Hydrolysis of diadenosine polyphosphates by nucleotide
pyrophosphatases/phosphodiesterases
Petra Vollmayer
1
, Timothy Clair
2
, James W. Goding
3
, Kimihiko Sano
4
,Jo¨ rg Servos
1
and Herbert Zimmermann
1
1
AK Neurochemie, Biozentrum der J. W. Goethe-Universitaet, Frankfurt am Main, Germany;
2
Laboratory of Pathology, NCI,
National Institutes of Health, Bethesda, Maryland, USA;
3
Department of Pathology and Immunology, Monash University,
Alfred Hospital, Prahran, Victoria, Australia;
4
Department of Pediatrics, Kobe University School of Medicine, Kobe, Japan
Diadenosine polyphosphates (Ap
n
As) act as extracellular
signaling molecules in a broad variety of tissues. They were
shown to be hydrolyzed by surface-located enzymes in an
asymmetric manner, generating AMP and Ap
n-1

1
,P
4
-tetraphosphate (Gp
4
G). Each of the three
enzymes hydrolyzed Ap
3
A, Ap
4
A, and Ap
5
A at comparable
rates. Gp
4
G was hydrolyzed by NPP1 and NPP2 at rates
similar to Ap
4
A,butonlyathalfthisratebyNPP3.
Hydrolysis was asymmetric, involving the a,b-pyrophos-
phate bond. Ap
n
A hydrolysis had a very alkaline pH opti-
mum and was inhibited by EDTA. Michaelis constant (K
m
)
values for Ap
3
Awere5.1l
M

Gs) and diguanosine poly-
phosphates (Gp
n
Gs) [10].
The diadenosine polyphosphates diadenosine
5¢,5¢¢¢-P
1
,P
3
-triphosphate (Ap
3
A), diadenosine 5¢,5¢¢¢-P
1
,P
4
-
tetraphosphate (Ap
4
A), and diadenosine 5¢,5¢¢¢-P
1
,P
5
-pen-
taphosphate (Ap
5
A) are stored in chromaffin granules at
millimolar concentrations together with noradrenaline
and other nucleotides such as ATP and ADP [11,12]. In
cholinergic synaptic vesicles, Ap
4

n
As were shown
to be metabolized and thus functionally inactivated by
enzymes associated with the cell surface or present in
body fluids [18]. Primary cleavage of Ap
n
As is asym-
metric, generating AMP and Ap
n-1
from Ap
n
A. Cellular
systems analyzed include blood [19], the adrenal medulla
Correspondence to H. Zimmermann, AK Neurochemie,
Biozentrum der J.W. Goethe-Universitaet, Marie-Curie-Str. 9,
D-60439 Frankfurt am Main, Germany.
Fax: + 49 69 79829606, Tel.: + 49 69 79829602,
E-mail: [email protected]
Abbreviations:Ap
4
, adenosine tetraphosphate; Ap
3
A, diadenosine
5¢,5¢¢¢-P
1
,P
3
-triphosphate; Ap
4
A, diadenosine 5¢,5¢¢¢-P

[20], vascular endothelial cells [21,22], synaptic membranes
[23,24], airway epithelia [25], and Xenopus oocytes [26].
At present, little is known concerning the molecular
identity of the enzymes involved. One group of enzymes
with the potential to hydrolyze Ap
n
As are NPP1, NPP2,
and NPP3, the three members of the ecto-nucleotide
pyrophosphatase/phosphodiesterase (E-NPP) family
(revised nomenclature) [27,28]. They reveal a surprisingly
broad substrate specificity and are capable of hydrolyzing
phosphodiester bonds of nucleotides and nucleic acids, and
pyrophosphate bonds of nucleotides and nucleotide sugars,
resulting in the release of nucleoside 5¢ monophosphates.
Both purine and pyrimidine nucleotides are hydrolyzed.
Substrates include ATP, NAD
+
,andp-nitrophenyl thymi-
dine 5¢ monophosphate (pNP-TMP) [29–31]. To date, three
related mammalian NPPs have been cloned and function-
ally characterized: NPP1 (plasma-cell differentiation anti-
gen-1, PC-1) [32]; NPP2 (autotaxin) [33] and its splice
variant, PD-Ia [34]; and NPP3 (gp130
RB13)6
, B10, PD-Ib)
[35–37]. NPP1 and NPP3 share 50% amino acid identity
and are more distantly related to NPP2 (39–41% identity).
They have a broad and partially overlapping tissue distri-
bution and may be co-localized within the same cell [37]. All
can exist as integral glycoproteins (of  120 kDa) of the

