ARTICLE
Received 11 May 2015 | Accepted 23 Oct 2015 | Published 1 Dec 2015
Ronald A. Albright1,*, Paul Stabach1,*, Wenxiang Cao2, Dillon Kavanagh1, Isabelle Mullen1,Alexander A. Braddock1, Mariel S. Covo1, Martin Tehan1, Guangxiao Yang1, Zhiliang Cheng3, Keith Bouchard3, Zhao-Xue Yu3, Stephanie Thorn4, Xiangning Wang4, Ewa J. Folta-Stogniew5, Alejandro Negrete6,Albert J. Sinusas4, Joseph Shiloach6, George Zubal7, Joseph A. Madri1, Enrique M. De La Cruz2& Demetrios T. Braddock1
Diseases of ectopic calcication of the vascular wall range from lethal orphan diseases such as generalized arterial calcication of infancy (GACI), to common diseases such as hardening of the arteries associated with aging and calciphylaxis of chronic kidney disease (CKD). GACI is a lethal orphan disease in which infants calcify the internal elastic lamina of their medium and large arteries and expire of cardiac failure as neonates, while calciphylaxis of CKD is a ubiquitous vascular calcication in patients with renal failure. Both disorders are characterized by vascular Mnckeburgs sclerosis accompanied by decreased concentrations of plasma inorganic pyrophosphate (PPi). Here we demonstrate that subcutaneous administration of an
ENPP1-Fc fusion protein prevents the mortality, vascular calcications and sequela of disease in animal models of GACI, and is accompanied by a complete clinical and biomarker response. Our ndings have implications for the treatment of rare and common diseases of ectopic vascular calcication.
1 Department of Pathology, Yale University School of Medicine, New Haven, Connecticut 06510, USA. 2 Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, Connecticut 06520, USA. 3 Alexion, Cheshire, Connecticut 06410, USA. 4 Department of Medicine and Diagnostic Radiology, Yale University School of Medicine, New Haven, Connecticut 06510, USA. 5 WM Keck Biotechnology Research Laboratory, Yale University School of Medicine, New Haven, Connecticut 06511, USA. 6 Biotechnology Core Laboratory, National Institute of Diabetes and Digestive Kidney Diseases, National Institutes of Health, Rockville, Maryland 20852, USA. 7 Z-Concepts LLC, East Haven, Connecticut 06512, USA. * These authors contributed equally to the work. Correspondence and requests for materials should be addressed to D.T.B. (email: mailto:[email protected]
Web End [email protected] ).
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 1
DOI: 10.1038/ncomms10006 OPEN
ENPP1-Fc prevents mortality and vascular calcications in rodent model of generalized arterial calcication of infancy
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006
Generalized arterial calcication of infancy (GACI) is an ultra-rare neonatal disease characterized by infantile onset of widespread arterial calcications in large and medium
sized vessels resulting in cardiovascular collapse and death in the neonatal period. The disease presents clinically with heart failure, respiratory distress, hypertension, cyanosis and cardiomegaly. The prognosis is grave, with older reports of an 85% mortality rate at 6 months1, while recently intensive treatment with bisphosphonates has lowered mortality to 55% at 6 months2. Tempering this apparent progress is the severe skeletal toxicity associated with prolonged use of etridonate in patients with GACI (ref. 3), the observation that the limited available data makes it difcult to determine if bisphosphonate treatment is truly protective or reects the natural history of the disease in less effected patients, and the ineffectiveness of bisphosphonates to prevent mortality in some patients even when instituted early47.
The overall incidence of GACI is rare, with 200 reported cases in the medical literature8 and a disease frequency of one in 391,000 (ref. 9). Although the disease was rst described by Bryant and White in 1901 (ref. 10), it was not until 2000 that Rutsch et al.11 noted that serum plasma inorganic pyrophosphate (PPi) levels and ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) enzymatic activity were signicantly impaired in GACI patients. ENPP1 (also known as PC-1) is the founding member of the ENPP or NPP family of enzymes, which
are characterized by phosphodiesterase activity12, and is a type II extracellular membrane bound glycoprotein located on the mineral-depositing matrix vesicles of osteoblasts and chondrocytes, as well as the vascular surface of cerebral capillaries13. ENPP1 is the primary source of extracellular PPi in the body, and hydrolyzes extracellular ATP into AMP and PPi (refs 12,14). PPi acts as a potent inhibitor of mineralization, presumably by occupying some of the Pi sites on the surface of nascent or growing hydroxyapatite crystals, thereby creating irregularities that slow or terminate crystal growth14,15. Rutsch et al. discovered that inactivating mutations in ENPP1 account for 75% of GACI patients2,11,16 and have contributed to recent observations that a sizable fraction of the remaining patients result from inactivating mutations in the ATP dependent membrane transporter ABCC6 (refs 1722). ABCC6 mutations result in decreased extracellular concentrations of nucleoside triphosphates through an unknown mechanism23, thereby limiting ENPP1s metabolism of ATP into extracellular PPi. The
identication of both ENPP1 and ABCC6 inactivating mutations as genetic etiologies for GACI therefore links the same biochemical pathway to unregulated tissue mineralization in both genetic variants of GACI.
Despite the multiple genetic etiologies and multifactorial nature of the expression, progression, and severity of medial wall vascular calcication, we hypothesized that disruption of
a b c
23 21 19 17 15 13 11
9 7 5
NPP1-WT (n=9)
100
50
0
0 20 40 60 80 Days elapsed
WT, n=14
Aortic Arch
R. Ventricle
Weight (g)
Percent survival
Mean lifespan
NPP1 asj/asj =58 days
asj, n=8
P<0.001
20 30 40 50 60 70
NPP-asj/asj (n=8)
Age (days)
(2)
d
f g
Papillary muscle
Scar tissue
Scar tissue
20 micrometers
20 micrometers
L.V.
e
10 m 10 m
Calcifications Calcifications
Normal myocardium
Figure 1 | Natural history study. (a) Daily weights of enpp1wt (cyan squares, n 9 animals) and enpp1asj/asj mice (green circles, n 8 animals) on the
acceleration diet in utero over a 70 day period. Mean average weights are plotted with s.ds denoted by error bars. A failure to thrive point is noted in the enpp1asj/asj cohort at day 26, when the weights diverge from enpp1wt. Death events are marked with red arrows. (b) Mean survival of enpp1asj/asj was 58
days. No deaths were observed in the enpp1wt cohort. Analysis by log-rank (MantleCox) test yields a w2 of 15.73 and P value of o0.0001. (c) enpp1asj/asj animals displayed dramatic calcications in heart and aorta visible on micro-CT scans. (d,e) Histology of enpp1asj/asj mice, aorta (Hematoxylin and Eosin
(H&E) and Alizarin red). Aortas of enpp1asj/asj mouse all possessed near circumferential calcications that were pervasive in the vascular walls, as illustrated by Alizarin red staining of the aortas. Scale bar, 20 mm (f). Histology of enpp1asj/asj mice, left ventricle. Extensive calcications surrounded by scar tissue revealing the presence of repeated, old and healed myocardial infarctions. Scale bar, 10 mm (g). Histology of enpp1asj/asj Mice, Septum (H&E). More typically, the enpp1asj/asj mice displayed small foci of calcications with surrounding scar tissue as seen here in the myocardial septum, also diagnostic of previous myocardial infarctions. Scale bar, 10 mm.
2 NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006 ARTICLE
Table 1 | Cardiovascular pathology, natural history study.
WT asj/asj Calcications in heart (CT/histology) 0/0 37%/100%
Calcications in aorta (CT/histology) 0/0 62%/100% Calcications in coronary arteries (histology) 0/0 100% Myocardial infarction (histology) 0 70%
CT, computed tomography; WT, wild-type.
NPP1ATP AMP + PPi
Crystallization
Inhibit
Inhibit
a
Activate
Pi
TNAP
Extracellular space
CD73
Pi + adenosine
ATP
ABCC6
ENPP1s extracellular purinergic metabolism accounts for the pathologic sequela and mortality associated with GACI and that enzyme replacement therapy with ENPP1 is a tractable therapeutic approach. To test this hypothesis, we employed ENPP1-asj mice (Jackson Laboratory), which carry an inactivating mutation in enpp1 such that homozygotes (enpp1asj/asj) have markedly reduced ENPP1 function. To enhance the disease phenotype, the mice were maintained on the an acceleration diet in utero24,25. Here we demonstrate that daily subcutaneous doses of a biologic agent comprising the extracellular domain of ENPP1 fused to the Fc region of human IgG1 eliminates the vascular calcications, mortality, myocardial infarctions, and sequela of GACI in enpp1asj/asj mice, and that this clinical response is accompanied by normalization of serum and tissue PPi concentrations.
