Soluble Ecto-5′-nucleotidase (5′-NT), Alkaline Phosphatase, and Adenosine Deaminase (ADA1) Activities in Neonatal Blood Favor Elevated Extracellular Adenosine

Abstract

Extracellular adenosine, a key regulator of physiology and immune cell function that is found at elevated levels in neonatal blood, is generated by phosphohydrolysis of adenine nucleotides released from cells and catabolized by deamination to inosine. Generation of adenosine monophosphate (AMP) in blood is driven by cell-associated enzymes, whereas conversion of AMP to adenosine is largely mediated by soluble enzymes. The identities of the enzymes responsible for these activities in whole blood of neonates have been defined in this study and contrasted to adult blood. We demonstrate that soluble 5'-nucleotidase (5'-NT) and alkaline phosphatase (AP) mediate conversion of AMP to adenosine, whereas soluble adenosine deaminase (ADA) catabolizes adenosine to inosine. Newborn blood plasma demonstrates substantially higher adenosine-generating 5'-NT and AP activity and lower adenosine-metabolizing ADA activity than adult plasma. In addition to a role in soluble purine metabolism, abundant AP expressed on the surface of circulating neonatal neutrophils is the dominant AMPase on these cells. Plasma samples from infant observational cohorts reveal a relative plasma ADA deficiency at birth, followed by a gradual maturation of plasma ADA through infancy. The robust adenosine-generating capacity of neonates appears functionally relevant because supplementation with AMP inhibited whereas selective pharmacologic inhibition of 5'-NT enhanced Toll-like receptor-mediated TNF-α production in neonatal whole blood. Overall, we have characterized previously unrecognized age-dependent expression patterns of plasma purine-metabolizing enzymes that result in elevated plasma concentrations of anti-inflammatory adenosine in newborns. Targeted manipulation of purine-metabolizing enzymes may benefit this vulnerable population.

Full text

and Ofer Levy
Laura Gelinas, Tobias Kollmann, Louis Bont
Bergelson, Sarah Burl, Beate Kampmann,
Taiese Bingham, Mirjam Belderbos, Ilana
Tresenriter, José Luis Millán, Anny Usheva,
Matthew Pettengill, Simon Robson, Megan
Adenosine
Blood Favor Elevated Extracellular
Deaminase (ADA1) Activities in Neonatal
Alkaline Phosphatase, and Adenosine
-NT),′-nucleotidase (5′Soluble Ecto-5
Immunology:
doi: 10.1074/jbc.M113.484212 originally published online July 29, 2013
2013, 288:27315-27326.J. Biol. Chem.
10.1074/jbc.M113.484212Access the most updated version of this article at doi:
.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the
Alerts:
When a correction for this article is posted• When this article is cited•
to choose from all of JBC's e-mail alertsClick here
Supplemental material:
http://www.jbc.org/content/suppl/2013/07/29/M113.484212.DC1.html
http://www.jbc.org/content/288/38/27315.full.html#ref-list-1
This article cites 72 references, 28 of which can be accessed free at
by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from by guest on September 22, 2013http://www.jbc.org/Downloaded from

Soluble Ecto-5ⴕ-nucleotidase (5ⴕ-NT), Alkaline Phosphatase,
and Adenosine Deaminase (ADA1) Activities in Neonatal
Blood Favor Elevated Extracellular Adenosine
*
□
S
Received for publication, May 8, 2013, and in revised form, July 26, 2013 Published, JBC Papers in Press, July 29, 2013, DOI 10.1074/jbc.M113.484212
Matthew Pettengill
‡§
, Simon Robson
§¶
, Megan Tresenriter
‡
, José Luis Millán
储
, Anny Usheva
§¶
, Taiese Bingham
‡§
,
Mirjam Belderbos**
1
, Ilana Bergelson
‡
, Sarah Burl
‡‡§§2
, Beate Kampmann
‡‡§§2
, Laura Gelinas
¶¶3
, Tobias Kollmann
储储4
,
Louis Bont**, and Ofer Levy
‡§5
From the
‡
Department of Medicine, Division of Infectious Diseases, Boston Children’s Hospital, Boston, Massachusetts 02115, the
¶
Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts 02215,
§
Harvard Medical School, Boston,
Massachusetts 02115, the
储
Sanford Children’s Health Research Center, Sanford-Burnham Medical Research Institute, La Jolla,
California 92037, the **Department of Pediatrics, University Medical Center Utrecht, 3584 CX Utrecht, The Netherlands, the
‡‡
Vaccinology Theme Group, Medical Research Council Unit, Fajara, The Gambia, and the
§§
Department of Pediatrics, Imperial
College London, London W2 IPG, United Kingdom, and the
¶¶
Experimental Medicine Program, Department of Medicine, and the
储储
Division of Infectious and Immunologic Diseases, Department of Pediatrics, University of British Columbia,
Vancouver, British Columbia V6T 1Z4, Canada
Background: Newborns have elevated plasma adenosine levels, which may influence their immunological function.
Results: Compared with adults, newborns have elevated plasma 5⬘-NT and alkaline phosphatase activities and lower adenosine
deaminase activity.
Conclusion: Soluble enzymes significantly influence extracellular purine metabolism in blood, and the levels of these enzymes
in newborns promote elevated adenosine.
Significance: Higher adenosine generation in newborn blood may promote an anti-inflammatory immunological status.
Extracellular adenosine, a key regulator of physiology and
immune cell function that is found at elevated levels in neonatal
blood, is generated by phosphohydrolysis of adenine nucleo-
tides released from cells and catabolized by deamination to
inosine. Generation of adenosine monophosphate (AMP) in
blood is driven by cell-associated enzymes, whereas conversion
of AMP to adenosine is largely mediated by soluble enzymes.
The identities of the enzymes responsible for these activities in
whole blood of neonates have been defined in this study and con-
trasted to adult blood. We demonstrate that soluble 5ⴕ-nucleotid-
ase (5ⴕ-NT) and alkaline phosphatase (AP) mediate conversion of
AMP to adenosine, whereas soluble adenosine deaminase (ADA)
catabolizes adenosine to inosine. Newborn blood plasma demon-
strates substantially higher adenosine-generating 5ⴕ-NT and AP
activity and lower adenosine-metabolizing ADA activity than adult
plasma. In addition to a role in soluble purine metabolism,
abundant AP expressed on the surface of circulating neonatal
neutrophils is the dominant AMPase on these cells. Plasma sam-
ples from infant observational cohorts reveal a relative plasma
ADA deficiency at birth, followed by a gradual maturation of
plasma ADA through infancy. The robust adenosine-generating
capacity of neonates appears functionally relevant because supple-
mentation with AMP inhibited whereas selective pharmacologic
inhibition of 5ⴕ-NT enhanced Toll-like receptor-mediated TNF-
␣
production in neonatal whole blood. Overall, we have character-
ized previously unrecognized age-dependent expression patterns
of plasma purine-metabolizing enzymes that result in elevated
plasma concentrations of anti-inflammatory adenosine in new-
borns. Targeted manipulation of purine-metabolizing enzymes
may benefit this vulnerable population.
Purine signaling nucleotides and nucleosides, primarily
adenosine triphosphate (ATP) and adenosine, are critical
pathophysiologic mediators. In the context of immunity and
inflammation, extracellular ATP (eATP)
6
is important for
T-cell activation (1, 2) and proliferation (3, 4), promoting neu-
*This work was supported, in whole or in part, by National Institutes of Health
(NIH) Grant 1R01AI100135-01 (to O. L.); NIH Training Grant T32 HD055148
(to M. P.); NIH Grants R01 HL094400, R21 CA164970, and P01 HL087203 (to
S. R.); NIH, NIDCR, Grant R01 DE012889 and NIH, NIAMS, Grant R01
AR053102 (to J. L. M.); and NIH, NIAID, Grant N01 AI50023 (to T. R. K.).
Author’s Choice—Final version full access.
□
S
This article contains supplemental Figs. 1–7.
1
Supported by a European Society for Pediatric Infectious Diseases travel grant.
2
Supported by the Medical Research Council, United Kingdom.
3
Recipient of a Frederick Banting and Charles Best Canada Graduate Scholar-
ship from the Canadian Institutes of Health Research.
4
Supported in part by a Burroughs Wellcome Career Award in the Biomedical
Sciences, a Michael Smith Foundation for Health Research Career Investi-
gator Award, and the AllerGen NCE Grants 07-A1A and 07-B2B.
5
The laboratory of this author is supported in part by Bill and Melinda Gates
Foundation Global Health Award OPPGH5284 and Grand Challenges
Explorations Award OPP1035192. To whom correspondence should be
addressed: Dept. of Medicine, Division of Infectious Diseases, Boston Chil-
dren’s Hospital, 300 Longwood Ave., Boston, MA 02115. Tel.: 617-919-
2904; Fax: 617-730-0254; E-mail: ofer.levy@childrens.harvard.edu.
6
The abbreviations used are: eATP, extracellular ATP; Ado, adenosine; eAdo,
extracellular Ado; ENTPD, ectonucleoside triphosphate diphosphohydro-
lase; AP, alkaline phosphatase; 5⬘-NT, ecto-5⬘-nucleotidase; TLR, Toll-like
receptor; TNAP, tissue-nonspecific alkaline phosphatase; APCP, adenosine
5⬘-(
␣
,
␤
-methylene)-diphosphate; HBSS, Hanks’ buffered salt solution;
EHNA, erythro-9-(2-hydroxy-3-nonyl)adenine hydrochloride; dipyridamole,
2,6-bis(diethanolamino)-4,8-dipiperidinopyrimido[5,4-d]pyrimidine; MFP,
microparticle-free plasma; RBC, red blood cell; ADA, adenosine deaminase;
TLC, thin layer chromatography.
THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 288, NO. 38, pp. 27315–27326, September 20, 2013
Author’s Choice © 2013 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.
SEPTEMBER 20, 2013•VOLUME 288 •NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27315

