In the past decades, genetic etiology of IBGC was suspected based on the identification of several familial clusters [8,9,10,11]. However, no genetic cause of IBGC was recognized before the identification in 2012 of SLC20A2 as the first associated gene [12]. Since then, six genes have been definitively linked with PFBC: four inherited in an autosomal dominant manner (SLC20A2, PDGFB, PDGFRB, and XPR1) and two displaying an autosomal recessive inheritance (MYORG and JAM2) [13,14,15,16,17]. Very recently, a novel gene (i.e., CMPK2) has been proposed to be associated with autosomal recessive brain calcifications and this awaits confirmation in independent studies [18]. A review of data from 555 genetically diagnosed patients with PFBC revealed the following frequency of mutations: SLC20A2~60%, MYORG~13%, PDGFB~13%, PDGFRB~6%, XPR1~6%, JAM2~2% [19,20]. A brief genetic characterization of each these genes is provided below in association with the genetic evidence deriving from knock-out or hypomorphic animal models.

SLC20A2 (chr. 8p11.21, 11 exons) encodes the sodium-dependent phosphate transporter 2 (652 amino acids), which plays a fundamental role in cellular phosphate transport. Several pathogenic variants have been reported, including missense, nonsense, splice-disruptive, small indels, and gross deletions. Pathogenic variants have been shown to impair uptake of phosphate (loss-of-function mechanism of disease) [12]. Interestingly, a complete Slc20a2 knockout mouse showed extensive, bilateral calcifications in the thalamus, basal ganglia, and cortex, recapitulating the human disease [21].

MYORG (chr. 9p13.3, one coding exon) encodes the myogenesis regulating glycosidase (putative) (714 amino acids), a protein located in the endoplasmic reticulum and nuclear membranes whose functions remain poorly defined, but are probably linked to protein glycosylation [22]. Many pathogenic variants have been described, including missense, nonsense, and small indels. The presence of biallelic nonsense mutations of MYORG in PFBC patients strongly suggests that a functional loss of MYORG is associated with brain calcifications. Indeed, Myorg-KO mice and zebrafish develop progressive brain calcifications [14,23].

PDGFRB (chr. 5q32, 23 exons) encodes the platelet-derived growth factor receptor beta (PDGFR-β) (1106 amino acids), a cell surface tyrosine kinase receptor for mesenchymal cell mitogens. Several pathogenic variants are known, including missense and start-loss mutations. Pdgfrb-deficient mice lack pericytes, essential for blood–brain barrier (BBB) formation, and display increased vascular permeability [24].

PDGFB (chr. 22q13.1, six exons) encodes the platelet-derived growth factor subunit B (241 amino acids), which is a potent mitogen for cells of mesenchymal origin, by binding and activating PDGF receptor tyrosine kinases such as PDGFR-β. A number of pathogenic variants have been reported in association with PFBC, including missense, nonsense, splice-disruptive, start loss, stop loss, and gross deletions. Hypomorphic Pdgfb mice develop brain calcifications that correlate with the degree of pericyte and BBB deficiency [17].

XPR1 (chr. 1q25.3, 15 exons) encodes the xenotropic and polytropic retrovirus receptor 1 (696 amino acids), which plays a role in phosphate homeostasis and mediates phosphate export from the cells. Few pathogenic variants have been reported, which include missense and splice disruptive. Remarkably, the majority of these are located in the SPX putative regulatory domain of the protein [15,25].

JAM2 (chr. 21q21.3, 10 exons) encodes the junctional adhesion molecule 2 (298 amino acids), which is localized at the tight junctions of both epithelial and endothelial cells and mediates heterotypic cell–cell interactions [26]. Being the most recently described gene, few pathogenic variants have been reported, including missense, start-loss, nonsense, and splice-disruptive. Pathogenic variants lead to reduction/absence of JAM2 protein, consistent with a loss-of-function mechanism. The brain calcification phenotype is replicated in the Jam2 complete knockout mouse [13].

Calcifications in the basal ganglia and other brain structures are also observed in several genetic diseases with normal calcium–phosphate metabolism. This clinically and genetically highly heterogeneous group of syndromes includes mitochondrial diseases (predominantly mtDNA mutations), Aicardi–Goutières syndrome (ADAR, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, TREX1, IFIH1), Coats plus syndrome (CTC1), leukoencephalopathy with calcifications and cysts (SNORD118), Cockayne syndrome (DDB2, ERCC1, ERCC2, ERCC3, ERCC4, ERCC5, ERCC6, ERCC8, GTF2H5, MPLKIP, POLH, XPA, XPC) and many others. Deep clinical phenotyping is crucial before genetic testing in this context [27]. In the absence of specific diagnostic clues for a particular form (see below) the best genetic approach is next-generation sequencing (e.g., gene panels or exome sequencing) [27]. A detailed genetic description of every one of the many genes involved in these syndromes is beyond the scope of this review. A complete list is provided in Table 1.

