Generation of a new Slc20a2 knockout mouse line as in vivo model for primary brain calcification

Abstract

Primary brain calcification (PBC) is a neurodegenerative disease that causes bilateral ectopic calcification in the brain. In this study, using newly generated Slc20a2 knockout (Slc20a2^−/−) mice, we establish an in vivo model for PBC. In contrast to heterozygous Slc20a2^+/− mice (9/9 animals) showing no obvious abnormalities, the homozygous Slc20a2^−/− mice exhibited severe calcification at 11 months of age (5/5 animals). Whilst smaller in size and number, the deposits were also detectable in 5-month-old Slc20a2^−/− mice (2/2 animals). By contrast, no obvious alterations were detectable in visceral organs, including the lung, kidney, liver, and spleen. Consistently, in PBC patients, despite the systemic mineral metabolic disturbance, calcification occurs only in a brain restricted manner. Hence, these observations suggest that our mouse model is capable of recapitulating certain aspects of human PBC etiology. In summary, our data suggested the utility of an in vivo PBC mouse model in understanding the pathological mechanisms behind brain calcification, which leads in development of novel therapeutics against PBC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13041-025-01240-8.

Patients with primary brain calcification (PBC) suffer from diverse clinical manifestations including Parkinsonism, cognitive disorder, and psychiatric symptoms. In the past decade, several causative gene mutations have been identified in SLC20A2, PDGFRB [1], PDGFB [2], XPR1 [3], MYORG [4], JAM2 [5], and NAA60 [6], whereas the etiology of PBC still remains largely unknown. It has been previously reported that Slc20a2 KO mice serve as a potential in vivo model for PBC [7]. In this report, we generated a new Slc20a2 KO mouse line using the genome editing technology, CRISPR/Cas9, and evaluated its suitability as a PBC model.

To disrupt Slc20a2 gene, we inserted a stop codon and an MluI site into the exon 3 sequences via the homology-directed repair mechanism mediated by Cas9/gRNA-induced DNA double strand break (DSB) following zygotic microinjection of ribonucleoprotein (Fig. 1A). The KO allele was confirmed by PCR genotyping and subsequent MluI digestion of PCR fragments (Fig. 1B). To check if any undesired off-target mutations were induced, we selected 14 potential DSB genomic locations using the Cas-OFFinder off-target prediction algorithm (http://www.rgenome.net/cas-offinder/) under the condition of satisfying up to 3 mismatches or 1 bulge plus up to 2 mismatches to the target sequence. As far as tested, we were unable to identify any mutations in these predicted off-target locations (Supplemental Table1 and Fig. 2). We used F2 mice in examinations of this study. The levels of Pit2 protein were diminished in Slc20a2^−/− mice compared with Slc20a2^+/+ mice in kidney and cerebellum (Supplemental Fig. 1D), which were detected by using specific antibody for Pit2 used in the previous studies [8, 9]. Consistent with the reported Slc20a2^−/−mice, our Slc20a2^−/− mice are smaller in size and show significantly decreased body weight, and the survival rate of Slc20a2^−/− mice was lower than that of the wild type (Supplemental Fig. 1G-I). At 11-months-old, whilst rarely detectable in Slc20a2^+/− mice (9/9 animals), the Slc20a2^−/− mice exhibited severe calcification (5/5 animals) in the brain. Consistent with previous reports [10, 11], calcium deposits in the Slc20a2^−/− mice were observed in the thalamus, hypothalamus, midbrain, pons, and cerebral cortex (Fig. 1C). The Slc20a2^−/− mice showed no detectable level of calcification in their visceral organs, including the liver, kidney, lung and spleen (Supplemental Fig. 1J).

Fig. 1.

