Quercetin Inhibits the Growth of Leukemic Progenitors and

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Quercetin Inhibits the Growth of Leukemic Progenitors and Induces the
Expression of Transforming Growth Factor-P1 in These Cells
By Luigi M. Larocca, Luciana Teofili, Simona Sica, Mauro Piantelli, Nicola Maggiano, Giuseppe Leone,
and Franco 0 . Ranelletti
We previously showed that quercetin (3,3',4',5,7 pentahydroxyflavone) inhibits ina
dose-dependent manner the
growth of acute leukemias and is able to enhance the antiproliferative activity of cytosine arabinoside. We show here
that quercetin inhibits the clonogenic activity of 20 of 22
acute leukemias (AL; 4 MI-AML, 3 MS-AML, 2 MS-AML, 3
M4-AML, 3 M5-AML, and 7 ALL). In the present report, we
show that the induction of transforming growth factor-PI
(TGF-PI) in leukemic blasts is one of the growth-inhibitory
mechanisms of quercetin in these cells.This observation
was supported by thefollowing data. (1)Quercetin-sensitive
leukemic blasts, when treated with quercetin, secrete large
F
LAVONOIDS AREA class of natural substances widely
distributed in the plant kingdom' that display a variety
of biologic actions.' Recently, we and others haveshown
(3,3',4',5,7-pentahydroxyflathat the flavonoidquercetin
vone) inhibits the growth
ofhuman leukemic cell lines'^'
and of acute myeloid leukemia (AML) and acute lymphoid
leukemia(ALL) primary leukemic blasts.' Moreover, we
have shown that quercetin is able toinhibit the colonyformation by AML and ALL progenitors.' Furthermore, quercetin
enhances the antiproliferative activity of cytosine arabinoside on the leukemic cell line HL60 and on fresh AML and
ALL progenitors,'likely by interactingwiththeso-called
type I1 estrogen binding sites (type I1 EBS)."'
The term of transforming growth factor-p (TGF-P) indicates a family of structurally related proteins with powerful
hematopoieticregulatory properties."' TGF-p I stimulates
or inhibitstheinvitro
growth of normalmyeloidprogenitors9. I?- 15 depending on their stage
of differentiation.
TGF-01 mightselectivelyact
as a negativeregulatoron
early hematopoietic progenitors,sparingmore
committed
Severalstudiesshowed
the inhibitoryeffect of
TGF-p1 on the growth of primary AML blasts in all samples
tested but one."^"
Because wehave previously observed that quercetin inhibits the growth of ovarian cancer cell line by inducing the
expression of TGF-P 1 ,2' we evaluated whether the modulation of TGF-PI production by quercetin could be a mechanism by whichthisflavonoid
exerts itsgrowth-inhibitory
effect on AML and ALL blasts.
From the lstituti di Anatomia Patologica, Serneiotica Medica and
Istologia, Universita Cattolica del Sacro Cuore, Romu, ltuly.
Submitted July 13. 1994; uccepted Januar?, 30, 199.5.
Supported by grants 40% and 60% from MURST and by Grant
No. ACRO 94,01098 PF39from CNR.
to LuigiM.Lurocca,MD,lstitutodi
Addressreprintrequest
Anatomiu,PatologicaUniversit2CuttolicaSC,Largo
F. Vito. I ,
Romu 00168, Italy.
The puhlicution costsof this article were defrayed in part page
by
chargepayment.Thisarticlemustthereforebeherebymarked
"advertisement" in accordance with 18 U.S.C. section 1734 solely t o
indicate this ,fact.
0 1995 by The American Society o j Hernatology.
0006-497//95/R512-0027$3.00/0
3654
amounts of TGF-P1 in the medium and show positivity for
TGF-PI-immunoreactive material in thecytoplasm. (2) At a
concentration of 8 pmol/L, antisense TGF-P1 oligonucleotides prevent the growth-inhibitory action of quercetin. (3)
Anti-TGF-p1 neutralizing monoclonal antibodies can prevent almost completely the growth-inhibitory activity of
quercetin. The analysis of quercetin-resistant casesconfirmed as well the central role of TGF-P1 in the growthinhibitory activity of quercetin. In conclusion, quercetin can
act as a cytostatic agent for leukemic cells by modulating
the production of TGF-PI.
0 1995 by The American Society of Hematology.
MATERIALS AND METHODS
Isolation cf leukemic blasts rrnd clonogenic as.say. Fresh leukemic blastswereisolatedfromheparinizedbone
marrow aspirates
obtained from 20 leukemic patients ( l 4 AML and 6 ALL) at diagnosis and from 2 patients in relapse ( I AML and I ALL). All samples
were obtained after informed consent was obtained. Diagnoses were
established according to the French-American-British recommendations?'.:" after cytologic examinations and cytochemical assays (peroxidase and naphtyl acetate esterase). Immunologic phenotype was
defined in ALLpatients.Aspiratesdiluted
with Hank'sBalanced
SaltSolution(HBSS)werelayeredontoFicoll-Hypaquedensity
gradient ( I ,077 g/mL,pH 7.6,292 mOsm/L) and centrifugedat 400g
for 30 minutes. Light-density mononuclear cells containedmore than
70% blast cells. In AML and in non-T ALL, T lymphocytes were
removed by rosetting with neuraminidase-treated sheep erythrocytes.
