Homologous recombination mediates stable Fah gene integration and phenotypic correction in tyrosinaemia mouse-model

Abstract

Aim: To stably correct tyrosinaemia in proliferating livers of fumarylacetoacetate-hydrolase knockout (Fah^-/-) mice by homologous-recombination-mediated targeted addition of the Fah gene.

Methods: C57BL/6 Fah^∆exon5 mice served as an animal model for human tyrosinaemia type 1 in our study. The vector was created by amplifying human Fah cDNA including the TTR promoter from a lentivirus plasmid as described. The Fah expression cassette was flanked by homologous arms (620 bp and 749 bp long) of the Rosa26 gene locus. Mice were injected with 2.1 × 10⁸ VP of this vector (rAAV8-ROSA26.HAL-TTR.Fah-ROSA26.HAR) via the tail vein. Mice in the control group were injected with 2.1 × 10⁸ VP of a similar vector but missing the homologous arms (rAAV8-TTR.Fah). Primary hepatocytes from Fah^-/- recipient mice, treated with our vectors, were isolated and 1 × 10⁶ hepatocytes were transplanted into secondary Fah^-/- recipient mice by injection into the spleen. Upon either vector application or hepatocyte transplantation NTBC treatment was stopped in recipient mice.

Results: Here, we report successful HR-mediated genome editing by integration of a Fah gene expression cassette into the "safe harbour locus" Rosa26 by recombinant AAV8. Both groups of mice showed long-term survival, weight gain and FAH positive clusters as determined by immunohistochemistry analysis of liver sections in the absence of NTBC treatment. In the group of C57BL/6 Fah^∆exon5 mice, which have been transplanted with hepatocytes from a mouse injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26.HAR 156 d before, 6 out of 6 mice showed long-term survival, weight gain and FAH positive clusters without need for NTBC treatment. In contrast only 1 out 5 mice, who received hepatocytes from rAAV8-TTR.Fah treated mice, survived and showed few and smaller FAH positive clusters. These results demonstrate that homologous recombination-mediated Fah gene transfer corrects the phenotype in a mouse model of human tyrosinaemia type 1 (Fah^-/- mice) and is long lasting in a proliferating state of the liver as shown by withdrawal of NTBC treatment and serial transplantation of isolated hepatocytes from primary Fah^-/- recipient mice into secondary Fah^-/- recipient mice. This long term therapeutic efficacy is clearly superior to our control mice treated with episomal rAAV8 gene therapy approach.

Conclusion: HR-mediated rAAV8 gene therapy provides targeted transgene integration and phenotypic correction in Fah^-/- mice with superior long-term efficacy compared to episomal rAAV8 therapy in proliferating livers.

Keywords: AAV8; Gene therapy; Liver based metabolic disease; Paediatric liver disease; ROSA26; Targeted integration.

Figure 1

Flowchart of treating mice. A: Vector map for rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR and for rAAV8-TTR.Fah. Fah cDNA is driven by the TTR promotor and for rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR located between the homologous arms of the Rosa26 locus. The vector was cloned into an AAV backbone; B: Scheme for the in vivo experiments. First-generation mice (C57BL/6 FAH^∆exon5 strain) were injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR (group 1, n = 4) or rAAV8-TTR.Fah (control group, n = 5). The NTBC treatment was stopped, and after 45 d, a partial hepatectomy was performed. In each group, one mouse was used as the donor for hepatocyte transplantation into C57BL/6 Fah^∆exon5 mice. These recipients were the second generation of mice in our study. NTBC treatment was discontinued after hepatocyte transplantation. TTR: Transthyretin promoter (liver specific); R26 HAL: Homologous arm left for target locus in Rosa26; HAR: Homologous arm right for target locus in Rosa26; ITR: Inverted terminal repeat.

Figure 2

Mice treated with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR. A: Weight graph and survival for first-generation mice (n = 4) injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR and 3 untreated controls (injected with sodium chloride). Continuous line = rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR mice, broken line = controls (same control mice as displayed at Figure 3A). Body weight is displayed as percentage of body weight at the time of virus injection or sodium chloride injection (controls). The timeline (x-axis) is displayed in days beginning with the day of virus/sodium chloride injection as day zero; B: FAH staining of liver tissue from controls (mice with sodium chloride injection) after death (100 × magnification); C: FAH staining of liver tissue from partial hepatectomy in mouse injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR (100 × magnification); D: Weight graph and survival for second-generation mice (continuous line), which were transplanted with one million hepatocytes from mice primarily injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR and controls (same control mice as displayed at Figure 3D) without hepatocyte transplantation (broken line). Body weight is displayed as percentage of body weight at time of hepatocyte transplantation. The timeline (x-axis) is displayed in days, beginning with the day of hepatocyte transplantation as day zero; E: FAH staining of liver tissue from a partial hepatectomy from a second-generation mouse, which received one million hepatocytes from mice primarily injected with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR (100 × magnification). 1/3PH: One third partial hepatectomy; HcTx: Hepatocyte transplantation.

Figure 3

Mice treated with rAAV8-TTR.Fah. A: Weight graph and survival for first-generation mice injected with rAAV8-TTR.Fah (n = 5) and 3 untreated controls (injected with sodium chloride). Continuous line = rAAV8-TTR.Fah mice, broken line = controls (same control mice as displayed at Figure 2A). Body weight is displayed as percentage of body weight at the time of virus injection or sodium chloride injection (controls). The timeline (x-axis) is displayed in days beginning with the day of virus/sodium chloride injection as day zero; B: FAH staining of liver tissue from controls (mouse with sodium chloride injection) after death (100 × magnification); C: FAH staining of liver tissue from a partial hepatectomy from a mouse injected with rAAV8-TTR.Fah (100 × magnification); D: Weight graph and survival for second-generation mice (continuous line), which were transplanted with one million hepatocytes from mice primarily injected with rAAV8-TTR.Fah and controls (same control mice as displayed at Figure 2D) without hepatocyte transplantation (broken line). Body weight is displayed as percentage of body weight at time of hepatocyte transplantation. The timeline (x-axis) is displayed in days, beginning with the day of hepatocyte transplantation as day zero; E: FAH staining of liver tissue from a partial hepatectomy in a second-generation mouse, which received one million hepatocytes from mice primarily injected with rAAV8-TTR.Fah (100 × magnification); F: FAH staining of liver tissue from a partial hepatectomy from the single second-generation mouse that showed cluster and weight gain.

Figure 4

Integration PCR gel electrophoresis. A representative gel picture from the analyses of genomic liver DNA, that was extracted from snap-frozen liver tissue harvested between 60-70 d after hepatocyte transplantation. Primers were located in the Rosa26 locus and in the FAH sequence of the donor DNA. Product could only be amplified if targeted integration occurred. The expected length of the PCR amplicon was 1107 bp. The PCR product was analysed utilizing agarose gel electrophoresis.

Figure 5

Sybr green qRT-PCR. Shown are the 2^-ddct values of two mice treated with rAAV8-ROSA26.HAL-TTR.Fah-ROSA26HAR and one mouse treated with rAAV8-TTR.Fah calculated on an untreated control mouse.