1. No category

From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
RAPID COMMUNICATION
Role of Interleukin-l5 in the Development of Human CD56' Natural Killer
Cells From CD34+ Hematopoietic Progenitor Cells
By Ewa Mrozek, Paul Anderson, and Michael A. Caligiuri
Human natural killer INK) cells are bone marrow (BM)-derived CD2+CD16+CD56+large granular lymphocytes (LGL)
that lack CD3 yet contain the
T-cell receptor (-chain (c-TCR).
NK cells provide requisite interferon-y (IFN-y) during the
early stages of infectionin several experimental animalmodels. A number of studies haveshown that human
CD3-CD56'
NK cells can be obtained from BM-derived CD34+ hematopoietic progenitor cells (HPCs) cultured in the presence of
interleukin-2 (IL-2) and an allogeneic feeder cell layer, or IL2 and other hematopoietic growth factors such as the c-kit
ligand (KL).The failure t o detect the IL-2 gene product within
the BM stroma and presence
the
of NK cells in IL-2-deficient
mice suggested that cytokines other than IL-2 may participate in NK cell differentiation from HPCs in vivo. IL-15 is a
cytokine which, while lacking any sequence homology t o IL2, can activate cells via the IL-2 receptor. Here we show that
human BM stromal cells express the IL-15 transcript, and
supernatants from long-term BM stromalcell cultures contain IL-15 protein. In vitro, CD3-CD56+ NK cells can be obtained from 21-day cultures of CD34+ HPCs supplemented
with IL-15 in the absence of IL-2, stromal cells, or other cytokinas. The addition of the KL t o these cultures had noeffect
on the differentiation of the CD3-CD56+ cytotoxic effector
cells, but greatly enhanced their expansion. The majority of
these cells lack CD2 and CD16, but do express I;-TCR. Similar
t o NK cells found in peripheral blood, the CD2-CD16-CD56'
NK cells grown in the presence of IL-15 were found t o be
potent producers of IFN-y in response t o monocyte-derived
cytokines. Thus IL-15, like KL, appears t o be produced by
BM stromalcells. IL-15 can induce CD34+ HPCst o differentiate into CD3-CD56+ NK cells, and KL can amplify this. Therefore, IL-15 may be a physiologically relevant ligand for NK
cell differentiation in vivo.
0 1996 by The American Societyof Hematology.
H
fold expansion ofhuman CD3-CD56' NK cells. Silva et al"
established a cell culture system without stromal cells that
allowed the CD34' HPCs to differentiate into N K cells in
the presence of L I P , 1L-2, and c-kitligand (KL). Most
recently, Shibuya et al' showed that cultures of HPCs with
KL alone orKL and IL-3 for7 days followedby the addition
of IL-2 for 21 days resulted in generation of CD3-CD56'
NK cells.
In each study, IL-2 was required for NK cell differentiation from CD34+ HPCs. Whilethis strongly suggests that at
leastafraction
of CD34' HPCsexpress functional IL-2
receptors(IL-2R), a n endogenoussource of IL-2for NK
cell differentiation in vivo has yet to be identified. Antigenactivated T cells are currently the only known source of IL2.' T-cell-deficientmice
have normal NK cell development," and NK cells can be foundin mice with a genetically
disrupted IL-2 gene."'," However, mice" and humans" who
lack the y< subunit of the 1L-2R (IL-2Ry) lack NK cells.
Collectively, these data suggest that cytokines other than IL2 which use IL-2Ry may be important in NK cell development.IL-15isanovel
cytokine thatlacks any sequence
homology with 1L-2, yet can activate T cells" and NK cells"
via components of the IL-2R.I' Moreover, the human IL- I5
gene was originally cloned from a human BM stromal cell
line, and its transcript is widely expressed in a number of
human tissues."
In the present report we analyzed human BM stromal cell
cultures for the IL- I5 gene product, investigated the role of
1L- 15 in the development of CD3-CD56' cells from purified
CD34+ HPCs, and further characterized this in vitro-generated, BM-derived population of NK cells.
UMAN natural killer (NK) cells are a distinct population of large granular lymphocytes (LGL)
that play
animportant role in the host's early immune response to
infection andmalignant transformation. Althoughthere
have beena number of important advances in our understanding of NK cell target recognition, signal transduction,
and cytokine production, the developmental sequence leading to generation of mature NK cells remains to be fully
'
There have been at least four studies demonstrating that
human NK cells, defined by LGL morphology, CDYCD56'
phenotype and non-MHC-restricted cytotoxicity, can be derived from bone marrow (BM) CD34' hematopoietic progenitor cells (HPCs) in the presence of recombinant interleukin-2 (IL-2). Lotzova etaI4 have shown that a long-term BM
culture (LTBMC) of CD34'CD33- cells in the presence of
10% fetal bovine serum and variable amounts of human IL2 resulted in the generation and 2- to 200-fold expansion of
CD3-CD56' NK cells. Miller et als demonstrated that culture of CD34'DR- HPCs in the presence of an allogeneic
irradiated BM stromal cell layer and IL-2 led to 147 t- 2 I -
From the Division of Medicine, the Departments of Hematologic
Oncology and Bone Marrow Transplantation, the Department of
Molecular Medicine. Roswell Park Cancer Instituie, Bufalo, NY;
and the Division of Tumor Immunology Dana Farber Cancer Institute, the Department of Rheumatology, Brigham and Women's Hospital, Harvard Medical School, Boston MA.
