|
|
 Previous Article | Table of Contents | Next Article 
Blood, Vol. 93 No. 12 (June 15), 1999:
pp. 4167-4178
Expression of p21Cip1/Waf1/Sdi1 and p27Kip1
Cyclin-Dependent Kinase Inhibitors During Human Hematopoiesis
By
Toshiyasu Taniguchi,
Hisako Endo,
Norio Chikatsu,
Kaoru Uchimaru,
Shigetaka Asano,
Toshiro Fujita,
Tatsutoshi Nakahata, and
Toru Motokura
From the Fourth Department of Internal Medicine, the Department of
Pathology, the Branch Hospital, School of Medicine, and the Departments
of Hematology/Oncology and Clinical Oncology, Institute of Medical
Science, University of Tokyo, Tokyo, Japan.
 |
ABSTRACT |
Expression of p21 and p27 cyclin-dependent kinase inhibitors is
associated with induced differentiation and cell-cycle arrest in some
hematopoietic cell lines. However, it is not clear how these inhibitors
are expressed during normal hematopoiesis. We examined various human
hematopoietic colonies derived from cord blood CD34+
cells, bone marrow, and peripheral blood cells using a quantitative reverse transcription-polymerase chain reaction assay, immunochemistry, and/or Western blot analysis. p21 mRNA was expressed increasingly over
time in all of the colonies examined (granulocytes, macrophages, megakaryocytes, and erythroblasts), whereas p27 mRNA levels remained low, except for erythroid bursts. Erythroid bursts expressed both p21
and p27 mRNAs with differentiation but expressed neither protein, whereas both proteins were expressed in megakaryocytes and peripheral blood monocytes. In bone marrow, p21 was immunostained almost exclusively in a subset of megakaryocytes and p27 protein was present
in megakaryocytes, plasma cells, and endothelial cells. In
megakaryocytes, reciprocal expression of p27 to Ki-67 was evident and
an inverse relationship between p21 and Ki-67 positivities was also
present, albeit less obvious. These observations suggest that a complex
lineage-specific regulation is involved in p21 and p27 expression and
that these inhibitors are involved in cell-cycle exit in megakaryocytes.
© 1999 by The American Society of Hematology.
| |
INTRODUCTION |
EUKARYOTIC CELL-CYCLE progression is
regulated by the cyclin-dependent kinases (CDKs), the activities of
which are controlled by multiple mechanisms such as cyclin binding, CDK phosphorylation and dephosphorylation, and binding of CDK inhibitors (CKIs).1 Of the two known families of CKIs, one is the
Cip/Kip family, consisting of p21Cip1/Waf1/Sdi1 (p21),
p27Kip1 (p27), and p57Kip2, and the other is
the INK4 family, which consists of p16INK4a,
p15INK4b, p18INK4c, and
p19INK4d.2
Among these CKIs, p21 and p27 are implicated in terminal
differentiation and proliferation in nonhematopoietic and hematopoietic cells. For example, p21 expression is induced by MyoD during skeletal muscle differentiation,3 and p21 expression in mouse embryo correlates with terminal differentiation of skeletal muscle, cartilage, skin, and nasal epithelium.4 p27 protein is expressed in a transient wave in developing myotomes of the mouse embryo.5 p27 protein progressively accumulates in oligodendrocyte precursor cells as they proliferate and is present at high levels in
oligodendrocytes.6 In the gastrointestinal tract, p21
expression and proliferation are inversely related.7,8 p27
is deeply involved in regulating normal T- and B-lymphocyte
proliferation.9-11 There is a mutually exclusive pattern of
staining for Ki-67 and p27 in human reactive lymphatic
tissues,12,13 whereas p21 is positive only in a few lymphoid cells in lymph nodes.7
In addition to the role as CKIs, p21 and p27 have functions of
promoting the association of CDK4 (or CDK6) with the D-type cyclins and
targeting CDK4 and cyclin D1 to the nucleus.14
p21-containing cyclin/CDK complexes exist in both catalytically active
and inactive forms.15 Monomeric p21 and p27 promotes the
assembly of active kinase complexes and, at higher stoichiometric
ratios, they inhibit CDK activity.14,16,17
Hematopoiesis is a process characterized by proliferation and
differentiation of cells derived from a small number of hematopoietic stem cells.18 Terminal differentiation is often associated
with loss of proliferation potentials and the final exit from cell cycle. p21 expression is triggered by multiple differentiation-inducing agents in various hematopoietic cell lines: HL-60,19-24
K562,20,21 U937,20,21,25 CMK,26
UT-7,27 and MEG-01s.28 p27 expression is also
induced by differentiation-inducing agents in some hematopoietic cell
lines: HL-60,23 U937,29 and
MEG-01s.28 However, cell-cycle controls can be
abnormal in cell lines as the result of cellular transformation
mechanisms. Information on p21 and p27 expression in normal
hematopoiesis is scanty30 and how p21 and p27 are involved
in normal hematopoiesis remains to be clarified.
In the present study, we directed our attention to expression patterns
of p21 and p27 in normal (primary) hematopoietic cells and terminally
differentiated blood cells. The correlation of cell proliferation with
p21 and/or p27 expression was given focus and the expression of Ki-67
antigen was also examined. We obtained evidence that the expression of
p21 and p27 is regulated in a lineage-specific manner and that both
proteins are highly expressed, particularly in megakaryocytes. In
megakaryocytes, p27 and probably p21 proteins are involved in terminal
exit from the cell cycle.
| |
MATERIALS AND METHODS |
Cytokines.
