1. Art and Diseño

From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
RAPID COMMUNICATION
Murine Thrombopoietin mRNA Levels Are Modulated by Platelet Count
By John M. McCarty, Katherine H. Sprugel, Norma E. Fox, Daniel E. Sabath, and Kenneth Kaushansky
The activity of the c-Mpl ligand on hematopoietic progenitors meets criteria expected for thrombopoietin (TPO). Bioassays have shown that blood TPO levels are inversely related to platelet mass. We sought to identify the molecular
basis for this regulation. To determine if TPO mRNA levels
respond to platelet demand, RNA from selected organs of
mice with high, normal, or low plateletcounts was subjected
to semiquantitative reverse transcriptase-polymerase chain
reaction. Although no differences in TPO mRNA levels between control and treated mice could be detected in liver
or kidney, TPO-specific bands were more intense after 25
to 30 polymerase chain reaction cycles in marrow-derived
mRNA from thrombocytopenic mice. The TPO-specific bands
were less intense in thrombocytotic mouse marrow and
spleen than control mouse marrow and spleen after 30 cycles. These data support the hypothesis that TPO levels are
regulated, at least in part, by modulating mRNA levels in
response to platelet demand.
0 1995 by The American Society of Hematology.
T
with the c-mpl cDNA (BAFJmpl), have been shown to correlate precisely with this in vivo activity.2’ Sera from mice
rendered thrombocytopenic by either irradiation or after administration of rabbit antimouse platelet sera (RAMPS) contain high levels of this Mpl ligand, which is maximal shortly
after the platelet nadir in these experiments. Similar results
have been reported from other laboratories.I4.”
Although regulation of TPO could occur at many levels,
two possible mechanisms by which serum TPO activity may
be regulated have been advanced in the literature. Based on
the finding that platelets bear receptors for TPO, Kuter et
allh.17 have proposed that TPO gene expression is constant
HE LIGAND FOR the proto-oncogene c-mpl has been
shown to meet a number of criteria expected of the
primary regulator of megakaryocyte and platelet production,
thrombopoietin (TPO). Althoughmany cytokines such as
interleukin-3 (IL-3), IL-6, IL- 1 1, and leukemia-inhibitory
factor (LIF) support the expansion of hematopoietic precursor cells and promote colony-forming unit-megakaryocyte
(CFU-Mk) colony formation,’-’ liquid marrow culture, marrow colony assays, and in vivo studies have shown that
these growth factors do not completely fulfill the operational
definition of a putative thrombopoietic protein.’.‘-* In contrast, using similar methods, the Mpl ligand displays these
expected activities; it supports megakaryocyte colony formation, increases megakaryocyte size and ploidy, induces the
expression of lineage-specific differentiation markers, and is
the most important regulator of megakaryocyte and platelet
prod~ction.~.’
These data support the contention thatMpl
ligand and plasma TPO are identical.
As the physiologic regulator of platelet production, TPO
levels wouldbe expected to vary inversely with platelet
demand. Using radioisotopic incorporation into platelets in
recipient animals as a bioassay for plasma TPO, Ode11et
a14. I O showed an increase in serum TPO activity in response
to decreasing platelet mass. These changes were associated
with increased megakaryocyte number, size, andploidy
within 6 to 24 hours and were maximal 96 hours after the
onset of acute thrombocytopenia.5.’’-’3
Bioassays measuring the proliferation of a factor-dependent murine cell line BAFh303, which was stablytransfected
From the Department of Medicine, Division of Hematology, and
the Departments of Laboratory Medicine and Medicine, Division
of Medical Genetics, University of Washington, Seattle, WA; and
Zymogenetics, Inc, Seattle, WA.
Submitted August I , 1995; accepted September 5, 1995.
Supported by National institutes of HealthGrants No. R01
CA31615 and R01 DK49855 ( K . K . )
Address reprint requests to John M. McCarty, MD, Box 357710,
Division of Hematology, University of Washington, Seattle, WA
98/95-7710.
The publicationcosts of this article were defrayed in partby page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 sole1.y to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-497i/95/8610-0044$3.00/0
3668
andthat serum levels are controlled by the platelet mass
through uptake and metabolism. Higher platelet counts
would result in increased TPO catabolism, leading to a lower
serum TPO levels. Conversely, a lower platelet mass would
have less capacity for TPO uptake and metabolism, resulting
in higher serum levels to act on megakaryocytes and their
precursors. Such a schema has been proposed for regulation
of serum macrophage colony-stimulating factor (M-CSF)
levels by monocytes.” Alternatively, serum levels could be
modulated by feedback regulation at the level of gene expression. Multiple studies have provided compelling evidence that serum erythropoietin levels are regulated by modulating expression of its gene.” TPO gene transcription could
be inhibited or TPO-specific mRNA destabilized by high
platelet counts; either of these processes could be countered
by physiologic regulators that sense thrombocytopenia.
These two models are not mutually exclusive, nor do they
encompass all possible mechanisms by which TPO serum
levels may be regulated. As a first step towards understanding the mechanisms that regulate TPO blood levels, we comparedthe levels of TPO-specific mRNAin the organs of
mice with high, low, and normal platelet counts.