pain were from Calbiochem (Schwalbach, Germany), and
aprotinin was from Roche (Mannheim, Germany).
Cell culture and cell transfection
Chinese hamster ovary (CHO) cells were cultured in HAM’s
F-12 medium containing 10% fetal bovine serum,
100 UÆmL
)1
penicillin and 100 lgÆmL
)1
streptomycin. They
were transfected by electroporation [41] with plasmid DNA
containing human NPP1 (NPP1pSVT7) [42] or rat NPP3
(GenBank
TM
accession number D30649; NPP3pcDNA3).
Transfection with the empty plasmid pSVT7 or pcDNA3
served as a control. Stable transfectants for NPP3 were
selected for neomycin resistance using 800 lgÆmL
)1
of
genectin G-418 sulfate (Invitrogen) and subcloned by
limiting dilution. Stably transfected cells were cultured in
the presence of 750 lgofgenectinG-418sulfatepermLof
culture medium. Sodium butyrate (6 m
M
)wasaddedtothe
culture medium of stably transfected cells 16 h before
preparation of membrane fractions or analysis of cell
surface-located enzyme activity.
Preparation of membrane fractions

supernatant fraction was centrifuged (100 000 g for 45 min
at 4 °C) and the pellets were resuspended in protease
inhibitor-containing buffer A prior to storage at )20 °C.
Protein was determined according to the method of
Peterson [43].
Measurement of nucleotide hydrolysis
The hydrolysis of substrates Ap
3
A, Ap
4
A, Ap
5
Aand
Gp
4
G was analyzed using membrane fractions containing
heterologously expressed NPP1 and NPP3, or the soluble
form of recombinant human NPP2. NPP2 was prepared
and isolated from a vaccinia virus lysate of BS-C-1 cells,
as described previously, through the concanavalin
A–agarose step [44], and stored in a solution containing
50 m
M
Tris, pH 7.5, 100 m
M
NaCl, 10 m
M
CaCl
2
,and

reversed-phase column (Jasco, Grob-
Umstadt, Germany) and eluted at 1.0 mLÆmin
)1
with the
mobile phase consisting of 50 m
M
potassium-phosphate
buffer (pH 6.4), 5 m
M
tetrabutylammonium hydrogen
sulfate and 10–18% methanol, depending on the nucleo-
tide analyzed. The absorbance at 260 nm was continu-
ously monitored and the nucleotide concentrations were
determined from the area under the absorbance peaks.
Hydrolysis rates represent initial rates.
2972 P. Vollmayer et al. (Eur. J. Biochem. 270) Ó FEBS 2003
pH dependence, ion dependence, and inhibitors
Two different buffer systems were applied for analyzing pH
dependence of Ap
3
A hydrolysis. For analysis from pH 6.0
to 7.5, a buffer solution containing 25 m
M
Hepes, 25 m
M
glycine, 140 m
M
NaCl, and 5 m
M
KClwasused.For

,
2m
M
CaCl
2
and 2 m
M
MgCl
2
, or 500 l
M
EDTA. Because
the storage solution of NPP2 contained CaCl
2
,150l
M
EGTA was added to the solution containing 2 m
M
MgCl
2
.
The substrate concentration was 200 l
M
Ap
3
AforNPP1,
NPP2 and NPP3
1
. The inhibitory effect on Ap
4

Cells were washed twice with phosphate-free saline solution
(140 m
M
NaCl, 5 m
M
KCl, 10 m
M
glucose, 20 m
M
Hepes,
pH 7.4) and subsequently incubated for 60 min at 37 °Cin
1 mL of identical saline solution containing 200 l
M
of the
substrate nucleotide, 1 m
M
MgCl
2
and 1 m
M
CaCl
2
.The
physiological saline solution collected from the cultures was
heat inactivated (95 °C for 4 min), followed by centrifuga-
tion (14 500 g for 20 min). Aliquots of the supernatant were
subjected to nucleotide analysis as described above.
Results
General biochemical properties of Ap
n