Recombinant human granulocyte colony-stimulating factor (G-CSF) was
provided by Chugai Pharmaceutical Co Ltd (Tokyo, Japan), recombinant
human interleukin-6 (IL-6) was from Ajinomoto Co Ltd (Kawasaki, Japan),
recombinant human stem cell factor (SCF) and recombinant human IL-3
were from Amgen Biologicals (Thousand Oaks, CA), and recombinant human
erythropoietin (EPO) and thrombopoietin (TPO) were provided by Kirin
Brewery Co Ltd (Tokyo, Japan).
Antibodies.
Antibodies used for immunostaining were as follows: anti-p21 (6B6;
Pharmingen, San Diego, CA), anti-p27 (F-8; Santa Cruz Biotechnology, Inc, Santa Cruz, CA), anti-p53 (DO-7; DAKOPATTS A/S, Glostrup, Denmark), and antihuman CD41 (5B12; DAKOPATTS A/S) monoclonal antibodies (MoAbs), which were used at 1:50 dilution. Another anti-p27
MoAb (G173-524; Pharmingen) was used at 1:400 dilution. Antihuman
glycophorin A (JC159, DAKOPATTS A/S) and anti-Ki-67 (MIB-1; Immunotech,
Marseille, France) MoAbs were used at 1:100 dilution. Rabbit antihuman
von Willebrand factor (vWF; A082; DAKOPATTS A/S) was used at 1:3,200
dilution. Rabbit antihuman  and  light chains (A191 and A193;
DAKOPATTS A/S) was used at 1:25,000 dilution. Anti-p21 (6B6) and
anti-p27 (G173-524) MoAbs, at 1:250 dilution, were also used in Western
blot analysis.
CD34+ cell preparation.
Human umbilical cord blood was obtained during normal full-term
deliveries, after the acquisition of written informed consent. Mononuclear cells (MNCs) were separated by Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden) density gradient centrifugation after depletion of phagocytes with silica (Immuno-Biological Laboratories Co,
Ltd, Fujioka, Japan).31 CD34+ cells were
purified from the MNCs, using Dynabeads M-450 CD34 and DETACHaBEAD CD34
(Dynal, Oslo, Norway). Flow cytometric analysis showed that 85% to
95% of the cells separated were CD34+.
Clonal cell culture.
For erythroid bursts, granulocyte colonies, and macrophage colonies,
purified CD34+ cells were incubated at concentrations of
250 cells/mL in methylcellulose culture, as reported.32,33
In brief, 1 mL of culture mixture contained cells,  -medium (GIBCO
BRL Life Technologies, Inc, Grand Island, NY), 0.9% methylcellulose
(Shinetsu Chemical, Tokyo, Japan), 30% heat-inactivated fetal bovine
serum (FBS; HyClone, Logan, UT), 1% crystallized and deionized
fraction V bovine serum albumin (BSA; Sigma, St Louis, MO), 0.05 mmol/L
2-mercaptoethanol (Wako Pure Chemical Industries, Osaka, Japan), 100 ng/mL SCF, 20 ng/mL IL-3, 80 ng/mL IL-6, 10 ng/mL G-CSF, and 2 U/mL
EPO. For megakaryocyte colonies, 1 mL of culture mixture contained
1,000 cells, -medium, 0.9% methylcellulose, 30% platelet-poor
plasma from a healthy adult volunteer, 1% BSA, 0.05 mmol/L
2-mercaptoethanol, 100 ng/mL SCF, and 5 ng/mL TPO. Individual colonies,
lifted under direct microscopic visualization, were suspended in 200 µL of phosphate-buffered saline (PBS) with 30% FBS. Half of the cell
suspension was spun in a cytocentrifuge (Cytospin 2; Shandon Southern
Instruments, Sewickley, PA) at 500 rpm for 5 minutes and processed for
May-Grünwald Giemsa staining. The other half was directly
subjected to RNA extraction.
Suspension culture.
One milliliter of culture mixture containing 10,000 to 15,000 purified
CD34+ cells, -medium (GIBCO BRL), 20% heat-inactivated
FBS (HyClone), 1% BSA, with or without 100 ng/mL SCF and 80 ng/mL IL-6
was incubated in a 24-well or 48-well tissue culture plate (Becton
Dickinson Labware, Lincoln Park, NJ) at 37°C in a humidified
atmosphere with 5% CO2.
Preparation of normal human peripheral blood (PB) monocytes,
lymphocytes, and granulocytes and bone marrow (BM) MNCs.
PB was obtained from healthy adult volunteers with informed consent.
BM aspirates from patients with non-Hodgkin's lymphoma, without BM involvement, were obtained after the acquisition of informed
consent. MNCs of PB or BM were prepared by Ficoll-Paque density
gradient centrifugation. PB MNCs were suspended in RPMI1640 medium
(GIBCO BRL) supplemented with 10% of heat-inactivated FBS (BioWhittaker, Walkersville, MD) and 60 mg/L kanamycin (Meiji Seika
Kaisha, Ltd, Tokyo, Japan) and incubated in an MSP plate (Japan
Immunoresearch Laboratories, Co, Ltd, Takasaki, Japan) at 37°C in a
humidified atmosphere with 5% CO2 for 1 hour. Nonadherent cells were used as lymphocytes. After four PBS washes, the plate was
incubated at 4°C for 30 minutes with ice-cold MSP-E buffer supplied
by the manufacturer. The adherent cells were detached by pipetting and
used as monocytes. Cytochemically, the lymphocytes were 95% to 96%
pure and monocytes were 88% to 89% pure (data not shown). PB
granulocytes were prepared from the pellet of Ficoll-Paque density
gradient centrifugation. Cells in the pellet were suspended in 3%
dextran/PBS and left standing for 30 minutes at room temperature for
erythrocyte sedimentation. Cells in the supernatant were resuspended in
0.2% NaCl for 1 minute on ice for hemolysis, followed by the addition
of 1.6% NaCl. After pelleting, cells were collected in PBS.