Northern blot analysis identified the liver, the kidney, and,
after longer exposure, the spleen as rich sources of TPOspecific transcripts.” Although marrowwasnottested
in
these initial studies, it was selected for study because of
its potential physiologic relevance. Using a semiquantitative
reverse transcription-polymerase chain reaction (RT-PCR)
technique:’ we sought to correlate TPO-specific mRNA levels from these tissues with platelet counts. Although no noticeable differences between control and thrombocytopenic
mice could be detected in liver or kidney, TPO-specific
bands were more intense in marrow-derived mRNA in both
Blood, Vol 86, No 10 (November 15),
1995:
pp 3668-3675
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3669
MURINE THROMBOPOIETIN MRNA REGULATION
RAMPS-treated and irradiated miceand in spleen-derived
mRNA in RAMPS-treated animals. The TPO-specific bands
were less intense in thrombocytotic mousemarrowand
spleen than in controls. The findings that TPO mRNA levels
are upregulated in marrow and spleen by thrombocytopenia
and may be downregulated by TPO induced thrombocytosis
support the hypothesis that TPO levels are regulated, at least
in part, by alterations in steady-state mRNA levels.
1:5 Liver
Liver
20 25 30
20 25 30
MATERIALS AND METHODS
Animal preparation. Female CS7B1/6 mice of S to 6 weeks of
age were studied in four separate groups. All groups had baseline
and posttreatment peripheral blood counts determined by Cell-Dyn
3000 counter (Abbott Laboratories, North Chicago, IL). After killing
the mice, liver, spleen. both kidneys, and femurs were obtained from
each mouse and flash frozen in liquid nitrogen.
Experiment I consisted of two mice. One mouse was exposed to
a single dose of SS0 Cy from a cesium source (Gammacell 40
Irradiator; Atomic Energy of Canada, Ltd. Kanata, Ontario, Canada)
and received 1.2 mg of carboplatinum (Bristol Meyers-Squibb Oncology Division, Princeton, NJ) intraperitoneally (IP). Another untreated mouse remained as control. Both mice were killed at day 13.
Experiment 2 consisted of three mice. One mouse was exposed
to a single dose of 600 Cy from a cesium source and received I .2 mg
of carboplatinum IP. A second mouse was treated with recombinant
murine TPO (Zymogenetics Inc, Seattle, WA) at 12.000 U IP daily
and a thirdwith TPO vehicle IP daily. All mice were killed after
16 days.
Experiment 3 consisted of three mice. One mouse received two
daily doses of 1 0 0 pL RAMPS (kindly supplied by G. Roth, VAMC.
Seattle, WA) IP andwaskilled
after 96 hours. One mouse was
treated with 12,000 U recombinant murine TPO IP daily for 7 days
and another with TPO vehicle alone for 7 days before death.
Experiment 4 consisted of 4 mice treated with daily doses of 175
pL RAMPS IP on day 1 and one untreated control mouse. Two of
the mice received 125 pL IP doses on subsequent days. Because the
maximal effect of RAMPS on platelets is seen by 6 hours, one
mouse was killed daily 6 hours after the RAMPS injection on days
I , 2, and 3. The fourth treated mouse died before the fourth dose
of RAMPS could be administered of an acute hemorrhagic diathesis.
The fifth mouse remained untreated on these successive dosing days.
RNA isolation. Tissues were disrupted in an 1 1 molL guanidium
thiocyanate solution using a Polymizer (Polytron Drives, Patterson,
NJ) tissue homogenizer. Total cellular RNAwas obtained using
the RNAgents Total Cellular RNA Isolation Kit (Promega Corp,
Madison. WI) according to the manufacturer’s directions.
RT-PCR. In batchedand parallel reactions, equal amounts of
total cellular RNA from each control and treated mouse organ were
subjected to reverse transcription with an oligo dT,, primer (Pharmacia Biotech Inc. Piscataway, NJ) using the Superscript II system
(GIBCO BRL, Gaithersburg, MD) according to the manufacturer’s
specifications. One tenth of the reverse transcription reaction was
subjected to PCR in parallel using Amplitaq (Applied Biosystems,
Foster City. CA). A 717-bp fragment was obtained using the oligonucleotides + S4 (5’-TCTGTCCAGCCCCGTAGGTC) +73 and +771
(S’-GTTCCATCCACAGGTCCGTG) +75 1. Batched reactions for
each reverse transcribed RNA species were separated into 30-pL
aliquots, subjected to at least three different numbers of PCR cycles,
and size fractionated on agarose gels. These same reverse transcription reactions were subjected to 20 cycles ofPCRwith
limiting
concentrations of primers specific for glyceraldehyde-3-dehydrogenase (GAPDH) +S70 (S-CAAAGTTGTCATCGATGACC) +SS0
and +376 (CCATGGAGAAGGCTGGGG) +394 to serve as an
RNA loading and reverse transcription control. The number of PCR
Fig 1. Semiquantitative RT-PCR can detect variations in TPO-specific mRNA levels. RT-PCR performed in parallel as described in the
Materials and Methods on liver total cellular RNA diluted 1:5 with
RNAse-free water (1:5 LIVER) and undiluted liver total cellular RNA
(LIVER).Negative and positive control lanes labeled 0 and +, respectively. Arrow marks the major band under study corresponding to
TPO 1 and TPO 2; the minor band corresponding to TPO 3 PCR product is seen beneath.
cycles were chosen so that the relative intensity of the bands seen
on agarose gels would belinearly related to the relative concentration
of the TPO-specific mRNA species present in the reverse transcription reactions.
