From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
Human Megakaryocytes Express Clusterin and Package It Without
Apolipoprotein A-1 Into a-Granules
By Jurg Tschopp, Dieter E. Jenne, Silvie Hertig. Klaus T. Preissner, Hans Morgenstern, Andr6-Pierre Sapino, and Lars French
Clusterin, a 70-Kd disulfide-linkedtwo-chain plasma glycoprotein circulates in blood as a high-density lipoprotein
particle and is highly induced after tissue injury and tissue
remodeling. In this study, peripheral blood leukocytes
were assayed for clusterinexpression. The proteinwas predominantly detectable in human platelets by immune cytochemistry. The content of clusterin was determined and
amounts to 2.5 f 1 . 3 pg/109 platelets, thus representing
about 2% of the blood pool. Clusterin purified from human
platelets had the same molecular weight as plasma clusterin under nonreducing conditions and was composed of
two disulfide-linked nonidentical subunits of the same
size. Both preparations were sensitive to reduction yielding the two subunits of 35 Kd. In contrast to plasma clusterin, the platelet form was not complexed to apolipoprotein A-I. By immunogold labeling, a-granule localizationof
clusterin was observed. Complete release of platelet clusterin occurred at optimal doses of A231 87, phorbol myristate acetate (PMA), and thrombin. Because clusterin
mRNA was detected by hybridization in situ in bone
marrow-derived megakaryocytes, platelet clusterin is
most likely produced and packaged into a-granules during
megakaryocyte development.
0 1993 by The American Society of Hematology.
H
sponds to N-linked carbohydrates. Ten cysteines give rise to
five disulfide bridges linking the two subunits.''
Like all other components of terminal complement complexes, clusterin circulates as a normal constituent in human blood plasma. Clusterin of plasma is specifically bound
to high density lipoproteins (HDL) and may represent an
important regulatory apolipoprotein (called apo-1) of
H D L s . ~ , ' ~Besides
-'~
liver and testis, the clusterin gene is
expressed in many other tissues and in different cell types,
including epithelial cells, eg, those lining the proximal tubules of the kidney,I5 neuronal cells,16 and in some cells of
the diffuse endocrine system.17
Transcription of the gene is temporarily highly upregulated in regressing, involuting, regenerating, or injured tissues. Clusterin levels are significantly increased in embryologic tissues undergoing apoptosis," in Alzheimer's disease
in the hippocampu~,'~
in epileptic foci,2oand in prostate2'
and mammary glandz2 after hormone withdrawal. Thus,
clusterin appears to be a transcript that is presumably involved in defensive responses to tissue damage and cellular
injury. In view of its high inducibility in damaged tissues,
clusterin has been regarded as a bona fide cell death
marker."
The aim of this study was to determine whether clusterin
is expressed and stored in circulating peripheral blood (PB)
cells. Immunohistochemical studies showed its abundant
presence in platelets, whereas other hematopoietic cell lineages were only slightly positive or negative. We report the
biochemical properties and subcellular location of clusterin
in a-granules and show by the method of in situ RNA:RNA
hybridization that clusterin mRNA is most abundantly expressed in a subpopulation of immature megakaryocytes.
UMAN CLUSTERIN (synonyms: complement lysis
inhibitor, CLI; sulfated glycoprotein 2, SGP-2; sp40,40; testosterone-repressedprostatic messenger 2, TRPM-2;
gp-111; apolipoprotein J, apo-J; gene name: CLI) was recently discovered as a integral component of the soluble
C5b-9 complement
which is assembled in the
fluid phase (plasma) from the complement proteins C5b to
C9 and vitronectin on activation of the complement cascade. Because binding of clusterin to terminal complement
complexes at the stage of nascent C5b-7 complexes abolishes their membranolytic p ~ t e n t i a l , ' ,clusterin
~,~
was also
called CLI.
Cloning and sequencing of cDNAs for human clust e r i n ' ~ ~established
,~
strong sequence homology with rat
SGP-27~8
and sheep c l u ~ t e r i n ,a~major
~ l ~ Sertoli cell-derived
GP, that had been characterized earlier in both rat and
sheep testis fluid. Clusterin was also found in human seminal plasma at high concentrations and for this reason the
term clusterin has been adopted by several investigators for
the human protein."
Clusterin from blood plasma and seminal plasma is a
heterodimer, consisting of two nonidentical subunits,
which, in the human, both display a molecular mass of approximately 35 Kd. Approximately 30% of the mass corre-
From the Institute ofBiochemistry, University of Lausanne, Epalinges. Switzerland; the Clinical Research Unit for Blood Coagulation and Thrombosis of the Max-Planck-Gesellschaft, Bad Nauheim, Germany; the Medical Biologie, Universitat des Saarlandes,
Homberg, Germany; and the Department qf Medicine, H6pital
Cantonal Universitaire, Geneva, Switzerland.