Granulocytes usually accounted for greater than 95%.
Cell lines.
A human leukemic cell line of pre-B phenotype, Nalm-6,34
was a gift from Dr T. Nakamura (First Department of Internal Medicine, University of Tokyo, Tokyo, Japan). A human megakaryoblastic leukemia cell line, MEG-01s, was a gift from Drs M. Ogura (Aichi Cancer Center,
Nagoya, Japan) and H. Saito (First Department of Internal Medicine,
Nagoya University, Nagoya, Japan).35 Cells were passaged in
RPMI1640 medium supplemented with 10% FBS (BioWhittaker) and 60 mg/L
kanamycin at 37°C in a humidified atmosphere with 5%
CO2.
RNA preparation and reverse transcription (RT).
Cells were directly subjected to the acid guanidinium
thiocyanate-phenol-chloroform (AGPC) method36 with 20 µg
yeast tRNA (GIBCO BRL) as a carrier and the extracted RNA was suspended
in 30 µL diethyl pyrocarbonate (Sigma)-treated water, boiled for 1 minute, and stored at  80°C until analysis. One sixth of RNA (5 µL) was placed in 20 µL of 1× RT buffer (10 mmol/L
Tris-HCl, 50 mmol/L KCl, 4 mmol/L MgCl2, 10 mmol/L
dithiothreitol, pH 8.3) with 250 µmol/L of each deoxyribonucleoside
triphosphate (dNTP), 1 µmol/L oligo(dT)15 (Boehringer
Mannheim, Tokyo, Japan), 20 U rRNasin ribonuclease inhibitor (Promega
Co, Madison, WI), 200 U Moloney murine leukemia virus
(M-MLV) reverse transcriptase (GIBCO BRL) and was
incubated at 37°C for 30 minutes. Total RNAs of the cell lines,
Nalm-6 and MEG-01s, and normal BM MNCs were extracted by the AGPC
method without yeast tRNA, and cDNA was synthesized with
oligo(dT)15 from total RNA, as described.37
Quantitative reverse transcription-polymerase chain reaction
(RT-PCR) assay.
Primers used are listed in Table 1.
Schematic presentation of design of the PCR primers is depicted in
Fig 1A. Expected sizes of the PCR products
are as follows: p21, 369 bp; p27, 469 bp; and  -actin, 640 bp. The
primers were synthesized by Greiner Japan (Tokyo, Japan). An aliquot (1 µL) of cDNA was placed in 20 µL of 1× PCR buffer (10 mmol/L
Tris-HCl, 50 mmol/L KCl, 1.5 mmol/L MgCl2, pH 8.3) with 200 µmol/L of each dNTP, 2 µCi [-32P]dCTP, primers
(p21AS, p27AS, S, and p27S at 200 nmol/L and ASp27S at 2 nmol/L),
and 0.5 U recombinant Taq DNA polymerase (Takara, Kyoto,
Japan). Reaction parameters were 94°C for 1 minute (first cycle, 5 minutes), 62°C for 2 minutes, and 72°C for 3 minutes (last
cycle, 10 minutes). After the indicated cycles were performed, 5 µL
of each PCR product was separated on a 4.5% polyacrylamide gel
followed by autoradiography. An optical scanner was used and densitometrical analysis was made using NIH Image software (NIH, Bethesda, MD). As negative controls, water instead of cDNA
or the products of the RT reactions without reverse transcriptase were
subjected to PCR and we confirmed no false-positive reaction.
View larger version (31K):
[in this window]
[in a new window]
| Fig 1.
Quantitative RT-PCR for relative expression levels of p21
and p27 with endogenously expressed -actin used as an internal
control. (A) Schematic presentation of primer setting on p21, p27, and
-actin sequences. Thick lines indicate coding regions and thin lines
represent truncated noncoding regions. Thick arrows indicate primers
used in the PCR. The common primer (p27S) is derived from the identical
region between p21 and p27 sequences, except for one mismatch in the
p21 sequence. The composite primer (ASp27S) consisted of p27S in the
5' part and -actin sequence in the 3' part. The common
primer (p27S) is shared in amplification of the three PCR products,
whereas p21AS, p27AS, and S are specific to p21, p27, and -actin
sequences, respectively. (B) Autoradiographs showing kinetics of the
PCR. An aliquot (5 µL, 0.1 µg RNA equivalent) of cDNA synthesized
from Nalm-6 cells RNA was placed in 100 µL of PCR reaction volume
with [ -32P]dCTP (10 µCi) and was subjected to PCR as
described in Materials and Methods. An aliquot (5 µL) of the reaction
mixture was taken at the indicated cycle. Arrows indicate the PCR
products corresponding to p21, p27, and -actin. (C)
[-32P]dCTP incorporation was plotted on a logarithmic
scale over number of cycles of PCR. Each symbol denotes the following:
( ) -actin; ( ) p27; ( ) p21. (D) Comparison between Northern
analysis and RT-PCR. RNAs extracted from MEG-01s cells stimulated with
TPA at 10 nmol/L were subjected to RT-PCR and to Northern analysis. The
gel for PCR products was dried and exposed to an x-ray film for 17 hours at 80°C with an intensifying screen. The membrane for
Northern analysis of -actin and p21 was exposed to x-ray films at
80°C with intensifying screens for 1 day and 7 days,
respectively. p27 mRNA was hardly detected by Northern analysis (data
not shown). (E) The amount of p21 transcripts standardized to -actin
determined by Northern analysis and by RT-PCR. The signal ratio of p21
over -actin at time 0 hour was defined as 1 U and the amounts of p21
transcripts standardized to -actin are plotted on a logarithmic
scale over time. Each symbol denotes the following: () Northern
analysis; () RT-PCR.
|
|
Northern analysis.
