Leukemia (2000) 14, 941–955
 2000 Macmillan Publishers Ltd All rights reserved 0887-6924/00 $15.00
www.nature.com/leu
CORRESPONDENCE
Arsenic trioxide (As2O3) gradually downregulates tissue factor expression without
affecting thrombomodulin expression in acute promyelocytic leukemia cells
TO THE EDITOR
Arsenic trioxide (As2O3) was shown to induce complete remission in
a high proportion of patients with relapsed and all-trans retinoic acid
(ATRA)-resistant acute promyelocytic leukemia (APL).1 The clinical
response to As2O3 is associated with the induction of nonterminal
cytodifferentiation and the activation of cysteine proteases (caspases)
that are characteristic of apoptosis. We have recently found that vitamin A derivatives, retinoids such as ATRA, exert anticoagulant effects
by upregulating thrombomodulin (TM) and downregulating tissue factor (TF) expression in APL and acute monoblastic leukemia cells.2,3
The effect of ATRA explains the rapid improvement of disseminated
intravascular coagulation syndrome (DIC), which is a complication,
observed in almost all APL patients. Since the mechanism of action
of As2O3 does not involve its binding to retinoic acid receptors (RARs),
in contrast to that of retinoic acids (RAs),4 we have examined the
regulation of TM and TF expression in APL cell line, NB4 cells treated
with As2O3 alone or in combination with ATRA. Our data generally
support a very recent report by Zhu et al,5 which has appeared in
this journal.
After we began treating NB4 cells with As2O3, we replaced half of
NB4 cell-suspended medium by fresh medium containing 10% fetal
calf serum and 1 ␮M As2O3 every 3 days. Such a culture condition
avoids serum starvation, makes cells more resistant to apoptosis
induced by As2O3 and keeps cell concentration rather more constant
than that previously reported in studies in vitro. Treatment of NB4
cells with 1 ␮M As2O3 for 14 days induced morphological changes,
such as condensed chromatin, smaller nuclei, a decreased
nuclei/cytoplasm ratio, and the appearance of granules in the cytoplasm. About 30% of the cells were found to resemble myelocyte
morphologically, and the maturation stage of the remainder of the
cells still corresponded to the promyelocyte stage. Less than 10% of
total cells appeared apoptotic showing nuclear fragmentation. More
than 90% of cells still kept alive by Trypan blue exclusion method.
Viable cell numbers were counted and adjusted.
We measured the changes in the total levels of TF and TM antigens
in NB4 cells incubated with 1 ␮M As2O3 (from Sigma, St Louis, MO,
USA). The TF antigen levels in cell lysates treated with As2O3 decreased
more gradually and modestly (1.37 ± 0.04 at 7 days, 0.81 ± 0.04 at 14
days vs control 2.20 ± 0.12 ng/107 cells) compared to those in cell lysates treated with 0.1 ␮M ATRA. ATRA induced a marked decrease in
TF expression at 1 day (0.40 ± 0.02 ng/107 cells) as reported before.2,3
On the other hand, As2O3 did not induce a change in TM antigen levels
even after treatment for 14 days (2.46 ± 0.16 vs control
2.69 ± 0.10 ng/107 cells). ATRA markedly increased the TM antigen
level (32.4 ± 4.9 ng/107 cells) at 1 day as reported.2,3
NB4 cells were treated with As2O3 and the cell surface TF and TM
activities were measured. Treatment of NB4 cells with As2O3 for 3 days
resulted in a decrease in cell surface TF activity, which occurred more
gradually, and not to the extent induced by 0.1 ␮M ATRA treatment for 1
day. ATRA promptly induced downregulation of TF activity as previously
reported.3 Treatment for a 7-day period was required to decrease the TF
activity to about one-third of the pretreatment levels of the NB4 cells
(Figure 1a).2,3 The percent decrease in cell surface TF activity after 7 days
was two-fold greater than that in TF antigen levels in total cell lysates
(Figure 1a). Cell surface TM activity was measured as the degree of acceleration of thrombin-catalyzed protein C activation. As2O3 stimulation did
not change the rate of protein C activation on the surface of NB4 cells
Correspondence: T Koyama, School of Allied Health Sciences, Faculty
of Medicine, Tokyo Medical and Dental University, 1–5–45 Yushima,
Bunkyo-ku, Tokyo 113–8519, Japan; Fax: 81–3–5803–5882
Received 21 October 1999; accepted 10 January 2000
Figure 1
Changes in TF and TM cofactor activity on the surface
of NB4 cells after exposure to As2O3. (a) NB4 cells were exposed to
1 ␮M As2O3 for 1, 3 or 7 days, or exposed to 0.1 ␮M ATRA for 1 day.
