Steel Factor (c-kit Ligand) Promotes the Survival of

1 downloads 0 Views 2MB Size Report
the survival of stem cells in culture in the absence of cell division. These studies ..... after transplantation according to the procedures described in Materi- als and ...
Steel Factor (c-kit Ligand) Promotes the Survival of Hematopoietic Stem/Progenitor Cells in the Absence of Cell Division By Jonathan R. Keller, Mariaestela Ortiz, and Francis W. Ruscetti It is known that the majority of primitive hematopoietic progenitors are in a noncycling quiescent state. In addition, normal hematopoietic progenitors and progenitor cell lines show an absolute dependence on growth factors for their survival in vitro, yet theeffect of growth factors on progenitor cell survival has not been separated from effects on both proliferation and differentiation. Using an in vitro assay system, we examined whether growth factors could promote the survival of stem cells in culture in the absence of cell division. These studies show that steel factor E L F ) and, to a lesser extent, interleukin-3 (IL-3) directly promoted the survival of elutriated bone marrow progenitor cells lcountercurrent centrifugal elutriation [CCEl-27) that are enriched for primitive hematopoietic progenitors that respond to the combination ofSLF plus IL-3. Furthermore, SLF promoted

the survival of short-term reconstituting cells (STRC), and long-term reconstituting cells (LTRC) with trilineage reconstitution potential in vivo. In comparison, granulocyte colony-stimulating factor (G-CSF), IL-6, leukemia inhibitory factor, IL-11, IL-l, granulocyte macrophage CSF (GM-CSF), and macrophage CSF (M-CSF) had no effect on the survival of these cells. In thepresence of mitotic inhibitors (nocodarole or aphidicolin), SLF promoted thesurvival of CCE-27progenitor cells that respond to the combination of SLF plus IL-3 in vitro and STRCs and LTRCs that are detected in vivo. Taken together, these data show that SLF can directly promote the survival of hematopoietic progenitor cells in the absence of cell division. This is a US government work. There are no restrictions on its use.

I

Institutes of Health, 1985). Mice were maintained in a pathogenfree environment. Bone marrow (BM) separation. BM cells (BMCs) were aspirated from the femurs of BALB/c mice with complete Iscove’s modified Dulbecco’s medium (IMDM) medium containing 10% fetal calf serum (FCS) 3 mg/mL glutamine, and antibiotics. Cells were further purified by countercurrent centrifugal elutriation (CCE). Briefly, 0.5 to I X 10’ BMCs in 10 to 20 mL complete IMDM was elutriated using a centrifuge (model J-6M; Beckman Instruments, Irvine, CA) with an elutriator rotor (model JE-6B; Beckman) and a standard chamber at a flow rate of 15 mL/min at 3,000 rpms. The flow rate was adjusted to 25 mL/min and cells were recovered from 25 to 27 mLlmin (CCE-27). In some experiments, CCE-27 cells were further purified using magnetic beads by incubating CCE-27 cells with RB68C5 (0.5 pg/l X 10“ cells) and RB6-6B2 (1.0 pg11 X IO6 cells) antibodies (gifts of R. Coffman, DNAX C o p , Palo Alto, CA), which recognize myeloid and B cells, respectively, in complete IMDM for 30 minutes at 4°C. Cells were washed and then incubated with sheep antirat IgG-coated magnetic beads (Dynal, Great Neck, NY) at a ratio of 40: I (beads:cells) and then incubated for 30 minutes at 4°C. Labeled cells were removed by a magnetic particle concentrator (Dynal) and unlabeled cells (lineage depleted, Lin-) were recovered in the supernatant. Fluorescence-activated cell sorting (FACS) separation of CCE27 cells. CCE-27 cells were incubated with biotin labeled anti-c-

T HAS BEEN SHOWN that approximately 10% of the pluripotent stem cell population is actively cycling during constitutive hematopoiesis andthatroughly 90% are quiescent.‘ In addition, the balance between quiescence and proliferation is tightly regulated. In particular, ithasbeen shown that the cycling status of the dormant stem cell population is altered in mice that are administered a sublethal dose of 5-fluorouracil in vivo (toxic to cycling progenitors), such that the majority of stem cells are in cycle 3 to 5 days after 5-fluorouracil treatment and, that by day 8, stem cells return to homeostatic cycling levels.* Although a number of growth factors have been identifiedthat promote the proliferation of progenitor cells, little is known about the mechanisms, conditions, or factors that regulate the survivalhiability of the dormant/quiescent pluripotential stem cell population. However, it is known that both normal progenitor cells and growth factor-dependent progenitor cell lines show an absolute dependence on growth factors for their survival in vitro,3yet the effect of growth factors on progenitor cell survival has not been separated from effects on both proliferation and differentiation. For example, several studies have indicated that interleukin-3 (IL-3), granulocyte colony-stimulating factor (G-CSF), IL-6, IL-4, IL-I, and steel factor (SLF) can promote the survival of various progenitor but these studies could not determine whether progenitor survival was a result of cell division leading to progenitor cell expansion andor differentiation. Thus, we developed an in vitro assay system to examine the growth factor requirements for the survival of primitive progenitor cells in the absence of cell division, and we have compared a variety of known hematopoietic growth factors. In this report, we found that SLF and, to a lesser extent, IL-3, promotes the survival of hematopoietic stedprogenitor cells in vitro in the absence of cell division. MATERIALS AND METHODS Mice. BALB/c, C57BL/6 (Ly 5.2) and their congenic C57BU6 (Ly 5.1) mice 8 to 12 weeks of age were obtained from the animal production area at the Frederick Cancer Research and Development Center (Frederick, MD). Animal care was provided in accordance with the procedures outlined in the Guide for Care and Use of Laboratory Animals (publication no. 86-23, Bethesda, MD, National Blood, Vol 86. No 5 (September l), 1995: pp 1757-1764