Submitted July 16, 1992; accepted February 12, 1993.
Supported by grants of the Swiss National Science Foundation
and the Sandoz Foundation.
Address reprint requests to Jiirg Tschopp, PhD, Institute of Biochemistry, University of Lattsanne, Ch des Boveresses 155, CH1066 Epalinges, Switzerland.
The publication costs ofthis 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 1993 by The American Society ofHematology.
0006-4971/93/8201-00I I$3.00/0
118
EXPERIMENTAL PROCEDURES
Chemicals. Triton X-100, enzyme inhibitors, and other common chemicals were all purchased from Sigma (Buchs, Switzerland). Phorbol myristate acetate (PMA), HEPES, and PIPES were
obtained from Calbiochem (Laufelfingen, Switzerland).
Platelel preparation. Fresh platelet-rich plasma was obtained
from the local blood center. Platelets were centrifuged at 500g for
20 minutes and resuspended in an equal volume of buffer containing 5 mmol/L PIPES, pH 6.8; 145 mmol/L NaCI; 4 mmol/L KCI;
0.5 mmol/L Na2HP0,; and 1 mmol/L EDTA. The washings were
Blow! VOI 82,NO 1 (July 1). 1993:pp 118-125
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
119
CLUSTERIN EXPRESSION IN MEGAKARYOCYTES
repeated two more times and the final pellet was resuspended in a
small volume of 20 mmol/L HEPES, pH 7.4; I37 mmol/L NaCI; 4
mmol/L KCI; 0.5 mmol/L Na,HPO,; and I mmol/L MgCl, to
yield a final concentration of IO9 platelets/mL. This suspension was
used within the following 2 hours.
Preparation of platelet releusate. Washed platelets ( I X IO9
cells/mL) were supplemented with 20 pmol/L calcium ionophore
A23 187 or 200 nmol/L PMA, and after 10 minutes stirring at 37°C
and the addition of a protease inhibitor cocktail LAP (leupeptin [5
pg/mL]; pepstatin [5 pg/mL]; antipain [5 pg/mL]), the reaction
mixture was centrifuged at 10,OOOg for 15 minutes at 4°C. The
supernatant was withdrawn and kept frozen at -70°C.
Purrfieation of clusterin. Clusterin was purified from platelets
using PMA-induced releasate obtained from washed platelets corresponding to 3 U of blood. The releasate was spun at 15,OOOg for 20
minutes. The supernatant was subjected to a immuno-affinitychromatography column (5 mL bed volume) with immobilized monoclonal anti-human clusterin antibodies’equilibrated in 10 mmol/L
Tris HC1, pH 7.4, containing 150 mmol/L NaCl (TBS). The adsorbed proteins were washed with 50-mL TBS, containing 0.5 NaCI
(final concentration). Clusterin was eluted with 0.2 mol/L glycine,
pH 2.8, and the collected fractions were immediately neutralized
with a I mol/L solution of Tris HCl, pH 8.0. Fractions containing
clusterin antigen were determined using the enzyme-linked immunosorbent assay (ELISA) described below, dialyzed against TBS,
and stored at -20°C until use.
Quantijication of clusterin in solubilized platelets. An ELISA
was used to quantify clusterin, which will be described in detail
elsewhere.23In brief, wells of microtiter plates (Nunc, MAXISORP,
Basel, Switzerland) were coated in the presence of 0.2% Tween-20
(SigmaChemical Co, St Louis, MO). Whereas most proteins do not
bind to the plastic in the presence of the detergent, clusterin is adsorbed in a dose-dependent fashion. Clusterin was detected using
the mouse anticlusterin antibody CL1-9, followed by an anti-mouse
IgG-peroxidase and the chromogenic substrate 2,2’-Azino-di-(3ethyl-benzthiazolinsulfonic acid) (Boehringer, Mannheim, Germany) according to standard procedures.
