proliferation of human CD34+ hematopoietic progenitor cells ...

From bloodjournal.hematologylibrary.org by guest on July 21, 2011. For personal use only.

1990 75: 2292-2298

Tumor necrosis factor-alpha strongly potentiates interleukin-3 and granulocyte-macrophage colony-stimulating factor-induced proliferation of human CD34+ hematopoietic progenitor cells C Caux, S Saeland, C Favre, V Duvert, P Mannoni and J Banchereau

Information about reproducing this article in parts or in its entirety may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://bloodjournal.hematologylibrary.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.

From bloodjournal.hematologylibrary.org by guest on July 21, 2011. For personal use only.

Tumor Necrosis Factor-alpha Strongly Potentiates Interleukin-3 and Granulocyte-Macrophage Colony-Stimulating Factor-Induced Proliferation of Human CD34+ Hematopoietic Progenitor Cells By Christophe Caux, Sem Saeland, Catherine Favre, Valerie Duvert, Patrice Mannoni, and Jacques Banchereau Previous studies have shown that tumor necrosis factors (TNFs) inhibit the proliferative effects of crude or purified colony-stimulating factors (CSFs) on low density human bone marrow cell fractions. In the present study we investigated the effects of TNFa on the growth of highly purified CD34' human hematopoietic progenitor cells (HPC) in response to recombinant CSFs. In short-term liquid cultures (5 t o 8 days), TNFa strongly potentiates interleukin3 (IL-3) and granulocyte-macrophage-CSF (GM-CSF)induced growth of CD34' HPC, while it has no proliferative effect per se. Within 8 days, the number of viable cells obtained in TNFa-supplemented cultures is threefold higher than in cultures carried out with IL-3 or GM-CSF alone. Secondary liquid cultures showed that the potentiating effect of TNFa on IL-3-induced proliferation of CD34' HPC does not result from an IL-3-dependent generation of

TNFa responsive cells. Limiting dilution analysis indicates that TNFa increases both the frequency of IL-3 responding cells and the average size of the IL-3-dependent clones. The potentiating effect of TNFa on IL-3- and GM-CSFdependent growth of CD34' HPC is also observed in day 7 colony assays. Under these short-term culture conditions, TNFa does not appear to accelerate cell maturation as a precursor morphology is retained. Finally, TNFa inhibits the relatively weak growth-promoting effect of granulocyteCSF (G-CSF), which acts on a more committed subpopulation of CD34' HPC different from that recruited by IL-3 and GM-CSF. TNFB displays the same modulatory effects as TNFa. Thus, TNFs appear t o enhance the early stages of myelopoiesis. 0 1990 by The American Society of Hematology.

T

clear cells, contains most of the hematopoietic progenitors, including multipotential self-renewing progenitors, as CD34' HPC can fully reconstitute lethally irradiated baboons.25In this context, we have investigated the effects of other growth regulators on CD34' progenitors. In the present study, using short-term proliferative assays of CD34' HPC, we found unexpectedly that T N F a , which has no growth-promoting effect per se, strongly enhances both IL-3- and GM-CSFdependent proliferation of CD34' HPC. We further demonstrated that the potentiating effect of T N F a on IL-3-induced proliferation of CD34+ H P C is direct and the consequence of both increased frequency and growth-rate of IL-3 responsive cells.

H E IN VITRO proliferation of human hematopoietic progenitor cells (HPC) depends upon the presence of exogenous growth factors such as interleukin-3 (IL-3), granulocyte-macrophage colony-stimulating factor (GMGSF). G-CSF, and M-CSF.' In contrast, soluble factors such as interferons (I F N s ) , ~ transforming .~ growth factor p (TGF-@),5-7 and tumor necrosis factors ( T N F S ) ~ .have ' ~ been shown to inhibit in vitro hematopoiesis, although their mode of action is not well characterized. T N F a is a protein produced by various cell types'' and that was originally shown to induce necrosis of tumors in vivo.I8 A cDNA clone coding for human T N F a has been isolated,' and the availability of recombinant preparations has revealed that this cytokine displays a broad spectrum of activities (reviewed in reference 17). Among those activities, T N F a , alone or in synergy with IFN-7, has been reported to inhibit in vitro hematopoietic colony formation by granulocytic and monocytic,' 1-13.'6e r y t h r ~ i d , ' ~ - "and . ' ~ multipotential progenitorsl0,I6when tested on total mononuclear bone marrow cells or fractions thereof depleted of mature cells.1°-'6 Recently, we and other^'^-*^ have shown that IL-3, GMCSF, and G-CSF can induce the proliferation of purified HPC expressing the CD34 antigen. This population, which represents 1% to 2% of bone marrow low density mononuFrom the Laboratory for Immunological Research. ScheringPlough (UNICET) Dardilly; and INSERM U119 and Centre Regional Anticancereux, Marseille, France. Submitted January 8, 1990; accepted February 16, 1990. C.C. is the recipient of a grant from the Fondation Mkrieux. Lyon. France. Address reprint requests to Christophe Caux. MD, ScheringPlough (UNICET), Laboratory for Immunological Research, 27 chemin des Peupliers, BP 11,69572 Dardilly, France. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C.section 1734 solely to indicate this fact. 0 1990 by The American Society of Hematology. 0006-4971/90/7512-0001%3.00/0 2292

