From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
Demonstration of Osteonectin mRNA in Megakaryocytes: The Use of the
Polymerase Chain Reaction
By Xavier
C.Villarreal,
Barbara W. Grant, and George L. Long
Platelets have been shown to release osteonectin on thrombin stimulation. The origin of platelet osteonectin was unclear as it may have been synthesized by megakaryocytes or
it may have been endocytosed from plasma as other platelet
a-granule constituents are. Platelet osteonectin has a larger
apparent molecular size than the bone species, although the
molecular basis for this difference has not been elucidated.
These two issues have been addressed here by (1)examining
the potential for osteonectin biosynthesis in human megakaryocytes by demonstrating the presence of osteonectin
mRNA in purified megakaryocyes, and (2) comparing the
coding portion of osteonectin transcript in megakaryocytes
t o the size of its bone counterpart. Because of the limitations
of cell population purity and in obtaining sufficient numbers
of megakaryocyte cells for Northern analysis, we have used
the polymerase chain reaction (PCR) to detect the presence
of human osteonectin mRNA in megakaryocyte and megakaryocyte-depleted bone marrow cells. Isolation of RNA,
cDNA synthesis, and PCR were performed on human osteosarcoma SaOS-2 cells, enriched megakaryocytes, and megakaryocyte-depleted cells. Restriction enzyme analysis of PCR
DNA products confirmed the identity of the products as
those encoding osteonectin for all three cell populations
studied. In addition, the sizes of DNA indicate that osteonectin genomic DNA, nuclear RNA, or altered transcript were not
amplified, and that the transcript from megakaryocytes is the
same size as that from bone cells. These data suggest that
the difference in protein size between platelet and bone
osteonectin is due t o posttranslational modification. To
overcome the possibility that megakaryocyte signal originated from contaminating cells (less than 5% by cell count),
all three cell populations were diluted t o less than one cell
per tube and PCR amplification was performed. Limiting
dilution analyses demonstrated the presence of osteonectin
mRNA in single megakaryocytes as well as in single cells
from the cell population depleted of megakaryocytes, suggesting the capacity for osteonectin biosynthesis in all cells
studied. The procedure we describe in this report can be used
t o examine specific characteristics of mRNA molecules in
heterogeneous cgll populations and in situations where only
small quantities of cells can be obtained.
o 1991 by The American Society of Hematology.
S
bone forms. The apparent larger size for platelet osteonectin may be due to either peptide cleavage at the carboxyterminus of bone osteonectin, differences in posttranslational protein modification, and/or differences in the mRNA
for the two proteins.
In tissues undergoing rapid proliferation and remodeling, including bone, it has been proposed that osteonectin
participates in extracellular matrix reg~lation.'~-'~
Unlike
bone and basement membrane tissues, platelets are anucleate and consequently not capable of undergoing rapid
proliferation and remodeling. Hence, the physiologic role
of osteonectin in platelets may be different from that in
tissues. Platelets secrete adhesion proteins, including fibrinogen," thrombospondin,18 and von Willebrand
that are involved in platelet aggregation and platelet
adherence to the vascular endothelium. Thrombospondin
has been reported to bind to platelet osteonectin and
thrombospondin-osteonectin complexes have been isolated
from thrombin activated platelets?' One intriguing possibility is that osteonectin via its interaction with thrombospondin may mediate the adhesion of platelets to each other and
to the vascular endothelium. Other possibilities may be a
role in regulating extracellular Caz+levels or as a protease
inhibitor, thereby indirectly or directly affecting complex
biologic processes such as blood coagulation.
In this study the polymerase chain reaction (PCR) was
used to detect and characterize the osteonectin transcript
in human megakaryocytes. Megakaryocytes were enriched
from the contaminating nonmegakaryocyte population using magnetizable beads coated with antibodies that react
against the megakaryocyte-specific cell-surface antigen,
IIbflIIa. The results of this study show identical osteonectin
transcript sizes for human osteosarcoma cell line SaOS-2,
megakaryocytes, and megakaryocyte-depleted populations.
Limiting dilution analysis suggests that osteonectin mRNA
TENNER et al' have reported the presence of osteonectin in platelets and its secretion on thrombin stimulation. Previous studies describe the purification of osteonectin homologues from bovine aortic cells (43K protein):
murine embryonic parietal endoderm (SPARC): and a
murine basement membrane tumor (BM-40): Two possible
biologic processes that may give rise to osteonectin in
platelets are endogenous synthesis in the megakaryocyte
precursor cell and/or endocytosis of exogenously produced
osteonectin circulating in plasma. Examples of platelet
proteins believed to be acquired via megakaryocyte endocytosis include: fibronectin: fibrinogen,6 and high molecular
weight kinin~gen.~
Endogenous syntheses of P-thromboglobulin,8 platelet factor-4: and thrombospondin" have been
reported for megakaryocytes.
