From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Demonstration of Antithrombotic Activity of Glomerular
Adenenos h e Dip hos p ha t ase
By Klaas Poelstra, Juul F.W. Bailer, Machiel J. Hardonk, and Winston W. Bakker
We have demonstrated that reduced glomerular adenosine
diphosphatase (ADPase) activity within the rat kidney is
associated with an increased thrombotic tendency. To establish a possible causal relationship between these intraglomerular events, experiments were conducted to inhibit adenosine diphosphate (ADP) degradation without influencing
other glomerular prothrombotic or antithrombotic mechanisms. Concurrently, we studied intraglomerular platelet
aggregation. Two ways of selective inhibition of glomerular
ADPase activity were applied: (1) by competitive substrates
(ie, uridine diphosphate [UDP]), and (2) by the nondegradable
ADP analogue ADP-p-S. Both strategies were used during ex
vivo alternate perfusion of kidneys with platelets and ADP (to
test intraglomerular thrombotic tendency). Each group (n = 6)
received different substrates or a combination of substrates.
A significant increase in platelet aggregation was observed in
kidneys after perfusion with platelets and ADP together with
the competitive substrate UDP as compared t o perfusions
with platelets and ADP alone (78.5% 2 9.8% ~ 2 7 . 9 %2 11.4%
glomeruli staining positive for platelets, P < .005). In contrast, UDP alone had no effect on platelet aggregation. Other
nucleoside polyphosphates (guanosine diphosphate and inosine triphosphate) were also effective as competitive substrates in the ex vivo perfusion model (n = 4). None of these
substrates was capable of increasing ADP-induced aggregation when studied in vitro. In addition, ADP- p-S also increased platelet aggregation in the perfusion model as
compared with native ADP (P < .005). These results show
that selective reduction of ADP degradation in intact kidneys
strongly promotes the intraglomerular proaggregatory condition. It can be concluded that glomerular ADPase exerts
potent antithrombotic activity within the normal rat kidney.
o 1991by The American Society of Hematology.
B
nate perfusion system, would be expected to inhibit ADP
degradation in intact kidneys due to the fact that excess of
UDP rather than ADP is converted by glomerular phosphatases.
A second approach to inhibit ADP degradation selectively deals with the application of an ADP analogue
(ADP-P-S) in the alternate kidney perfusion system. ADPp-S is able to activate platelets but inhibits phosphatase
activity by steric inhibition of the active site.” In this case
also enhanced thrombus formation can be expected in
intact rat kidneys following alternate perfusion with platelets and ADP-P-S because, in contrast to perfusions with
native ADP, the aggregation stimulus cannot be removed
by the glomerular enzyme.
The results confirm the presence of nucleoside phosphatase activity, ie, ADPase and UDPase, cytochemically
as well as biochemically within rat glomeruli and demonstrate the inhibitory effect of ADP-P-S on both ADPase
and UDPase activity. Furthermore, data obtained by perfusion studies prove the antiaggregatory action of glomerular
ADPase and suggest, in view of the importance of ADP as a
~”
role for
mediator of platelet aggregation in V ~ V O , ~ , ’a~ major
this enzyme in preventing platelet aggregation and subsequent thrombus formation in the rat kidney in vivo.
LOOD VESSEL walls possess the capacity to interfere
directly with platelet aggregation, in addition to the
anticoagulatory and anti-inflammatory mechanisms.’ Endothelial cells produce prostacyclin (PGI,) with potent antithrombotic a~tivity?~
whereas the amplification signal
adenosine diphosphate (ADP)4 is removed by membraneassociated ADPase activity.’.’ Although in vitro studies
have demonstrated the potent antiaggregatory activity of
ADPase?.’ this antithrombotic mechanism has received
relatively little attention. Recently, however, ADPase activity has been demonstrated in the glomerular basement
membrane (GBM) of the rat kidney: and it was suggested
that this enzyme subserves an important antithrombotic
Thus, in rat kidneys with reduced ADPase activity
(elicited within 24 hours by local radiation or adriamycin
injection), platelet aggregation could be easily induced with
alternate ex vivo perfusion of platelets and ADP.9
However, in the models described, it is difficult to prove a
causal relationship between reduced ADPase activity on
the one hand and increased thrombotic tendency on the
other hand, as other prothrombotic or antithrombotic
systems may also be affected by X-irradiation or adriamycin
(ADR) treatment. It has been shown for instance that
ADR-induced nephropathy is also associated with an increased intraglomerular thromboxane A, (TXA,) production.” Therefore, we now design experiments that enable us
to study the antithrombotic role of glomerular ADPase,
without influencing other vessel wall-associated hemostatic
mechanisms, using two strategies. The first strategy to
inhibit glomerular ADP degradation selectively was based
on the observation that the enzyme exhibits nucleoside
polyphosphatase activity; in addition to ADP, adenosine
triphosphate (ATP), inosine triphosphate (ITP), or uridine
diphosphate (UDP) can also serve as substrate for the
enzyme.’ This observation opened the possibility of applying these as competitive substrates for ADPase activity in
an ex vivo kidney perfusion system. An excess of UDP, for
instance, supplemented to the ADP solution in the alterBlood, Vol78, No 1 (July I ) , 1991: pp 141-148
From the Department of Pathology, University of Groningen, The
Netherlands.