Submitted November 14, 1995; accepted December 21, 1995.
Supported in part by the Coleman Leukemia Research Fund.
Address reprint requests to Michael A. Caligiuri, MD, Rowel1
Park Cancer Institute, Buffalo, N Y 14263.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in uccordunce with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
000~-4971/96/8707-0054$3.00/0
2632
MATERIALSANDMETHODS
PuriJication of' human BM progenitors. BM was aspirated from
the posterior iliac crest of healthy adult donors after obtaining informed consent. BM mononuclear cells (MNC) were separated by
density centrifugation over a Ficoll-Hypaque (Sigma Diagnostics,
St Louis, MO) gradient at 400g for 30 minutes. CD34' cells were
Blood, Vol 87, No 7 (April l), 1996: pp 2632-2640
PMENT
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
CELL
ROLE OF IL-15 IN NK
2633
1
2
3
4 5
-67
P-"
Fig 1. (A) Nested RT-PCR
analysis of IL-15 gene expression. Lane 1, water control; lane
2, lipopolysaccharide-activated
human monocytes; lane 3,
mouse cell line P-618; lane 4
through 6, human BM stromal
cells from three different donors.
Below: @-actin control for human (lanes 1, 2,4, 5, and 6) and
mouse (lane 3) samples. (B)
Southern blot analysis for detection of the 11-2 RT-PCR product.
Lane 1, water control; lane 2,
PHA-activated T cells; lane 3, K562 cell line; lanes 4 through 6,
human BM stromal cells from
t h e same three donors used in
(A); lane 7, manufacturer's positive cDNA control. Below: p-actin control.
B
1
2
3
4 5 6 7
11-2
1 2 3
P-actin
purified from MNC by panning, using flasks coated with anti-CD34
monoclonal antibody (MoAb) (Applied lmmune Sciences, Santa
Clara, CA) or by affinity chromatography with the Ceprate device
(Cellpro, Bothel, WA), following the manufacturer's instructions.
The enriched cells were next stained with anti-CD34-PE. anti-CD2FlTC (Becton Dickinson, San Jose, CA) and anti-CDI6-FITC (Caltag Laboratories, San Francisco, CA) and sorted for the
CD34'CD2-CD16- (CD34*LlN-) population on a FACStar Plus
cell sorter (Becton Dickinson). Purity was always ~ 9 7 % .
Long-term suspension culture of CD34' HPCs. Sorted
CD34'LlN- HPCs were cultured in 96-well U-bottomed microplates
at a concentration of 2 X IO4 cells in 200 pL of RPM1 1640 (GIBCO,
Grand Island, NY), 10% heat-inactivated human AB (HAB) serum
4 5
6
1
(C-Six Diagnostics, Mequon, WI), andantibiotics. The experimental
cultures were supplemented with 1 0 0 n@mL of recombinant (r) IL15 andlor 1 0 0 nglmL ofrKL (Immunex Corp. Seattle, WA). For
one set of experiments, CD34'LlN- HPCs were cultured in 230 1Ul
mL of rIL-2 (Hoffman LaRoche, Nutley, NJ; specific activity of
1.53 X 10' Ulmg) and 1 0 0 ng/mL of rKL. Plates were incubated at
37°C in 5% CO2 humidified atmosphere for 21 days. At day 7, day
14,and every 3 days thereafter, halfofthe culture mediumwas
replacedwith fresh medium containing 10% HABandthe same
concentration of growth factors. After 21 days, the cells were enumerated and analyzed for morphology, cytotoxicity, phenotype, and
cytokine production.
Assessment of immunophenotype. cytoxiciw. and proliferation.
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
MR~ZEK,ANDERSON, AND CALIGIURI
2634
A
CD3-FITC
E!
IL-l5+KL
MSlg-FITC
The cells generated in 21day cultures were first incubated with
nonreactive mouse Ig for 10 minutes on ice and then stained with
directly conjugated MoAbs: NKH-l(anti-CD56)-PE (Coulter Immunology, Hialeah, FL) and anti-CD16-RTC and/or Leu-5b(anti-CD2)FITC and/or Leu-4(anti-CD3)-RTC (Becton Dickinson). Background fluorescence was determined on cells stained with directly
conjugated nonreactive MoAbs (Becton Dickinson). Cells were analyzed using an MNC gate on a FACScan with the Lysis I1 software
program (Becton Dickinson). Cytoplasmic staining for I;-TCR on
digitonin-permeabilized CD56+ BM-derived NK cells was performed as previously described." Cytotoxic activity of cultured cells
CD3-FITC
Fig 2. Phenotypic and quantitative assessment of cells d e
rived fromCD34' HPCs cultured
for 21 days under indicatedconditions. (A) Phenotypic assessment of CD34' HPCs after a 21day culturein the presence of KL
alone, 11-15 alone, or a combination of11-15 and KL. (B) See next
page. Fold increase in absolute
number of CD34'HPCs after a
21-day culture in the presence of
medium alone, 11-15 alone, KL
alone, or a combination of 11-15
and KL. (C) Absolute number of
CD56' cells cultured under conditions shown in (B). The absolute number of CD56' cells was
calculated by multiplying the total number of viable
cells by the
percentage of cells expressing
CD56 antigen on flow cytometric
analysis. Results represent the
mean k SE for threeseparate experiments.