MATERIALS AND METHODS

Hematopoietic factors. Recombinant human IL-326 (provided by Dr S. Tindall, Scbering-Research, Bloomfield, N J ) and G-CSF (provided by Dr N. Arai, DNAX, Palo Alto, CA) were purified to homogeneity. Specific activities of IL-3 and G-CSF were 5 x 10' U/mg and 2 x lo6 U/mg, respectively; one unit corresponding to half maximal proliferative activity as determined by (methyl--'H) thymidine (3H-TdR; specific activity 25 Ci/mmol; CEA Saclay, France) uptake by human HPC in liquid cultures. IL-3 and G-CSF were used at a saturating concentration of 10 ng/mL (50 U/mL) and 25 ng/mL (50 U/mL), respectively. Human GM-CSFZ7(provided by Drs P. Trotta and T.L. Nagabhushan, Schering-Research) was expressed in Escherichia coli and subsequently purified to homogeneity. Specific activity was 2 x IO6 U/mg, as previously defined.*' GM-CSF was used at a saturating concentration of 100 ng/mL (200 U/mL). Supernatant of Cos 7 cells transfected with the cDNA of human TNFP was used at a concentration of 0.5% (vol/vol), and mock supernatant derived from Cos 7 cells transfected with unrelated cDNA was used as negative control (provided by Dr N. Arai, DNAX, Palo Alto, CA). Recombinant purified T N F a was purchased from Genzyme (Boston, MA) (specific activity 2 x lo7 U/mg). In most experiments, it was used at a saturating concentration of 25 ng/mL (500 U/mL). Collection and fractionation of cord blood and bone marrow cells. Umbilical cord blood and bone marrow samples were collected according to institutional guidelines. Cord blood was collected during normal full-term deliveries in glass vessels containing preservBlood, Vol 75, No 12 (June 15), 1990: pp 2292-2298

From bloodjournal.hematologylibrary.org by guest on July 21, 2011. For personal use only. 2293

TNFa POTENTIATES PROLIFERATION OF CD34’ CELLS

results were expressed as mean counts per minute (cpm) standard deviation (SD). Progenitor assays in semisolid medium. Semisolid cultures in methylcellulose were carried out by plating 5 x lo’ cells per tissue culture grade 35 mm Petri dish (Corning, NY), in the presence of 1 IU purified recombinant human erythropoietin (Epo; Amgen, Thousand Oaks, CA; specific activity, 7 x lo4 IU/mg), as previously described in detail.” Duplicate dishes were plated in each experiment, and after 7 days of incubation at 37OC and 5% CO,, colonies (250 cells) were counted using phase-contrast microscopy. Limiting dilution studies. Limiting dilution studies were done in 96-well round-bottom microtest tissue culture plates (Nunc) in Iscove’s modified Dulbecco’s medium (IMDM) containing 30% FBS, mol/L 2-mercaptoethanol, 0.5% wt/vol BSA neutralized with NaHCO, and L-glutamine and antibiotics as indicated above. For each cell concentration used, ranging from 0.5 to 40.5 cells per well, and for each factor tested, aliquots of 50 pL were seeded in 120 to 360 wells. After 5 to 7 days of culture (37OC, 5% CO,), the positive wells were scored using an inverted microscope, and for selected cell concentrations, the number of cells in each clone was determined. For calculation of clone frequencies, the data were plotted as the log of the fraction of nonresponding wells versus the number of cells plated per well. Linear regression analysis was applied to the data to calculate the best slope.28Confidence intervals around the slope were calculated using a form of the Student’s t test. Conformance of the data to linearity was tested by using a x2 analysis: a x2 derivedprobability P > .05 indicating single-hit curves, thus demonstrating activity on a single limiting cell type. Average clone size was calculated as the ratio of (total cell number enumerated in 120 wells seeded at 1.5 cells/well)/(number of responding cells present in the 120 wells observed).28Calculation of average clone size was done on triplicates of 120 wells.