Submitted July 27,1990; accepted February 19, 1991.
Supported by grants of the Dutch Kidney Foundation (C86-062).
Address reprint requests to K Poelstra, PhD, Department of
Pathology, University of Groningen, Oostersingel63, 9713 E Z Groningen, The Netherlands.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section I734 solely to
indicate this fact.
0 1991 by The American Society of Hematology.
0006-497119117801-0017$3.00l0
141
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
POELSTRA ET AL
142
MATERIALS AND METHODS
Animals
Female PVGic rats (200 to 220 g), 3 months old, were used
throughout the study.
5’-0-(2-thiodiphosphate)(ADP-P-S; Boehringer Mannheim
GmbH, Mannheim, Germany) or a combination of substrates.”
Inorganic phosphate was subsequently measured according to the
method of Chandrarajan and Klein.” Phosphate release was
related to the protein content of the suspension, measured according to the method of Lowry et
Preparationof Platelet Suspensions
Platelet-rich plasma (PRP) was prepared from citrated blood
drawn from healthy volunteers as described previously.’ Platelets
were kept in polyolefin bags (PL732; Baxter, Maarssen, The
Netherlands) for a maximum of 5 days at 22°C and in a shaker to
prevent aggregation. Before perfusion, PRP was centrifuged at
1,000g (15 minutes) and volumes were adjusted to a platelet
number of 2.7 x lo9 cells/mL. Platelet aggregability was routinely
tested using an aggregometer. Batches not showing a standard
response on ristocetin were discarded. Each batch of PRP was used
for four exvivo perfusion studies.
Aggregometer Tests
Substrates were tested in vitro for their potential to induce
platelet aggregation using an aggregometer (Biodata Corp, Hatboro, PA; model PAP-4) according to standard procedure^.'^
Platelet suspensions were prepared by centrifugation (I@ 10
minutes) of fresh citrated blood drawn from healthy volunteers.
ADP and ADP-p-S were used in concentration of 5 FmoliL,
whereas UDP, guanosine diphosphate (GDP), and ITP were
added in a final concentration of 50 or 500 pmol/L as indicated in
the Results section.
Ex Evo Perfusion
Experimental Design
Left kidneys of PVGic rats were perfused at 37°C according to
the method of Hoyer et all4 under halothane, O,iN,O, anesthesia.
After removal of the blood with saline, kidneys were perfused with
4 mL endotoxin (1 pgimL), 1 mL substrate in saline (see “Experimental Design”), 4 mL human PRP (2.7 x lo9cellsiml), and again
with 1 mL substrate and 4 mL PRP using a peristaltic pump (flow
rate 2 mL/min). Total perfusion time was 8 minutes and after
perfusion kidney specimens were processed for histochemistry as
described below.
There were three different sets of experiments. Experiment A
was performed to demonstrate glomerular phosphatase activity
cytochemically as well as biochemically and to demonstrate the
applicability of competitive substrates and ADP-f3-S as inhibitors
of glomerular ADPase activity. Experiments B and C were conducted to study intraglomerular platelet aggregation ex vivo during
selective inhibition of glomerular ADP degradation by competitive
substrates and ADP-p-S, respectively.
Experiment A. Glomeruli isolated from intact kidneys were
incubated with the substrates ADP, UDP, ADP-P-S, or a combination of substrates. During the latter incubations ADP-p-S functioned as a steric inhibitor for the active site of glomerular ADPase
and UDPase. For each isolation procedure (n = 5) six kidneys
were used, harvested from three rats. Additionally, three rats were
used for cytochemical demonstration of phosphatase activity.
Experiment B. The effect of ADPase inhibition by competitive
substrates on intraglomerular thrombotic tendency was assessed in
two groups of rats; a group with normal and a group with reduced
glomerular ADPase activity induced by a single adriamycin injection (Adriblastina, Farmitalia, Belgium; 7.5 mgkg body weight
intravenously [IV]), 48 hours before the perfusion studies.25
Thrombotic tendency was assessed by alternate kidney perfusion
ex vivo with platelets and substrates.’ Each platelet batch was used
for four perfusion studies using different substrates, excluding
variations in platelet responsiveness between experimental groups.
Kidneys were perfused with either 10 pg/mL ADP (group I; n = 6),
10 pgimL ADP plus 100 pgimL UDP (group 11; n = 6), or 100
pg/mL UDP (group 111; n = 6) as substrates or no substrate at all
(group IV; n = 6). In an additional set of experiments (n = 4), the
effect of other nucleoside diphosphates and triphosphates (GDP
and ITP) was tested in the ex vivo perfusion system.
Moreover, direct effects of UDP, GDP, and ITP on platelet
aggregation were also studied using an aggregometer (n = 5).
Experiment C. The effect of the ADP analogue ADP-p-S on
platelet aggregation was studied in vitro using an aggregometer,
whereas the effect of ADP-p-S on intraglomerular thrombotic
tendency was studied in intact ludneys during alternate perfusion
studies. Rats (n = 6) received standard alternate perfusion with
either 10 pg/mL ADP-p-S or 10 pg/mL native ADP (n = 6).