ResultsNatural history study of GACI in enpp1asj/asj mice. When fed an acceleration diet, the daily weights of enpp1asj/asj mice diverged from wild-type (WT) siblings pairs at day 26, when the ENPP1-asj/asj mice experienced a failure to thrive event and began to lose weight (Fig. 1a). After day 26 the enpp1asj/asj animals displayed progressive stiffness and reductions in physical activity. All of the enpp1asj/asj animals died between days 3571, with a median lifespan of 58 days (Fig. 1b). The presence of calcications in enpp1asj/asj and enpp1wtj mice was evaluated via postmortem by micro-computed tomography (CT) scans (Fig. 1c) and histologic sections taken from the heart, aorta and kidneys (Fig. 1dg). Approximately one-third of the enpp1asj/asj mice had visible calcications in their hearts, and two-thirds had calcications in their aortas visible by micro-CT imaging (Table 1). These percentages increased to 100% on histologic examination, which also showed that many of the animals had dramatic nearly circumferential calcications in their aortic walls (Fig. 1d,e). Histologic examination also revealed that 100% of the coronary arteries possessed arterial wall calcications and that 70% of the animals had focal or conuent areas of myocardial necrosis consistent with myocardial infarction (Fig. 1f,g). Conversely, the enpp1wt mice displayed none of these abnormalities. These ndings demonstrate that the animal model recapitulates GACI in humans, which is characterized by prominent calcications of the large and medium sized arteries and a cardiac demise.
Design and characterization of therapeutic. To produce soluble, recombinant ENPP1 for in vivo use, we modied our previous expression construct26 by fusing ENPP1 to the Fc domain of IgG1 (hereafter referred to as hENPP1-Fc and mENPP1-Fc for the human and mouse protein isoforms, respectively, Fig. 2b) and expressed the fusion protein in stable mammalian (HEK293) cell lines (Supplementary Note 1). The effect of expressing the hENPP1-Fc in mammalian, as compared to our previous studies in insect cells26, altered the MichaelisMenton kinetics by reducing (tightening) the Michaelis constant KM for ATP
substrate by over two orders of magnitude, while also reducing (slowing down) the kcat by a factor of 23 (Fig. 3ac). The altered
MichaelisMenton constants could reect either the change in expression system from baculovirus-infected insect cells to mammalian cells or the addition of the Fc tag. To differentiate between these two possibilities, we expressed hENPP1 without an Fc tag in mammalian (HEK293) cell lines to directly compare the kinetics of ENPP1 with and without an Fc tag. We found that the kinetic constants of the hNPP1-Fc and hNPP1 enzymes expressed in mammalian cells were essentially identical (Fig. 3d), suggesting that the alteration in kinetic constants was due to the change in expression systems.
To understand the pharmacokinetics of ENPP1-Fc we determined plasma concentrations of drug following a single subcutaneous injection over time. Fifteen mice were administered a single subcutaneous dose of 10 mg kg 1 and blood was collected from tail veins to quantitate ENPP1-Fc in the plasma.
The concentration ENPP1-Fc reaches a maximal plasma concentration of B300 nM (Cmax) within 8 h (tmax) and remains elevated at 100 nM for 72 h after dosing (Fig. 4a). The integral of the plotted concentrationtime curve, or area under the curve, is 9 mM h 1. Plotting the data as fraction of drug absorbed over time allows for the determination of the elimination (ke) and
absorption (ka) constants by tting the data to the equation for the total systemic absorption of a drug administered at a
ATP Intracellular space
b
925 Fc IgG
NPP1
NPP2
1 77 98
1 28 36
Figure 2 | Metabolic pathways and therapeutic design. (a) Metabolic pathways of interest. ENPP1 converts extracellular ATP into AMP and PPi,
TNAP converts PPi into Pi, and CD73 converts AMP into adenosine and Pi. ABCC6 is a membrane transporter that increases the extracellular concentration of NTP via an indirect unknown mechanism (shown as a dashed line). Loss of function mutations in TNAP result in familial hypophosphatasia. Loss of function mutations in ENPP1 result in GACI, loss of function mutations in ABCC6 result in pseudoxanthomatous elasticum, and loss of function mutations in CD73 results in a disease of arterial and joint calcication termed ACDC. TNAP, tissue-nonspecic alkaline phosphatase. (b) Design of ENPP1 protein therapeutic. To produce a soluble recombinant protein, a segment of the extracellular region of NPP2 containing a furin cleavage site was substituted into ENPP1 (ref. 26) and the Fc portion of IgG1 was appended to the c-terminus.
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 3
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006
a b
300
150
250
200
[AMP] (M)
[AMP] (M)
100
150
100
50
50
0
0 5 10 15 20 25 30 35
0 0 10 20 30 40
Time (min)
Time (min)
c d
7
8
sf9
6
6
5
Initial rate
(s1 per enzyme)
4
Initial rate (s1 per enzyme)
[Initial ATP] (M) [Initial ATP] (M)
4
3
2
2
HEK293, Fc
HEK293, no Fc
1
0 0 100 200 300 400 500
0 1 10 100 1,000
Figure 3 | Steady state kinetics of hENPP1-Fc. Time courses of AMP formation measured by HPLC analysis after addition of (a) 50 nM hNPP1 puried from sf9 cells to (from bottom to top) 3.9 (black), 7.8 (red), 15.6 (green), 31.3 (blue), 62.5 (cyan), 125 (magenta), 250 (dark yellow) and 500 (navy) mM
ATP or (b) 10 nM hNPP1 from HEK293 cells to (from bottom to top) 7.8 (black), 15.6 (red), 31.3 (green), 62.5 (blue), 125 (cyan) and 250 (magenta) mM ATP. The smooth curves though the data are ts obtained by non-linear kinetic time course analysis39. (c) Comparison of ENPP1 expressed in insect cells with ENPP1-Fc expressed in mammalian cells. NPP1 (sf9, black circles) is human ENPP1 produced in baculovirus-infected cells while NPP1 (HEK293, Fc and blue squares) is human ENPP1-Fc produced in HEK293 cells. The initial rate of ATP cleavage by ENPP1-Fc from HEK293 cells is essentially the same at [ATP] greater than 7.8 mM, yielding a kcat (the average of the rates Z7.8 mM) of 3.4 (0.4) s 1 per enzyme. The initial rate at 2.0 mM ATP concentration is about a half of the kcat value. We therefore estimate a KM B 2 mM for ATP hydrolysis by hNPP1-Fc protein. (d) ATP concentration dependence of the initial hydrolysis rate of ENPP1 (no Fc) puried from HEK293 cells. The reaction displays a kcat of 3.46 (0.44) s 1 per enzyme, estimated from the average of all ATP concentrations Z2 mM. At 1 mM initial ATP (the rst data point), the hydrolysis rate is slightly less than kcat values. We therefore estimate a KM o2 mM for ATP hydrolysis by ENPP1 (no Fc) puried from HEK293 cells.
subcutaneous depot at time zero,
F
kaka ke
: 1
This analysis yield a ke 0.1070.016 h 1 and a ka 0.048
0.008 h 1, which yields an elimination half life (t1/2) 6.5 h
(Fig. 4b).
The activity of ENPP1-Fc was noted to diminish over a 30 day period when stored at 4 C, but the enzyme could be frozen at
80 C and retain nearly complete activity on thawing (Fig. 4c). The enzyme was therefore stored as a frozen stock solution after purication until needed. The oligomerization state of ENPP1-Fc at 1 mM and 100 nM was determined by size exclusion chromatography coupled with light scattering, refractive index, and absorbance (ultraviolet) detection (SEC-MALLS/RI/UV)27, which allows for a determination of the molecular weight (MW) of glycosylated proteins from the relationship of molecular weight and the ratio of laser light scattering, ultraviolet and refractive index (RI) signals in the absence of knowledge of the extend of glycosylation28. The results of the SEC-MALLS analysis revealed that the MW of the ENPP1-Fc protein represents a dimer over the concentration range of 100 nM1 mM, with an experimentally determined MW of 27438 KDa (Fig. 4d). This agrees well with the calculated MW of 252 KDa for an ENPP1-Fc dimer. The experimental approach also allows for an independent estimation of the glycosylation level of the fusion protein, which was determined to be B5% according to the method of Hayashi et al.29
Following purication, ENPP1-Fc was dialysed into PBS supplemented with Zn and Ca2 (PBSplus) concentrated to between 5 and 7 mg ml 1 and frozen at 80 C in aliquots of
200500 ml. Aliquots were thawed immediately before use, and the specic activity of the solution was adjusted to31.25 a.u. ml 1 (or E0.7 mg ml 1 depending on the preparation) by dilution in PBSplus.