RNAs extracted from MEG-01s cells stimulated with
12-o-tetradecanoylphorbol-13-acetate (TPA; Sigma) at 10 nmol/L
were subjected to Northern analysis, as described.28
Immunocytochemical study.
For immunostaining of p21, p27, and Ki-67, cytosmears were fixed with
10% formaldehyde in PBS at room temperature for 10 minutes. After a
wash in PBS, the cytosmears were treated with 100% methanol for 10 minutes at 20°C and washed twice with Tris-buffered saline (TBS). For immunostaining of CD41 and glycophorin A, cytosmears were
air-dried for 2 hours, fixed with 100% acetone at 4°C for 1 minute, and washed with TBS twice. Primary antibody in 3% BSA/TBS was
incubated overnight at 4°C in a humidified chamber. After washing
with TBS, cells were stained with universal DAKO APAAP kit (DAKOPATTS
A/S), as described by the manufacturer, and were counterstained with hematoxylin.
Western blot analysis.
Cells were lyzed with 1× sample buffer (60 mmol/L Tris-HCl, 2%
sodium dodecyl sulfate [SDS], 0.1 mol/L dithiothreitol, pH 6.8) and
boiled for 5 minutes. Protein concentration was determined by
spectrophotometry using BCA Protein Assay Reagent (Pierce, Rockford,
IL).38 A total of 30 µg protein per lane was loaded on
15% SDS-polyacrylamide gel and subjected to Western blot analysis, as
described.28
Immunohistochemical study.
Human BM aspirates were retrieved from the archives of the Department
of Pathology, the Branch Hospital, University of Tokyo, School of
Medicine. These materials had been fixed in formalin and embedded in
paraffin using standard methods. Three-micrometer-thick sections were
dried on MAS-coated glass slides (Matsunami Glass Ind, Ltd, Kishiwada,
Japan), deparaffinized with xylene, and soaked in PBS. Slides were
transferred to citrate buffer (pH 6.0) and heated in a microwave oven 5 times for 5 minutes. After cooling at room temperature for 1 hour and
two washes in TBS, the slides were incubated overnight at 4°C with
the indicated primary antibody diluted in 3% BSA/TBS in a humidified
chamber. After washing with TBS, staining with universal DAKO APAAP kit
was performed as described by the manufacturer. Nuclei were
counterstained with hematoxylin. For negative controls, mouse IgG was
used. For p27, a human reactive lymph node sample served as a positive control.
Sequential double immunohistochemical staining.
Sequential double immunohistochemical staining was performed using the
alkaline-phosphatase/anti-alkaline-phosphatase (APAAP) procedure,
followed by the avidin-biotinylated peroxidase complex (ABC) method.
After the above-mentioned antigen retrieval, the slides were incubated
for 10 minutes in a 3% H2O2 solution in PBS
for blocking endogenous peroxidase. After the APAAP procedure, the
slides were treated three times for 30 minutes with 0.1 mol/L glycine-HCl buffer (pH 2.2). After three washes in PBS, the slides were
covered with normal serum contained in a Vectastain ABC kit (Vector
Laboratories, Burlingame, CA) for 30 minutes at room temperature and
then incubated overnight at 4°C with the indicated antibody in a
humidified chamber. The sections were stained using the Vectastain ABC
kit, as described by the manufacturer. Peroxidase activity was
visualized by applying diaminobenzidine chromogen, containing 0.015%
H2O2. For negative controls, mouse IgG served
as a substitute for the primary antibody or the primary antibody was omitted.
| |
RESULTS |
Quantitative RT-PCR.
To monitor p21 and p27 mRNAs expression levels in a small amount of
material such as a colony generated from a CD34+ cell, we
first developed a quantitative RT-PCR assay for relative expression
levels of p21 and p27 mRNAs with endogenously expressed -actin used
as an internal control. To circumvent difficulties in conventional
competitive RT-PCR assays, coamplification of targets (p21 and p27) and
control (-actin) in a tube was designed (Fig 1A). Simple
coamplification of control and target DNAs can correct for variation in
tube-to-tube amplification efficiency if the data are obtained during
the exponential phase of the PCR reaction. However, -actin is
expressed at a level much higher than those of the target genes and
comparable coamplification can be difficult because of rapid saturation
of control reactions (data not shown). Therefore, we designed a PCR
using a composite primer at a low concentration to reduce the -actin
signals at a constant ratio to the level comparable to the target signals.
We designed a common primer, p27S, derived from a homologous sequence
between p21 and p27 and a composite primer, ASp27S, the 3'
part of which is derived from the sequence of -actin and the
5' part of which is the same as the common primer,
p27S. The ASp27S primer is used in the amplification of -actin
cDNA, but not p21 or p27, because there is a large region of
-actin-specific sequence in the 3' part. The three specific
primers for p21, p27, and -actin are p21AS, p27AS, and S,
respectively. Amplification of -actin cDNA must be initiated with
S and ASp27S, and then p27S takes the place of ASp27S, because
we use ASp27S at a lower concentration. Final products
for -actin were generated by virtue of p27S and S. We confirmed
the identity of the PCR products by direct sequencing (data not shown).
We found exponential amplification phases of three genes overlapped to
26 cycles with a cDNA from cell lines used as a template (Fig
1B and C) and to 32 cycles with a cDNA from cells of colonies and
PB used as a template (data not shown). Therefore, PCR was run for 21 cycles for analysis of cell lines and for 30 cycles for cells from
colonies and PB.