After T-cell depletion, the percentage of blasts in the resulting cell
population was greater than 90%, as assessed by morphologic, cytoenLymatic. andimmunophenotypicanalysis.Blastsweresuspended at I X 10' cells/mL in Iscove's modified Dulbecco's medium
(IMDM: Flow Laboratories, Irvine, UK) containing 0.9% methylcelx lo-' mol/L2-mercaptoethanol
lulose (Sigma, St Louis, MO). 2
(Sigma), 20% heat-inactivated fetal calf serum (HI-FCS; Flow Labs),
and 10% phytohemagglutinin (PHA)-leukocyte-conditioned medium, prepared accordingtothe method of Aye ct al." Aliquots
of 0 . 1 mL were plated in 96-microwell flat-bottom plates (Becton
Dickinson, Lincoln Park. NJ). All plates were incubated at 37°C i n
a fully humidified S% C02-9S% air atmosphere. Quercetin (Aldrich.
Steinheim,Germany)wasadded
at the indicated concentration at
the start of cultures and then every 2 days from an absolute ethanol
stock solution. Ethanol (vehicle) concentration never exceeded I %
(vol/vol) in untreated and treated cultures. Each samplewas cultured
in quadruplicate. After 6 to 10 days, aggregates consisting of 10 o r
more cells were counted as colony-forming unit-leukemic (CFU-L).
Result are expressed as the number of colonies (mean of 4 plates)
per IO'cells plated. To verify the leukemic natureof CFU-ALL, cells
i n coloniesfromALLpatientswerestained
with Wright-Giemsa.
peroxidase, and nonspecific esterase.
TGF-81 u r d TGF-82 NSSCZJ. Purified ( > 9 5 % ) primary leukemic
blasts (2 X 10" cells/mL) were cultured, for various lengths of time.
i n 35 X 10 mm Petri dishes (Falcon; Becton Dickinson) in RPM1
1640 medium (GIBCO Biocult, Paisley, UK) supplemented with 2
mmolL L-glutamine, 1 0 0 U/mL penicillin, and 2% HI-FCS i n the
presence of l wmol/L quercetin. An equivalent amount of ethanol
was added to cells as a control.At the end of incubation the medium
was collected with siliconized pipet tips, combined with Apoprotin
(final concentration, 0.04 trypsin inhibitor units/mL: Sigma), centrifuged at 8001: for 15 minutes in siliconized tubes to remove cellular
Blood, Vol 85, No 12 (June 15).
1995:
pp 3654-3661
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
QUERCETININDUCESTGF-P1
3655
IN LEUKEMICBLASTS
Table l.Quometin InhibitoryCapacity of Colony Formation and
Clonogonic Efficiency in Acute Leukemic Subtype8
Case
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Leukemic Cell
Subtype
ID 50%.
ipmoVLl
Clonogenic
Efficiencyt
M1
M1
M1
M1
M2
M2
M2
M3
M3
M4
M4
M4
M5
M5
M5
ALL
ALL
ALL
ALL
ALL
ALL
8.0
30.0
4.0
Resistant
10.0
Resistant
3.0
0.3
0.5
0.6
0.1
0.06
0.2
0.08
0.8
0.01
0.04
0.07
0.4
91
87
154
100
15
8
* The ID 50% was
65
186
125
132
53
46
13
6
18
12
41
23
80
12
21
0.03
0.01
calculatedfrom the analysis ofthe dose-response
curves.
t Number of CFU-L perlo'cells plated. Results shownare the mean
of quadruplicate cultures
in the presence of vehicle
alone (1% ethanol,
vol/vol). The standard deviationfor each value was less
than 10% and
was omitted.
debris, and then stored at -80°C until use. To evaluate the concentration of TGF-p1 and TGF-p2 in conditioned medium, we useda
sensitive radioimmunoassay (New England Nuclear Research products, Dupont, Boston, MA) and an immunoenzymatic method (R &
D System, Minneapolis, MN) strictly specific for TGF-p1 and TGFp2 detection, respectively. Each assay was performed according to
the instructions of the manufacturer. To accurately quantify TGF-p1 ,
samples were activated using a two-step acidificatiodneutralization
method. Briefly, 100 pL samples were incubated in the presence of
10 pL 1.2 NHCl for 15minutes at m m temperature &er vortexing.
Then, 20 pL of 0.5 m o m HEPEWO.72 m o m NaOH was added to
neutralize. The measured TGF-p1 value was multiplied by1.3 to
correct for the dilution. This procedure allowes the quantitation of
total (active + latent) TGF-PI.
