Weighing the evidence for the roles of plasma versus local pyrophosphate in ectopic calcification disorders

Abstract

Ectopic calcification is characterized by inappropriate deposition of calcium mineral in non-skeletal connective tissues and can cause significant morbidity and mortality, particularly when it affects the cardiovascular system. Identification of the metabolic and genetic determinants of ectopic calcification could help distinguish individuals at greatest risk of developing these pathological calcifications and could guide development of medical interventions. Inorganic pyrophosphate (PP_i) has long been recognized as the most potent endogenous inhibitor of biomineralization. It has been intensively studied as both a marker and a potential therapeutic for ectopic calcification. Decreased extracellular concentrations of PP_i have been proposed to be a unifying pathophysiological mechanism for disorders of ectopic calcification, both genetic and acquired. However, is reduced plasma concentrations of PP_i a reliable predictor of ectopic calcification? This perspective article evaluates the literature in favor and against a pathophysiological role of plasma versus tissue PP_i dysregulation as a determinant of, and as a biomarker for, ectopic calcification.

Ectopic calcification and the role of inorganic pyrophosphate

Inorganic pyrophosphate (PP_i) was first identified in 1961 as a potent inhibitor of in vitro calcium phosphate precipitation.⁽¹⁾ It was subsequently found that PP_i is a normal component of urine and plasma, important for the prevention of calcification of collagen, elastin and other nucleation sites in vivo.^(2,3) Subcutaneous injection of PP_i inhibited vitamin D-induced aortic calcification in rats.⁽⁴⁾ PP_i was shown to be a potent, mineral-binding, small molecule inhibitor of crystal nucleation and growth.⁽⁵⁾ Micromolar concentrations of PP_i were sufficient to inhibit the deposition of hydroxyapatite crystals in connective tissues in the presence of millimolar concentrations of calcium (Ca) and phosphate (P_i).⁽⁶⁾ Although the mechanism by which PP_i inhibits hydroxyapatite crystal growth is not entirely known, PP_i is thought to adsorb specifically to crystal growth sites, thus preventing further apposition of mineral ions. This mechanism of inhibiting mineralization has led to the description of PP_i as a physiological “water-softener” that prevents pathological soft tissue calcification.^(7,8) By contrast, degradation of PP_i to P_i in bones and teeth by tissue-nonspecific alkaline phosphatase (TNAP) enables the physiological mineralization of these tissues.⁽⁹⁾

PP_i is a by-product of many intracellular metabolic reactions. PP_i is also produced in the extracellular matrix of most tissues, where it acts as a potent inhibitor of mineral nucleation and growth. The ABCC6, ENPP1, ANKH, CD73, and TNAP proteins are key players of extracellular PP_i metabolism (Figure 1). Under physiological conditions, ABCC6 mediates the release of ATP primarily from hepatocytes into the extracellular space, whereas ANKH extrudes ATP from cells in a wide variety of tissues including joints, bones, and kidney. Extracellular ATP is converted into AMP and PP_i by ENPP1, an ectonucleotide pyrophosphatase/phosphodiesterase 1.^(10,11) ENPP1 is the principal enzyme that hydrolyzes extracellular ATP to AMP and PP_i, as circulating levels of PP_i are low or absent in mice and GACI type 1 patients with genetic deficiency of ENPP1. ABCC6 activity is responsible for about 60–70% of plasma PP_i concentrations while ANKH accounts for about 20% of plasma PP_i.^(12,13) CD73 converts AMP to P_i and adenosine, the second of which inhibits TNAP-mediated hydrolysis of PP_i to P_i. TNAP is the key enzyme involved in the breakdown of PP_i. TNAP is also a very potent ATPase, able to generate 3 molecules of P_i as ATP is hydrolyzed sequentially into ADP, AMP and adenosine.^(9,14,15) Loss-of-function mutations in the TNAP protein cause hypophosphatasia (HPP), a rare inherited disorder of bone and mineral metabolism manifesting osteomalacia and rickets due to excessive levels of extracellular PP_i.⁽¹⁶⁾ Thus, extracellular PP_i homeostasis is regulated by a balanced action of ABCC6, ANKH, and ENPP1 that promote PP_i synthesis, and the opposing action of TNAP that catalyzes the degradation of ATP, ADP, AMP and PP_i (Figure 1).