trophil adhesion (to endothelial cells (5)) and degranulation (6,
7), reactive oxygen species production (8), and multiple other
proinflammatory immune cell functions (9). Extracellular
adenosine (eAdo) is primarily generated by the metabolism of
eATP and has an opposing profile of immunoregulatory effects
from the precursor molecule; eAdo inhibits neutrophil-endo-
thelial adhesion (10, 11) and degranulation (12), reactive oxy-
gen species production (13, 14), and macrophage production of
proinflammatory/Th1-polarizing cytokines (IL-12p70, TNF-
␣
)
(9) as well as T-cell proliferation and effector functions (15, 16).
eATP is primarily transported outside of cells by vesicular traf-
ficking or secreted via pannexin-1 channels (17), where it can
then stimulate P2 purinergic receptors and also serve as a
source of eAdo through sequential dephosphorylation by several
types of ectonucleotidases leading to P1 (Ado) receptor signaling
(18). Thus, the enzymes involved in extracellular purine metabo-
lism critically regulate whether eATP release results in largely P2
receptor-mediated proinflammatory sequelae or P1 receptor
stimulation and an anti-inflammatory response.
There are four families of ectonucleotidases (19): ENTPDs
(including ENTPD1 (CD39); substrates ATP and ADP), ecto-
nucleotide pyrophosphatase/phosphodiesterases (substrates
ATP and ADP), alkaline phosphatases (APs; substrates ATP,
ADP, and AMP), and ecto-5⬘-nucleotidase (5⬘-NT (CD73); sub-
strate AMP). Additionally, it was recently shown that adenylate
kinase (AK1) influences extracellular adenine nucleotide pools
via transphosphorylation (20). 5⬘-NT and AP are both attached
to the plasma membrane by a glycosylphosphatidylinositol
anchor and are found in plasma, possibly as a consequence of
glycosylphosphatidylinositol-specific phospholipase activity
(21–24). Of note, eAdo can be taken up into cells via nucleoside
transporters (25) or be metabolized by adenosine deaminase
(primarily ADA1, whereas the product of another gene
(CECR1), referred to as ADA2, has little activity near physio-
logical concentrations of Ado), generating inosine. Extracellu-
lar ADA1 is found either soluble in fluids or cell-associated via
binding to CD26 (26) or Ado receptors (27). Increased ecto-
nucleotidase activity would both diminish eATP exposure and
enhance eAdo production, shifting subsequent cellular responses
to an anti-inflammatory profile. In addition, modified levels of
ADA1 activity or cellular adenosine transport could promote
eAdo retention. An ectopurine enzyme profile that favors the
metabolism of eATP and production and retention of eAdo
should then mediate dampening of inflammatory and Th1-po-
larizing immune responses.
Innate immune responses in human newborns are distinct
from those of adults (28), favoring higher production of Th2-
and Th17-polarizing cytokines (e.g. IL-4 and IL-6) and dimin-
ished production of proinflammatory/Th1-polarizing cyto-
kines (e.g. TNF-
␣
and IL-12p70) in response to Toll-like receptor
(TLR) agonists or whole microbes in vitro (29–31). Impaired
neonatal production of key cytokines, such as TNF-
␣
, may con-
tribute to impaired mobilization and activation of phagocytes
and to impaired vaccine responses (28). Interestingly, the polar-
ized neonatal cytokine profile may be influenced by soluble fac-
tors in newborn plasma (29, 32, 33), which contains signifi-
cantly higher levels of eAdo than adult plasma (32). In the
current study, we assessed whether the difference in eAdo in
newborn blood was due to differential levels of blood ectoen-
zyme expression. We found that direct eAdo-generating
enzymes and ADA activities in whole blood were primarily sol-
uble, not cell- or microparticle-associated. Remarkably, new-
born whole blood ex vivo generated more and metabolized less
eAdo than adult blood from exogenous purine precursors (ADP
or AMP), due primarily to elevated soluble 5⬘-NT and tissue-
nonspecific alkaline phosphatase (TNAP)-mediated AMPase
activity and lower soluble ADA1 activity, respectively.
MATERIALS AND METHODS
Reagents—Adenosine 5⬘-(
␣
,
␤
-methylene)-diphosphate (APCP)
was obtained from Sigma. EDTA and Hanks’ buffered salt solu-
tion with (HBSS⫹) or without (HBSS⫺) calcium and magne-
sium were from Invitrogen. MLS-0038949 (34) was purchased
from EMD Millipore (Billerica, MA). erythro-9-(2-Hydroxy-3-
nonyl)adenine hydrochloride (EHNA) and 2,6-bis(diethanolamino)-
4,8-dipiperidinopyrimido[5,4-d]pyrimidine (dipyridamole)
were acquired from Tocris Cookson (Bristol, UK). Ficoll-
Paque, Percoll, ammonium hydroxide, and 2-propanol were
purchased from Fisher. 8-Aminoguanosine was obtained from
Santa Cruz Biotechnology, Inc. Other reagents and sources are
listed with assay descriptions.
Blood Collection—Peripheral blood was collected after
informed consent from healthy adult volunteers according to
Boston Children’s Hospital Institutional Review Board-ap-
proved protocols (mean age 31.3 years), and newborn cord
blood (mean gestational age 38.9 weeks) was collected immedi-
ately after elective cesarean section delivery (epidural anesthe-
sia) of the placenta. Births to HIV-positive mothers were
excluded. Human experimentation guidelines of the United
States Department of Health and Human Services, the Brigham
and Women’s Hospital, Beth Israel Medical Center, and Boston
Children’s Hospital were observed, following protocols
approved by the local institutional review boards. Number of
repeats (n) indicates number of independent experiments. For
primary human cells (or plasma), no subject was studied more
than once in each of the different experiments. Blood was col-
lected into syringes containing a final concentration of 20
units/ml heparin (Sagent Pharmaceuticals, Schaumberg, IL)
and used within2hofcollection.
Blood Plasma Collection from International Cohorts—Blood
plasma was collected in a cross-sectional study (as described
previously (35)) at birth and at 1, 2, 3, 4, 6, 9, and 12 months at
the Medical Research Council Sukuta Health Centre in Fajara,
The Gambia. Briefly, for every study subject, written informed
consent of a parent/guardian was obtained. Children were
excluded if they had any signs of intercurrent infection. Neo-
nates were also excluded if they had a low birth weight (⬍2.5 kg)
or were a twin. Venous blood was collected into tubes contain-
ing heparin at 7.5 units/ml blood and centrifuged at 1500 rpm
for 10 min, and plasma was removed and frozen for future anal-
ysis. A separate longitudinal study at birth and at 1 and 2 years
of age from which plasma was obtained was approved by the
Institutional Ethics Review Board at both the University of
Washington and the University of British Columbia, and details
of this cohort and subject recruitment have been described pre-
viously (36). Briefly, following written and informed consent,
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27316 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013