Immunohistochemical analysis. Primary leukemic blasts (2 X
lo6celldmL) were cultured for 12hours in 35 X 10 mm Petri dishes
(Falcon; Becton Dickinson) in RPM1 1640 medium supplemented
with 2 mmom L-glutamine, 100 U/mL penicillin, and 2% HI-FCS
at 37°C in a fully humidified 5% C02-95% air atmosphere in the
presence of 1 p m o m quercetin. An equivalent amount of ethanol
was added to control cells. At the end of culture, cells were incubated
for 10 minutes in Caz+/MgZc-freecold PBS containing 0.02%ethylenediaminetetracetic acid, harvested by gently pipetting, and washed
three times with phosphate-buffered saline (PBS) supplemented with
2 mg/mL ofbovine serum albumin (Flow Labs). Cytospins prepared
with Shandon Cytospin (Shandon, Cheshire, UK) were fixed with
4% para-formaldehyde in PBS for 10 minutes at room temperature
(RT) and permeabilized with cold (-20T) methanol for 10 minutes.
Cells were then washed with PBS and incubated for 1 hour at RT
with an anti-TGF-p1 mouse monoclonal antibody (MoAb; R & D
System; clone TB 21; 10 pg/mL) or with an unrelated mouse MoAb
isotype-matched control. Hydrogen peroxide, normal goat blocking
serum, biotinylated Igs, avidin-biotin complex, and3-amino-9ethylcarbazole substrate solutions were used according to the manufacturer (ABC ELITE detection system; Vector, Burlingame, CA).
Cells were lightly counterstained with Mayer's hematoxylin and
mounted with CrystallMount (Biomeda, Foster City, CA). In the
negative controls, no immunostaining was detectable. The positivity
of the reactions were independently assessed in blind fashion by two
pathologists; 100 consecutive cells in three or more fields were
counted. The overall intensity of the staining was arbitrarily scored
as negative or positive. The results were expressed as the percentage
of positively stained cells regardless of intensity.
Inhibition of TGF-p1 synthesis by antisense oligonucleotide.
The sequences of the phosphorothioate oligonucleotides were as
follows: TGF-p1 antisense, 5'-CCCGGAGGGCWCGGGGGA3'; TGF-p1 sense, 5'-TCCCCCmCGCCCTCCGGG-3'; TGFp 1 missense, 5'-GGCGAGCGAGTGAGCGCGCGG-3"4(ATG ini-
n
A
Fig 1. Tim.cwrw of TGF-p1
production by AML"4 (AI and
ALL IB) blnta culturedinthe
presence (0)or absence (0)
of 1
pmol/L quercetin. Blasts(2x W
d l r l m L ) were cultured for the
indicated time with or without
quercetin; at tho end of the culture period, tho conditioned m a
dia
were
hanrosted
and
the
amount of total (activated Iatent) TGF-p1 was evaluatedby a
radioimmunoassay er repotted
in Materials and Methoda.
+
s
e
Sa
W
m
0
8
16 24 32 40 48
(hr)
B1
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
LAROCCA ET AL
3666
tiation codon or its complement underlined in the sense and antisense
sequences). Phosphorothioate oligonucleotides were purchased high
performance liquid cbmatography (HPLQ-puriiied from MedProbe (Oslo, Norway). Cells were incubated
in the presence of oligonucleotides at the final concentration of 8 pmol/L, essentially as
described by others for evaluation of differentgenesexpresion.'^^^^ Oligonucleotides or control medium were added to the
cells in IMDM in the absence of serum. Two hours later, 10% HIFCS was added to the liquid culture. After 15 to 17 hours, oligonucleotides or control medium were added again andcells were plated
as described abovein a semisolid culture
medium containing quercetin at various concentrations(0.001 to 10 pmol/L) or vehicle alone.
Cm-Ls were scored after 6 to 10 days.
TGF-@activity neutralizing experiments. To test the prevention
of the growth inhibitory effectof quercetin by a monoclonal antiTGF-01, $2,-83 neutralizing antibody (anti-TGF-ps MoAb;
Genw e ) , leukemic blasts were plated, as reported above in semisolid
culture medium containing the anti-TGF-PS MoAb
andor quercetin
at the indicated final concentrations.Matched control cultures contained unrelated MoAb. CFU-Ls were scored after 6 to 10 days.
RESULTS
Inhibition of CFU-L by quercetin. As shown in Table 1,
thecolonyformationbyleukemic
cells was inhibited by
quercetin in all but 2 cases (patients no. 4 and 6)that were
resistant to quercetin at a concentration up to 100 pmol/L.
Confirming previous reports:2o blasts with high clonogenic
efficiency were less sensitive to quercetin than those with a
low clonogenic capacity.In fact, the quercetin concentration
inhibiting growth by 50% (ID 50%) positively correlated
with the clonogenic efficiency of leukemic cells ( r = .81; n
= 19; P < .001). A"M3, "4,
and -M5 and commonALL were highly sensitive to quercetin, with an ID 50%
ranging from 0.1 to 0.01 pmol/L. &-M1 and
-M2 were
less sensitive to quercetin, including the 2 resistant cases.
Secretion of TGF-p1 by leukemic blasts treated with quercetin. To ascertain whether quercetin is able to stimulate
the productionof TGF-B1 and TGF-p2, the amount of these
cytokines were measured at various time points during 48
hours of culture in 1 &-M4 case (patientno. 10; Fig 1A)
and in 1ALL case (patientno. 16; Fig 1B).A " M 4 blasts
constitutively released total (active
latent) TGF-p1 that
was evaluable in the medium after 24 hours of culture. In
the presence of 1 pmol/L quercetin, a peak of TGF-p1 release was evident after 4 hours (Fig lA), followedbya
decrease at 12 hours to about 42% of the peak value. The
concentration of TGF-B1 did not further change until the
end of culture period. The cell number was stable during
the culture period both in quercetin-treated and in control
cultures.
