Protein replacement therapy partially corrects the cholesterol-storage phenotype in a mouse model of Niemann-Pick type C2 disease

Abstract

Niemann-Pick type C2 (NPC2) disease is a fatal autosomal recessive neurovisceral degenerative disorder characterized by late endosomal-lysosomal sequestration of low-density lipoprotein derived cholesterol. The breach in intracellular cholesterol homeostasis is caused by deficiency of functional NPC2, a soluble sterol binding protein targeted to the lysosomes by binding the mannose-6-phosphate receptor. As currently there is no effective treatment for the disorder, we have investigated the efficacy of NPC2 replacement therapy in a murine gene-trap model of NPC2-disease generated on the 129P2/OlaHsd genetic background. NPC2 was purified from bovine milk and its functional competence assured in NPC2-deficient fibroblasts using the specific cholesterol fluorescent probe filipin. For evaluation of phenotype correction in vivo, three-week-old NPC2(-/-) mice received two weekly intravenous injections of 5 mg/kg NPC2 until trial termination 66 days later. Whereas the saline treated NPC2(-/-) mice exhibited massive visceral cholesterol storage as compared to their wild-type littermates, administration of NPC2 caused a marked reduction in cholesterol build up. The histological findings, indicating an amelioration of the disease pathology in liver, spleen, and lungs, corroborated the biochemical results. Little or no difference in the overall cholesterol levels was observed in the kidneys, blood, cerebral cortex and hippocampus when comparing NPC2(-/-) and wild type mice. However, cerebellum cholesterol was increased about two fold in NPC2(-/-) mice compared with wild-type littermates. Weight gain performance was slightly improved as a result of the NPC2 treatment but significant motor coordination deficits were still observed. Accordingly, ultrastructural cerebellar abnormalities were detected in both saline treated and NPC2 treated NPC2(-/-) animals 87 days post partum. Our data indicate that protein replacement may be a beneficial therapeutic approach in the treatment of the visceral manifestations in NPC2 disease and further suggest that neurodegeneration is not secondary to visceral dysfunction.

Figure 1. Effect of purified bovine NPC2 on cholesterol accumulation in wild type and NPC2^−/− fibroblasts.

(A) SDS-PAGE of 5 µg purified NPC2 resolved in a 10–20% gradient gel under nonreducing conditions and stained with Coomassie Brilliant Blue (lane 1). Molecular mass markers are shown on the left (lane M). (B) Upper row human- (Hu) and lower row murine (Mu) fibroblasts cultivated for 48 hours in complete medium (DMEM + 10% FBS). Wild type fibroblasts (left panels), NPC2^-/- fibroblasts (middle panels), NPC2^-/- fibroblasts supplemented with 600 nM NPC2 (right panels). Cells were fixed with 10% phosphate buffered formalin, pH 7.4 and stained with Filipin III and visualized using fluorescence microscope.

Figure 2. Immune response to NPC2 in 129P2 wild type- and NPC2^−/− mice.

NPC2 coated microtiter wells were incubated with serial dilutions of immunized serum as indicated. Bound antibodies were detected by TRIFMA as described in Materials and Methods. An arbitrary concentration of anti-NPC2 antibodies was set to 1000 mU/ml in positive control serum prepared by subcutaneously injections of NPC2 with Freund's complete adjuvant as immune potentiator. Sera from saline treated healthy mice were similarly tested and served as negative controls. (A) Anti-NPC2 antibody concentration in intraperitoneal injected wild type mice, (B) Anti-NPC2 antibody concentration in intravenous injected wild type mice, and (C) Anti-NPC2 antibody concentration in intravenous NPC2 treated NPC2^−/− mice. Data represents mean values of triplicate wells.

Figure 3. Effect of NPC2 treatment on animal body weight.

(A) Males (n = 4) and (B) females (n = 6), respectively, of saline-treated wild type mice (•), saline-treated NPC2^−/− mice (○), and NPC2 treated NPC2^−/− mice (▾). The mice were weighed weekly from P21 to P87. Each animal was injected twice weekly with saline or NPC2 (5 mg/kg). Values are means ± SEM.

Figure 4. Effect of NPC2 replacement therapy on systemic organs and serum cholesterol storage.

Saline treated wild type mice (black bars), saline treated NPC2^−/− mice (light gray bars), and NPC2 treated NPC2^−/− mice (5 mg/kg) (dark gray bars) were injected twice weekly during the evaluation period (P21 to P87). Post mortem total cholesterol concentrations were determined in liver, spleen, kidney, lung, and serum. Each bar represents the mean ± SEM measure for 6 animals in each of the experimental groups done in duplicate. Bars not sharing a letter within a given panel are significantly different (P<0.05).

Figure 5. Histochemical and immunohistochemical analysis of NPC2 replacement therapy in murine liver sections.

Staining of representative tissue sections of 87 days old saline treated wild type mice (left panels), saline treated NPC2^−/− mice (middle panels), and NPC2 treated NPC2^−/− mice (right panels). Hematoxylin-eosin (H&E) staining (first row), immunohistochemical localisation of antigen F4/80 positive macrophages (brownish) (second row), and Masson's trichrome staining to detect collagen (blue) (third row). Lipid laden macrophages (Kupffer cells) are clearly visible and prominent in liver section from saline treated NPC2^−/− mice, whereas only a minority of the macrophages in liver sections from NPC2 treated NPC2^−/− mice are correspondingly loaded. Data are representative of three separate experiments. n = 3 animals in each experimental group. Scale bars represent 80 µm.

Figure 6. Histochemical analysis of NPC2 replacement therapy in murine spleen sections.

Hematoxylin-eosin (H&E) staining of spleen from saline treated wild type mice (left panel), saline treated NPC2^−/− mice (middle panel), and NPC2 treated NPC2^−/− mice (right panel). Massive accumulation of lipid droplets was most prominent observed in the spleens of saline treated NPC2^−/− mice. Data are representative of three separate experiments. n = 3 animals in each experimental group. Scale bares represent 100 µm.

Figure 7. Histochemical and immunohistochemical analysis of NPC2 replacement therapy in murine lung sections.

Staining of representative tissue sections from saline treated wild type mice (left panels), saline treated NPC2^−/− mice (middle panels), and NPC2 treated NPC2^−/− mice (right panels). Hematoxylin-eosin (H&E) lung staining (first row), immunohistochemical localisation of antigen F4/80 positive macrophages (brownish) (second and third row), and PAS-staining to detect the presence of glycogen (dark purple) (fourth row). Data are representative of three separate experiments. n = 3 animals in each experimental group. Scale bars represent 30 µm except in second row where it equals 100 µm.

Figure 8. Effect of NPC2 replacement therapy on murine brain cholesterol storage.

Total cholesterol levels in cerebellum, cortex, and hippocampus was measured postmortem in saline treated wild type mice (black bars), saline treated NPC2^−/− mice (light gray bars), and NPC2 treated NPC2^−/− mice (dark gray). Each bar represents the mean ± SEM for 6 animals in each of the three groups. Bars not sharing a letter within a given panel are significantly different (P<0.05).

Figure 9. Histochemical and immunohistochemical staining of cerebellum.

Staining of representative cerebellum sections from saline-treated wild type mice (left panels), saline treated NPC2^−/− mice (middle panels), and NPC2 treated NPC2^−/− mice (right panels). Hematoxylin-eosin (H&E) staining (first row), immunohistochemical GFAP-staining for detection of astrocytes (brownish) (second row), and immunohistochemical localisation of antigen F4/80 positive macrophages (brownish) (third row). Data are representative of three separate experiments. n = 3 animals in each experimental group. Scale bars represent 30 µm.