was determined by a standard "Cr-release cytotoxicity assay against
the NK-sensitive K562 and NK-resistant COL0 205 cell lines as
previously described,lx at an effector-to-target ratio of four. Cell
proliferation was measured bymethyl-['H]
thymidine incorporationIx and cell enumeration was performed by vital dye exclusion
using a standard hemocytometer. Results for cytotoxicity and proliferation assays represent the mean ? SE of triplicate wells, and a
Student's paired r-test was used where indicated.
BM stromal cell cultures. Fresh BM MNC were suspended in
75-cm2 tissue culture flasks at a concentration of 0.5 to 1.0 X 10"
cells/mL in minimum essential a-medium with 15% fetal calf serum
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
2635
ROLE OF IL-15 IN NK CELL DEVELOPMENT
nedium
IL-15
KL
C
T
Mdlum
!L-15
Fig 2
IL-15
KL
+
KL
(Cont'd).
(GIBCO) and 1 X
m o m hydrocortisone (Upjohn, Irving, TX)
according to the method of Wetzler et aI.l9 Cultures were maintained
at 37°C inhumidified 5% CO2. Medium was replaced and nonadherent cells removed every 7 days. Cultures reached confluency between
4 and 6 weeks, at which time they were washed twice to remove
nonadherent cells and then treated with 0.25% trypsin (GIBCO) for
I O minutes to remove the adherent layer for RNA extraction. Under
these conditions the adherent layer was found to contain 3% to 5%
monocytes/macrophages. Culture supernatants obtained immediately
before initiating the process of adherent cell removal were collected
and stored at -70°C.
RNA extraction, polymerase chain reaction (PCR) amplification,
andSouthern analysis of PCR product. Total RNA was prepared
from whole cells using RNAzol in accordance with the method of
Chomczynski.**For all reverse transcription PCRs (RT-PCR), firststrand synthesis was performed at 37°C for l hour in 30 pL reaction
mixture containing 10 pmol/L of downstream primer, 1 pg of total
RNA as the template, 200 U Superscript I1 RT reverse transcriptase
and dNTPs (GIBCO) at 0.2 mmol/L each, followed by chilling on
ice for 5 minutes. Two microliters from the IL-2 RT reaction was
amplified by PCR with the human IL-2 amplimer set from Clontech
Laboratories (Palo Alto, CA), following the manufacturer's recommendation. To confirm the presence of an RT-PCR product specific
for the IL-2 gene, Southern blot analysis was performed using
5'ATCCCAAACTCACCAGGATGCTCA3' as a 32P-labeledprobe
as described.*' Two microliters of the human IL- 15 RT reaction was
amplified by PCR with the following set of primers: S'AACTGG-
GTGAATGTAATAAGTGA"ITG3' (upstream) and S'AGTGTTGATGAACATTTGGACAATATG3' (downstream). Amplification
conditions for each cycle were: 1 minute at 94°C 1 minute at 55"C,
and l minute 30 seconds at 72°C. A total of 35 cycles were followed
by a final l-minute extension at72°C. A second round ofPCR
amplification with the nested set of primers S'AAAA'ITGAAGATCTTATTCAATCT3' (upstream) and 5'AAAACTCTGCAAAAA'ITCTITAAT3' (downstream) was performed underidentical conditions. The products were separated on 2% agarose gels in
Tris-Acetate-EDTA buffer (Boehringer Mannheim, Indianapolis, IN)
containing 0.2 pglmL ethidium bromide. The integrity of RNA samples was verified by amplification of the @actin transcript. Peripheral
blood (PB) MNC activated with 10 p g / d of phytohemaglutynin
(Sigma) and 1 0 0 IU of rIL-2 served as a positive control for expression of IL-2 mRNA, whereas PB monocytes activated with 10 pg/
mL of LPS (Sigma) served as a positive control for expression of
IL-15 rnRNA. Nucleotide sequence analysis of the human IL-15
PCR product showed that the appropriate mRNA species was amplified.