Histochemistryand Cytochemistry
Tissue processing. Immediately after ex vivo perfusion with
platelets and ADP, kidney specimens were fixed in 1%paraformaldehyde and embedded in plastic according to standard procedures.” For cytochemical demonstration of ADPase and UDPase
activity, kidneys were prepared according to standard procedures
described elsewhere.I6
Enzyme cytochemistry. Phosphatase activity was demonstrated
at the electron-microscopical level by staining of 30-km vibratome
sections using the cerium-based method.” Sections were subsequently postfixed with OsO, and embedded in Epon 812 (Serva,
Heidelberg, Germany) according to standard procedures.’6
Platelet aggregation. Platelet aggregation was detected in 2-bm
plastic sections by demonstrating fibrinogen binding to human
platelets perfused through the rat kidney. On activation platelets
expose membrane receptors for fibrinogen,Ixleading to the binding
of fibrinogen present in PRP.I9 Subsequent conversion of fibrinogen is inhibited by citrate. Fibrinogen is demonstrated using
indirect immunoperoxidase methods with rabbit-antihuman fibrinogen (Behringwerke AG, Marburg, Germany) according to standard methods.15
Biochemistiy
Glomerular nucleoside polyphosphatase activity was assayed in
suspensions of glomeruli of normal rats. Kidneys were perfused
with saline for 1 minute using a peristaltic pump (140 mm Hg
pressure) and immediately snap frozen in freon ( - S O T ) for
storage. Glomeruli were subsequently isolated from the cortex
using standard sieving techniques2’with sieves of 180, 150, and 200
mesh, respectively. The whole isolation procedure was performed
in 0.15 mol/L NaCl and at 4°C to preserve viability of glomeruli.
Suspensions purified up to 90% as tested by lightmicroscopy were
used for determination of enzyme activity. Phosphatase activity was
assessed by measuring the amount of phosphate released by
glomeruli after incubation with 2.5 mmoliL ADP, UDP, adenosine-
QuantitativeAnalysis
Results of ex vivo perfusion studies were evaluated by counting
the percentage of glomeruli per kidney, positive for platelet
aggregates, which are readily detectable in plastic embedded rat
tissue after staining for human fibrinogen. We scored the total
number of glomeruli in each section containing aggregates, ie,
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANTITHROMBOTIC ACTIVITY OF GLOMERULAR ADPASE
143
more than one fibrinogen positive platelet clump, irrespective of
their size, rather than to apply an arbitrary scale. This scoring
resulted in a highly reproducable scoring procedure, although it
overestimates the overall aggregation in kidneys with small aggregates. From each kidney, 75 glomeruli from three kidney sections
were evaluated and the results expressed as the arithmetic means
(+.SD)of six kidneys per group. Data were analyzed by Wilcoxon’s
test and considered significant at P < .05.
RESULTS
Experiment A
Photomicrographs of cytochemical staining for glomerular nucleoside polyphosphatase activity using the ceriumbased method are shown in Fig 1. After incubation with
ADP, reaction product can be demonstrated in normal rat
kidneys along epithelial and endothelial plasma membranes as well as throughout the GBM. Incubation of
kidney sections with UDP gives rise to the same localization
of reaction product as compared to incubations with ADP
(Fig 1B). In contrast, when ADP-p-S is used as the
substrate, no reaction product at all can be demonstrated in
kidney sections of normal rats (Fig 1C).
Figure 2 shows the results of biochemical studies of
glomerular phosphatase activity. The results demonstrate a
glomerular phosphate release of 2.63 & 0.22 mmol/L of
inorganic phosphate (PI) per milligram of protein during
incubation with ADP, which is similar to glomerular phosphate release after incubation with UDP (2.64 e 0.53
mmol/L Pi/mg protein). In contrast, after incubations with
ADP-(3-S as substrate no inorganic phosphate following a
30-minute incubation period can be detected (0.00 e 0.00
mmol/L P,/mg protein). Moreover, after incubation of
glomeruli with ADP-p-S in combination with ADP,. a
significant reduction in the amount of free phosphate is
demonstrated as compared to incubations with ADP
(0.04 2 0.04 mmol/L Pi/mg protein; P < .005, Wilcoxon).
This effect of ADP-p-S is also observed when UDP is
applied as substrate (0.06 e 0.02 mmol/L Pi/mg protein;
P < .005, Wilcoxon).
,. ”
tP
Experiment B
Intraglomerular platelet aggregates after alternate perfusion experiments are demonstrated by staining for human
fibrinogen (Fig 3). As can be seen in Fig 3B, perfusions with
the combination of ADP with an excess of UDP lead to a
strong increase in intraglomerular staining as compared to
perfusions with ADP alone (Fig 3A). If present, intraglomerular platelet aggregates were small in the glomeruli of
ADP-perfused kidneys, in contrast to the large aggregates
found in kidneys perfused with the combination of ADP
and UDP. Intraglomerular platelet aggregates in UDP and
saline perfused kidneys (groups I11 and IV, respectively)
were generally small, as in ADP-perfused kidneys.
Quantitative evaluation of intraglomerular platelet aggregation with UDP as the competitive substrate is demonstrated in Fig 4. Platelet aggregation was analyzed by
scoring the percentage of glomeruli per kidney section
positive for human fibrinogen (see Materials and Methods).
