In land-based turbot nurseries that transfer 60–150 g juveniles to seawater grow-out cages, the live Edwardsiella tarda strain EIBAV1 vaccine API is compounded into an aqueous injectable suspension for intraperitoneal vaccination immediately before transport. The injection route is selected because the handling event is already required for grading and size sorting, and the edwardsiellosis challenge risk rises with transport-associated confinement and skin abrasion. Aseptic compounding is performed under EU GMP Annex 1 isolator conditions because the live bacterial vaccine cannot be terminal-filtered through 0.22 µm membranes without quantitative loss of the antigen. The freeze-dried API, standardised to a release titre of not less than 1×10^10 CFU/g, is reconstituted at 2–8°C with sterile phosphate-buffered saline at a ratio of 10 g API per 1,000 doses, yielding a uniform suspension titred to 1×10^8 CFU per 0.1 mL. The compounding vessel is a 50 L stainless-steel jacketed tank with bottom-mount magnetic stirring at 120 rpm; the suspension is transferred to the filling line through single-use silicone tubing of 2.4 mm inner diameter. A peristaltic filling pump calibrated gravimetrically every 15 minutes delivers 0.1 mL ± 5% into 100 mL Type I borosilicate glass vials under Grade A laminar airflow; bromobutyl stoppers and aluminium flip-off caps are applied at 2,000 vials/h. Each lot is released against identity, live titre, extraneous contaminant absence, and residual moisture criteria under Ph. Eur. 0062, and stability is tracked under VICH GL3. The terminal finished product is a refrigerated injectable suspension for turbot, stored and shipped at 2–8°C. Published data for this exact injection configuration is limited, so the batch release specification relies on challenge model potency rather than in-vitro physico-chemical indicators alone.

What Bath Immersion Parameters Govern Larval Uptake in Turbot Nurseries?

For turbot larvae between 1 g and 5 g, immersion vaccination is selected over injection because the animals are too small for reliable intraperitoneal delivery and the hatchery already operates batch-handling tanks with controlled water chemistry. The liquid or reconstituted vaccine concentrate is diluted at 1:100 v/v into clean seawater in a 1,000 L conical fibre-reinforced plastic immersion tank; the seawater is conditioned to 18–20°C, salinity 30–32 g/L, dissolved oxygen 6–8 mg/L, and pH 7.8–8.1. Two ceramic oxygen diffusers operating at 8 L/min keep the vaccine uniformly dispersed, and fish are introduced at a maximum biomass of 15 kg/m³. The target bath titre is 1×10^8 CFU/mL, verified before fish introduction by spread-plating on tryptic soy agar with 2% w/v sodium chloride and incubating at 37°C for 24 h. Immersion duration is limited to 60 s; extension beyond 90 s has not been shown to improve mucosal uptake in production-scale turbot hatcheries and raises oxygen consumption. Post-immersion water is treated with 10 mg/L available chlorine for 30 minutes before discharge under national aquatic animal health permit conditions. The terminal product is a 5 L high-density polyethylene jerrican containing the live vaccine concentrate at 1×10^10 CFU/mL, compliant with Ph. Eur. 0062 and Regulation (EU) 2019/6 Annex II. Field records from nursery operations indicate that immersion below 16°C produces weaker and more variable serological responses; published data for this specific strain in immersion challenge is limited beyond those production observations.