ative-free grade-1 sodium heparin (Sigma, St Louis, MO) at a final concentration of 20 IU/mL. Marrow was obtained by iliac crest aspiration from adult bone marrow transplant donors free of hematological disease. Light-density mononuclear cells were isolated by Ficoll-Hypaque gradient separation at a density of 1.077 g/mL. Mononuclear fractions were depleted of adherent cells by overnight incubation at 37OC in RPMI 1640 medium (GIBCO, Grand Island, NY) supplemented with 1% wt/vol tissue culture grade bovine serum albumin (BSA; Boehringer Mannheim, FRG). Cells bearing CD34 antigen were isolated from nonadherent mononuclear fractions through positive selection by indirect immune “panning” using anti-My10 monoclonal antibody (MoAb) (HPCA1; Becton-Dickinson,Mountain View, CA), as reported When the purity of the isolated cells, determined by FACScan flow cytometric analysis, was below 95%, a second purification step was performed using immunomagnetic beads. Briefly, after overnight incubation at 37OC, the CD34+ “panned” cells were resuspended at lo7 cells/mL in a cocktail of MoAbs (Leu-M3 [CD14], Leu5b [CD2], Leu3 [CD4], Leulla [CD16], and Leu16 [CD20] from Becton-Dickinson, at 2 pg/mL each) for 15 minutes, washed twice, and incubated for 30 minutes with immunomagnetic beads ( lo7 beads/mL) coated with anti-mouse immunoglobulins (Dynabeads M450; Dynal, Oslo, Norway). The beads were removed using a magnet, and the CD34’ cells were recovered in suspension. Thus, in all experiments, the isolated cells were 95% to 99% CD34’, as judged by staining with the anti-My10 MoAb. Cell cultures in liquid medium. Cell cultures were established in the presence of IL-3 (10 ng/mL) or IL-3 plus TNFa (25 ng/mL) in medium consisting of RPMI 1640 supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (FBS; Flow Laboratories, Irvine, UK), 10 mmol/L HEPES, 2 mmol/L L-glutamine, 5 x 10 -’mol/L 2-mercaptoethanol, penicillin (100 U/mL), and streptomycin (100 pg/mL); hereafter designated complete medium. CD34‘ cells were seeded in 24-well culture plates (Linbro; Flow Laboratories, McLean, VA) for expansion. Optimal conditions were maintained by splitting these cultures every 4 to 5 days. At different timepoints as indicated, cells were collected for counting and establishment of secondary cultures. For initiation of secondary cultures, cells were washed 4 times and further incubated for 3 hours in complete medium in order to eliminate growth factors. Subsequent proliferation assays were then carried out as described below. For proliferative assays, cells were distributed in 96-well roundbottom microtest tissue culture plates (Nunc, Roskilde, Denmark) at 5 x 10’ cells/100 pL complete medium and incubated (37°C. 5% CO,) with saturating concentrations of factors for 6 days. After incubation, cells were pulsed with 1 pCi of ’H-TdR for 6 hours, harvested, and counted. Tests were carried out in triplicate, and

-

m Fig 1. TNFa potentiates IL3-induced proliferation of highly purified CD34’ HPC. (A) Kinetics of ’H-TdR uptake by CD34+ HPC cultured in IL-3 (10 ng/mL) (-0-I# TNFa (25 ng/mL; -+-I, IL-3 TNFa (-W-), or medium alone (-0-1. (B) Viable cell count kinetics of CD34’ HPC cultured with IL-3 (-Ci-) or IL-3 TNFa (-&I. Cell numbers represent cumulative values from bulk cultures maintained in optimal conditions by splitting on day 4 in the presence of factors.

RESULTS

TNFa potentiates IL-3-induced proliferation of CD34+ HPC. The effects of TNFa on CSF-induced proliferation of C D 3 4 + HPC were examined in a liquid culture system. Results shown in Fig 1A indicate that TNFa (25 ng/mL) strongly enhances the IL-3-dependent proliferation of C D 3 4 HPC, whereas it has no proliferative effect per se. T h e enhancement of ’H-TdR incorporation was observed at all times tested. Figure 1B shows t h a t the enhanced ’H-TdR incorporation (Fig 1A) reflects a n increased cell proliferation, since more viable cells a r e recovered in IL-3 plus TNFa +

A

1001

+

25

+

1

0

2

’

1

4

.

1

6

days of culture

.