Seqrtencing qfRT-PCR prohcfs. To confirm the identity of the
PCR products. representative RT-PCR reactions were electrophoresed in agarose gels, the bands were excised, and DNA was recovered using Geneclean (Bio 101, Vista, CA) according to the manufacturer‘s recommended protocol. After subcloning into pGEM-T
(Promega Corp. Madison, WI), positive clones were pickedand
grown out and plasmids were isolated by alkaline lysis followed by
polyethylene glycerol precipitation. Using a primer that recognized
the l7 (S’-TAATACGACTCACTATAGGG) or the SP6 (S‘-GATTTAGGTGACACTATAG) promoter sequence of pGEM-T, several
clones were sequenced using the Taq Dye-Deoxy Dye Terminator
a 370A
Cycle Sequencing Kit (Applied Biosystems) andrunon
automated Sequencer (Applied Biosystems) to verify that they were
derived from mTPO.
RESULTS
TPO mRNA detected by RT-PCR is predominantly TPO
1. Gurney et al.’2 report two TPO isoforms, TPO2 and
TPO 3, that differ from full length TPO 1 by 12-bp and 1 12bp deletions, respectively, leading to truncation in the fourth
exon. Of eight clones sequenced, six contained sequences
consistent with full-length TPO 1 and two werefound to
contain the 12-bp deletion that characterizes TPO 2. The
truncated form, TPO 3, could be differentiated by size using
these primer pairs and could onlybe appreciated after 30
cycles of PCR. The minor band seen in Fig 1 is consistent
with TPO 3. Thus, the RT-PCR method on which this study
is based predominantly recognizes the active isoform TPO
1 with some minor contribution of the poorly secreted TPO
2 and TPO 3 forms.
E e c t qf experimental conditionsonperipheralblood
counts. The results of posttreatment complete blood counts
from each set of mice are shown in Table 1. Both radiation
and RAMPS treatment regimens resulted in moderate to severe thrombocytopenia, with the mostprofound decrease
seen after successive doses of RAMPS in experiment 4. All
concentrations of recombinant TPO (rTP0) used in this
study induced substantial thrombocytosis.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
McCARTY ET AL
3670
Table 1. Effect of Experimental Conditions on Peripheral Blood
Counts of Study Mice as Measured on Coulter Counter
Experimental
Group
1
2
3
4
PLT
Condition
Control
XRT/Carboplatinum
Control
RAMPS 100 U lP/7 d
TPO 12,000 U lP/7 d
Control
XRT/Carboplatinum
TPO 12,000 U IP/16 d
Control
RAMPS 6 h
RAMPS 30 h
RAMPS 54 h
WBC
1335.041.2 2.0
0.2
67.0
1184.041.3 4.0
65.039.1
4.1
3244.0
5.3
775.049.0 13.5
25.0
7.5
3915.038.0 10.0
945.0
5.5
80.0
2.0
20.0
2.0
8.0
0.6
HCT
5.7
44.6
<5
46.0
29.0
25.0
6.0
Experimental groups as designated in the Methods and Materials
section.
Abbreviations: WBC, white blood count per microliter; HCT, hematocrit; PLT, platelet count per microliter.
Semiquantitative RT-PCR can detect variations in TPOspecijc mRNA levels. The semiquantitative RT-PCR technique used in this study capitalizes on the fact that PCR
bands derived from more abundant mRNA species will be
detected after fewer cycles of RT-PCR than those that are
present in lower concentrations. To show the ability of this
method to detect differences in mRNA concentrations, we
performed the technique on RT-PCR reactions containing
decreasing concentrations of liver RNA. Figure 1 compares
the TPO-specific PCR band intensities using two serial fivefold dilutions of liver RNA. As expected, a more intense
TPO-specific band is seen at 25 and 30 cycles in the fivefold
more concentrated sample. Below the major PCR band, the
minor band contributed by the TP03 isoform at 30 cycles
can be seen. This established that this technique could be
used to detect differences in TPO-specific mRNA. Although
RNAineach
experimental group was subjected to these
same conditions in parallel, comparison of band intensity
was found to be more efficient if photos of the gels were
cut out and rearranged, as seen in Fig 2.