Kelm and Mann"~'*have recently noted a difference in
molecular weight of -3,000 between platelet and bone
osteonectin (bovine and human) when analyzed by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE). Amino acid sequencing of platelet osteonectin
established that amino-termini are identical in platelet and
From the Departments of Biochemistry and Medicine, University of
Vermont,Burlington.
Submitted January 2,1991; accepted May 3,1991.
Supported by US Public Health Service Grants HL 38899, HL
34282, HL 01607, GCRCRR109, and C 0 6 HL39745.
Address reprint requests to George L. Long, PhD, Department of
Biochemistry, University of Vermont, Given Bldg, Burlington, VT
05405.
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 1734 solely to
indicate this fact.
0 I991 by The American Society of Hematology.
0006-4971I91/7805-OO23$3.OOiO
1216
Blood, Vol78, No 5 (September 1). 1991: pp 1216-1222
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
OSTEONECTIN MRNA IN MEGAKARYOCYTES
is present in megakaryocytesand also in the megakaryocytedepleted cell populations.
MATERIALS AND METHODS
Megakaryocyte cell culture and purification. Iliac crest bone
marrow was obtained from normal volunteers who had given
informed consent consistent with institutional guidelines. Nonadherent light density bone marrow cells were prepared and partially
depleted of platelets as previously described.” Cells were cultured
at lo6 cells/mL in Iscove’s modified Dulbecco’s medium (GIBCO
Laboratories, Grand Island, NY) supplemented with 10% fetal
bovine serum (FBS; Hyclone Laboratories, Logan, UT) and
recombinant human IL-3 10 U/mL (a gift of S. Clark, Genetics
Institute, Cambridge, MA) for 10 days at 37”C, 5% CO,. Megakaryocytes were separated from other cultured cells using the immunomagnetic procedure described by the manufacturer (Dynal, Fort
Lee, NJ) with appropriate modifications. Magnetizable particles
coated with goat antimouse IgG were obtained from Dynal and
incubated with a murine monoclonal specific for the megakaryocyte/
platelet specific glycoprotein IIb/IIIa complex (1.6 p,g antibody
HP-lD/mg beads), a gift of W. Nichols (Rochester, MN). Before
use, the beads were washed in phosphate-buffered saline (PBS;
0.441 mmol/L KH,PO,, 1.33 mmoW NqHPO,, and 137 mmoVL
NaCI, pH 7.4) three times by collecting them onto a Dynal magnet
and decanting the supernatant. Cultured cells were suspended in
culture medium at 2 x lo7cells/mL and cooled on ice to 4°C. Beads
were added at a ratio of 10 beads per megakaryocyte (calculated
from cytospin cell differentials of cultured marrow). The mixture
was incubated on ice with occasional shaking for 15 minutes and
then diluted to 2 x lo6 cells/mL with cold medium to minimize
entrapment of negative cells on magnetic collection. Separation of
cells labeled with magnetizable particles was achieved on the
magnet by placing the tube containing the cell suspension in
contact with the magnet for 2 minutes, decanting the supernatant;
and repeating the suspension, magnet separation, and decanting
steps two more times to wash the remaining cells. Cell yield was
determined using a hemacytometer. From 5 X lo7 cultured cells
containing 0.5% to 2% recognizable megakaryocytes the yield of
bead-positive cells was typically 50,000 to 100,000. Megakaryocyte
purity was determined morphologically on stained cytocentrifuge
preparations, and cells used for these studies contained large
multinucleate cells with purple granular cytoplasm. Contaminating
cells as judged by morphologic criteria were macrophages and
granulocytes. Nonmegakaryoqte preparations containing primarily macrophages and granulocytes were taken from decanted
supernatant and their cell yields and purities (50.1% megakaryocyte contamination) were estimated in a similar fashion.
Cell culture of SaOS-2 and Jurkat cell lines. SaOS-2 cells were
grown in Eagle’s minimum essential media and Jurkat cells in
RPMI supplemented with 10% FBS (Hyclone Laboratories), 50 U
of penicillin, 50 pgimL streptomycin, and 5 &mL of gentamycin
(antibiotics purchased from GIBCO Laboratories). Cells were
grown in a humidified atmosphere of 95% air/5% CO,. Jurkat cells
grown in suspension to densities of 5 X 105 cells/mL were used.
SaOS-2 cells were grown in suspension as aggregates for 3 days to a
density of 7 X 105cells/mL. SaOS-2 cells were dispersed by treating
with trypsin (0.01%, GIBCO Laboratories) and 1 mmoVL ethylene
diamine tetraacetate (EDTA). Jurkat and SaOS-2 cells were
counted using a hemacytometer.
RNA isolation. Total RNA was isolated by the method of
Chomczynski and Sacchi.’ Approximately 10,OOOcells were used to
extract RNA (large-scale purification). Small-scale RNA isolations
were also performed in limiting dilution analysis (-0.37 cells).