Therapeutic proof of concept in enpp1asj/asj mice. Dosing was performed according to activity units (a.u.) per kg animal weight to account for variations in specic activity in different protein preparations. The specic activity of the enzyme varied with each protein preparation, and because the clinical response was noted to be highly dependent on enzyme specic activity, we rejected protein preparations with specic activities of o40 a.u. mg 1.
To establish initial dosing levels for the proof of concept study we performed dose escalation trials in limited numbers of animals (12 per dose level). While both the human and mouse version ENPP1 were used in the dose escalation trials, the proof of concept study was performed with the mouse isoform of ENPP1-Fc (mENPP1-Fc). enpp1asj/asj mice were dosed daily on the fourteenth day of life with subcutaneous injections of mENPP1-Fc and weekly with intra-peritoneal injections of GK 1.5, the latter added to minimize immune rejection of recombinant protein30. Subcutaneous doses of mEnpp1-Fc at 500 a.u. kg 1 qD demonstrated a strong early response in weight with an absence of the observed failure to thrive crisis observed in undosed enpp1asj/asj animals (Supplementary Fig. 1). This dose corresponded to between 610 mg kg 1 of ENPP1-Fc, depending on the specic activity of the protein preparation.
On the basis of the results of the dose escalation trials we chose to enrol a cohort of 8 enpp1asj/asj animals dosed with mENPP1-Fc at 500 a.u. kg qD and weekly intra-peritoneal injections with GK
e ket e kat
4 NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006 ARTICLE
a
500
Plasma [ENPP1-Fc] (nM)
b
Fraction absorbed
0.3
0.2
0.1
0.0
400
300
200
100
0
0
20 40 60 80 0 20 40 60 80
Time (hours) Time (hours)
c
d
Initial absorption change
rate (OD s1 )
0.0005
0.0004
0.0003
0.0002
0.0001
0.5 1.0 1.5 2.0 Added protein stock (l)
Abs 280 nm
1.41.21.00.80.60.40.2
0.2
0.0
10 12 14 16 18 Elution volume (ml)
300
200
100
0
440 220 156
Mw (kDa)
Frozen Unfrozen
Figure 4 | Characterization of ENPP1-Fc. (a) Pharmacokinetics of absorption and excretion. 15 C57B6 mice were injected with 10 mg kg 1 ENPP1-Fc and blood was collected at the indicated times. The concentration of ENPP1-Fc was estimated from the activity units of enzyme as measured by TMP assay and each data point represents measurement in 5 animals except for time 0, which represents 15 animals. The mean value is plotted, and error bars represent
s.d. in the measurements. A Cmax of E300 nM is reached at 8 h, and the area under the curve is calculated as 9 mM h 1. (b) To derive the pharmacokinetic constants of drug absorption and elimination we plotted the fraction of dosed drug absorbed in the plasma from a single 10 mg kg 1 subcutaneous dose over time and t the resulting curve with the equation F
kk k
e k t e k t
to obtain the elimination and absorption constants ke and ka, respectively.
The resulting curve yielded values for ke 0.1070.016 h 1 and ka 0.0480.008 h 1. The goodness of the t yielded an R2 0.7573 and an absolute
sum of squares 0.01. (c) Stability of ENPP1 therapeutic. ENPP1-Fc Ap3A activity was seen to be stable to freeze-thaw cycle in PBS following storage at 80 C. (d) Molecular weight of ENPP1-Fc determined by SEC-MALLS/RI/UV. Weight average molar masses (Mw) are plotted for analyses of ENPP1-Fc protein at various concentrations. Lines correspond to UV traces of the protein eluting from the SEC column monitored at 280 nm (left axis). Molar masses were recorded every sec across the elution prole (circles; right axis); for clarity only every 20th measurement of molar mass is plotted. ENPP1-Fc protein was analysed in a concentration range from 100 nM (dashed line, empty circles) to 1 mM (solid line, lled circles) measured at the apex of the eluting peak.
The elutions position of protein standards used for validation of SEC-MALLS/UV/RI performance are marked. The SEC-MALLS/UV/RI analysis yielded MW of 27438 KDa with a glycosylation estimate of 0.05 g of sugars per gram of polypeptide.
1.5 (Fig. 5). We included a control groups (enpp1wt and enpp1asj/asj) dosed daily with vehicle and weekly with GK 1.5 in an identical manner as the dosed cohort, and the study duration was shortened to 55 days. All 8 treated enpp1asj/asj animals survived the full 55 days of the trial, with a dramatic clinical response observed in treated animals (Supplementary Movie 1), while the median lifespan of the untreated enpp1asj/asj animals decreased from 58 to 35 days in the therapeutic trial, perhaps resulting from the weekly intra-peritoneal injections of the GK 1.5 immunosuppressant. The untreated enpp1asj/asj animals also all experienced a failure to thrive crisis at day 26, followed by weight loss and mobility restriction progressing variably to paralysis and death over the next 30 days. All but one untreated enpp1asj/asj animal expired over the 55 day trial, while in contrast all treated enpp1asj/asj mice gained weight comparable to the enpp1wt mice and displayed no signs of reduced mobility or stiffness (Supplementary Movie 1).
At the conclusion of the study, 100% of the enpp1asj/asj mice treated with vehicle displayed calcications in their hearts, aortas and coronary arteries, and 77% of the animals displayed histologic evidence of myocardial infarction (Table 2). In most cases this took the form of small areas of myocardial cell necrosis and single cell drop out in the vicinity of the cardiac calcications (Figs 5d and 6fh), but in two animals (22%) there were large, full thickness myocardial infarctions in the free wall of the right ventricle (Fig. 6ce). Myocardial brosis in the myocardial tissue
adjacent to regions of coronary artery calcication was a common nding (Fig. 6g,h), illustrating that ischaemia from coronary artery calcication likely accounts for a substantial burden of the myocardial disease. Some untreated enpp1asj/asj animals displayed dramatic calcications of coronary arteries, heart, ascending and descending aorta (Supplementary Movie 2). In contrast, none of the enpp1asj/asj animals treated with ENPP1-Fc displayed cardiac, arterial, or aortic calcication on histology or post-mortem micro-CT (Table 2 and Figs 5e, 6b and 7a). Although renal calcications are not a feature of human GACI, the enpp1asj/asj mouse model is noted to have calcications in the kidneys. Similar to previous ndings, we noted light calcications in the kidneys of E60% of the WT animals and 100% of the dosed enpp1asj/asj mice. The calcications in these animals were centred in the renal medulla. 100% of the undosed enpp1asj/asj animals had heavy, extensive calcications, centred in the outer medulla, with extension into the renal cortex. In comparison to previous reports, the renal calcications in the dosed animals seem to approximate renal calcications seen in the heterozygous enppWT/asj animals.
Biomarker response. In addition to survival, daily animal weights, and terminal histology, treatment response was also assessed via post-mortem high-resolution micro-CT scans to image vascular calcications, plasma PPi concentrations, and
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 5
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006
a
NPP1-WT + Vehicle n=8
22
20
14
8
18
16
Weight (g)
NPP1-asj/asj + 500 au/Kg qD n=8
NPP1-asj/asj + Vehicle n=8
12
10
5
20 40 50
30
Rt. coronary
Age (days)
b
c
Percent survival
100
50
0
0 20 40 60 Days elapsed
Mean SurvivalNPP1 asj/asj + vehicle
NPP1 asj/asj + 35 days
500 au/Kg n=8
NPP1 asj/asj + Vehicle n=8
NPP1-asj/asj untreated
NPP1-asj/asj treated
P<0.0001
d
Intracardiac calcifications
L.V.
e
L.V.
20 m
20 m
Papillary muscles
Papillary muscles
Figure 5 | Proof of Concept Study. (a) Daily animal weights. The mean daily weights of enpp1wt (cyan squares, n 8), treated enpp1asj/asj (purple
triangles, n 8), and untreated enpp1asj/asj mice (green circles, n 8) are
plotted with s.d.s denoted by error bars. Dosing with eNPP1-Fc and weighing commenced on day 14. Treatment consisted of daily injections of 10 mg kg 1 ENPP1-Fc formulated in PBSplus and weekly injections of Gk 1.5.