Liver
28
25
31
Kidney
25 28 31
Spleen
28
25
31
TPO-spec@ mRNA increases afrer irradiatiodcarboplatinum therapy. Figure 2 displays the results of the semiquantitative RT-PCR on RNA samples from experiment 1. The relative intensity of TPO-specific bands in the control and treated
reactions are similar through 31 cycles of PCR of liver- and
kidney-derived mRNA. However, in treated marrow. a more
intense band can be seen after28 cycles compared with control,
indicating a higher TPO mRNA concentration after irradiation.
A less dramatic increase can also
be seen in treated spleen after
31 cycles. Cornparison of PCR bands for GAPDH from each
organ-derivedmRNAshowsthatequalamounts
of reverse
transcribedmRNAwerepresent
in each PCRreaction. To
extend these results, we performed a second experiment, this
time includinga rTP0-treated animal. Figure 3 confirms earlier
results that no significant differences be
candiscerned in control
myelosuppressed or rTPO-treated liver or kidney samples from
experiment 2. The presence of more prominent bands in radiated versus control marrow and spleen lanes as well as absent
bands in the rTPO-treated marrow and spleen lanes support the
conclusion that TPO mRNA can be modulated in response to
changes in platelet counts.
TPO-specijc mRNA responds to changes in platelet
counts mediated by RAMPS. To study the regulation of
TPO mRNA levels further, we used a second independent
method for induction of acute thrombocytopenia. Comparison of the TPO-specific PCR bands from vehicle, rTP0-,
and RAMPS-treated mice in Fig 4 again shows that thrombocytopenia failed to alter TPO mRNA levels in liver or
kidney. As was seen after myelosuppression-inducedthrombocytopenia, a substantial increase in band intensity can be
seen in the marrow and spleen of RAMPS-treated mice. In
addition, a decrease inband intensity isseen in marrow
derived from the rTPO-treated animal. Comparison of control and rTPO-treated spleen is inconclusive given the low
intensity of the control band. These results suggest that, in
marrow and, to a lesser extent, in spleen, TPO mRNA levels
increase in response to acute thrombocytopenia and decrease
with thrombocytosis.
The RAMPS-treated mice in experiment 4 achieved the
most profound thrombocytopenia of all those tested. Examination of the relative band intensities shown in Fig 5 again
Marrow
2825
cDNA
31
LC Lx Kc K x Sc Sx MC Mx 0
Fig 2. TPO-specific mRNA increases after irradiationlcarboplatinumtherapy. Results of semiquantitative RT-PCR on liver, kidney, spleen,
and marrow-derived total cellular RNA from untreated and carboplatinumlradiation-treatedmice as described in the Materials and Methods.
Reactions were run for 25, 28, or 31 cycles. The positive control lane is labeled cDNA. GAPDH control reactions after 20 cycles are included
untreated kidney (Kc), treated kidney [ K x ) , untreated spleen (Sc), treated spleen
for each RNA studied: untreated liver (LC), treatedliver (h),
6x1, untreated marrow (MC),treated marrow (Mx),and negative control ( 0 ) .
OPOIETIN
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
MURINE
3671
Liver
Kidney
Spleen
Marrow 0/cDNA
Vehicle
WCarbo
TPO
25 28 31
25 28 31
28
25
31
25 28 31
1$
GAPDH
LC L x Lt Kc Kx Kt Sc Sx St MC MxMt
0
Fig 3. TPO-specific mRNA responds t o changes in platelet counts mediated by irradiationlcarboplatinum(XRTlCarbol treatment. Results
of semiquantitative RT-PCR on liver, kidney, spleen, and marrow-derived totalcellular RNA from untreated, RAMPS-, and recombinant mTPOtreated miceas described in the Materials andMethods. Reactions were run for25,28, or 31 cycles. Negative control (01and positive control
lanes IcDNA) are shown. GAPDH control reactions after 20 cycles are included for each RNA sample: untreated liver (LC), XRTICarbo-treated
liver (h),
TPO-treated liver (Lt), untreated kidney (Kc), XRTICarbo-treated kidney (Kx), TPO-treated kidney (Kt), untreated spleen (Scl, XRTl
Carbo-treated spleen (Sx), TPO-treated spleen (St), untreated marrow (MC),XRTlCTx-treated marrow (Mx), TPO-treated marrow (Mtl, and
negative control (0).
ment could possibly introduce a sampling bias. Because the
shows that no changes in TPO-specific mRNA can be detected in liver or kidney derived lanes, but that bands show
same results were seen with either radiation of RAMPS inincreased intensity in RAMPS-treated marrow and spleen.
duction of thrombocytopenia, it became important to know
Although this effect can be seen as early as 6 hours after
whether these manipulations altered the cellularity or compoRAMPS administration, it is most apparent after 30 and 54
sition of marrow. Figure 6 shows photomicrographs of huhours of acute thrombocytopenia.
