Serum alkaline phosphatase levels accurately reflect cholestasis in mice

This work was supported by a grant (P24809‐B19) from the Austrian Science Foundation (to P.F.). All experiments have been approved by the local animal care and use committees according to criteria outlined by the National Academy of Sciences (BMWF‐66.010/0045‐II/10b/2010 and GZ‐BMWF‐66.010/0012‐II/3b/2014). Potential conflict of interest: Prof. Trauner consults for, advises, is on the speakers' bureau of, and received grants from Falk. He consults for and advises Albireo, Genfit, Intercept, and Phenex. He is on the speakers' bureaus of Gilead, MSD, and Roche. He received grants from Intercept and Albireo. To the Editor: Well‐characterized preclinical animal models are of utmost importance to explore the complex pathophysiology and novel treatment modalities in thus far difficult‐to‐treat cholangiopathies, such as primary sclerosing cholangitis (PSC) and primary biliary cirrhosis (PBC).1 However, there is an ongoing discussion regarding the suitability of serum tests to monitor cholestasis in such models. Therefore, we compared serum samples of four different well established mouse models for sclerosing cholangitis, including lithocholic acid (LCA)‐fed, 3,5‐diethoxycarbonyl‐1,4‐dihydrocollidine (DDC)‐fed mice as a model for toxic/chemically induced cholangitis, multidrug resistance protein‐2 (Mdr2; Abcb4) knockout mice (Mdr2−/−), representing a genetic model with characteristic features of sclerosing cholangitis,1 and common bile‐duct‐ligated mice (CBDL) in comparison to their respective controls. Aliquots of serum samples collected during harvesting for previous experiments have been kept frozen at −80°C. After 4‐fold dilution of sera, alanine aminotransferase (ALT), alkaline phosphatise (ALP), serum bilirubin, and serum bile acids (SBAs) were measured enzymatically on a Cobas 501 analyzer (Roche Diagnostics, Mannheim, Germany). For ALP, the reagent, ALP2 (ALP IFCC Gen.2.), was used according to the recommendations of the International Federation of Clinical Chemistry (IFCC).3 For statistical analyses, mouse models were summarized as toxic/chemically induced (4‐ and 7‐days LCA‐fed, and 1‐, 4‐, and 8‐weeks DDC‐fed mice), genetically induced (8‐weeks‐ and 4‐months‐old Mdr2−/− mice) and surgically induced cholestasis (CBDL for 7 days and 6 and 8 weeks) with respective controls for every group. Statistical analysis included Student t test and Spearman's correlation coefficient, using IBM SPSS Statistics 21 (IBM Corp., Armonk, NY). A P value <0.05 was considered significant. Serum parameters for each experimental group are reported as arithmetic means ± standard deviation (Table 1). With the exception of the DDC‐ and LCA‐fed group, where, in general, high standard deviations for serum liver tests were observed, ALP levels significantly correlated with SBA levels, which are highly specific and sensitive for cholestasis. We herein show that ALP is a sensitive parameter to accurately monitor cholestasis in mouse models of chemically, genetically, and surgically induced cholestatic liver injury. However, there is a wide variation in the degree of cholestasis, which is also reflected by variations in SBA levels, which follow closely those of ALP serum levels. Given that bile acid derivatives are increasingly tested as potential therapeutic agents in mouse models of cholestatic liver diseases, measurement of ALP might be seen as advantageous, because it is unlikely to be confounded by diet or treatment. In conclusion, ALP represents a specific marker to accurately assess cholestasis in mice. Table 1 - Serum Parameters of Mouse Models for Cholestasis Laboratory Parameters ALT (U/L) ALP (U/L) Bilirubin (mg/dL) SBA (μmol/L) Toxic/chemically induced cholestasis Controls (n = 4) 20 ± 9 113 ± 49 0.07 ± 0.04 5 ± 2 DDC feeding 1w (n = 3) 176 ± 121 (P = 0.155) 628 ± 594 (P = 0.272) 0.17 ± 0.13 (P = 0.297) 43 ± 30 (P = 0.160) DDC feeding 4w (n = 4) 344 ± 319 (P = 0.135) 807 ± 382 (P = 0.011)a 0.35 ± 0.17 (P = 0.020)a 76 ± 45 (P = 0.052) DDC feeding 8w (n = 4) 360 ± 423 (P = 0.206) 660 ± 662 (P = 0.197) 0.12 ± 0.06 (P = 0.194) 140 ± 115 (P = 0.100) LCA feeding 4d (n = 3) 1,001 ± 796 (P = 0.051) 613 ± 317 (P = 0.109) 0.80 ± 0.67 (P = 0.198) 200 ± 144 (P = 0.143) LCA feeding 7d (n = 4) 683 ± 532 (P = 0.088) 481 ± 249 (P = 0.027)a 0.20 ± 0.22 (P = 0.287) 82 ± 42 (P = 0.010)a Genetically induced cholestasis Mdr2 WT 8w (n = 4) 64 ± 9 114 ± 5 0.10 ± 0.05 9.5 ± 1.8 Mdr2 KO 8w (n = 4) 452 ± 163 (P = 0.017)a 235 ± 81 (P = 0.058) 0.14 ± 0.05 (P = 0.315) 64 ± 44 (P = 0.087) Mdr2 WT 4m (n = 4) 76 ± 13 73 ± 7 0.04 ± 0.03 11 ± 5 Mdr2 KO 4m (n = 4) 357 ± 139 (P = 0.007)a 196 ± 76 (P = 0.047)a 0.16 ± 0.06 (P = 0.010)a 41 ± 36 (P = 0.152) Surgically induced cholestasis SOP 7d (n = 4) 20 ± 9 20 ± 24 0.03 ± 0.04 7 ± 6 CBDL 7d (n = 4) 497 ± 256 (P = 0.010)a 337 ± 72 (P = 0.000)a 8.0 ± 1.4 (P = 0.002)a 583 ± 117 (P = 0.000)a SOP 6w (n = 4) 43 ± 4 79 ± 11 0.08 ± 0.03 13 ± 6 CBDL 6w (n = 4) 306 ± 397 (P = 0.277) 1,083 ± 668 (P = 0.057) 9.0 ± 8.2 (P = 0.116) 462 ± 349 (P = 0.042)a SOP 8w (n = 3) 48 ± 14 76 ± 11 0.04 ± 0.07 7 ± 3 CBDL 8w (n = 4) 119 ± 80 (P = 0.199) 1,374 ± 824 (P = 0.045)a 12 ± 8 (P = 0.054) 845 ± 692 (P = 0.096) Correlation Coefficients ALT (U/L) ALP (U/L) Bilirubin (mg/dL) All groups SBA, μmol/L 0.77b 0.85b 0.78b Toxic/chemically induced cholestasis SBA, μmol/L 0.59b 0.43 0.42 Genetically induced cholestasis SBA, μmol/L 0.89b 0.95b 0.26 Surgically induced cholestasis SBA, μmol/L 0.12 0.73b 0.41 Laboratory parameters are expressed means ± standard deviation; correlation coefficient according to Spearman for all groups (including control groups) and each experimental group (without respective controls).aP < 0.05, statistically significant difference between intervention and corresponding control group.bP < 0.01, statistically significant correlation between respective serum parameters.Abbreviations: WT, wild type; KO, knockout; SOP, sham operating procedure; d, days; w, weeks; m, months.