2.6.3. Ex-Vivo Liver Perfusion Protocol

The NMP-protocol lasted 150 min and divided into two phases. The first 30 min was called "rewarming", followed by 120 min of normothermic perfusion (Figure 3). In brief, after 30 min of static cold storage, the graft was connected to the NMP and the perfusion began. The portal flow was set at 5 mL/min and was increased by 5 mL/min every 5 min up to 30 mL/min or portal pressure of 8 cmH2O. At the beginning of the rewarming phase, the heat exchanger was set at 30 ◦C and the temperature was raised up to 40 ◦C to obtain a liver temperature of 37 ◦C after 30 min. During the normothermic phase temperature was maintained at 37 ◦C and portal flow remained unchanged.


**Figure 3.** Schematic overview of NMP protocol. PVF, portal vein flow.

#### *2.7. Perfusate Analysis*

Perfusate samples were collected soon after grafting ex-situ reperfusion and then every 30 min during the normothermic phase. The concentration of gas, metabolites, hyaluronan (HA), electrolytes, hemoglobin, and cells were evaluated. Pre-liver stopcock was used for perfusate sampling, while post liver perfusate was collected directly from IVC. Bile was collected by gravity directly into a test tube and weighted.

#### *2.8. Sample Processing and Analysis*

Perfusate samples were centrifugated at 1500 rpm for 15 min at 4 ◦C (Haereus Multifuge 33R, Thermo Fisher Scientific, Whaltham ). Supernatants were collected and stored at −20 ◦C for mediator concentration evaluation, whereas cell pellets were suspended in erythrocyte lysis buffer (0.155 M NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA, pH 7.4; all from MilliporeSigma, Frankfurter, Germany) and incubated for 10 min at 4 ◦C. A 10-min centrifugation at 2000 rpm was then performed. Recovered supernatants were used for free hemoglobin assessment, whereas pellets were suspended in 0.25 mL 13 PBS solution (MilliporeSigma, Frankfurter, Germany) for cell count and characterization.

Using a gas analyzer, acid-base balance (pH, pCO2, pO2), electrolytes (K<sup>+</sup>, Na+, Cl−, Ca2<sup>+</sup>), and metabolites (glucose and lactate) were evaluated (ABL 800 Flex; A. De Mori Strumenti, Milan, Italy).

At the beginning and end of the normothermic phase, perfusate samples were investigated for hepatocyte integrity (alanine-aminotransferase, ALT; aspartate-aminotransferase, AST; Lactate dehydrogenase, LDH), cholangiocellular damage (ALP, GGT), and blood count.

The concentration of total HA in outflow perfusate was determined by ELISA using a commercially available kit (sensitivity, 0.068 ng/mL; low standard, 0.625 ng/mL; R&D Systems, Minneapolis, MN, USA). The procedure was validated in preliminary experiments by using HA preparations with a known concentration and known average molecular mass (Select-HA HiLadder and LoLadder; Hyalose, Oklahoma City, OK, USA) [21–24].

Sodium Citrate concentration was measured by enzymatic assay using spectrophotometry (Citrate Assay Kit, Sigma-Aldrich, Saint Louis, USA). Millipore Amicon Ultra 0.5 mL 20 KDa were used to remove proteins. Thereafter, deproteinized samples were diluted 1:2 with citrate assay buffer and centrifugated at 14,000× *g*.
