitrogen content of protein.
The N content of biofloc was calculated according to the following equation: Where Nbiofloc is total N mass in the biofloc; DWbiofloc is the total dry weight harvest yield of biofloc (g), 35% is the average protein content of biofloc (Tubin et al., 2020), 16% is the average nitrogen content of protein.
The N content of tilapia was calculated according to the following equation: Ntilapia = WWtilapia× 25% × 59% × 16% (5) Where Ntilapia is total N mass in the tilapia; WWtilapia is the wet weight of tilapia (g), 25% the dry matter content of tilapia, 59% is the average protein content of tilapia (Khanjani et al., 2021), 16% is the average nitrogen content of protein.
The N content of tubificidae was calculated according to the following equation: Nworm = WWworm × 40% × 57% × 16% (6) Where Nworm is total N mass in the tubificidae; WWworm is the wet weight of tubificidae (g), 40% is the dry matter content of tubificidae, 57% is the average protein content of tubificidae (Yan et al., 2004), 16% is the average nitrogen content of protein.

Text S2. Library construction
DNA from all samples was amplified in triplicate by PCR for library construction.The 50-µL PCR mixture contained 25 µL of 2 × Premix Taq (Takara Biotechnology, Dalian Co. Ltd., China), 1 µL of each primer (10 µM), and 3 µL of template DNA (20 ng/µL).Additional ddH2O was added to reach a final volume of 50 µL.PCR was conducted using a Bio-Rad S1000 PCR thermal cycler (Bio-Rad Laboratory, CA, USA).The thermocycling steps were as follows: initial denaturation at 94°C for 5 minutes, followed by 30 cycles of 94°C for 30 seconds, 52°C for 30 seconds, and 72°C for 30 seconds, with a final extension at 72°C for 10 minutes.The quality of the library was assessed using a Qubit 2.0 Fluorometer (Thermo Fisher Scientific, MA, USA).

Text S3. Real-time Quantitative PCR
All specific primers for target genes were synthesized by Tyhygene Biotech Co., Ltd.(Wuhan, China).Each 20 µL reaction mixture consisted of 10 µL SYBR Premix (Takara, China), 0.5 µL DNA template, 0.5 µL each of forward and reverse primers, and 8.5 µL of sterile water.The forward and reverse primers for target genes, along with the annealing temperatures, are listed in Table S1.All plasmids containing the target genes were provided by Tyhygene Biotech Co., Ltd.(Wuhan, China) and subjected to a 10-fold serial dilution to serve as DNA templates for the construction of standard curves (R² > 0.99).RT-qPCR was performed using a CFX96 thermal cycler (Bio-Rad, USA).Relative abundance, defined as the absolute number of genes normalized to the weight of total suspended solid (TSS), was used in this study to compare the differences in denitrifying genes.

Text S4. Illumina Miseq Sequencing, Quality Control Assembly, Amplicon Sequence Variants (ASVs) Clustering, and Taxonomy Annotation
After extracting genomic DNA from the samples, the DNA of the conserved region was amplified using specific primers (341F/806R) with barcodes.The PCR products were gelpurified and quantified using a QuantiFluorTM fluorometer.Subsequently, the purified amplicons were equimolarly pooled for sequencing library construction.The library was paired-end sequenced (2×250) on an Illumina Miseq platform by Novogene (Beijing, China), following standard protocols.

Figure S5 .
Figure S5.Daily feeding amount for each fish tanks.

Figure S6 .
Figure S6.The impact of aeration rates and worm densities on biofloc biomass growth and worm predation, respectively.(a) Linear regression analysis of biofloc biomass growth under different aeration rates.(b) Linear regression analysis of biofloc biomass predation by worm under different worm densities.The slopes of linear regression represent the biofloc biomass growth rates and worm predation rates, respectively.

Table S1 .
Growth performance of tilapia reared in fish tanks of BRAS and BWRAS.

Table S2 .
Water quality, feed nitrogen input, and fish growth performance in aquaculture system using biofloc technology.