ALL blasts did not show constitutiverelease of total (active latent) TGF-p1 during the 48-hourperiod of culture.
+
+
Fig 2. Irnrnunolocalization of TGF-p1 in AML-M5 (A and B) and in ALL (C and D) blasts culturedfor 12 hours with (B and D) or without (A
and C) 1 prnollL quercetin. Original magnification x 800.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
QUERCETININDUCESTGF-P1
3657
IN LEUKEMIC BLASTS
Table 2. Effect of Quercetin on TGF-B1 Expression in Blast-Enriched Bone Marrowa From AML (baeNo. 14)
and ALL (Care No. 20) Leukemic Patianta
ALL
AML-M5
Cell Lineage
(3%)t
Granulocytic
Monocytic
Erythroid
(-1
(2%)
(95%)
Blast
Untreated"
Treated
51 t W
95 t 5
20 t 4
20 2 4
60 t 9
85 t 8
-
-
Untreated
Treated
(2%)t
(1%)
60t5
823
98 t 4
40 t 7
(1%)
36 5 4
0
56 2 5
82 t 6
(96%)
Cells were cultured as outlined in Materials and Methods for 12 hours with vehicle alone (ethanol 0.1%, vol/vol) or 1 pmol/L quercetin.
t Percentage of each celltype as assessed by cytoenzymatic and immunophenotypic analysis.
Percentage of TGF-01 immunoreactivecells. Results are expressed as the mean t SD of two independent quadruplicate counts.
*
In the presence of 1 pmol/L quercetin, ALL blasts showed
a peak of TGF-P1 release after 12 hours of culture (Fig lB),
followed by a decrease at 24 hours to about 50% of the peak
value. The number of ALL blasts after 48 hours of culture
was 80% of the initial number of cells plated both in quercetin-treated and in control cultures.
The concentration of TGF-P2 was below the detection
levels of the enzyme-linked immunosorbent assay (ELISA)
sensitivity in all samples tested (data not shown).
Immunolocalization of TGF-PI. To determine if quercetin enhances the intracellular levels of TGF-P1 , immunohistochemical assays using an anti-TGF-Dl MoAb were performed on blasts from 1 patient with AML-M5 (patient no.
14) and 1 with ALL (patient no. 20) cultured for 12 hours
with or without 1 pmol/L quercetin. As shown in Fig 2, in
both cases quercetin enhances the intracellular content of
anti-TGF-Bl immunoreactive materials. The immunoreac-
100m
0
1
tive product was localized in the cytoplasmic compartment.
The blasts of the patient with AMGM5 expressed higher
basal levels of TGF-P1 than the blasts of the patient with
ALL. The presence of vehicle (ethanol) in the culture medium did not modify the basal level of TGF-p1 expression
(data not shown). As shown in Table 2, the percentage of
blasts in leukemic bone marrow preparations was 295%.
With due caution given the low number of cells observed,
it could be noted that in the remaining contaminating population (1) most segmented and immature granulocytes showed
constitutive expression of TGF-P1 not modified by quercetin
treatment and (2) in both monocytic and erythroid cell lineages quercetin increased the percentage of TGF-P1 immunoreactive cells.
Effects of antisense TGF-PI oligonucleotide and of antiTGF-PI, -P2, $3 neutralizing MoAbon the inhibition effect
of CFU-L by quercetin. The aim of our study was to evaluate the role of TGF-P1 in quercetin-dependent inhibition of
CFU-L formation. To inhibit TGF-P1 production we used
an antisense oligonucleotide. An anti-TGF-PS MoAb was
used to neutralize TGF-PS produced by leukemic blasts. A
80 -
60-
rw
0
g 4020 -
- 0
10-'O10-~IO-*10"
io-'
None
Q concentration (M)
Fig 3. CFU-ALL inhibition byq u a d n alone (0)or in the preaonce
of 8 pmol/L antisenae (A)or miaaonae (0)
TGF-B1 oligonucleotides
is dependent onthe concentration of quercetin. Cdla 110' cella) were
treeted with quercetin and TGF-B1 oligoa as reported in Matoriala
and Methods.R w l t a ahown ora the mean of four replicatea for aach
variable. Standarddoviatbns were baa than 10% and were omitted.
The number of control CFU-ALL was 41 k 3 for 10' cells plated.
Q
Q+antisense
Treatments
Fig 4. TGF-B1 production bythe r m e ALL bloats shown in Fig 3
cultured for 12 houra in the pr#snco of vehicle alono Inona) or 1
pmol/L quercetin (a) or lpmol/L quercetin plua8 pmol/L antiunae
TGF-B1 oligoa (Q a n t i r n r ) . At the end of the Culture period, the
conditioned mediawere hanreated and
the amount of total (ectivatod
+ Iatont) TGF-B1 waa evaluated bya radioimmunoasaayaa reported
in Materiala end Methods.