cord or peripheral blood was collected via sterile venipuncture
into tubes containing 143 USP units of sodium heparin (Vacu-
tainer Tubes, BD Biosciences). Blood was centrifuged at 1250
rpm for 10 min, and plasma was removed and frozen at ⫺80 ºC
until further analysis.
Plasma and Hemocyte Preparation—Plasma was prepared
sequentially by centrifuging whole blood at 500 ⫻gfor 5 min
and carefully collecting the upper layer (platelet-rich plasma).
Additional centrifugation at 3,000 ⫻gfor 30 min yielded plate-
let-poor plasma that could be spun at 16,000 ⫻gfor 30 min to
collect microparticle-free plasma (MFP). Washed hemocytes
were generated by centrifuging whole blood at 500 ⫻gfor 5 min
and replacing the plasma volume with HBSS⫹, followed by two
more cycles of centrifugation and supernatant replacement
with HBSS⫹. All centrifugations were performed at ambient
temperature (24 °C).
Neutrophil and Erythrocyte Isolation—Neutrophils were iso-
lated as described previously (37). Briefly, whole blood was lay-
ered onto Hypaque-Ficoll and centrifuged at 1100 ⫻gfor 30
min at room temperature (brake off), and the peripheral blood
mononuclear cell/cord blood mononuclear cell layer and liquid
phase were carefully removed. The upper portion of the RBC/
granulocyte layer was carefully collected and resuspended in
PBS before 1:1 dilution with a 3% dextran (Pharmacosmos)
solution (in saline) followed by repeated inversion (10 times)
prior to allowing the cells to settle for 20 min at room temper-
ature. The leukocyte-rich upper layer was added to new tubes
for centrifugation at 500 ⫻gfor 10 min. Erythrocytes were lysed
by resuspending the pellet in H
2
O for 25 s, followed by imme-
diate tonicity restoration with 2⫻PBS. Cells were spun again at
500 ⫻g, and the resulting pellet was resuspended in HBSS⫺
and spun again at 500 ⫻gfor 10 min. The final pellet was
resuspended in HBSS⫹, and cell density was determined by
hemocytometer and adjusted as necessary with HBSS⫹to a
density of 4 ⫻10
6
cells/ml (viability, as measured by trypan blue
exclusion during quantitation, was ⬎90% for each sample).
Erythrocytes were collected from the remaining cell pellet fol-
lowing RBC/granulocyte layer removal from Hypaque-Ficoll
separation and were washed three times with HBSS⫹(with
centrifugation at 500 ⫻gfor 5 min) before resuspension in an
equal volume of HBSS⫹(500
␮
l of RBC cell pellet ⫹500
␮
lof
HBSS⫹). The purity of erythrocytes (RBCs) collected by this
method was evaluated by the absence of leukocytes in
Wright-stained smears. RBC density was also determined by
hemocytometer.
Enzyme Assays and Thin Layer Chromatography—Enzy-
matic modification of nucleotide and nucleoside substrates was
evaluated utilizing [
14
C]ADP (PerkinElmer Life Sciences),
[
14
C]AMP, and [
14
C]adenosine (Moravek Biochemicals). Sub-
strates were added to prewarmed samples at a concentration of
0.5, 5.0, 50, or 200
␮
M, as indicated, during gentle vortexing
before incubating at 37 °C in a dry bath for specified times. The
50
␮
Msubstrate concentration was selected for the majority of
the samples to model putative responses during inflammatory
tissue damage, during which elevated levels of extracellular
purine substrate may be present, and also for technical consid-
erations to aid in the detection of metabolic products. Plasma
enzyme activity measurements utilizing a saturating concen-
tration of substrate (200
␮
M) and collection at multiple time
points (times indicated) were used to calculate rates of activity
in plasma and were within the linear range of activity. Samples
with non-saturating levels of substrate are displayed as accu-
mulated reaction product at the indicated time point. Reactions
were terminated by adding 3 volumes of a solution containing
50 mMEDTA (Invitrogen) and 30
␮
MEHNA (ADA inhibitor;
Tocris) in HBSS (Invitrogen). Controls for analytes of interest
([
14
C]adenosine or [
14
C]inosine; Moravek) were prepared at
assay-appropriate concentration ranges, and controls and sam-
ples were then applied to silica gel matrix thin layer chromatog-
raphy plates (Sigma-Aldrich) and migrated until the solvent
(6:3:1, 2-propanol/ammonium hydroxide/distilled H
2
O) front
reached 4 cm beyond the application site. Plates were dried and
incubated in a storage phosphor screen cassette overnight, after
which the screens were analyzed on a GE Storm 860 Imager.
Analytes were quantified relative to controls by densitometry
utilizing ImageJ software (version 1.43u; National Institutes of
Health). For assays utilizing 0.5
␮
M[
14
C]adenosine substrate,
plates were incubated in a storage phosphor screen cassette for
14 days before analysis.
Flow Cytometry—Whole blood was washed (2 ml of blood ⫹
10 ml of HBSS⫹; centrifuged at 300 ⫻gfor 5 min; 10 ml super-
natant removed and the remainder reconstituted) to remove
the majority of plasma. Fluorochrome-conjugated monoclonal
antibodies against CD3 (phycoerythrin-Cy7), CD19 (APC-
Cy7), CD14 (Alexa Fluor 700), CD66b (PerCP Cy5.5), TNAP
(phycoerythrin), and CD73/5⬘-NT (APC) (all from BD Biosci-
ences); CD39/ENTPD1 (FITC; eBioscience); and ADA1 (Calbi-
ochem; labeled in-house with Pacific Blue APEX antibody
labeling kit per manufacturer’s recommendations (Invitrogen),
which was also performed for an isotype control antibody) or
appropriate phycoerythrin-, APC-, and FITC-conjugated iso-
type control antibodies (BD Biosciences and eBiosciences) were
added directly to 200
␮
l of washed whole blood at room tem-
perature for 30 min. Samples were added to BD Facs lysing
solution for 10 min and centrifuged at 500 ⫻gfor 5 min. Super-
natants were removed, and cell pellets were washed with PBS
and centrifuged again. Cell pellets were finally resuspended in
1% paraformaldehyde (VWR) and analyzed on a BD LSR II flow
cytometer within 48 h. Data were collected uncompensated,
and compensation, doublet exclusion, and data analysis were
performed with FlowJo software (version 9.2; TreeStar).
ADA Chromogenic Assay—ADA1 and ADA2 activities in
plasma samples were determined with an adenosine deaminase
assay kit per the manufacturer’s instructions (Diazyme Labora-
tories, Poway, CA), run in duplicate with or without EHNA (20
␮
M). ADA2 is not EHNA-sensitive; thus, activity in EHNA-
containing wells was considered to reflect ADA2 activity.
ADA1 activity was calculated by subtracting ADA2 activity
from total ADA activity.
Alkaline Phosphatase Isoform Assay—Blood was collected as
described previously, but without anti-coagulant, and left at
ambient temperature until coagulation (up to 2 h). Serum sam-
ples were then sent to Mayo Medical Laboratories for evalua-
tion of alkaline phosphatase isoform content by electrophoresis
through alkaline-buffered agarose gels with or without wheat
germ lectin for separation, essentially as described previously
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
SEPTEMBER 20, 2013• VOLUME 288• NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27317

(38), followed by an enzymatic assay with a chromogenic sub-
strate, 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetra-
zolium in aminomethyl propanol buffer, pH 10.0, utilizing the
Sebia Hydragel 7 and 15 ISO-PAL systems (Sebia, Norcross,
GA). Results were reported in units/liter for bone, liver type 1,
and liver type 2 AP. Intestinal AP was detected in 1 adult sample
(10 units/liter). No placental or germ cell AP was reported.
Whole Blood Stimulation with Staphylococcus epidermidis—
Experiments assessing bacteria-induced cytokine production
employed S. epidermidis 1457, a clinical strain from a patient
with a central catheter infection that was previously isolated by
Mack et al. (39). Anti-coagulated whole blood was incubated
with APCP (an inhibitor of 5⬘-NT) or buffer control before
stimulation with 10
7
live S. epidermidis/ml, lipopolysaccharide
(LPS; 1
␮
g/ml), or vehicle control for 4 h during end-over-end
rotation at 37 °C. Supernatants were collected following cen-
trifugation and stored at ⫺80 °C prior to measurement of
TNF-
␣
, IL-6, IL-1
␤
, IL-23, and IL-10 by ELISA per the manu-
facturer’s recommendations (BD Biosciences for TNF-
␣
and
IL-10; eBioscience for IL-6, IL-23, and IL-1
␤
).
Plasma alkaline phosphatase activity was evaluated in dietha-
nolamine buffer with 1 mMMg
2⫹
,20
␮
MZn
2⫹
and 2.7 mM
p-nitrophenyl phosphate (Sigma) at pH 9.8 or 7.4, as indicated,
at room temperature. 195
␮
l of buffer was added to 5
␮
lof
sample in a 96-well plate (Corning Inc.), and absorbance at 405
nm was evaluated at 5, 10, and 15 min after reaction initiation.
p-Nitrophenyl phosphate dephosphorylation product nitro-
phenyl phosphate was compared with serially diluted nitro-
phenyl phosphate controls of known concentrations, and the
rate of nitrophenyl phosphate generation was calculated for
each interval (and averaged). AP activity was calculated as
units/liter, which is
␮
mol of nitrophenyl phosphate generat-
ed/min/liter of plasma (
␮
mol/min/liter). Samples were run
in duplicate.
RESULTS
Blood ATPase and ADPase Activities Are Largely Cell-associ-
ated, whereas AMPase Activity Is Largely Plasma-based—Given
the importance of Ado as an endogenous immunoregulatory
molecule and the previously discovered difference in Ado con-
centration between newborn and adult blood plasma (29, 32),
we sought to determine if different levels of enzymatic activity
in whole blood contribute to the Ado bias in newborn plasma.
To this end, radiolabeled ADP ([
14
C]ADP) was added to whole
blood, washed hemocytes, or MFP in the presence of 100
␮
M
dipyridamole (to prevent cellular uptake of Ado), and at indi-
cated times, blood-derived supernatants were collected for sep-
aration by thin layer chromatography (TLC) as described under
“Materials and Methods.” In agreement with previous work
employing adult blood (40), these studies indicated that in both
neonatal cord and adult peripheral blood, dephosphorylation of
ADP to generate AMP was primarily accomplished by cell-as-
sociated enzymes, but dephosphorylation of AMP to generate
Ado mostly occurred in the soluble phase of blood (Fig. 1A). We
did not detect significant differences in total AMPase activity in
the different preparations of plasma (platelet-rich plasma,
platelet-poor plasma, or MFP) but utilized MFP to more com-
pletely demonstrate the soluble nature of the enzymatic activity
(supplemental Fig. 1). Also, all blood was processed within 2 h
of phlebotomy, during which time there was not significant
shedding of AP or 5⬘-NT into the soluble phase of the blood
(supplemental Fig. 1).
Leukocyte Expression of Purine Ectoenzymes by Flow Cy-
tometry—To determine which cell types in blood expressed the
candidate purine-regulating enzymes of interest, we measured
CD39 (ENTPD1), CD73 (5⬘-NT), TNAP, and ADA1 and cell
type markers CD3, CD19, CD14, and CD66b by polychromatic
flow cytometry as described under “Materials and Methods”
(supplemental Fig. 2). In comparison with adults, neonates
demonstrated the following differences in subpopulations: (a)
CD19⫹B cells had lower CD73 expression, in agreement with
previous work (41), and higher TNAP (supplemental Fig. 2B),
and (b) CD66b⫹granulocytes had similar levels of CD39 and
no detectable CD73 but significantly more TNAP and less
ADA1 (supplemental Fig. 2C). Similar levels of CD39 were
detected on newborn and adult CD14⫹monocytes, whereas
CD73, TNAP, and ADA1 were not detected on these cells (sup-
plemental Fig. 2D).
Robust Neonatal Cord Blood Ado Generation Reflects Greater
Cell ATPase/ADPase Activity, Greater Plasma AMPase Activ-
ity, and Lower Soluble ADA—Newborn blood generated signif-
icantly more Ado from ADP than adult blood. ADPase activity
in blood is primarily mediated by cellular ENTPD1, and our
flow cytometry studies (supplemental Fig. 2) suggested that
these activity rates are higher in newborns, due to the well doc-
umented elevated leukocyte densities of newborns (42) rather
than higher per cell activity. Of note, the robust generation of
Ado in newborn blood was also in part due to higher AMPase
FIGURE 1. Cell-associated enzymes dominate newborn and adult blood
ADPase activity, whereas soluble enzymes dominate blood AMPase and
ADA activities and are differentially expressed in newborns and adults.
Whole blood, hemocytes, or plasma (MFP) were incubated with 50
␮
M
[
14
C]ADP for 5 or 15 min (A). MFP was incubated with 50
␮
M[
14
C]AMP and 20
␮
MEHNA (B)or50
␮
M[
14
C]adenosine (C) for 15 min prior to measurement of
conversion to other purine metabolites by TLC. Shown is one representative
of five (A) or six (Band C) independent experiments.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27318 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013