As can be seen in Fig- 4, supplementation
of excess UDP to
..
the ADP solution significantly increases intraglomerular
0
C
Fig 1. Localization of enzyme activity can be seen along epithelial
and endothelial cells as well as throughout the glomerular basement
membrane after incubation with ADP (A) or UDP (6).whereas reaction
product is completely absent after incubation with ADP-P-S (C). C,
capillary lumen; U, urinary space; Ep, epithelial cell. Original magnification x 42.500.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
POELSTRA ET AL
144
an aggregometer. As can be seen, none of these substrates
per se induces platelet aggregation in vitro. The aggregation response on ADP in combination with UDP or GDP in
the ratio 1:lO is similar to the aggregation response on ADP
alone, only in the case of ITP a reduction is observed.
However, when ADP in combination with UDP, GDP, or
ITP is used in the ratio 1:lOO a significant reduction in the
maximal aggregation response is observed in vitro in all
cases.
3.4
T
e
2.8
-
2.2
-
-
n 1.8 >
E
a
-
1.4
I
E
-
-
10.8
-
0.2
-
-
I1
ADP
UDP
Experiment C
ADP-Bs
ADPlADP-Bs
UDP/ADP-Bs
substrate
Fig 2. Phosphatase activity in suspensions of glomeruli obtained
from normal rats. Results represent the mean f SD of five experiments. ADP-pS is not hydrolyzedand significantly reduces both ADP
and UDP degradation (substrate concentrations 2.5 mmol/L). ' P <
.005 as compared with ADP, Wilcoxon.
platelet aggregation in intact kidneys as compared to
perfusions with ADP alone (78.5% 2 9.8% positive glomeruli v 27.9% 2 11.4%, respectively; P < .005, Wilcoxon),
whereas the substrate UDP alone leads to a similar percentage of fibrinogen-positive glomeruli as in saline-perfused
kidneys (25.1% 2 5.3% v 21.7% t 5.1%, respectively). In
contrast, thrombotic tendency in kidneys of ADR-treated
rats is significantly increased when ADP is used as the
substrate as compared with kidneys of untreated rats
(69.8% 2 9.1%; P < .005, Wilcoxon). UDP supplementation to the perfusate also increases intraglomerular platelet
aggregation in ADR-treated rats (87.9% 2 7.8%). However, the increased aggregation in these kidneys is not
observed when UDP alone is used as the substrate: 16.0%
2 9.6%, which is comparable with the percentage of
fibrinogen positive glomeruli in saline-perfused kidneys of
ADR-treated rats (25.6% 2 2.9%).
To discriminate between effects of UDP as a nucleoside
polyphosphatase substrate on the one hand and on platelets
on the other hand, aggregation responses were also studied
in vitro using an aggregometer. Typical aggregation responses on the stimuli are depicted in Fig 5. It is shown that
UDP does not induce platelet aggregation nor does it
interfere with ADP-induced platelet aggregation when the
UDP concentration is 10 times the ADP concentration.
Additionally, other nucleoside diphosphates and triphosphates, ie, GDP and ITP, were also tested in the ex vivo
kidney perfusion model (n = 4). Perfusion of these substrates in combination with ADP also induced an increase
in intraglomerular platelet aggregation as compared to
perfusion experiments with ADP alone (45.0% 2 13.4%
and 56.5% 2 18.5% fibrinogen-positive glomeruli for GDP
and ITP respectively, v 26.4% 2 9.91% positive glomeruli
after ADP perfusion (P < .025).
Results of experiments with nucleoside diphosphates and
triphosphates in vitro are summarized in Table 1. In this
table the maximal aggregation response of platelets on
ADP is comDared to aggregation responses on ADP in
combination with either UDP, GDP, or ITP in vitro using
I
I
I
Cytochemical and biochemical studies using the ADP
analogue ADP-p-S as the substrate for glomerular phosphatase activity demonstrate that this substrate is not
hydrolyzed by the glomerular enzyme (Fig 1) and inhibits
glomerular ADPase activity (Fig 2). However, when this
Fig 3. Rat kidney stained for human fibrinogen after alternate
perfusion ex vivo with human PRP and ADP (10 pg/mL; [A]) or ADP
UDP (10 respectively 100 d m L ; [B]). Increasedstainingfor fibrinogenpositive cells after addition of UDP. Original magnificaion x 350.
+
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANTITHROMBOTIC ACTIVITY OF GLOMERULAR ADPASE
145
Table 1. Aggregation Induced by Nucleotide Polyphosphates In Vltro
Substrate
ADP
UDP
ADPIUDP
ADPIUDP
GDP
ADPIGDP
ADPIGDP
ITP
ADPIITP
ADPIITP
ADP ADPNDP UDP
saline
ADP ADPAJDP UDP
saline
Fig 4. lntraglomerular platelet aggregation as reflected by staining
for fibrinogen after kidney perfusion according t o the standard
protocol (see Experimental Design). Columns represent arithmetic
means (2SD) of percentages of positive stained glomeruli of six rats
per group (dark set of columns: untreated rats; light columns:
ADR-treated rats). In untreated rats a significant increase is observed
after ADP/UDP perfusion as compared with ADP, UDP, or saline alone
(*P< .005). In kidneys of ADR-treated rats, increased fibrinogen
staining is observed after ADP perfusions as well as after ADP/UDP
perfusions as compared t o perfusions with UDP or saline (**P< .05,
Wilcoxon).