MgCl
2
,2m
M
CaCl
2
plus 2 m
M
MgCl
2
2
,or500l
M
EDTA. As shown in
2
Fig. 1,
the catalytic rates were similar in the presence of Ca
2+
or Mg
2+
. Activity was not further stimulated by the
simultaneous addition of Ca
2+
and Mg
2+
, while EDTA
greatly reduced the catalytic activity. In the following,
catalytic activity was measured in the presence of 1 m
M
CaCl

)1
Æmg pro-
tein
)1
(n ¼ 4) for NPP3, and 8.0 ± 0.5 l
M
and
8.6 ± 0.8 nmolÆmin
)1
ÆmL
)1
(n ¼ 2) for NPP2.
Substrate specificity and product pattern
We subsequently analyzed the ability of NPP1 to NPP3
to hydrolyze various dinucleoside polyphosphates and
determined the pattern of product formation. In order to
avoid potential product inhibition, the enzyme concen-
tration was adjusted to obtain low catalysis rates. Table 1
compares the hydrolysis rates of the three enzymes for
the dinucleotides investigated. For each of the enzymes,
catalyticrateswerenormalizedtothesubstrateAp
4
A. In
the case of NPP1 and NPP3, substrates were added to
the surface of transfected intact and viable cells, or to
membrane fractions obtained from transfected cells.
Ap
3
A, Ap
4

M
MgCl
2
,or500l
M
EDTA. In the case of NPP2, 150 l
M
EGTA was added to the 2 m
M
MgCl
2
-containing solution to chelate CaCl
2
contained in the storage
solution. Substrate concentrations were 200 l
M
. The 100% values
correspond to 9.3 ± 1.0 nmolÆmin
)1
Æmg protein
)1
(NPP1), 13.7 ±
0.2 nmolÆmin
)1
ÆmL
)1
(NPP2), and 26.6 ± 11.0 nmolÆmin
)1
Æmg pro-
tein

AMP, Ap
4
AtoATPandAMP,andAp
5
AtoAp
4
and
AMP. Similarly, Gp
4
G was hydrolyzed to GTP and
GMP (results not shown).
Differential effect of inhibitors
We further investigated the possibility that the enzymes
differ regarding their sensitivity to potential inhibitors that
have previously been reported to affect hydrolysis of
diadenosine polyphosphates [18]. Ap
4
A(200l
M
)was
chosen as the substrate (Fig. 3). Cibacron Blue (100 l
M
)
strongly inhibited the hydrolysis of Ap
4
Abyallthree
enzymes. PPADS, at 100 l
M
, inhibited the hydrolysis of
Table 1. Substrate preference of NPP1, NPP2 and NPP3. Activity was determined at the surface of viable and transfected CHO cells or using

)1
Æmg protein
)1
(NPP3, surface activity and activity of
membrane fractions, respectively) and 15.6 ± 0.3 nmolÆmin
)1
ÆmL
)1
(NPP2). Values represent means ± SD of three experiments with duplicate
determinations in each (NPP1, NPP3) or mean values ± range of two experiments (NPP2). ND, not determined.
Substrate
NPP1 (relative rate) NPP2 (relative rate) NPP3 (relative rate)
Cell surface Membrane fraction Soluble Cell surface Membrane fraction
Ap
4
A1 1 1 1 1
Ap
3
A 1.21 ± 0.03 1.06 ± 0.05 1.02 ± 0.02 1.19 ± 0.03 1.31 ± 0.08
Ap
5
A 0.98 ± 0.07 1.20 ± 0.13 1.07 ± 0.01 1.05 ± 0.05 1.54 ± 0.17
Gp
4
G ND 1.20 ± 0.20 1.14 ± 0.01 ND 0.45 ± 0.01
Fig. 2. Hydrolysis of diadenosine polyphosphates and product formation by heterologously expressed NPP1, NPP2 and NPP3. NPP1 and NPP3 were
analyzed after expression in CHO cells. The substrates Ap
3
A, Ap
4