To further confirm the quantitative nature of the PCR, we compared the
RT-PCR assay with Northern analysis using RNAs extracted from MEG-01s
cells stimulated with TPA. The addition of 10 nmol/L TPA upregulated
p21 mRNA expression and induced cell cycle arrest at the G1-S boundary
in MEG-01s cells.28 As shown in Fig 1D and E, the two
methods gave similar results and the quantitative nature of the PCR was reconfirmed.
p21 and p27 expressions in colonies generated from human cord blood
CD34+ cells.
We monitored p21 and p27 mRNA expression levels in various colonies
derived from CD34+ cells using the above-mentioned
quantitative RT-PCR method. A total of 116 colonies were analyzed.
Figure 2 shows the time course of relative
expression levels of p21 and p27 mRNAs. In all four lineages
(granulocytes, macrophages, megakaryocytes, and erythroblasts), p21
mRNA expression increased over time up to day 15. The increased levels
of p21 mRNA in erythroid bursts and megakaryocyte colonies were similar
to those observed in differentiated MEG-01s cells. p27 mRNA levels
remained low during the observed period in all lineages except
erythroid bursts. The gradual increase of p27 mRNA up to day 15 was
evident only in erythroid bursts.
View larger version (35K):
[in this window]
[in a new window]
| Fig 2.
p21 and p27 mRNA expression during clonal culture of
CD34+ cells. Human umbilical cord CD34+
cells were incubated in methylcellulose culture containing cytokines or
in suspension culture as described in Materials and Methods. From day 9 to 18, the indicated colonies were harvested for RNA extraction and
cytosmeared for lineage confirmation (see Table 2). Purified
CD34+ cells and cells in suspension culture of
CD34+ cells were harvested for RNA extraction. RT-PCR was
performed as described in Materials and Methods and each PCR product
was separated on a 4.5% polyacrylamide gel followed by
autoradiography. (A through E) Autoradiographs of RT-PCR products.
Arrows indicate the PCR products corresponding to p21, p27, and
-actin. (F through J) p21 and p27 mRNA expression levels in the
indicated colonies are standardized to -actin determined by
densitometry and are plotted over time in culture. The mean ± SE
(bar) is shown. () p21; () p27. Numbers of colonies analyzed are
denoted in parentheses. Seven cord blood samples were used for these
studies.
|
|
We also examined p21 and p27 mRNA levels in purified CD34+
cells. The p27 mRNA levels were low, but these cells expressed
unexpectedly high levels of p21 mRNA that were decreased after 12 hours
of incubation in suspension culture with or without additional
cytokines (Fig 2E and J). Because the elevated levels of p21 mRNA might be related to the purification process, whether they reflect in vivo
levels remains to be investigated.
We next immunocytochemically examined p21 and p27 protein expression in
these colonies on day 15 of the culture from 7 and 6 subjects,
respectively. We confirmed cell lineages by May-Grünwald Giemsa
staining and immunostaining with anti-CD41 and anti-glycophorin A
antibodies (Table 2). As shown in
Fig 3 and Table 2, some megakaryocytes and
macrophages were positive for p21, although there were few p21-positive
erythroblasts and granulocytes. p21 staining was always restricted to
nuclei and with none in the cytoplasm. Some megakaryocytes were
positive for p27 (Fig 3), although there were few p27-positive
erythroblasts, macrophages, and granulocytes. Therefore, the proteins
expression did not necessarily correlate with their mRNA levels.
View larger version (79K):
[in this window]
[in a new window]
| Fig 3.
p21 and p27 expression in cells of colonies. Cells of
colonies on day 15 of the culture generated from human cord blood
CD34+ cells were immunostained using anti-p21 antibody,
6B6 (A through E), and anti-p27 antibody, F-8 (F through J).
Immunostaining was performed using the APAAP method as described in
Materials and Methods. Nuclei were counterstained with hematoxylin.
Positive cells showed red nuclear staining. (A and F) megakaryocytes,
(B and G) erythroblasts, (C and H) macrophages, (D and I) granulocytes,
and (E and J) MEG-01s cells harvested 48 hours after the addition of 10 nmol/L TPA (positive control; original magnifications: [A] and [F],
×200; [B] through [E] and [G] through [J], ×50).
|
|
p21 and p27 expressions in PB and BM MNCs.
To confirm the correlation of cells in the colonies and PB cells, we
examined PB cells of each lineage by RT-PCR and Western blot analysis
using PBs from 4 normal individuals (Fig
4). Lymphocytes expressed both p21 and p27 mRNAs, but only a
substantial amount of p27 protein as reported by other
workers.9,39 Monocytes expressed mainly p21 mRNA and much
less p27 mRNA (Fig 4A) and expressed both proteins, albeit weakly
(Fig 4B). Consistent with the results of Western analysis,
immunocytochemistry showed that both proteins were present in monocytes
and that PB lymphocytes expressed only p27 protein (Fig 4D through F
and I through K and Table 3). Both proteins
were detected in nuclei. In PB granulocytes, p21 or p27 was not
detected immunocytochemically (Fig 4G and L).
View larger version (69K):
[in this window]
[in a new window]
| Fig 4.
p21 and p27 expression in PB cells. (A)
Autoradiographs of RT-PCR products. Normal BM MNCs (lane 1), PB MNCs
(lane 2), lymphocytes (lane 3), and monocytes (lane 4) were analyzed
for p21 and p27 mRNA expression by RT-PCR. Arrows indicate the PCR
products corresponding to p21, p27, and -actin. (B) Western blot
analysis. Each lane was loaded with 30 µg of total protein from
MEG-01s cells harvested 48 hours after 10 nmol/L TPA addition (positive
control, lane 1), PB MNCs (lane 2), lymphocytes (lane 3), and monocytes
(lane 4). Western blot analyses with anti-p21 (6B6) and anti-p27
(G173-524) antibodies were performed as described in Materials and
Methods. Data were confirmed in 3 more independent experiments. Cells
were immunostained using anti-p21 antibody, 6B6 (C through G), and
anti-p27 antibody, F-8 (H through L). Immunostaining was performed as
described in Materials and Methods using the APAAP method. Nuclei were
counterstained with hematoxylin. Positive cells showed red nuclear
staining. (C and H) MEG-01s cells harvested 24 hours after the addition
of 10 nmol/L TPA (positive control), (D and I) PB MNCs, (E and J)
lymphocytes, (F and K) monocytes, and (G and L) granulocytes (original
magnifications: [C] and [H], ×50; [D] through [G] and [I]
through [L], ×200).
|
|
Immunohistochemistry of human BM.