ADP-p-S is applied to stimulate platelet aggregation in
vitro, a significant aggregation response is observed (Fig 6).
In addition, platelet aggregation ex vivo is significantly
increased in normal kidneys perfused with ADP-P-S as
compared to aggregation in kidneys perfused with native
ADP. As can be seen in Fig 7, the average percentage of
fibrinogen-positive glomeruli after ADP-p-S perfusion is
61.6% ? 13.8%, whereas after perfusion with native ADP
this percentage is 27.2% ? 9.7% (P < .005).
.
..
..
.. ..
..
..
..
..
.
... ... ... ... ... ... ... ... ... . ..
I
.d . 3 . ~ . 3 . - . 1
..
.
.
. .. . . ... . ... . . .. . . .. . . ... . . .. . . .. . . ..
... ... ... ... ... ... ... ... ... ...
.. .. .. .. .. .. .. .. .. ..
.
.
.
.
.
.
~
.
1
.
~
.
1 IJDP
U
Ratio
(ADP):(C.S )
% Aggregation'
71.8 f 3.43
0.3 f 0.05t
65.7 2 4.72
30.1 f 11.2t
0.2 f 0.03t
70.3 f 4.51
27.0 f 6.45t
0.2 2 0.05t
59.8 f 7.32$
19.8 2 5.15t
1:lO
1:lOO
1:lO
1:lOO
1:lO
1:lOO
*Arithmetic means (n = 5) of the maximal aggregation response
(rSD) observed in citrated human PRP during continuous recording
sessions using an aggregometer. ADP is used as an agonist in a
concentration of 5 pmolIL, and the final concentration of competitive
substrates (C.S.) when used in combination with ADP is 50 or 500
pmollL (as indicated) and 50 pmol1L when added alone.
tP < .005(Wilcoxon) as compared with ADP-induced aggregation.
SP < .01 (Wilcoxon) as compared with ADP-induced aggregation.
DISCUSSION
In this study, experiments were performed to elucidate
the role of glomerular ADPase activity as a platelet inhibitory mechanism without influencing other antithrombotic
mechanisms or prothrombotic modalities. Therefore, glomerular ADPase activity was selectively inhibited by using
competitive substrates on the one hand and an ADP
analogue on the other hand in an ex vivo kidney perfusion
model.
Biochemical and cytochemical data confirm the presence
of both ADPase and UDPase activity in the glomeruli of
rats. As reported previously,8 both substrates give rise to a
similar localization of reaction product throughout the
filtration barrier at the electron-microscopical level (Fig 1).
Biochemical studies with isolated glomeruli are in agree.. . . .. . . .. . . .. . . .. . . .. . . .. . . ... . ... . ... . ... . . . . ... . ... . . .
. . . . . . . . . . . . . . .
.. .. ..
..
: ah*
ty,.
'
..
'
' '
.
'
.. .
: . .?OX. ; . .
.. .. ..
.
.: . .jr,.j.
.
+?*.:..
.. .. ..
.:"w*':'.
'
. . .
i
:
:
:
I**.
ADPos
. . .
:.
.si.:.
.
. .. ._
t
I
.
.
. . . . . .
.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
: . . g p z . j . . ;. . j . . j . . j . . j , . j . . j . . j . . j
..j..
.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
. . I @ ? . . . . . .. . . ... . ... . . .. . . .. . . .. . . ... . ... . . .. . . .. . . .. . . ... . .
.. .. .. .. .. .. .. .. .. .. . . . . . . . . . .
. . . . . . ... . . .. . . .. . . ... . ... . . .. . . .. . . .
. . .. . .. . . . . . .. .
.. ' *60%
.. .j.. , .. .. .. .. .. .. .. .. .. .. .. .. ..
. . . . . . .
. . . .
:,, .**
'
/!I
:&
.. ........................................
'
--- 1 IIINUTE ---
..............................
. . .. .. .. .. .. .. .. ..
~
ADP
.. .. .. ... ... ... ... ... ... ... .. .. .. .. . .
:..cor.:..:..:..:..:..:..:..:..
.. .. . . . . . . . . . .. .. .. ..
. . . . . . . . . .
: . .jp,. j . . ; . . j . . ; . . j . . j . . j . .':. .
.. .. . .. .. .. .. .. .. .. ..
.. .. .. .. .. .. .. .. .. ..
Fig 5. Representative aggregation response of human PRP in vitro
with the stimulus ADP (5 pmol/L), ADP in combination with UDP (5
respectively 50 pmol/L), or UDP alone (50 pmol/L). Changes in
aggregation were continuously recorded after the addition of agonist
using an aggregometer. It can be seen that UDP does not induce
platelet aggregation nor does UDP interfere with ADP-induced aggregatlon.
..............................