Figure 1. Schematic illustration of the synthesis and hydrolysis of PP_i, a potent inhibitor of pathologic tissue calcification.

The underlying mechanisms of ectopic calcification remain poorly understood despite the current understanding of a seemingly linear biochemical pathway revealed by the coordinated actions of several proteins regulating the homeostasis of extracellular PP_i. Autosomal recessive diseases, PXE, ankylosis, GACI, ACDC, and HPP, are caused by loss-of-function mutations in ABCC6, ANKH, ENPP1, CD73 and TNAP proteins, respectively. ABCC6 and ANKH facilitate ATP release from hepatocytes and peripheral cells, respectively, to extracellular milieu. Extracellularly released ATP is converted by ENPP1into AMP and PP_i, the latter being an anti-calcification factor. CD39 competes with ENPP1 for the extracellular ATP and converts ATP into ADP and further ADP to AMP and two molecules of the pro-calcification molecule P_i. CD73 converts AMP to P_i and adenosine, the latter inhibits TNAP, an enzyme that hydrolyzes ATP and PP_i. TNAP also hydrolyzes ATP to ADP, ADP to AMP, and AMP to adenosine. Deficiencies in ABCC6, ANKH, ENPP1 and CD73 proteins lead to reduced plasma PP_i levels promoting pathologic calcification. By contrast, deficiency in TNAP causes defective bone mineralization in HPP due to excessive plasma PP_i levels. An equilibrium between concentrations of extracellular PP_i and P_i regulates ectopic calcification and normal skeletal mineralization.

Several genetic disorders that feature ectopic calcification are associated with dysregulated extracellular PP_i homeostasis.⁽¹⁷⁾ They include pseudoxanthoma elasticum (PXE), generalized arterial calcification of infancy (GACI), ankylosis, and arterial calcification due to deficiency of CD73 (ACDC), all transmitted via autosomal recessive modes. PXE (OMIM 264800) is characterized by late-onset yet progressive accumulation of calcium mineral at ectopic sites, including the skin, eyes, and cardiovascular system.⁽¹⁸⁾ The classical forms of PXE are caused by loss-of-function mutations in the ABCC6 gene, encoding a transmembrane efflux transporter ABCC6 protein expressed primarily in the liver. GACI is an extremely severe, early-onset vascular calcification disease often diagnosed by prenatal ultrasound revealing calcium deposits in the fetal heart and arteries.⁽¹⁹⁾ Based on the mutated genes, GACI is classified into two types, GACI type 1 (OMIM 208000) and GACI type 2 (OMIM 614473) which are associated with inactivating mutations in the ENPP1 and ABCC6 genes, respectively. Regardless of the two types, most patients with GACI die before six months of age.^(20,21) In contrast to GACI, ankylosis and ACDC (OMIM 211800) are adult-onset calcification disorders of elderly individuals. Patients with ankylosis develop calcification of tissues and poorly perfused bodily fluids, such as cartilage, intervertebral disc, and synovial fluid of joints.⁽²²⁾ Patients with ACDC develop arterial calcification in the lower extremities and calcification in joint capsules of the hands and feet causing severe pain.⁽²³⁾ Ankylosis and ACDC are caused by biallelic inactivating mutations in the ANKH and NT5E genes, encoding ANKH (ANK in mice) and CD73 proteins, respectively.

