Therapeutic liver repopulation by transient acetaminophen selection of gene-modified hepatocytes

Abstract

Gene therapy by integrating vectors is promising for monogenic liver diseases, especially in children where episomal vectors remain transient. However, reaching the therapeutic threshold with genome-integrating vectors is challenging. Therefore, we developed a method to expand hepatocytes bearing therapeutic transgenes. The common fever medicine acetaminophen becomes hepatotoxic via cytochrome p450 metabolism. Lentiviral vectors with transgenes linked in cis to a Cypor shRNA were administered to neonatal mice. Hepatocytes lacking the essential cofactor of Cyp enzymes, NADPH-cytochrome p450 reductase (Cypor), were selected in vivo by acetaminophen administration, replacing up to 50% of the hepatic mass. Acetaminophen treatment of the mice resulted in over 30-fold expansion of transgene-bearing hepatocytes and achieved therapeutic thresholds in hemophilia B and phenylketonuria. We conclude that therapeutically modified hepatocytes can be selected safely and efficiently in preclinical models with a transient regimen of moderately hepatotoxic acetaminophen.

Fig. 1.. Principles of APAP metabolism and APAP selection.

(A) Pathways of APAP metabolism. Cypor is required for the metabolism of APAP to the hepatotoxic metabolite NAPQI. N-acetylcysteine (NAC) detoxifies NAPQI. (B) Mechanism of APAP selection. Disruption of Cypor renders hepatocytes resistant to the hepatotoxic effects of APAP, resulting in a selective advantage with APAP administration. RNAi, RNA interference.

Fig. 2.. APAP selection of Cypor-deficient hepatocytes.

(A) Schematic of the pX330-Cypor gRNA vector. (B) ALT in APAP-treated mice 6 hours after APAP injection for mice treated with pX330-Cypor gRNA (n = 10) or negative control (n = 8). (C) Percent Cypor-deficient hepatocytes measured by in-del analysis of whole-liver homogenate gDNA in mice that received a negative control pX330 vector lacking a gRNA (n = 3), pX330-Cypor gRNA and no APAP (n = 7), or pX330-Cypor gRNA with APAP treatment (n = 5). (D) Cypor immunofluorescent staining (red) in liver sections from mice that received pX330-Cypor gRNA with or without APAP treatment. Arrow indicates zone 3 expression of Cypor. Scale bars, 200 µm. DAPI, 4′,6-diamidino-2-phenylindole. (E) Top: Cypor immunofluorescent staining and GFP expression in liver sections from mice treated with a Cypor shRNA GFP-expression transposon vector and APAP selection (n = 3). Bottom: Negative control: Mouse treated with a transposon expressing an shRNA against an unrelated gene (Hpd). Scale bars, 100 µm. Data are means ± SD. *P < 0.05 and **P < 0.01 by Mann-Whitney U test.

Fig. 3.. Selection of hF9-expressing lentiviral vectors.

(A) Schematic of the lentiviral vector containing a U6-driven Cypor gRNA and hF9 driven by the liver-specific mTTR promoter (LV-hF9-Cypor gRNA). LTR, long terminal repeat; WPRE, woodchuck hepatitis virus posttranscriptional regulatory element. (B) hF9 concentrations in transgenic Cas9 mice treated with LV-hF9-Cypor gRNA over the course of APAP selection. (C) Quantification of Cypor-deficient hepatocytes by in-del analysis in APAP-selected (n = 4) and unselected (n = 4) Cas9 mice treated with LV-hF9-Cypor gRNA. (D) Cypor and hF9 liver immunohistochemistry in Cas9 mice treated with LV-hF9-Cypor gRNA followed by APAP selection. Scale bars, 100 µm. (E) Schematic of the lentiviral vector expressing Cypor shRNA and hF9 (LV-hF9-Cypor shRNA). (F) hF9 concentrations in wild-type mice (n = 4) treated with LV-hF9-Cypor shRNA during APAP selection compared to unselected mice (n = 4). (G) Cypor and hF9 liver immunohistochemistry in wild-type mice treated with LV-hF9-Cypor shRNA followed by APAP selection. Scale bars, 100 µm. Data are means ± SD. *P < 0.05 by Mann-Whitney U test.

Fig. 4.. APAP selection in a PKU mouse model.

(A) Schematic of selectable lentiviral vector expressing human PAH. (B) Serum Phe concentrations in female and male Pah^enu2/enu2 mice. All mice received a selectable PAH-expressing lentiviral vector as neonates, followed by either APAP selection (female n = 3, male n = 4) or no further treatment (female n = 3, male n = 3). Dashed line represents the therapeutic threshold of 360 µM. (C) Coat color comparison of wild-type (WT), APAP-selected Pah^enu2/enu2, and unselected Pah^enu2/enu2 mice. (D) Pah enzyme activity in liver homogenate as percentage of wild type. (E) Clonal expansion of Cypor-deficient hepatocytes in liver from a selected mouse. Scale bars, 200 µm. (F) Hematoxylin and eosin staining of liver sections in APAP-selected mice. Data are means ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s two-tailed t test assuming equal variance.

Fig. 5.. Safety and stability of APAP selection.

(A) Final blood hF9 concentrations and estimated percent Cypor-deficient hepatocytes in fully (n = 4) and partially (n = 13) APAP-selected mice. Dashed line represents the therapeutic hF9 threshold. (B) Representative Cypor and Cyp2E1 immunohistochemistry in a partially selected mouse (selection halted at 3181 ng/ml of blood hF9). Dotted lines surround Cypor-positive regions. Scale bars, 200 µm. (C) Experimental timeline to test long-term stability of Cypor-deficient hepatocytes. (D) Estimated percent Cypor-deficient hepatocytes in mice that received continued weekly APAP (n = 4) or no continued APAP (n = 4) over a 42-week period. (E) Representative Cypor immunohistochemistry 42 weeks after the last APAP dose. Dotted lines surround Cypor-negative regions. Scale bars, 200 µm. (F) ALT concentrations (6 hours after injection) in mice receiving pX330-Cypor gRNA and either intraperitoneal APAP (n = 19) or on an APAP-containing diet (measured 4 hours into the light cycle) (n = 10). (G) Percent Cypor-deficient hepatocytes in mice treated with pX330-Cypor gRNA followed by no further treatment (n = 7), intraperitoneal APAP injections (n = 5), or APAP diet (n = 13). (H) Representative Cypor immunohistochemistry in a mouse treated with APAP diet for 6 weeks showing areas of Cypor-negative hepatocytes. Scale bars, 200 µm. Data are means ± SD. **P < 0.01 and ***P < 0.001 by Mann-Whitney U test.