(A) Genomic editing strategy for the mouse Slc20a2 gene was presented. (B) Genotyping of the Slc20a2 knockout allele was determined. (C, D) Sagittal sections (C) and coronal sections of the brain at the hippocampus–amygdala level (D) using Kossa staining were performed. (E, F) Deposits near cortical vessels (E) and in the cortical parenchyma (F) in 11-month-old (−/−) mice were determined using HE staining. (G) Cerebral cortex using KB staining in 11-month-old (−/−) mice was performed. (H–J) Deposits stained using Kossa (H), Alcian Blue (I), and Berlin Blue stains in 11-month-old (−/−) mice (J) were performed. (K–L) Inflammatory changes in the cortex, glial fibrillary acidic protein (K), Iba-1 (L) were determined. Scale bars: C-D: 1 mm, E–G: 50 μm, (G inset), H–J: 2 μm, K-L: 50 μm

Compared to the mice in the previous paper, our mice are simple modifications that leave the original gene sequence as much as possible, so there is no insertion of reporter genes or drug resistance genes as in the previous report. The difference is that the possibility of phenotypic effect due to the insertion of a foreign gene for knockout is lower than that of the preceding mouse, which is an advantage of our mice. As we found slight reduction of Slc20a2 mRNA levels in Slc20a2^+/− mice (Supplemental Fig. 1E) and Pit2 protein expression in Slc20a2^+/− mice was reduced to about 50% of that in Slc20a2^+/+ mice (Supplemental Fig. 1F), the lack of calcification in the heterozygous mice doesn’t appear to be mediated by a dosage compensation effect, whereby defective protein and/or mRNA expression from the knockout allele might be compensated by the expression from wild-type allele. Given the reduction of Scl20a2 expression, it is tempting to assume that the bulk phosphate transport activity of the cells expressing Slc20a2 might be also decreased in the Slc20a2^+/− mice. If this would be the case, it is conceivable that it would require longer time period until calcification would become evident in the Slc20a2^+/− mice.

In the Slc20a2^−/− mice, deposits were observed mainly at the peri-vascular space and the brain parenchyma, where they formed concentric patterns (Fig. 1E and F). In contrast, neurons surrounding these deposits appeared to be unaffected (Fig. 1G). These deposits were positive not only for von Kossa, Alcian blue staining (for calcium detection) but also for Berlin blue staining (for iron detection) (Fig. 1H–J), suggesting the deposit formation may be resulting from defective homeostatic regulation of both calcium and iron ions in the brain.

Glial changes against the deposits were also observed in 11-month-old Slc20a2^−/− mice (Fig. 1K and L). In addition, some astrocytes take deposition into them reaching their process (Fig. 1K), and inflammatory changes appear to concomitantly progress gradually. These observations are consistent with the pathological abnormalities seen in PBC patients [12]. However, in sharp contrast with the previous report showing the presence of gliosis and its associated neuronal death in the brain autopsy samples of PBC patients [13], neither neurodegeneration, which is evident by the accumulation of phosphorylated tau and α-synuclein, (Supplemental Fig. 1K), nor neuronal destruction was detectable within the region surrounding the brain deposits in 11-month-old Slc20a2^−/− mice (Supplemental Fig. 1L). Although the pathological mechanism of neurodegeneration in PBC patients remained largely elusive, the fact that the Slc20a2^−/− mice exhibit severe calcification without showing obvious neurodegeneration suggests that calcification doesn’t represent a primary cause of neurodegeneration.

In conclusion, we successfully generated a new Slc20a2 KO mouse model. This approach confirmed the utility of an in vivo PBC model to promote drug discovery research and unravel the molecular mechanisms behind brain calcification in humans.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

I.H. and H.Kurita designed the study. H.Kurita and H.Kitaura wrote the manuscript. H.Kurita, H.Kitaura, K.N., T.M., and M.O. performed experiments and analyzed the data. K.O., M.I., A.K., M.O., and I.H. edited the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by grants from Science and Technology of Japan (Basic Research (C) (25K15455)) (to H.Kurita), the Ministry of Education, Culture, Sports, Science and Technology of Japan (Basic Research (B) (17H04198)) (to I.H.) and Japan Agency for Medical Research and Development (AMED) (20ek0109313s0203) (to A.K.), (20ek0109313s0103) (to M.O.), and (20ek0109313h003) (to I.H.) in Japan.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Mouse studies were approved by the Animal Experiment Committees of Gifu Pharmaceutical University, Japan (Approval number: 2018 − 152) and the Committee for Animal Research and Welfare of Gifu University (Approval number: 2020-063).

Consent for publication

None.

Competing interests

The authors declare no competing interests.