--- 1 IiINUlE ---
- t
Fig 6. In vitro aggregation response in human PRP after addition of
ADP (5 pmol/L) or the ADP analogue ADP-p-S (5 pmol/L) as detected
using an aggregometer. It Is shown that both stlmuli induce aggregation.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
146
POELSTRA ET AL
80
.-
i
70
3
-g . 6 0
0
9
50
E
40
C
gJ
30
e
220
.-
e=
10
0
ADP
ADP-IM
Fig 7. Intraglomerular platelet aggregation as reflected by intracapillar fibrinogen staining after alternate perfusion ex vivo with PRP
and 10 pg/mL ADP or 10 pg/mL ADP-p-S. Columns represent arithmetic means (-cSD) of the percentage of positive glomeruli per kidney
section of six rats per group. Kidneys perfused with ADP-p-S show
significantly more aggregation as compared to kidneys perfused with
native ADP ("P< ,005. Wilcoxon).
ment with these results (Fig 2); glomerular UDPase activity
is of the same magnitude as glomerular ADPase activity.
Moreover, ADPase and UDPase activities are strongly
inhibited by ADP-p-S, indicating that ADP-P-S is a steric
inhibitor of the active site of both ADPase and UDPase.
From these results it can be deduced that UDP binds to the
same binding site as ADP and, therefore, UDP can be used
as a competitive substrate to inhibit glomerular ADPase
activity. ATP, a known competitive substrate for ADP
binding sites," cannot be used here because the enzymatic
degradation products (ADP, adenosine monophosphate,
and adenosine) influence platelet aggregation
ADP-P-S was applied in this biochemical experiment as an
inhibitor of the receptor site because this ADP analogue is
nondegradable by glomerular phosphatase activity (Fig 2);
therefore, phosphate release during the incubation period
can exclusivelybe attributed to ADPase or UDPase activity.
From these results it can be concluded that UDP can be
used as a competitive substrate to inhibit glomerular ADP
degradation and, secondly, that ADP-P-S is an inhibitor of
glomerular phosphatase activity.
The first principle of enzyme inhibition was tested using
the alternate kidney perfusion system described previ0us1y.~Using this model, intraglomerular platelet aggregation can be studied without involvement of the coagulation
cascade. Human platelets are used in this bioassay to
evaluate thrombotic tendency because they can be readily
detected in rat tissue using immunologic techniques and
because they can be obtained in sufficient amounts for
perfusion studies. Human platelets, like rat platelets, aggregate in response to ADP.'~
As illustrated in Fig 3, abundant intraglomerular platelet
aggregation occurs in normal kidneys (ie, kidneys with
normal ADPase activity) exclusively when UDP is supplemented in excess to the platelet stimulating agonist (ie,
ADP). Also, quantitative evaluation of the data shows a
significant increase in platelet aggregation in normal kidneys exclusively when UDP is combined with ADP in the
perfusion system (Fig 4),whereas after perfusion with UDP
alone this increase is not observed. Moreover, this proaggregatory effect of UDP in combination with ADP is only
observed ex vivo; in vitro UDP does not induce platelet
aggregation even at high concentrations (50 kmol/L) nor
does it interfere with ADP-induced platelet aggregation
(Fig 5). Therefore, the aggregation-increasing effect of
UDP ex vivo cannot be attributed to direct effects on
platelets. Therefore, it can be concluded that specific
inhibition of ADP degradation occurs by excess of UDP,
and this subsequently leads to a significant increase in
intraglomerular platelet aggregation in intact kidneys.
This principle is also clear in kidneys with reduced
ADPase activity after ADR
(Fig 4, right set of
columns). These rats show a significant increase in platelet
aggregation after perfusion with the substrate ADP as
compared to untreated rats, which is in agreement with
previous studies,' but UDP alone has no effect. This
demonstrates that an ADP-dependent antithrombotic mechanism is impaired in ADR-treated rats and other mechanisms, ie, glomerular TXA, production," are not involved in
the present study 48 hours after the injection. The increased urinary thromboxane excretion of ADR-treated
rats" is observed in the full-blown nephrotic condition.
Additional experiments with other nucleoside polyphosphates, ie, GDP and ITP, demonstrated that the effect of
competitive substrates on ex vivo platelet aggregation was
not restricted to UDP, although GDP and ITP are less
effective as compared with UDP. The reason for this is not
clear, but a low glomerular GDPase activity has also been
observed histochemically (personal observations, October
1988). The reduced effect of ITP ex vivo as compared with
UDP may be related to a competitive effect of ITP on the
platelet ADP receptors as has also been reported for ATP,26
thus impairing ADP-induced aggregation. ITP impairs
ADP-induced aggregation in vitro at lower concentrations
(ratio 1:lO) as compared with UDP and GDP (Table 1).
Furthermore, it can be seen in Table 1 that all competitive substrates applied in this study impair ADP-induced
aggregation at high concentrations in vitro (ratio 1:100),
probably by a competitive effect on the platelet ADP
receptor. Moreover, it is clear that none of the substrates
applied in this study as a competitive substrate is able to
induce aggregation directly, nor do they increase ADPinduced aggregation in vitro as compared with ADP alone.
Therefore, results of ex vivo perfusion studies cannot be
explained by direct effects of these nucleotides on platelets.