Fromthe Biological Carcinogenesis and Developmental Program, Program Resources Inc/DynCorp, and Luboratory of Leukocyte Biology, Biological Response Mod$ers Program. National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, MD. Submitted February 6, 1995; accepted April 20, 1995. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government. Address reprint requests to Jonathan R. Keller, PhD, Biological Carcinogenesis and Development Program, PRUDynCorp. PO Box B, Frederick, MD 21702-1201. 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 15: U.S.C. section 1734 solely to indicate this fact. This is a US government work. There are no restrictions on its use. 0006-497//95/8605-00I9$0.00/0 1757

KELLER, ORTIZ, AND RUSCETTI

1758

Protocol: Progenitor Cell Survival In Soft Agar

+ Score For Colony FormationAner 7 Days

Fig 1. Summary of the assay for progenitor cellsurvival in soft agar.

kit or isotype matched control antibodies (Pharmingen. San Diego. CA) at S00 ngll X IO6 cellsll00 pL of medium for 30 minutes at 4°C. r h e cells were washed in complete IMDM twice. and the cell pellet was resuspended in the same medium at 1 X 10" cells per 1 0 0 pL containing 1 p g of streptavidin-phycoerythrin(SA-PE) and incubated for 30 minutes at 4°C. The cells were washed two times in complete IMDM and resuspended in the same medium to approximately 2 to S X IO'cellslmL for FACS (Becton Dickinson. Mountain View. CA). Grorcv/t,focrors. Cultures were supplemented withmedium plus or minus predetermined optimal doses of purified murine SLF (gift of S. Gillis, Immunex Corp. Seattle. WA), or murine IL-3 (mulL3). human leukemia inhibitory factor (huL1F). huIL-l I , murine IL6 (muIL-6) (Peprotech. Rocky Hill. NJ), murine granulocyte-macrophage colony-stimulating factor (muGM-CSF: giftof T. Boone, Amgen Corp. Thousand Oaks. CA), human G-CSF (huG-CSF. gift of L. Souza, Amgen), huIL-l (Hoffman-LaRoche, Nutley. NJ). or human macrophage-CSF (huM-CSF, gift of M. Geier. Cetus Corp). Soji crgar co/orly-forrnirtg c~srcry. CCE-27 cells wereplatedat 2.5 X Io" cellslmLl3S mm Luxpetri dishes (Miles Laboratories. Inc. Naperville. IL) in complete IMDM containing 0.35% seaplaque agarose (FMC Bioproducts, Rockland, ME). Dishes were incubated in a fully humidified atmosphere at 37°C. 5%- CO2 and then scored for colony formation after 7 to 10 days (colony-forming unit culture, CFU-C). For the survival assays. the soft agar plates were pulsed with 200 pL IMDM containing SLF plus IL-3 (final concentration: IL-3.30 nglmL; SLF, 100 nglmL) and then scored for colony formation 7 to 10 days after the cytokine pulse (summarized in Fig l ). Sin,p/e cell ussuy.r. Separated cells were seeded in Terasaki plates, 60 wells per plate (Nunc. Damstrup. Denmark) at a concentration of I cell per well in 20 pL of medium containing purified growth factors. For each determination. a total of 300 cells were examined. Wells were scored for proliferation ( > l 0 cellslwell) after 6 to IO days of incubation at 37°C and S% COz. -'H-tI~.vmicline.incorporation ctssoys. CCE-27 and Lin- CCE-27 cells were cultured in 96-well microtiter plates in 100 pL of complete IMDM and incubated at 37°C. 5% CO2 for the indicated times. The wells were pulsed with I pCi of "-thymidine (6.7 Ci/mmol. New England Nuclear, Boston. MA) overnight andthen harvested for liquid scintillation counting after 28, 48. and 72 hours. Background (medium alone) 'H-thymidine incorporation after 24 hours was consistently higher than after 48 hours or 72 hours because freshly aspirated BM contains cells already in cycle. BM trcrttsplantcttion erncl nsscrys ,for short-term crtld /ong-ternt Irentatopoieric reconstitrrtiort. CS7BL/6 (Ly S. I ) recipient mice were exposed 10 Gy total body lethal irradition ('"Cesium irradiation) 3

