Lifetime correction of genetic deficiency in mice with a single injection of helper-dependent adenoviral vector

Abstract

Ideally, somatic gene therapy should result in lifetime reversal of genetic deficiencies. However, to date, phenotypic correction of monogenic hyperlipidemia in mouse models by in vivo gene therapy has been short-lived and associated with substantial toxicity. We have developed a helper-dependent adenoviral vector (HD-Ad) containing the apolipoprotein (apo) E gene. A single i.v. injection of this vector completely and stably corrected the hypercholesterolemia in apoE-deficient mice, an effect that lasted the natural lifespan of the mice. At 2.5 years, control aorta was covered 100% by atherosclerotic lesion, whereas aorta of treated mice was essentially lesion-free. There was negligible toxicity associated with the treatment. We also developed a method for repeated HD-Ad vector administration that could be applied to organisms, e.g., humans, with life spans longer than 2-3 years. These studies indicate that HD-Ad is a promising system for liver-directed gene therapy of metabolic diseases.

Figure 1

Structures of Ad vectors. L-ITR and R-ITR, left and right Ad inverted terminal repeat sequence, respectively; HPRT, intron region of human genomic hypoxanthine phosphoribosyltransferase stuffer sequence; C346, cosmid C346 human genomic stuffer sequence (8); Ψ, Ad packaging sequence; CMV-P, cytomegalovirus promoter; PCK-P, phosphoenolpyruvate carboxykinase promoter; m-cE, mouse apoE cDNA; m-gE, mouse genomic apoE DNA; h-Aint, human apoA-I intron 1; b-G(A), bovine β-globin polyadenylation signal; SV(A), SV40 polyadenylation signal.

Figure 2

Comparison of FG-Ad and different HD-Ads. (a) Restriction mapping of shuttle and HD-Ad vectors. Lane 1, PmeI-digested pΔ28-cE; lane 2, HD-Ad-cE; lane 3, PmeI-digested pΔ21-gE; lane 4, HD-Ad5-gE (serotype 5); lane 5, HD-Ad2-gE (serotype 2). (b) Southern blot analysis of helper virus contamination. Lane 1, helper virus serotype 5; lane 2, PmeI-digested pΔ28-cE; lane 3, HD-Ad-cE; lane 4, PmeI-digested pΔ21-gE; lane 5, HD-Ad5-gE; lane 6, HD-Ad2-gE; lane 7, helper virus serotype 2. The upper and lower bands correspond to the left arm and the right arm, respectively. (c) Plasma cholesterol levels in apoE^−/− mice following injections of Ad vectors. Cholesterol levels in apoE Ad-treated groups were significantly lower than those in the DB group with a P < 0.001 except the following: HD-Ad5-cE at 140 days (P < 0.01) and FG-Ad5-cE at 84 days (P < 0.05) and after 84 days (not significant). (d) Western blot of plasma apoE in different experimental groups. Treatment groups are identified in the left column. (e) Plasma anti-Ad5 neutralizing antibody titers. *, P < 0.001.

Figure 3

Readministration of Ad vectors of different serotypes. (a) Plasma cholesterol levels following readministration of Ad vectors of different serotypes. *, P < 0.001. (b) Plasma apoE levels measured by ELISA. Arrowhead, vector readministration; ■, plasma apoE in C57BL/6 mice (4.4 ± 0.7 mg/dl). (c) Western blot analysis of plasma apoE in mice treated by different Ad vectors. (d) Plasma anti-Ad5 neutralizing antibody titer following readministration of HD-Ads of different serotypes. Closed bars, FG-Ad5-cE followed by HD-Ad5-gE (n = 5); open bars, FG-Ad5-cE followed by HD-Ad2-gE (n = 5). *, P < 0.005 (FG-Ad5, 0 vs. 28 days); **, P < 0.002 (FG-Ad5, 0 vs. 28 days); and ***, P < 0.01 (FG-Ad5 at 112 days vs. HD-Ad5–2nd injection at 140 days).

Figure 4

Lifetime correction of hypercholesterolemia in apoE^−/− mice by HD-Ads. (a) Plasma cholesterol levels in mice injected with different Ads. (Upper) Days after birth; (Lower) days after injection. ●, HD-Ad5-gE (1.5 × 10¹¹ particles per mouse, n = 10); ○, HD-Ad[5-cE → 2-gE], HD-Ad5-cE (1.5 × 10¹¹ particles per mouse) followed by HD-Ad2-gE (1.0 × 10¹¹ particles per mouse) (n = 10); ▾, FG-Ad5-cE (1 × 10¹¹ particles per mouse, n = 10); ▿, DB control (n = 10); ■, plasma cholesterol level of normal C57-wt mice (n = 10). Boldface arrow, reinjection of HD-Ad2-gE. During the 2.5-year experimental period, mice died of different causes as denoted by the symbols above the cholesterol values. Similar symbols were used to denote the mouse groups as those used for plasma cholesterol. Number of deaths are as indicated; causes of death were as follows: *, accidental death (accidents during handling/bleeding); small arrows, anesthetic death (especially in the course of changing anesthetic agent from methoxyflurane to isoflurane); and crosses, natural deaths. (b) Plasma apoE levels. ●, HD-Ad5-gE; ○, HD-Ad[5-cE → 2-gE]; ▾, FG-Ad5-cE; and ■, C57-wt. Bold arrow, reinjection of HD-Ad2-gE. (c) Plasma liver enzyme activities. (Upper) Plasma aspartate aminotransferase and alanine aminotransferase activities after an initial injection. ●, HD-Ad5-gE; ○, HD-Ad5-cE; ▾, FG-Ad5-cE; and ▿, DB control. (Lower) Plasma aspartate aminotransferase and alanine aminotransferase activities after readministration of HD-Ad2-gE (n = 5) in HD-Ad5-cE-injected mice. DB, mice reinjected with dialysis buffer (n = 5). *, P < 0.001; **, P < 0.005. (d) FPLC profile of total plasma cholesterol in apoE^−/− mice after injection of different Ad vectors. The individual lipoprotein classes are very-low density lipoprotein and remnants (nos. 3–8), intermediate and low-density lipoproteins (nos. 9–18), and HDL (nos. 22–27). HDL profiles are magnified inside a window in each panel. ●, HD-Ad5-gE; ○, HD-Ad5-cE; ▾, FG-Ad5-cE; and ▿, DB control.

Figure 5

Tissue analyses at 817 days after HD-Ad injection (age 901 days). (a) Tissue distribution of vector-derived DNA, apoE mRNA, and protein. L1, liver of C57-wt; L2, liver of DB control; L3, liver of HD-Ad[5-cE → 2-gE]-treated mice. Lanes 1–7 are from the HD-Ad5-gE treated mice. Lane 1, liver; lane 2, spleen; lane 3, lung; lane 4, kidney; lane 5, brain; lane 6, small intestine; lane 7, muscle. (b) Immunohistochemistry of liver sections 817 days after HD-Ad treatment. (c) Aortas of HD-Ad-treated and DB control mice stained with Oil Red O.