A direct comparison of all three heterologously expressed
members of the E-NPP family revealed that they are capable
of hydrolyzing the diadenosine polyphosphates Ap
3
Ato
Ap
5
A. This was true both for the membrane-bound enzymes
NPP1 and NPP3 and for the soluble NPP2. In addition, all
three enzymes hydrolyzed Gp
4
G. This suggests that they
are generally capable of hydrolyzing physiologically released
dinucleoside polyphosphates, including the vasoactive
adenosine polyphosphoguanosines and diguanosine poly-
phosphates [10]. Hydrolysis was asymmetric and involved
the a,b-pyrophosphate bond, resulting in the production of
AMP and Ap
n-1
as the primary hydrolysis products.
Hydrolysis of Ap
n
As had a very alkaline pH optimum,
comparable to that of alkaline phosphatases [18]. Dinucleo-
tide hydrolysis was inhibited by the divalent metal cation
chelator, EDTA, as previously described for other substrates
of the enzymes [29,30]. K
m
values for NPP1, NPP2 and
NPP3 were in the low micromolar range for Ap

4
A) dinucleoside
tetraphosphatase isolated from rat liver [48] shares its
general functional properties with members of the E-NPP
family. As NPP1, NPP2, and NPP3 are all expressed in liver
[49] it is difficult to assign the functional properties of the
enzyme fraction identified to a specific member of the
E-NPP family. Ap
3
AandAp
4
A are hydrolyzed asymmet-
rically by human plasma, and an enzyme chromatographing
at 230 kDa, with properties very similar to those of NPP1,
has been purified from plasma [50]. In human serum, three
isoenzymes splitting Ap
4
AandAp
3
A have been identified,
with K
m
values for Ap
4
A in the range of 2–10 l
M
[51]. These
isoenzymes may relate to the isoforms of the E-NPP family
present in serum [18]. In contrast to ATP, Ap
n

rat C6 glioma cells was found to hydrolyze Ap
4
A, Gp
4
G
and NAD
+
, compatible with the catalytic properties of
NPP1 [53]. Human airway epithelia contain mRNA enco-
ding all three forms of NPPs [54]. Diadenosine polyphos-
phates applied to the apical epithelial surface were
hydrolyzed, as would be expected for NPPs, but catalytic
activities cannot be assigned to an individual enzyme.
Not all Ap
n
A-hydrolyzing enzymes described share the
catalytic properties of NPPs. Whereas the diadenosine
polyphosphate hydrolase activity in the blood [50], in
synaptic membranes from the Torpedo electric organ, or on
chromaffin cells are activated by Ca
2+
ions, the enzyme on
cultured endothelial cells is inhibited by Ca
2+
[21,22]. The
enzyme from adrenomedullary vascular endothelial cells
also has a considerably lower K
m
for Ap
n

),
heparin, and heparan sulfate (both 0.1 mgÆmL
)1
) were applied at an
Ap
4
A concentration of 200 l
M
. The 100% values correspond to
64.6 ± 16.1 nmolÆmin
)1
Æmg protein
)1
(NPP1), 12.3 ± 0.3 nmolÆ
min
)1
ÆmL
)1
(NPP2), and 109.0 ± 24.4 nmolÆmin
)1
Æmg protein
)1
(NPP3). Values represent means ± SD of three experiments with
triplicatedeterminationsineach(NPP1)ormeanvalues±rangeof
two experiments with triplicate determinations in each (NPP2, NPP3).
Ó FEBS 2003 Hydrolysis of diadenosine polyphosphates (Eur. J. Biochem. 270) 2975
differential inhibition by PPADS and suramin can be used
to assign Ap
n
A hydrolysis in individual tissues to defined

broad substrate specificity, including dinucleoside poly-
phosphates, nucleoside triphosphates, nucleotide sugars,
NADH and coenzyme A [60]. Our results suggest that
NPP1, NPP2, and NPP3 are major enzyme candidates for
the hydrolysis of extracellular dinucleoside polyphosphates
released in vertebrate tissues.
Acknowledgements
This work was supported by grants from the Deutsche Forschungs-
gemeinschaft (SFB 269, A4), from the Foerderfonds der Chemischen
Industrie (to H. Zimmermann) and by the National Health and
Medical Research Council of Australia (to J. W. Goding).
References
1. McLennan, A.G. (1992) Ap
4
A and Other Dinucleoside Polyphos-
phates. CRC Press, Boca Raton.
2. Miras-Portugal, M.T., Gualix, J., Mateo, J., Dı
´
az-Herna
´
ndez, M.,
Go
´
mez-Villafuertes, R., Castro, E. & Pintor, J. (1999) Diadeno-
sine polyphosphates, extracellular function and catabolism. Prog.
Brain Res. 120, 397–409.
3. Hoyle, C.H.V., Hilderman, R.H., Pintor, J.J., Schlu
¨
ter, H. &
King, B.F. (2001) Diadenosine polyphosphates as extracellular

