10 degree C static storage of porcine donation after circulatory death livers improves biliary viability and mitigates ischemia-reperfusion injury

Abstract

Optimized static cold storage has the potential to improve the preservation of organs most vulnerable to ischemia-reperfusion injury. Data from lung transplantation suggest that storage at 10 °C improves mitochondrial preservation and subsequent allograft function compared with conventional storage on ice. Using a porcine model of donation after circulatory death, we compared static storage of livers at 10 °C to ice. Livers (N = 5 per group) underwent 10 hours of storage followed by 4 hours of normothermic machine perfusion (NMP) for real-time allograft assessment. Allografts were compared using established NMP viability criteria, tissue immunostaining, and tissue metabolomics. Livers stored at 10 °C demonstrated lower portal venous vascular resistance and greater hepatic artery vasoresponsiveness. Lactate clearance during NMP was similar between the groups. Livers stored at 10 °C showed favorable biochemical parameters of biliary viability, including greater bile volume, pH, and bicarbonate. Metabolomics analysis revealed increased aerobic respiration, improved electron transport chain function, and less DNA damage after reperfusion of livers stored at 10 °C. Static storage of donation after circulatory death livers with extended cold ischemic time at 10 °C demonstrates superior allograft function with evidence of improved biliary viability and mitochondrial function compared with ice. These data suggest that storage at 10 °C should be considered for translation to clinical practice.

Keywords: 10 °C static storage; biliary viability; donation after circulatory death; ischemia-reperfusion injury; liver transplantation; organ preservation.

Figure 1.

Experimental overview. (A) Schematic of experimental design with N = 5 per experimental group. (B) Backbench preparation of the porcine liver with cannulas placed in the inferior vena cava (IVC), hepatic artery (HA), portal vein (PV), and common bile duct (CBD). (C) Components of the normothermic machine perfusion platform. Q, flow probe; P, pressure sensor. Schematic created using BioRender.com.

Figure 2.

Evaluation of vascular viability during normothermic machine perfusion. (A) Portal vein (PV) vascular resistance measured throughout normothermic machine perfusion (NMP) with corresponding calculation of the area under the curve. N = 5 in each experimental group. (B, C) Assessment of α-1 adrenergic responsiveness after 4 hours of NMP. Hepatic artery (HA) flow (B) and vascular resistance (C) after administration of 4 mg phenylephrine. Percent change from baseline is reported in adjacent graphs for each parameter. Data are presented as mean ± SD. For B and C, N = 5 for ice and N = 4 for 10 °C groups given flow probe malfunction in 1 experiment. *P < .05, ***P < .001.

Figure 3.

Hepatocellular viability assessed over 4 hours of normothermic machine perfusion. (A, B) Serum lactate measured during normothermic machine perfusion (NMP) (A) with calculation of lactate clearance between 0 and 4 hours of NMP (B). (C) Aspartate and alanine aminotransferase (AST, ALT) measured at start and end of NMP. (D) Total bicarbonate added to perfusate during NMP to maintain pH > 7.25. (E, F) Serum potassium (E) and glucose consumption (F) throughout NMP. Data are presented as mean ± SD. N = 5 in each experimental group. All biomarkers reflect values taken from the inferior vena cava.

Figure 4.

Evaluation of biliary functional and regenerative capacity. (A-C) Biochemical evaluation of bile produced during normothermic machine perfusion (NMP) including rate of bile production (A), bile pH (B), and bile bicarbonate (C). N = 5 for 10 °C group, N = 4 for ice group after excluding 1 liver due to hemobilia. (D, E) Immunofluorescence staining of the bile duct after 4 hours of NMP. (D) Identification of the peribiliary vascular plexus through staining of endothelial marker CD31 with quantification of CD31⁺ cells. (E) Evaluation of progenitor and mature cholangiocytes through staining of transcription factor SOX9 and mature biliary epithelium marker CK7, respectively. Insets highlight evaluation of co-expression of SOX9 and CK7 in the peribiliary glands with quantification of cells expressing both markers. N = 5 per experimental group for D and E. Data are presented as mean ± SD. Representative images for Livers 1–5 in each group are in Supplementary Figures 5 and 6; negative immunofluorescence controls are shown in Supplementary Figure 9. *P < .05, **P < .01, ****P < .0001. Scale bar = 100 μm.

Figure 5.

Differentially abundant tissue metabolites in livers perfused after 10 °C storage compared with ice. (A) Heat map of metabolites from liver parenchymal tissue with greatest difference in abundance between ice and 10 °C storage groups after 4 hours of normothermic machine perfusion. (B) Relative abundance of metabolites related to glycolysis, tricyclic acid (TCA), glutathione, and malonate metabolism. (C) Correlation of dimethylmalonic acid, precursor to succinate dehydrogenase inhibitor malonate, with anaerobic metabolite lactate and glycolysis metabolite phosphoenolpyruvate. Data are presented as mean ± SD. N = 5 in each experimental group. *False discovery rate adjusted P < .05.

Figure 6.

Correlation of metabolic changes and real-time assessment during normothermic machine perfusion. (A) Potential mechanism of observed metabolic differences between livers stored on ice compared to 10 °C: after reperfusion, impaired ETC function leads to tricyclic acid (TCA) cycle dysfunction with decreased aerobic metabolism of acetyl coA and increased reliance on anaerobic glycolytic pathways. (B) Peak oxygen consumption achieved during normothermic machine perfusion (NMP). (C) Correlation of tissue lactate with bicarbonate requirement to maintain pH > 7.25 during NMP and serum lactate at end of NMP. Data are presented as mean ± SD. N = 5 in each experimental group. *P < .05.

Figure 7.

DNA damage after reperfusion of livers stored on ice compared with 10 °C. (A) Metabolites of nucleotide breakdown that were significantly greater within liver parenchyma after reperfusion of livers stored on ice compared to 10 °C. (B, C) Representative images of immunohistochemistry for terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) after 4 hours of NMP (B) with corresponding quantification (C). (D) Correlation of nucleotide metabolites and TUNEL⁺ tissue staining. Data are presented as means ± SD. N = 5 in each experimental group. Representative images for Livers 1–5 in each group are in Supplementary Figure 8. *False discovery rate adjusted P < .05; **P < .01. Scale bar = 100 μm.