While we don’t understand which components of PP_i metabolism act locally and which influence plasma PP_i, these hereditary disorders, PXE, GACI, ankylosis, and ACDC, are all characterized by reduced circulating levels of PP_i, and reduced concentrations of this important physiological inhibitor of mineralization is considered responsible for ectopic calcification in soft connective tissues.⁽¹⁷⁾ This perspective article summarizes the role of circulating PP_i in genetic and acquired disorders associated with ectopic calcification. We refer to systemic levels of PP_i as measured by plasma PP_i and by local PPi we mean PP_i produced in situ at the cellular level. While the plasma PP_i assay has been optimized, it still remains largely a research tool at the moment, although it may become more widely available in the near future. To our knowledge however, there are no assays that can measure extracellular PP_i in vivo or in tissue sections, so this remains a major difficulty in this field. In our research papers, we have sometimes measured PP_i output in primary cultures as a surrogate for local production of PP_i.^(24–26) Due to the lack of standardization of blood collection, processing, and assay method, there are inconsistencies in the plasma PP_i concentrations reported by different laboratories, which might obscure the interpretation of the role of plasma PP_i in ectopic calcification.

Experimental models where plasma PP_i was a reliable predictor of ectopic calcification

Vascular calcification in GACI correlates with plasma PP_i levels

Enpp1^−/− knockout mice (Enpp1^tm1Gdg), a rodent model of human GACI, have essentially no detectable plasma PP_i and develop extensive aortic calcification by two months of age.⁽²⁷⁾ Allografts of aorta from wild-type mice into Enpp1^−/− mice resulted in calcification of the donor tissue, but the amount of calcification was less than the adjacent endogenous Enpp1^−/− aorta. Abnormal aortic tissue from Enpp1^−/− mice showed no further calcification after transplantation into wild-type mice. These results showed that normal circulating levels of PP_i overcame the absence of local PP_i and prevented vascular calcification, but also that the role of local PP_i production in the vascular tissue could partially prevent calcification in the presence of low plasma PP_i.⁽²⁷⁾

Genetic overexpression of human ENPP1 and administration of an ENPP1-Fc enzyme restore plasma PP_i levels and completely prevent ectopic calcification in Enpp1 mutant mice

Homozygous Enpp1^asj mice, so-named because they develop a peculiar joint stiffness as they age (“ages with stiffened joints,” asj), represent a useful model of human GACI.⁽²⁸⁾ These mice harbor a p.V246D missense mutation in ENPP1 resulting in absence of ENPP1 protein. Accordingly, homozygous Enpp1^asj mice had low plasma PP_i levels, widespread ectopic calcification, defects in bone quality and mineralization, and early mortality.⁽²⁸⁾ Crossing the Enpp1^asj mice with transgenic mice ubiquitously overexpressing a human ENPP1 transgene produced Enpp1^asj mice that had wild-type plasma concentrations of PP_i and which did not develop ectopic calcification or bone mineralization defects.⁽²⁹⁾

Similarly, enzyme replacement therapy with injections of recombinant ENPP1-Fc enzyme in Enpp1^asj and Enpp1^ttw (the “tip toe walking” mouse is another model of GACI due to genetic deficiency of ENPP1) mice normalized plasma levels of PP_i, prevented ectopic calcification, improved bone mineralization, and prevented mortality.^(30,31) ENPP1-Fc enzyme replacement therapy also prevented neointimal proliferation and improved blood pressure and cardiovascular functions in Enpp1^ttw and Enpp1^asj−2J mice (the asj-2J mutation is allelic to asj).^(32,33) An ENPP1-Fc enzyme, INZ-701, is currently being studied in a phase 1/2, multi-center, open-label, first-in-human clinical trial in GACI type 1 patients (ClinicalTrials.gov identifier NCT04686175).

Human ABCC6 transgene restores plasma PP_i and prevents ectopic calcification in Abcc6^−/− mice

Abcc6^−/− knockout mice serve as a model of PXE. After expressing low level of human ABCC6 transgene in the liver, the Abcc6^−/− mice had a modest 27% increase in plasma PP_i levels that led to a significant reduction in acute dystrophic cardiac calcification and chronic skin calcification.⁽³⁴⁾ Similarly, injection of young Abcc6^−/− mice with recombinant adenovirus to achieve liver-targeted expression of human ABCC6 cDNA prior to the development of ectopic calcification was associated with sustained hepatic expression of human ABCC6 protein for up to 4 weeks, which normalized plasma PP_i levels and prevented tissue calcification.⁽³⁵⁾

Plasma PP_i levels correlate with age of onset and severity of ectopic calcification in genetic disorders of PXE, GACI, and ACDC