hours before the intravenous injection of CS7BU6 (Ly-5.2) donor cells that were resuspended in phosphate-buffered saline (PBS). Short-term reconstituting cells are defined as those cells that allow a lethally irradiated mouse to survive for 40 to 60 days. To detect long-term reconstituting cells, BM cells and thymocytes were procured from recipient mice 4 to 6 months after transplantation for FACS analysis. Hematopoietic reconstitution of donor and host (recipient) cells was determined by two color FACS analysis using monoclonal antibodies A-20-1.7 (recognizes host derived Ly 5.1 + cells) and104-2.1 (recognizes donor derived Ly 5.2' cells) or isotype-matched control antibodies that were developed with fluorescein isothiocyanate (F1TC)-conjugated affinity-purified goat antimouse IgG2a (Southern Biotechnology Associates, Inc.Birmingham, AL). Todetermine the percentage of lineage-specific cells of host or donor origin, B-lineage cells and granulocytes in the BM were also stained with biotin-conjugated RA3-6B2 (B220) or biotinconjugated RB6-8CS. respectively (Pharmingen) or isotype-matched control antibodies and developed with Neutrolite Avidin-RPE (Southern Biotech Assoc Inc). To determine the percentage of Tlineage cells in the thymus, cells were also stained with biotinconjugated anti-Thy-l (Pharmingen) or isotype-matched control antibodies and developed with Neutrolite avidin-RPE. The level of donor reconstitution for each lineage was averaged plus or minus standard error for each group consisting of at least three to five mice.

RESULTS

The effect qf hematopoietic Rrowtl1.factor.yon the survival of counter current elutriated BM cells. BMCs were separated by CCE at a flow rate of 27 mL/min (CCE-27) to

enrich for cells with low-density and blast-like characteristics.' These cells represent 5% to 7% of the total unfractionated BM cell input. This population was also selectedbecause they are enriched in comparison with unseparated BMCs for progenitors, which form large colonies in soft agar in response to SLF plus IL-3, but form fewer colonies in response to single growth factors such as IL-3, M-CSF, GM-CSF,G-CSF, and SLF (5 +- I , 5 2 I , 2 +- I , 2 2 I , and 0 colonies, respectively) (Table l ) . Furthermore, similar to normal BMCs, the CCE-27 BMC population contains short-term marrow reconstituting cells (STRC) or cells that protect lethally irradiated recipients 30 to 60 days after transplantation, as well as long-term reconstituting cells (LTRC) thatare capable of trilineage (granulocyte, B-cell, and T-

1759

SURVIVAL OFSTEMIPROGENITORCELLS

Table 1. Colony-Stimulating Factor-Induced Growth of CCESeparated Murine BM Cells CFU-C

Factor*

Unseparated B M t (7.5 x 10' cells/plate)

CCE-27t (2.5 x 10' cells/plate)

Medium IL-3 GM-CSF G-CSF M-CSF SLF IL-3 +57SLF

0 30 2 4 3 2 2 2 5 5 2 0.5 2 2

0 5+1 522 221 221 0 33 2 4

Growth

29 15 52

Cultures were supplemented with medium with or without muSLF (100 ng/mL), mulL-3 (30 ngImL), muGM-CSF (20 ng/mL), huG-CSF (50 nglmL), or huM-CSF (50 nglmL). Plates were scored for colony growth after 7 to 10 days.CFU-G,CFU-GM, and CFU-M were detected in cultures supplemented with G-CSF, IL-3, GM-CSF, or M-CSF. In addition, cultures supplemented with SLF plus IL-3 promoted CFU-Mix (containing granulocytes, macrophages, mast cells, megakaryocytes, or eosinophils). These results were repeated in three separate experiments. t Unseparated BMC or BMC purified by CCE, recovered at 27 m U min (fraction27). were plated in softagar colony assays according to the procedures described in Materials and Methods.

cell) reconstitution in lethally irradiated recipients greater than 4 to 6 months after transplantation (Table 2).7 To determine which growth factors promote the survival of progenitor cells in vitro, CCE-27 cells were plated in soft agar assays supplemented with single growth factors and then were cultured for 24,48,72, and 96 hours before stimulation with the combination of SLF plus IL-3 (SLFBL-3) (outlined in Fig 1). Saturating concentrations of SLF, IL-1, IL-11, or IL-6 alone did not induce colony formation of CCE-27 cells up to 10 days in culture, whereas IL-3, GMCSF, G-CSF, and M-CSF induced few colonies (Table 1). Furthermore, SLF plus IL-3 induced 33 ? 4 colonies and this response was not significantly increased by the addition of any of the growth factors examined as a third cytokine. Using this assay, greater than 90% of the CCE-27 progenitor cells that give rise to colonies in soft agar in response to the combination of SLF/IL-3 at the initiation of the culture survived for 24 and 48 hours in the presence of SLF and gradually declined by 96 hours (Fig 2). These progenitors show an absolute requirement for growth factors, as no progenitor survival was seen in cultures containing medium alone after 24 hours or longer. In comparison, other early acting factors including IL-6, L E , IL-11, and IL-l showed little or no effect on survival even after 24 hours and were comparable with medium controls at 48, 72, and 96 hours (Fig 2). IL-3 showed a reduced but statistically significant positive effect on CCE-27 cell survival after 24 and 48 hours, and G-CSF had an effect at 24 hours, whereas GM-CSF and M-CSF had no effect on survival (not shown in Fig 2). These results could have been affected by the concentration of cytokines because we were pulsing the plates with IL-3/SLF and thereby effectively doubling the concentration of either cytokine. To rule out this possibility, CCE-27 cells were incubated with IL-3 for 48 hours and then were pulsed with SLF alone or SLF/IL-3 or cultures were incubated with

