Target-Cell-Directed Bioengineering Approaches for Gene Therapy of Hemophilia A

Abstract

Potency is a key optimization parameter for hemophilia A gene therapy product candidates. Optimization strategies include promoter engineering to increase transcription, codon optimization of mRNA to improve translation, and amino-acid substitution to promote secretion. Herein, we describe both rational and empirical design approaches to the development of a minimally sized, highly potent AAV-fVIII vector that incorporates three unique elements: a liver-directed 146-nt transcription regulatory module, a target-cell-specific codon optimization algorithm, and a high-expression bioengineered fVIII variant. The minimal synthetic promoter allows for the smallest AAV-fVIII vector genome known at 4,832 nt, while the tissue-directed codon optimization strategy facilitates increased fVIII transgene product expression in target cell types, e.g., hepatocytes, over traditional genome-level codon optimization strategies. As a tertiary approach, we incorporated ancient and orthologous fVIII sequence elements previously shown to facilitate improved biosynthesis through post-translational mechanisms. Together, these technologies contribute to an AAV-fVIII vector that confers sustained, curative levels of fVIII at a minimal dose in hemophilia A mice. Moreover, the first two technologies should be generalizable to all liver-directed gene therapy vector designs.

Keywords: AAV; codon optimization; factor VIII; hemophilia; promoter design; vector optimization.

Figure 1

Promoter Design and Testing (A) Novel, liver-directed synthetic promoters were constructed by the assembly of discrete units designed for high transcriptional activity in the context of the liver. Straight lines represent DNA of no known function. HNF, hepatocyte nuclear factor; AbpNat, α-microglobulin/bikunin precursor native sequence; AbpShort, human α-microglobulin/bikunin precursor shortened sequence; AFP, α-fetoprotein; TSS, transcription start site. (B) To determine the relative strength of promoter designs, promoters driving the expression of ET3 were transfected into HepG2 cells, and fVIII activity was measured 48 hr post-transfection. Activity is displayed relative to HLP (*p < 0.001 by one-way ANOVA and the Holm-Sidak method). (C) To determine the relative strength of the promoters in vivo, plasmid DNA was delivered by hydrodynamic injection to hemophilia A mice, and plasma fVIII activity was measured by fVIII chromogenic assay (n = 3–4 per group; *p < 0.001 by one-way ANOVA and the Holm-Sidak method). Data are presented as the mean ± sample SD.

Figure 2

Liver-Directed Codon Optimization (A) The fold difference in CUB of each codon in the liver CUB table with respect to the same codon in the human CUB table, as defined in formula 2. (B) The absolute difference codon usage bias of each codon in the liver CUB table versus the respective codon in the human CUB table, as defined by formula 3. (C) The correlation of the ratio of CUB for the set of liver genes to that of the entire human genomic coding sequence against the correlation of the ratio of CUB for ET3-LCO to that of ET3-NoCo. Points represent individual codon species. Open circles denote the 6 codon species with the lowest ratio of ET3-LCO to ET3-NoCo. Due to overlapping positions, two open circles appear as a single point. (D) Plasmid DNA encoding codon-optimized variants of ET3 were co-transfected into HepG2 cells with either GFP expression plasmid or a plasmid encoding the six tRNAs predicted to be most limiting of the expression of ET3-NoCo (*p < 0.05 by two-tailed t test).

Figure 3

Expression and Tissue Specificity of HCO versus LCO (A and B) An53-HCO and An53-LCO were transfected into (A) HepG2 cells and (B) BHK cells. fVIII activity from conditioned media was measured 48 hr after transfection (*p < 0.05 by two-tailed t test). (C) fIX-LCO and fIX-HCO plasmids were transfected into HepG2 cells supplemented with vitamin K. fIX activity was measured 48 hr after transfection (*p < 0.05 by two-tailed t test). Data are presented as the mean ± sample SD.

Figure 4

Expression and Tissue Specificity of LCO, MCO, and NoCo Designs (A and B) HepG2 cells were transfected with (A) ET3-LCO-, ET3-MCO-, and ET3-MCO-encoding plasmids and (B) HSQ-LCO-, HSQ-MCO-, and HSQ-NoCo-encoding plasmids. fVIII activity in the conditioned media was measured 48 hr after transfection. In (A), *p = 0.011 and **p = 0.006, and in (B), *p = 0.006 and **p = 0.005 by one-way ANOVA and the Holm-Sidak method. (C and D) BHK cells were transfected with (C) ET3-LCO-, ET3-MCO-, and ET3-MCO-encoding plasmids and (D) HSQ-LCO-, HSQ-MCO-, or HSQ-NoCo-encoding plasmids. In (C), *p < 0.001, and in (D), *p = 0.007 and **p = 0.004 by one-way ANOVA and the Holm-Sidak method. Data are presented as the mean ± sample SD.

Figure 5

In Vivo Expression of Codon-Optimized Designs (A–C) Shown here: (A) An53-HCO and An53-LCO; (B) ET3-LCO, ET3-MCO, and ET3-NoCo; and (C) HSQ-LCO, HSQ-MCO, and HSQ-NoCo plasmids were hydrodynamically injected into hemophilia A mice (n = 3–4 for all groups). Plasma levels of fVIII activity were measured 24 hr after transfection. In (A), *p < 0.05 by two-tailed t test, and in (B), *p = 0.05 and **p = 0.004 by one-way ANOVA and the Holm-Sidak method. NS, not significant. (D) Linearized ET3 and HSQ-LCO and NoCo were hydrodynamically injected into hemophilia A mice. Plasma levels of fVIII activity were measured for 6 weeks post-injection. Data are presented as the mean ± sample SD.

Figure 6

AAV Vector Delivery of Tissue Optimized Transgene Cassettes (A) Schematics of AAV2/8 transgene designs. Straight lines represent DNA sequence of no known function. (B) AAV2/8-HLP-An53-LCO and AAV2/8-HLP-An53-HCO were delivered intravenously to hemophilia A mice at a dose of 1 × 10¹¹ vg/kg (n = 4 per group). (C) AAV2/8 HCB-HSQ-LCO, AAV2/8-HCB-ET3-LCO, and AAV2/8-HLP-V3co were delivered intravenously to hemophilia A mice at a dose of 1 × 10¹¹ vg/kg (n = 3–4 per group). (D) A long-term, dose-finding experiment was performed with varying does of AAV2/8-HCB-MVM-ET3-LCO (n = 3 for all other doses). Data are presented as the mean ± sample SD.