Measurement of cytokine production. CD34' BM HPCs,
CD3-CD56b"ghrPBNK cells, and 21-day cultures of CD56+LINNK cells were each sorted (297% purity) by flow cytometry using
a FACStar Plus (Becton Dickinson) and plated at 50,000 cells/well
in RPMI with 10% HAB. After overnight incubation in the absence
of any cytokines, cells were costimulated with 100 nglmL IL-15 and
10 U of rhuIL-l2 (Genetics Institute Inc, Cambridge, MA: specific
activity 4.5 X lo6 Ulmg). After 72 hours supernatants were collected
and frozen at -70°C until assayed simultaneously. Commercial
enzyme-linked immunosorbent assay (ELISA) kits wereused for
measuring granulocyte-macrophage colony-stimulating factor (GMCSF), tumor necrosis factor-a (TNF-a)(R & D Systems, Minneapolis, MN), and IFN-y (GIBCO). Supernatants from BM stromal cell
cultures were thawed and concentrated IO-fold using Centricon Concentrators (Amicon, Beverly, MA). RPMI 1640 with 10% HAB and
RPM1 1640 with 10% fetal bovine serum (FBS) were used as negative controls and were also concentrated 10-fold. All samples were
then assayed for IL-15 protein using our human L - 1 5 ELISA,Z2
with a sensitivity of 10 pg/mL. Identical samples were assayed for
human IL-15 by a commercially available ELISA (BioSource International, Camarillo, CA) with a sensitivity of 11 pg/mL, and were
assayed for IL-2 by ELISA (Endogen, Inc, Cambridge, MA) with a
sensitivity of <6 pg/mL.
Immunobloting analysis for the expression of C-TCR prorein.
Western analysis for c-TCR was performed as previously described." Briefly, after 3-week cultures of HPCs in the presence of
IL- 15 and KL or IL-2 and KL, sorted CD56+LIN- cells were solubilized in NP-40 lysis buffer (1% NP-40, 150 m m o E NaC1, I mmoll
L EDTA, 1 m m o E PMSF, 50 m m o E Tris, pH 8.0) for 30 minutes
on ice, then centrifuged for 30 minutes at 12,OOOg. Supernatants
were diluted 1:l with sample buffer (2% sodium dodecyl sulfate
[SDS], 10% glycerol, 0.1 m o m Tris HC1, pH 6.8, 0.02% bromphenol blue), separated on 10% SDS polyaclylamide gels, transferred
to nitrocellulose, andprobedwith the anti-<-TCR MoAb TIA-2.
Controls included CD3-CD56b"sh' N K cells sorted from PB (positive), and the COL0 205 cell line (ATCC, Rockville, MD; negative).
RESULTS
Human IL-15and IL-2gene products in human stromal
cells. We analyzed human BM stromal cell cultures for the
presence of the IL-2and IL-I5gene products. As shown in
Fig 1, transcript for IL-15 wasdetected in stromal cells
from three of three individuals tested by RT-PCR, whereas
transcript for IL-2 wasnotdetected by RT-PCR. Further,
supernatants collected from these cultures were concentrated
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
2636
IO-fold and tested for the presenceof IL- 15 and I t - 2 protein
by ELISA. After correction for concentration, IL- 15 protein
was detected in two of the three samples at a concentration
of 3.7 pg/mL and 1.6 pg/mL, respectively, while undetectable in the third sample and in negative controls. In contrast,
IL-2 protein wasnotfound in any of the samples by an
ELISA that could have detected <0.6 pg/mL in these concentrated supernatants.
Effect of IL-15 and KL on the generution qf NK cells from
CD34' HPCs. A purified population of CD34'
HPCs was
cultured for 21 days in thepresence of KLalone,IL-15
alone, or a combination of KL and IL-IS. The results are
summarized graphically in Fig 2A-C. Cells cultured for 21
days in medium alone neither expanded nor
differentiated
intoCD56' NK cells (not shown).CD34' HPCs cultured
with KL had a 5.7 2 0.6-fold increase in absolute cell nwnber but did not develop CD56 surface antigen expression or
LGL morphology. CD34' HPCs cultured for 21 days with
IL- 1S alone had minimal ( I . 1 +- 0.3-fold) increase in absolute cell number, yet did acquire CD56 antigen expression
(-40% to S % ) , and LGL morphology. CD34' HPCs cultured for 21 days in the presence of IL-l5 and KL showed
cellnumber with -40% to
a 6.5 5 0.7-foldincreasein
85% of cells expressing CD56with LGL morphology. Thus,
while IL- IS alone can induce CD34+ HPCs to express CD56
surface antigen anddevelopLGL
morphology during 21
days of culture, IL- 1S and KL are required for differentiation
and significant expansion of the CDS6' NK population. This
was confirmedin follow-upstudies where CD34' HPCs
were cultured underdifferentconditions
and assayed for
proliferation on days I , 7, and 21 by [?H]-thymidineincorporation. As shown in Fig 3, by day 7 there was a significant
difference in 'H-TdR incorporation by cells cultured in IL15 alone compared to cells cultured with KL or KL and ILIS ( P < .OOS).
CD56 antigenexpressionwasanexcellentpredictor
of
NK cytotoxic activity. CD34' HPCs cultured for 21 days in
IL-IS alone or 1L-15 and KL showed an equivalent amount
of cytotoxic activity against K562 tumor cell targets when
corrected for cell number (Fig 4A). The same was true of
the NK-resistant target, C O L 0 205 (not shown). In contrast,
no cytotoxic activity was seen from CD34+ HPCs cultured
for 21 days in KL alone.