activity in neonatal blood plasma. To selectively measure
AMPase activity, [
14
C]AMP was added to whole blood, washed
hemocytes, or MFP in the presence of dipyridamole and 20
␮
M
EHNA (to inhibit deamination of Ado), and the production of
Ado was measured by TLC. Although hemocytes did have some
AMPase activity, the importance of soluble enzymes for the
metabolism of AMP in blood was confirmed (Fig. 1B). Similarly,
although detectable metabolism of [
14
C]Ado to inosine
occurred on hemocytes, substantially greater activity was
observed in plasma (Fig. 1C; all conditions included dipyrida-
mole). The ADA activity associated with cells and in the plasma
was completely EHNA-sensitive, indicating that ADA1 was
responsible and not ADA2 (data not shown). AMPase and ADA
activity assays, conducted in the absence of dipyridamole, dem-
onstrated that newborn plasma had significantly more AMPase
activity on average (Fig. 2A), and also significantly less ADA
activity (Fig. 2B), than adult plasma, both of which could con-
tribute to greater Ado generation in newborn blood. Plasma
AMP to adenosine conversion and adenosine deamination
were also measured in larger sets of samples with a 50
␮
Msub-
strate concentration, which may be more relevant in the con-
text of tissue damage in vivo (supplemental Fig. 3) (16, 17).
Additionally, plasma adenosine deamination was measured
with 0.5
␮
Msubstrate (supplemental Fig. 4), closer to the basal
plasma adenosine levels (32). Preparations of plasma that
included microparticles (platelet-poor plasma) and platelets
(platelet-rich plasma) did not have AMPase activity (supple-
mental Fig. 1A) or ADA activity
7
distinguishably different from
that of MFP, which was used to evaluate the activity of soluble
enzymes that were not associated with particles or platelets.
Plasma microparticles do incorporate ENTPD1 (CD39), the
activity of which influences endothelial cell activation (43).
Soluble AP and CD73 Comprise the Majority of Blood
AMPase Activity—To assess the relative contribution of soluble
and cell-associated enzymes to blood AMPase activity,
[
14
C]AMP (50
␮
M) was added to either cell (i.e. hemocyte) or
plasma blood fractions, with or without the addition of inhibi-
tors of 5⬘-NT (APCP 100
␮
M) or TNAP (MLS-0038949 (44) 100
␮
M), and conversion to Ado was assessed. Of note, in both new-
born and adult blood studied ex vivo, only a minority of blood
AMPase activity was cell-associated. Upon incubation with
100
␮
MMLS-0038949, 100
␮
MAPCP, or both and 50
␮
M
[
14
C]AMP, newborn hemocytes demonstrated a significantly
higher conversion of AMP to adenosine than adult hemocytes
(Fig. 3). Experiments were performed at volumetric equiva-
lence (i.e. not at equivalent cell numbers), and differences in cell
density, known to be higher in newborn blood (42), probably
contribute to observed differences in purine metabolism. Sur-
prisingly, AP was responsible for a slight majority of newborn
hemocyte-associated AMPase activity, with 5⬘-NT responsible
for the remainder (Fig. 3, Aand C). In contrast, adult hemocytes
demonstrated a reverse pattern, with 5⬘-NT responsible for a
slight majority of total activity. In neutrophils isolated from
peripheral venous blood, AMP conversion to adenosine was
largely due to AP activity, with modest 5⬘-NT activity (Fig. 3, B
and D). Neutrophil AP-mediated ecto-AMPase activity was sig-
nificantly higher on newborn cells than on adult cells. RBCs had
very low levels of AMPase activity (data not shown) but repre-
sent the considerable majority of hemocyte ecto-ADA activity.
Newborn and adult RBC ecto-ADA1 activity rates per cell, as
assessed by measuring conversion of [
14
C]Ado to inosine as
described under “Materials and Methods,” were not signifi-
cantly different (data not shown).
The Majority of Blood AMPase Activity Is Soluble—To deter-
mine which enzymes were responsible for the AMPase activity
in plasma, the conversion of [
14
C]AMP (added at 50
␮
M)toAdo
in plasma (MFP) was assessed with or without the addition of
inhibitors of 5⬘-NT (APCP; 100
␮
M) or TNAP (MLS-0038949
(44); 100
␮
M) as described under “Materials and Methods.”
Although 5⬘-NT was primarily responsible for soluble AMPase
activity, TNAP also significantly contributed (Fig. 4A). Because
previous work (40) utilizing 1
␮
MAMP substrate had ruled out
a role for AP in plasma AMPase activity, we evaluated whether
AP contributed to AMPase activity only at higher substrate
concentrations. Although at a 5
␮
MAMP concentration (i.e.
basal AMP levels), AP did not substantially contribute (Fig. 4B),
at a 100
␮
M(i.e. pathophysiologic) AMP concentration, AP sig-
nificantly contributed to plasma AMPase activity (Fig. 4C). In
fact, the plasma AMPase rate, calculated with a substrate con-
centration of 200
␮
M, demonstrates a significant positive rela-
tionship with plasma alkaline phosphatase activity (supplemen-
tal Fig. 5). These observations suggest that plasma AP, known to
be higher at birth (45), may play a substantial role in blood Ado
generation when extracellular levels of purine nucleotides are
substantially elevated during tissue damage or in a hypoxic
environment. Thus, soluble AP appears to be critically involved
in blood Ado production at elevated levels of extracellular
purine precursors and to a greater extent in newborn than in
adult blood; reduction in AMPase activity with TNAP inhibitor
(MLS-0038949) for newborn plasma was 0.70 ⫾0.56
␮
mol/
7
M. Pettengill, S. Robson, M. Tresenriter, J. L. Millán, A. Usheva, T. Bingham, M.
Belderbos, I. Bergelson, S. Burl, B. Kampmann, L. Gelinas, T. Kollmann, L.
Bont, and O. Levy, unpublished observations.
FIGURE 2. Neonatal cord blood plasma demonstrates high AMPase and
lower relative ADA activities. A, soluble plasma (MFP) AMPase activity was
determined by adding 200
␮
M[
14
C]AMP for 5, 10, and 15 min in the presence
of EHNA before reaction termination, and subsequent TLC was quantified by
densitometry with the rate calculated based on the change in product
between 5 and 10 min and between 10 and 15 min (which were equivalent
and were averaged; n⫽16 each population; Student’s ttest; *, p⬍0.05).
B, soluble plasma (MFP) ADA activity was determined by adding 200
␮
M
[
14
C]adenosine for 30 and 60 min prior to reaction termination and TLC
separation (n⫽12 each population; Student’s ttest; *, p⬍0.05, rate
calculated based on the change in product between 30 and 60 min). Error
bars, S.E.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
SEPTEMBER 20, 2013• VOLUME 288• NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27319

liter/min compared with 0.31 ⫾0.13
␮
mol/liter/min (p⬍0.05;
Student’s ttest; n⫽11 each population; shown as mean ⫾S.D.,
calculated from the Fig. 4Adata set). We characterized the
expression of AP isoforms at birth and found that neonatal
plasma contains greater levels of TNAP (as expected (46)) and a
distinct TNAP isoform profile compared with adult plasma
(supplemental Fig. 6). The majority of total AP at birth was the
bone isoform of TNAP, although relative to adult plasma, neo-
natal plasma also contained high concentrations of the liver
type 2 isoform of TNAP and lower liver TNAP isoform 1. We
also confirmed that higher alkaline phosphatase activity in neo-
natal as compared with adult plasma was also noted when
assayed at pH 7.4, further indicating the physiologic relevance
of these findings.
7
Age-dependent Increase in ADA during Infancy—We next
characterized the ontogeny of soluble ADA1 and 5⬘-NT in two
cohorts of infant plasma samples: (a) plasma collected in a
cross-sectional study at birth and at 1, 2, 3, 4, 6, 9, and 12
months and (b) plasma collected in a separate longitudinal
study at birth and at 1 and 2 years of age. ADA activity was
assayed as described previously with [
14
C]Ado substrate. Of
note, the plasma adenosine deamination was significantly
higher at 1 and 2 years of age than at birth (Fig. 5A). In the
cross-sectional study of West African (Gambian) infants,
FIGURE 3. Newborn neutrophils express relatively high AMPase activity. Newborn washed hemocytes (Aand C) or isolated peripheral blood neutrophils
(4 ⫻10
6
/ml) (Band D) were incubated with or without inhibitors of TNAP (MLS0038949; 100
␮
M) and 5⬘-NT (APCP; 100
␮
M) before the addition of 50
␮
M
[
14
C]AMP for 15 min in the presence of EHNA before reaction termination, and subsequent TLC separation was quantified by densitometry. A,n⫽6 for both
newborn and adult; B,n⫽5 for newborn and n⫽6 for adult. C, one representative of six independent experiments; D, one representative of five independent
experiments. A, newborn versus adult control condition; Student’s ttest; ***, p⫽⬍0.001; all other conditions compared with appropriate population (newborn
or adult) control by analysis of variance with Bonferroni’s multiple comparison test (〫) with p⬍0.01. B, newborn versus adult Student’s ttest; *, p⬍0.05; **,
p⬍0.01; ***, p⬍0.001, all other conditions compared with population control by analysis of variance with Bonferroni’s multiple comparison test, with p⬍0.01
unless indicated as not significant (ns). Error bars, S.E.
FIGURE 4. TNAP and 5ⴕ-NT are key contributors to soluble plasma AMPase activity. The contributions of TNAP and 5⬘-NT were determined by the addition
of a 100
␮
Mconcentration of the selective inhibitor MLS0038949 or APCP, respectively. 50
␮
M[
14
C]AMP was added in the presence of EHNA, and after 5 min,
the reaction was terminated prior to TLC separation and quantified by densitometry (A;n⫽11). TNAP did not influence total AMPase activity in plasma at low
concentrations of substrate (B;5
␮
M[
14
C]AMP, 1 min, n⫽8) but did significantly contribute at high concentrations (C; 100
␮
M[
14
C]AMP; 5 min; n⫽8; Student’s
ttests; *, p⬍0.05; **, p⬍0.01; ***, p⬍0.001). Error bars, S.E.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27320 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013