1

8

0

2

4

6

days of culture

8


CAUX ET AL

cultures when compared with IL-3 cultures. After 4 days and 8 days, approximately two- and threefold as many viable cells, respectively, were recovered when T N F a was added to cultures. The TNFa-induced enhancement of IL-3-mediated cell proliferation is dose dependent (Fig 2), with maximal stimulation observed at approximately 2.5 ng/mL TNFa. Interestingly, a combination of 2.5 to 25 ng/mL T N F a with a half maximal dose of IL-3 (0.25 ng/mL) results in a proliferative response higher than that obtained with an optimal dose of IL-3 used alone (2.5 ng/mL). TNFa and @ enhance the proliferation of CD34+ HPC induced by IL-3 and GM-CSF, but inhibit that induced by C-CSF. As GM-CSF and, to a lesser extent, G-CSF are also able to induce the proliferation of CD34+ HPC,2"-22.24 we wondered whether T N F a would also enhance cell proliferation induced by these factors. Thus, CD34+ HPC were cultured with optimal concentrations of various factors used alone or in dual combinations. Data shown in Table 1 indicate that T N F a is able to enhance the GM-CSFdependent proliferation of CD34' HPC, and, as observed with IL-3, T N F a approximately doubles the level of 3H-TdR incorporation. The recovery of viable cells is also enhanced (data not shown). In contrast, T N F a inhibits the G-CSFdependent proliferation of CD34+ HPC. Finally, TNFB displays the same activities as T N F a on CSF-induced proliferation of CD34+ HPC. IL-3 does not induce the generation of cells proliferating in response to TNFa. Since we previously showed that preculturing CD34+ HPC with IL-3 results in a subsequent enhancement of the proliferative responses to GM-CSF and G-CSF,"." we wondered whether IL-3 would induce the appearance of a cell population proliferating in response to T N F a alone. Thus, CD34+ HPC were precultured for various periods of time with IL-3, and proliferative responsiveness to T N F a and CSFs was subsequently assayed as measured by 'H-TdR incorporation in secondary liquid

h

r? 0 7

X

0

Y

m

c

P =I

10-2

10-1

100

101

102

TNF ngiml Fig 2 . Dose dependence of the potentiating effect of TNFa on IL-3-induced CD34+ HPC proliferation. CD34' HPC ( 5 x lo3)were cultured in TNFa (-+-I, TNFa 25 ng/mL IL-3 I-.-). TNFa 2.5 ng/mL IL-3 (-0-1, TNFa 0.25 ng/mL IL-3 (-A-)*or TNFa 0.025 ng/mL IL-3 (-A-1. 3H-TdR uptake was measured on day 6 Of culture.

+

+

+

+

Table 1. Effects of TNFs in Combination With CSFs on the Proliferation of CD34' HPC in Liquid Culture ~

Factors

Medium

TNFa

Medium IL-3 GM-CSF G-CSF

3.5 t 1.0 53.6 t 6.8 20.2 t 2.5 10.5 1.9

5.5 2.8 95.2 i 11.4 41.9 t 4.2 5.1 t 2.1

*

*

TNFO

4.5 i 0.6 91.7 k 7.2 40.9 6 . 8 0.4 4.9

* *

~

IL-3

53.6 53.6 66.1 63.0

t 6.8 k

6.8

t 3.8 t 3.8

Values show 3H-TdR uptake (CPM x k SD) obtained from one representative experiment of three and expressed as mean t SD of triplicate wells. Proliferation was measured at day 6 ; TNFs and all CSFs were used at saturating concentrations, as described in Materials and Methods.

cultures. Results shown in Fig 3 indicate that T N F a alone has no direct proliferative effect on IL-3 precultured cells, while the cells remain responsive to IL-3, GM-CSF, and G-CSF. The same results were obtained when preculturing CD34+ HPC with IL-3 plus T N F a (data not shown). TNFa increases the frequency and size of IL-3 responsive clones. We next addressed whether the potentiating effect of T N F a on IL-3-dependent cell growth was due to an increased frequency of IL-3 responding cells, to an enhanced growth rate of cells already responding to IL-3, or to both. To answer this question, clonogenic day 7 CFU assays in methylcellulose and limiting dilution studies in liquid medium were carried out with CD34+ HPC. First, day 7 CFU, from cord blood or bone marrow CD34+ HPC (Table 2) show that T N F a enhances the total number of colonies induced by IL-3 and GM-CSF, but blocks the G-CSF-dependent generation of colonies. In addition, colonies generated in response to combinations of IL-3 and T N F a or GM-CSF and T N F a were larger than those generated in the presence of IL-3 or GM-CSF alone (not shown). Second, limiting dilution studies were done where CD34+ HPC were seeded at various concentrations into wells of microtitration plates with IL-3 and with or without TNFa. The number of growing clones was determined after 6 days of culture. As shown in Fig 4, T N F a increases the frequency of IL-3 responding cells (1 of 3.2 cells generates clones in the presence of IL-3 and T N F a , whereas 1 of 4.5 cells generates clones in the presence of IL-3 alone). The combination of 1L-3 and T N F a generates 20% to 40% more clones than IL-3 alone (three separate experiments). The fact that the potentiating effect of T N F a is seen at very low cell concentration (0.5 cells per well) and the linear slope of the frequency analysis curves demonstrates that T N F a acts directly on the CD34 ' HPC. We further determined the number of cells in positive wells (seeded with 1.5 cells) to estimate clone size (Fig 5). Cultures done in the presence of T N F a and 1L-3 yielded a higher number of large clones (greater than 32 cells) and a lower number of small clones (less than 32 cells) than those done with IL-3 alone. This is illustrated by a shift of clone size distribution towards the larger size clones (Fig 5) and by a higher average clone size in the combination of IL-3 plus T N F a (59 cells per clone, experiment 1 and 53 cells per clone, experiment 2) than in IL-3 alone (34 cells per clone, experiment 1 and 32 cells per clone, experiment 2).