In situ hybridization. The clusterin sense and antisense probes,
prepared from PGEM-4 CLI, containing the full-length clusterin
cDNA,‘ were transcribed in vitro in the presence of 30 pmol/L
[3H]-labeled uridine triphosphate (UTP) and 30 pmol/L [3H]labeled CTP (cytidine 5’-triphosphate,40 and 20 Ci/mmol, respectively; Amersham International, Zurich, Switzerland) and reduced
to an average size of 50 to 100 nucleotides by mild alkaline hydrolysis as previously described.24
Human spinal bone marrow (BM) aspirates were smeared onto
poly-L-lysine (Sigma)coated microscope slides, fixed in 4% glutaraldehyde in phosphate-bufferedsaline (PBS)for 4 minutes rinsed in
PBS, and storedin 70%ethanol at 4°C until analyzed. In situ hybridizations were performed with minor modifications of the procedure
previously de~cribed.’~
Briefly, 0.4 to 1 X IO6 cpm of [’HI-labeled
cRNAs were applied to each section in 60 pL of hybridization mixture. After graded ethanol dehydration, sections were immersed in
NTB-2 emulsion (Eastman Kodak, Lausanne, Switzerland), diluted 1:l in deionized water. After 12- to 15-week exposure, they
were developed in Kodak D-19 developer, fixed in 30%Na-thiosulfate, and counterstained in methylene blue or May-Griinwald solution (Fluka, Buchs, Switzerland).
Hybridizations with sense RNA probes were performed as controls of specificity. For each experiment, control tissue sections of
human testes were also included, the Sertoli cells containing large
amounts of clusterin mRNA. Microphotographswere taken with a
photomicroscope (Zeiss, Oberkochen, Germany), equipped with
an immersion dark-field condensor, using Kodak Ektachrome 50
color film.
Immunohistochemistry. For immunoperoxidase stainings,
freshly isolated platelets were cytocentrifuged at 600 rpm for 10
minutes allowed to dry, and kept at room temperature overnight.
The cells were permeabilized with 0. I % Triton X-100. A three layer
biotin-avidin-peroxidase technique was used25in combination with
monoclonal antibodies (MoAbs) to human clusterin (CLI-9). The
peroxidase activity was shown with the chromogen 3-amino-9-ethylcarbazol.”
Immuno-electronmicroscopy. Washed human platelets (2 X
I 08/mL) from hirudin-supplemented blood were prefixed with 50
mL fixative (6% paraformaldehyde, 0.1% glutaraldehyde in 0.1
mol/L phosphate buffer, pH 7.2). After centrifugation, the pellets
were postfixed with 3% paraformaldehyde, 0.1% glutaraldehyde in
phosphate buffer, washed, and resuspended in a solution containing 7.5% (wt/vol) polyvinylpyrrolidone K25 (Fluka AG, Buchs,
Switzerland)and 2 mol/L sucrose in PBS. For inclusion, droplets of
platelet-rich suspensions were placed on 5% gelatine at 3 7 T , and
after cooling, blocks were plunge-frozen in liquid propane at
- 180°C (KF80; Reichert, Vienna, Austria). Ultrathin frozen sections (FC4 D; Reichert-Jung, Vienna, Austria) were mounted with
2.3 mol/L sucrose droplets on pioloform-coated Ni-grids.
After washing in PBS containing 0.05 mol/L glycine, the grids
were incubated with primary antibody AC8 (IgG rabbit-antihuman clusterin; 5 pL/100 pL PBS) for 60 minutes at 20°C. After
washing with PBS containing 0.5% (wt/vol) bovine serum albumin
(BSA) (Aurion/Biotrend, Koln, Germany), secondary labeling with
goat antibody (IgG goat-antirabbit GAR EM IO; Aurion/Biotrend;
20 pL/mL PBS) was performed for 60 minutes at 20°C. Finally, the
grids were stained with 2% (wt/vol) ammoniummolybdate. Controls were done with rabbit IgG antihuman fibrinogen (Dakopatts,
Hamburg, Germany) as primary antibody or without primary antibody.
Other methods. Gel electrophoretic analysis was performed accordingto L;immli,z6using 10%sodium dodecyl sulfate (SDS)-polyacrylamide gels. After electrophoresis, proteins were transferred
onto nitrocellulose using the buffer system of Towbin et aL2’Immunoblots were processed using the alkaline-phosphatase detection
system. The apolipoprotein (apo) A-I specific rabbit antiserum was
purchased from Behring (Zurich, Switzerland).
RESULTS
Association of clusterin with human platelets. Clusterin circulates in human plasma as an apoA-I containing
HDL complex that most likely originates from human
liver. These HDL-like complexes as well as clusterin found
in seminal plasma have been extensively characterized.
The clusterin gene is also transcribed in a regulated manner by resident cells ofother organ tissues in the extravascular compartment. To investigate the possible presence and
storage of clusterin in circulating hematopoietic cells, immunocytochemistry on PB cells was performed using an
indirect peroxidase technique. Two different clusterin-specific MoAbs yielded the same immunostaining pattern.