SLF for 48 hours and then pulsed with IL-3 alone or SLF/ IL-3 (Table 3). No significant difference in colony formation was observed in either experiment indicating that increasing the cytokine concentrations above the saturating amounts does not alter the survival results obtained. The direct effect of hematopoietic growth factors on the survival of hematopoieticprogenitor cells. Because the frequency of SLF plus IL-3 responsive progenitors in the CCE27 cell population was too low to detect in single cell assays, CCE-27 cells were further purified by FACS using antibodies that recognize c-kit (Fig 3). In this regard, it has recently been shown that the majority of detectable pluripotential stem cells in the BM are c-kit+ in similarly elutriated BMC pop~lations.~ CCE-27 cells with high side scatter were not sorted because we have determined that they are greater than 90% to 95% mature granulocytes using granulocyte-specific monoclonal antibodies (RB6-8C5') and morphology (data not shown) (Fig 3A). The cells with low side scatter were sorted for high expression of c-kit antigen (see gate selection onFig3C) and were greater than90%to95% c-kit+ on reanalysis (Fig 3D). The FACS profile for the staining of an unrelated isotype-matched control is shown inFig 3B. In soft agar colony assays, greater than90%to 95% of the SLF/IL-3-responsive progenitors in the CCE-27 fraction are c-kit+ with little or no colony formation detected in the ckit- fraction (data not shown). Furthermore, the CCE-27 c-kit+ cells are enriched for pluripotential stem cells in comparison with unfractionated CCE-27 cells (Table 2). CCE27 c-kit- cells (up to 5 X lo5cells per mouse) did not contain radioprotective cells and similar to saline control mice were moribund within 10 to 15 days after irradiation and transplantation (data not shown). To determine the direct effects of cytokines on survival, c-kit+ CCE-27 cells were plated at 1 cell per 20 pL in Terasaki plates in the presence or absence of SLF, IL-3, G-CSF, or medium alone, and then pulsed with 10 pL of SLFBL-3 after 24 and 48 hours (the final concentration of cytokines Table 2. STRCs and LTRCs in CCE BM Cell Populations STRCt

Cell Fraction*

Normal BMCs Saline CCE-27 cells

LTRC (>4 rnodt % Donor Cells

No. of Cells

40- to 60-

Injected

Day Survival

Granulocvte

B Cell

T Cell

313 013 8111 8111 015 415 315 015

93 t 4 NA 87 2 8 82 6 NA 83 2 13 56 C 18 NA

83 2 7 NA 96 2 2 75 2 11 NA 73 2 11 54 2 15 NA

98 2 2 NA 99 2 1 99 2 1 NA 78 2 15 62 i 6 NA

2 X 105 0 1 x lo5 5 x 10' 1 X 104 CCE-27 c-kit+ 1 x lo4 5 x 103 1 X 103

+

Abbreviation: NA, not applicable. * Normal BMCs, CCE-27 cells,and CCE-27-c-kitt cells were obtained as described in Materials and Methods. t Cell populations were injected into lethally irradiated recipients at the indicated concentrations and monitored forsurvival for the first 40 to 60 days (STRC).Surviving mice were examined for donorv host reconstitution of granulocytes, B cells, and T cells more than4 months after transplantation according to the procedures described in Materials and Methods. Mice that were moribund were not applicable.

1760

KELLER,ORTIZ,

AND RUSCETTI

t

Fig 2. The effect of growth factors onthe survival of IL-3 plus SLF-responsive progenitors. CCE-27 BMCs were plated in soft agar at a density of 2.5 x 10' cells/mL/35-mm plate. Cultures were supplemented with medium with or without muMGF ELF) (100 ng/mL), mulL-3 (30 ng/mL), huLlF (200 ng/mL), hulL-l1 (200 nglmL), mulL-6 (200 nglmL), muGMCSF, huG-CSF (50 nglmL),hulL-l(20nglmL), or huM-CSF (50 nglmL). Soft agar plates were pulsed with 200 pL IMDM containing SLF plus IL-3 (final concentration:IL-3, 30 ng/mL; SLF, 100 nglmL) at the indicated times. Plates were scored for colony formation 7 days after pulse.

Pulse WithIL-3 Plus SLF (Hours)

was not altered in this assay). Isolated c-kit+ CCE-27 cells proliferated to form small colonies in response to single growth factors such that SLF, IL-3, and G-CSF induced 4, 9, and 8 colonies per 300 single cells, respectively (SLF- or G-CSF-induced colonies contained 50 cells or less) (Table 4). In comparison, the combination of SLF/IL-3-induced 33 colonies/300 single cells seeded and cells plated in medium alone showed no growth. Combining the results of four separate experiments, between 80% and 90% of the single cells that could proliferate in response to SLFAL-3 atthe initiation of the culture survived in the presence of SLF for 24 hours, and 60% to 75% survived after 48 hours. IL-3 also directly promoted the survival of isolated c-kit+ CCE-27 cells, but to a lesser extent than SLF after 24 and 48 hours, and GCSF directly promoted survival after24 hours, but not at 48 hours and little or no survival was observed in cultures

Table 3. Effect of Growth Factor Concentration on Progenitor Cell Survival CFU-C* Growth Factort Preincubation