Untreated enpp1asj/asj and enpp1wt received daily injections of PBSplus and weekly injections of GK 1.5. Deaths in the untreated enpp1asj/asj cohort are
denoted by red arrows on the day of death. No deaths were noted in the
enpp1wt or the treated enpp1asj/asj cohort. (b) Survival curves, proof of
concept study. The survival of treated and untreated enpp1asj/asj animals are
plotted as a solid red line and dashed black line, respectively. Analysis by Log-rank (MantleCox) test yields a w2 of 13.18 and P value of 0.003.
(c) Phenotypic comparison, treated and untreated enpp1asj/asj mice. There is
a dramatic size difference in the treated and untreated animals, and a marked difference in the mobility and health of the animals, best seen in Supplementary Movie 1. (d) Left ventricle histology, untreated asj/asj mouse Hematoxylin and Eosin (H&E),displaying large focus of calcications and micro-infarctions in the free wall. Scale bar, 20 mm. (e) Left ventricle histology, treated enpp1asj/asj mice (H&E). None of the treated enpp1asj/asj
mice displayed abnormal L. ventricular histology. Scale bar, 20 mm.
99mTc PPi (99mPYP) uptake (Fig. 7af and Table 2). The biochemical and physiologic response was complete as measured by all of these parameters. None of the WT or treated enpp1asj/asj animals were noted to possess any vascular calcications via micro-CT, in contrast to the dramatic calcications noted in the aortas, coronary arteries, and hearts of the untreated enpp1asj/asj cohort (Fig. 7a). In addition, serum PPi concentrations of treated enpp1asj/asj animals (E5.2 mM) were elevated to WT levels
(4.4 mM) and signicantly above untreated enpp1asj/asj levels (o0.5 mM) (Fig. 7b).
99mPYP, an imaging agent typically employed in cardiac imaging and bone remodelling, was used as a marker for treatment response. It is sensitive to areas of unusually high-bone rebuilding activity since it localizes to the surface of hydro-xyapatite and then may be taken up by osteoclasts. One would expect increased 99mPYP uptake in animals lacking functional ENPP1 since they have reduced plasma [PPi] and correspondingly elevated rates of mineralization. To test this hypothesis, we performed weekly in vivo 99mPYP imaging in enpp1wt and undosed enpp1asj/asj animals to detect differences in 99mPYP uptake between the sibling pairs (Fig. 7c,d). We limited analysis of 99mPYP uptake to the head, which is comprised of both enchondral bone (skull) and soft tissue (vibrissae), which are known sites of ectopic calcication in enpp1asj/asj mice. This also simplied data analysis as the head does not overlap with internal organs showing transient 99mPYP uptake (such as the bladder, heart and diaphragm) during the 180 camera rotation that occurs during data collection.
Weekly serial imaging of enpp1wt and untreated enpp1asj/asj animals demonstrated that, as expected, uptake of 99mPYP in the heads as per cent injected dose was greater in enpp1asj/asj animals than in enpp1wt animals, and that changes in 99mPYP uptake within experimental groups did not vary signicantly over the course of the study (Fig. 7c,d). We therefore chose to measure
99mPYP uptake in treated and untreated enpp1asj/asj animals at two time pointsdays 3035 and at the completion of the study (days 5065). Comparison of these experimental groups demonstrates that ENPP1-Fc treatment returned 99mPYP uptake in GACI mice to WT levels (Fig. 7e,f), suggesting that ENPP1-Fc treatment is able to abrogate unregulated tissue, vibrissae and skull mineralization in enpp1asj/asj mice by raising the extracellular PPi concentrations.
Reappearance of calcications following cessation of dosing. To address questions regarding the reappearance of calcications following the cessation of dosing, we enrolled two animals in a limited dosing trial in which enpp1asj/asj animals were dosed with hENPP1-Fc starting on day 14 and ending on day 27. On day 28 and thereafter the animals were dosed with PBSplus and the
appearance of vascular calcications were followed with weekly in vivo CT scans (Fig. 8a). Both animals were free of calcications until day 64, when one animal developed calcications in the heart, which were noted on day 79 to progress to the aorta, spleen, kidney, and liver. The second animal developed renal calcications on day 79. Surprisingly both animals remained alive past day 84. The limited dosing study demonstrates that calcications reappear following the cessation of dosing and that treatment in the 1427 day window signicantly extends survival in enpp1asj/asj animals maintained on the acceleration diet in utero.
DiscussionDiseases of ectopic tissue calcication range from ultra-rare diseases such as GACI to nearly ubiquitous maladies in the aging population such as hardening of the arteries. The genetic aetiology of human GACI suggests that the lethal arterial calcications result from impairment of extracellular purinergic metabolism, either through loss of function mutations in ENPP1 or upstream reductions in nucleotide triphosphates metabolized by ENPP1 into extracellular PPi. Here we demonstrate that daily subcutaneous injections of ENPP1-Fc fusion protein eliminate the mortality, cardiac and arterial calcications, and other sequela of disease in rodent models of GACI. ENPP1 is the enzyme
6 NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006 ARTICLE
Table 2 | Cardiovascular pathology, proof of concept study.
WT Vehicle asj/asj vehicle asj/asj mENPP1-Fc
Calcications heart (CT/Histology) 0/0 55%/100% 0/0 Calcications aorta (CT/Histology) 0/0 66%/100% 0/0 Calcications in coronary arteries (CT/histology) 0/0 43%/100% 0/0 Myocardial infarction (Histology) 0/0 77% 0 Renal calcications 60% (light) 100% (heavy) 100% (light)
CT, computed tomography; WT, wild-type.
responsible for the generation of extracellular PPi, and we demonstrate that ENPP1-Fc raises plasma pyrophosphate levels from the nearly undetectable levels present in enpp1asj/asj mice to concentrations comparable to those seen in WT sibling pairs. Our ndings suggest that ENPP1-Fc may be effective in other diseases of ectopic calcication in which plasma PPi concentrations are decreased.
Extracellular PPi is a potent inhibitor of mineralization. ENPP1 is the major producer, and tissue-nonspecic alkaline phosphatase (TNAP) is the primary degrader, of extracellular PPi (Fig. 2a). In blood, [PPi] is B320 times lower than [Pi], therefore small changes in [PPi] have large effects in the [Pi]/[PPi] ratio, whereas similar magnitude changes in [Pi] would not. If the [Pi]/[PPi] ratio is the mechanism governing hydroxyapatite mineralization, as has been proposed, attention should be given to plasma [PPi] in patients suffering from disorders in ectopic mineralization.
The contribution of AMP generation and purinergic signalling to the suppression of vascular calcication is incompletely understood but has been addressed previous studies of enpp1asj/asj mice using aortic allografts of WT into enpp1asj/asj mice and vice versa31. These studies demonstrated that normal levels of extracellular pyrophosphate were sufcient to prevent vascular calcication over the entire surface of a transplanted enpp1asj/asj aortic allograft. The suppression of calcication in aortic allografts was throughout the entire length of the allograft with no gradient of suppression of calcication observed. Because purinergic signalling is believed to be autocrine or paracrine in nature (and not systemic), the authors concluded that purinergic signalling was either not important for vascular calcication suppression or was not substantially altered by ENPP1 deciency. Our observations that enpp1asj/asj mice dosed with ENPP1-Fc are free of vascular calcications and have normal plasma PPi concentrations supports the notion that plasma PPi levels may be primarily responsible for suppression of vascular calcication in ENPP1 deciency.
Reduced plasma PPi levels are also present in vascular calcication associated with end stage renal disease (ESRD)31,32. One in ten adults suffer from chronic kidney disease (CKD), and it is estimated that 80% of patients with CKD on dialysis and 4783% of patients with CKD not on dialysis possess vascular calcications compromising their quality of life and endangering their health. Vascular calcications associated with ESRD contributes to poor outcomes by increasing pulse pressure, causing or exacerbating hypertension, and inducing or intensifying myocardial infarctions and strokes. Most patients with ESRD do not die of renal failure, but from the cardiovascular complications of ESRD, and it is important to note that many very young patients with ESRD on dialysis possess coronary artery calcications. The histologic subtype of vascular calcication associated with CKD is known as Mnckeburgs sclerosis, which is a form of vessel hardening in which calcium deposits are found in the muscular layers of the medial vascular wall. This form of calcication is histologically distinct from
intimal or neo-intimal vascular wall calcication commonly observed in atherosclerosis but identical to the vascular calcications observed in human GACI patients, and in the rodent models of the disease described herein. The similar pathophysiology of vascular calcication present in GACI and CKD suggests that these disorders may be treated by a common therapeutic.