TNFa POTENTIATES PROLIFERATION OF CD34' CELLS

T

40

30

Fig 3. The potentiating affect of TNFa on IL-3-induced proliferation of CD34+ HPC is not a consequence of an IL-3-dependent induction of TNFa responding cells. Purified CD34' HPC were precultured with IL-3 (10 ng/mL); and at various time points, cells were harvested, washed, processed as described in Materials and Methods, and subsequently assayed for secondary proliferative GM-CSF ( 0 ) .G-CSF responsiveness t o IL-3 (E@, TNFa (a),or medium alone (0).Responsiveness was measured by 3H-TdR uptake, 6 days after initiation of secondary cultures.

20

10

(m),

0 0

12

5

days of pre-culture in IL-3

DISCUSSION

This study demonstrates that tumor necrosis factors strongly potentiate the proliferative effects of IL-3 and GM-CSF on CD34' HPC isolated from cord blood and bone marrow. Within 8 days, the number of cells generated in TNFa-supplemented liquid cultures of CD34+ HPC is three times higher than in cultures carried out with IL-3 alone. This reflects a true potentiating effect as T N F a per se does not act as a growth factor for CD34+ HPC. In our hands, T N F a appears to be the most potent activator of IL-3dependent proliferation of CD34+ HPC, since combinations of IL-3 with GM-CSF, G-CSF, IL-1 (data not shown), and IL-6 (data not shown) are always less potent than the combination of IL-3 and TNFcu. As expected, TNFP, which uses the same receptor as TNFa," also potentiates IL-3induced proliferation of CD34' HPC. As shown by secondary liquid cultures, the observed potentiation is not the consequence of an IL-34ependent

generation of cells proliferating in response to T N F a alone. Rather, colony assays and limiting dilution studies indicate that the effect of T N F a results from both an increased frequency and an increased growth rate of IL-3 responding cells. Furthermore, limiting dilution analysis demonstrates a direct effect of T N F a on its target cells. Although T N F a strongly potentiates the proliferative effects of IL-3, no morphologic differences were observed in terms of cell maturation between day 7 colonies generated with IL-3 and those generated with IL-3 and T N F a (data not shown). In number of cells / well 0

5

10

15

20

VI

E 0

e

.-

E

F

.-

0

Table 2. Effects of TNFa in Combination With CSFs on the Induction of Day 7 CFU From Cord Blood and Bone Marrow CD34+ HPC

0

n

E c 0

Factors Experiment 1 Medium TNFa Experiment 2 Medium TNFa

Medium

IL-3

GM-CSF

G-CSF

L

c 0

I-

0 0 0 0

24 1 245 335 343

169 141 259 251

20 24 0 0

4 8 7 2

240 222 310

154 134 258 276

21 15 2 1

356

Duplicate dishes plated with 5 x lo3 cells in the presence of Epo (1 U/mL) and CSFs with or without TNFa (25 ng/mL) or in medium alone were scored for number of colonies on day 7 (colonies 5 5 0 cells). For CSF concentrations, see Materials and Methods. Experiment 1, cord blood; Experiment 2, bone marrow.

Y

2 Fig 4. TNFa increases the frequency of IL-3 responsive CD34' HPC. Replicate wells (1201 were seeded at 13.5 cells per well, (2401 a t 4.5 cells per well, and (3601a t 1.5 and 0.5 cells per well. The presence of clones was assessed on day 6. Single-hit kinetics were obtained, and the slopes were drawn by linear regression analysis. The shaded region represents the 95% confidence limit. The clone frequencies obtained in this experiment were 1:4.5 = 0.22 ( P = .65) for IL-3 ( 4 - 1 and 1:3.2 = 0.31 ( P = .45) for IL-3 TNFa (-0-1. TNFa or medium alone did not induce the development of any clones. The experiment presented here is representative of three.