The diagnosis of GACI is often made by prenatal or perinatal ultrasound and/or CT imaging that shows echo brightness or calcium deposition in vascular tissues. ENPP1-deficient patients with GACI type 1 have plasma PP_i levels that are 0–10% of healthy controls,⁽³³⁾ indicating that most or all PP_i in plasma is derived from ENPP1-mediated hydrolysis of extracellular ATP and other nucleoside triphosphates. Due to extensive arterial calcification, GACI patients have a 55% mortality rate before the age of 6 months.⁽³⁶⁾ By contrast, PXE is late-onset with the skin, eye, and cardiovascular involvements not noticeable until the patients are in their 20s or 30s, and the majority of PXE patients have a normal life expectancy. Individuals diagnosed with PXE have plasma PP_i levels at 30–40% of wild-type controls.⁽¹⁰⁾ ACDC is an adult-onset disorder in which ectopic calcification primarily affects the joints and lower extremity arteries. While plasma PP_i levels have not been reported in patients with ACDC, circulating levels of PP_i in the CD73-deficient mouse model of ACDC are approximately 50% of that in wild-type controls.⁽³⁷⁾ These observations suggest that the varying degree of reduced plasma PP_i concentrations correlates, in general, with the onset and severity of the ectopic calcification phenotypes in PXE (60–70% reduction; late onset), GACI (90–100% reduction; prenatal or perinatal onset), and ACDC (50% reduction; adult onset) (Figure 1).

PP_i and bisphosphonates prevent ectopic calcification in preclinical and clinical studies

Administration of PP_i and non-hydrolyzable PP_i analogs such as the bisphosphonates prevented ectopic calcification, attesting to the importance of PP_i as a major regulator of hydroxyapatite deposition. Oral administration or subcutaneous injection of the first generation bisphosphonate etidronate increased bone mass and significantly reduced skin calcification in Abcc6^−/− and Enpp1^asj mice.^(38,39) Daily intraperitoneal injections of PP_i or etidronate that led to transient increases of plasma PP_i inhibited acute dystrophic cardiac calcification and prevented spontaneous skin calcification in Abcc6^−/− mice.⁽³⁴⁾ Oral administration of the tetra-sodium salt of PP_i to Abcc6^−/− mice and Enpp1^ttw mice by ad libitum access to supplemented drinking water attenuated skin, vascular, and cardiac calcification, and this PP_i salt is readily absorbed in humans.⁽⁴⁰⁾ Additionally, pregnant heterozygous Enpp1^ttw mice that drank tetra-sodium PP_i-containing water produced homozygous Enpp1^ttw offspring with significantly reduced skin and vascular calcification when assessed after weaning.⁽⁴⁰⁾ Delivery of di-sodium and potassium salt of PP_i in gelatin capsule provides an effective PP_i delivery system in humans and avoids excess sodium intake, which was previously found to cause gastro-intestinal discomfort.⁽⁴¹⁾ Although an inadvertent discovery, increasing the amount of PP_i in the chow to 7.1 times greater than the concentration in standard chow attenuated dystrophic cardiac calcification and skin calcification in Abcc6^−/− mice, providing additional evidence that dietary PP_i can modulate ectopic calcification.⁽⁴²⁾

Etidronate inhibited femoral artery calcification and subretinal neovascularization events in a double-blinded single-center one-year study consisting of 74 adult PXE patients.⁽⁴³⁾ A post-hoc analysis showed that etidronate significantly stopped calcification progression in all vascular beds except for the coronary arteries.⁽⁴⁴⁾

Plasma PP_i levels are low in other genetic and acquired disorders of ectopic calcification