+
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
LAROCCA ET AL
3658
120 ]
1
120
100
AI
100
B(
4
E 80
2
2 60
rcl
80
S8
60
% 40
40
0
20
20
0
0
0
0.1
1
10
Q concentration (PM)
120
g
4
cl
100
120 1
2
4
80
100
80
360
3 60
%l
E
9
0
0.1
l
10
Q concentration (PM)
+4
40
O
8
20
0
(m),
40
20
0
0
0.1
1
10
Q concentration (PM)
preliminary dose/response curve showed that oligonucleotide concentrations greater than 10 pmol/L were toxic (data
not shown). A final concentration of 8 pmol/L was used in
all experiments. To showthat antisense TGF-PI oligonucle&dab could inhibit quercetin-induced cytokine production,
1 case of flavonoid-sensitive ALL (ID SO% -0.4 pmolL)
in which antisense TGF-P1 oligonucleotides prevented the
action of quercetin (Fig 3) was analyzed for the production
of TGF-P1 in the presence of TGF-P I antisense oligonucleotides. As shown in Fig 4, TGF-P1 antisenseoligonucleotides
could prevent the quercetin-induced cytokine production.
Figure S shows the effectof TGF-P1 antisense oligonucleotideand neutralizing anti-TGF-psMoAb on quercetindependent inhibition of leukemic blasts growth. In I AMLM4 (patient no. IO, Fig 5A) and in 1 ALL (patientno. 19,
Fig SB) the addition of antisense but not sense(data not
shown) or missense TGF-P1 oligonucleotides prevented the
inhibitoryaction of quercetin. The effect of the TGF-P1
antisense oligonucleotides was proportional to the flavonoid
concentrations. The prevention of quercetin effects by antiTGF-PS MoAb was concentration dependent, resulting in a
total prevention at an antibody Concentration of 100 pg/mL.
In the quercetin-resistant AML-M2 case (patient no. 6, Fig
SC), TGF-/3 1 antisense oligonucleotides produced an inhibition greater than SO% on CFU-L formation that was independent of the quercetinconcentrationused.
Moreover, antiTGF-PS MoAb inhibited in a dose-dependent manner CFULformation, in this case with a maximum activityat 100
pg/mL anti-TGF-0s MoAb. In the quercetin-resistant AMLM1 case (patient no. 4, Fig SD), neither TGF-P1 antisense
oligonucleotides nor anti-TGF-Ps MoAb produced any evident effect on CFU-L. An isotype-matched MoAb was used
Fig 5 . Effect of antisense
TGF-p1 oligos and neutralizing
anti-TGF-psMoAb
on quercetin-mediated inhibition of leukemic blasts. (AI AML-M4 (patient
no. 10 of Table 1); (B) ALL (patient no. 19of Table ll;(Cl AMLM 2 (patient no. 6 of Table 1); (Dl
AML-M1 (patient no. 4 of Table
1). Cells were plated atlo5 cells/
mL, as reported in Materialsand
Methods, in the presence of vehicle alone (ethanol l % , vol/voll
or quercetin (.l at the indicated
concentrations. Eight micromoles of antisense (0)
and missense oligonucleotides ( B ) were
used alone or in combination
with various quercetin concentrations.
Anti-TGF-ps
MoAb
were added at 25 pg/mL (€4). 50
pg/mL
and 100 pg/mL (0)
with or without quercetin at various concentrations. Thenumber of control CFU-L are reported
in
Table
1.
Results were expressed as the mean of quadruplicate cultures. Standard deviations were less than 10% and
were omitted.
0
0.1
1
10
Q concentration (PM)
as a control for anti-TGF-0s MoAb; no effects on CFU-L
formationwas observed even at the highestconcentration
used (100 pg/mL; data not shown).
The addition of TCF-Pl antisense oligonucleotides prevented thc inhibitory effect of 1 pmollL quercetin in all t.he
quercetin-sensitive cases (Fig 6). The extent of the effect of
TGF-P1antisense oligonucleotidesvaried amongthe patients and did not seem to be related to the sensitivity of the
leukemic blasts to quercetin.
DISCUSSION
Although originally described as a negative regulator of
normal myeloid progenitor cell growth, recent studies have
shown that TGF-01 is a bifunctional regulator of hematopoietic cells. TGF-P1 can either
or
stim~late'~.''.'~
the growth of murine and human hematopoietic progenitor
cells; in particular, TGF-P1 is a potent inhibitor of primitive
hematopoietic progenitors, whereas the growth of more committed progenitor cells is either not affected or stimulated
by TGF-PI."'