1217
Volumes of reagents used to extract RNA were adjusted proportionately to those used by Chomczynski and Sacchi.=
cDNA synthesis. Synthesis of cDNA used reagents and buffers
supplied by Bethesda Research Laboratories (BRL; Bethesda,
MD) in reaction volumes of 20 pL. For large-scale RNA preparations the following concentrations of reagents were used: 10
p,mol/L of each deoxynucleotide triphosphate (dNTF’), 1 ng oligo(dT)lZ.ls. For small-scale RNA extractions the cDNA buffer
conditions were: 10 nmol/L of each dNTP, 1 pg of oligo(dT),,,,.
After heat denaturation for 5 minutes at 65”C, 100 U of Moloney
Murine leukemia virus reverse transcriptase was added. Samples
were incubated for 45 minutes at 37°C. Digestion of RNA in
mRNA-cDNA hybrids was accomplished by the addition of 1 U of
RNAse H and digesting for 1 hour at 37°C. Two ammonium
acetate-isopropanol precipitations were performed to remove
excess dNTPs and oligo (dT)lz.18.Samples were rinsed with 70%
cold ethanol and dried under vacuum. Large-scale cDNA preparations were resuspended in 100 )LLof diethylpyrocarbonate (DEPC)treated water and stored at 4°C before PCR. Small-scale cDNA
syntheses were used directly for PCR.
Oligonucleotide synthesis. Oligonucleotides used as primers for
PCR were synthesized using 2-cyanoethyl phophoramidite derivatives on an Applied Biosystems (ABI; Foster City, CA) DNA
Model 381A synthesizer. Oligonucleotides were purified on an
HPLC unit (AB1 Model 1 5 2 4 polynucleotide separation system)
using a column packed with 7-km particle size Aquapore RP-300
resin. The manufacturer’s recommended buffer and gradient
conditions for purification of oligonucleotides were followed.
Purified oligonucleotides were dried down and resuspended in 100
KL of DEPC-treated water. Concentrations were determined by
absorbance readings at 260 nm and diluted to 4 pmol/L.
PCRpn’mem. Primers used to amplify osteonectin consisted of
five sets spanning the coding region of osteonectin. The numbering
of nucleotides (nts) is according to the cDNA sequence reported
elsewhere.” Primer set 1was composed of nts 58 through 87 and
nts 969 through 936, ivc (inverse complement); set 2: nts 58 through
87 and nts 504 through 487, ivc; set 3: nts 182 through 199 and nts
782 through 765, ivc; set 4: nts 651 through 671 and nts 969 through
936, ivc; and primer set 5: nts 811 through 828 and nts 929 through
912, ivc. Primer combination 5 was used to amplify osteonectin
probe for Southern hybridizations. Primers for p-actin were
constructed to the human species reported by Ponte et al.= The
oligonucleotides were constructed to nts 146 through 166 and 650
through 627, ivc, as reported elsewhere.=
Osteonectin recombinant plasmids. Recombinant plasmids containing osteonectin inserts (pHVON-9-2 and pHVON-1.5 RI)
were used as templates to determine primer combinations for use
in PCR, and have been described.” Plasmids were linearized by
PstI digestion and diluted to 1 ngipL concentrations before PCR
amplification.
PCR amplification of cDNA. The reagents used for PCR were
from Perkin-Elmer Cetus (Emeryville, CA). Amplification reaction
mixtures for PCR consisted of preparing reaction solutions of
dNTPs, primers, buffer, and water. Final reaction volumes were
100 pL. For large-volume cDNA PCR reactions, four reaction
solutions were made using different osteonectin primer sets.
Reaction solutions contained 200 pmoW of each dNTP (final
molarity), 1X PCR buffer (10 mmoW Tris-HC1, pH 8.3,50 mmol/L
KCI, and 1.5 mmol/L MgCl,) and 20 pmol each of each set of
osteonectin primers. Two microliters of resuspended cDNA was
added. For limiting dilution studies the reaction solutions were
composed of 50 Kmol/L of each dNTP (final molarity), 1X PCR
buffer, and 10 pmol of p-actin and osteonectin (set 4) primers. A
programmed Perkin-Elmer Cetus thermal cycler was used. All
samples were heated at 94°C for 10 minutes. After initial heating,
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1218
VILLARREAL, GRANT, AND LONG
0.5 U of AmpliTaq DNA polymerase (Perkin-Elmer Cetus) was
directly pipetted into the tubes, and 100 )LL of mineral oil was
overlayed onto the surface, followed by thermal cycling. Cycling
conditions were: denaturation at 92°C for 45 seconds, annealing at
56°C for 45 seconds, and extension at 72°C for 60 seconds.
Thirty-five cycles were performed.
