Engineering memory T cells as a platform for long-term enzyme replacement therapy in lysosomal storage disorders

Abstract

Enzymopathy disorders are the result of missing or defective enzymes. Among these enzymopathies, mucopolysaccharidosis type I is a rare genetic lysosomal storage disorder caused by mutations in the gene encoding alpha-L-iduronidase (IDUA), which ultimately causes toxic buildup of glycosaminoglycans (GAGs). There is currently no cure and standard treatments provide insufficient relief to the skeletal structure and central nervous system (CNS). Human memory T (Tm) cells migrate throughout the body's tissues and can persist for years, making them an attractive approach for cellular-based, systemic enzyme replacement therapy. Here, we tested genetically engineered, IDUA-expressing Tm cells as a cellular therapy in an immunodeficient mouse model of MPS I. Our results demonstrate that a single dose of engineered Tm cells leads to detectable IDUA enzyme levels in the blood for up to 22 weeks and reduced urinary GAG excretion. Furthermore, engineered Tm cells take up residence in nearly all tested tissues, producing IDUA and leading to metabolic correction of GAG levels in the heart, lung, liver, spleen, kidney, bone marrow, and the CNS, although only minimal improved cognition was observed. Our study indicates that genetically engineered Tm cells hold great promise as a platform for cellular-based enzyme replacement therapy for the treatment of mucopolysaccharidosis type I and potentially many other enzymopathies and protein deficiencies.

Keywords: CRISPR-Cas9; Hurler syndrome; gene therapy; genome engineering; homology-directed repair; lysosomal storage disorders; memory T cells; mucopolysaccharidosis type I.

Graphical abstract

Figure 1

Efficient transgene insertion at the AAVS1 locus of human Tm cells using CRISPR-Cas9 and rAAV (A) Schematic of recombinant adeno-associated virus (rAAV) alpha-L-iduronidase (IDUA) expression cassette encoding homology arms targeting the adeno-associated virus integration site 1 (AAVS)1 locus and MND promoter driving expression of IDUA-T2A-RQR8. (B) Quantification of RQR8 expression via flow cytometry in engineered Tm and non-engineered Tm cells. (C) Line graph depicting IDUA activity over time in the supernatant of engineered and control Tm cells (N = 2 independent donors).

Figure 2

Short-term efficacy of engineered Tm cells to secrete IDUA and ameliorate waste substrates in vivo (A) Flow cytometry results of human CD45RO+ cells in total peripheral blood leukocytes in mice over time. (B) Fluorometric assay of murine plasma IDUA activity over a 10-week period. (C) Flow cytometry results of human CD45RO+ cells within organ tissues. (D) Fluorometric assay of tissue lysate IDUA activity levels at week 10 in multiple organs. (E) Tissue GAG content of tissue lysates at week 10 post treatment. (F) Urine GAG content at week 0, 6, and 10 post Tm cell treatment. All figures: heterozygous (black), untreated MPS I (red), and Tm cell-treated MPS I (blue) NSG mice. N = 5 mice each cohort. Statistics analysis: two-way ANOVA ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (F) Error bars represent mean with SD.

Figure 3

Engineered Tm cells persist and continually produce IDUA within NSG mice for up to 22 weeks (A) Flow cytometry results of human CD45RO cells showing total leukocytes in the peripheral blood of Tm cell-treated NSG-MPS I mice. (B) Graph of fluorimetric assay of murine plasma IDUA activity over 22 weeks from treated and control cohorts of mice. (C) Graph of RQR8-stained cells within the human CD45RO+ population of peripheral blood leukocytes as measured by flow cytometry. All figures: heterozygous (black) and Tm cell-treated MPS I (blue) NSG mice. n = 12 mice in each cohort. Statistical analysis: mixed effects analysis (A-C) ∗p < 0.05, ∗∗p < 0.01.

Figure 4

Engineered Tm cells provide IDUA systemically and reduce GAG accumulations across multiple tissues, ameliorating histopathological hallmarks of MPS I (A) Flow cytometry results of human CD45RO within organs of Tm cell-treated NSG-MPS I mice. (B) Fluorometric assay of tissue lysate IDUA activity levels at week 22 post human Tm treatment. (C) GAG content of tissue lysates at week 22 post human Tm cell treatment. (D) Urine GAG content at 0, 6, 12, and 18 weeks post Tm cell treatment. (A–D) Heterozygous (black), NSG-MPS I (red), and Tm cell-treated NSG-MPS I (blue) NSG mice (n = 12 mice each cohort). (E) Representative histological stained images of human CD3 (top), human IDUA (middle), and LAMP-1 (bottom) of mouse cohorts. (F) Quantification of human IDUA immunohistochemistry within liver samples (n = 4 each cohort). (G) Quantification of LAMP-1 immunohistochemistry within liver samples (n = 4 each cohort). Statistical analysis: two-way ANOVA (A–D), one-way ANOVA (F and G) ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 5

Engineered Tm cells have the capability to ameliorate pathological MPS I hallmarks within the central nervous system (A) Representative histological images of human CD3 staining within brain tissue. Black arrows, CD3 immunopositive cells; gray arrows, astrocytes (top). Representative histological images of human IDUA within brain tissue (middle). Representative histological images of LAMP-1 staining within brain tissue. Black arrows, neurons; gray arrows, astrocytes (bottom). (B) Quantification of LAMP-1 immunohistochemistry within brain tissue samples (n = 4 each cohort). (C) Barnes maze time to escape of heterozygous (black), untreated NSG-MPS I (red), and Tm cell-treated NSG-MPS I mice (blue) (n = 9). Statistical analysis: one-way ANOVA (B), two-way ANOVA (C) ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.