+


5

u)

CAUX ET AL

151

T

10

.c

0

z

P

E

5

3

c

0 8-15

A

16-31

32-63

64-127128-255

number of cells per clone

I

T

* 0

t

P

E

3

c

8-15

B

16-31

32-63

64-127 128-255

number of cells per clone

Fig 5. TNFa increases the size of IL-3-dependent clones. Clone size distribution obtained in limiting dilution experiments done with CD34' HPC seeded at 1.5 cells per well in the presence of IL-3 (.I or 11-3 TNFa (@I. After 6 days of culture, the size of each clone was determined by counting using an inverted microscope. TNFa or medium alone did not induce the development of any clones (28 cells per well). The clones were classified into sections of arbitrarily defined numbers. The numbers plotted correspond to the mean of triplicates of 120 wells observed. The average clone size calculated on triplicates of 120 wells were (A): Experiment 1 , 3 4 k 3 for IL-3 and 59 k 4 for IL-3 TNFa: and ( 8 ) Experiment 2.32 k 4 for IL-3 and 53 + 2 for IL-3 TNFa.

+

+ +

this context, cells generated in short-term liquid culture (8 days) in the presence of T N F a and IL-3 express a phenotype and a precursor morphology similar to that of cells generated with IL-3 alone (data not shown). T N F a was also shown to potentiate GM-CSF-induced proliferation of CD34+ HPC in liquid or semisolid cultures with parameters comparable with those observed with IL-3. In contrast, T N F a displays an antagonist effect on the proliferation of CD34+ HPC induced by G-CSF. In this context, we and other investigator^^^^^'.^^.^^ have previously shown that GM-CSF and IL-3 recruit largely overlapping CD34 + cell populations, including multipotential progenitors, while G-CSF acts on a subpopulation of CD34' committed progenitors. Thus, the potentiating and antagonistic effects of T N F a on CSF-induced CD34+ HPC proliferation may be related to the heterogeneity of C S F

responding target cells, which remain to be phenotypically defined. T N F a has generally been described as an inhibitor of in vitro hematopoiesis.8,'' However, such studies reported that the degree of inhibition mediated by T N F a was variable14.'' and depended on the cellular composition of the bone marrow fractions tested and on the type of conditioned medium used as the source of CSFs. These inhibitory effects of T N F a may be indirect through the release of factors by accessory cells," factors that may act directly or in concert with T N F a to inhibit hematopoie~is.~-'~*'' Thus, the use of heterogeneous bone marrow cell fractions or conditioned medium as a source of CSFs3' in previous studies could explain the apparent discrepancies with our current results, obtained with recombinant factors and CD34' HPC, a population that is highly enriched for progenitor cells and devoid of accessory cells. In keeping with the generally described inhibitory effects of T N F a , we have found (in other experiments to be submitted) that T N F a can block cell growth in long-term cultures (14 to 35 days) of CD34' HPC, when well-differentiated cells are present in the cultures. Our data on the potentiating effect of T N F a in short-term cultures are supported by the recent observations that T N F a can enhance GM-CSF-induced in vitro growth of cells from some acute myeloid leukemia (AML) As previously shown for IL-3 and GM-CSF,20,33 such observations further indicate similarities in cytokine responsiveness of AML cells and normal hematopoietic cells. Additional studies are required to determine the mechanism of the potentiation observed in our present investigation. As T N F a has been shown to upregulate the expression of the IL-2 receptor on activated T cells,34it is tempting to speculate that T N F a might upregulate the expression of IL-3 and GM-CSF receptors on CD34+ HPC. However, T N F a may also mediate its effect by activating a specific second messenger, thus amplifying mitotic signals given by IL-3 or GM-CSF alone. Conversely, the inhibitory effect of T N F a on G-CSF-dependent proliferation of CD34+ HPC may result from downregulation of the G-CSF receptor or from a blocking of intracellular pathways activated by G-CSF. TNFa, primarily identified as a cytotoxic factor was later demonstrated to display growth-stimulatory properties on different cell type^.^^-^' Such a capacity to display both proliferative and antiproliferative effects is indeed shared by many factor^.'^ In this study, we showed that TNFa, originally described as an inhibitor of hematopoiesis, strongly enhances the early stages of myelopoiesis. This finding should be taken in consideration when clinical use of cytokines is proposed in order to control the proliferation of leukemias or tumors. ACKNOWLEDGMENT

We are grateful to N. Courbiere, M. Vatan, and C. MCnCtrier for editorial assistance; 1. Durand for flow cytometric analysis; Dr J.P. Favier, Professor R.C. Rudigoz, and colleagues for providing umbilical cord blood samples; Professor P. HervC and colleagues for providing bone marrow samples; and Dr J.A. Waitz for support and for critical reading of the manuscript.