Limiting dilution and Poisson distribution analyses. SaOS-2,
megakaryocyte-enriched, and megakaryocyte-depleted cells were
diluted to less than 1cellil00 pL and delivered in 100-pL volumes
to 500 pL Eppendorf tubes. Within each tube, RNA isolation,
cDNA synthesis, and PCR were performed individually. Osteonectin positive reactions were confirmed on reaction with osteonectin
probe in Southern hybridizations. The mean number of cells per
tube was calculated by the Poisson distribution, using the general
formula F, = (m')(e-")/a, where a is r factorial, F,is the probability
of precisely r cells per tube, and m is the mean number of cells per
tube. To solve for m, the Poisson distribution is equated to the zero
term, ie, the number of negative reactions is r and F, is the fraction
of negative PCR reactions/total PCR reactions. Letting s denote F,
the equation now becomes s = (e-"'). Rearranging and solving for
m yields: m = -In s.
Restriction enzyme digestion. Restriction enzymes and buffers
were purchased from BRL and Boehringer-Mannheim (BM,
Indianapolis, IN). After chloroform extraction and sodium acetateethanol precipitations, PCR products were divided into three
separate tubes. Ten units of restriction enzymes were added. The
untreated controls contained 1X digestion buffer (React Buffer 2,
BRL). Samples were incubated for 60 minutes at 37°C. After
incubation, samples were sodium acetate-ethanol precipitated, and
products were resolved on 3.5% polyacrylamide gel in 1X Trisborate-EDTA (TBE; 89 mmol/L Tris-HC1, pH 8.3, 89 mmol/L
borate, and 2 mmol/L EDTA) running buffer.
Southem hybridization. Osteonectin probe was generated using
primer set 5. This 118-bp product is internal to set 4 primers used
to amplify osteonectin in limiting dilution analysis. The 118-bp
PCR product was separated on a 3.5% polyacrylamide gel and
crush-eluted according to the procedure of Maxam and Gilbert?6
The osteonectin probe was labeled using nonradioactive dNTPs,
buffers, and enzymes according to the instructions supplied with
the Genius (BM) and Photogene (BRL) kits. The PCR products
from small-scale syntheses were resolved on 1.2% agarose gels in
1X TBE running buffer. The method of Smith and Summers" was
used to transfer DNA onto MSI (Micron Separation Inc, Westboro, MA) nylon (Magnagraph) membranes. Hybridization conditions used were those of the manufacturer.
Computeranalysis. Restriction fragment lengths, data searches,
and primer structure analyses were performed using the University
of Wisconsin software package version 5?* Computing was performed on an IBM PS2 Model 60 microcomputer emulating a
VT102 terminal and attached to the University VAX 8600 mainframe, using MS-DOS Kermit version 2.29.
RESULTS
Amplijication of osteonectin cDNA. As a positive cell
control human osteosarcoma SaOS-2 was used. Osteonectin cDNA from SaOS-2 has been previously characterized
by Villarreal et al.24As purification and distribution studies
of osteonectin have suggested its localization to cells
displaying adherent p r ~ p e r t i e s , ~ -it~ ~seemed
' ~ ~ * ~ logical to
select a cell line for a possible negative control which grows
in suspension. We chose to use the Jurkat lymphoblastoid
cell line.
Initial attempts to amplify the region of SaOS-2 osteonectin cDNA from start codon to stop codon (nts 58 through
58100
200
300
400
500
6M)
700
800
900969
l b p
P
I
H
HE
A
K
1
I 1
I
I
P t i
I
P
1
2
H
HE
A
I 1
I
A
I
1
3
K
4
Fig 1. Schematic diagram of osteonectin cDNA regions of amplification. Top line represents the beginning of translation (nts 58
through 60, start codon ATG) to the end of translation (nts 967
through 969, stop codon TAA). Numberingof nucleotides is according
to Villarreal et aI.= Numbers at the right of each horizontal line
represent the primer sets used for amplification. Primer set 1 covered
the entire target area and primer sets 2,3, and 4 span subfragments of
453 bp (5' region), 601 bp (central portion), and 325 bp (3' end),
respectively. Primer set 3 overlaps primer sets 2 end 4. Single letter
abbreviations used for restriction endonuclease sites are: A, A w l ; E,
EcoRI; H,Hinfi; K,&nl; and P,Pvull.
969, Fig 1)failed. Unsuccessful amplification may be due to
DNA secondary structure difficulties resulting within the
cDNA or the primer combination used (primer set 1).
Nucleotide composition analysis showed an average GC
content of -57% in this area of DNA (nts 58 through 969).
We decided to test primer combinations that spanned
different regions of the coding segment of osteonectin.
Primer sets were tested on PstI digested recombinant
plasmids containing SaOS-2 osteonectin cDNA 0.54 kb
(pHVON-9-2) and 1.5 kb (pHVON-1.5 RI) in length. A
composite linear diagram of osteonectin and areas amplified by each primer combination are illustrated in Fig 1.
Primer sets 1 and 4 cover the 5' and 3' ends of osteonectin
cDNA (corresponding to NH,-terminus and COOHterminus, respectively). Primer set 3 overlaps primer set 2
by 115 nucleotides and primer set 4 by 85 nucleotides, as
shown in Fig 1.