The second approach to demonstrate the importance of
intraglomerular ADP-degrading activity in the prevention
of intraglomerular platelet aggregation is the use of a
nondegradable ADP analogue. ADP-p-S is not hydrolyzed
by the glomerular enzyme," as is also demonstrated in Fig
2, but it is still capable of inducing platelet aggregation in
vitro (Fig 6). While native ADP does not induce a significant intraglomerular platelet aggregation in the intact
kidney ex vivo, it can be seen in Fig 7 that the nondegrad-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
147
ANTITHROMBOTIC ACTIVITY OF GLOMERULAR ADPASE
able ADP-f3-S is able to induce a significantly increased
platelet aggregation. These observations confirm that glomerular degradation of ADP is important for the prevention
of intraglomerular platelet aggregation in the present
model.
Taken together, these results demonstrate that selective
reduction of glomerular ADPase activity, either by competitive substrates o r by ADP-P-S, strongly enhances intraglomerular thrombotic tendency, as reflected by increased platelet aggregation in the rat kidney. Apparently glomerular
ADPase exerts potent antithrombotic activity within intact
rat kidneys by inhibiting platelet aggregation and subsequent thrombus formation.
A major advantage of this ex vivo model, as compared
with the in vivo situation, is the possibility of impairing
glomerular ADPase without influencing other antithrombotic mechanisms such as the eicosanoid synthesis. In vivo,
reduction of glomerular ADPase activity, for instance by
ADR injection, local X-irradiati~n,’~
or induction of acute
g l o m e r ~ l o n e p h r i t i scan
, ~ ~ hardly be achieved without influencing these mechanisms or induction of aspecific damage
t o the vascular wall, which may facilitate platelet aggregation. Nevertheless, the question may emerge as to whether
intraglomerular ADPase is also important in vivo as an
antithrombotic mechanism. However, the role of membraneassociated ADPase in vivo is directly related t o the importance of ADP as a stimulator of platelet aggregation in vivo.
Although platelet aggregation in vivo can be initiated by a
variety of stimuli, including TXA,, platelet activating factor,
collagen, and thrombine, ADP plays a major role in
amplifying the hemostatic res~onse.~
Moreover, at least the
rat hemostatic response has been shown t o depend more on
ADP than on TXA2.l3 Also, in experimental glomerulonephritis in vivo, a strong association between intraglomerular
platelet aggregation and decreased ADPase activity has
been shown,,’ supporting the notion that glomerular ADPase is a prominent antiaggregatory principle within glomeruli of the rat kidney in vivo.
In particular, because glomerular fenestrated endothelium enables a close contact between the bloodstream and
the collagen of the GBM, platelet activation may readily
occur after a slight perturbation of the capillary wall.
Therefore, antithrombotic mechanisms within the glomerular capillary wall may b e very important. Moreover, platelet
aggregation can induce damage within g l o m e r ~ l i , ~stimu~~”
late the coagulation cascade: or potentiate the inflammatory response.3o33’
These observations stress the importance
of a powerful antithrombotic mechanism within the intact
glomerular capillary wall. In addition to other vessel wallassociated anticoagulatory mechanisms (ie, the fibrinolytic
system and heparan sulphate p r o t e o g l y c a n e ~ or
~ ) endothe,~
ADPase activity
lial PGI, and NO p r o d ~ c t i o n , ’glomerular
probably fullfills a major antithrombotic function in the rat
kidney.
ACKNOWLEDGMENT
We thank H. Wierenga and H.R.A. Meiborg for performing the
microphotography, and J. Donga for his excellent technical assistance. The collaboration of Dr C.Th. Smit-Sibinga, head of the
bloodbank “Groningen, Drenthe,” is greatly appreciated.
REFERENCES
1. Marcus AJ,Weksler BB, Jaffe EA, Broekman MJ: Synthesis
of prostacyclin from platelet-derived endoperoxides by cultured
human endothelial cells. J Clin Invest 66:979,1980
2. Gaetano de G, BerletC V, Cerletti C Prostaglandins, thromboxanes and platelet function, in Gryglewsky RJ, Stock G (eds):
Prostacyclin and Its Stable Analogue Iloprost. New York, NY,
Springer-Verlag,1987,p 25
3. Radomski MW, Palmer RMJ, Moncada S: The antiaggregating properties of vascular endothelium: Interactions between prostacyclin and nitric oxide. Br J Pharmacol92:639,1987
4. Mustard JF, Packham MA, Kinlough-Rathbone R L Mechanisms in thrombosis, in Bloom AL, Thomas DP (eds): Haemostasis
and Thrombosis. New York, NY, Churchill Livingstone, 1987, p
618
5. Gordon EL, Pearson JD, Slakey L L The hydrolysis of
extracellular adenine nucleotides by cultured endothelial cells
from pig aorta; feed forward inhibition of adenosine production at
the cell surface. J Biol Chem 261:15496, 1986
6. Glasgow JG, Schade R, Pitlick FA: Evidence that the ADP
hydrolysis by human cells is related to thrombogenic potential.