TNFa POTENTIATES PROLIFERATION OF CD34+ CELLS REFERENCES 1. Metcalf D: The molecular control of cell division, differentiation commitment and maturation in haemopoietic cells. Nature 339:27, 1989 2. Broxmeyer HE, Lu L, Platzer E, Feit C, Juliano L, Rubin BY: Comparative analysis of the influences of human 7,a and @ interferons on human multipotential (CFU-GEMM), erythroid (BFU-E), and granulocyte-macrophage (CFU-GM) progenitor cells. J Immunol131:1300,1983 3. Mamus SW, Beck-Schroeder S, Zanjani ED: Suppression of normal human erythropoiesis by gamma-interferon in vitro: Role of monocytes and T-lymphocytes. J Clin Invest 75:149, 1985 4. Murphy M, Loudon R, Kobayashi M, Trinchieri G: Interferon and lymphotoxin, released by activated T cells, synergize to inhibit granulocyte-monocyte colony formation. J Exp Med 164:263, 1986 5. Ohta M, Greenberger JS, Anklesaria P, Bassols A, MassaguC J: Two forms of transforming growth factor-@ distinguished by multipotential haematopoietic progenitor cells. Nature 329:539, I987 6. Ottman OG, Pelus LM: Differential proliferative effects of transforming growth factor @ on human hematopoietic progenitor cells. J Immunol 140:2661, 1988 7. Sing GK, Keller JR, Ellingsworth LR, Ruscetti FW: Transforming growth factor @ selectively inhibits normal and leukemic human bone marrow cell growth in vitro. Blood 72:1504, 1988 8. Gray PW, Aggarwal BB, Benton CV, Bringman TS, Henzel WJ, Jarrett JA, Leung DW, Moffat B, Ng P, Svedersky LP, Palladino MA, Nedwin GE: Cloning and expression of cDNA for human lymphotoxin, lymphokine with tumour necrosis activity. Nature 312:721, 1984 9. Pennica D, Nedwin GE, Hayflick JS, Seeburg PH, Derynck R, Palladino MA, Kohr WJ, Aggarwal BB, Goeddel DV: Human tumour necrosis factor: Precursor structure, expression and homology to lymphotoxin. Nature 312:724, 1984 10. Broxmeyer HE, Williams DE, Lu L, Cooper S, Anderson SL, Beyer GS, Hoffman R, Rubin BY: The suppressive influences of human tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: Synergism of tumor necrosis factor and interferon-gamma. J Immuno1 136:4487, 1986 11. Peetre C, Gullberg U, Nilsson E, Olsson I: Effects of recombinant tumor necrosis factor on proliferation and differentiation of leukemic and normal hematopoietic cells in vitro: Relationship to cell surface receptor. J Clin Invest 78:1694, 1986 12. Tobler A, Munker R, Heitjan D, Koeffler H P In vitro interaction of recombinant tumor necrosis factor a and all-transretinoic acid with normal and leukemic hematopoietic cells. Blood 70:1940, 1987 13. Murase T, Hotta T, Saito H, Ohno R: Effect of recombinant human tumor necrosis factor on the colony growth of human leukemia progenitor cells and normal hematopoietic progenitor cells. Blood 69:467, 1987 14. Munker R, Koeffler HP: In vitro action of tumor necrosis factor on myeloid leukemia cells. Blood 69:1102, 1987 15. Barber KE, Crosier PS, Gillis S, Watson JD: Human granulocyte-macrophage progenitors and their sensitivity to cytotoxins: Analysis by limiting dilution. Blood 70:1773, 1987 16. Murphy M, Perussia B, Trinchieri G: Effects of recombinant tumor necrosis factor, lymphotoxin, and immune interferon on proliferation and differentiation of enriched hematopoietic precursor cells. Exp Hematol 16:131, 1988 17. Beutler B, Cerami A: The biology of cachectin/TNF. A