To guard against osteonectin cDNA amplification by
recombinant plasmids harboring osteonectin inserts (cDNA
or genomic DNA), contamination controls were run. Contamination controls contained all reagents for PCR with the
only difference being omission of osteonectin DNA template.
Restriction enzyme digestions. To establish the identity of
osteonectin, two restriction enzyme digests and one undigested incubation were performed with each PCR primer
combination product. The intron positions and relative
sizes of exons for human osteonectin have been reported
elsewhere." Any products migrating with higher molecular
sizes would be the result of either genomic DNA, unprocessed nuclear mRNA, or altered transcript amplification.
The sizes for uncut and digested fragments were estimated
by electrophoretic mobilities on 3.5% polyacrylamide gels
(Fig 2, A through C). The bands for uncut and those
liberated by restriction enzyme cleavage were in agreement
with the predicted sizes, and are listed in Table 1, and are
consistent with those for fully processed bone osteonectin
cytoplasmic mRNA.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
OSTEONECTIN MRNA IN MEGAKARYOCYTES
1219
Table 1. Analyses of SaOS-2, Megakaryocyte, and
Nonmegakaryocyte PCR Products
Fragment Size
(bp)
Primer Set*
2
RestrictionEnzymet
Predicted
Estimated*
(-)
453
363,90
300,153
60 1
304,149
359,242
257.233,56
325
269,56
181,144
450
355,86
306,149
600
305,145
350,235
250,234.50
320
270,50
185,145
Hinfl
Pvull
3
(-)
Pvull
EcoRl
Hinf I
4
(-1
Aval
O n1
*See Materials and Methods section and Fig 1 for details on specific
primer sets.
tlnitial PCR products without digestion (-1.
*Based on mobility on 3.5% polyacrylamide gels compared with
commercial size markers (see Fig 2).
In the digestion lanes faint bands migrating as the same
size as uncut bands are detected. These bands may represent undigested osteonectin DNA or DNA other than for
osteonectin. Bands can also be seen in the contamination
control lanes (no. 15 for all primer sets; Fig 2, A through C).
However, no products are seen migrating with the same
mobility as osteonectin. These bands probably represent
nonspecific primer annealing and subsequent amplification
of "primer-dimer'' artifact^.^' Primer-dimer products are
duplex DNA composed of the primers used in PCR. The
contamination controls suggested that the amplified signals
in test reactions were indeed osteonectin cDNA, synthesized from its cellular mRNA template.
Osteonectin cDNA was amplified from all cell types
examined, including Jurkat cells. The significance of this
observation was not pursued further.
Limiting dilution analyses and cDNA PCR. The possibility existed that PCR osteonectin products in the previous
studies were derived from the -5% contaminating nonmegakaryocyte cells. To resolve this issue, limiting dilution
analyses were performed. The enriched megakaryocyte and
nonmegakaryocyte cells as well as a positive cell control
(SaOS-2) were diluted and aliquoted into 18 different tubes
at less than 1 cell per tube. Isolation of RNA, cDNA
synthesis, and PCR was done on the contents of each tube.
Thus, each tube represented an independent trial. Amplification was performed by addition of p-actin and osteonectin primers. Efficiency of cDNA synthesis was monitored by
the addition of p-actin primers. Osteonectin primer set 3
was used for PCR reactions.
To monitor accidental amplification by exogenous P-actin and osteonectin templates, PCR contamination controls
were run. Products were run and separated on 1.2%
agarose gels (Fig 2, D through F). The DNA product sizes
expected for @-actinand osteonectin are 505 bp and 325 bp,
respectively. In some lanes three bands can be visualized:
an upper band migrating as a 517 through 506 bp product, a
344-bp product, and a band migrating as less than 100 bp.
The upper band corresponds to that predicted for @-actin,
the middle band osteonectin, and the lower band is believed
to be primer-dimer. The contamination lane showed only a
band migrating with a size less than 100 bp and is believed
to be primer-dimer. The absence of any bands greater than
100 bp in the contamination control lane dismisses the
possibility that exogenous sources of p-actin and osteonectin DNA were amplified.
The intensities of the osteonectin signals in the SaOS-2
and megakaryocyte gels appears to be greater than those
observed for nonmegakaryocytes. However, in the absence
of further quantitation the assertion that more copies of
osteonectin mRNA are present in SaOS-2 and megakaryocytes is not justified.
Southern hybridizations and Poisson dishibution analyses.