Thromb Res 13:255,1978
7. Lieberman DE, Lewis GP, Peters TJ: A membrane-bound
enzyme in rabbit aorta capable of inhibitingadenosine-diphosphateinduced platelet aggregation. Lancet 2:330, 1977
8. Hardonk MJ, Kalicharan D, Hulstaert CE: Cytochemical
demonstration of ATPase activity in the rat kidney basement
membrane using the cerium-based method. Acta Histochem Suppl
Band XXXI S:253,1985
9. Bakker WW, Willink EJ, Donga J, Hulstaert CE, Hardonk
MJ: Antithrombotic activity of glomerular adenosine diphos-
phatase in the glomerular basement membrane of the rat kidney. J
Lab Clin Med 109:171,1987
10. Remuzzi G, Imberti L, Rossini M, Morelli C, Carminati C,
Cattaneo GM, Bertani T: Increased glomerular thromboxane
synthesis as a possible cause of proteinuria in experimental
nephrosis. J Clin Invest 75:94, 1985
11. Goody RS, Eckstein F, Schirmer RH: The enzymatic synthesis of thiophosphate analogues of nucleotides. Biochim Biophys
Acta 276:155,1972
12. Born GVR, Kratzer MAA: Source of adenine nucleotides
responsible for haemostatic platelet aggregation. J Physiol 324:
52P, 1981 (abstr)
13. Begent NA, Born GVR: Further evidence that haemostasis
depends more on ADP than on thromboxane A, in the rat. J
Physiol341:36P, 1983 (abstr)
14. Hoyer JR, Mauer SM, Michael AF: Unilateral renal disease
in the rat. I. Clinical, morphologic, and glomerular mesengial
functional features of the experimental model produced by renal
perfusion with aminonucleoside.J Lab Clin Med 85:756, 1975
15. Goor H van, Harms G, Gerrits PO, Kroese FGM, Poppema
S, Grond J: Immunohistochemical antigen demonstration in plastic
embedded lymphoid tissue. J Histochem Cytochem 36:115, 1988
16. Hulstaert CE, Kalicharan D, Hardonk MJ: Cytochemical
demonstration of phosphatases in the rat liver by a cerium-based
method in combination with osmium tetroxide and potassium
ferrocyanidepostfixation. Histochemistry 78:71,1983
17. Veenhuis M, van Dijken JP, Harder W: A new method for
the cytochemical demonstration of phosphatase activities in yeasts
based on the use of cerous ions. FEMS Microbiol Lett 9:285, 1980
18. Legrand C, Dudernard V, Nurden AT: Studies on the
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
148
mechanism of expression of secreted fibrinogen on the surface of
activated human platelets. Blood 73:1226, 1989
19. Suzuki H, Kinlough-Rathbone RL, Packham MA, Tanoue
K, Yamazaki H, Mustard JF: Immunocytochemical localization of
fibrinogen during thrombin-induced aggregation of washed human
platelets. Blood 71:1310,1988
20. Blau E, Michael AF: Rat glomerular basement membrane
composition and metabolism in amino nephrosis. J Lab Med 77:97,
1971
21. Lowry OH, Colowick SP, Kaplan NO: Micromethods for the
assay of enzymes, in Colowick SP, Kaplan NO (eds): Methods in
Enzymology, vol IV. San Diego, CA, Academic, 1957, p 366
22. Chandrarajan J, Klein L Determination of inorganic phosphorus in the presence of organic phosphorus and high concentrations of proteins. Anal Biochem 72:407, 1976
23. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ: Protein
measurement with the folin phenol reagent. J Biol Chem 193:265,
1951
24. Born GVR: Strong inhibition by 2-chloroadenosine of the
aggregation of blood platelets by adenosine diphosphate. Nature
202:95, 1964
25. Bakker WW, Kalichran D, Donga J, Hulstaert CE, Hardonk
POELSTRA ET AL
MJ: Decreased ATPase activity in adriamycin nephrosis is independent of proteinuria. Kidney Int 31:704,1987
26. Jefferson JR, Harmon JT, Jamieson G A Identification of
high-affinity (& 0.35 pmol/L) and low-affinity (Kd 7.9 pmol/L)
platelet binding sites for ADP and competition by ADP analogues.
Blood 71:110,1988
27. Poelstra K, Hardonk MJ, Koudstaal J, Bakker WW: Intraglomerular platelet aggregation and experimental glomerulonephritis. Kidney Int 37:1500,1990
28. Cameron JS: Platelets in glomerular disease. Annu Rev Med
35:175, 1984
29. Clark WF, Friesen M, Linton AL, Lindsay RM: The platelet
as a mediator of tissue damage in immune complex glomerulonephritis. Clin Nephrol6:287,1976
30. Nachman RL, Weksler BB: The platelet as an inflammatory
cell, in Glynn LE, Houck JC, Weissmann G (eds): Handbook of
Inflammation, vol2. New York, NY, Elsevier, 1980, p 145
31. Marcus AJ, Broekman MJ, Safier LB, Ullman HL, Islam N,
Serhan CN, Rutherford LE, Korchak HM, Weissmann G: Formation of leucotrienes and other hydroxy acids during plateletneutrophil interactions in vitro. Biochem Biophys Res Commun
109:130,1982
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1991 78: 141-148
Demonstration of antithrombotic activity of glomerular adenosine
diphosphatase [published erratum appears in Blood 1991 Oct
15;78(8):2163]
K Poelstra, JF Baller, MJ Hardonk and WW Bakker
Updated information and services can be found at:
http://www.bloodjournal.org/content/78/1/141.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.