meri taken from the same mice used in experiments 1 and
RAMPS treatment does not affect marrow cellularity or
4. Compared with the normally100% cellular marrow of
the control mouse in Fig 6A, the myelosuppressed mouse
composition. In the present study we used two well-established methods to induce severe thrombocytopenia, ie, myelo- marrow was hypocellular, with a predominance of marrow
stromal cells, histiocytes, and plasma cells (Fig 6B). As exsuppressive therapy and antiplatelet antisera. These methods
may affect the cellular composition of spleen and marrow;
pected, the administration of rTP0 resulted in substantial
thus, it is possible that the induction of thrombocytopenia
megakaryocyte expansion (Fig 6C). However, examination
of the humeri taken from each of the RAMPS-treated mice
may have altered the relative proportion of TPO-producing
shown in Fig 6D through F showed no change in overall
cells in these organs. The end result could be an artificial
enrichment for TPO mRNA. RT-PCR surveys of primary
cellularity or cellular composition compared with the control
cell cultures and of cell lines performed in our laboratory
mouse. On the marrow corresponding to the 54 hours of
have shown that stromal elements are a source of TPORAMPS exposure, which is when the biologic effect of TPO
specific mRNA in marrow, suggesting that relative enrichis first seen in various biologic and marrow colony assays,
Liver
Fig 4. TPO-specific mRNA responds t o changes in platelet
counts mediated byRAMPS. Results
RT-PCR
on liver, kidney, spleen, and marrow-derivedtotal cellular RNA
from untreated, RAMPS, and recombinant mTPO-treated mice
as described in the Materials and
Methods. Reactions were run for
25, 28, or 31 cycles. Negative
control (0)
and positive control
lanes (cDNA) are shown. GAPDH
control reactions after 20 cycles
included for each RNA sample.
Kidney
Spleen
Marrow 0/cDNA
Vehicle
RAMPS
TPO
28
25
31
28
25
31
LC L r LtKcKrKt
28
25
31
Sc Sr StMC
25 28 31
Mr M t 0
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
McCARTY ET AL
3672
Liver
Kidney
Spleen
Marrow
0/cDNA
Vehicle
6 Hrs RAMPS
I
30 Hrs RAMPS
54 Hrs RAMPS
20 25 30
20 25 30
20 25 30
20 25 30
GAPDH
LcLbL30L54KcK6K30K56ScS6S30S54McM6M30M.54
Fig 5. TPO mRNA increases can be detected 6 hours after RAMPS administration. Results of semiquantitative RT-PCR on liver, kidney,
spleen, and marrow-derived total cellular RNAfrom untreated and RAMPS-treated mice as describedin the Materials and Methods. Reactions
were run for 20, 25, or 30 cycles. Negative control ( 0 )and positive control lanes (cDNA) are shown. GAPDH control reactions after 20 cycles
are included for each RNA sample: untreated liver (LC), 6-hour RAMPS-treated liver (L6). 30-hour RAMPS-treated liver IL30).--hour RAMPStreated liver (L541, untreated kidney (Kc), 6-hour RAMPS-treated kidney (K6). 30-hour RAMPS-treated kidney (K30). %-hour RAMPS-treated
kidney (K54). untreated spleen (Scl, 6-hour RAMPS-treated spleen (S6), 30-hour RAMPS-treated spleen (S301,54-hour RAMPS-treated spleen
6541, untreated marrow (MC), 6-hour RAMPS-treated marrow (M6), 30-hour RAMPS-treated marrow (M30), and 54-hour RAMPS-treated
marrow (M54).
a substantial increase in megakaryocyte number and size can
be seen (Fig 6F). These data would indicate that the increase
in TPO-specific mRNA after RAMPS-induced thrombocytopeniaisnot due to a gross change in the composition or
cellularity of the marrow.
TPO qfects on marrowcomposition
nnd cellularity.
Fromthe photomicrograph of marrow derived fromthe
rTPO-treated mouse seen in Fig 6, substantial expansion of
the megakaryocyte compartment without change in overall
cellularity is evident. On these animals, the megakaryocyte
mass accounts for nearly 30% of the marrow space. Because
these cells contribute a significant proportion of the mRNA
in these tissues, it also became important to know whether
megakaryocytes contained TPO-specific mRNA. Using the
same experimental conditions used in this study, TPO
mRNA could not be detected from purified megakaryocytes
despite 80 cycles of RT-PCR (data not shown). However,
this implies that the reduction in band intensity seen in the
marrow and spleen from rTPO-treated animals might be explained, at least in part, by relative dilution from megakaryocyte-derived mRNA.
DISCUSSION
The gene for the Mpl ligand, TPO, has recentlybeen
cloned by a number of groups. Work by Wendling et al,”
has shown that virtually all of the thrombopoietic activity
of thrombocytopenic plasmacanbe accounted for by the
Mpl ligand, thereby supporting earlier studies that suggest
that TPO protein levels are inversely related to platelet
counts. In this study, we sought to determine whether TPOspecific mRNA levels vary in tissues and could therefore
account for changes in serum TPO concentration. To do so,
we developed a semiquantitative RT-PCR-based assay that
could easily detect TPO-specific transcripts in liver, kidney,
spleen, and marrow. Because mRNA was made from whole
tissue extracts, itis impossible for usto determine from
this study which cells in these tissues are the site of TPO
production. It is possible that a common cell type found in
all these tissues, such as endothelial cells, fibroblasts. or
tissue macrophages, is the source of TPO production. Surveys of several cell lines, including primary and transformed
endothelial cells and fibroblasts, have shown them to express
TPO message and protein. Although some inferences can be
made from RT-PCR in cell lines, the cellular localization of
TPO awaits the completion of in situ hybridization experiments, which are already underway.