To this point, the current medical strategies treating ectopic vascular calcication of CKD are typically ineffective and are directed at reducing hyperphosphatemia and minimizing serum calcium concentrations. Continuous intra-peritoneal infusions of exogenous PPi substantially inhibits vascular calcications in rodent models of ESRD, but the rapid hydrolysis of PPi in vivo prevents translation of this therapy to the clinic33. Attempts treating vascular calcications in rodent models of ESRD with non-hydrolyzable PPi analogues (bisphosphonates) required doses above those known to inhibit bone formation, and the anticipated bone toxicity coupled with a renal mechanism of clearance discourages the use of bisphosphonates as therapeutic agents in CKD34. It appears that we are able to circumvent both the poor pharmacokinetics of PPi and the bone toxicity of bisphosphonates by fusing the PPi generating enzyme ENPP1 to the Fc domain of IgG1, and demonstrated that the biologic is effective in GACI. We suspect that it may also be effective in vascular calcication associated with CKD.
Methods
ENPP1-asj GACI mouse model. Animal care and maintenance were provided through Yale University Animal Resource Center at Yale University (New Haven). All procedures were approved by the Animal Care and Use Committee of Yale University and complied with the US National Institutes of Health guide forthe care and use of laboratory animals. Heterozygous enpp1asj/ (genotype
C57BL/6J-Enpp1asj/GrsrJ, Jackson Laboratory stock number 012810) breeding pairs were maintained on the acceleration diet (TD.00442, Harlan Laboratories, Madison WI) throughout the entire experiment and food and water were delivered ad libitum. The animal colony was housed in pathogen free conditions. All experimental animals were maintained on the acceleration diet in utero through completion of the study. Litters were genotyped on day 8 and weaned at day 21. Following weaning, sibling pairs were sequentially divided into cohorts as described below and enrolled in experimental trials. Animals were consecutively enrolled in experimental trials without regard to gender, and the gender of the experimental animals was not recorded. On the basis of the trial dosing we estimated that statistically signicant survival differences could be determined between dosed and undosed animals using a sample size of eight animals in each experimental group. We used identical breeding pairs throughout the study, and all experimental groups were enrolled sequentially beginning with the enpp1wt and undosed enpp1asj/asj cohorts and ending with the dosed enpp1asj/asj cohort. Enrolment of the dosed and undosed enpp1asj/asj cohorts spanned 4 months. Once the enrolment of an experimental group began both sexes of the appropriate genotype were consecutively enrolled in an experimental cohort with the exclusion of severely runted animals weighing o5.5 g at 14 days of life. Following weaning, all experimental animals were housed with littermates to allow for cooperative grooming and nesting. Experimentalists were not blinded during the study.
ENPP1-Fc design. Human and mouse NPP1 (Human: NCBI accession NP_006199; Mouse: NCBI accession NP_03839) modied to express soluble, recombinant protein as described previously26 were fused to IgG1 by subcloning into pFUSE-h IgG1-Fc1 or pFUSE-m IgG1-Fc1 plasmids (InvivoGen, San Diego, CA), respectively. The nal human protein sequence is listed in Supplementary Fig. 1.
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 7
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006
a b
a b
WT
Plasma PPi (uM)
15
10
5
0
* *
*
uM PP i
20 m
20 m
c d
Untreated asj/asj Treated asj/asj
WT
asj/asj treated
asj/asj untreated
Infarcted free wall
R. Ventricular cavity
Healthy myocardium
Calcifications
Infarcted free wall
c d
25
25
20
asj/asj
20
15
%ID/g
%ID/g
WT
15
10
10
R. Ventricle
Viable myocardium
5
n=36 n=21
20 m 10 m
20 m 10 m
10 m 10 m
5
0
WT
e Infarcted free wall
Healthy myocardium
f
0 0 20 40 60 80
asj/asj
Age (days)
e
Day 3035 Day 5065
n=7 n=7 n=3
f
25
* *
25
*
*
20
20
15
15
%ID/g
10
%ID/g
10
5
5
n=6 n=11
n=6
Treated asj/asj
Intracardiac calcations
0 WT
Scar tissue
Calcified coronary arteries
0 WT
Untreated asj/asj
g
Scar
h
Untreated asj/asj
Treated asj/asj
Healthy myocardium
Scar
Figure 7 | Biomarkers of disease response. (a) Postmortem high-resolution micro-CT scans revealed extensive calcications in untreated enpp1asj/asj mice in the hearts, coronary arteries, and ascending and
descending aortas, but absolutely no calcications in these organs in the treated enpp1asj/asj cohort or in enpp1wt mice. For further imaging,
Supplementary Movie 2. (b) Plasma [PPi] in enpp1wt and treated and untreated enpp1asj/asj animals revealed that treatment with mENPP1-Fc
increased [PPi] in enpp1asj/asj mice to WT levels, and well above the nearly undetectable levels present in untreated enpp1asj/asj mice. *Po0.0015,
Students two-tailed t-test. (c,d) Per cent uptake of injected 99mPYP in
heads of WT and asj/asj animals. The per cent uptake of 99mPYP in heads
of animals in the natural history study were recorded weekly in the WT and asj/asj animals on the acceleration diet, demonstrating that 99mPYP uptake
remains nearly constant over an 80 day period following birth, but differs markedly between the two experimental groups. (d) In the natural history study, the average 99mPYP uptake in heads of enpp1wt animals was around 15% of injected dose over the 80 day period, while the PYP uptake in enpp1asj/asj animals was around 20% *Po0.001, Students two-tailed t-test.
(e,f) 99mPYP uptake. The per cent 99mPYP uptake in the heads of all experimental groups was recorded in the middle of the study (days 3035, in e) and at the end of the study (days 5065, in f). enpp1wt and treated enpp1asj/asj animals had per cent uptake in the skulls around 15%, while the
untreated enpp1asj/asj cohort was at or above 20%. *Po0.001, Students two-tailed t-test.
Figure 6 | Representative histology, proof of concept study. (a,b) Aorta (Alizarin red). Untreated enpp1asj/asj mice displayed nearly circumferential
aortic calcications (a), while treated enpp1asj/asj mice did not (b). Scale bar,
20 mm. (c) Untreated enpp1asj/asj mice, right (R) ventricle Hematoxylin and Eosin (H&E). Two untreated enpp1asj/asj mice had large, conuent,
myocardial infarctions in the free wall of the R. ventricle. Scale bar, 20 mm. All treated enpp1asj/asj mice displayed normal R. ventricle myocardium
(not shown). (d) Untreated enpp1asj/asj mice, R. ventricle (H&E) higher
power view of the boxed area in panel c. Scale bar, 10 mm. (e) Untreated enpp1asj/asj mice, R. ventricle (Trichrome) Trichrome stains of the R.
ventricle in the same animal demonstrates the infarcted free wall. Scale bar, 20 mm. (f) Untreated enpp1asj/asj mice, myocardial septum (H&E). Nearly all animals (77%) displayed intracardiac calcications surrounded by scar tissue, as demonstrated in this animal in the myocardial septum. Scale bar, 10 mm. (g) Untreated enpp1asj/asj mice, coronary arteries (H&E). All untreated enpp1asj/asj mice had coronary calcications, with most displaying
circumferential calcications in coronary arteries surrounded by scar tissue,
diagnostic of ischaemia and myocardial infarction. Scale bar, 10 mm.
(h) Untreated enpp1asj/asj mice, coronary arteries (Trichrome). Trichrome
stains of coronary artery regions of untreated enpp1asj/asj mice
demonstrates increased brosis associated with vascular wall calcications (blue colour and labelled scar), demonstrating the myocardial injury in the animals. Scale bar, 10 mm.