Finally, we immunohistochemically examined p21 and p27 expression in in
vivo BM cells from 6 different individuals. Typical results of BM are
shown in Figs 5,
6, and 7.
Because of rarity of megakaryocytes in normal BM, we also used BM from
essential thrombocythemia (ET) patients for analysis of megakaryocytes. p21 immunoreactivity was present in the nuclei of a subset of megakaryocytes (Figs 5A and D and 6A). Most cells comprising BM, namely
erythroid and myeloid cells, did not stain for p21 antigen. According
to double staining using anti-p21 and anti-vWF antibodies, p21
positivity in megakaryocytes in 2 ET patients (31.4% to 42.4%) was
similar to 32.0% to 44.7% (n = 2) in normal BM. When we
immunohistochemically investigated p53 protein expression in normal BM,
we found that normal BM cells, including megakaryocytes, were p53
negative (data not shown). Therefore, p21 may be expressed in
megakaryocytes through p53-independent mechanisms, as noted in other
cell types.4,19-22,24,40
View larger version (102K):
[in this window]
[in a new window]
| Fig 5.
p21, p27, and Ki-67 staining of BM. Normal BM
(A, B, and C) and BM from an ET patient (D, E, and F) was immunostained
using anti-p21 antibody, 6B6 (A and D); anti-p27 antibody, G173-524 (B
and E); and anti-Ki-67 antibody, MIB1 (C and F). Immunostaining was
performed using the APAAP method, as described in Materials and
Methods. Positive cells showed red nuclear staining. Although weak
cytoplasmic staining in some megakaryocytes was visible when using the
MoAb G173-524, it was not reproducible with another antibody F-8
(original magnification × 50).
|
|
View larger version (49K):
[in this window]
[in a new window]
| Fig 6.
p21 and Ki-67 expression in megakaryocytes. Serial
sections of BM from an ET patient were immunostained for p21 (A) and
Ki-67 (B). Positive cells showed red nuclear staining. Arrowheads
indicate megakaryocytes: ( ) strongly positive; ( ) moderately
positive; ( ) weakly positive; ( ) negative (original magnification × 50). (C) Percentage of Ki-67 negative () and weakly (+),
moderately (+), and strongly (++) positive megakaryocytes
subdivided according to p21 positivity ([] none, [+] weak,
[+] moderate, [++] strong). Numbers of counted cells are
shown in parentheses.
|
|
View larger version (48K):
[in this window]
[in a new window]
| Fig 7.
p27 and Ki-67 expression in megakaryocytes. Serial
sections of BM from an ET patient were immunostained for p27 (A) and
Ki-67 (B). Positive cells showed red nuclear staining. Arrowheads
indicate megakaryocytes: () strongly positive; () moderately
positive; () weakly positive; () negative (original magnification × 50). (C) Percentage of Ki-67 negative () and weakly (+),
moderately (+), and strongly (++) positive megakaryocytes
subdivided according to p27 positivity ([] none, [+] weak,
[+] moderate). There was no p27 strongly positive megakaryocyte.
Numbers of counted cells are shown in parentheses.
|
|
Weak to moderate p27 immunoreactivity was evident in nuclei of a subset
of megakaryocytes (Figs 5B and E and 7A). Weak cytoplasmic staining in
some megakaryocytes was also visible when using the MoAb, G173-524.
However, it was not reproducible with another antibody, F-8. Although
it is not clear whether the cytoplasmic staining was specific, there
might be a significance of cytoplasmic localization of p27
protein.41 Double staining using anti-p27 and anti-vWF
antibodies (data not shown) showed that approximately half the number
of megakaryocytes expressed p27 protein, both in normal BM (51.35 to
77.9%, n = 2) and in BM from ET patients (42% to 52%, n = 2),
although the proportions were somewhat overestimated because of
endothelial cells (see below). Some small cells in BM showed strong
nuclear staining (Fig 5B and E). Because p27 mRNA levels increased
exclusively in erythroid bursts, we examined p27 positivity in
erythroid cells. However, glycophorin A-positive cells were negative
for p27 (data not shown). Double staining using anti-p27 and anti-
and - light chains antibodies showed the small p27-strongly positive
cells to be plasma cells (data not shown); these cells were
p21-negative and no Ki-67-positive plasma cells were found, as
reported.42 p27 is involved in inhibiting normal B-cell
proliferation,11,43 but to our knowledge, there are no
reports regarding a role for p27 in normal plasmacytic maturation or in
proliferation. It will be of interest to determine if tumor cells of
plasma cell dyscrasias show aberrant p27 expression. Vascular
endothelial cells and lymphoid aggregates were also p27-positive (data
not shown). Findings in the subset of the cells with nuclear staining
of p27, using either of the two MoAbs (G173-524 and F-8), were consistent.
We used the anti-Ki-67 antibody, MIB1, as a marker of proliferation.