To confirm the identity of osteonectin, Southern hybridizations were performed on all three agarose gels. The
osteonectin probe is internal to the primers used to amplify
osteonectin, to avoid probe binding to artifactual PCR
products resulting from false priming. Hybridization was
specific to the middle 344-bp bands, confirming its identity
as osteonectin DNA (Fig 2, G through I). Table 2 lists the
positive and negative osteonectin PCR results for the three
cell populations examined. Of 18 independent trials for
each cell population, 9 positive reactions for SaOS-2, 7
positive signals for megakaryocyte-enriched cells, and 10
positive reactions for megakaryocyte-depleted cells were
observed. The mean number of cells calculated by the
Poisson distribution was less than 1 for all three cell
populations (Table 3). Assuming an upper limit of 5%
nonmegakaryocyte contamination in PCR reactions from
megakaryocyte samples, the calculated probability that the
product in any one tube is from nonmegakaryocyte cDNAs
was 15%. The possibility that the product for all seven
PCR positive reactions were from nonmegakaryocyte
cDNAs is therefore -1/6.6 x 10'. As the megakaryocyte
contamination in the megakaryocyte-depleted population
was I0.1%, the probability of amplification of megakaryocyte cDNAs in these cases was even lower. Therefore, it can
be concluded statistically that osteonectin mRNA is present
in both the megakaryocyte and nonmegakaryocyte cells
tested.
DISCUSSION
The observations by Stenner et al' that platelets contain
osteonectin, and by Kelm and Mann" that platelet osteonectin differs in molecular size from bone osteonectin, have
raised questions both about its origin and about the basis
for the size difference. Osteonectin may appear in platelets
as a result of endogenous synthesis in the megakaryocyte
cell precursor or endocytosis from plasma into megakaryocytes. Possible explanations for the size difference are
peptide cleavage of bone osteonectin, posttranslational
modification of platelet osteonectin, or differences in the
osteonectin transcript in megakaryocytes. An increased size
of megakaryocyte osteonectin transcript or deletion of
counterpart bone transcript were eliminated as biologic
events producing different sizes of osteonectin protein by
studies reported in this communication involving analyses
of uncut and restriction enzyme digested PCR products.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1220
VILLARREAL, GRANT, AND LONG
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
OSTEONECTIN MRNA IN MEGAKARYOCYTES
1221
Tqble 2. Summary of Limiting Dilution PCR Amplification
Cell Type
SaOS-2
Megakaryocytes
Nonmegakaryocytes
No. Positive*
9
7
10
No. Negative
9
11
8
Total No. of
Amplifications
18
18
18
*Stainable PCR product bands and Southern hybridizations were
used to score positive osteonectin signals (Fig 2, D through F). See the
t e a for details.
Thus, size differences must occur at the posttranscriptional
level. Consistent with our conclusion, Kelm and Mann12
have suggested that the carbohydrate structure of platelet
osteonectin is responsible for its larger apparent molecular
size. Initial studies by Kelm and Mann12 indicate that
platelet osteonectin reacts with the lectin LCA (lens culinaris agglutinin) and is cleaved by treatment with N-Glycanase, suggesting that a complex-type carbohydrate addition
occurs on platelet osteonectin. Bone osteonectin treated
with the glycosidase Endo-H results in a lower molecular
size. Bone osteonectin reacts with concanavalin A. These
results suggest that bone has a high-mannose type carbohydrate addition. Removal of carbohydrate components from
bone and platelet osteonectin by N-Glycanase exposure
results in both species having identical electrophoretic
mobilities on SDS-PAGE. These data suggest that complextype oligosaccharide addition found in platelet osteonectin
and not found in bone osteonectin is at least in part
responsible for its molecular size difference.
As 5% contaminating nonmegakaryocyte cells were
present in the bead-purified human bone marrow megakaryocytes used in these studies it could not be conclusively
stated that osteonectin mRNA was endogenous to megakaryocytes. We were able to overcome our inability to
obtain sufficient numbers of purified megakaryocytes from
a heterogenous cell population by the use of PCR amplification. The numbers of megakaryocytes purified will not yield
sufficient quantities of mRNA for Northern analysis. However, the amplification capabilities of PCR overcomes
problems of low mRNA abundance. PCR technology allows
several million-fold amplification of DNA, thus providing
investigators with a tool to detect and manipulate trace
quantities of nucleic acids. However, any PCR osteonectin
products synthesized may have been amplified from the
5% contaminating nonmegakaryocyte cells. Therefore,
limiting dilution analysis coupled with PCR was used to
determine if osteonectin mRNA resides in megakaryocytes
andlor nonmegakaryocytes. In all cases, including both
-
-
Table 3. Poisson Distribution Analyses
Fraction of Negative
PCR Reactions
Mean No. of
Cell Type
(SI*
Cellflube
(m)t
SaOS-2
Megakaryocytes
Nonmegakaryocytes
0.50
0.39
0.44
0.69
0.49
0.81
*Negative osteonectin PCR reactions were based on absence of
staining PCR bands and failure to bind osteonectin probe in Southern
hybridizations (Fig 2, G through I, and Table 2) and expressed as the
fraction of negative osteonectin PCR signalshotal number of PCR
reactions.