Protein production with shaking asks. Stable transfections of the ENPP1-Fc were established in HEK293 cells under zeocin selection. Briey, adherent HEK293 cells were adapted for suspension growth by progressive dilution of fetal bovine serum from 10 to 1% over 56 passages35. Zeocin selection (450 mg ml 1)
was maintained throughout the adaptation process, and cultures were supplemented with 1% insulin-transferrin-selenium (Mediatech 25800-CR), and cells were passaged when they reached conuence in 175 cm2 tissue culture asks. Adapted cells were used to seed liquid culture growths in FreeStyle
medium (Gibco, Waltham MA) in shaker asks at 37F and 5% CO2, agitated at 120 r.p.m. with high humidity. Cell density in liquid culure was maintained between 0.5 and 2.5 106 cells per ml and zeocin selection was maintained in
liquid culture at 450 mg ml 1 until the cells reached a volume of 4 l in liquid culture, and then reduced to 225 mg ml 1 thereafter. The culture was gradually expanded to 8 l and then maintained for another 12 days to accumulate extracellular protein. During the maintenance phase, cultures were supplemented with CD EfcientFeed C AGT (Gibco #A13275-05) to enhance protein production.
Protein production with bioreactor. Cells were propagated in a 10 l bioreactor equipped with dissolved oxygen and pH control. Dissolved oxygen was kept at
8 NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006 ARTICLE
a
eNPP1-WT (10)
eNPP1-asj/asj (2)
Age (days)
Mass (g)
End dosing hNPP1-Fc
Start dosing hNPP1-Fc(500 au/Kg + GK 1.5)
Animal 1: calcifications
appear in heart Animal 1: calcifications in heart Aorta, kidney, spleen, liver Animal 2: calcifications appear in kidney.
12
11
10
9
8
7
6
5 10 20 30 40 50 60 Days
b
NPP1 asj/asj +inactive hENPP1+GK1.5 (n=3)
NPP1 asj/asj+ vehicle (n=9)
Mass (g)
(2)
Figure 8 | Limited dosing study. (a) To determine if calcications reappear following cessation of dosing, two enpp1asj/asj animals
were dosed daily with hENPP1-Fc between days 1427 (green arrows), followed by daily dosing with PBSplus. Rejection of human version of ENPP1-Fc (hENPP1-Fc) was suppressed during the limited dosing period with weekly doses of GK 1.5. The average daily weights are of enpp1wt (blue triangles, n 10)
and enpp1asj/asj (red squares, n 2) mice are plotted with the error bars denoting s.d.s. The dosed enpp1asj/asj animals were followed for reappearance of
vascular and organ calcications by weekly in vivo CT scans. Calcications in both animals were eventually observed (cyan arrows). One animal developed calcications in the heart on day 64, which progressed to the aorta, liver, kidney and spleen by day 79. The second animal developed renal calcications on day 79. (b) Negative control experiment. To demonstrate that ENPP1-Fc enzyme activity is essential for therapeutic effect, and that weekly GK 1.5 administration does not alter the natural history disease, enpp1asj/asj mice were dosed daily with 10 mg kg 1 inactive hENPP1-Fc, and weekly with
GK 1.5 (blue diamonds, n 3). The average daily weights are plotted compared with enpp1asj/asj mice dosed with vehicledaily PBSplus and weekly GK
1.5(green triangles, n 9) and the error bars denote s.d.s. All three enpp1asj/asj mice dosed with inactive ENPP1-Fc experienced a drop in weight and
mortality (deaths denoted by blue arrows) similar to enpp1asj/asj mice dosed with vehicle (deaths denoted by red arrows), demonstrating that neither
inactive ENPP1-Fc nor GK 1.5 extends survival.
40% air saturation by supplying the culture with mixture of air and oxygen not exceeding 3 l minute at an agitation rate of 80 RPM. pH was controlled at 7.4 by sparging CO2 when the pH was higher than 7.4. Culture growth was followed by measuring cell number, cell viability, glucose and lactate concentrations. Final yields for both methods of production were B5 mg of puried ENPP1-Fc per liter of culture.
Protein purication. The liquid cultures were centrifuged at 4,300g for 15 min and the supernatants were ltered through a 0.2 mm membrane and concentratedvia tangential low using a Pellicon3 0.11 m2 Ultracell 30 kD cassette (Millipore, Billerica MA). The concentrated supernatant was loaded onto a protein-A column, then washed using 50 mM Tris pH8, 150 mM NaCl2, 1 mM ZnCl2 and 1 mM
CaCl2. The recombinant protein was eluted using Elution Buffer (150 mM Sodium Citrate; pH 4.3, 150 mM NaCl2, 1 mM ZnCl2 and 1 mM CaCl2), then
immediately neutralized using a 1/10th volume of 1 M Tris pH 9.2, 150 mM NaCl2,
3 mM ZnCl2 and 3 mM CaCl2. Fractions containing enzymatic activity were pooled and dialysed against PBSplus buffer (1 PBS buffer pH 7.4, 11 uM ZnCl2, 20 uM
CaCl2), then concentrated to B6 mg ml 1, distributed into small aliquots and stored at 80 C. Resulting protein samples were then tested with Pierce LAL
Chromogenic Endotoxin Quantitation Kit (cat. 88282) to verify that all were free of endotoxin.
Enzymology. The steady state hydrolysis of ATP by human NPP1 was determined by high performance liquid chromatography (HPLC; refs 26,36). Briey, enzyme reactions were started by addition of 10 or 50 nM NPP1 to varying concentrations of ATP in the reaction buffer containing 20 mM Tris, pH 7.4, 150 mM NaCl, 4.5 KCl, 14 uM ZnCl2, 1 mM MgCl2 and 1 mM CaCl2. At various time points, 50 ml reaction solution was removed and quenched with an equal volume of 3 M formic acid. The quenched reaction solution was loaded on a C-18 (5 mm 250 4.6 mm)
column (Higgins Analytical) equilibrated in 15 mM ammonium acetate (pH 6.0) solution and eluted with a 020% methanol gradient. Substrate and products were monitored by ultraviolet absorbance at 259 nm and quantied according to the integration of their correspondent peaks and standard curves.
Size exclusion chromatography coupled with static laser light scattering. The light scattering data were collected using a Superose 6, 10/30, HR Size Exclusion Chromatography (SEC) column (GE Healthcare, Piscataway, NJ), connected to high performance liquid chromatography system (HPLC), Agilent 1200, (Agilent Technologies, Wilmington, DE) equipped with an autosampler. The elution from SEC was monitored by a photodiode array (PDA) UV/VIS detector, UV, (Agilent Technologies, Wilmington, DE), differential refractometer, RI, (OPTI-Lab rEx Wyatt Technology, Santa Barbara, CA), static and dynamic, multiangle laser light scattering (LS) detector (HELEOS II with QELS capability, Wyatt Technology, Santa Barbara, CA). The SEC-MALLS/UV/RI system was equilibrated in PBSplus buffer at the ow rate of 0.3 ml/min. Two software packages were used fordata collection and analysis: the Chemstation software (Agilent Technologies, Wilmington, DE) controlled the HPLC operation and data collection from the multi-wavelength ultraviolet /VIS detector, while the ASTRA software (Wyatt Corp., Santa Barbara, CA) collected data from the refractive index detector, the light scattering detectors, and recorded the ultraviolet trace from the PDA detector. The weight average molecular weights, Mw, were determined across the entire elution prole in the intervals of 1 s from static light scattering (LS) measurement using three detector approach2729.
Vehicle. Mouse ENPP1-Fc (mENPP1-Fc) was formulated in vehicle such that the volume of vehicle delivered was 16 ml vehicle per gram of body weight. Vehicle consisted of americanBio 10X PBS (Stock# AB11072) diluted to 1 with
endotoxin free water and supplemented with 14 mM CaCl2 and 14 mM ZnCl2.
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 9
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006
Dosing. Animals in the proof of concept study (Figs 57) were dosed either with vehicle or with mouse ENPP1-Fc (mENPP1-Fc) formulated in vehicle. Mice were dosed with daily subcutaneous injections starting on day 14 at dose levels of500 a.u. kg 1 mENPP1-Fc. Animals in the limited dosing study (Fig. 8a) were dosed with human ENPP1-Fc (hENPP1-Fc) formulated in vehicle beginning on the fourteenth day of life and ending on the twenty-seventh day of life. To tolerize the mice to human protein, the mice were dosed intraperitoneally with 75 mg of GK 1.5 (eBioscience, San Diego, CA) on the thirteenth day of life, and 50 mg of GK 1.5 on the twentieth day of life. Between days 28 and 83 the mice received vehicle. Inactive hENPP1-Fc protein (Fig. 8b) was generated by exposure to EDTA. The specic activity hENPP1-Fc following EDTA exposure was o 20 a.u. mg 1, or o50% of the activity of native protein.