The majority of leukocytes and erythrocytes were clusterin-negative (data not shown), whereas thrombocytes
stained strongly positive after immune-peroxidase visualization of bound antibodies. When purified platelets were
analyzed by immunocytochemistry, most cells scored positive (Fig 1). The staining signal was highly specific and
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
TSCHOPP ET AL
120
L
'1
P
only low background was observed when the specific
MoAb was omitted.
Actii~ation-dependent rcJ/casc(f chisterin .finin plate/els.
Different agonists were used to release clusterin from activated human thrombocytes. The CaZ' ionophore A23 187 at
a concentration of 20 pmol/L led to the complete release of
clusterin into the supernatant (Fig 2. R). This treatment also
led to the parallel release of platelet-factor 4 (data not
shown). No remaining clusterin was detected in washed pellets (Fig 2, P) after degranulation. Almost 100% release was
Clu
kDa
PR
PR P R
F T CluRP
also observed with different agonists. ie, 200 nmol/L PMA
and I O nmol/L thrombin. Immunoblot analysis of the supernatant from activated platelets showed one band ofclusterin antigen. migrating at 70 Kd. indistinguishable from
plasma-derived clusterin (Clu). The amount of antigen detected in the supernatant of thrombin-stimulated platelets
(R) was similar to that obtained after freezing and thawing
(Fig 2. F) or solubilization with 1% Triton X-100 (Fig 2, T).
Isolation and cliaractcrization c?Splatelet clusterin. A n
MoAb column was used to purify clusterin from platelets by
affinity chromatography. PMA-induced releasates corresponding to 3 U of blood were harvested and passed over a
clusterin-antibody affinity column. After extensive washing, platelet clusterin was eluted with a 0.2 mol/L glycine
buffer, pH 2.8. as previously described for plasma clusterin.
When the fractions were analyzed by SDS-polyacrylamide
gel electrophoresis (PAGE) under nonreducing conditions
followed by Coomassie blue staining (Fig 3A). platelet clusterin migrated as a single band and showed the same apparent molecular weight of 70 Kd as the plasma-derived clusterin.
Purified platelet clusterin from fraction 13 (Platelet C)
was further analyzed by immunoblot analysis under nonreducing (Fig 3B. NR) and reducing conditions (Fig 3B, R).
Without reduction. fraction 13 contained a single immunereactive protein of approximately 70 Kd that comigrated
with plasma-derived clusterin. As shown in Fig 3B. clusterin
from human platelets migrated in multiple. closely spaced
bands of about 35 Kd after reduction with dithiothreitol
(DTT), again similar to serum-derived clusterin that was
shown to be proteolytically processed into two subunits of
identical size during biosynthesis. However. a minor frac-
I
6 8 4 0
45
Fig 1. lmmunostaining of clusterin in platelets.
(A) Platelets purified from human blood were permeabilized and treated with a monoclonal anti-human clusterin antibody, biotinylated antimouse
IgG, and avidin-peroxidase conjugate with appropriate washes. Palatelets are brightly labeled in
contrast to the control in which the monoclonal anticlusterin antibody has been left out (B) or an isotype-matched irrelevant antibody (anti-T-cell receptor, data not shown) has been used. Original
magnification X 300.
1
-A
23187
PMA
Thrombin
Thrombin
Fig 2. Release of clusterin from platelets by secretagogues. Supernatents (R) and washed pellets (P) of platelets treated with the
Ca2+ ionophore A231 87 (20 pmol/L), the protein kinase C activator
P M A (200 nmol/L) and thrombin (1nmol/L) were analyzed by Western blot analysis. Thrombin-induced releasate was harvested from
platelets of two individuals. Also shown are platelets solubilized
with 1%Triton X-1 00 (T) or by three cycles of freezing and thawing
(F). Purified plasma clusterin (Clu) is included as control.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
CLUSTERIN EXPRESSION IN MEGAKARYOCYTES
121
A
G
15
10
20
Fraction number
C
B
identify substoichiometric amounts of apoA-I in preparations of platelet clusterin.
Qiiunlifitution ofpk11e4~1
c~lri.stc~tYn. A novel ELISA was
used to measure the amount of clusterin in biologic samples. Platelets were washed in physiologic salt buffers and
lysed in 0.2% Triton X-100. The amount of clusterin that
adsorbs to microtiter plates after blocking with 0.2% Tween20. was shown to be linear in the range between 5 ng/mL
and 100 ng/mL clusterin and was quantitated immunologically with the use of peroxidase-conjugated antimouse antibodies. In control experiments in which an irrelevant MoAb
was used. low background signals were obtained with platelet extracts. Standard curves were constructed with purified
clusterin and plasma pools of healthy blood donors. In normal subjects. the clusterin content of human platelets
amounts to 2.5 I .3 &IO9 platelets (n = 5). PMA-induced
releasates showed similar values (data not shown).
