Hematopoietic stem cell gene therapy for the multisystemic lysosomal storage disorder cystinosis

Abstract

Cystinosis is an autosomal recessive metabolic disease that belongs to the family of lysosomal storage disorders (LSDs). The defective gene is CTNS encoding the lysosomal cystine transporter, cystinosin. Cystine accumulates in all tissues and leads to organ damage including end-stage renal disease. Using the Ctns(-/-) murine model for cystinosis, we tested the use of hematopoietic stem and progenitor cells (HSPC) genetically modified to express a functional CTNS transgene using a self-inactivating-lentiviral vector (SIN-LV). We showed that transduced cells were capable of decreasing cystine content in all tissues and improved kidney function. Transduced HSPC retained their differentiative capabilities, populating all tissue compartments examined and allowing long-term expression of the transgene. Direct correlation between the levels of lentiviral DNA present in the peripheral blood and the levels present in tissues were demonstrated, which could be useful to follow future patients. Using a new model of cystinosis, the DsRed Ctns(-/-) mice, and a LV driving the expression of the fusion protein cystinosin-enhanced green fluorescent protein (eGFP), we showed that cystinosin was transferred from CTNS-expressing cells to Ctns-deficient adjacent cells in vitro and in vivo. This transfer led to cystine decreases in Ctns-deficient cells in vitro. These data suggest that the mechanism of cross-correction is possible in cystinosis.

Figures

Figure 1

Transduction efficiency and stability of the transgene. (a,b) Blood cell engraftment in Ctns−/− mice transplanted with pCCL-eGFP–transduced HSPC. (a) Percentage of eGFP-expressing cells measured in red cell-lysed peripheral blood at different time points in eight Ctns−/− mice transplanted with pCCL-eGFP–transduced HSPC (from #3439 to 4100; n = 8) and one non-treated Ctns−/− mouse control (#3423). The transduction efficiency was over 50% for all the mice except one, even reaching 90% for two of them. eGFP expression was stable for up to 9 months. (b) Flow cytometry analysis of the hematopoietic lineage contribution of eGFP-expressing cell in the blood for mouse 4158 compared with a non-transplanted mouse control. Transduced HSPC are capable of generating T-lymphocytes, B-lymphocytes, and monocytes. (c) In vivo luciferase imaging in Ctns−/− mice transplanted with pCCL-LUC–transduced HSPC (n = 8). Representative picture of two Ctns−/− mice transplanted with pCCL-LUC–transduced HSPC (3515 and 3517) compared with a non-transplanted Ctns−/− mouse control (3518) at 12 months post-transplantation. Live animals were injected with luciferin and luciferase expression was acquired using the IVIS imaging system. The luminescence signal intensities (total photon flux) were quantified and represented in the histogram. eGFP, enhanced green fluorescent protein; HSPC, hematopoietic stem and progenitor cell; LUC, luciferase.

Figure 2

Tissue integration of ex vivo transduced bone marrow-derived cells after transplantation. Representative confocal microscopy pictures of tissue sections from Ctns−/− mice transplanted with pCCL-eGFP–transduced HSPC (n = 8) and controls. Nuclei are stained by DAPI (blue) and direct eGFP fluorescence is seen in green. (a) Representative sections of the spleen of mice transplanted with pCCL-eGFP–transduced HSPC (left picture) compared with non-transplanted control (middle picture) show abundant eGFP-expressing cells exclusively in the transplanted mice. F-actin intermediate filament staining by Bodipy-Phalloidin showing tissue structure is seen in red. In the right picture, the immunostaining for eGFP (red channel) of liver from non-transplanted control did not show any unspecific staining. (b) Representative sections of liver from mice transplanted with pCCL-eGFP–transduced HSPC immunostained with an anti-GFP antibody (seen in red). eGFP expressing bone marrow-derived eGFP-expressing cells colocalize with the anti-GFP immunostaining, proving the specificity of the green fluorescence. F-actin intermediate filament staining by Alexa Fluor 647-Phalloidin is seen in orange in the merge picture. (c–e) Immunostaining for F4/80. Representative sections of Ctns−/− mice transplanted with pCCL-eGFP–transduced HSPC show eGFP expressing bone marrow-derived cells coexpressing F4/80 in the (c) liver, (d) the brain, and the (e) kidney demonstrating their differentiation into Kupffer cells, microglial cells, and inflammatory dendritic cells, respectively. Bars: 20 µm (in a,c,e); 10 µm (b,d). DAPI, 4′,6-diamidino-2-phenylindole; eGFP, enhanced green fluorescent protein; HSPC, hematopoietic stem and progenitor cell.