-diadenosine 5¢-
tetraphosphate induces nitric oxide release from bovine aortic
endothelial cells. FEBS Lett. 427, 320–324.
7. Steinmetz,M.,Janssen,A.K.,Pelster,F.,Rahn,K.H.&Schlatter,
E. (2002) Vasoactivity of diadenosine polyphosphates in human
small mesenteric resistance arteries. J. Pharmacol. Exp. Ther. 302,
787–794.
8. Van der Giet, M., Schmidt, S., Tolle, M., Jankowski, J., Schlu
¨
ter,
H., Zidek, W. & Tepel, M. (2002) Effects of dinucleoside poly-
phosphates on regulation of coronary vascular tone. Eur. J.
Pharmacol. 448, 207–213.
9. Gabriels, G., Rahn, K.H., Schlatter, E. & Steinmetz, M. (2002)
Mesenteric and renal vascular effects of diadenosine polyphos-
phates (Ap
n
A). Cardiovasc. Res. 56, 22–32.
10. Ralevic, V., Jankowski, J. & Schlu
¨
ter, H. (2001) Structure–activity
relationships of diadenosine polyphosphates (Ap
n
As), adenosine
polyphospho guanosines (Ap
n
Gs) and guanosine polyphospho
guanosines (Gp
n
Gs) at P2 receptors in the rat mesenteric arterial

7
-di (adenosine-5)-heptaphosphate from
human platelets. J. Biol. Chem. 274, 23926–23931.
16. Ralevic, V. & Burnstock, G. (1998) Receptors for purines and
pyrimidines. Pharmacol. Rev. 50, 413–492.
17. Dı
´
az-Herna
´
ndez,M.,Pintor,J.,Castro,E.&Miras-Portugal,
M.T. (2001) Independent receptors for diadenosine pentaphos-
phate and ATP in rat midbrain single synaptic terminals. Eur. J.
Neurosci. 14, 918–926.
18. Zimmermann, H. (2001) Ecto-nucleotidases. In Handbook of
Experimental Pharmacology. Purinergic and Pyrimidergic Signal-
ling. (Abbracchio, M.P. & Williams, M., eds), pp. 209–250.
Springer Verlag, Heidelberg.
19. Lu
¨
thje, J. & Ogilvie, A. (1988) Catabolism of Ap
4
AandAp
3
Ain
whole blood. The dinucleotides are long-lived signal molecules in
the blood ending up as intracellular ATP in the erythrocytes. Eur.
J. Biochem. 173, 241–245.
20. Ramos, A., Pintor, J., Miras-Portugal, M.T. & Rotlla
´
n, P. (1995)

A by rat brain synaptosomes.
Braz.J.Med.Biol.Res.31, 1529–1532.
25. Picher, M. & Boucher, R.C. (2000) Biochemical evidence for an
ecto-alkaline phosphodiesterase I in human airways. Am. J.
Respir. Cell. Mol. Biol. 23, 255–261.
26. Aguilar, J.S., Reyes, R., Asensio, A.C., Oaknin, S., Rotlla
´
n, P. &
Miledi, R. (2001) Ectoenzymatic breakdown of diadenosine
polyphosphates by Xenopus laevis oocytes. Eur. J. Biochem. 268,
1289–1297.
27. Zimmermann, H., Beaudoin, A.R., Bollen, M., Goding, J.W.,
Guidotti, G., Kirley, T.L., Robson, S.C. & Sano, K. (2000) Pro-
posed nomenclature for two novel nucleotide hydrolyzing enzyme
families expressed on the cell surface. In Ecto-ATPases and
Related Ectonucleotidases (Van Duffel, L. & Lemmens, R., eds),
pp. 1–8. Shaker Publishing BV, Maastricht.
28. Zimmermann, H. (2001) Ectonucleotidases: some recent devel-
opments and a note on nomenclature. Drug Dev. Res. 52, 44–56.
29. Bollen,M.,Gijsbers,R.,Ceulemans,H.,Stalmans,W.&Stefan,
C. (2000) Nucleotide pyrophosphatases/phosphodiesterases on the
move. Crit.Rev.Biochem.Mol.Biol.35, 393–432.
30. Goding, J.W. (2000) Ecto-enzymes: physiology meets pathology.
J. Leukoc. Biol. 67, 285–311.
31. Zimmermann, H. (2000) Extracellular metabolism of ATP and
other nucleotides. Naunyn Schmiedeberg’s Arch. Pharmacol. 362,
299–309.
32. Rebbe, N.F., Tong, B.D., Finley, E.M. & Hickman, S. (1991)
Identification of nucleotide pyrophosphatase/alkaline phospho-
diesterase I activity associated with the mouse plasma cell