The pathophysiology of PFBC is profoundly linked to the loss of integrity of the BBB and should be distinguished from the calcification processes that take place in secondary brain calcifications, in which calcium deposition develops via two main pathways: dystrophic calcification resulting from membrane disruption and uncontrolled calcium entry (hypoxic ischemic injury, intrauterine infections), and calcium–phosphate metabolism alterations (pseudohypoparathyroidism and pseudopseudohypoparathyroidism, mainly linked to loss of function mutations of GNAS, involved in the intracellular transmittal pathway of PTH) [28].

Considering the functions of the known causative genes, the pathology of PFBC can be ascribed to altered phosphate transportation or impaired permeability through the BBB, which both result in increased inorganic phosphate (Pi) levels in CSF and brain interstitial fluid. The excess of Pi induces its deposition in the form of calcium hydroxyapatite (Ca₁₀[PO₄]6[OH]₂) in the vascular extracellular matrix (Figure 1).

The reasons for the greater basal ganglia vulnerability to calcium deposition are not completely understood but may depend on the higher expression of the causative genes in the basal ganglia neurovascular system, as well as on the peculiar vascularization of these areas [29].

SLC20A2 and XPR1 operate directly in inorganic phosphate transportation: the former, mainly expressed by neurons, astrocytes, and vascular smooth muscle cells (VSMC), imports extracellular Pi into the cell mediating its transportation from CSF to blood; the latter exports intracellular Pi out of cells such as neurons, astrocytes, and microglia [15]. Mutations of these transporters lead to the accumulation of extracellular Pi, due to reduced Pi internalization in SLC20A2 mutants, and to intracellular accumulation with subsequent down-regulation of uptake in XPR1 mutants. It has been hypothesized that persistent elevation in extracellular Pi may induce intracranial calcification, possibly via trans-differentiation of vascular smooth muscle cells (VSMCs) into osteoblast-like cells [30].

Conversely, PDGFRB, PDGFB, MYORG, and JAM2 operate in the brain neurovascular unit (made of neurons, astrocytes, endothelial cells, pericytes, and VSMCs) and indirectly impair the BBB permeability [31]. PDGFR-β is expressed mainly by pericytes and VSMCs, while PDGFB is predominantly released by the endothelium, enabling the recruitment of PDGFR-β-expressing cells and thus promoting the wrapping of the cerebral vessel [32]. Deficiency of PDGF-B/PDGFR-β signaling causes alteration in pericyte recruitment and endothelial cell morphology, leading to increased vascular permeability. MYORG is predominantly expressed in astrocytes and localized in the endoplasmic reticulum; though presumed to act as an α-glucosidase, its molecular function remains largely unknown [22]. Astrocytic endfeet connect the endothelial cells to the extracellular matrix and neurons, thus exerting an important function in the maintenance of the BBB. Speculatively, intracellular ER alterations may impair astrocytes’ properties and conformation, causing BBB disruption. JAM2 is a key component of the tight junctions between endothelial cells in the neuro-vascular unit. JAM2 loss-of-function causes disarrangement of tight junctions, leading to high Pi leakage from the blood to the brain tissue [13].

Concerning genetic brain calcifications other than PFBC, at least three additional pathogenetic mechanisms haver been identified:

• Cerebral microangiopathies and gliosis, resulting from the malfunctioning of cellular systems fundamental in vessels’ stability and/or angiogenesis, such as small nucleolar RNA involved in ribosome biogenesis in leukoencephalopathy with calcification and cysts (LCC) [33], telomeres in Coats plus syndrome, nucleases in Aicardi–Goutières syndrome and type IV collagen in COL4A1-related disease [33,34,35,36].
• Reaction to microgliopathy and neuronal loss, often seen in more complex encephalopathies such as Cockayne syndrome where microscopic examination has highlighted the presence of both iron and calcium deposits, or the NRROS-related disorders, Adams–Oliver syndrome and Nasu–Hakola disease [37].
• Mitochondriopathies, where brain calcification may be attributed to co-present endocrine alterations (hypoparathyroidism) and/or to progressive cell degeneration caused by mitochondrial dysfunction in the brain [38].

The underlying pathogenetic mechanisms of ectopic/dystrophic calcifications of non-cerebral soft tissues are extremely different and are mainly associated with systemic mineral imbalance (i.e., hyperparathyroidism) or tissue alteration and necrosis (i.e., myositis ossificans) [39].