The vast majority (>95%) of CD56' cells generated in
21-day culture of HPCs in the presence of IL-I S , with or
without KL, had LGLmorphology with high-density surface
expression of CD56 (CD56br'gh'),and all lacked expression
of CD34. Lineage marker analysisreproducibly showed that
the BM-derivedCDS6h"gh' cells didnot expressCD2or
CD16 surface antigens (Fig 4B), in contrast to the majority
of PB NK cells.'." The CD56hr'ghh'LIN"
phenotype is similar
to that observed for CD34' HPCs cultured in the presence
of IL-2.4.S.".'d
1;-TCRprotein expression in the CD56'LIN- BM-derived
N K cells. 1;-TCR is found in PB NK cells and is biochemically and functionally linkedto CD16.2s"7 C-TCR protein
expression was analyzed in CD56'LIN- cells derived from
CD34' HPCs of three different donors,cultured for 21 days
in KL + IL-IS. (-TCR was also analyzed in CDS6'LIN
MROZEK, ANDERSON,AND CALIGIURI
cells derived from CD34' HPCs of three different donors,
cultured for 21 days in KL + IL-2. Both populations were
found to expressC-TCR protein by immunoblot analysis (Fig
S ) and on flow cytometricanalysis of permeabilizedcells
(not shown). CD56'LIN- NK cells cultured in KL and 1L2 appeared toexpress less c-TCR than that detected i n
CD56+LIN- NK cells cultured in KL and L I S .
Cyrokine production by CD56' LIN BM-derirvd N K cells.
To date, the cytokine production by CD56'LlN NK cells
derived fromCD34*HPCs
has not been investigated.
Costimulation of PB NK cells with monocyte-derived cytokines 1L-12 and IL-15 provides an optimal stimulus for NK
of INF-y,
TNF-a,
and
GM-CSF.I5
cell
production
CDS6+LIN~-cells weresorted
from 21-day cultures of
CD34' HPCs in IL-15 and KL. CD3"CD56hr''h' PB NK cells
andCD34' BM HPCs were also sorted from freshblood
and BM, respectively. Cells were next incubated overnight
in tnedium alone and then costimulated with IIL-l2 and rlLIS for 72 hours. Supernatants werecollectedandassayed
for cytokine production by ELISA. Results are displayed in
Table 1 and show that CD56'LIN"BM-derived
NK cells
produced significant amounts of INF-y a s well as moderate
amounts of TNF-a and GM-CSF. compared with PB NK
cells. In contrast,freshlyisolated
CD34-HPCs produced
minimal amounts o f all three cytokines.
DISCUSSION
A number of studies have now demonstrated that NK cells
can be generated in vitro from CD34' human BM HPCs in
LTBMC with IL-2.4".'4 Endogenous IL-2 isthought to be
only transiently produced by antigen-activated T cells, which
themselvesexpress ahigh-affinityIL-2R
and require the
lymphocytotrophic hormone for clonal expansion.x Further,
NK cells can develop in the absence of T cells or IL-2,'"''
but not in the absence of IL-2Ry."." Collectively these data
suggest thatIL-2 is probably not required and notsufficiently
available for NK cell differentiation in vivo, and that other
cytokines whichuse IL-2Ry are likely to be important in
this process. Recently, human
IL-IS has been cloned from
12
IL-lS+UL
10
a
6
a
U
t
4
I
0
2
n
Day 0
Day 7
Day 21
Fig 3. %-iTdR
thymidine incorporation by CD34+ HPCs cultured
for 1 day, 7 days, or 2 1 days in the presence of medium alone, IL-15
alone, KL alone, or a combination of 11-15 and KL. Each result represents the mean CPM k SE of three separate experiments.
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
ROLE OF IL-15 IN NK CELLDEVELOPMENT
A
"1
2637
7
Medium
IL-15
KL
IL-lS+KL
Mslg-FITC
CDB-FITC
Fig 4. (AI Cytotoxic activity of CD34' HPCs cultured for 21 days
in medium alone, IL-15 alone, KL alone or combination of IL-15 and
KL. The effector-to-target ratio was4. Results represent the mean f
SE for five separate experiments. (B1 Phenotypic characterizationof
cells generated from a 21-day culture of CD34' HPCs in the presence
of both IL-15 and KL. Results displayed are representative of 10 separate experiments.
a human BM stromal cell line," and has been shown to use
the IL-2RP and IL-2Ry protein subunits for binding and
signal transduction in T cells and NK cells.'".'" Despite lacking any sequence homology with IL-2, IL-15 has thus far
been shown to share virtually identical bioactivities with IL2. Moreover, IL-15 transcript is widely expressed in a number of tissues. These data led us to hypothesize that 1L-IS
could be involved in human N K cell differentiation from
BM-derived HPCs, and the current study was performed to
test this hypothesis.
Several novel observations have been made in this report.
