Two hypomorphic alleles of mouse Ass1 as a new animal model of citrullinemia type I and other hyperammonemic syndromes

Abstract

Citrullinemia type I (CTLN1, OMIM# 215700) is an inherited urea cycle disorder that is caused by an argininosuccinate synthetase (ASS) enzyme deficiency. In this report, we describe two spontaneous hypomorphic alleles of the mouse Ass1 gene that serve as an animal model of CTLN1. These two independent mouse mutant alleles, also described in patients affected with CTLN1, interact to produce a range of phenotypes. While some mutant mice died within the first week after birth, others survived but showed severe retardation during postnatal development as well as alopecia, lethargy, and ataxia. Notable pathological findings were similar to findings in human CTLN1 patients and included citrullinemia and hyperammonemia along with delayed cerebellar development, epidermal hyperkeratosis, and follicular dystrophy. Standard treatments for CTLN1 were effective in rescuing the phenotype of these mutant mice. Based on our studies, we propose that defective cerebellar granule cell migration secondary to disorganization of Bergmann glial cell fibers cause cerebellar developmental delay in the hyperammonemic and citrullinemic brain, pointing to a possible role for nitric oxide in these processes. These mouse mutations constitute a suitable model for both mechanistic and preclinical studies of CTLN1 and other hyperammonemic encephalopathies and, at the same time, underscore the importance of complementing knockout mutations with hypomorphic mutations for the generation of animal models of human genetic diseases.

Figure 1

Expression of ASS1 in mouse tissues. Several tissues from wild-type FVB/N adult mice were analyzed by IHC using a monoclonal antibody against the C terminus of human ASS (BD 611700). Note the high expression of ASS1 in liver, intestine (villi), kidney (tubules), and bone marrow. In the skin, ASS1 is expressed in the keratinocytes of the basal epidermis and outer root sheath. All magnifications, ×200.

Figure 2

Homozygous Ass1^bar and Ass1^fold and compound heterozygous mice exhibit variable degrees of growth retardation and alopecia. A, Left: Growth retardation and alopecia at P9 in Ass1^bar/Ass1^bar (bottom) and Ass1^fold/Ass1^fold (middle) compared with control littermate (top). Right: Growth retardation and alopecia at P10 in Ass1^bar/Ass1^bar (right) and Ass1^bar/Ass1^fold (middle) compared with littermate control (left) (all mice on C57BL/6 background). B, Left: At P21, Ass1^fold/Ass1^fold (top) and Ass1^bar/Ass1^fold (middle) are almost the size of wild-type littermates (bottom) and exhibit a full coat of pelage, albeit sparse. Right: An adult FVB/N-bar/bar mouse exhibiting substantial reduction in body size and generalized alopecia when compared with littermate control (bottom). C, Top: Mean weight values of FVB-+/+ (n = 15), FVB-bar/bar (n = 15), C57BL/6-+/+ (n = 12), C57BL/6-bar/bar (n = 10), C57BL/6-fold/fold (n = 12), and C57BL/6-bar/fold (n = 10), showing growth retardation in mutant mice during the first 3 weeks of life, particularly pronounced in bar/bar mice. P < 0.0001 for C57BL/6-+/+ vs. C57BL/6-bar/bar; P = 0.0002 for FVB-+/+ vs. FVB-bar/bar (pairwise comparisons using t-tests). Bottom: Survival curves over the first 30 days of life show differences in death rates between FVB-bar/bar (n = 45), C57BL/6-bar/bar (n = 42), C57BL/6-fold/fold (n = 22), and C57BL/6-bar/fold (n = 12) mutant mice. At P30, differences between bar/bar and bar/fold genotypes were highly statistically significant (P < 0.001).

Figure 3

Histology and immunohistochemistry of FVB/N-Ass1^bar/Ass1^bar mouse epidermis at P15. A: Hematoxylin-eosin (H&E) staining of normal (left) and mutant (right) dorsal skin. Although hair follicles are in anagen, hair shafts do not penetrate the epidermis and keratin is accumulated inside the dilated follicular infundibulum (arrow). Mutant skin is severely dysplastic, markedly hyperkeratotic, and lacks a normal hypodermal fat layer. B: Note the overexpression of basal cell marker K14 in mutant epidermis. C: Note that in back skin from mutant mice, K6 expression is present in the interfollicular epidermis, a feature not found in age- and sex-matched littermate controls. D and E: Note the marked increase in the expression of the late marker proffilagrin/filaggrin and adhesion molecule desmoglein in mutant epidermis. Immunohistochemistry was performed with specific antibodies as described in Materials and Methods. All magnifications, ×100.

Figure 4

Morphological changes in the cerebellum of FVB/N-Ass1^bar/Ass1^bar mice at P7 and P14. A: H&E and Ki-67 immunostaining (inset) of a sagittal section of normal (left) and mutant (right) cerebella. Note the reduced cellularity of the IGL and poorly defined ML in bar/bar mice at P7 compared with controls. B: Activated caspase-3 immunostaining at P7 showing increased apoptosis in the EGL and IGL of bar/bar mice (arrowheads). C: At P14, Pax6 immunostaining shows a thicker EGL in bar/bar mutant mice compared with controls (arrowheads). D: Seventy-two hours after a single BrdU injection, there are still many granule cells labeled in the EGL or migrating through the ML in bar/bar mutant mice (P14). E: GFAP immunostaining shows disorganization, low density, and shortening of Bergmann glial fibers in the cerebellum of bar/bar mice at P14. Immunohistochemistry was performed with specific antibodies as described in Materials and Methods. Magnifications, ×40 (A) and × 100 (B–E). IGL indicates internal granule layer; EGL, external granule layer; ML, molecular layer.

Figure 5

Nitrotyrosine and NOS1 in cerebella from C57BL/6-Ass1^fold/Ass1^fold mice at P8 and MDA at P14. A: Nitrotyrosine immunostaining (in red) of a sagittal section of normal (left) and mutant (right) cerebella shown at ×100 and ×400 (inset). Nuclei are shown in blue. Arrowheads point at areas with nitrotyrosine in the EGL and IGL. B: NOS1 immunostaining (in green) shown at ×100. Arrowheads point at areas with positive staining. C: MDA immunostaining at P14 showing increased expression in the EGL of fold/fold cerebellum (arrowheads) at ×400. Immunohistochemistry was performed with specific antibodies as described in Materials and Methods (magnifications, ×100 and ×400). IGL indicates internal granule layer; EGL, external granule layer; ML, molecular layer; NOS1, neuronal NO synthase; MDA, malondialdehyde.

Figure 6

Improvement in mutant mice treated with sodium benzoate and L-arginine. A: All bar/bar mice, carrying the more severe allele, circumvented the neonatal crisis when treated with daily intraperitoneal injections of sodium benzoate (0.1 g/kg body weight) and L-arginine (1 g/kg body weight). Without treatment, 50% of homozygous bar/bar mice die before day 12. After three weeks of daily injections, mutant mice started to grow hair (left), and some mice exhibited a full hair coat (although sparse) after four weeks of treatment (right). B: Cerebellar morphology in treated mutant mice appeared to be corrected at P14. Top: Cerebellum from untreated bar/bar mouse showing thick EGL, poorly defined ML and reduced cellularity of the IGL. Middle: Treated bar/bar mouse showing thin EGL (with increased proliferation of granule cells when compared with wild type), well-defined ML, and almost normal cellularity of the IGL. Bottom: Cerebellum from wild-type mouse. Ki-67 immunohistochemistry was performed with specific antibodies as described in Materials and Methods. Magnifications, ×40.