Characterization of an immunodeficient mouse model of mucopolysaccharidosis type I suitable for preclinical testing of human stem cell and gene therapy

Abstract

Mucopolysaccharidosis type I (MPS-I or Hurler syndrome) is an inherited deficiency of the lysosomal glycosaminoglycan (GAG)-degrading enzyme alpha-l-iduronidase (IDUA) in which GAG accumulation causes progressive multi-system dysfunction and death. Early allogeneic hematopoietic stem cell transplantation (HSCT) ameliorates clinical features and extends life but is not available to all patients, and inadequately corrects its most devastating features including mental retardation and skeletal deformities. To test novel therapies, we characterized an immunodeficient MPS-I mouse model less likely to develop immune reactions to transplanted human or gene-corrected cells or secreted IDUA. In the liver, spleen, heart, lung, kidney and brain of NOD/SCID/MPS-I mice IDUA was undetectable, and reduced to half in heterozygotes. MPS-I mice developed marked GAG accumulation (3-38-fold) in these organs. Neuropathological examination showed GM(3) ganglioside accumulation in the striatum, cerebral peduncles, cerebellum and ventral brainstem of MPS-I mice. Urinary GAG excretion (6.5-fold higher in MPS-I mice) provided a non-invasive and reliable method suitable for serially following the biochemical efficacy of therapeutic interventions. We identified and validated using rigorous biostatistical methods, a highly reproducible method for evaluating sensorimotor function and motor skills development. This Rotarod test revealed marked abnormalities in sensorimotor integration involving the cerebellum, striatum, proprioceptive pathways, motor cortex, and in acquisition of motor coordination. NOD/SCID/MPS-I mice exhibit many of the clinical, skeletal, pathological and behavioral abnormalities of human MPS-I, and provide an extremely suitable animal model for assessing the systemic and neurological effects of human stem cell transplantation and gene therapeutic approaches, using the above techniques to measure efficacy.

FIGURE 1. Genotyping by PCR

The wild-type (WT) allele produced a 550 bp PCR product whereas the disrupted MPS-I allele produced a 350 bp PCR product. Heterozygous animals (Het) demonstrated both products. The last lane on the right shows a DNA size ladder.

FIGURE 2. Phenotypic features of wild type (WT), heterozygous and NOD/SCID/MPS-I (MPS) mice

A. Facial features. M: males; F: females. MPS-I mice show coarsened fur and facial features, shortened snout, and broadened face resulting in lack of protrusion of eyes. B. Radiograph of the skull: the arrow indicates the thickened zygomatic bone in MPS-I mice.

FIGURE 3. Sensorimotor and learning assessment by the RotaRod test

Balance times on the Rotarod for male and female animals are shown separately, for the first (pre-training) and fifth (post-training) tests. Wild-type (WT): grey columns; heterozygous: hatched columns; MPS-I: black columns. The columns show the balance times at each speed (5, 15, 25 and 35 rpm) expressed as the mean +/- SE. Comparison between MPS-I and the other (WT or heterozygous) genotypes: *P<0.02; **P<0.001. MPS-I mice of both genders had impairment of motor coordination, evidenced by lower balance times at higher speeds (≥15 rpm) when compared to WT or heterozygous mice on the initial (pre-training) tests. The MPS-I mice also demonstrated a relative inability to acquire motor coordination skills even after 5 training tests, evidenced by their lower balance times at higher speeds (≥25 rpm) on the post-training tests. This inability to improve motor coordination was more severe in males than in female MPS-I mice, evidenced by the lack of any improvement between the pre- and post-training balance time at 35 rpm in MPS-I males.

FIGURE 4. Urinary excretion of GAGs

Total GAGs and creatinine were measured in a 24 h urine collection, to calculate the relative GAG excretion in μg GAG/mg creatinine. Urine samples were collected from 8 animals of each genotype. Data is shown as the mean +/- SE. Comparison between MPS-I and WT or heterozygous mice: P <0.0002. Excretion of GAGs by heterozygous mice was equivalent to that by WT mice, whereas excretion of GAGs by MPS-I mice was consistently higher.

FIGURE 5. Tissue IDUA enzyme activity

Samples of each of the indicated tissues were obtained from 10 animals of each genotype, except 9 animals each for MPS-I liver and spleen. Data is shown as the mean +/- SE. WT: grey columns; heterozygous: hatched columns; MPS: *no detectable activity. Comparison between MPS-I and WT or heterozygous mice: P <0.0001 for all tissues. Comparison between WT and heterozygous mice: brain: P <0.0001; heart: P <0.001; liver: P <0.005; kidney: P <0.009; lung: P <0.04 and spleen: P <0.002. IDUA activity was undetectable in all tissues from MPS-I mice, and was about ½ of WT in all examined tissues of heterozygous mice.

FIGURE 6. Tissue accumulation of GAGs

Total GAGs and protein were measured in each tissue sample, to calculate the GAG content in μg GAG/mg protein. Samples of each tissue were obtained from 8-13 animals of each genotype. Data is shown as the mean +/- SE. WT: grey columns; heterozygous: hatched columns; MPS: black columns. Comparison between MPS-I and WT or heterozygous mice: P <0.0003 for all tissues except MPS-I brain vs WT brain: P <0.008. The GAG content of different tissues from WT mice was about 5-15 μg/mg protein. In contrast, there were marked differences in the GAG content of different tissues from MPS-I mice, being lowest in the brain (3-fold higher than WT) and highest in the liver (38-fold higher than WT).

FIGURE 7. GM₃ ganglioside accumulation in the brain

Immunohistochemical staining showing accumulation of GM₃ gangliosides (brown staining cells) in all the examined regions of the brains of MPS-I mice, most notably in the brainstem and striatum. There was no detectable accumulation in the brains of WT or heterozygous mice. Magnification: 400x; bar: 25 μM.