First, we were able to show that long-term cultures of human
BM stromal cells obtained from normal individuals constitutively express transcript for human IL-IS and secrete IL-IS
protein into culture supernatants. We have not identified the
specific cellular components ofBM stroma responsible for
IL-IS production. However unstimulated CD14' monocytes,
comprising -S% of cultures, do not produce detectable ILIS protein,'' suggesting that other stromal cells are likely to
be involved. We were unable to detect 11,-2 mRNA in BMderived stromal cells using RT-PCR. consistent with an earlier report," and we did not find II,-2 protein in concentrated
stromal cell supernatants. Thus, IL-IS is :I ligandthat not
only binds to components of the1L-2R.butisproduced
within the marrow stroma, making it a more physiologically
relevant ligand for the 1L-2Rthoughttohe
constitutively
expressed on CD34" hematopoietic progenitors.','" Definitive proof of a role for 1L-IS in N K cell development will
have to await gene disruption studies in the mouse.
Results presented here demonstrate for the first time that
IL-IS and KL are the only two cytokines required for the
differentiation and expansion of the CDS6'LIN- N K cells
from CD34' HPCs. Others have shown that BM stromal
cells produce boththe membrane-bound and soluble isoforms of KI, transcript.'".'' Thus, it seems plausible that the
CD34" HPC subset which differentiates into a CDS6' NK
cell could coexpress IL-2Rpy and c-kitprotein receptors
either constitutively or early in their differentiation pathway.
receiving coactivation signals from their respective ligands
produced by the human stromal cells. IL-IS and KL have
relatively distinct functions in N K cell development from
CD34" HPCs. In the absence of IL-2, IL-IS appears to be
essential for CD%' antigen expression, LGL morphology,
and functional differentiation. Given the low NK precursor
frequency previously reported within a CD34' HPC subset,>.'?and our results demonstrating a modest expansion of
CDS6' N K cells from CD.34' HPCs in the presence of ILI S done. it is likely that IL- I S can induce some proliferation
in the absence of KL. In contrast, KL alone has virtually no
influence on N K cell differentiation from CD34' HPCs. yet
can clearly amplify the expansion of the CDS6'. NK population in the presence IL-IS. Therefore. 1L-IS may serve primarily to promote differentiation of CD34' HPCs that are
already committed as NK cell precursors, whereas KL may
serve to amplify this response. Additionally. I L - IS may serve
to promote less committed HPC populations toward N K cell
differentiation. Further elucidation of the effects IL- I S may
have on this pathway are likely to require the identification
and characterization of the CD34' HPC that is committed
to N K cell differentiation.
The synergy seen between IL- IS andKL in promoting
N K cell differentiation and expansion from CD34.' HPCs in
vitro is supported by our earlier identification of an NK cell
subset found in PR which is unique among human lymphocytes in its constitutive expression of1L-2R andc-kit.>'
Alone, KL functions as a survival factor for the c-kit' N K
cells,3" and significantly enhances IL-2- or IL-IS-induced
NK cell proliferation'3 (and W.E. Carson. M.A. Caligiuri,
unpublished observation, July 1994). Interestingly, this subset of c-kit' PR N K cells appears to be less differentiated
than the majority of N K cells hecause they maintain a high
proliferative potential and either lack CD16 or express it in
low density. The more ahL;ndant CD2*CD16'CDS6""" PB
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
2638
AND CALlGlURl
I 2 3 4 5 6 7 8 9 1 0
30 -
0.6 1.0 0.50.250.71.4 1.1 1.4 1.1 11.3
Number of Cells / Specimen (x106)
Fig 5. Characterization ofc-TCR in CD56'LIN- NK cells derived from CD34'HPCs. Western blot analysis of cell lysates using an MoAb
reactive against 5-TCR (TIA-2). The cell number contained in each lysate is shown below each lane. Lanes 1 through 3, CD56'LlN- NK cells
generated from BM CD34' HPCs in the presence of 11-15 and KL. Lanes 4 through 6, CD56'LIN- NK cells generated from BM CD34' HPCs in
the presence of IL-2 and KL. Lanes 7 and 8, CD56"igh' PB NK cells. Lanes 9 and 10, COL0 205 cell line.
NK cells neither express c-kit nor show significant 'H-TdR
allylinked to 6-TCR in NK
Therefore, it was someincorporation in the presence of IL-2 or IL-lS.'s~'' Thus, it
what surprising to find that the CDS6'LIN- NK cells derived
appears that the minor c - ~ ~ ~ + C D ~ + C D I ~ - PB
CD
NK
S ~ ~ "from
~ ~ ' CD34'HPCs in the presence of IL- l5 or IL-2 do
cell subset is intermediately differentiated between the
indeed express the 6-TCR protein in the absence of these
BM-derived population and the more abundant c-kiractivation antigens. Their expression does appear lower than
the CD2'CDS6h"Fh'N K cells in PB which either lack CD16
CD2'CD16'CDS6''i"' NK population.