motility-stimulating protein. J. Biol. Chem. 267, 2524–2529.
39. Belli, S.I., van Driel, I.R. & Goding, J.W. (1993) Identification and
characterization of a soluble form of the plasma cell membrane
glycoprotein PC-1 (5¢-nucleotide phosphodiesterase). Eur. J. Bio-
chem. 217, 421–428.
40. Meerson, N.R., Delautier, D., Durand-Schneider, A. M., Moreau,
A., Schilsky, M.L., Sternlieb, I., Feldmann, G. & Maurice, M.
(1997) Identification of B10, an alkaline phosphodiesterase of the
apical plasma membrane of hepatocytes and biliary cells, in rat
serum: increased levels following bile duct ligation and during the
development of cholangiocarcinoma. Hepatology 27, 563–568.
41. Kegel, B., Braun, N., Heine, P., Maliszewski, C.R. & Zimmer-
mann, H. (1997) An ecto-ATPase and an ecto-ATP diphos-
phohydrolase are expressed in rat brain. Neuropharmacology 36,
1189–1200.
42. Belli, S.I. & Goding, J.W. (1994) Biochemical characterization of
human PC-1, an enzyme possessing alkaline phosphodiesterase I
and nucleotide pyrophosphate activities. Eur. J. Biochem. 226,
433–443.
43. Peterson, G.L. (1977) A simplification of the protein assay method
of Lowry et al. which is more generally applicable. Anal. Biochem.
83, 346–356.
44. Clair,T.,Lee,H.Y.,Liotta,L.A.&Stracke,M.L.(1997)Auto-
taxin is an exoenzyme possessing 5¢-nucleotide phosphodiesterase/
ATP pyrophosphatase and ATPase activities. J. Biol. Chem. 272,
996–1001.
45. Umezu-Goto, M., Kishi, Y., Taira, A., Hama, K., Dohmae, N.,
Takio, K., Yamori, T., Mills, G.B., Inoue, K., Aoki, J. & Arai, H.
(2002) Autotaxin has lysophospholipase D activity leading to
tumor cell growth and motility by lysophosphatidic acid produc-

Ain
human plasma. Purification and characterization of a glycoprotein
complex with 5¢-nucleotide phosphodiesterase activity. Eur. J.
Biochem. 149, 119–127.
51. Lu
¨
thje, J. & Ogilvie, A. (1987) Catabolism of Ap
4
AandAp
3
Ain
human serum. Identification of isoenzymes and their partial
characterization. Eur. J. Biochem. 169, 385–388.
52. Gasmi, L., Cartwright, J.L. & McLennan, A.G. (1998) The
hydrolytic activity of bovine adrenal medullary plasma mem-
branes towards diadenosine polyphosphates is due to alkaline
phosphodiesterase-I. Biochim. Biophys. Acta 1405, 121–127.
53. Grobben, B., Anciaux, K., Roymans, D., Stefan, C., Bollen, M.,
Esmans, E.L. & Slegers, H. (1999) An ecto-nucleotide pyro-
phosphatase is one of the main enzymes involved in the extra-
cellular metabolism of ATP in rat C6 glioma. J. Neurochem. 72,
826–834.
54. Picher, M. & Boucher, R.C. (2001) Metabolism of extracellular
nucleotides in human airways by a multienzyme system. Drug Dev.
Res. 52, 66–75.
55. Rodriguez-Pascual, F., Torres, M., Rotlla
´
n, P. & Miras-Portugal,
M.T. (1992) Extracellular hydrolysis of diadenosine polyphos-
phates, Ap