plasma adenosine deamination was significantly increased by 3
months of age as compared with birth (cord; Fig. 5B). A chro-
mogenic ADA assay utilizing high concentrations of Ado sub-
strate revealed that total ADA and ADA1 activities were signif-
icantly lower in newborn plasma, in agreement with our assay at
more physiological concentrations of substrate (Fig. 5, C–E).
Activity in neonatal plasma of ADA2, a protein with weak
deaminase activity (47) that probably functions rather as a
growth factor (48), was not significantly different between new-
born and adult plasma. No significant changes were observed in
soluble 5⬘-NT activity in plasma during the first 2 years of life in
the above described infant cohorts (data not shown).
Inhibition of Blood 5⬘-NT Activity Enhances whereas the
Addition of AMP Inhibits TLR-mediated TNF-
␣
Production in
Blood—To assess the potential functional relevance of robust
Ado-generating activity in human newborn cord blood, we
tested the effect of selective inhibition of 5⬘-NT. To this end,
human newborn cord and adult peripheral blood were incu-
bated with vehicle control or with the 5⬘-NT-selective inhibitor
APCP. Because our prior studies suggested that Ado may par-
ticularly affect polarization of TLR2-mediated cytokine pro-
duction (32), we assessed the impact of 5⬘-NT inhibition on
cytokine induction by S. epidermidis, a Gram-positive patho-
gen that causes neonatal bacteremia and signals via TLR2 (49).
Consistent with a role for Ado in inhibiting neonatal TNF-
␣
induction, the addition of APCP significantly and selectively
enhanced S. epidermidis-induced production of the proinflam-
matory/Th1-polarizing cytokine TNF-
␣
in newborn whole
blood (supplemental Fig. 7). In contrast, S. epidermidis-in-
duced production of IL-1
␤
and of the Th17-polarizing cyto-
kines IL-6 and IL-23 and of the anti-inflammatory cytokine
IL-10 was not affected by the addition of APCP, indicating cyto-
kine-specific Ado effects on neonatal leukocytes. It should be
noted that S. epidermidis expresses a 5⬘-NT homologue (50),
AdsA, which is also inhibited by APCP and would have contrib-
uted to AMPase activity in this assay. A modest enhancement of
newborn TNF-
␣
in response to LPS (TLR4) in the presence of
APCP did not reach significance, and the other cytokines tested
were unaffected (data not shown). TNF-
␣
in control whole
blood samples was low but detectable and was not significantly
different between newborn and adult samples (data not shown).
Additionally, we have evaluated the ability of AMP or Ado
to modulate the production of proinflammatory cytokine
(TNF-
␣
) in a 96-well plate format whole blood assay. Newborn
cord or adult peripheral blood was diluted with RPMI and
treated with LPS (TLR4 agonist; 100 ng/ml) or mock-treated
with RPMI in the presence or absence of 100
␮
MAMP or Ado.
Additional wells were treated again with 100
␮
MAMP or Ado
2 h after the initial stimulation (other wells received an addition
of RPMI at this time point, and all wells were mixed gently).
After a total of4hofincubation, cells were pelleted by centrif-
ugation at 500 ⫻gfor 5 min, and supernatants were evaluated
by ELISA for TNF-
␣
(Fig. 6). Because no receptor has been
described for AMP, the effect of AMP is presumed to be medi-
FIGURE 5. Soluble plasma ADA levels increase significantly during the first year of life. A, soluble ADA activity was determined by the addition of 50
␮
M
[
14
C]adenosine for 5 min to microparticle-free plasma collected at 0, 1, and 2 years of age from 12 subjects (paired Student’s ttests; 0 versus 1, p⬍0.001; 0 versus
2, p⬍0.0001; 1 versus 2, p⬍0.01). B, a cohort of platelet-rich plasma samples from a previous study, including samples from cord blood and 1, 2, 3, 4, 6, 9, and
12 months of age (1 sample/subject), were processed to MFP and incubated with 50
␮
M[
14
C]adenosine for 15 min (analysis of variance with Bonferroni’s
multiple comparison test; 0 versus 3 months, p⬍0.01; 0 versus 4 months, p⬍0.05). Total ADA (C), ADA1 (D), and ADA2 (E) activity were evaluated by a
chromogenic assay as described under “Materials and Methods.” Total ADA (n⫽11 newborn, n⫽14 adult; Student’s ttest; **, p⬍0.01) (C) and ADA1 activity
(n⫽11 newborn; n⫽14 adult; Student’s ttest; *, p⬍0.05) (D) were significantly lower in newborn samples. Error bars, S.E.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
SEPTEMBER 20, 2013• VOLUME 288• NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27321

ated by Ado following dephosphorylation of AMP. AMP and
Ado addition diminished TNF-
␣
production. In samples
treated twice, AMP and Ado were significantly more inhibitory
in newborn cord than in adult peripheral blood. The increased
sensitivity in the newborn population to AMP may be due to
elevated AMPase activity or greater sensitivity to the metabolic
product Ado, whereas increased sensitivity to Ado may also be
related to a deficiency in the newborn samples of ADA, result-
ing in slower clearance of Ado. We evaluated the plasma
AMPase activity from the same blood samples used in Fig. 6Ato
determine if there was a correlation between plasma AMPase
activity and TNF-
␣
production. There was a significant, nega-
tive relationship between plasma AMPase activity and whole
blood TNF-
␣
production; samples with higher AMPase activity
produced less TNF-
␣
in response to LPS (Fig. 6B).
DISCUSSION
Our study has characterized Ado-generating and metaboliz-
ing activity in human neonatal blood. eATP and eAdo signaling
impacts and regulates a wide variety of cellular functions, and
thus the regulation of these molecules in the extracellular space
is critical. Because eAdo is elevated in newborn blood, which
potentially influences differential immunological function in
this population (28), we assessed whether newborn eAdo regu-
lation in blood differed from that in adults and identified the
enzymes involved. We found that ATPase and ADPase activity
in whole blood is primarily mediated by cell-associated
ENTPD1 (CD39) (51), and although such hemocyte activity was
higher by volume in newborn blood (Fig. 1A), this may reflect
greater numbers of leukocytes at birth (52). Thus, eATP and
extracellular ADP dephosphorylation appears to be higher in
newborn blood due to elevated numbers of leukocytes, which
express similar levels of ENTPD1 per cell compared with adult
cells. AP significantly contributes to cell-associated AMPase
activity and, at elevated levels of AMP (ⱖ50
␮
M), comprises the
majority of cellular AMPase activity in blood at birth (Fig. 4, B
and C). Although we found modest levels of AP expression on
subsets of circulating lymphocytes by flow cytometry (supple-
mental Fig. 1A), CD66b ⫹granulocytes (primarily neutrophils)
express abundant AP but low CD73 at birth (for expression, see
supplemental Fig. 1C; for activity, see Fig. 3B). Because puriner-
gic signaling and ectopurine metabolism are critical for several
aspects of neutrophil function (9), including chemotaxis (53),
future work will focus on the functional implications of greater
granulocyte-associated AP expression at birth.
In marked contrast to the dephosphorylation of ATP and
ADP in blood that is largely cell-based, extracellular dephos-
phorylation of AMP in newborn cord and adult peripheral
blood is primarily accomplished by soluble plasma factors, spe-
cifically soluble 5⬘-NT and APs (see model in Fig. 7). Of note,
5⬘-NT is also expressed on endothelial cells (11), and in vivo,
this may significantly impact blood AMPase activity.
There are four human AP genes: ALPL (tissue-nonspecific
AP, high levels in bone, kidney, and liver but widely expressed in
other tissues as well), ALPP (placental AP), ALPPL2 (germ cell
AP), and ALPI (intestinal AP) (45). We found that the majority
of total AP at birth was the bone isoform of the ALPL gene
product (TNAP), although relative to adult plasma, neonatal
plasma also contained relatively high concentrations of liver
type 2 isoform TNAP and lower liver TNAP isoform 1 (supple-
mental Fig. 6). Serum AP levels, predominantly the bone iso-
form of TNAP early in life, are elevated during infancy and
childhood compared with adulthood (54). In agreement with
our results (Fig. 3B), TNAP did not measurably contribute to
AMP hydrolysis in blood at 1
␮
MAMP substrate (40). However,
at elevated levels of the purine substrate AMP (⬎50
␮
M), AP
significantly contributed to purine metabolism (Fig. 3Cand
supplemental Fig. 5). There appears to be a wide range of
(patho)physiologically relevant purine concentrations in vivo.
Low concentrations of eATP (0.01–0.1
␮
M) are released from
resting cells, and transient moderate increases in eATP (to ⬍5
␮
M) mediate critical cell signaling functions (17). Larger
releases of eATP (e.g. in the context of cellular damage) consti-
tute a danger signal, reaching ⬃10–20
␮
Min rat and human
blood following vascular injury (55) and ⬎100
␮
Min murine
FIGURE 6. Extracellular AMP metabolism leads to diminished LPS-induced TNF-
␣
production in newborn and adult blood (A). Newborn cord or adult peripheral
blood was diluted 1:1 (final) with RPMI and treated with LPS (100 ng/ml) or mock-treated with RPMI and treated with 100
␮
MAMP or Ado once (at start) or twice
(second addition at 2 h), as indicated, and all wells were mixed gently at the 2 h time point. After a total of4hofincubation (37 °C, 5% CO
2
, 96-well round bottom
plates), supernatants were evaluated by ELISA for TNF-
␣
(n⫽13 newborn, n⫽15 adult; Student’s ttest; *, p⬍0.05). Whole blood samples with higher plasma
AMPase activity produce less TNF-
␣
in response to LPS (B). Plasma AMP dephosphorylation to adenosine was determined as in Fig. 3; n⫽11 for each
population; statistical analysis by linear regression, significantly non-zero slope for adult and combined populations; *, p⬍0.05; newborn population not
significant. Error bars, S.E.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27322 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013