Where oral boostering is required after on-growing turbot have been moved to net-pens, the API is processed into a low-temperature feed top-dressing rather than compressed tablets or hard capsules. Tablet and capsule formats are not deployed in commercial turbot oral vaccination because uniform intake across a feeding shoal cannot be assured and the formulation energy required for compression or capsule filling would reduce live bacterial titre. The freeze-dried API is first pre-blended with pharmaceutical-grade sodium alginate binder at an API-to-binder ratio of 1:4 w/w in a jacketed ribbon blender with 3 mm rotor clearance, operating at 10 rpm for 180 s and maintained at 4°C. The pre-blend is de-agglomerated through a pin mill at 800 rpm and top-coated onto 3.5 mm extruded fish feed pellets at a rate of 1 kg blend per 1,000 kg feed, targeting 1×10^7 CFU/g of finished feed. Coating is carried out in a low-shear rotary coater at 4°C and 12 rpm using a 2% w/w fish gelatine solution sprayed through 350 µm nozzles at 0.35 MPa; coated granules are flash-cooled to 2°C and sieved between 1.0 mm and 4.0 mm to reduce separation during transport. A 10 g feed sample from each batch is homogenised in buffered peptone water and plated for live titre verification before bagging. The terminal finished product is a 20 kg foil-lined polyethylene bagged oral premix granule, stored and transported at 2–8°C. Compliance follows Regulation (EU) 2019/6 when the product is a veterinary premix; where a member state or non-EU jurisdiction classifies the top-dressing as a feed hygiene matter, Regulation (EC) 183/2005 applies to the feed operation.

Lyophilisation Robustness and Cold-Chain Powder Storage

The most sensitive downstream operation for this live vaccine API is lyophilisation because Edwardsiella tarda loses viability sharply when primary drying proceeds above -25°C or when residual moisture exceeds 2.0%. For stable dry powder supply, the bulk API is resuspended in a lyoprotectant matrix at a ratio of 1:3 w/v; the matrix consists of 5% w/v sucrose, 2% w/v sodium glutamate, and 1% w/v porcine gelatine in water for injection. Freeze-dryer shelves are pre-cooled to -40°C for 120 minutes; primary drying is held at -20°C ± 2°C and 0.100 mbar ± 0.020 mbar for 24 h; secondary drying is ramped from -20°C to 20°C over 12 h at 0.050 mbar. In-process viability is measured by plate count at 6-hour intervals, and validation batches exceeding 0.5 log CFU viability loss are rejected for downstream formulation release. The resulting cake is backfilled with dry nitrogen to a residual moisture of ≤2.0% and sealed in 10 mL Type I borosilicate vials with chlorobutyl stoppers and aluminium seals. Industrial transfer campaigns have shown that uncontrolled ice nucleation in pilot freeze dryers without controlled nucleation introduces batch-to-batch viability variation of up to ±0.7 log CFU; production dryers are therefore specified with a minimum shelf temperature uniformity of ±1°C and a vacuum control tolerance of ±0.020 mbar. Release testing is performed under Ph. Eur. 0062 and EU GMP Annex 1, with identity, live titre, moisture, and vacuum retention as core release parameters. The terminal product is a lyophilised powder for reconstitution into injectable, immersion, or oral formulations and is shipped at 2–8°C in insulated containers with validated 72 h cold-life.

When quarantine-introduced broodstock require immersion vaccination before release into RAS

For high-value turbot broodstock entering a recirculating aquaculture system, immediate injection may be withheld to reduce handling-associated stress, and a short-duration immersion protocol is applied within a segregated quarantine loop. The API concentrate is pre-diluted 1:50 v/v in a separate 50 L stainless-steel mixing vessel with gentle aeration at 1 L/min, then metered into a 500 L quarantine tank to a final bath titre of 2×10^8 CFU/mL. The quarantine tank is maintained at 16–18°C, pH 7.8–8.0, and ozone residual below 0.01 mg/L; pre-dilution water is temperature-matched to the main tank within ±0.5°C to avoid thermal shock. Exposure is capped at 90 s, and 20% of the water volume is exchanged immediately after exposure to reduce live vaccine release into the quarantine loop. The terminal product is a 10 L low-density polyethylene immersion solution container compliant with Regulation (EU) 2019/6 where used within the Union, and with national aquatic animal health import permits elsewhere. The quarantine loop segregation is maintained for 72 h to avoid transient nitrification disruption in the main RAS biofilter. Published data for this specific broodstock quarantine configuration is limited; therefore each batch is released with challenge-model potency data rather than field efficacy claims, and the user must verify compatibility with ongoing antibacterial treatments because live vaccine viability is suppressed by simultaneous oxytetracycline immersion.