Medium IL-3 SLF SLF/IL-3

Cytokine Pulse No Pulse

SLF

0 6 2 1 1 2 1 32 2 4

0 13 2 3

IL-3

0 51-2

0

28 2 2

ND

ND

SLF/IL-3

0 12 2 1 30 t 1 35 2 2

* CCE-27 cells were plated in soft agar assays and scored as described in the legend for Fig2. t IL-3 and SLF or IL-3/SLF were added to the soft agar cultures at at the concentrations the initiation of the experiment ("no pulse") described in Fig 2. Furthermore, 48 hours after incubation in SLF or IL-3, the cultures were pulsed with 200 pL of SLF or IL-3 alone or IL3/SLF at the concentrations also described in the legend for Fig 2. These results were repeated in two other experiments.

containing medium alone (Table 4). Thus, SLF and, to a lesser extent IL-3 and G-CSF, directly promote the survival of CCE-27 c-kit+ progenitors that respond to the combination of SLF/IL-3. However, at this point, the effect of cytokines on proliferation and survival have not been separated. The effect of mitotic inhibitors on SLF- and IL-3-mediated survival o f hematopoietic progenitors. Althoughno visible colony formation was observed in soft agar colony assays and little or no proliferation was observed in Terasaki wells in response to SLF alone, it was possible that some proliferation had occured andthus, it was not possible to separate the effects of growth factors on proliferation from cell survival. In this regard, proliferation assays showed that SLF promoted low, but statistically significant "-thymidine incorporation of CCE-27 cells, whereas medium, IL-6, LIF, and IL-l had no effect (Fig 4A). Because insufficient numbers of c-kit+ CCE-27 were obtained by FACS separation for 'H-thymidine incorporation assays, CCE-27 cells were further purified by removing B220' B cells and RB6-8C5' myeloid cells (Lin-CCE-27) that was 10% to 20% of initial cell population and also contained greater than 90% of the SLFfiL-3 responding progenitors. Lineage-negative CCE-27 also proliferated significantly in response to SLF (Fig 4B). Thus, survival of progenitor cells in culture might have been a consequence of cell division. Therefore, to determine whether cell division was required for survival, the effects of SLF on cell survival were compared in the presence or absence of two mitotic inhibitors, nocodazole (inhibits G2/M transition8)or aphidicolin (blocks in S phase).' CCE-27 BMCs were cultured for 24 or 48 hours in medium plus or minus SLF, IL-3, or G-CSF in the presence or absence of mitotic inhibitors and then washed and replated in soft agar cultures containing SLFAL-3 (Table 5). There were 30 t 2 IL-3/SLF-responsive progenitors in 2.5 x lo4 cells at the initiation of the culture. Similar to the results obtained in soft agar assays above (Fig 2), greater

SURVIVAL OFSTEM/PROGENITORCELLS

1761

A

Fig 3. FACS separation of CC€-27 c-kit+ cells. CCE-27 cells were labeled with anti-c-kit monoclonal antibodies and separated according t o the procedures outlined in Materials and Methods. (A) Forward and side scatter properties of CC€-27cells andthe gate window set for sorting. (B and C) lsotype control anti-c-kit-stained or CCE-27 cells, respectively, and the gate window (C) for separating c-kit+ cells. (D) Reanalysis of the c-kit+ FACS-sorted cells.

.-l

Table 4. Direct Effect of Growth Factors on theSurvival of SLF Plus IL-3 Responsive c-kit+ Progenitors No. of Responders1300 Single Cells Pulse With SLF Plus IL-3 (h)

No Pulse With IL-3ISLF

48

24 ~~

33

Medium SLF/IL-3 SLF IL-3 G-CSF

0 2 2

421 9 2 2 8 2 1

5b

160

150

260

.

2! l

Side Scatter

than 90% of the SLF/IL-3 responsive progenitors detected at the initiation of the assay were recovered from CCE-27 cells in liquid cultures in the presence of SLF, of these, S4% were recovered from liquid cultures in SLF plus nocodazole and 71% in SLF plus aphidicolin after 24 hours (Table 5). After 48 hours, 20% to 2S%of the SLF/IL-3 responsive progenitors in the CCE-27 fraction were recovered from liquid cultures containing SLF plus nocodazole or aphidicolin compared with the controls. In comparison, IL-3 showed some effect on survival after 24 hours in the presence of nocodazole or aphidicolin and little or no effect after 48 hours, whereas G-CSF and other early acting factors (data not shown) had no effect. To show that nocodazole and aphidicolin had inhibited proliferation, CCE-27 cells were cultured with growth factors in the presence or absence of nocodazole or aphidicolin in 'H-thymidine incorporation assays. Both aphidicolin and