Two models of TPO gene regulation have been proposed.
In the first, serum levels are maintained solely by platelet
uptake and metabolism of TPO. In the second, platelet levels
are sensed. resulting in appropriately increased or decreased
levels of TPO gene expression. The phenotype of a recently
reported NF-E2 knockout mouse lends credence to the assertion that platelet metabolism of TPO is not the sole determinant of serum TPO levels. In the NF-E2 -” phenotype,
although platelet counts are low, serum TPO levels remain
normal.” TPO levels in these transgenic mice are therefore
discordant with those expected from a purely catabolic
model. The precise impact of the NF-E2”- phenotype on
the modulation of TPO mRNA levels in response to thrombocytopenia cannot be determined completely, whereas no
difference in liver TPO mRNA levels was seenbetween
control and mutant mice; spleen- and marrow-derived RNA
was not tested in their study.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
MURINE THROMBOPOIETIN MRNA REGULATION
3673
Fig 6. Effect of irradiationlcarboplatinum. RAMPS, and TPO administration on marrow composition and cellularity. Photomicrographs of
hematoxylin and eosin stained humeral cross-sections from mice killed in experiments 1 and 4. (A) Control mouse; (B) rnyelosuppressed and
carboplatinurn-treatedmouse; (C) recombinant murine TPO at 12,000 U IP per day-treated mouse; (D) 6-hour RAMPS-treated mouse; (E) 30hour RAMPS-treated mouse; and (F)54-hour RAMPS-treated mouse. Original magnifications: (A and C through F) x 10, (B) x 50.
Further support for an inductive model of TPO gene regulation can be found in the observations of Wendling et al"
in RAMPS-treated mice. Whereas a modest increase in serum TPO activity could be detected at the time of platelet
nadir in a BAFdmpl proliferation assay, peak serum activity
was seen at 24 hours after RAMPS administration. These
same observations were also reported by Hunt et al.'4 In a
murine model of immune-mediated thrombocytopenia, peak
serum TPO levels were detected not at the time of platelet
nadir, which occurred at 8 hours, but rather at 24 hours after
treatment. From their findings, Hunt et al" considered the
delayed activity peak to represent the time required for TPO
gene induction, as well as TPO protein production and accumulation. This response profile is consistent with the induction of TPO-specific mRNA in the RAMPS-treated mice
reported in the present study.
The results reported here indicate at least one molecular
basis for TPO regulation. The semiquantitative method used
in this study shows that differences in mRNA levels detected
in some organs must be due to either transcriptional or posttranscriptional regulatory mechanisms. From Northern blot
analysis, it is obvious that the liver and the kidney are the
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3674
two most predominant sites of TPO mRNA.'4,20However,
these two tissues do not exhibit modulation of TPO mRNA
in response to changes in platelet counts in our study. In
the spleen and marrow, perhaps themore physiologically
relevant tissues for thrombopoiesis, TPO-specific mRNA
levels were found to be inversely related to platelet counts.
Although this does not exclude a role for platelet catabolism
in maintaining appropriate serum TPO levels, it does provide
evidence for an alternative mechanism regulating TPO in
serum in response to changes in platelet counts. Thus, at
least in marrow and spleen, mRNA is either transcriptionally
induced or stabilized. However, the magnitude by which
TPO mRNA levels change in both spleen and marrow cannot
be assessed for comparison with the degree by which TPO
protein activity and levels change in serum. Correlation of
serum TPO activity and serum and cellular protein levels
with mRNA production from these tissues is required to
address quantitative aspects of TPO expression. It is anticipated that the modulation of mRNA levels seen in mouse
spleen may not be seen in human splenic tissues because of
the greater functional role of the spleen as a site of hematopoiesis in the mouse. Nonetheless, these findings in mice
can likely be extrapolated to human marrow.
Studies of the mRNA isoforms reported by Gurney et aIz2
have shown that only the full-length TPO mRNA is secreted
and accounts for the serum TPO activity seen in thrombocytopenia. It is important to note that the majority of mRNA
that comprise the major PCR products in this study encode
for the full-length biologically active form of TPO, with only
a minor contribution of the nonsecreted, inactive TPO 2
form. TPO 3 could be detected only in insignificant amounts.
Changes in the relative ratios of these forms may well represent a regulatory mechanism; this hypothesis may be tested
by subcloning and sequencing the RT-PCR products from
each of these tissues in the presence and the absence of a
thrombocytopenic stimulus.