tThe mean number of cells per tube was calculated by equating the
Poisson distribution to the zero term and solving for m, where m =
-In s (see Materials and Methods, Limiting Dilution and Poisson
Distribution Analyses).
megakaryocytes and nonmegakaryocyte cells, cDNA for
osteonectin was amplified. These results strongly suggest
that osteonectin mRNA is present in megakaryocytes and
nonmegakaryocytes. Thus, the potential for endogenous
osteonectin protein synthesis resides in both populations of
cells. Recently, preliminary metabolic studies of osteonectin biosynthesis in megakaryocyte-enriched and megakaryocyte-depleted cells have been reported?' The osteonectin
synthesized by megakaryocyte cells migrated in a manner
identical to that of platelet osteonectin and demonstrated
the same glycosidase sensitivity." In contrast, nonmegakaryocyte osteonectin displayed an electrophoretic mobility and
glycosidase susceptibility similar to that of bone osteonectin.3l The metabolic labeling and PCR studies taken together strongly suggest that platelet osteonectin is synthesized by megakaryocytes, and is not the product of uptake
from plasma.
The protocol we describe in this report is of general value
and it could be used in a variety of other situations to
address issues concerning the presence and sizes of particular mRNAs in heterogeneous populations of cells. In
addition, it would be advantageous in situations where
limited quantities of desired cells can be obtained and
purified.
ACKNOWLEDGMENT
The authors are grateful to John A. Robinson and Laura S.
Oakley for providing megakaryocytes and SaOS 2 and Jurkat cell
cultures. The secretarial assistance of Jean Kenyon is acknowledged.
<
Fig 2. Electrophoresis of PCR products. Panels (A) through (C) refer t o cDNAamplification by primer combinations 2,3, and 4, respectively, run
on 3.5% polyacrylamide gels. The following lane descriptions apply t o panels (A) through (C): DNA markers, lanes I(BRL 1-kb ladder) and 8 (BRL
+Xl74/Hae///I; lane 15, contamination control; lanes 2 through 4. megakaryocyte cDNA; lanes 5 through 7, SaOS-2 cDNA; lanes 9 through 11,
nonmegakaryocyte cDNA; and lanes 12 through 14, Jurkat cDNA. Single letter abbreviations used: U, uncut; A, A d ; E, EcoRI; H,Hinfl; K, O n l ;
and P,pVull. Marker sizes (bp) for I-kb ladder are listed alongside each panel. Panels (D) through (F) present agarose gel electrophoresis, 1.2%. of
PCR amplification products from limiting dilution t o less than 1cell. Panels (D) through (F) refer t o amplification from SaOS-2, megakaryocytes,
and nonmegakaryocytes, respectively. The following lane descriptions apply t o all three panels: lane 1, BRL 1-kb ladder; lanes 2 through 19
represent entire PCR reaction from each cell population sample. Two marker sizes (bp) are listed alongside each panel. In each positive case the
show Southern hybridizations of products
upper and lower bands correspond t o f3-actinand osteonectin cDNAs, respectively. Panels ( G )through (I)
from limiting dilution PCR amplification. Panels ( 0 )through (I)represent gel transfers from SaOS-2, megakaryocyte, and nonmegakaryocyte PCR
correspond t o gel panels (D), (E), and (F), respectively.
reactions, respectively, probed with an osteonectin cDNAfragment. Panels (GI,(HI, and (I)
Top arrow signifies position on gel of top band (f3-actin). middle arrow osteonectin band location, and bottom arrow primer-dimer position.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1222
VILLARREAL, GRANT, AND LONG
REFERENCES
1. Stenner DD, Tracy RP, Riggs BL, Mann KG: Human
platelets contain and secrete osteonectin, a major protein of
mineralized bone. Proc Natl Acad Sci USA 83:6892,1986
2. Sage H, Bomstein P: Characterization of a novel serum
albumin-binding glycoprotein secreted by endothelial cells in
culture. J Biol Chem 259:3993,1984
3. Mason IJ, Taylor A, Williams JG, Sage H, Hogan BLM:
Evidence from molecular cloning that SPARC, a major product of
mouse embryo parietal endoderm, is related to an endothelial cell
“culture shock” glycoprotein of Mr 43,000. EMBO J 5:1465,1986
4. Dziadek M, Paulsson M, Aumailley M, Timpl R: Purification
and tissue distribution of a small protein (BM-40) extracted from a
basement membrane tumor. Eur J Biochem 161:455,1986
5. Mosseson MW, Amrani D L The structure and biologic
activities of plasma fibronectin. Blood 56:154,1980
6. Handagama P, Rappolee DA, Werb 2, Levin J, Barnton DM:
Platelet a-granule fibrinogen, albumin, and immunoglobulin G are