Loculizrrt ion of clir.sto.in in n-pun I ilcs f7.r irninI inoc.lc.cIron tnicro.sc.opy. To identify the subcellular localization
of clusterin in human platelets. acryl-embedded ultrathin
sections of platelets were studied by immunoelectron microscopy using monoclonal anti-clustein antibodies in conjunction with immunogold-labeled second antibodies. Because of their characteristic density and size. n-granules
were easily distinguishable from other types of platelet organelles. The cu-granules have diameters of approximately
300 nm. As can be seen in Fig 4. resting platelets with an
oval section profile showed dense labeling in the n-granules.
preferentially located in the periphery of the granules.
whereas moderately dense labels were seen within the surface-connected membrane system. Labeling above background of the plasma membrane was also found. A similar
pattern was observed for fibrinogen (Fig 4C).
Biosvnt l1c~si.sof’ clr ist w i n bj nicyukurj*ocyte x Because
clusterin is a relatively abundant G P (50 to 100 pg/mL) in
plasma and only a small portion of clusterin per milliliter
blood is associated with washed platelets, we wondered
whether clusterin is taken up into platelets from the plasma
pool via the endocytic pathway as has been shown in the
case of albumin. IgG. and fihronectin.’* or whether clusterin gene is actually synthesized and directly packaged into
n-granules at a certain stage of megakaryocyte diferentiation. To establish clusterin transcription in platelet precursors, cytospin BM cell preparations were hybridized to a
tritium [3H]-labeled clusterin cRNA probe. Specifically hybridizing transcripts were shown by emulsion autoradiography. Mature and intermediate megakaryocytes were identified by their morphology using light microscopy (Fig 5).
Expression ofclusterin mRNA was detected in a population
of megakaryocytes with multilobed nuclei whose cytoplasm
equaled the nucleus in size (Fig 5A and B). Occasionally,
cells with a bilobed or trilobed nucleus did not stain positive
for clusterin mRNA. suggesting that in more immature
megakaryocytes the clusterin mRNA is absent or scarce (Fig
5C). There was also little or no expression in platelets (data
not shown). A low amount of clusterin expression was observed in and between other BM cells (Fig 5A. B. and C). but
the cell types expressing low amounts of clusterin have not
been identified. Hybridization with the sense strand cRNA
*
NR
R
NR
Fig 3. Purification of platelet clusterin. Clusterin was isolated
from PMA-mediated releasateof platelets bya one-step immunoaffinity chromatography with immobilized mouse antihuman clusterin. After washing was done to elute nonspecifically bound proteins, clusterin was eluted by lowering the pH to 2.8 and 1-mL
fractions were collected. Half of the lyophilized sample was analyzed by SDS-PAGE under nonreducing conditions and stained with
Coomassie blue (A). The outer left lane corresponds to purified
plasma-clusterin ( 5 pg). Fraction 13 (PlateletC) was analyzed by
Westem blot under nonreducing (NR) and reducing (R)conditions
and compared with plasma clusterin (PlasmaC) using a clusterinspecific MoAb. In C, both forms of clusterin were assayed by immunoblot for the presence of apoA-l with a monospecific anti-apoA-l
antiserum.
tion ofthe purified protein was resistant to reduction. probably representing incompletely processed precursor molecules of clusterin.
The platelet and plasma forms of human clusterin differed strikingly in their molecular composition. As shown in
Fig 3C, clusterin in human platelets is not associated with
apoA-I, whereas clusterin of human plasma occurs as a
stable HDL-like complex containing apoA-I. The 28-Kd
band of apoA-I was detected only in plasma-derived clusterin, but was not detected in affinity-purified platelet clusterin by the sensitive technique of immunoblotting(Fig 3C).
Absence of apoA-I in preparations of platelet clusterin excludes the possibility that contaminating plasma is the
source of platelet-associated clusterin. Preferential uptake
of clusterin from human plasma or preferential loss of
apoA-I into plasma seems unlikely because we failed to
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
122
TSCHOPP ET AL
Fig 4. Localizationof clusterin in resting human platelets. (B and D)Localization of clusterin in resting human platelets. Gold labels are
associated with a-granules (G, one granule is encircled by white dots in [B]) and mostly associated with the periphery). Some gold particles
are also discernable at the plasma membrane (arrowheads) and within the surface connected system (SCS);
M, mitochondria. (A) The
control shows a platelet section without primary antibody (anticlusterin) added. ( C )This control shows a platelet section labeled with rabbit
IgG antihuman fibrinogen antibodies. Gold particles are found in a-granules (G), on the membranes of the surface connected system (S)
and
with moderate density at the platelet surface. M,mitochondria.
probe generated significantly lower levels of silver grain
density (Fig 5D).