The absence of CD2 antigen surface expression on the
or express it in low density. Nonetheless, this observation
BM-derived NK cells cultured in the presence of IL-15 is
lends support to the recent report that other NK cell molecules such as the target recognition structure p58 can associin striking contrast to virtually all CDS6' NK cells found in
ate with <-TCR,'X and may indeed be expressed onthe BMPB.' However, a CD2-CD1 6-CDS6h"Ch'N K cell of maternal
derived NK cells.
origin has been identified within the human decidua during
early pregnancy:
and the same cell can be foundwithin
A number of experimental animal models have demonfetal liver (L.L. Lanier, personal communication, December
strated that NK cells have a pivotal role in providing IFN1995). These data suggest that additional differentiation facy during infections and septic hock.'^.'" Interestingly, the
tor(s) are likely to be important prior to NK cell migration
current report provides thefirst
documentation that the
CDS6'LIN- N K cells derived from CD34' HPCs can prointo the circulation. It is possible that undefined thymic stroduce substantial amounts of IFN-y in response to two monomal cell factors required for high-density CD2 expression
cyte-derived cytokines, IL-l2 and IL-IS, quite comparable
on T cell^"^^'^ are also found in BM stroma and are required
tothat produced by mature PB N K cells.15 The abundant
for CD2 expression on PBNK cells.2yThe presence of a
production of IFN--y at this presumably early stage ofNK
CD2+CD16-CDS6' N K cell subset in PB suggests that acquisition of CD16 may occur in the peripheral circulation
cell differentiation, ie, in the absence of CD2 and CD16
during subsequent steps of NK cell maturation.
surface antigen expression, would lend support to this function being highly conserved and hence possibly morecritical
Expression of the 6-TCR protein has not previously been
during the early innate immune response to infection.
analyzed in CDS6'LIN-BM-derived
NK cells. CD2 and
CD16 are both N K cell activation antigens, and we have
In summary, this study provides the first in vitro evidence
previously shown that CD16 is biochemically and functionof a possible role for IL-IS in human NK cell development.
We identified the /L-15gene product in human BM stromal
cells that lack IL-2, and demonstrated that CDS6'LIN- N K
cells can be derived from CD34' HPCs in the presence of
Table 1. Cytokine Production by CD34' HPCs, CD56'LlN- BMIL-IS and the absence of IL-2. The synergy between IL-15
Derived NK Cells, and CD3-CD56"" PB NK Cells
and KL in expanding the CDS6' NK population isconsistent
Type of Cells
with IL-2R and c-kit expression on a circulating subset of
CD34' HPCs
CD56'LIN BM NK
CD56b''0PB
h' NK
Cytokine.
NK cells normally found in PB. Morphology, CD56 and 6TCR expression, cytotoxicity, and cytokine production of
TNF-a
46.6 2 16.1
269.3 -c 20.3
8 2 7.4
these cells are all most consistent with the N K cell lineage.
734 5 207
GM-CSF
11 f 1.8
284.4 5 67.3
28.1 2 8.5
3,520 2 760
5,650 -c 650
The existence of the CD2-CD16-CDS6h"Fh'phenotype in
IFN-y
certain fetal and adult organ tissues suggests this cell may
Reported in pglrnL,following 72-hour costirnulationwith 3 nglrnL
be fully differentiated in some instances. However, the acofrhulL-12and100nglrnLof
rhull-15. Cytokineproductionin the
quisition of CD2 and CD16 on the vast majority of PB NK
absence costirnulation was not detectable.
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
2639
ROLE OF IL-15 IN NK CELL DEVELOPMENT
cells suggests that additional, as of yet undefined factors are
required for the complete differentiation of PB NK cells
from BM precursor cells.
ACKNOWLEDGMENT
We thank Drs Elaine Thomas, Michael Widmer, and Douglas
Williams at Immunex for rhuKL and rhuIL-15, Dr Stanley Wolf at
Genetics Institute for rhuIL-12, Melanie Marien and Dr William
Carson for developing and performing the human E-15 ELISA, Dr
Robert Baiocchi and Matthew Strout for their assistance with the
RT-PCR assay, Adam Oberkircher for performing the Southern analysis, and Modi Wetzler for additional technical assistance. We thank
Drs Kenneth Grabstein, Meir Wetzler, Jerome Ritz, Lewis Lanier,
and Thomas Tomasi for helpful suggestions.
REFERENCES
1. Robertson MJ, Ritz J: Biology and clinical relevance of human
natural killer cells. Blood 76:2421, 1990
2. Lanier LL, Spits H, Phillips JH: The developmental relationship between NK cells and T cells. Immunol Today 13:392, 1992
3. Spits H, Lanier LL, Phillips JH: Development of human T and
natural killer cells. Blood 85:2654, 1995
4. Lotzova E, Savary CA, Champlin RE: Genesis of human oncolytic natural killer cells from primitive CD34+CD33- bone marrow
progenitors. J Immunol 1505263, 1993
5. Miller JS, Verfaillie C, McGlave P: The generation of human
natural killer cells from CD34+/DR- primitive progenitors in longterm bone marrow culture. Blood 80:2182, 1992
6. Silva MRG, Hoffman R, Srour EF, Ascensao JL: Generation
of human natural killer cells from immature progenitors does not
require marrow stromal cells. Blood 84:841, 1994
7. Shibuya A, Nagayoshi K, Nakamura K, Nakauchi H: Lymphokine requirement for the generation of natural killer cells from