Growth Factor Preincubation

d

0 ND 27 -c 1 19 t 2 22 t l

0 ND 25 t 3 19 2 1 8 2 2

nocodazole inhibited the proliferation of IL-3ISLF-induced proliferation of CCE-27 (Fig S). Thus, among the early acting factors, SLF can best promote the survival of progenitor cells in the absence of cell division. The effect of cytokines on the sundval of short-term and long-term reconstituting cells. To determine whether SLF and IL-3 could promote the survival of STRCs and LTRCs in the CCE-27 cell population, CCE-27 cells were plated in liquid culture in the presence or absence of cytokines or in the presence or absence of nocodazole and then transplanted into irradiated recipients after a 36-hour incubation. SLF promoted the survival of CCE-27 cells with short-term reconstituting activity and long-term reconstituting activity in 14of 15 mice in experiments no. 1, 2,and 3, whereas no mice survived with cells injectedfrom cultures incubated with medium alone (Table 6). Furthermore, SLF also promoted the survival of STRCs and LTRCs in six of IS mice in the presence of nocodazole (Table 6). In comparison however, in experiment no. l (Table 6), three of five mice survived when injected with cells cultured in IL-3; these cells had greatly reduced LTRC activity (by percentage of donor reconstitution) in comparison with cells cultured in SLF (Table 6). In addition, irradiated recipient mice did not survive when injected with cells cultured in IL-3 in experiment no. 2 and 3. Finally, none of the irradiated recipients survived when injected with cells cultured in IL-3 plus nocodazole or cells cultured in either IL- I or G-CSF (Table 6). Therefore, similar to the results obtained above, SLF can promote the survival of both STRCs and LTRCs in the absence of cell division.

DISCUSSION CC€-27BMCs were obtained and were further purified by FACS separation as described in Materials and Methods. CC€-27 c-kit' cells Of the nine cytokines examined known to effect hematowere seeded into Terasaki plates at 1 cell per 20 pL and cultured at poietic progenitor cell growth, only SLF and, to a lesser 37°C. 5% CO2.Cultures were supplemented in medium with or without extent, IL-3 can promote the survival of SLF/IL-3 responsive cytokines as described in Fig 2. Individual wells were pulsed with 10 CCE-27 BM progenitors in the absence ofcell division. pL IMDM containing SLF plus IL-3 (final concentration: SLF, 100 ngl Furthermore, these cytokines. directly promote the survival mL; IL-3,30 nglmL) at the indicated timesand scored for growth, > l 0 of the c-kit' subpopulation of CCE-27 cells that contain cells per well, after 7 days. These results are representative of four separate experiments. greater than 95% of the SLF/IL-3 responsive cells in the

1762

loo

KELLER,ORTIZ,

A CCE Fratton27

l

70

P

sLF

ILJ

10

AND RUSCETTI

B CCE L W F m t h 27

l -

L

24

72

48

ydLun

~rrrt~-m~t~h.~uy(~oun)

total CCE-27 cell population and are enriched for pluripotential long-term reconstituting stem cells detected in vivo. In this regard, we show for the first time that SLF promotes the survival of LTRCs in the absence of cell division. However, in the LTRC assay, weused the total CCE-27 cell population, and therefore, although the data suggest that we are affecting the survival of the c-kit+ population, because of the heterogeneity of the stem cell compartment, we cannot state which subpopulation of this compartment survives without further purification. In agreement with these results, previous studies have shown that SLF mediates the survival of primordial germ cells and melanocytes in the absence of proliferation.'".'' In addition, although soluble SLF promoted germ cell survival in these studies, the germ cells showed improved and longer survival on feeder cells that expressed the transmembrane form of SLF. In this regard, although SUSLd mice (which only produce the soluble form of SLF)

Fig 4. The effect of hematopoietic growth factorson the proliferation of CCE-27 and LinCCE-27 cells. Cells,CCE-27 (1 x 10') in (A) and Lin-CCE-27 (5 x 109 in (B), were plated in 96-well microtiter plater and incubated at 37°C for the indicated times. Cultures were supplemented with the indicated cytokines at concentrations described in Fig 2. The wells were pulsed with 3H-thymidine andprocuredfor liquid scintillation counting as described in Materials and Methods.

show impaired development of a variety of stem cells resulting in sterility and lack of pigmentation, they remain viable, suggesting that hematopoiesis can be maintained by soluble SLF, whereas the survival of primordial germ cells and melanoblasts We would like to compare the effects of soluble versus transmembrane SLF on stem cell survival using stromal cell lines that only produce the transmembrane form of SLF; however, the results from such an experiment must be viewed with caution, because survival factors in addition to SLF produced by the stromal cells could contribute to stem cell survival. Because SLF promotes the survival of a portion of the progenitor cells in our assay in the absence of cell division, further characterization of the survival requirements for stendprogenitor cell populations will include assessing the effects of multiple hematopoietic growth factors and extracellular matrix proteins. In this regard, it is important to note

Table 5. Effect of Mitotic Inhibitors on Growth Factor-Mediated Survival of IL-3 Plus SLF Responsive Progenitors No. of CFU-CS

24 hrs

48 hrs

Growth Factor Preincubation

-

Nocodazole

Aphidicolin

Medium SLF IL-3 G-CRF

2 2 1 28 2 1 18 ? 1 71-1

1 ? 1 15 ? l 6 2 2 1 -+ 0.5

2+1 20 i 1 71-1 21-1

~

0 5 + 27 1 c4 20 1- 1 5+1

Nocodazole

0 6 t l 0 0

Aphidicolin

0 2 2 1 1 5 1

CCE-27 were obtained and directly plated in soft agar assays as described in Materials and Methods or were plated in liquid culture in complete IMDM at 5 X lo5 cells per mL and incubated at 37°C.5% CO,. The liquid cultures were supplemented with growth factors at the concentrations described in Fig 2 in the presence or absence of 40 ng/mL nocodazole or 250 ng/mL aphidicolin. This predetermined concentration of nocodazole does not significantly effect cell viability compared with controls and completely inhibits colony formation in soft agar assays. Cells were removed from the liquid cultures at the indicated times, washed three times to remove nocodazole or aphidicolin, and then replated in soft agar assays at 2.5 x 10' cells/mL using the initial cell input to plate in the soft agar assays. Colony formation was scored after 7 to 10 days.