Cell surface TF activity (assay of 106 cells) was determined on the
basis of normal plasma-based one-stage recalcification clotting time
and quantitatively measured by reference to standard curves constructed using human placenta TF as described previously.3 (b) Cell
surface TM cofactor activity in cell suspensions was determined as
described previously.2,3 Basal ⌬OD405 nm/min levels were 0.033 ±
0.002/min/5 × 106 cells (0.72 ± 0.04 APC nM/min/106 cells). These
assays were repeated independently five times and the results are
expressed as the mean ± s.d. The difference between the TF cofactor
activity on the surface of NB4 cells treated with As2O3 for 7 days and
that of untreated cells is statistically significant (P = 0.0005).
(Figure 1b). As2O3 treatment of THP-1 monoblastic leukemia cells for 7
and 14 days did not decrease TF activity on the cell surface.
The downregulation of TF induced by ATRA was partially abolished
by pretreatment of NB4 cells with As2O3 for a 1-day period
(Figure 2a). Such treatment inhibited the upregulation of TM to an
extent about one-third of that induced by ATRA (Figure 2b).
Figure 2
Effect of As2O3 and ATRA in combination on the TF (a)
and TM (b) activities on the surface of NB4 cells. NB4 cells were
pretreated with 1 ␮M As2O3 for 1 day, and subsequently exposed to
As2O3 and 0.1 ␮M ATRA for 1 more day (As + ATRA). The effects of
As2O3 treatment for a 2-day period and ATRA treatment for a 1-day
period of NB4 cells are also shown. These assays were repeated independently four times. The inhibitory effect of pretreatment with As2O3
on TM upregulation by ATRA was significant (P = 0.02).
Correspondence
942
Figure 3
RT-PCR analysis of TF and TM mRNAs in NB4 cells
treated with As2O3. Total RNAs were extracted from cultured NB4
cells after exposure to 1 ␮M As2O3 for 7 or 14 days. RT-PCR analysis
of TF (a) or TM (b) mRNA was performed as previously described.3
Base pair markers are shown to the left. Glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) mRNA was used as a control for assessing
the quantity of mRNA.
To examine the changes in TF and TM mRNA levels, NB4 cells
were treated with As2O3 for 7 or 14 days. In cells treated with As2O3,
the level of expression of TF mRNA decreased more gradually than
that observed in cells treated with ATRA (Figure 3a),3 and As2O3 did
not induce any change in the level of expression of TM mRNA
(Figure 3b).
Treatment of NB4 cells with 0.1 ␮M ATRA for 1 day markedly
reduced the level of expression of TF mRNA (Figure 4a, lane 2),
whereas treatment with 1 ␮M As2O3 for 2 days only slightly reduced
the TF mRNA level (Figure 4a, lane 3). The downregulation of TF
mRNA expression induced by ATRA was partially inhibited by pretreatment of the cells with As2O3 (Figure 4a, lane 4). Expression of
TM mRNA was markedly increased in NB4 cells treated with ATRA
for 1 day (Figure 4b, lane 2), whereas the level of expression of TM
mRNA did not change in cells treated with As2O3 for 2 days
(Figure 4b, lane 3). The upregulation of TM mRNA expression induced
by ATRA was apparently inhibited by pretreatment with As2O3
(Figure 4b, lane 4). According to Western blotting, treatment with
0.1 ␮M ATRA appreciably reduced the level of promyelocytic leukemia (PML)/RAR␣ fusion protein, but the level of PML/RAR␣ decreased
more markedly after treatment with 1 ␮M As2O3. The downregulation
of PML/RAR␣ expression induced by As2O3 was partially inhibited by
cotreatment with ATRA.
During the period of As2O3 therapy, the levels of the markers indica-
tive of coagulopathy and hyperfibrinolysis fibrinogen/fibrin degradation
products (FDP) and D-dimers gradually decreased and thus clinical
improvement of DIC was also observed.1,5 Whereas APL patients
treated with ATRA show rapid improvement of coagulopathy within a
few days, DIC in patients treated with As2O3 improves in 1 week or
more (personal communication by Dr Zhou Jin, First College of Harbin
Medical University, Harbin, People’s Republic of China). The effect of
As2O3 in inducing modest TF downregulation without TM upregulation,
as shown in this study, seems to be compatible with the gradual and
slow improvement of DIC in patients treated with As2O3.