This antibody recognizes an antigen present in the nuclei of
continuously cycling cells in G1, S, G2, and M phases, but not in G0
phase,44,45 and is a reliable marker of cell cycling in BM
samples.42,46 Most of the cells in BM were positive for Ki-67 (Figs 5C and F, 6B, and 7B), as noted by other
investigators.42 To clarify the relationship between cell
cycling status and expression of p27 or p21 in in vivo megakaryocytes,
serial sections of BM from an ET patient were stained alternately for
Ki-67 and p21 (Fig 6) or for Ki-67 and p27 (Fig 7). We scored
p21, p27, and Ki-67 immunostaining of each megakaryocyte by the
intensity ([] none, [+] weak, [+] moderate, or
[++] strong). At least 200 cells identified in more than three
consecutive sections were counted. Regardless of p21 expression levels,
approximately half the number of megakaryocytes expressed no Ki-67
antigen. However, the percentage of Ki-67 strongly positive
megakaryocytes decreased with the increase in p21 intensity (Fig 6);
thus, there was inverse relationship between p21 and Ki-67 expression
in a subset of megakaryocytes. A clear inverse relationship was also
noted between p27 and Ki-67 expression (Fig 7). Almost all of the
p27-positive megakaryocytes were negative or weakly positive for Ki-67,
and almost all of the p27-negative megakaryocytes were moderately or
strongly positive for Ki-67. These observations strongly suggest that
p27 and probably p21 expression is tightly linked to cell cycling
status in megakaryocytes.
| |
DISCUSSION |
We made use of quantitative RT-PCR to evaluate p21 and p27 mRNA levels
in greater than 100 colonies. The results were reproducible and
comparable to findings in Northern analysis. Because each assay is
performed in a tube and includes an internal control, a control signal
confirms the appropriateness of samples and reaction. The method will
be applicable to other genes in multiple samples of small quantities.
According to our assay, p21 mRNA levels were elevated over time in all
colonies of the lineage examined. The increased levels were comparable
to those of differentiated and cell-cycle-arrested MEG-01s cells (Figs
1D and 2). Half lives of p21 transcripts were not altered during
differentiation of MEG-01s cells (T.T. and T.M., unpublished
data). In other differentiation models, mechanisms for
elevation of p21 mRNA were often transcriptional24,26,30,47 and p53-independent.4,19-22,24,40 Accordingly, normal BM
cells never immunostained for p53. Therefore, p21 mRNA expression
during hematopoiesis may be regulated in a p53-independent and
lineage-nonspecific manner.
Both p21 and p27 proteins were expressed in a subset of megakaryocytes
in normal human BM. An inverse relationship between p27 and Ki-67
suggests that p27 protein is expressed in cell-cycle arrested
megakaryocytes. However, reciprocal expression of p21 to Ki-67 was not
so obvious as that of p27, although the relationship between p21 and
Ki-67 strong positivities was clearly inverse. The presence of
p21-negative and Ki-67-negative megakaryocytes suggests that p21
expression is transient compared with continuous p27 expression in
cell-cycle arrested cells. The proportion of p21-positive cells among
Ki-67 moderately and strongly positive cells was larger than that of
p27-positive cells, an observation that may reflect p21 expression
earlier than p27 during cell-cycle exit. The hypothetical timing of p21
and p27 expression along the cell-cycle exit of megakaryocytes is shown
in Fig 8. p21 may act early and transiently
in the course of cell-cycle exit, and p27 may be subsequently
upregulated. In support of this, the transient expression of p21 was
noted in keratinocytes.48,49 Although the expression of p21
is increased in postmitotic cells immediately adjacent to the
proliferative compartment, the expression is decreased in terminally
differentiated primary keratinocytes.48 We previously reported that MEG-01s treated with TPA showed megakaryocytic
differentiation and expressed both p21 and p27 proteins in association
with a G1-phase arrest.28 Both p21 and p27 were present in
cyclin E-associated complexes, the histone H1 and Rb kinase activities
of which were then inactivated. In this differentiation model, p21
protein expression preceded that of p27 in analysis of synchronized
cell population.28 It remains to be elucidated whether p21
and p27 are indeed associated with any cyclin/CDK complex in the course
of cell-cycle arrest of normal megakaryocytes.
Various studies using leukemic cell lines showed that p21 and p27 may
be involved in differentiation and polyploidization of megakaryocytes.
For example, the ectopic expression of p21 or p27 leads to induction of
megakaryocytic differentiation of CMK cells.26
Overexpression of p21 results in an increase in ploidy of UT-7 cells,
which suggests that p21 may be implicated in polyploidization via
suppression of cdc2 activity at mitosis.27 However,
reciprocal expression of p21 and p27 to Ki-67 in in vivo megakaryocytes
does not support these views. Recent studies using normal
megakaryocytes showed that polyploidization of megakaryocytes is due to
abortive mitosis due to alterations in the regulation of mitotic
exit50,51 and that cdc2 is active in endoreduplicating megakaryocytes.51 Our observations together with these
reports suggest that suppression of cdc2 kinase activity by p21 (or
p27) is unlikely to be the mechanism of polyploidization in
megakaryocytes and that p21 and p27 have functions after
endoreduplication is completed.
In mice lacking p21, red and white blood cells in PB are
normal,52 although the absolute numbers of marrow
progenitors are significantly decreased.53 Meanwhile, in
mice lacking p27, complete blood count is normal, although the numbers
of hematopoietic progenitors are increased significantly.54
Although either p21 or p27 may be dispensable for hematopoietic
differentiation, it remains to be determined how hematopoiesis,
especially megakaryocytopoiesis, would be affected by knockout of both
p21 and p27.
p27 mRNA levels remained low during the observed period in
megakaryocyte colonies, even though p27 protein was expressed in a
subset of megakaryocytes. This finding is consistent with knowledge that the p27 protein level is regulated mainly by protein degradation steps through the ubiquitin-dependent proteolytic pathway and that p27
mRNA and protein levels do not coincide in many
situations.55-57 On the contrary, the elevation of p27 mRNA
in erythroid bursts, without detectable proteins in BM erythroblasts,
is intriguing. Although p27 regulation is frequently posttranslational,
p27 mRNA is indeed upregulated in some situations and is usually
accompanied by protein expression.58-60 The elevation of
p27 mRNA levels was unique to erythroid lineage and suggests the
presence of erythroid-specific regulation of p27 mRNA expression.