DISCUSSION
Human clusterin is found in plasma and in various tissues
and cell lineages ranging from epithelial cells and hepato-
cytes to neuroendocrine cells. In this study. we identify clusterin in a-granules of platelets and show most abundant
clusterin expression in a subpopulation of immature megakaryocytes using in situ RNA:RNA hybridization. Our
observations strongly suggest that the clusterin gene is
transcribed during megakaryocyte development at the ap-
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
CLUSTERIN EXPRESSION IN MEGAKARYOCYTES
123
Fig 5. Localization of clusterin mRNA in megakaryocytes
by in situ hybridization of the
'H-labeled clusterin cRNA probe
to human BM smears. (Left
panels) Light field micrographs.
(Right panels) The corresponding autoradiograph dark field
micrographs. The megakaryocyte displayed in A and B contain high amounts of clusterin
mRNA and are representative
of the majority of megakaryocytes on a smear. Occasional
megakaryocytes are weakly or
not stained as the one shown
in C. The megakaryocyte displayed in D was hybridized with
the sense probe.
propriate stage that permits translation and packaging of
proteins into a-granules. Thus. we established that in vivo
megakaryocyte precursors possess the capacity to synthesize
the clusterin that we identified in their cytoplasmic gran-
ules. Although not formally proven. our data support the
view that clusterin is packaged into n-granules in conjunction with the transcription of the clusterin gene. Compared
with the levels of clusterin circulating as an HDL complex
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
TSCHOPP ET AL
124
in human plasma 1 (50 to 80 pg/mL), the amounts ofclusterin in platelets are relatively low and account for only 1 pg
clusterin per milliliter of blood (4 X 10' platelets) that represents about 2% of total clusterin. At present, we cannot rule
out the possibility that at least some clusterin is taken up
through endocytosis of plasma clusterin.
Indirect evidence underscores the biologic significance of
platelet clusterin. First of all, the concentration of clusterin
within platelets is about fivefold higher than in human
plasma. One milliliter of platelets (1.04 X IO") contains
about 250 p g clusterin. About the same concentration has
been reported for von Willebrand factor ( v W F ) ,a~biologi~
cally significant a-granule protein that is also synthesized by
megakaryocytes. Secondly, clusterin of platelets is distinct
from plasma clusterin in that it is not associated with apoAI. Both findings strongly contradict the objection that clusterin could be sequestered into a-granules from plasma entirely through a mechanism of bulk flow endocytosis.
Albumin or a I-antitrypsin that are known to follow this
nonspecific route of entry have at least a IO-fold higher concentration in plasma than in platelets.29
Purification and characterization of platelet clusterin
showed that its two-chain structure and glycosylation pattern is conserved between plasma and platelet clusterin.
Clusterin is known to undergo intensive posttranslational
glycosylation that may differ in certain tissues or cell types.
For instance, the rat analogue of clusterin secreted by Sertoli
cells and found in seminiferous tubule fluid is composed of
monomers of M, 47,000 and 34,000, whereas the epididymal protein exhibits subunits of M, 40,000 and 29,000.30
Because molecular weight differences between the plasma
and platelet form have not been detected by one-dimensional SDS-PAGE, we suggest that the glycosylation pattern
of the two subunits of dusterin is similar, if not identical,
with plasma clusterin. As for plasma clusterin, we found
that a small fraction of the protein in a-granules is not processed into the two-chain form during biosynthesis. Thus,
absence of proteolytic processing does not seem to prevent
correct sorting and packaging of clusterin into platelets.
Likewise, the single-chain form of clusterin has been reported to take the same apical route as the two-chain form
in polarized Madin-Darby canine kidney (MDCK) cell^.'^
Platelet a-granules are the principal intracellular reservoir of proteins destined for release duringprimary hemostasis at the site of vessel wall injury. Many of these granule
proteins contribute to tissue remodeling and wound healing. Clusterin was localized to these storage organelles as
shown by electronmicroscopy after immuno-gold labeling
of intracellular clusterin. This subcellular location is consistent with its release from platelets induced by different secretagogues such as thrombin, PMA, and ionophores. All of
these stimulators triggered almost complete release into the
extracellular medium.