primary mediator of the host response. Annu Rev Immunol 7:625, 1989 18. Carswell EA, Olf LJ, Kassel RL, Green S, Fiore N, Williamson B: An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA 72:3666,1975 19. Saeland S, Caux C, Favre C, Aubry JP, Mannoni P, PCbusque MJ, Gentilhomme 0, Otsuka T, Yokota T, Arai N, Arai KI, Banchereau J, de Vries JE: Effects of recombinant human interleukin-3 on CD34-enriched normal hematopoietic progenitors and on myeloblastic leukemia cells. Blood 72:1580, 1988 20. Saeland S, Caux C, Favre C, Duvert V, Ptbusque MJ, Mannoni P, de Vries JE: Combined and sequential effects of human IL-3 and GM-CSF on the proliferation of CD34' hematopoietic cells from cord blood. Blood 73:1195, 1989 21. Caux C, Favre C, Saeland S, Duvert V, Mannoni P, Durand D, Aubry JP, de Vries JE: Sequential loss of CD34 and class I1 MHC antigens on purified cord blood hematopoietic progenitors cultured with IL-3: Characterization of CD34-, HLA-DR+ cells. Blood 74:1287,1989 22. Leary A, Yang YC, Clark CC, Gasson JC, Golde DW, Ogawa M: Recombinant gibbon interleukin 3 supports formation of human multilineage colonies and blast cell colonies in culture: Comparison with recombinant human granulocyte-macrophage colony-stimulating factor. Blood 70: 1343, 1987 23. Bot FJ, Dorssers L, Wagemaker G, Lowenberg B: Stimulating spectrum of human recombinant multi-CSF (IL-3) on human marrow precursors: Importance of accessory cells. Blood 71 :1609, 1988 24. McNiece I, Andrews R, Stewart M, Clark S, Boone T, Quesenberry P: Action of interleukin-3, G-CSF, and GM-CSF on highly enriched human hematopoietic progenitor cells: Synergistic interaction of GM-CSF plus G-CSF. Blood 74:110, 1989 25. Berenson RJ, Andrews RG, Bensinger WI, Kalamasz D, Knitter G, Buckner D, Bernstein ID: CD34+ marrow cells engraft lethally irradiated baboons. J Clin Invest 81:951, 1988 26. Otsuka T, Miyajima A, Brown N, Otsu K, Abrams J, Saeland S, Caux C, de Waal Malefijt R, de Vries JE, Meyerson P, Yokota K, Gemmel L, Rennick D, Lee F, Arai N, Arai KI, Yokota T: Isolation and characterization of an expressible cDNA encoding human IL-3. J Immunol 140:2288, 1988 27. Lee F, Yokota T, Otsuka T, Gemmell L, Larson N, Luh J, Arai K, Rennick D: Isolation of cDNA for a human granulocytemacrophage colony-stimulating factor by functional expression in mammalian cells. Proc Natl Acad Sci USA 82:4360, 1985 28. Lefkovits I, Waldmann H: Limiting dilution analysis of cells in the immune system. Cambridge, UK, Cambridge University Press, 1985, p 8 1 29. Emerson SG, Yang YC, Clark SC, Long MW: Human recombinant granulocyte-macrophage colony stimulating factor and interleukin 3 have overlapping but distinct hematopoietic activities. J Clin Invest 82: 1282, 1988 30. Lu L, Walker D, Broxmeyer HE, Hoffman R, Hu W, Walker E: Characterization of adult human marrow hematopoietic progenitors highly enriched by two-color cell sorting with My10 and major histocompatibility class I1 monoclonal antibodies. J Immunol 139: 1823,1987 31. Hoang T, Levy B, Onetto N, Haman A, RodriguezCimadevilla JC: Tumor necrosis factor 01 stimulates growth of the clonogenic cells of acute myeloblastic leukemia in synergy with granulocyte-macrophage colony-stimulating factor. J Exp Med 170:15, 1989 32. Mannoni P: Unpublished data, January 1990


33. Pebusque MJ, Fay C, Lafage M, Semptrt C, Saeland S, Caux C, Mannoni P: Recombinant human IL-3 and G-CSF act synergistically in stimulating the growth of acute myeloid leukemia cells. Leukemia 3:200, 1989 34. Scheurich P, Thoma B, Ucer U, Pfizenmaier K: Immunoregulatory activity of recombinant human tumor necrosis factor (TNF)a: Induction of T N F receptors on human T cells and TNF-amediated enhancement of T cell responses. J Immunol 138:1786, 1987 35. Vilcek J, Palombella VJ, Henriksen-DeStefano D, Swenson C, Feinman R, Hirai M, Tsujimoto M: Fibroblast growth enhancing

CAUX ET AL

activity of tumor necrosis factor and its relationship to other polypeptide growth factors. J Exp Med 163:632, 1986 36. Kehrl JH, Miller A, Fauci AS: Effect of tumor necrosis factor a on mitogen-activated human B cells. J Exp Med 166:786, 1987

37. Digel W, Stefanic M, Schoniger W, Buck C, Raghavachar A, Frickhofen N, Heimpel H, Porzsolt F: Tumor necrosis factor induces proliferation of neoplastic B cells from chronic lymphocytic leukemia. Blood 73:1242, 1989 38. Sporn MB, Robert AB: Peptide growth factors are multifunctional. Nature 332:217, 1988