Since As2O3 may neither interact nor activate a RA-responsive
element in the TM gene, it may not be able to upregulate TM
expression. The effects of As2O3 could be due to the rapid modulation
and degradation of PML/RAR␣ fusion protein induced by As2O3. As
a result of the disappearance of the PML/RAR␣ fusion protein, which
might have some stimulatory effect on TF gene expression in the APL
cell line NB4, the level of TF expression may gradually decrease.
Whereas ATRA has a positive- or negative-acting effect by inducing
the formation of complexes with its receptors, As2O3 does not form
such complexes to directly regulate TM and TF gene expression. The
repression of TF by As2O3, therefore, progresses more slowly than that
in the case of ATRA.
For the effect on cell differentiation induced by ATRA, the presence
of the PML/RAR␣ fusion protein is rather necessary. Also, it has been
reported that As2O3 inhibits ATRA-induced, rapid differentiation of
NB4 cells when applied in vitro in combination with ATRA.6 In our
study, pretreatment with As2O3 partially abolished the rapid TM upregulation and TF downregulation effects of ATRA. The rapid disappearance of the PML/RAR␣ fusion protein might interfere with the action
of ATRA on TM and TF regulation. These phenomena indicate that
As2O3 influences the RA-induced pathway for regulation of these
genes. Whereas the modulation of TF by As2O3 appears to occur in
parallel with cell differentiation or apoptosis, further investigations are
necessary to elucidate the precise mechanism involved in these
functions.
Acknowledgements
This work was supported in part by research grants from the Atsuko
Ouchi Memorial Fund, Tokyo Medical and Dental University, from
the Research Foundation for Clinical Pharmacology.
M Ohsawa1
T Koyama1
M Shibakura1
S Kamei1
S Hirosawa2
1
School of Allied Health Sciences, and
First Department of Internal Medicine,
Tokyo Medical and Dental University,
Tokyo, Japan
2
References
Figure 4
RT-PCR analysis of TF (a) and TM (b) mRNAs in NB4
cells treated with As2O3 and/or ATRA. Total RNAs were extracted
from cultured NB4 cells after exposure to 0.1 ␮M ATRA for 1 day (lane
2), 1 ␮M As2O3 for 2 days (lane 3), or ATRA and As2O3 for 1 day after
1-day preincubation with As2O3 (lane 4).
Leukemia
1 Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qui QY, Zhu J, Tang
W, Sun GL, Yang KQ, Chen Y, Zhou L, Fang ZW, Wang YT, Ma
J, Zhang P, Zhang TD, Chen SJ, Chen Z, Wang ZY. Use of arsenic
trioxide (As2O3) in the treatment of acute promyelocytic leukemia
(APL): II. Clinical efficacy and pharmacokinetics in relapsed
patients. Blood 1997; 89: 3354–3360.
2 Koyama T, Hirosawa S, Kawamata N, Tohda S Aoki N. All-trans
retinoic acid (ATRA) upregulates thrombomodulin and downregulates tissue factor expression in acute promyelocytic leukemia
cells. Blood 1994; 84: 3001–3009.
3 Shibakura M, Koyama T, Saito T, Miyasaka N, Kamiyama R, Hirosawa S. Anticoagulant effects of synthetic retinoids mediated via
different receptors on human leukemia and umbilical vein endothelial cells. Blood 1997; 90: 1545–1551.
4 Wang Z-G, Rivi R, Delva L, König A, Scheinberg DA, GambacortiPasserini C, Gabrilove J, Warrell RP, Pandolfi PP. Arsenic trioxide
and melarsoprol induce programmed cell death in myeloid leukemia cell lines and function in a PML and PML-RAR␣ independent
manner. Blood 1998; 92: 1497–1504.
5 Zhu J, Guo WM, Yao YY, Zhao WL, Pan L, Cai X, Ju B, Sun GL,
Correspondence
Wang HL, Chen SJ, Chen GQ, Caen J, Chen Z, Wang ZY. Tissue
factors on acute promyelocytic leukemia and endothelial cells are
differently regulated by retinoic acid, arsenic trioxide and chemotherapeutic agents. Leukemia 1999; 13: 1062–1070.
6 Shao W, Fanelli M, Ferrara FF, Riccioni R, Rosenauer A, Davison
K, Lamph WW, Waxman S, Pelicci PG, LoCoco F, Avvisati G,
Testa U, Peschle C, Gambacorti-Passerini C, Nervi C, Miller WH
Jr. Arsenic trioxide as an inducer of apoptosis and loss of
PML/RAR␣ protein in acute promyelocytic leukemia cells. J Natl
Cancer Inst 1998; 90: 124–133.