Damage of the inner endothelial layer triggers platelet
adhesion and release of platelet factors to the medial smooth
muscle cells. As a response to platelet- and plasma-derived
factors, smooth-muscle cells change their morphologic phenotype and migrate into the intima forming multi-layered
nodules. These phenotypic alterations of smooth muscle
cells in vivo resemble the morphologic transformation of
cultured smooth muscle cells in vitro. As shown previ~ u s l y clusterin
,~~
mRNA are low in monolayer cultures of
smooth muscle cells, but increase strongly during formation
of multilayers and muscle-cell nodules. Thus, local synthesis and secretion of clusterin is probably initiated in smooth
muscle cells after activation and proliferation at the site of
injury. Co-ordinate upregulation of clusterin by two independent delivery mechanisms, degranulation of platelets,
and local de novo synthesis would be consistent with a putative biologic role in remodeling and regeneration of injured
blood vessel walls.
However, how clusterin contributes to lesion repair at the
biochemical level remains to be investigated. Clusterin, in
particular the cell-derived form that is not associated with
HDLs, may function as a local carrier for poorly soluble
growth-promoting peptides or lipophilic substances. Alternatively, it may be involved in homotypic or heterotypic
cell-to-cell interactions in view of its cell-aggregating activity. Because clusterin inhibits the assembly ofthe lytic terminal complement complex on cellular membranes, it may
reduce complement activation at the site of vessel wall damage and may protect endothelial and smooth muscle cells
from inappropriate complement attack.
ACKNOWLEDGMENT
The authors thank Magali Schreier, M. Allegnni, and M. Freiwald for their technical assistance as well as Dr P. Beris who helped
us interpret the medullary aspirates and collect the material for
hybridizations. The helpful comments of Drs R. Etges, T.
Wiedmer, and P. Sims are also gratefully acknowledged.
REFERENCES
1. Jenne DE, Tschopp J: Molecular structure and functional
characterization of a novel human cytolysis inhibitor (CLI) found
in blood and seminal plasma. Identity to sulfated glycoprotein 2
(SGP-2), a constituent of rat testis fluid. Proc Natl Acad Sci USA
86:7123, 1989
2. Murphy BF, Saunders JR, O'Bryan MK, Kirszbaum L,
Walker ID, d'Apice A J F SP-40,40 is an inhibitor of C5b-6-initiated haemolysis. Int Immunol 1551, 1989
3. Choi N-H, Nakano Y ,Tobe T, Mazda T, Motowo T: Incorporation of SP-40,40 into the soluble membrane attack complex
(SMAC, SC5b-9) of complement. Int Immunol2:413, 1990
4. Choi N-H, Mazda T, Tomita M: A serum protein SP40,40
modulates the formation of membrane attack complex of complement on erythrocytes. Mol Immunol 26935, 1989
5. de Silva HV, Harmony JA, Stuart WD, Gil CM, Robbins J:
Apolipoprotein J: Structure and tissue distribution. Biochemistry
295380, 1990
6. Kirszbaum L, Sharpe JA, Murphy B, d'Apice AJ, Classon B,
Hudson P, Walker, ID: Molecular cloning and characterization of
the novel, human complement-associated protein, SP-40,40: A link
between the complement and reproductive systems. EMBO J
8:71 I, 1989
7. Sylvester SR, Skinner MK, Griswold MD: A sulfated glycoprotein synthesized by Sertoli cells and by epididymal cells is a
component of the sperm membrane. Biol Reprod 31:1087, 1984
8. Collard MW, Griswold MD: Biosynthesis and molecular
cloning of sulfated glycoprotein 2 secreted by rat Sertoli cells. Biochemistry 26:3297, 1987
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
CLUSTERIN EXPRESSION IN MEGAKARYOCYTES
9. Blaschuk OW, Burdzy K, Fritz IB: Purification and characterization of a cell-aggregating factor (clusterin), the major glycoprotein in ram rete testis fluid. J Biol Chem 258:7714, 1983
IO. Cheng CY, Chen CL, Feng ZM, Marshall A, Bardin CW: Rat
clusterin isolated from primary Sertoli cell-enriched culture medium is sulfated glycoprotein-2 (SGP-2). Biochem Biophys Res
Commun I55:398, 1988
I 1. Jenne DE, Tschopp J: Clusterin: The intriguing guises of a
widely expressed glycoprotein. TIBS 17:154, 1992
12. Jenne DE, Lowin B, Peitsch MC, Bottcher A, Schmitz G,
