Blood, Vol. 95 No. 5 (March 1), 2000:
pp. 1880-1881
CORRESPONDENCE
 |
Letter |
To the editor:
Protein C inhibitor in platelets?
In a recent article, Prendes et al1 described the
presence and localization of protein C inhibitor (PCI) in blood
platelets. The concentration of PCI antigen that was determined in
platelet lysates was 160 ng/2 × 109 platelets. Although
this figure was determined with an extremely accurate ELISA, it was 28 times the previously reported concentration of PCI in
platelets.2 Also, in contrast to what the authors suggest,
blood contains approximately 2 × 108 platelets/mL, which
implies that the PCI concentration that was determined in the platelet
lysate could be due to a contamination of the sample with 0.3% of
plasma. Without a proper check for contamination of the platelet lysate
with plasma proteins, it is nearly impossible to accurately determine
the exact platelet concentration of PCI. Nevertheless, the calculations
above already indicate that platelets can hardly influence the
concentration of PCI in plasma (5.8 µg/mL).
A second point of concern is the determination of "PCI activity
antigen."1 The authors postulate that platelets contain 14% active PCI antigen relative to plasma. Assuming 100% reactivity of PCI in plasma, the active PCI antigen concentration in 1 mL of blood
(equivalent to 2 × 108 platelets) is 810 ng/mL, more
than 50 times the PCI antigen concentration in platelets (16 ng/2 × 108 platelets). This could be explained if the
"PCI activity antigen" assay was not specific for PCI. If true,
this conclusion would undermine the validity of other conclusions in
the paper, because the same antibody was also used for
immunofluorescence and immunoelectronmicroscopy data.
Finally, the use of PCR to determine mRNA of a protein in platelets
should be used cautiously if no control with leukocytes is included.
Platelet preparations are nearly always contaminated with leukocytes,
and because these contain at least 1000-fold more RNA than platelets,
signals are very likely to be from leukocytes instead of platelets.
The physiological role of protein C inhibitor is not yet known.
Characterization of knockout mice will be invaluable in determining the relevance of PCI for hemostasis and other systems. However, the
role of platelet PCI in hemostasis, if any, is likely to be marginal
because of the extremely low concentration in platelets.
Joost C. M. Meijers
Department of Vascular Medicine, Academic Medical
Center, Amsterdam, The Netherlands
| |
References |
1.
Prendes MJ, Bielek E, Zechmeister-Machhart M, et al.
Synthesis and ultrastructural localization of protein C inhibitor in human platelets and megakaryocytes.
Blood.
1999;94:1300-1312[Abstract/Free Full Text].
2.
Nishioka J, Ning M, Hayashi T, Suzuki K.
Protein C inhibitor secreted from activated platelets efficiently inhibits activated protein C on phosphatidylethanolamine of platelet membrane and microvesicles.
J Biol Chem.
1998;273:11,281-11,287[Abstract/Free Full Text].
Response:
Protein C inhibitor does exist in
platelets
In our study,1 which was criticized by Dr
Meijers, we wanted to determine whether or not platelets contain
protein C inhibitor (PCI). It was not the aim to analyze to what extent
platelet PCI might influence the overall PCI activity in blood. Using
different approaches we can show that PCI is present in platelets. We
can furthermore show that PCI contained in platelets originates from synthesis in megakaryocytes as well as from uptake by platelets. Therefore the comment that "it is nearly impossible to accurately determine the exact platelet concentration of PCI" is not valid, because we did not only determine PCI in platelet lysates, but also
used different methods to show PCI synthesis and uptake. We agree with
Dr Meijers that even at the 28-fold higher concentration (as compared
to the values published by Nishioka et al2) platelet PCI
can hardly influence the concentration of PCI in plasma, but that was
never postulated. This assumption is also supported by the observation
that the concentrations of PCI in serum are not higher than in plasma
(what would be expected if significant amounts of PCI were released
from activated platelets). However, at sites of platelet accumulation,
and especially if PCI is bound to stimulating substances (e.g.
proteoglycans, phospholipids), platelets might contribute to an
increase in the local PCI activity.
The remark concerning the concentration of active PCI in platelet
lysates ("second point of concern") is incorrect. We assume that
Dr Meijers might have misread the respective paragraph in the
"Results" section of our article. We determined active PCI antigen in platelet lysates, which were derived from a platelet suspension containing an approximately 10-fold higher platelet concentration (2 × 109 platelets/mL) than blood. These
platelet lysates (which had a PCI antigen concentration of 160 ng/mL)
contained 14% (=812 ng/mL) of the active PCI antigen present in plasma
(5.8 µg/mL). Therefore, the specific activity of platelet PCI is not
50-fold, but only 5-fold higher than the specific activity of plasma
PCI. This is stated in our article in "Results" as well as in the
"Discussion" sections. We have furthermore discussed some
possible explanations for the higher specific activity of platelet PCI
as compared to plasma PCI, such as the possibility that platelet PCI
might be bound to substances stimulating its activity (eg,
proteoglycans). This is not unlikely, since we have shown previously
that cellular proteoglycans can stimulate the interaction of PCI with
urokinase.3 On the other hand, the active PCI antigen assay
used in the present report is also based on the interaction of PCI with
urokinase. However, we apologize if our data were not presented clearly
enough. We hope that it is now clear that our antibody detects PCI and not other antigens, and we therefore cannot understand the last sentence of the second paragraph of Dr Meijers' letter, on the use of
this antibody, which was published several years
ago.4
As far as reverse transcription polymerase chain reaction (RT-PCR) is
concerned, we are of course aware that the problem of contamination of
platelet preparations with leukocytes can never be completely ruled
out. However, this is stated in our article and we have therefore also
confirmed PCI-synthesis in megakaryocytes by in situ hybridization of
bone marrow sections.1
We completely agree that, in fact, the physiological role of PCI is not
yet known. We therefore also hope that transgenic mice will help
clarifying the role of PCI in hemostasis and in other systems. In fact,
PCI (
/) mice recently generated by Dr P. Uhrin from our
department don't show any obvious disturbances of the hemostatic
system (unpublished observations).
Maria J. Prendes
Bernd R. Binder
Margarethe Geiger
Department of Vascular Biology and Thrombosis
Research, University of Vienna, Vienna, Austria
| |
References |
1.
Prendes MJ, Bielek E, Zechmeister-Machhart M, et al.
Synthesis and ultrastructural localization of protein C inhibitor in human platelets and megakaryocytes.
Blood.
1999;94:1300-1312.
2.
Nishioka J, Ning M, Hayashi T, Suzuki K.
Protein C inhibitor secreted from activated platelets efficiently inhibits activated protein C on phosphatidylethanolamine of platelet membrane and microvesicles.
J Biol Chem.
1998;273:11,281-11,287.
3.
Geiger M, Priglinger U, Griffin JH, Binder BR.
Urinary protein C inhibitor. Glycosaminoglycans synthesized by the epithelial kidney cell line TCL-598 enhance its interaction with urokinase.
J Biol Chem.
1991;266:11851-11857[Abstract/Free Full Text].
4.
Ecke S, Geiger M, Resch I, et al.
Inhibition of tissue kallikrein by protein C inhibitor. Evidence for identity of protein C inhibitor with the kallikrein binding protein.
J Biol Chem.
1992;267:7048-7052[Abstract/Free Full Text].