Clearance rate of ENPP1-Fc. Briey, 15 male C57BI/6J mice aged 8 weeks or older were administered a single subcutaneous dose of 10 mg kg 1 of mENPP1-Fc.
Blood was collected from tail veins into heparin-containing tubes before the injection and at different time points post injection. Mice were divided into three groups for blood collection: (1) 1 h, 8 h, 48 h, 96 h. (2). 3 h, 24 h, 45 h and 72 h. (3). 6 h and 32 h. Enpp1 activity in plasma were measured with pNP-TMP assay in the following buffer: 100 mM Tris HCl (pH 9.0), 500 mM NaCl, 5 mM MgCl2, 0.05%
(vol/vol) Triton X-100 and 4 mM pNp-TMP. Plasma activity units were converted into protein mass using an empirically determined conversion factor of 0.332 units per mg protein. To derive plasma concentrations and fraction of subcutaneous drug absorbed into plasma, the total blood volume of each mouse was estimated at1.5 ml. Clearance rate constants were derived as detailed in the text.
Denition of activity unit. Animals were dosed according to total activity of enzyme delivered rather than concentration of enzyme to account for varying activity among batches of enzyme used. One activity unit (1 a.u.) is dened as pM of pNP-TMP substrate hydrolyzed min 1 mg 1 enzyme. The activity assay was performed in a buffer consisting of 50 mM Tris pH9, 150 mM NaCl, 0.1 mM ZnCl2,
0.1 mM CaCl2, 0.1 mM MgCl2. The activity of acceptable protein preparations varied between 40 and 43 a.u. mg 1, and preparations with o40 a.u. mg 1 were discarded. A dose of 500 a.u. kg 1 corresponds to between 6 and 10 mg kg 1, depending on the specic activity of the protein preparation.
Quantication of plasma PPi. Animals were terminally bled retro-orbitally using heparinized, micropipets, and the blood was immediately dispensed into heparin-treated eppendorf tubes and placed on wet ice. The samples were spun in a 4 C pre-cooled microcentrifuge at 4,000 r.p.m. for 5 min, and plasma was collected and diluted in one volume of 50 mM Tris-Acetate pH 8.0. Plasma was then ltered
through a 300 KDa membrane via ultracentrifugation (NanoSep 300 K, Pall Corp., Ann Arbour, MI) and frozen at 80 C. Pyrophosphate was quantitated using
standard three-step enzymatic assays using uridine 5 diphospho[14C]glucose to record the reaction product, uridine 5 diphospho[14C]gluconic acid37. Briey, a reaction mixture (100 ml) containing 5 mM MgCl2, 90 mM KCL, 63 mM Tris-HCL (pH 7.6), 1 nmol NADP , 2 nmol glucose 1,6-diphosphate, 400 pmol uridine
5-diphosphoglucose, 0.02 mCi uridine 5 diphospho[14C]glucose, 0.25 units of uridine 5-diphosphoglucose pyrophosphorylase, 0.25 units of phosphoglucose mutase, 0.5 units of glucose 6-phosphate dehydrogenase, and inorganic pyrophosphate (50200 pmol) is incubated for 30 min at 37 C. The reaction is terminated by the addition of 200 ml of 2% charcoal well suspended in water.
The mixture is kept on ice, vortexed three times over 15 min and claried by centrifugation for 5 min. An aliquote of 200 ml of supernatant is then counted in scintillation solution.
In vivo 99mPYP imaging. The bone imaging agent 99mTc-pyrophosphate (Pharmalucence, Inc) was evaluated in cohorts of animals using a preclinical microSPECT/CT hybrid imaging system with dual 1 mm pinhole collimators (X-SPECT, Gamma Medica-Ideas)38. Each animal was injected intraperitoneally with 25 mCi of the radiolabelled tracer and imaged 11.5 h after injection. A CT scan (512 projections at 50 kVp, 800 uA and a magnication factor of 1.25) was acquired for anatomical co-localization with the SPECT image. The SPECT imaging was acquired with 180 per collimator head in a counter-clockwise rotation, 32 projections, 60 s per projection with an ROR of 7.0 cm, FOV of 8.95 cm and an energy window of 140 keV20. CT images were reconstructed with the FLEX X-O CT software (Gamma Medica-Ideas) using a ltered back-projection algorithm. SPECT images were reconstructed using the FLEX SPECT software(5 iterations, 4 subsets) and subsequently fused with the CT images and analysed using the AMIRA software and ofine in-house script. Data were corrected for decay and injected dose to achieve % injected dose.
Quantication of 99mPYP uptake. For the 99mPYP murine scans, the animals were imaged two hours postinjection. The resulting SPECT scans were imported into NIHs ImageJ image processing software and regions of interest were drawn around each animals head (target organ) and whole body. Per cent injected activity (PIA), often referred to as per cent injected dose was calculated by comparing the ratio of counts in the head to the counts in the whole body, and expressed as per
cent injected dose to give a measure as of the afnity with which the radiotracer is taken up by the region of interest (head). The total counts in each scan were taken as the whole body measure of injected dose.
References
1. Moran, J. J. Idiopathic arterial calcication of infancy: a clinicopathologic study. Pathol. Annu. 10, 393417 (1975).
2. Rutsch, F. et al. Hypophosphatemia, hyperphosphaturia, and bisphosphonate treatment are associated with survival beyond infancy in generalized arterial calcication of infancy. Circ. Cardiovasc. Genet. 1, 133140 (2008).
3. Otero, J. E. et al. Severe skeletal toxicity from protracted etidronate therapy for generalized arterial calcication of infancy. J. Bone Miner. Res. 28, 419430 (2013).
4. Stuart, G., Wren, C. & Bain, H. Idiopathic infantile arterial calcication in two siblings: failure of treatment with diphosphonate. Br. Heart J. 64, 156159 (1990).
5. Stuart, A. G. Idiopathic arterial calcication of infancy and pyrophosphate deciency. J. Pediatr. 123, 170171 (1993).
6. Galletti, S. et al. Generalized Arterial Calcication of Infancy: Fatal Clinical Course Associated with a Novel Mutation in ENPP1. JIMD Rep. 1, 2327 (2011).
7. Kalal, I. G., Seetha, D., Panda, A., Nitschke, Y. & Rutsch, F. Molecular diagnosis of generalized arterial calcication of infancy (GACI). J. Cardiovasc. Dis. Res. 3, 150154 (2012).
8. Chong, C. R. & Hutchins, G. M. Idiopathic infantile arterial calcication: the spectrum of clinical presentations. Pediatr. Dev. Pathol. 11, 405415 (2008).
9. Ferreira, C., Ziegler, S. & Gahl, W. in GeneReviewss (eds Paragon, R. A. et al.) 34 (University of Washington, 2014).
10. Bryant, J. & White, W. A case of calcication of the arteries and obliterative endarteritis, associated with hydronephrosis, in a child aged six months. Guys Hosp Rep. 55, 1728 (1901).
11. Rutsch, F. et al. PC-1 nucleoside triphosphate pyrophosphohydrolase deciency in idiopathic infantile arterial calcication. Am. J. Pathol. 158, 543554 (2001).
12. Bollen, M., Gijsbers, R., Ceulemans, H., Stalmans, W. & Stefan, C. Nucleotide pyrophosphatases/phosphodiesterases on the move. Crit. Rev. Biochem. Mol. Biol. 35, 393432 (2000).
13. Harahap, A. R. & Goding, J. W. Distribution of the murine plasma cell antigen PC-1 in non-lymphoid tissues. J. Immunol. 141, 23172320 (1988).
14. Terkeltaub, R. Physiologic and pathologic functions of the NPP nucleotide pyrophosphatase/phosphodiesterase family focusing on NPP1 in calcication. Purinergic Signal. 2, 371377 (2006).
15. Addison, W. N., Azari, F., Sorensen, E. S., Kaartinen, M. T. & McKee, M. D. Pyrophosphate inhibits mineralization of osteoblast cultures by binding to mineral, up-regulating osteopontin, and inhibiting alkaline phosphatase activity. J. Biol. Chem. 282, 1587215883 (2007).
16. Rutsch, F. et al. Mutations in ENPP1 are associated with idiopathic infantile arterial calcication. Nat. Genet. 34, 379381 (2003).
17. Sakata, S., Su, J. C., Robertson, S., Yin, M. & Chow, C. W. Varied presentations of pseudoxanthoma elasticum in a family. J. Paediatr.Child H. 42, 817820 (2006).