Plasma PP_i concentrations are low in other clinical conditions in which ectopic calcification occurs, such as chronic kidney disease,^(45–47) systemic sclerosis,⁽⁴⁸⁾ and progeria.⁽⁴⁹⁾ The reduced plasma PP_i in a mouse model of progeria is caused by increased activities of TNAP and CD39 that enhance PP_i hydrolysis and reduce the availability of ATP for PP_i generation. As a corollary, intensive treatment with the combination of ATP, the TNAP inhibitor levamisole, and the CD39 inhibitor ARL67156, prevented vascular calcification and increased longevity in these mice.⁽⁴⁹⁾ PP_i and bisphosphonates also reduce arterial calcification in animals with renal failure,^(50,51) and pharmacological TNAP inhibition led to an increase in plasma PP_i that prevented aortic calcification in a murine model of chronic kidney disease-mineral and bone disorder.⁽⁵²⁾

Experimental models where plasma PP_i was not a reliable predictor of ectopic calcification

Patients who are ENPP1 heterozygotes or ABCC6 homozygotes display completely different phenotypes despite similarly reduced plasma PP_i levels

Recent studies suggest that some patients who are heterozygous for ENPP1 variants manifest early-onset osteoporosis.^(53,54) These individuals have plasma PP_i levels similar to ABCC6 homozygotes with PXE.⁽⁵³⁾ In addition, ABCC6-PXE patients have similar reductions in plasma PP_i levels to ABCC6-GACI type 2 patients.⁽⁵⁵⁾ Furthermore, although most patients with biallelic ENPP1 variants develop GACI type 1, some patients instead manifest typical PXE despite having similarly low plasma PP_i levels.⁽⁵⁶⁾ Plasma PP_i does not fully explain the extent of ectopic calcification and divergent phenotypes of these patients, suggesting the involvement of local, cell-specific effects of ENPP1 and ABCC6 on various tissues, and the compounding effects of systemic factors such as plasma Ca, Pi, and Mg, and fetuin-A levels.^(57–59)

Genetic overexpression of human ENPP1 and administration of an ENPP1-Fc enzyme restore plasma PP_i levels but do not completely prevent ectopic calcification in Abcc6^−/− mice

Crossing Abcc6^−/− mice with transgenic mice carrying the human ENPP1 cDNA generates Abcc6^−/− mice with normal plasma levels of PP_i. These mice still developed ectopic calcification although it was significantly less than the Abcc6^−/− mice.⁽²⁹⁾ The same results were observed in Abcc6^−/− mice after administration of ENPP1-Fc, a recombinant ENPP1 enzyme. Ectopic calcification was not completely prevented even with a high dose of ENPP1-Fc that caused plasma PP_i levels to exceed those of wild-type mice.⁽⁶⁰⁾ These results suggest that ENPP1 did not correct the underlying ABCC6 deficiency, and suggest that local PP_i levels may be determinant in those tissues affected by ectopic calcification in the PXE mice.

Novel mouse models carrying the p.H362A and p.T238A ENPP1 missense mutations develop distinct phenotypes albeit low plasma PP_i levels

The characterization of the Enpp1^asj mouse model of GACI has shed light on the catalytic activity of the ENPP1 protein through its ability to hydrolyze ATP to generate PP_i, thereby regulating soft tissue calcification and bone mineralization. In addition to ATP, ENPP1 was recently found to hydrolyze extracellular cyclic GMP-AMP (2’,3’-cGAMP), an agonist for the stimulator of interferon genes (STING) that is evaluated as a target in cancer immunotherapy.⁽⁶¹⁾ The distinct roles of ENPP1’s catalytic activities on ATP and cGAMP on plasma PP_i, ectopic calcification, bone mineralization, and immune responses were recently analyzed using two mouse models with different missense mutations in the ENPP1 protein, p.H362A and p.T238A, that were generated by CRISPR-Cas9 genome editing.^(62,63) These mutant mice have striking biochemical and phenotypical differences from the Enpp1^asj mice (Table 1). Because the p.V246D mutation in the Enpp1^asj mouse leads to a deficiency of ENPP1 protein, these mice are unable to hydrolyze both ATP and cGAMP. By contrast, mutant mice with p.H362A or p.T238A mutation express the mutant ENPP1 proteins at levels that are similar to the wild-type protein in normal mice. When analyzed in vitro, the ENPP1 p.H362A mutant cannot hydrolyze cGAMP while maintaining the hydrolysis activity of ATP to produce AMP and PP_i.⁽⁶²⁾ By contrast, ex vivo hydrolysis of ATP was similarly reduced in plasma from Enpp1^asj and the p.H362A mutant mice, and plasma concentrations of PP_i were comparably reduced in both mutant mouse lines relative to wild-type mice. Remarkably, despite having plasma levels of PP_i that are as low as in the Enpp1^asj mice, the p.H362A mutant mice do not exhibit ectopic calcification and defective bone mineralization seen in Enpp1^asj mice. Exogenously added ATP was, however, degraded in the liver lysate of the p.H362A mice, suggesting that tissue PP_i rather than circulating PP_i prevents the development of ectopic calcification in these mice.⁽⁶²⁾ While cGAMP is not responsible for the devastating vascular calcification phenotype in GACI, it was shown to be responsible for antiviral immunity and systemic inflammation.⁽⁶²⁾ The ENPP1 p.T238A mutant lacks catalytic activity for hydrolysis of both ATP and cGAMP, allowing for ENPP1 protein signaling to be studied uncoupled from its enzymatic activity. These mice have plasma PP_i as low as in the Enpp1^asj mice, yet they develop normal trabecular bone microarchitecture, have favorable bone biomechanical properties, and develop less severe joint calcification than Enpp1^asj mice.⁽⁶³⁾