not synthesized by rat and mouse megakaryocytes. J Clin Invest
86:1364,1990
7. Schmaier AH, Zuckerberg A, Silverman C, Kuchibholta J,
Tuszynski GP, Colman R W High-molecular weight kininogen. A
secreted platelet protein. J Clin Invest 71:1477,1983
8. Levine SP: Secreted platelet proteins as markers for pathological disorders, in Phillips DR, Schuman MA (eds): Biochemistry of
Platelets. San Diego, CA, Academic, 1986, p 377
9. Ryo R, Nakoff A, Huang SS, Ginsberg MH, Deuel T F New
synthesis of a platelet-specific protein: Platelet factor 4 synthesis in
a megakaryocyte-enriched rabbit bone marrow culture system. J
Cell Biol96:515,1983
10. Lawler J: The structural and functional properties of thrombospondin. Blood 67:1197, 1986
11. Kelm RJ Jr, Mann KG: Human platelet osteonectin: Release, surface expression, and partial characterization. Blood
75:1105,1990
12. Kelm RJ Jr, Mann KG: The collagen binding specificity of
bone and platelet osteonectin is related to differences in glycosylation. J Biol Chem 266:9632,1991
13. Termine JD, Kleinman HK, Whitson SW, Conn KM, McGarvey ML, Martin GR: Osteonectin, a bone-specific protein
linking mineral to collagen. Cell 26:99,1981
14. Tung PS, Domenicucci C, Wasi S, Sodek J: Specific immunohistochemical localization of osteonectin and collagen types I and
111in fetal and adult porcine dental tissues. J Histochem Cytochem
33:531, 1985
15. Mann KR, Deutzmann M, Paulsson M, Timpl R: Solubilization of protein BM-40 from a basement membrane tumor with
chelating agents and evidence for its identity with osteonectin and
SPARC. FEBS Lett 218:167,1987
16. Sage H, Vernon RB, Funk SE, Everitt EA, Angello: SPARC,
a secreted protein associated with cellular proliferation, inhibits
cell spreading in vitro and exhibits Ca’+-dependent binding to the
extracellular matrix. J Cell Biol 109341,1989
17. Marguerie GA, Plow EF, Edgington TS: Human platelets
possess an inducible and saturable receptor specific for fibrinogen.
J Biol Chem 254:5357,1979
18. Wolff R, Plow EF, Ginsberg MH: Interaction of thrombospondin with resting and stimulated platelets. J Biol Chem 261:
6840,1986
19. Fugimoto T, Ohara S, Hawiger J: Thrombin induced exposure and prostacyclin inhibition of the receptor for factor VIIIhon
Willebrand factor on human platelets. J Clin Invest 254:5357,1979
20. Parker RI, Gralnick HR: Identification of platelet glycoprotein IIbIIIa as the major binding site for released platelet von
Willebrand factor. Blood 68:732,1986
21. Clezardin P, Malaval L, Ehrensperger A-S, Delmas PD,
Dechavanne M, McGregor J L Complex formation of human
thrombospondin with osteonectin. Eur J Biochem 175:275,1988
22. Grant BW, Nichols WL, Solberg LA, Yachimiak DJ, Mann
KG: Quantitation of human in vitro megakaryocytopoeisis by
radioimmunoassay. Blood 69:1334,1987
23. Chomczynski P, Sacchi N: Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162156,1987
24. Villarreal XC, Mann KG, Long GL: Structure of human
osteonectin based upon the analysis of cDNA and genomic
sequences. Biochemistry 28:6483,1989
25. Ponte P, Sun-Yu N, Engel J, Gunning P, Kedes L Evolutionary conservation in the untranslated regions of actin mRNA’s:
DNA sequence of a human beta-actin cDNA. Nucleic Acids Res
12:1687,1984
26. Maxam AM, Gilbert W Sequencing end-labeled DNA with
base-specific chemical cleavages. Methods Enzymol65:499,1980
27. Smith GE, Summers MD: The bi-directional transfer of
DNA and RNA to nitrocellulose or diazobenzyloxymethyl paper.
Anal Biochem 109:123,1980
28. Devereux J, Herbil P, Smithies 0: A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Res
12:387, 1984
29. Wasi S, Otsuka K, Yao K-L, Tung PS, Aubin JE, Sodek J,
Termine JD: An osteonectin-like protein in porcine periodontal
ligament and its synthesis by periodontal fibroblasts. Can J Biochem Cell Biol62:470,1984
30. Gelfand DH, White TJ: Thermostable DNA polymerases, in
Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds): PCR Protocols. A Guide to Methods and Applications. Academic, San Diego,
CA, 1990, p 129
31. Hair GA, Kelm RJ Jr, Mann KG, Grant BW: Megakaryocytes synthesize platelet osteonectin. Blood 76:1822a, 1990 (abstr)
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1991 78: 1216-1222
Demonstration of osteonectin mRNA in megakaryocytes: the use of
the polymerase chain reaction
XC Villarreal, BW Grant and GL Long
Updated information and services can be found at:
http://www.bloodjournal.org/content/78/5/1216.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.