1763

SURVIVAL OFSTEMIPROGENITORCELLS

30 25 Fig 5. The effect ofnocodazole and aphidicolin on SLF plus IL-3-induced proliferation of CCE-27 cells. CCE-27 cells were cultured at 1 x lo5 cellrlmL of medium in S w e l l microtiter plates in the presence (0) or absence (0)of nocodazole (20 nglmL) or presence (A) of aphidicolin (250 nglmL)and incubated at 37°C fortheindicated times. Cells werepulsed with 'H-thymidine and procured for liquid scintillation countingas described in Materials andMethods. presented The is data minute counts mean per as the f SD. The results are representative of three separate experiments.

20 15 P

1 g

2

10

&

5 0

48

24

0

72

0

48

24

72

Harvest 3H-Thymldlne A88ay (Hours)

here, that while the other factors examined in our assays did not promote progenitor survival, they could induce CCE27 cell colony formation in vitro in combination with SLF indicating that the receptors for these growth factors were present on the cells. Thus, this in vitro assay system will allow us to determine the additional requirements needed to promote stem cell survival. Furthermore, Lansdorp et all6 have shown that it is possible to monitor the survival of CD34+ human progenitors in the absence of cell division using the tracking dye PKH26; thus, we would like to compare our biologic survival data with FACS data using PKH26. It is known that stem cells and their differentiated progeny can be maintained in vitro on normal stroma and stromal cell lines." Recently, it has been shown that the addition of

an antibody that blocks the interaction of c-kit with SLF abrogates hematopoietic development both in vitro and in vivo suggesting that the presence of SLF is essential to maintain hematopoietic de~elopment.'~.'~ Interestingly, stem cell levels were maintained in these cultures, thus the survival of primitive stem cells can also be mediated by factors other than SLF. In this regard, there is some evidence that suggests that the FLT-3 ligand might play a role in the proliferation/ survival of CD-34+ progenitors, and therefore, we would like to examine the effects of K T - 3 in our system.*' In agreement with our results, Katayama et a16 have recently examined the survival requirements of a different BM cell population obtained from mice previously injected three times with 5-FU and found that, of a number of growth

Table 6. Effect of Growth Factors on theSurvival of CCE-27 LTRCs and STRCs

No. of Growth Factor

Cells Injected

LTRC % Donor Reconstitution Nocodazole

40-60 d Survival

Granulocyte ~

Experiment no. 1 Medium Medium SLF SLF IL-3 IL-3 Experiments no. 2 and 3 Medium SLF SLF IL-3 IL-3 IL-l G-CSF