Previous studiessuggested that the growth of most human
leukemic celllinesand of the vast majority ofcells from
AML patients was inhibited by TGF-P1.'y-27 The datapresented here show that the bioflavonoid quercetin inhibits the
growth of AML and ALL blasts in vitro by the induction of
TGF-P1 production. This finding is based on the following
observations: (1) quercetin enhances the intracellular content
and the secretion of TGF-01 by sensitive leukemic blasts;
(2) the inhibitory action of quercetin can be partially blocked
by TGF-01 antisenseoligonucleotidesandcan
be almost
totallyabolished
by neutralizing anti-TGF-PI, $2, -03
MoAb.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
3659
QUERCETININDUCES TGF-Dl IN LEUKEMICBLASTS
100
M3 - M5
M1 - M2
ALL
0
0
80
60
40
20
0
I
~~
I
I
I
O-2
I
I
m
m
6
8
4
0
A
I
I
I
v3
E
Fig 6. Antisense TGF-p1 oligonucleotides prevent the inhibitory actionof quercetin (1 pmol/L) on CFU-L. Cells were plated at lo6 cells/
mL, as reported in Materials and Methods,in the presence of quercetin elone(Q) or in combinationwith 8 pmol/L antisense (as) or missense
(mSI oligonucleotides. The number of control CFU-L were reportedin Teble 1. Results were expressed as the mean of quadruplicate cultures.
Standerd deviations wereless than 10% and were omitted. Patient nos.: panelM1-M2, (0)1, (0)2, (0)3, (D)4, (A) 5, (A)6, ( 0 )
7; panel M3M5, (0)8, IO) 9,101 10, (D) 11, (A) 12, (A)13, ( 0 )
14, ( + l 15; panel LLA, (0)16, (0)17, ( 0 )18,(D) 19, (A) 20, (AI 21.
Because it has been reported that retinoic acid, which also
induces TGF-P1 production, is able to upregulate TGF-01
receptor expression on the HL-60 cell line,33it seems interesting to investigate whether quercetin can act in a similar
fashion.
The differences between TGF-01 antisense oligonucleotides and anti-TGF-Ds MoAb in preventing the effect of
quercetin could be explained as follows. (1) Neutralizing
antibodies but not antisense oligonucleotides can block the
effect of TGF-01 already present in quercetin untreated cells
as shown by immunohistochemistry. If this is true, quercetin
could stimulate the secretion in addition to the synthesis of
TGF-01. (2) TGF-Pl antisense oligonucleotides are not able
to completely inhibit TGF-01 synthesis, as shown in the
case of the ALL presented in Fig 4. (3) The neutralizing
anti-TGF-0s MoAb used in this study is able to neutralize
TGF-/33 in addiction to TGF-01. It isthen possible that
quercetin could stimulate the release of TGF-P3 as well.
TGF-02 is not involved in the quercetin-dependent growth
inhibition because it is not produced by quercetin-treated
leukemic blasts.
Two patients (nos. 4 and 6) behaved differently from all
other patients. Particularly, in both cases leukemic blasts
were resistant to the growth-inhibitory effect of quercetin.
In patient no. 4, because quercetin induces TGF-P1 produc-
tion in leukemic blasts (data not shown), this resistence could
be dependent on the unresponsiveness of these cells to TGFD l . The growth of leukemic blasts from patient no. 6 was
stimulated by TGF-@l.Actually, antisense oligonucleotides,
blocking the synthesis of TGF-Pl, induced a reduction of
more than 50% of the clonogenic activity in this case. Furthermore, 100 p g / d of neutralizing anti-TGF-@ MoAb
inhibited almost abolished CFU-L activity (Fig 5C). In this
case, quercetin did not significantly increase the release of
TGF-P1 above background (data not shown) and was unable
either to stimulate or inhibit the leukemic cell growth. These
data are in accordance with a recent report’’ about heterogeneous responses of leukemic cell lines and primary leukemic
blasts to the growth-regulatory action of TGF-@l.
Although the mechanism of the antiproliferative activity
of quercetin remains to be fully clarified, there is evidence
suggesting that the action of this substance is probably mediated by its interaction with the so-called type II EBS.3*6.34-36
Indeed, our data indicate that the quercetin ID 50% in sensitive leukemic cells is compatible with the dissociation constant (kd) of type I1 EBS in leukemic blasts! This possibility
is also supported by the observation that the antiestrogen
tamoxifene, which binds to type I1 EBS,37 induces TGF0 1 production in estrogen-receptor-positive MCF-7 human
breast cancer cells3’ as well as in estrogen-receptor-negative
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
3660
LAROCCA ET AL
human fetal fib rob last^.'^ For these reasons, it seems interesting to investigatewhether tamoxifene could inhibit leukemic
cell growth as well.