Figure 3

Effect of pCCL-CTNS in cystine content levels in vitro and in vivo. (a) The upper panel shows the expression levels of the human CTNS transgene in pCCL-CTNS–transduced Ctns−/− fibroblasts compared with pCCL-eGFP–transduced and control Ctns−/− fibroblasts as negative controls, and 293T, which are human cells, as positive control. The lower panel shows that cystine content is significantly reduced in pCCL-CTNS–transduced Ctns−/− fibroblasts compared with both pCCL-eGFP–transduced and control Ctns−/− fibroblasts. **P < 0.0001. (b) Cystine content levels (nmol half cystine/mg protein) in Ctns−/− mice (males and females) transplanted with either pCCL-CTNS–transduced Ctns−/− HSPC (pCCL HSC) or WT congenic HSPC (WT HSC) compared to control non-treated Ctns−/− mice (KO) at 4 and 8 months post-transplant in the different tissues tested. Each point represents individual data from each mouse and the mean value is represented by the median line of the diamonds; the extremities of the diamonds represent the 95% confidence intervals in each group. Significant decrease in cystine levels is observed in pCCL-CTNS–treated mice in the majority of tissues compared with control Ctns−/− mice. *P < 0.05 compared with non-treated Ctns−/− mice, ^P < 0.05 compared with pCCL-CTNS–treated Ctns−/− mice. (c) Kidney cystine content (nmol half cystine/mg protein) in non-treated male Ctns−/− mice (KO), transplanted with pCCL-CTNS–transduced Ctns−/− HSPC (pCCL HSC) or with WT congenic HSPC (WT HSC) at 8 months post-transplant. *P < 0.05 compared with non-treated Ctns−/− mice. GFP, green fluorescent protein; HSC, hematopoietic stem cell; KO, knockout; WT, wild-type.

Figure 4

Cystine crystal quantification in kidney sections from male Ctns−/−mice. Kidney sections stained with methylene blue in ethanol were used to observe cystine crystals. Low magnification pictures were analyzed using ImageJ software. Abundant cystine crystals were observed in kidney sections from non-treated Ctns−/− males (n = 9) in contrast to pCCL-CTNS–treated males (n = 8). Wild-type mice (n = 4) were used as negative control for our method of cystine crystal detection using ImageJ. Error bars are defined as mean ± SD, *P < 0.05.

Figure 5

LM-PCR analysis of the Ctns−/− mice transplanted with pCCL-CTNS–transduced HSC. Two gels are depicted, the first line of each gel is the ladder and the last lane represents a non-transplanted mouse control. Each lane shows the amplicons corresponding to the distribution of the lentiviral vector insertion sites obtained by LM-PCR in the spleen of each mouse tested (#1–13). Two dominant amplicons (arrow) were observed in one of the transplanted mice (#2). HSC, hematopoietic stem cell; LM-PCR, ligation-mediated PCR.

Figure 6

In vitro study of cross-correction. Ctns−/− fibroblasts were stably transduced by a lentivrial vector driving the expression of the fusion protein cystinosin-eGFP (pCCL-CTNSeGFP). (a–c) Demonstration of the functionality of the fusion protein. (a) Cystinosin-eGFP seen in green is targeted to the lysosomes, colocalizing with the LysoTracker staining seen in red. Bar: 10 µm. (b) Western blot analysis with an anti-GFP antibody revealed bands between 70 and 85 kb in the pCCL-CTNSeGFP–transduced fibroblasts. (c) Cystine measurements showed a significant cystine decrease in the CTNSeGFP-expressing fibroblasts compared with Ctns−/− fibroblast controls. *P < 0.0001. (d,e) Demonstration of the cross correction in vitro by coculture of pCCL-CTNSeGFP–transduced fibroblasts and DsRed Ctns−/− fibroblasts seen in red. (d) Green vesicles can be seen in the DsRed-expressing fibroblasts, demonstrating the transfer of cystinosin. Bar: 5 µm. (e) Cystine content was significantly decreased in the sorted DsRed-expressing cells cocultured pCCL-CTNSeGFP compared with the sorted DsRed-expressing cells cultured alone. Error bars are defined as mean ± SD. *P < 0.0001. eGFP, enhanced green fluorescent protein.

Figure 7

In vivo transfer of cystinosin from CTNS-expressing bone marrow-derived cells to Ctns-deficient host cells. Representative confocal microscopy pictures of liver sections from (a,b) Ctns−/− mice transplanted with pCCL-CTNSeGFP–transduced DsRed Ctns−/− HSPC (n = 5) and (c,d) Ctns−/− mice transplanted with non-transduced DsRed Ctns−/− HSPC as controls (n = 2). Nuclei are stained by DAPI (blue), DsRed-expressing bone marrow-derived cells are seen in red and cystinosin-eGFP fusion proteins are seen in green. Green intracellular vesicles are observed exclusively in Ctns−/− mice transplanted with pCCL-CTNSeGFP–transduced DsRed Ctns−/− HSPC (as shown in a,b). Large eGFP-positive vesicular structures are observed in DsRed-expressing bone marrow-derived cells but also discrete green vesicles can be seen in the host cells, demonstrating a transfer of the cystinosin from CTNS-expressing cells to the adjacent host cells (arrows in a,b). (a,c) F-actin intermediate filament staining by Alexa Fluor 647-Phalloidin is seen in orange and shows the cell boundaries. (b,d) Immunostaining with the lysosomal marker LAMP-2 is seen in orange. eGFP-positive vesicles observed in both DsRed-expressing cells and host cells colocalize with the anti-LAMP-2 staining (as shown in b). Bars: 10 µm. DAPI, 4′,6-diamidino-2-phenylindole; eGFP, enhanced green fluorescent protein.