943
Cellular cytotoxic drug sensitivity in children with acute leukemia and Down’s
syndrome: an explanation to differences in clinical outcome?
TO THE EDITOR
During the last decade several groups have reported that acute
myeloid leukemia (AML) in children with Down’s syndrome (DS) has
a significantly better outcome than AML in other children.1,2 In acute
lymphoblastic leukemia (ALL), on the other hand, the clinical outcome tends to be worse for DS than for non-DS children.3,4 The reason
for this difference between AML and ALL is not known.
In an ongoing study we prepared leukemic cells at diagnosis from
62 children with AML and 222 children with ALL for test of in vitro
sensitivity to cytotoxic drugs. Five children with AML (8%) and five
children with ALL (2.3%) also had DS, figures similar to those previously reported in population-based studies, and this enabled us to
compare drug sensitivity in tumor cells from DS and non-DS children.
Leukemic cells from fresh bone marrow or blood samples were
assessed for their in vitro drug sensitivity by the fluorometric microculture cytotoxicity assay (FMCA) against a panel of drugs used in the
current Nordic protocols for AML2 and ALL,5 respectively. The FMCA
is a non-clonogenic assay, very similar to the more widely used MTT
assay,6 and is based on the measurement of fluorescence generated
from hydrolysis of fluorescein diacetate to fluorescein by cells with
intact plasma membranes.7 A survival index (SI) was calculated as the
fluorescence signal of cells after 72 h of drug incubation as a percentage of that of unexposed controls.
Leukemic cells from DS children with AML were significantly more
sensitive (P = 0.03–0.002) than blast cells from non-DS children to
cytosine arabinoside (Ara-C), doxorubicin, dexamethasone (see Figure
1a) and amsacrine (not shown). A trend in the same direction was
found for etoposide (VP16) and 6-thioguanine. DS patients with AML
were significantly younger (mean 2.1 years) than non-DS patients (7.9
years), but a statistically significant difference was found also when
comparing the DS samples with those of age-matched controls, and
there was no clear correlation between age and in vitro sensitivity
among non-DS patients with AML.
Three Down patients had FAB subtype M1 and two had M7 AML.
In contrast, leukemic cells from DS children with ALL were significantly less sensitive (P = 0.02–0.001) than cells from non-DS patients
to dexamethasone and asparaginase, with a similar trend for vincristine, doxorubicin and Ara-C (Figure 1b). The mean age for DS children
with ALL was 10.8 years and for non-DS patients 6.5 years (NS). Two
DS patients fulfilled standard risk and three intermediate risk criteria
according to the Nordic ALL protocol.
Findings similar to ours have previously been described in DS children with AML,8,9 while in ALL a single study on six DS patients
showed no significant differences for most drugs tested (ALL cells of
DS patients were slightly more sensitive to anthracyclines).9
The inferior clinical outcome in ALL has been attributed to excessive therapy-related toxicity in DS children, but this would not explain
the difference between AML and ALL. Our data, which have to be
confirmed in larger series of patients, indicate that at least part of the
explanation for the differences in clinical outcome between DS and
non-DS children should be sought among mechanisms for tumor cell
drug resistance. The fact that the difference in drug sensitivity between
DS and other children was diametrically opposed for AML and ALL,
suggests cell lineage-specific rather than general cellular alterations
Correspondence: G Lönnerholm, Akademiska Barnsjukhuset, SE75185 Uppsala, Sweden; Fax: 46 18 665853
Received 20 October 1999; accepted 22 December 1999
Figure 1
In vitro sensitivity of leukemic blast cells from children
with Down’s syndrome and acute myeloid leukemia (AML; n = 5),
upper panel, or acute lymphoblastic leukemia (ALL; n = 5), lower
panel, compared to non-Down children with leukemia (n = 57 and
217, respectively). SI denotes surviving cells, in percentage of
untreated control, after 72 h incubation with cytosine arabinoside
0.5 ␮g/ml (Ara-C), doxorubicin 0.5 ␮g/ml (Dox), etoposide 5 ␮g/ml
(VP16), 6-thioguanine 10 ␮g/ml (6-TG), dexamethasone 7.1 ␮g/ml
(Dexa), vincristine 0.5 ␮g/ml (Vcr), asparaginase 10 U/ml (Asp). The
results are presented as mean values plus s.e.m. The statistical significance levels (Student’s t-test) are indicated; NS, non-significant.
associated with the presence of three copies of chromosome 21 as
suggested by Taub et al.8 Presently, we do not understand the
mechanism(s) behind the differences in drug sensitivity between DS
and non-DS children, but we think our data point at an area for further
fruitful research.
Leukemia