REFERENCES
1. Kilhnau J: The flavonoids, a class of semi-essential food com-
ponents: Their role in human nutrition. World Rev Nutr Diet 24: 117,
1976
2. Gabor M: Szent-Gyorgyi and the bioflavonoids: New results
and perspectives of pharmacological research into benzo-pyrone derivatives, in Cody V, Middleton E Jr, Harborne JB, Beretz A (eds):
Plant Flavonoid Biology and Medicine 11: Biochemical, Cellular,
and Medical Properties. New York, NY, Liss, 1988, p 1
3. Scambia G , Ranelletti FO, Benedetti Panici P, Piantelli M,
Rumi C, Battaglia F, Larocca LM, Capelli A, Mancuso S: Type
I1 estrogen binding sites in a lymphoblastoid cell line and growth
inhibitory effect of estrogen, antiestrogen and bioflavonoids. Int J
Cancer 46: 11 12, 1990
4. Yoshida M, Yamamoto M, Nikaido T: Quercetin arrests human
leukemic T-cells in late G, phase of the cell cycle. Cancer Res
52:6676, 1992
5. Post JFM, V m a RS: Growth inhibitory effects of bioflavonoids and related compounds on human leukemic CEM-Cl and
CEM-C7 cells. Cancer Lett 67:207, 1992
6. Larocca LM, Piantelli M, Leone G, Sica S , Teofili L, Benedetti
Panici P, Scambia G, Mancuso S , Capelli A, Ranelletti FO: Type I1
oestrogen binding sites in acute lymphoid and myeloid leukaemias:
Growth inhibitory effect of oestrogen and flavonoids. Br J Haematol
75:489, 1990
7. Larocca LM, Teofili L, Leone G, Sica S , Pierelli L, Menichella
G , Scambia G, Benedetti Panici P, Ricci R, Piantelli M, Ranelletti
FO: Antiproliferative activity of quercetin on normal bone marrow
and leukemic progenitors. Br J Haematol 79:562, 1991
8. Teofili L, Pierelli L, Iovino MS, Leone G, Scambia G, De
Vincenzo R, Benedetti Panici P, Menichella G, Macri E, Piantelli
M, Ranelletti FO, Larocca LM: The combination of quercetin and
cytosine arabinoside synergistically inhibits leukemic cell growth.
Leuk Res 16:497, 1992
9. Keller JR, Jacobsen SEW, Dubois CM, Hestdal K, Ruscetti
FW: Trasforming growth factor$: A bidiretional regulator of hematopoietic cell growth. Int J Cell Cloning 102, 1992
10. Bonewald L F Can transforming growth factor beta be useful
as a protective agent for pluripotent hematopoietic progenitor cells?
Exp Hematol 20:1249, 1992
11. Jacobsen SEW, Ruscetti FW, Roberts AB, Keller JR: TGFp is a bidirectional modulator of cytokine receptor expression on
murine bone marrow cells. Differential effects of TGF-PI and TGFp3. J Immunol 151:4534, 1993
12. Aglietta M, Stacchini A, Severino A, Sanavio F, Ferrando
ML, Piacibello W: Interaction of transforming growth factor-beta 1
with hematopoietic growth factors in the regulation of human normal
and leukemic myelopoiesis. Exp Hematol 17:296, 1989
13. Jacobsen SEW, Keller J R , Ruscetti FW, Kondaiah P, Roberts
AB, Falk LA: Bidirectional effects of TGF-P on colony-stimulating
factor induced human myelopoiesis in vitro: Differential effects of
distinct TGF-P isoforms. Blood 78:2239, 1991
14. Hatzfeld J, Li ML, Brown EL, Sookdeo H, Levesque JP,
O’Toole T, Gurney C, Clark SC, Hatzfeld A: Release of early hematopoietic progenitors from quiescence by antisense transforming
growth factor p1 or Rb oligonucleotides. J Exp Med 174:925, 1991
15. Busuker I, Neddermann KM, Petty BA, Schacter B, Spitalny
GL, Tepper MA, Pasternak RD: In vivo regulation of hemopoiesis
by transforming growth factor p 1. stimulation of GM-CSF- and
M-CSF-dependent murine bone marrow precursors. Exp Hematol
20:431, 1992
16. Hino M, Tojo A, Miyazono K, Urabe A, Takaku F Effects
of type P transforming growth factors on hematopoietic progenitor
cells. Br J Haematol 70:143, 1988
17. Ottmann OG, Pelus LM: Differential proliferative effects of
transforming growth factor-p on human hematopoietic progenitor
cells. J Immunol 140:2661, 1988
18. Piacibello W, Ferrero D, Sanavio F, Badoni R, Stacchini A,
Severino A, Aglietta M: Responsiveness of highly enriched CFUGM subpopulations from bone marrows, peripheral blood, and cord
blood to hematopoietic growth inhibitors. Exp Hematol 19:1084,
1991
19. Tessier N, Hoang T: Transforming growth factor P inhibits
the proliferation of the blast cells of acute myeloblastic leukemia.
Blood 72:159, 1988
20. Nara N, Tohda S, Nagata K, Suzuki T, Yamashita Y: Inhibition of the invitro growth of blast progenitors from acute myeloblastic leukemia patients by transforming growth factor p (TGF-P).