tumor interstitia (56). In vitro EC
50
values for the P2X and P2Y
eATP receptors have been estimated to be as low as 0.05
␮
Mfor
P2X
1
and as high as 780
␮
Mfor P2X
7
(9), which plays a critical
role in inflammatory responses to intracellular pathogens (57,
58). eAdo is generated primarily via metabolism of eAMP from
extracellular ADP and eATP released from platelets and cells
(9, 17). EC
50
values for the adenosine receptors have been esti-
mated to be as low as 0.18
␮
M(A
1
) and as high as 64.1
␮
M(A
2B
)
(9). eAdo concentrations in the murine central nervous system
are typically in the 0.01–0.1
␮
Mrange but upon systemic insult
increase into the ⬎1
␮
Mrange (59), having been measured as
high as 40
␮
M(60). Working within the limits of detection of the
thin layer chromatography assay, we have herein studied enzy-
matic regulation of extracellular purines in blood and blood
plasma ex vivo utilizing substrate concentrations at 0.5
␮
M
eAdo, close to basal plasma eAdo concentrations (⬃0.15 and
0.05
␮
MeAdo in newborns and adults, respectively (32)). For
the purposes of measuring enzymatic rates at elevated and
saturating concentrations, we have also characterized these
plasma enzymes at 50 and 200
␮
MeAdo and eAMP, which are
relatively high purine substrate concentrations that might be
achieved locally during pathophysiologic states of injury and
inflammatory tissue damage. Overall, our results indicate that
neonatal purine-metabolizing enzymatic activity is signifi-
cantly and consistently different from that of adults across a
range of substrate concentrations, such that neonatal plasma
will generate more eAdo and metabolize less of that eAdo to
inosine, resulting in the higher ambient eAdo concentrations.
Our results are probably valid in that (a) age-dependent matu-
ration of ADA expression was observed across multiple and
diverse newborn and infant cohorts in Boston, The Gambia
(West Africa), and British Columbia, and (b) they are consistent
with the higher newborn than adult plasma eAdo concentra-
tions measured ex vivo (32).
Overall, enzymatic generation of eAdo is higher in newborn
blood than in adult blood, and enzymatic deamination of eAdo
is lower, favoring elevated levels of eAdo in this population (see
model in Fig. 7). Of note, we also demonstrate that higher
soluble AMPase activity correlates with lower LPS-induced
proinflammatory TNF-
␣
production in blood stimulated
with LPS (Fig. 6B). Thus, the endogenous purine enzyme
profile may critically regulate innate immune responses.
Moreover, consistent with our prior studies (32), neonatal leu-
kocytes appeared to be more sensitive to the inhibitory effects
of Ado in that addition of AMP resulted in more profound
inhibition of LPS-induced TNF production in newborn than in
adult blood. Our demonstration that inhibition of 5⬘-NT activ-
ity with APCP enhances TNF production induced by S. epider-
midis, a bacterium that activates human leukocyte cytokine
production via TLR2 (49), also suggests that such eAdo gener-
ation may be functionally relevant, contributing to suppression
of proinflammatory/Th1-polarizing cytokine while not influ-
encing Th17-polarizing cytokines, such as IL-6, and anti-in-
flammatory cytokines, such as IL-10 (supplemental Fig. 7).
Such patterns of low Th1 but preserved Th2/Th17/anti-inflam-
matory fetal/neonatal cytokine production have been observed
in vitro and in vivo (35, 61, 62). This cytokine skew may serve
to limit fetal-maternal alloimmune reactions and excessive
inflammation upon initial colonization in early life but may also
render the newborn more susceptible to infections with intra-
cellular pathogens and impair vaccine responses.
The immunosuppressive activity of Ado is abrogated by
ADA-mediated deamination of Ado to generate inosine. ADA1
(ADA) is primarily responsible for the deamination of Ado at
FIGURE 7. Distinct purine metabolism at birth results in higher ambient adenosine concentrations. This figure presents a model that synthesizes both
published and novel information from this current study highlighting age-specific differences in sequential cell- and plasma-based generation of Ado from ATP
and of deamination of Ado to inosine (Ino). A, newborn blood contains a high density of CD39/ENTPD1-expressing leukocytes, contributing to high ATPase and
ADPase activity (i); high plasma concentration of 5⬘-NT and AP that drive conversion of AMP to Ado (ii); and relative deficiency of ADA, resulting in a net increase
in ambient adenosine concentrations and capacity to generate adenosine following release of purine substrates (iii). B, in marked contrast, adult blood features
lower density of CD39/ENTPD1-expressing leukocytes (i), lower plasma 5⬘-NT and AP activity (ii), and relatively greater ADA activity, resulting in lower basal
adenosine concentrations (iii).
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
SEPTEMBER 20, 2013• VOLUME 288• NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27323