B Cell

T Cell

-

~~~

X

105 105 X 105 X 105 X 105 X 105

-

015

-

-

X

+

015

-

-

-

515 215 315 015

95 z 3 91 1- 5 61-5

77 t 15 95 f 2 20 t 10

t 11 98 t 1 t 11

-

-

-

5 5

X

105 105 105 105 105

5 5

X 105

-

105

-

5 5 5 5 5 5

5 5 5

X

X X

X

X

-

+ -

+ -

+ -

+

015 9110 4110 015 015 015 015

-

88 34

-

-

98 f 1 83 t 2

87 _f 8 87 t 6

-

85 t 11 87 f 14

-

-

-

-

-

-

-

-

-

-

-

-

CC€-27 cells from donor C571BL6 (Ly 5.2) mice were cultured at 5 x IO5 cells per mL in complete IMDM in the presence or absence of cytokines at the concentrations indicated in Fig 2, and nocodazole was added to some of the cultures at the concentration indicated in Table 5. After a 36-hour incubation, cells were removed from culturedishes and washed twice and resuspended in 200 @LPES. Cells were injected into recipient mice with each mouse receiving the contents of one well, and recipient mice were analyzed for long- and short-term reconstitution according to the procedures outlined in Materials and Methods.

1764

KELLER, ORTIZ, AND RUSCETTI

factors tested, both SLF and IL-3 could promote the survival of progenitors that respond to the combination of IL-3 plus IL-6 plus erythropoietin in vitro. In comparison, Leary et a12’ have examined the growth factor survival requirements ofpurifiedhuman CD34+ HLA-DR- marrowprogenitors that give rise to blast cell colonies (induced in response to IL-3 plus IL-6)andshowedthat,althoughnoneofthese progenitors survived the delayed addition of growth factors, both IL-3 and GM-CSF could support their survival, whereas SLF, IL-6, and G-CSF werenoteffective.2’Incontrast, Brandt e t a122 have shown that IL-3, but not SLF, could support the survival of CD34+ HLA-DR+ (more committed progenitors), whereas SLF, but not IL-3, could promote the survival of CD34+ HLA-DR- c-kit’ progenitors (more primitive progenitors).22 Although these authors did not rigorouslyruleoutpotentialindirecteffects(usingsingle cell assays) or some effects on progenitor proliferation, both results suggest that SLF a n d o r IL-3 promote stem cell survival, and that the differences in the survival requirements may be a resultofstedprogenitor cell heterogeneity.” Therefore,wehaveinitiatedexperimentstocomparethe survival requirements of a number of separated a n d o r e n riched stedprogenitor cell populations. ACKNOWLEDGMENT We thank Dr Dan L. Longo for review of the manuscript and John Wine, Louise Finch, and Jeanette Higgins for their excellent technical assistance.

REFERENCES 1. Ogawa M: Differentiation and proliferation of hematopoietic stem cells. Blood 81:2844, 1993 2. Harrison DE, Lerner CP: Most primitive hematopoietic stem cells are stimulated to cycle rapidly after treatment with 5-fluorouracil. Blood 78:1237, 1991 3. Koury MJ: Programmed cell death (apoptosis) in hematopoiesis. Exp Hematol 20391, 1992 4. Williams DE, Broxmeyer HE: Interleukin-la enhances the in vitro survival of purified murine granulocyte-macrophage progenitor cells in the absence of colony stimulating factors. Blood 72:1608, 1988 5. Bodine DM, Crosier PS, Clark SC: Effects of hematopoietic growth factors on the survival of primitive stem cells in liquid suspension culture. Blood 78:914, 1991 6. Katayama N, Clark SC, Ogawa M: Growth factor requirement for survival in cell cycle dormancy of primitive murine lymphohematopoietic progenitors. Blood 81:610, 1993 7. Orlic D, Fischer R, Nishikawa S, Nienhuis AW, Bodine DM: Purification and characterization of heterogeneous pluripotent hematopoietic stem cell populations expressing high levels of c-kit receptor. Blood 82:762, 1993 8. De Brabander MJ, Van de Veire RML, Aerts FEM, Borgers

M, Janssen PA: The effects of methyl (5-(2-thienylcarbonyl)-II benzimidazol-2-yl) carbamate a new synthetic drug interfering with microtubules on mammalian cells cultured in vitro. J Cancer Res 36:905, 1976 9. Spadari S, Sala F, Pedrali-Noy G: Aphidicolin and eukaryotic DNA synthesis. Trends Biochem Sci 7:29, 1982 10. Dolci S, Williams DE, Irnst MK, Resnick JL, Brannan Cl, Lock LF, Lyman SD, Boswell HS, Donovan PJ: Requirement for mast cell growth factor for primordial germ cell survival in culture. Nature 352309, 1991 I I . Steel KP, Davidson DR, Jackson U: TRP-2/DT, a new early melanoblast marker, shows that steel growth factor (c-kit ligand) is a survival factor. Development 115:111 1 , 1992 12. Silvers WK: The Coat Colours of Mice: A Model for Mammalian Gene Action and Interaction. New York, NY, Springer, 1979, p 243 13. Russell ES: Hereditary anemias of the mouse: A review for geneticists. Adv Genet 20:357, 1979 14. Flanagan JG, Chan DC, Leder P: Transmembrane form of thekit ligand growth factor is determined by alternative splicing and is missing in the SLd mutant. Cell 64:1025, 1991 15. Brannan CI, Lyman SD, Williams DE, Eisenman J, Anderson DM, Cosman D, Bedell MA, Jenkins NA, Copeland NG: SteelDickie mutation encodes a c-kit ligand lacking transmembrane and cytoplasmic domains. Proc Natl Acad Sci USA 88:4671, 1991 16. Lansdorp PM, Dragowska W: Maintenance of hematopoiesis in serum-free bone marrow cultures involves sequential recruitment of quiescent progenitors. Exp Hematol 2 1 :1321, 1993 17. Dexter TM, Allen TD, Lajtha LG: Conditions controlling the proliferation of haemopoietic stem cells in vitro. J Cell Physiol 91:335, 1977 18. Kodama H, Nose M, Yamaaguchi Y, Tsunoda J, Suda T, Nishikawa S, Nishikawa S: In vitro proliferation of primitive hemopoietic stem cells supported by stromal cells: Evidence for the presence of a mechanism(s) other than that involving c-kit receptor and its ligand. J Exp Med 176:351, 1992 19. Wineman JP, Nishikawa S, Muller-Sieburg C E Maintenance of high levels of pluripotent hematopoietic stem cells in vitro: Effect of stromal cells and c-kit. Blood 81:365, 1993 20. Small D, Levestein M, Kim E, Carow C, Amin S, Rockwell P, Witte L, Burrow C, Ratajczak MZ, Gewirtz AM, Civin CI: STKI , the human homolog of FLK-2/FL.T-3, is selectively expressed in CD-34’ human bone marrow cells and is involved in the proliferation of early progenitorktem cells. Proc Natl Acad Sci USA 91 :459. 1994 21.Leary AG, Zeng HQ, Clark SC, Ogawa M: Growth factor requirements for survival in Go and entry into the cell cycle of primitive human hemopoietic progenitors. Proc Natl Acad Sci USA 89:4013, 1992 22. Brandt JE, Bhalla K, Hoffman R: Effects of interleukin-3 and c-Kit ligand on the survival of various classes of human hematopoietic progenitor cells. Blood 83:1507, 1994 23. Civin CI, Banquerigo ML, Strauss LC, Loken MR: Antigenic analysis of hematopoiesis. IV. Characterization of MY IO-positive progenitor cells in normal human bone marrow. Exp Hematol 15:10, 1987