Tschopp J: Clustenn (complement lysis inhibitor) forms a high
density lipoprotein complex with apolipoprotein A-I in human
plasma. J Biol Chem 266: 11030, 1991
13. de Silva HV, Stuart WD, Park YB, Mao SJ, Gil CM, Wetterau JR, Busch SJ, Harmony JA: Purification and characterization
of apolipoprotein J. J Biol Chem 265:14292, 1990
14. James RW, Hochstrasser AC, Borghini 1, Martin B, Pometta
D, Hochstrasser D: Characterization of a human high density lipoprotein-associated protein, NA I /NA2. Identity with SP-40,40, an
inhibitor of complement-mediated cytolysis. Arterioscler Thromb
I1:645, 1991
15. Hartmann K, Rauch J, Urban J, Parczyk K, Diel P, Pilarsky
C, Appel D, Haase W, Mann K, Weller A, et al: Molecular cloning
of gp 80, a glycoprotein complex secreted by kidney cells in vitro
and in vivo. A link to the reproductive system and to the complement cascade. J Biol Chem 266:9924, I99 1
16. May PC, Lampert Etchells M, Johnson SA, Poirier J, Masters JN, Finch CE: Dynamics of gene expression for a hippocampal
glycoprotein elevated in Alzheimer’s disease and in response to experimental lesions in rat. Neuron 5:831, 1990
17. Krisch K, Buxbaum P, Horvat G, Krisch I, Neuhold N,
Ulrich W, Srikanta S: Monoclonal antibody HISL-19 as an immunocytochemical probe for neuroendocrine differentiation. Its application in diagnostic pathology. Am J Pathol 123:100, 1986
18. Buttyan R, Olsson CA, Pintar J, Chang C, Bandyk M, Ng
P-Y, Sawczuk IS: Induction of the TRPM-2 gene in cells undergoing programmed death. Mol Cell Biol9:3473, 1989
19. Duguid JR, Bohmont CW, Liu N, Tourtellotte WW:
Changes in brain gene expression shared by scrapie and Alzheimer
disease. Proc Natl Acad Sci USA 86:7260, 1989
125
20. Danik M, Chabot JG, Mercier C, Benabid AL, Chauvin C,
Quiron R, Suh M: Human gliomas and epileptic foci express high
levels of a mRNA related to rat testicular sulfated glycoprotein 2, a
purported marker of cell death. Proc Natl Acad Sci USA 88:8577,
1991
2 1. Kyprianou N, English HF, Isaacs JT: Programmed cell death
during regression of PC-82 human prostate cancer following androgen ablation. Cancer Res 50:3748, 1990
22. Rennie PS, Bruchovsky N, Buttyan R, Benson M, Cheng H:
Gene expression during the early phases of regression of the androgen-dependent Shionogi mouse mammary carcinoma. Cancer Res
48:6309, 1988
23. Hogasen K, Mollnes TE, Tschopp J, Harboe M: Quantitation of vitronectin and clusterin: Pitfalls and solutions in enzyme
immunoassays for adhesive proteins. J Immunol Methods 160: 107,
1993
24. Sappino A-P, Huarte J, Vassali J-D, Belin D: Sites of synthesis of urokinase and tissue-type plasminogen activators in the murine kidney. J Clin Invest 87:962, 199 I
25. Guesdon JL, Ternynck T, Avrameas S: The use of avidinbiotin interaction in immunoenzymatic techniques. J Histochem
Cytochem 27:1131, 1979
26. Lammli UK: Cleavage of structural proteins during the assembly of the head of Bacteriophage T4. Nature 227:680, 1970
27. Towbin H, Staehelin T, Gordon J: Electrophoretic transfer
of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc Natl Acad Sci USA 76:4350,
I979
28. Harrison P: The origin and physiological relevance of alphagranule adhesive proteins. Br J Haematol 74:125, 1990
29. George JN: Platelet immunoglobulin G: Its significance for
the evaluation of thrombocytopenia and for understanding the origin of alpha-granule proteins. Blood 76:859, 1992
30. Sylvester SR, Morales C, Oko R, Griswold M D Localization
of sulfated glycoprotein-2 (clusterin) on spermatozoa and in the
reproductive tract of the male rat. Biol Reprod 45: 195, 199 1
3 I . Diemer V, Hoyle M, Baglioni C, Millis AJT: Expression of
porcine complement inhibitor mRNA in cultured aortic smooth
muscle cells: Changes during differentiation in vitro. J Biol Chem
2675257, 1992
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1993 82: 118-125
Human megakaryocytes express clusterin and package it without
apolipoprotein A-1 into alpha-granules
J Tschopp, DE Jenne, S Hertig, KT Preissner, H Morgenstern, AP Sapino and L French
Updated information and services can be found at:
http://www.bloodjournal.org/content/82/1/118.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.