From the results obtained in this study some generalities
can be made about the time course of the TPO mRNA response to thrombocytopenia. From experiment 4,in which
mice were killed 6 hours after each successive dose of
RAMPS, it appears that a small but discernible increase in
TPO mRNA can be detected in spleen and marrow as early
as 6 hours after the onset of acute thrombocytopenia. This
increase continues to 30 hours and, although no difference
can be seen in this study between 30 and 54 hours after the
induction dose of RAMPS, the band intensity remains higher
than control through this time period. Data from the myelosuppressed mice that were killed after day 16 showed an
increased level of TPO mRNA, suggesting that, in this
model, TPO mRNA remains high in marrow and spleen as
long as the platelet count remains low. Thus, although the
response of TPO mRNA in marrow andspleen is fairly rapid,
by 6 hours, with peak levels by 30 hours, it is also longlasting, suggesting an ongoing signal that results in higher
TPO mRNA levels. Likewise, the reduction in TPO mRNA
levels in response to exogenously administered rTPO-induced thrombocytosis can be seen for 16 days. Although it
is not certain that this is due solely to modulation of mRNA
levels or due to relative dilution of TPO mRNA by increased
McCARTY ET AL
amounts of megakaryocyte RNA, it is consistent with the
data from the thrombocytopenic animals. This issue can be
further addressed by studying mice hypertransfused with
platelets to avoid disturbing the megakaryocyte compartment
of the marrow. Further studies examining the effect of platelet recovery after thrombocytopenia on TPO mRNA levels
in this myelosuppressed model would also be informative.
The nature or the source of the signal that allows tailoring
of TPO mRNA levels to platelet demand is not clear from
our data and is the basis of ongoing study.
It is also apparent from examination of the peripheral
blood counts performed on the mice from these studies that,
in addition to thrombocytopenia, carboplatinudradiation as
well as RAMPS treatment affects the peripheral red blood
cell and whiteblood cell compartments. Because several
additional cytokine levels would be expected to be elevated
in these animals (eg, erythropoietin, GM-CSF, and G-CSF),
it is possible that the stimulus for TPO mRNA induction may
not be due to isolated thrombocytopenia alone. However, in
the RAMPS-treated mouse of experiment 1, no effect is seen
on either the white blood cell count or hematocrit. Thus, it
appears that TPO &A
can be induced in the absence of
anemia or leukopenia.
This study provides an initial insight into the molecular
mechanisms that regulate expression of the TPO gene. Current and future efforts are directed at more precise quantitation and the cellular localization of TPO mRNA. The development of in vitro models of TPO production and studies
including nuclear run-on assays and mFWA half-life determinations will be required to confirm the constitutive nature
of TPO expression in liver and kidney and to determine
relative contributions of transcriptional and posttranscriptional mechanisms that regulate TPO mRNA levels in response to changes in platelet mass. It is hoped that a better
understanding of the molecular processes that regulate this
important mediator of thrombopoiesis will lead to greater
insights into the process of hematopoiesis as a whole.
ACKNOWLEDGEMENT
TheauthorsthankDr Steve Collins for his contributions to the
preparation of this manuscript and Jackie Humes for her technical
advice on the preparation of the murine models.
REFERENCES
1. Hill W,Warren MK, LevinJ, Gauldie J: Evidence that interleukin-6 does not play a role in the stimulation of platelet production
after induction of acute thrombocytopenia. Blood 80:346, 1992
2 . Debili N, Masse JM,Katz A, Guichard J, Breton-Gorius J,
Vainchenker W: Effects of the recombinant hematopoietic growth
factorsinterleukin-3,interleukin-6,stem cell factor,andleukemia
inhibitoryfactor on themegakaryocyticdifferentiation of CD34'
cells. Blood 82:84, 1993
3. Kaushansky K, Lok S, Holly R D , Broudy VC, Lin N, Bailey
MC, Forstrom J W , Buddle MM, Oort PJ, Hagen FS,Roth GJ,Papayannopoulou T, Foster DC: Promotionof megakaryocyte progenitor
expansion and differentiation by the c-Mpl ligand thrombopoietin
[see comments]. Nature 369:568, 1994
4. Ode11 lT, JacksonCW,FridayTJ,CharshaDE:Effects
of
thrombocytopenia on megakaryocytopoiesis. Br J Haematol 17:91,
1969
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3675
MURINE THROMBOPOIETINMRNAREGULATION
5. Odell TT, Jackson CW, Murphy J R Platelet recovery after