Leukemia 3572, 1989
21. Kerangueven F, Sempere C, Tabilio A, Mannoni P: Effects of
transforming growth factor P, tumor necrosis factor alpha, interferon
gamma and LIF-HILDA on the proliferation of acute myeloid leukemia cells. Eur Cytokine Net 1:99, 1990
22. Taetle R, Payne C, Dos Santos B, Russell M, Segarini P
Effect of transforming growth factor PI on growth and apoptosis of
human acute myelogenous leukemia cells. Cancer Res 53:3386, 1993
23. de Vos S, Brach MA, Asano Y , Ludwig W-D, Bettelheim P,
Gruss H-J, Henmann F: Transforming growth factor-81 interferes
with the proliferation-inducing activity of stem cell factor in myelogenous leukemia blasts through functional down-regulation of the ckit proto-oncogene product. Cancer Res 53:3638, 1993
24. Scambia G, Benedetti P, Ranelletti FO, Ferrandina G,De
Vincenzo R, Piantelli M, Masciullo V, Bonanno G, kola G, Mancuso
S: Quercetin enhances transforming growth factor P1 secretion by
human ovarian cancer cells. Int J Cancer 57:211, 1994
25. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton
DAG, Gralnick HR, Sultan C: Proposal for the classification of acute
leukaemias. Br J Haematol 33:451, 1976
26. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton
DAG, Gralnick HR, Sultan C: Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French-Amencan-British Cooperative Group. Ann Intern Med 103:626, 1985
27. Aye MT, Niho Y, Till JE, McCulloch EA: Studies of leukemic
cell populations in culture. Blood 44:205, 1974
28. Gewirtz A M , Calabretta B: A c-myb antisense oligodeoxynucleotide inhibits normal human hematopoiesis in vitro. Science
242:1303, 1988
29. Broxmeyer HE, Lu L, Hangoc G, Cooper S , Hendrie PC,
Ledbetter JA, Xiao M, Williams DE, Shen F-W: CD45 cell surface
antigens are linked to stimulation of early human myeloid progenitor
cells by interleukin 3 (IL-3), granulocytdmacrophage colony-stimulating factor (GM-CSF), a GM-CSFIIL-3 fusion protein, and mast
cell growth factor (a c-kit ligand). Exp J Med 174:447, 1991
30. Larocca LM, Teofili L, Sica S , Piantelli M, Ranelletti F o ,
Iovino MS, Leone G: Bioflavonoids: A possible toolin the therapy
of acute leukemias, in Kaspers GJL, Pieters R, Twentyman PR,
Weisenthal LM, Veerman AJP (eds): Drug resistance in leukemia
and lymphoma. Chur, Switzerland, Harwood Academic, 1993, p
27 1
3 1. Keller JR, Mantel C, Sing G: Transforming growth factor P 1
selectively regulates early hematopoietic progenitors and inhibits the
growth of IL-3 dependent myeloid leukemia cell lines. J Exp Med
168:737,1988
32. Sing G, Keller JR, Ellingsworth L, Ruscetti F W : Trans-
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
QUERCETIN INDUCES TGF-81 IN LEUKEMIC BLASTS
forming growth factor p selectively inhibits normal and leukemic
human bone marrow cell growth in vitro. Blood 72:1504, 1988
33. Falk LA, De Benedetti F, Lohrey N, Birchenall-Roberts MC,
Ellingsworth LW, Faltynek CR, Ruscetti Fw: Induction of Transforming growth factor-p1 (TGF-@l)receptor expression and TGFp1 protein production in retinoic acid-treated HL-60 cells: Possible
TGF-p1 mediated autonine inhibition. Blood 77:1248, 1991
34. Markaverich BM, Robens RR, Alejandm MA, Johnnan GA,
Middleditch BS, ClarkN:Bioflavonoid interaction with rat uterine
type I1 binding sites and growth inhibition. J Steroid Biochem
3071,
1988
35. Scambia G, Ranelletti FO, Benedetti Panici
P, Piantelli M,
Bonanno G, De Vincenzo R, Ferrandina G, Rumi C, Larocca LM,
Mancuso S : Inhibitory effect of quercetin on OVCA 433 cells and
presence oftype 11oestrogen binding sites
in primary ovarian tumors
and cultured cells. Br J Cancer 62:942, 1990
36. Ranelletti FO, RicciR,LaroccaLM,Maggiano
N, Capelli
A, Scambia G, Benedetti PaniciP, Mancuso S,Piantelli M: Growth-
3661
inhibitory effectof quercetin and presenceof type Il estrogen-binding sites in human colon-cancer cell lines and primary colorectal
tumors. Int J Cancer 50486, 1992
37. Scambia G, Ranelletti FO, Benedetti Panici
P, Piantelli M,
BonannoG,DeVincenzoR,FerrandinaG,Pierelli
L, Capelli A,
Mancuso S: Quercetin inhibits the growth of a multidrug-resistant
estrogen-receptor-negativeMCF-7 human breast-cancer cell line expressing type II estrogen-binding sites. Cancer Chemother Phannacol 28:255, 1991
38. Knabbe C, Lippman ME, Wakefield LN, FlandersKC, Kasid
A,DerynckR,Dickson
RB: Evidencethattransforminggrowth
factor p is a hormonally regulated negative growth factor in human
breast cancer cells. Cell 48:417, 1987
39. Colletta AA, WakefieldLM,Howell FV, van Roozndaal
KEP, Danielpour D, Ebbs SR, SpornM B , Baum M: Anti-oestrogen
induce the secretion of active transforming growth factor beta from
human fetal fibroblasts. Br J Cancer 62:405, 1990
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1995 85: 3654-3661
Quercetin inhibits the growth of leukemic progenitors and induces
the expression of transforming growth factor-beta 1 in these cells
LM Larocca, L Teofili, S Sica, M Piantelli, N Maggiano, G Leone and FO Ranelletti
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/12/3654.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.