18. Le Boulanger, G. et al. An unusual severe vascular case of pseudoxanthoma elasticum presenting as generalized arterial calcication of infancy. Am. J. Med. Genet. A 152A, 118123 (2010).
19. Bertrand, J. et al. Decreased levels of nucleotide pyrophosphatase phosphodiesterase 1 are associated with cartilage calcication in osteoarthritis and trigger osteoarthritic changes in mice. Ann. Rheum. Dis. 71, 12491253 (2012).
20. Li, Q., Schumacher, W., Jablonski, D., Siegel, D. & Uitto, J. Cutaneous features of pseudoxanthoma elasticum in a patient with generalized arterial calcication of infancy due to a homozygous missense mutation in the ENPP1 gene. Br. J. Dermatol. 166, 11071111 (2012).
21. Li, Q. et al. Mutations in the ABCC6 gene as a cause of generalized arterial calcication of infancy: genotypic overlap with pseudoxanthoma elasticum.J. Invest. Dermatol. 134, 658665 (2014).22. Nitschke, Y. et al. Generalized arterial calcication of infancy and pseudoxanthoma elasticum can be caused by mutations in either ENPP1 or ABCC6. Am. J. Hum. Genet. 90, 2539 (2012).
23. Jansen, R. S. et al. ABCC6 prevents ectopic mineralization seen in pseudoxanthoma elasticum by inducing cellular nucleotide release. Proc. Natl Acad. Sci. USA 110, 2020620211 (2013).
24. Li, Q. et al. Mutant Enpp1asj mice as a model for generalized arterial calcication of infancy. Dis. Model Mech. 6, 12271235 (2013).
25. Li, Q., Chou, D. W., Price, T. P., Sundberg, J. P. & Uitto, J. Genetic modulation of nephrocalcinosis in mouse models of ectopic mineralization: the Abcc6(tm1Jfk) and Enpp1(asj) mutant mice. Lab. Invest. 94, 623632 (2014).
10 NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10006 ARTICLE
26. Albright, R. A. et al. The molecular basis of purinergic signal metabolism by ecto-nucleotide pyrophosphatase/phosphodiesterases 4 and 1 and implications in stroke. J. Biol. Chem. 289, 32943306 (2013).
27. Folta-Stogniew, E. Oligomeric states of proteins determined by size-exclusion chromatography coupled with light scattering, absorbance, and refractive index detectors. Methods Mol. Biol. 328, 97112 (2006).
28. Arakawa, T., Langley, K. E., Kameyama, K. & Takagi, T. Molecular weights of glycosylated and nonglycosylated forms of recombinant human stem cell factor determined by low-angle laser light scattering. Anal. Biochem. 203, 5357 (1992).
29. Hayashi, Y., Matsui, H. & Takagi, T. Membrane protein molecular weight determined by low-angle laser light-scattering photometry coupled with high-performance gel chromatography. Methods Enzymol. 172, 514528 (1989).
30. Merino, J. et al. Autoimmune syndrome after induction of neonatal tolerance to alloantigens: effects of in vivo treatment with anti-T cell subset monoclonal antibodies. J. Immunol. 139, 14261431 (1987).
31. Lomashvili, K. A., Narisawa, S., Millan, J. L. & ONeill, W. C. Vascular calcication is dependent on plasma levels of pyrophosphate. Kidney Int. 85, 13511356 (2014).
32. ONeill, W. C., Sigrist, M. K. & McIntyre, C. W. Plasma pyrophosphate and vascular calcication in chronic kidney disease. Nephrol. Dial. Transplant 25, 187191 (2010).
33. ONeill, W. C., Lomashvili, K. A., Malluche, H. H., Faugere, M. C. & Riser, B. L. Treatment with pyrophosphate inhibits uremic vascular calcication. Kidney Int. 79, 512517 (2011).
34. Lomashvili, K. A., Monier-Faugere, M. C., Wang, X., Malluche, H. H. & ONeill, W. C. Effect of bisphosphonates on vascular calcication and bone metabolism in experimental renal failure. Kidney Int. 75, 617625 (2009).
35. Xiao, S., White, J. F., Betenbaugh, M. J., Grisshammer, R. & Shiloach, J. Transient and stable expression of the neurotensin receptor NTS1:a comparison of the baculovirus-insect cell and the T-REx-293 expression systems. PloS One 8, e63679 (2013).
36. Albright, R. A. et al. NPP4 is a procoagulant enzyme on the surface of vascular endothelium. Blood 120, 44324440 (2012).
37. Cheung, C. P. & Suhadolnik, R. J. Analysis of inorganic pyrophosphate at the picomole level. Anal. Biochem. 83, 6163 (1977).
38. Dobrucki, L. W. et al. Serial noninvasive targeted imaging of peripheral angiogenesis: validation and application of a semiautomated quantitative approach. J. Nucl. Med.e 50, 13561363 (2009).
39. Cao, W. & De La Cruz, E. M. Quantitative full time course analysis of nonlinear enzyme cycling kinetics. Sci. Rep. 3, 2658 (2013).
Acknowledgements
These studies were nancially supported by research grants from Alexion Pharmaceuticals Inc., The Connecticut Bioscience Innovation Fund, and Yale University Department of Pathology. DTB acknowledges helpful conversations with Michael Hodsdon (Eli Lily), Elias Lolis (Yale), and John Louis (NIH/NIDDK/Laboratory of Chemical Physics), and Dr Jon Morrow (Yale) for his enthusiastic support and encouragement throughout the course of this study. This material is based onwork supported in part by the State of Connecticut under the Connecticut Bioscience Innovations Fund. Its contents are solely the responsibility of the authors and do not necessarily represent the ofcial views of the State of Connecticut or Connecticut Innovations, Incorporated.
Author contributions
R.A.A., P.S., W.C., G.Y., Z.C., K.B., Z-X.Y., S.T., E.J.F.-S., A.J.S., J.S., G.Z., J.A.M., E.M.D.L.C. and D.T.B. designed the research. R.A.A, P.S., W.C., D.K., I.M., A.A.B., M.S.C., M.T., G.Y., S.T., X.W., E.J.F.-S., A.N., A.J.S., J.S., G.Z., E.M.D.L.C. and D.T.B performed the research. R.A.A., P.S., W.C., G.Y., Z.C., K.B., Z.-X.Y., S.T., X.W., E.J.F.-S., A.J.S., J.S., G.Z., J.A.M., E.M.D.L.C. and D.T.B. Analysed the data. R.A.A., P.S., W.C., Z.C., S.T., E.J.F.-S., A.J.S, J.S., G.Z., E.M.D.L.C. and D.T.B. wrote the papers.
Additional information
Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications
Web End =http://www.nature.com/ http://www.nature.com/naturecommunications
Web End =naturecommunications
Competing nancial interests: R.A.A. and D.T.B. are co-inventors of a pending patent application owned by Yale University related to these studies. Z.C., K.B. and Z.-X.Y. are employees of Alexion Pharmaceuticals, Inc. The remaining authors declare no competing nancial interest.
Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/
Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions/
Web End =reprintsandpermissions/
How to cite this article: Albright, R. A. et al. ENPP1-Fc prevents mortality and vascular calcications in rodent model of generalized arterial calcication of infancy. Nat. Commun. 6:10006 doi: 10.1038/ncomms10006 (2015).
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Web End =http://creativecommons.org/licenses/by/4.0/
NATURE COMMUNICATIONS | 6:10006 | DOI: 10.1038/ncomms10006 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 11
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright Nature Publishing Group Dec 2015
Abstract
Diseases of ectopic calcification of the vascular wall range from lethal orphan diseases such as generalized arterial calcification of infancy (GACI), to common diseases such as hardening of the arteries associated with aging and calciphylaxis of chronic kidney disease (CKD). GACI is a lethal orphan disease in which infants calcify the internal elastic lamina of their medium and large arteries and expire of cardiac failure as neonates, while calciphylaxis of CKD is a ubiquitous vascular calcification in patients with renal failure. Both disorders are characterized by vascular Mönckeburg's sclerosis accompanied by decreased concentrations of plasma inorganic pyrophosphate (PPi ). Here we demonstrate that subcutaneous administration of an ENPP1-Fc fusion protein prevents the mortality, vascular calcifications and sequela of disease in animal models of GACI, and is accompanied by a complete clinical and biomarker response. Our findings have implications for the treatment of rare and common diseases of ectopic vascular calcification.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer