Efficient intrathymic gene transfer following in situ administration of a rAAV serotype 8 vector in mice and nonhuman primates

Abstract

The thymus is the primary site of T-cell development and plays a key role in the induction of self-tolerance. We previously showed that the intrathymic (i.t.) injection of a transgene-expressing lentiviral vector (LV) in mice can result in the correction of a T cell-specific genetic defect. Nevertheless, the efficiency of thymocyte transduction did not exceed 0.1-0.3% and we were unable to detect any thymus transduction in macaques. As such, we initiated studies to assess the capacity of recombinant adeno-associated virus (rAAV) vectors to transduce murine and primate thymic cells. In vivo administration of AAV serotype 2-derived single-stranded AAV (ssAAV) and self-complementary AAV (scAAV) vectors pseudotyped with capsid proteins of serotypes 1, 2, 4, 5, and 8 demonstrated that murine thymus transduction was significantly enhanced by scAAV2/8. Transgene expression was detected in 5% of thymocytes and, notably, transduced cells represented 1% of peripheral T lymphocytes. Moreover, i.t. administration of scAAV2/8 particles in macaques, by endoscopic-mediated guidance, resulted in significant gene transfer. Thus, in healthy animals, where thymic gene transfer does not provide a selective advantage, scAAV2/8 is a unique tool promoting the in situ transduction of thymocytes with the subsequent export of gene-modified lymphocytes to the periphery.

Figure 1

Influence of encapsidated genome configuration and capsid serotype on in vivo thymus transduction efficiency. Single-stranded adeno-associated virus (ssAAV) and self-complementary AAV (scAAV) vectors based on the AAV2 backbone and expressing the enhanced green fluorescent protein (eGFP) transgene were pseudotyped with capsid proteins of serotypes 1, 2, 4, 5, and 8. 1 × 10¹⁰ vector genomes (vg)–containing particles of the indicated ss and scAAV vectors were injected directly into the thymi of 3- to 4-week-old mice. Thymocytes were harvested from killed animals 30 days following AAV administration. In vivo expression of eGFP was assessed by quantitative reverse transcription–PCR in triplicate samples and all data were normalized to hypoxanthine-guanine phosphoribosyl transferase (HPRT). Normalized means ± SD of arbitrary units (AU) are shown.

Figure 2

Kinetics of transgene expression and phenotype of transduced cells following in vivo intrathymic administration of scAAV2/8. (a) Mice were injected intrathymically with 4 × 10¹¹ vector genomes–containing scAAV2/8 particles. Thymi were harvested from killed animals at days 3, 10, and 30 postinjection and the percentages of eGFP⁺ thymocytes were assessed by flow cytometry. The means ± SD of three animals are shown for each time point (b) Flow cytometry data showing the percentages of eGFP⁺ cells as a function of side scatter (SSC) are presented at days 0, 3, 10, and 30 (top row). The phenotype of the eGFP⁺ cells was assessed as a function of CD4 and CD8 expression, allowing double-negative (CD4⁻/CD8⁻), double-positive (CD4⁺/CD8⁺) and mature single-positive cells (CD4⁺, CD8⁺) to be distinguished (middle row). Double-negative (DN) cells were further phenotyped on the basis of CD44 and CD25 expression distinguishing DN1 (CD44⁺CD25⁻), DN2 (CD44⁺CD25⁺), DN3 (CD44⁻CD25⁺), and DN4 (CD44⁻CD25⁻) subsets (bottom row). All dot plots are compared to the total thymocyte profiles (left panels). (c) Immunohistochemical analyses of thymic stromal cells from mice injected with scAAV2/8 vectors 10 days previously. Enhanced green fluorescent protein (eGFP) expression was evaluated in the medulla (A and D). Epithelial and dendritic cells were visualized by staining with a wide-spectrum anti-CTK pAb (B), and an anti-S100 pAb (E), respectively. Overlays of eGFP with CTK and S100 immunostaining are shown in C and F, respectively. Arrows indicate colocalization. Bar = 50 µm. PBS, phosphate-buffered saline; scAAV, self-complementary adeno-associated virus.

Figure 3

Thymocytes transduced by in situ injection of a scAAV2/8 vector differentiate and migrate to the periphery. (a) Expression of the enhanced green fluorescent protein (eGFP) transgene was monitored in the lymph nodes (LNs) and spleens of mice intrathymically injected with the scAAV2/8 vector. Expression was assessed by quantitative reverse transcription–PCR at days 3, 10, and 30, and was normalized to hypoxanthine-guanine phosphoribosyl transferase. Normalized means ± SD of arbitrary units (AU) are shown for duplicate samples. (b) The percentages of LN and spleen cells expressing the eGFP transgene were also monitored by flow cytometry. The means ± SD of three animals are shown for each time point. (c) Following intrathymic scAAV2/8 administration, the phenotypes of transduced lymphocytes was monitored in LNs by assessing eGFP expression in CD3 and CD19 populations as compared to mock-transduced mice (phosphate-buffered saline) (upper panels). The expression of the eGFP transgene within the CD3 subset was further evaluated by CD4 and CD8 staining. The relative ratio of CD4 and CD8 cells within the nontransduced (eGFP⁻) and transduced (eGFP⁺) fractions are shown in representative dot plots (bottom panels). scAAV, self-complementary adeno-associated virus.

Figure 4

In vivo administration of scAAV2/8 in a macaque thymus results in thymocyte transduction. (a) Three-year-old cynomolgus macaques were injected intrathymically in the right lobe with 2 × 10¹³ vector genomes–containing scAAV2/8 particles in two or three points. Ten days following injection, thymi were removed under general anesthesia and four sections were made from each lobe, shown here in the schematic diagram and labeled as A to D. (b) Transduction efficiencies in the various sections were assessed by quantitative PCR of the enhanced green fluorescent protein (eGFP) transgene. All values were normalized to Epo and the mean of eGFP copies per genomic DNA ± SD are shown for samples from the two macaques. (c) Flow cytometry data showing the percentages of eGFP⁺ cells as a function of side scatter (SSC) in the section B and A from right (right panels) and left (left panels) lobes, respectively, from a representative macaque injected 10 days previously with the scAAV2/8 vector (top row). The phenotypes of eGFP⁺ thymocytes in comparison to the total thymocyte population are shown as a function of CD4 and CD8 expression. scAAV, self-complementary adeno-associated virus.

Figure 5

In situ thymus injection of a scAAV2/8 vector in macaques results in the differentiation and migration of transduced thymocytes to peripheral lymphoid organs. (a) Expression of the enhanced green fluorescent protein (eGFP) transgene was monitored following in vivo intrathymic administration of a scAAV2/8 vector in macaques (right lobe). Expression was assessed by quantitative PCR in lymph nodes and spleen 30 days following scAAV2/8 injection. Expression was normalized to Epo and means of eGFP copies per genomic DNA ± SD are shown for samples from the two macaques. (b) Localization of eGFP⁺ cells in splenic corpuscles. A, HES staining with: GC, germinal center; M, mantle; CA, central arteriole. B, eGFP⁺ cells are located in the GC. C, CD4 labeling. D, merge of eGFP signal with CD4 labeling (colocalizations are indicated by arrows). Bar = 100 µm. scAAV, self-complementary adeno-associated virus.