physiological levels of substrate (19), whereas both ADA1 and
ADA2 (CECR1) have co-stimulatory and growth factor activi-
ties. Extracellular ADA is found either soluble in fluids or cell-
associated via binding to CD26 (26), or Ado receptors (27).
ADA1 expression is not well characterized in newborns unless
it is entirely absent, as in ADA deficiency causing severe com-
bined immunodeficiency (63). We have discovered a physio-
logic deficiency of soluble plasma ADA at birth and character-
ized the age-dependent maturation of plasma ADA expression
across two international cohorts, revealing that levels of this key
enzyme gradually rise during the first year of life. Whereas the
uncommon genetic deficiency of ADA severely impairs multi-
ple immune responses in effected individuals, our discovery of a
relative physiologic ADA deficiency that is apparently normal
at birth and early infancy suggests a potential mechanism that
may contribute to the general susceptibility of healthy new-
borns to bacterial, mycobacterial, and viral infections (64). Of
note, a similar profile (elevated plasma AP (65), 5⬘-NT (66), and
eAdo (66) and lower total ADA (but not lower ADA1) (67)) is
found during pregnancy, another functionally distinct immu-
nologic state associated with impaired Th1-polarizing cytokine
responses and enhanced susceptibility to intracellular patho-
gens (68, 69).
Our results may also have translational implications. First, a
great deal of preclinical biomedical research directed at devel-
opment of immunomodulatory agents, such as immune
response modifiers, adjuvants, and vaccines, employs in vitro
assays of leukocytes. To the extent that eAdo exerts relevant
immunomodulatory effects, such efforts to model age-specific
immune responses in vitro should take into account that
plasma demonstrates age-specific Ado-generating activity.
Accordingly, rather than culturing the leukocytes to be com-
pared in artificial media or heat-treated pooled adult sera, as is
often done, such assays may be optimally conducted by employ-
ing age-specific plasma matched to the age of the individuals
from which the leukocytes are isolated. From a therapeutic
standpoint, multiple purine analogues, such as adenosine, caf-
feine, and imiquimod, are used clinically, and several others are
in biopharmaceutical development. Imidazoquinolines, such as
the FDA-approved topical antiviral agent imiquimod, are
small synthetic purine analogues that activate TLR7 and/or
TLR8, which also interact with the Ado system (70 –72). Cer-
tain imidazoquinolines that are relatively more refractory to
the inhibitory effects of Ado, such as the TLR7/8 agonist
R-848 (resiquimod), are particularly effective in activating
human neonatal monocytes and monocyte-derived dendritic
cells, suggesting potential as neonatal immune response modi-
fiers and/or vaccine adjuvants (73). To the extent that enhance-
ment of neonatal immune responses may be beneficial, as has
been demonstrated in neonatal murine studies (74), agents that
are relatively refractory to Ado inhibition or that inhibit AP
and/or 5⬘-NT (thereby reducing generation of Ado) may
enhance neonatal host defense (75).
Acknowledgments—We thank Christy Mancuso for technical support
as well as Dr. Michael Wessels for several helpful discussions regard-
ing the manuscript.
REFERENCES
1. Schenk, U., Westendorf, A. M., Radaelli, E., Casati, A., Ferro, M., Fum-
agalli, M., Verderio, C., Buer, J., Scanziani, E., and Grassi, F. (2008) Puri-
nergic control of T cell activation by ATP released through pannexin-1
hemichannels. Sci. Signal. 1, ra6
2. Filippini, A., Taffs, R. E., and Sitkovsky, M. V. (1990) Extracellular ATP in
T-lymphocyte activation. Possible role in effector functions. Proc. Natl.
Acad. Sci. U.S.A. 87, 8267–8271
3. Baricordi, O. R., Ferrari, D., Melchiorri, L., Chiozzi, P., Hanau, S., Chiari,
E., Rubini, M., and Di Virgilio, F. (1996) An ATP-activated channel is
involved in mitogenic stimulation of human T lymphocytes. Blood 87,
682–690
4. Baricordi, O. R., Melchiorri, L., Adinolfi, E., Falzoni, S., Chiozzi, P., Buell,
G., and Di Virgilio, F. (1999) Increased proliferation rate of lymphoid cells
transfected with the P2X
7
ATP receptor. J. Biol. Chem. 274, 33206–33208
5. Sud’ina, G. F., Mirzoeva, O. K., Galkina, S. I., Pushkareva, M. A., and
Ullrich, V. (1998) Involvement of ecto-ATPase and extracellular ATP in
polymorphonuclear granulocyte-endothelial interactions. FEBS Lett. 423,
243–248
6. Cockcroft, S., and Stutchfield, J. (1989) ATP stimulates secretion in hu-
man neutrophils and HL60 cells via a pertussis toxin-sensitive guanine
nucleotide-binding protein coupled to phospholipase C. FEBS Lett. 245,
25–29
7. Balazovich, K. J., and Boxer, L. A. (1990) Extracellular adenosine nucleo-
tides stimulate protein kinase C activity and human neutrophil activation.
J. Immunol. 144, 631–637
8. Chen, Y., Shukla, A., Namiki, S., Insel, P. A., and Junger, W. G. (2004) A
putative osmoreceptor system that controls neutrophil function through
the release of ATP, its conversion to adenosine, and activation of A2
adenosine and P2 receptors. J. Leukoc. Biol. 76, 245–253
9. Bours, M. J., Swennen, E. L., Di Virgilio, F., Cronstein, B. N., and Dagnelie,
P. C. (2006) Adenosine 5⬘-triphosphate and adenosine as endogenous
signaling molecules in immunity and inflammation. Pharmacol. Ther.
112, 358– 404
10. Cronstein, B. N., Levin, R. I., Belanoff, J., Weissmann, G., and Hirschhorn,
R. (1986) Adenosine. An endogenous inhibitor of neutrophil-mediated
injury to endothelial cells. J. Clin. Invest. 78, 760–770
11. Eltzschig, H. K., Thompson, L. F., Karhausen, J., Cotta, R. J., Ibla, J. C.,
Robson, S. C., and Colgan, S. P. (2004) Endogenous adenosine produced
during hypoxia attenuates neutrophil accumulation. Coordination by ex-
tracellular nucleotide metabolism. Blood 104, 3986–3992
12. Bouma, M. G., Jeunhomme, T. M., Boyle, D. L., Dentener, M. A., Voitenok,
N. N., van den Wildenberg, F. A., and Buurman, W. A. (1997) Adenosine
inhibits neutrophil degranulation in activated human whole blood. In-
volvement of adenosine A2 and A3 receptors. J. Immunol. 158,
5400–5408
13. Cronstein, B. N., Kramer, S. B., Weissmann, G., and Hirschhorn, R. (1983)
Adenosine. A physiological modulator of superoxide anion generation by
human neutrophils. J. Exp. Med. 158, 1160–1177
14. Cronstein, B. N., Rosenstein, E. D., Kramer, S. B., Weissmann, G., and
Hirschhorn, R. (1985) Adenosine. A physiologic modulator of superoxide
anion generation by human neutrophils. Adenosine acts via an A2 recep-
tor on human neutrophils. J. Immunol. 135, 1366–1371
15. Cso´ka, B., Himer, L., Selmeczy, Z., Vizi, E. S., Pacher, P., Ledent, C., Deitch,
E. A., Spolarics, Z., Ne´meth, Z. H., and Hasko´ , G. (2008) Adenosine A2A
receptor activation inhibits T helper 1 and T helper 2 cell development
and effector function. FASEB J. 22, 3491–3499
16. Sitkovsky, M. V., and Ohta, A. (2005) The “danger” sensors that STOP the
immune response. The A2 adenosine receptors? Trends Immunol. 26,
299–304
17. Trautmann, A. (2009) Extracellular ATP in the immune system. More
than just a “danger signal”. Sci. Signal. 2, pe6
18. Robson, S. C., Se´vigny, J., and Zimmermann, H. (2006) The E-NTPDase
family of ectonucleotidases. Structure function relationships and patho-
physiological significance. Purinergic Signal. 2, 409– 430
19. Yegutkin, G. G. (2008) Nucleotide- and nucleoside-converting ectoen-
zymes. Important modulators of purinergic signalling cascade. Biochim.
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27324 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013

