0.45 μm syringe filter

100 mg of pulverized tissue from each sample was vortexed with 1 mL dH₂O for 30 seconds. The particulate was removed using 25 mm 0.45 μm syringe filters (VWR) and filtered liquid transferred into HPLC vials (Thermo Scientific). Each sample was analyzed using Bio-Rad Aminex HPX-87H column (300 mm × 7.8 mm, 9 µm particle size; Bio-Rad Laboratories, Hercules, California, USA), attached to a Varian Pro Star 230 HPLC system (Varian Inc., California, USA) equipped with UV and Refractive Index (RI) detectors. The column and RI detector were maintained at 65 °C and 55 °C, respectively. For each sample, the injection volume was set at 10 µL and 0.005 M sulfuric acid solution was used as an eluent. The samples were eluted at a flow rate of 0.6 mL/min over 50 min. Organic acids (citric and malic) were detected using UV detector at 210 nm and sugars (glucose and fructose) were detected using the RI detector. Different organic acids and sugars were identified by comparing the retention time of the peaks with known individual standards (Sigma Aldrich, St. Louis, Missouri, USA) run under the same conditions. Quantification was achieved using the external standard method.

ZIKV strain MR766 was obtained from ATCC and propagated in Vero cells for 4 days. Viral supernatant was filtered via 0.45 μM syringe filters (VWR). Virus titers were determined by a standard plaque assay as described previously^{49 (link),50 (link)}. Human fetal brain slice culture was cultured for 4 days before incubating with the indicated PFUs of ZIKV. Two hours later, tissues were washed once with phosphate buffered saline (PBS) and maintained in regular culture medium. Infected tissues were fixed for further image studies 48 hours after infection. For primary human brain monolayer infection, cells were incubated with the indicated PFUs of ZIKV for 2 hours. Cells were then washed with PBS once and maintained in regular medium. Previous studies have demonstrated that high ZIKV titers could be detected in fetal brain tissues, placenta and umbilical cords^{11 (link),13 (link)}. In these studies, real-time PCR was extensively used to determine viral titers in human maternal and fetal samples. Although the standard plaque assay was used for viral titering, we have standardized ZIKV titers to RNA copy numbers by real-time PCR. The titers use in our studies were comparable to those measured in human fetal brains.

Enriched salmon juice was used as a growth medium for cocultivation of LAB and target strains. Although not directly comparable to the conditions in a salmon fillet, salmon juice was chosen as a growth medium to mimic the nutritional composition of the fish. The juice was prepared from fresh salmon loins obtained from a local retailer, as described by Wiernasz, Cornet, Cardinal, Pilet, Passerini, and Leroi [27 (link)]. In brief, 500 g of salmon loin was blended with 1 L of distilled water, boiled for 2 min, filtered through a 185 mm folding filter (Schleicher & Schuell, Dassel, Germany), and sterilized at 100 °C for 30 min. Sterile salmon juice was stored at −40 °C. Before use, 90 mL of salmon juice was supplemented with 10 mL of 1 M K₂HPO₄/KH₂PO₄ buffer solution (Merck, Darmstadt, Germany) at pH 6.7, 1 g of D-glucose (Merck, Oslo, Norway), and 1.5 g of NaCl (VWR, Leuven, Belgium) to prevent growth inhibitory conditions in the model system. After enrichment, the medium was sterilized through a 0.45 μm syringe filter (VWR, Oslo, Norway).