Table 1.

Biochemical findings and clinical features of mouse models with different mutations in the Enpp1 gene

Characteristics	Mouse models with Enpp1 mutations
	p.V246D (asj)	p.H362A	p.T238A
Biochemical features:
Is ENPP1 protein made	No	Yes	Yes
Hydrolysis of ATP	No	Yes	No
Hydrolysis of 2’3’-cGAMP	No	No	No
Plasma PPi	↓	↓	↓
Plasma Pi	↓	Similar to WT mice	↓
Plasma FGF23	↑	Not reported	↑
Plasma PTH	↑	Not reported	↑
Phenotypical features:
Breeding and lifespan	Difficulty breeding beyond 2–3 mo	Normal lifespan and readily breed	Increased mortality beyond 20 wks
Stiffened posture	Yes	No	Less severe than asj mice
Ectopic calcification	Wide spread, many tissues affected	Very minimal (only muzzle, ear, and airway analyzed)	Less severe than asj mice (only joints of forepaws analyzed)
Vascular stenosis	No	Not reported	Not reported
Bone (μCT, histomorphometry)	↓BMD (trabecular and cortical bone)	Not reported	Normal trabecular bone, Moderate changes of cortical bone
Bone (mechanic property)	↓	Not reported	Maximal load similar to WT mice, stiffness and total work less severe than asj mice
Calvarial mineralization	↓	Not reported	↑
Wnt inhibitor Sfrp1 expression	↑	Not reported	↓

Overexpression and inhibition of TNAP drives and inhibits ectopic calcification, respectively, without altering plasma PP_i levels

TNAP is a known regulator of eutopic and ectopic calcification. The role of TNAP in soft tissue and bone mineralization through hydrolysis of PP_i has been well studied. TNAP deficiency occurs in patients with hypophosphatasia (HPP), a genetic disorder characterized by impaired mineralization of the skeleton and teeth featuring rickets and early loss of teeth in children or osteomalacia and dental problems in adults caused by accumulation of extracellular PP_i concentrations.⁽¹⁶⁾ As a corollary, one would postulate that increased TNAP expression would enhance PP_i hydrolysis and thereby reduce plasma PP_i levels, with consequent development of ectopic calcification. The characterization of three TNAP transgenic mice point to a more complicated scenario. Transgenic mice in which overexpression of human TNAP targeted to osteoblasts had serum alkaline phosphatase activity 10–20 times of wild-type mice. These mice exhibited increased bone mineralization, yet their plasma PP_i levels were not different from wild-type mice.⁽⁶⁴⁾ The vascular smooth muscle cell-specific and endothelial cell-specific human TNAP transgenic mice had serum alkaline phosphatase activity that was 4–20 and 20–35 times of the wild-type mice, respectively.^(65,66) These mice had normal plasma PP_i levels, yet they developed severe arterial calcification.^(65,66) The results suggest that normal plasma PP_i levels do not explain the occurring of ectopic calcification, and suggest instead that local tissue levels of PP_i determine hydroxyapatite deposition in calcification-prone tissues. The lack of an apparent effect of those higher levels of plasma TNAP on plasma PP_i could be attributed in part to the fact that TNAP is physiologically inhibited by normal plasma concentrations of P_i.⁽⁶⁷⁾