Biophys. Acta 1783, 673–694
20. Yegutkin, G. G., Wieringa, B., Robson, S. C., and Jalkanen, S. (2012) Me-
tabolism of circulating ADP in the bloodstream is mediated via integrated
actions of soluble adenylate kinase-1 and NTPDase1/CD39 activities.
FASEB J. 26, 3875–3883
21. Lehto, M. T., and Sharom, F. J. (1998) Release of the glycosylphosphatidy-
linositol-anchored enzyme ecto-5⬘-nucleotidase by phospholipase C. Cat-
alytic activation and modulation by the lipid bilayer. Biochem. J. 332,
101–109
22. Strohmeier, G. R., Lencer, W. I., Patapoff, T. W., Thompson, L. F., Carlson,
S. L., Moe, S. J., Carnes, D. K., Mrsny, R. J., and Madara, J. L. (1997) Surface
expression, polarization, and functional significance of CD73 in human
intestinal epithelia. J. Clin. Invest. 99, 2588–2601
23. Low, M. G. (1987) Biochemistry of the glycosyl-phosphatidylinositol
membrane protein anchors. Biochem. J. 244, 1–13
24. Solter, P. F., and Hoffmann, W. E. (1999) Solubilization of liver alkaline
phosphatase isoenzyme during cholestasis in dogs. Am. J. Vet. Res. 60,
1010–1015
25. Molina-Arcas, M., Casado, F. J., and Pastor-Anglada, M. (2009) Nucleo-
side transporter proteins. Curr. Vasc. Pharmacol. 7, 426– 434
26. De Meester, I., Vanham, G., Kestens, L., Vanhoof, G., Bosmans, E., Gigase,
P., and Scharpe´, S. (1994) Binding of adenosine deaminase to the lympho-
cyte surface via CD26. Eur. J. Immunol. 24, 566–570
27. Herrera, C., Casado´, V., Ciruela, F., Schofield, P., Mallol, J., Lluis, C., and
Franco, R. (2001) Adenosine A2B receptors behave as an alternative an-
choring protein for cell surface adenosine deaminase in lymphocytes and
cultured cells. Mol. Pharmacol. 59, 127–134
28. Levy, O. (2007) Innate immunity of the newborn. Basic mechanisms and
clinical correlates. Nat. Rev. Immunol. 7, 379–390
29. Levy, O., Zarember, K. A., Roy, R. M., Cywes, C., Godowski, P. J., and
Wessels, M. R. (2004) Selective impairment of TLR-mediated innate im-
munity in human newborns. Neonatal blood plasma reduces monocyte
TNF-
␣
induction by bacterial lipopeptides, lipopolysaccharide, and imi-
quimod, but preserves the response to R-848. J. Immunol. 173, 4627–4634
30. Nguyen, M., Leuridan, E., Zhang, T., De Wit, D., Willems, F., Van Damme,
P., Goldman, M., and Goriely, S. (2010) Acquisition of adult-like TLR4 and
TLR9 responses during the first year of life. PLoS One 5, e10407
31. Willems, F., Vollstedt, S., and Suter, M. (2009) Phenotype and function of
neonatal DC. Eur. J. Immunol. 39, 26–35
32. Levy, O., Coughlin, M., Cronstein, B. N., Roy, R. M., Desai, A., and Wes-
sels, M. R. (2006) The adenosine system selectively inhibits TLR-mediated
TNF-
␣
production in the human newborn. J. Immunol. 177, 1956–1966
33. Belderbos, M. E., Levy, O., Stalpers, F., Kimpen, J. L., Meyaard, L., and
Bont, L. (2012) Neonatal plasma polarizes TLR4-mediated cytokine re-
sponses towards low IL-12p70 and high IL-10 production via distinct
factors. PLoS One 7, e33419
34. Dahl, R., Sergienko, E. A., Su, Y., Mostofi, Y. S., Yang, L., Simao, A. M.,
Narisawa, S., Brown, B., Mangravita-Novo, A., Vicchiarelli, M., Smith,
L. H., O’Neill, W. C., Milla´n, J. L., and Cosford, N. D. (2009) Discovery and
validation of a series of aryl sulfonamides as selective inhibitors of tissue-
nonspecific alkaline phosphatase (TNAP). J. Med. Chem. 52, 6919– 6925
35. Burl, S., Townend, J., Njie-Jobe, J., Cox, M., Adetifa, U. J., Touray, E.,
Philbin, V. J., Mancuso, C., Kampmann, B., Whittle, H., Jaye, A., Flanagan,
K. L., and Levy, O. (2011) Age-dependent maturation of Toll-like recep-
tor-mediated cytokine responses in Gambian infants. PLoS One 6, e18185
36. Corbett, N. P., Blimkie, D., Ho, K. C., Cai, B., Sutherland, D. P., Kallos, A.,
Crabtree, J., Rein-Weston, A., Lavoie, P. M., Turvey, S. E., Hawkins, N. R.,
Self, S. G., Wilson, C. B., Hajjar, A. M., Fortuno, E. S., 3rd, and Kollmann,
T. R. (2010) Ontogeny of Toll-like receptor mediated cytokine responses
of human blood mononuclear cells. PLoS One 5, e15041
37. Nauseef, W. M. (2007) Isolation of human neutrophils from venous blood.
Methods Mol. Biol. 412, 15–20
38. Rosalki, S. B., and Foo, A. Y. (1984) Two new methods for separating and
quantifying bone and liver alkaline phosphatase isoenzymes in plasma.
Clin. Chem. 30, 1182–1186
39. Mack, D., Fischer, W., Krokotsch, A., Leopold, K., Hartmann, R., Egge, H.,
and Laufs, R. (1996) The intercellular adhesin involved in biofilm accumu-
lation of Staphylococcus epidermidis is a linear
␤
-1,6-linked glucosamin-
oglycan. Purification and structural analysis. J. Bacteriol. 178, 175–183
40. Coade, S. B., and Pearson, J. D. (1989) Metabolism of adenine nucleotides
in human blood. Circ. Res. 65, 531–537
41. Thompson, L. F., Ruedi, J. M., O’Connor, R. D., and Bastian, J. F. (1986)
Ecto-5⬘-nucleotidase expression during human B cell development. An
explanation for the heterogeneity in B lymphocyte ecto-5⬘-nucleotidase
activity in patients with hypogammaglobulinemia. J. Immunol. 137,
2496–2500
42. Schelonka, R. L., Yoder, B. A., desJardins, S. E., Hall, R. B., and Butler, J.
(1994) Peripheral leukocyte count and leukocyte indexes in healthy new-
born term infants. J. Pediatr. 125, 603–606
43. Banz, Y., Beldi, G., Wu, Y., Atkinson, B., Usheva, A., and Robson, S. C.
(2008) CD39 is incorporated into plasma microparticles where it main-
tains functional properties and impacts endothelial activation. Br. J.
Haematol. 142, 627–637
44. Sergienko, E., Su, Y., Chan, X., Brown, B., Hurder, A., Narisawa, S., and
Milla´n, J. L. (2009) Identification and characterization of novel tissue-
nonspecific alkaline phosphatase inhibitors with diverse modes of action.
J. Biomol. Screen 14, 824– 837
45. Millán, J. L. (2006) Mammalian Alkaline Phosphatases: From Biology to
Applications in Medicine and Biotechnology, Wiley-VCH, Weinheim,
Germany
46. Lockitch, G., Halstead, A. C., Albersheim, S., MacCallum, C., and Quigley,
G. (1988) Age- and sex-specific pediatric reference intervals for biochem-
istry analytes as measured with the Ektachem-700 analyzer. Clin. Chem.
34, 1622–1625
47. Zavialov, A. V., and Engstro¨m, A. (2005) Human ADA2 belongs to a new
family of growth factors with adenosine deaminase activity. Biochem. J.
391, 51–57
48. Zavialov, A. V., Yu, X., Spillmann, D., and Lauvau, G. (2010) Structural
basis for the growth factor activity of human adenosine deaminase ADA2.
J. Biol. Chem. 285, 12367–12377
49. Strunk, T., Power Coombs, M. R., Currie, A. J., Richmond, P., Golenbock,
D. T., Stoler-Barak, L., Gallington, L. C., Otto, M., Burgner, D., and Levy,
O. (2010) TLR2 mediates recognition of live Staphylococcus epidermidis
and clearance of bacteremia. PLoS One 5, e10111
50. Thammavongsa, V., Kern, J. W., Missiakas, D. M., and Schneewind, O.
(2009) Staphylococcus aureus synthesizes adenosine to escape host im-
mune responses. J. Exp. Med. 206, 2417–2427
51. Deaglio, S., and Robson, S. C. (2011) Ectonucleotidases as regulators of
purinergic signaling in thrombosis, inflammation, and immunity. Adv.
Pharmacol. 61, 301–332
52. Proytcheva, M. A. (2011) Diagnostic Pediatric Hematopathology, pp. 5–20,
Cambridge University Press, Cambridge, UK
53. Junger, W. G. (2011) Immune cell regulation by autocrine purinergic sig-
nalling. Nat. Rev. Immunol. 11, 201–212
54. Crofton, P. M. (1987) Properties of alkaline phosphatase isoenzymes in
plasma of preterm and term neonates. Clin. Chem. 33, 1778–1782
55. Born, G. V., and Kratzer, M. A. (1984) Source and concentration of extra-
cellular adenosine triphosphate during haemostasis in rats, rabbits and
man. J. Physiol. 354, 419– 429
56. Pellegatti, P., Raffaghello, L., Bianchi, G., Piccardi, F., Pistoia, V., and Di
Virgilio, F. (2008) Increased level of extracellular ATP at tumor sites. In
vivo imaging with plasma membrane luciferase. PLoS One 3, e2599
57. Coutinho-Silva, R., Correˆa, G., Sater, A. A., and Ojcius, D. M. (2009) The
P2X
7
receptor and intracellular pathogens. A continuing struggle. Puri-
nergic Signal. 5, 197–204
58. Tschopp, J., and Schroder, K. (2010) NLRP3 inflammasome activation.
The convergence of multiple signalling pathways on ROS production?
Nat. Rev. Immunol. 10, 210–215
59. Pedata, F., Corsi, C., Melani, A., Bordoni, F., and Latini, S. (2001) Adeno-
sine extracellular brain concentrations and role of A2A receptors in ische-
mia. Ann. N.Y. Acad. Sci. 939, 74–84
60. Hagberg, H., Andersson, P., Lacarewicz, J., Jacobson, I., Butcher, S., and
Sandberg, M. (1987) Extracellular adenosine, inosine, hypoxanthine, and
xanthine in relation to tissue nucleotides and purines in rat striatum dur-
ing transient ischemia. J. Neurochem. 49, 227–231
61. Angelone, D. F., Wessels, M. R., Coughlin, M., Suter, E. E., Valentini, P.,
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
SEPTEMBER 20, 2013• VOLUME 288• NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27325

Kalish, L. A., and Levy, O. (2006) Innate immunity of the human newborn
is polarized toward a high ratio of IL-6/TNF-
␣
production in vitro and in
vivo.Pediatr. Res. 60, 205–209
62. Kollmann, T. R., Crabtree, J., Rein-Weston, A., Blimkie, D., Thommai, F.,
Wang, X. Y., Lavoie, P. M., Furlong, J., Fortuno, E. S., 3rd, Hajjar, A. M.,
Hawkins, N. R., Self, S. G., and Wilson, C. B. (2009) Neonatal innate TLR-
mediated responses are distinct from those of adults. J. Immunol. 183,
7150–7160
63. Hirschhorn, R., and Beratis, N. G. (1973) Letter. Severe combined immu-
nodeficiency and adenosine-deaminase deficiency. Lancet 2, 1217–1218
64. Remington, J. S., Klein, J. O., Wilson, C. B., Nizet, V., and Maldonado, Y.
(2011) Infectious Diseases of the Fetus and Newborn Infant, pp. 80–191,
7th Ed., Elsevier, Philadelphia
65. Sussman, H. H., Bowman, M., and Lewis, J. L., Jr. (1968) Placental alkaline
phosphatase in maternal serum during normal and abnormal pregnancy.
Nature 218, 359–360
66. Yoneyama, Y., Suzuki, S., Sawa, R., Otsubo, Y., Power, G. G., and Araki, T.
(2000) Plasma adenosine levels increase in women with normal pregnan-
cies. Am. J. Obstet. Gynecol. 182, 1200–1203
67. Yoneyama, Y., Suzuki, S., Sawa, R., Otsubo, Y., Miura, A., Kuwabara, Y.,
Ishino, H., Kiyokawa, Y., Doi, D., Yoneyama, K., and Araki, T. (2003)
Serum adenosine deaminase activity and its isoenzyme pattern in women
with normal pregnancies. Arch. Gynecol. Obstet. 267, 205–207
68. Jamieson, D. J., Theiler, R. N., and Rasmussen, S. A. (2006) Emerging
infections and pregnancy. Emerg. Infect. Dis. 12, 1638–1643
69. Weetman, A. P. (1999) The immunology of pregnancy. Thyroid 9,
643–646
70. Gao, Z. G., Ye, K., Go¨blyo¨ s, A., Ijzerman, A. P., and Jacobson, K. A. (2008)
Flexible modulation of agonist efficacy at the human A3 adenosine recep-
tor by the imidazoquinoline allosteric enhancer LUF6000. BMC Pharma-
col. 8, 20
71. Scho¨n, M. P., and Scho¨ n, M. (2007) Imiquimod. Mode of action. Br. J.
Dermatol. 157, 8–13
72. Scho¨n, M. P., Scho¨ n, M., and Klotz, K. N. (2006) The small antitumoral
immune response modifier imiquimod interacts with adenosine receptor
signaling in a TLR7- and TLR8-independent fashion. J. Invest. Dermatol.
126, 1338–1347
73. Philbin, V. J., Dowling, D. J., Gallington, L. C., Corte´s, G., Tan, Z., Suter,
E. E., Chi, K. W., Shuckett, A., Stoler-Barak, L., Tomai, M., Miller, R. L.,
Mansfield, K., and Levy, O. (2012) Imidazoquinoline Toll-like receptor 8
agonists activate human newborn monocytes and dendritic cells through
adenosine-refractory and caspase-1-dependent pathways. J. Allergy Clin.
Immunol. 130, 195–204.e9
74. Wynn, J. L., Scumpia, P. O., Winfield, R. D., Delano, M. J., Kelly-Scumpia,
K., Barker, T., Ungaro, R., Levy, O., and Moldawer, L. L. (2008) Defective
innate immunity predisposes murine neonates to poor sepsis outcome but
is reversed by TLR agonists. Blood 112, 1750–1758
75. Power Coombs, M. R., Belderbos, M. E., Gallington, L. C., Bont, L., and
Levy, O. (2011) Adenosine modulates Toll-like receptor function: basic
mechanisms and translational opportunities. Expert Rev. Anti Infect. Ther.
9, 261–269
Anti-inflammatory Purine Metabolism Profile in Newborn Blood
27326 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 288•NUMBER 38 •SEPTEMBER 20, 2013