However, as p27 protein was absent, it may be dispensable in erythropoiesis.
Recently, Dao et al61 reported that CD34+ cells
from human BM expressed p27 protein and that reduction in levels of p27
protein using antisense oligonucleotides to p27 coupled with
transforming growth factor- (TGF-) neutralization
induces cell-cycle entry and increases retroviral transduction of
primitive human hematopoietic cells. We detected only low levels of p27
mRNA in cord blood CD34+ cells. The difference may be due
to posttranslational regulation again. We suppose that primitive
hematopoietic cells have elevated p27 protein levels, which are
decreased in cycling cells and increased again with cell-cycle exit
especially in megakaryocytes, although the mRNA levels remain low.
Steinman et al30 reported that myeloid maturation of
umbilical cord blood CD34+ cells is associated with an
increase in p21 expression at RNA and protein levels. We observed that
p21 mRNA expression increased with time up to day 15 in granulocyte
colonies, findings consistent with their report. However, in our
system, neither BM myeloid lineage cells nor PB granulocytes were
positive for p21 protein and only a few p21-positive cells were
detected in granulocyte colonies. Although one would need to exclude
the possible expression of p21 protein in granulocytes, p21 protein
levels in myeloid cells are much lower than in megakaryocytes.
p21 mRNA expression increased with time in macrophage colonies and some
of the macrophages immunostained for p21 antigen. Accordingly, PB
monocytes expressed p21 mRNA and protein. Consistent with these
observations, several cell lines are seen to express p21 mRNA and
protein during monocytic differentiation19-21,25,62 and
some alveolar macrophages of the lung are p21-positive.63 p27 protein levels are also elevated during monocytic differentiation in some cell lines.23,29 However, the p27 protein level was low in PB monocytes in our study and p27 protein was not detected in
macrophages in in vitro colonies.
By way of summary, although p21 mRNA levels increased over time in any
lineage of hematopoietic colonies, p21 protein levels were elevated
only in limited lineages (megakaryocyte and monocyte/macrophage). p27
protein was detected only in megakaryocytes, monocytes, and lymphoid
cells (lymphocytes and plasma cells), whereas p27 mRNA levels were
elevated only in erythroid bursts. These observations suggest that
complex lineage-specific regulation mechanisms are involved in p21 and
p27 protein expression in human hematopoiesis. In addition, reciprocal
expression of p21 and p27 to Ki-67 suggests that, in megakaryocytes,
both p21 and p27 may be involved in terminal exit from the cell cycle.
| |
ACKNOWLEDGMENT |
The authors thank Dr A. Noda for p21 cDNA plasmid, Dr J. Massague for
p27 cDNA plasmid, M. Yoshikawa for technical assistance, and M. Ohara
for language assistance.
| |
FOOTNOTES |
Submitted October 8, 1998; accepted February 16, 1999.
Supported in part by grants from the Ministry of Education, Science,
Sports and Culture of Japan.
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.
Address reprint requests to Toru Motokura, MD, Fourth Department of
Internal Medicine, University of Tokyo, School of Medicine, 3-28-6 Mejirodai, Bunkyo-ku, Tokyo 112-8688, Japan.
| |
REFERENCES |
1.
Morgan DO:
Principles of CDK regulation.
Nature
374:131, 1995[Medline]
[Order article via Infotrieve]
2.
Sherr CJ, Roberts JM:
Inhibitors of mammalian G1 cyclin-dependent kinases.
Genes Dev
9:1149, 1995[Free Full Text]
3.
Halevy O, Novitch BG, Spicer DB, Skapek SX, Rhee J, Hannon GJ, Beach D, Lassar AB:
Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD.
Science
267:1018, 1995[Abstract/Free Full Text]
4.
Parker SB, Eichele G, Zhang P, Rawls A, Sands AT, Bradley A, Olson EN, Harper JW, Elledge SJ:
p53-independent expression of p21Cip1 in muscle and other terminally differentiating cells.
Science
267:1024, 1995[Abstract/Free Full Text]
5.
Zabludoff SD, Csete M, Wagner R, Yu X, Wold BJ:
p27Kip1 is expressed transiently in developing myotomes and enhances myogenesis.
Cell Growth Differ
9:1, 1998[Abstract]
6.
Durand B, Gao FB, Raff M:
Accumulation of the cyclin-dependent kinase inhibitor p27/Kip1 and the timing of oligodendrocyte differentiation.
EMBO J
16:306, 1997[Medline]
[Order article via Infotrieve]
7.
Fredersdorf S, Milne AW, Hall PA, Lu X:
Characterization of a panel of novel anti-p21Waf1/Cip1 monoclonal antibodies and immunochemical analysis of p21Waf1/Cip1 expression in normal human tissues.
Am J Pathol
148:825, 1996[Abstract]
8.
el-Deiry WS, Tokino T, Waldman T, Oliner JD, Velculescu VE, Burrell M, Hill DE, Healy E, Rees JL, Hamilton SR, Kinzler KW, Vogelstein B:
Topological control of p21WAF1/CIP1 expression in normal and neoplastic tissues.
Cancer Res
55:2910, 1995[Abstract/Free Full Text]
|
TOP
ABSTRACT
INTRODUCTION