When TNAP was overexpressed instead via administration of an AAV8-hTNAP-D₁₀ vector targeting TNAP expression in the bone, serum alkaline phosphatase activity was increased 500–800 times in the wild-type and Alpl^−/− mice.⁽⁶⁸⁾ Plasma PP_i was then significantly suppressed in these mice, yet there was no ectopic calcification in the kidneys, aorta, coronary arteries, and brain in a 70-day follow-up.⁽⁶⁸⁾ Similar findings were reported for asfotase alfa, a recombinant and bone-targeted TNAP enzyme currently used for the treatment of patients with HPP.⁽⁶⁹⁾ This, however, is not unexpected given the documented dual requirement for the local co-expression of TNAP in the context of a collagenous matrix as necessary and sufficient to mineralize any extracellular matrix.⁽⁷⁰⁾

Genetic and pharmacologic inhibition of TNAP via Abcc6^−/−Alpl^+/− mice (with ~50% reduction of serum alkaline phosphatase activity) and the pharmacological TNAP inhibitor SBI-425, respectively, showed significant inhibition of ectopic calcification in Abcc6^−/− mice.⁽⁷¹⁾ In an independent study, SBI-425 also attenuated tissue calcification in Abcc6^−/− mice without altering bone mineralization.⁽⁷²⁾ In these studies, plasma PP_i levels were not altered, suggesting the contribution of local events, such as increased tissue levels of PP_i, to the therapeutic effects of TNAP inhibition.

Plasma PP_i levels do not correlate with the degree of ectopic calcification and disease severity in PXE patients

A study was performed with 107 French PXE patients and 26 healthy controls investigating the relationship between plasma PP_i levels, coronary calcification, lower limbs arterial calcification, and disease severity by Phenodex score.⁽⁷³⁾ The results of this study show that plasma PP_i concentration is lower in PXE patients compared to healthy controls and that arterial calcification is strongly dependent on age but not on plasma PP_i levels.⁽⁷³⁾ Correlations of plasma PP_i levels and serum alkaline phosphatase activities are reported with conflicting results in PXE patients.^(73,74)

Plasma PP_i is not a determinant of serum T50 measuring serum calcification propensity in PXE patients

Serum samples from a cohort of 57 Belgian PXE patients were evaluated via the T50 serum calcification propensity assay.⁽⁷⁵⁾ T50 was found to be an independent predictor of ocular, vascular, and overall disease severity in PXE. Multivariate regression analysis identified serum levels of fetuin-A, phosphorus, and magnesium but not PP_i as significant determinants of T50.⁽⁷⁵⁾ The serum T50 score, however, does not measure apatite formation, the site of action of PP_i. Furthermore, this test was done in serum, which does not reflect plasma PP_i due to release of PP_i from platelets. Thus, PP_i would not be expected to alter the serum calcification propensity score.

Conclusions and future prospects

Identifying genetic and metabolic determinants for ectopic calcification is of paramount importance toward understanding the drivers and inhibitors of the ectopic calcification process in general. Unquestionably, PP_i is a critical mediator of ectopic calcification. There are, however, emerging data suggesting that circulating PP_i may not be a good proxy for extracellular PP_i levels in tissues prone to calcification, and that changes in local PP_i concentrations may exert a profound role controlling ectopic calcification. Unfortunately, there is currently no method for determining the level of extracellular PP_i in tissues. Such an assay, if developed, could help answer some of the unexplained findings. In addition to PP_i, ectopic calcification is also influenced by Ca, P_i, Mg, and fetuin-A levels in the blood and diet and variations in those may also affect comparisons between studies and models.

Data availability statement

Data sharing not applicable - no new data generated.