induction of acute thrombocytopenia. Proc Soc Exp Biol Med
148:829, 1975
6. Williams N: Is thrombopoietin interleukin 6? Exp Hematol
19:714, 1991
7. Neben T Y , Loebelenz J, Hayes L, McCarthy K, Stoudemire J,
Schaub R, Goldman SJ: Recombinant human interleukin-l 1 stimulates megakaryocytopoiesis and increases peripheral platelets in normal and splenectomized mice. Blood 81:901, 1993
8. Metcalf D, Waring P, Nicola NA: Actions of leukaemia inhibitory factor on megakaryocyte and platelet formation. Ciba Found
Symp 167:174,1992
9. Gurney AL, Carver-Moore K, de-Sauvage FJ, Moore M W :
Thrombocytopenia in c-Mpl-deficient mice. Science 265:1445,
1994
10. Odell TT, Jackson CW, Friday TJ, Du KY: Assay of megakaryocytopoiesis in thrombocytopenic rats. Br J Hematol 21:233,
1971
11. Kuter DJ, Rosenberg RD: Regulation of megakaryocyte
ploidy in vivo in the rat. Blood 75:74, 1990
12. Jackson CW, Edwards CC: Platelet production capacity at
intervals after acute thrombocytopenia. Scand J Haematol 17:285,
1976
13. Miura M, Jackson CW, Lyles SA: Increases in circulating
megakaryocyte growth-promoting activity in the plasma of rats following whole body irradiation. Blood 63:1060, 1984
14. de Sauvage FJ, Hass PE, Spencer SD, Malloy BE, Gurney
Spencer SA, Darbonne WC, Henzel WJ, Wong SC, Kuang
WJ, Oles U,Hultgren B, Solberg LA, Goeddel DV, Eaton DL:
Stimulation of megakaryocytopoiesis and thrombopoiesis by the cMpl ligand [see comments]. Nature 369:533, 1994
15. Bartley T D , Bogenberger J, Hunt P, Li YS, Lu HS, Martin
F, Chang MS, Samal B, J.L. N, Swift S, Johnsom MJ, Hsu RY,
Parker VP, Suggs S, Skrine J D , Merewether LA, Clogston C, Hsu
E, Hokom MM, Homkol A, Choi E, Pangelinan M, Sun Y, Mar V,
McNinch J, Simonet L, Jacobsen F, Xie C, Shutter J, Chyte H, Basu
R, Selander L, Trollinger D, Sieu L, Padilla D, Trail G, Elliot G,
Izumi R, Covey T, Crouse J, Garcia A, Xu W, Del Castillo J, Biron
J, Cole S, Hu MCT, Pacifici R, Ponting I, Saris C, Wen D, Yung YP,
Lin H, Bosselman RA: Identification and cloning of a megakaryocyte
growth and development factor that is a ligand for the cytokine
receptor Mpl. Cell 77:1117, 1994
AL.
16. Kuter DJ, Beeler DL, Rosenberg RD: The purification of
megapoietin: A physiological regulator of megakaryocyte growth
and platelet production. Proc Natl Acad Sci USA 91:11 104, 1994
17. Kuter DJ, Rosenberg RD: The reciprocal relationship of
thrombopoietin (c-Mpl ligand) to changes in the platelet mass during
busulfan-induced thrombocytopenia in the rabbit. Blood 85:2720,
1995
18. Bartocci A, Mastrogiannis DS, Migliorati G, Stockert RJ,
Wolkoff AW, Stanley ER: Macrophages specifically regulate the
concentration of their own growth factor in the circulation. Proc
Natl Acad Sci USA 84:6179, 1987
19. Goldberg MA, Gaut CC, Bunn HF: Erythropoietin mRNA
levels are governed by both the rate of gene transcription and posttranscriptional events. Blood 77:271, 1991
Kuijper JL, Lofton-Day
20. Lok S, Kaushansky K, Holly
CE, Oort PJ, Grant FJ, Heipel MD, Burkhead SK, Kramer J M , Bell
AM, Sprecher CA, Blumberg H, Johnson R, Prunkard D, Ching
AFT, Mathewes SL, Bailey MC, Forstrom J W , Buddle MM, Osbom
SG, Evans SJ, Sheppard PO, Presnell SR, O’Hara PJ, Hagen FS,
Roth GJ, Foster DC: Cloning and expression of murine thrombopoietin cDNA and stimulation of platelet production in vivo [see comments]. Nature 369:565, 1994
21. Shivdasani RA, Rosenblatt MF, Zucker-Franklin D, Jackson
CW, Hunt P, Saris CJM, Orkin SH: Transcription factor NF-E2 is
required for platelet formation independent of the actions of thrombopoietinMGDF in megakaryocyte development. Cell 81:695, 1995
22. Gurney AL, Kuang WJ, Xie MH, Malloy BE, Eaton DL,
de-Sauvage FJ: Genomic structure, chromosomal localization, and
conserved alternative splice forms of thrombopoietin. Blood 85:981,
1995
23. Wendling F, Maraskovsky E, Debili N, Florindo C, Teepe
M, Titeux M, Methia N, Breton-Gorius J, Cosman D, Vainchenker
W: cMpl ligand is a humoral regulator of megakaryocytopoiesis [see
comments]. Nature 369:571, 1994
24. Hunt P, Li YS, Nichol JL, Hokom MM, Bogenberger JM,
Swift SE, Skrine JD, Hornkol AC, Lu H, Clogston C, Merewether
LA, Johnson MJ, Parker V, Knudten A, Farese A, Rou YH, Garcia
A, Stead R, Bosselman RA, Bartley TD: Purification and biologic
characterization of plasma-derived megakaryocyte growth and development factor. Blood 82:540, 1995
RD.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
1995 86: 3668-3675
Murine thrombopoietin mRNA levels are modulated by platelet count
JM McCarty, KH Sprugel, NE Fox, DE Sabath and K Kaushansky
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/10/3668.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.