CD34+ hematopoietic progenitor cells. Blood 85:3538, 1995
8. Smith KA: Interleukin-2: Inception, impact and implications.
Science 280:1169, 1988
9. Dorshkind K, Pollack SB, Bosma MJ, Phillips RA: Natural
killer (NK) cells are present in mice with severe combined immunodeficiency (scid). J Immunol 134:3798, 1985
10. Schorle H, Holtschke T, Hunig T, Schimpl A, Horak I: Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting. Nature 352:621, 1991
11. Kiindig TM, Schorle H, Bachmann MF, Hengarten H, Zinkernagel RM, Horak I: Immune responses in interleukin-2-deficient
mice. Science 262:1059, 1993
12. DiSanto JP, Muller W, Guy-Grand D, Fischer A, Rajewsky
K Lymphoid development in mice with a targeted deletion of the
interleukin 2 receptor y chain. Proc Natl Acad Sci USA 92:377,
1995
13. Noguchi M, Huafang Y, Rosenblatt HM, Filipovich AH, Adelstein S, Modi WS, McBride OW, Leonard WJ: Interleukin-2 receptor y chain mutation results in X-linked severe combined immunodeficiency in humans. Cell 73:147, 1993
14. Grabstein KH, Eisenman J, Shanebeck K, Rauch C, Srinivasan S, Fung V, Beers C, Richardson J, Schoenborn MA, Ahdieh M,
Johnson L, Alderson MR, Watson JD, Anderson DM, Gin JG: Cloning of a T cell growth factor that interacts with the p chain of the
interleukin-2 receptor. Science 264:965, 1994
15. Carson WE, Gin JG, Lindemann MJ, Linen ML, Ahdieh M,
Paxton R, Anderson D, Eisenman J, Grabstein K, Caligiuri MA:
Interleukin (IL)15 is a novel cytokine that activates human natural
killer cells via components of the IL-2 receptor. J Exp Med 1801395,
1994
16. Gin JG, Ahdieh M, Eisenman J, Shanebeck K, Grabstein K,
Kumaki S, Namen A, Park LS, Cosman D, Anderson D: Utilization
of the p and y chains of the IL-2 receptor by the novel cytokine
IL-15. EMBO J 13:2822, 1994
17. Anderson P, Blue ML, O’Brien C, Schlossman S F Monoclonal antibodies reactive with the T cell receptor chain: Production
and charakterization using a new method. J Immunol143:1899,1989
18. Caligiuri MA, Zmuidzinas A, Manley TJ, Levine H, Smith
K, Ritz J: Functional consequences of interleukin 2 receptor expression on resting human lymphocytes. Identification of a novel natural
killer cell subset with high affinity receptors. J Exp Med 171:1509,
1990
19. Wetzler M, Kurzrock R, Taylor K, Spitzer G, Kantarjian H,
Baiocchi G, Ku S, Gutterman JU, Talpaz M: Constitutive and induced expression of growth factors in normal and chronic phase
chronic myelogenous leukemia Ph’ bone marrow stroma. Cancer
Res 505801, 1990
20. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal Biochem 162:156, 1987
21. Oberkircher AR, Strout MP, Herzig GP, Fritz PD, Caligiuri
MA: Description of an efficient and highly informative method for
evaluation of hematopoietic chimerism following allogeneic bone
marrow transplantation. Bone Marrow Transplant 16:695, 1995
22. Carson WE, Ross ME, Baiocchi RA, Marien M, Baioni N,
Grabstein K, Caligiuri MA: Endogenous production of interleukin
15 by activated human monocytes is critical for optimal production
of interferon-y by natural killer cells in vitro. J Clin Invest 96:2578,
1995
23. Lanier LL, Le AM, Civin CI, LokenMR. Philips JH: The
relationship of CD16 (Leu-11) and Leu-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes. J Immunol 136:4480, 1986
24. Krajewski CA, MrrizekE, Caligiuri MA: Generation and
expansion of CD3-CD56+ non-MHC-restricted cytotoxic effectors
from human hematopoietic precursors; requirement for c-kit ligand
and interleukin-2. Proc Am SOCClin Oncol 13:294, 1994
25. Anderson P, Caligiuri MA, Ritz J, Schlossman S F CD3negative natural killer cells express TCR as part of a novel molecular complex. Nature 341:159, 1989
26. Anderson P, Caligiuri MA, O’Brien C, Manley T, Ritz 3,
Schlossman S F Fcy receptor type I11 (CD16) is included in the
NK receptor complex expressed by human natural killer cells. Proc
Natl Acad Sci USA 87:2274, 1990
27. Lanier LL, Yu G, Philips JH: CO-association of CD36 with
a receptor (CD16) for IgG Fc on human natural killer cells. Nature
342:803, 1989
28. Cluitmans FHM, Esendam BHJ, Landegent JE, Willemze R,
Falkenburg JHF: Constitutive in vivo cytokine and hematopoietic
growth factor gene expression in the bone marrow and peripheral
blood of healthy individuals. Blood 85:2038, 1995
29. Miller JS, Alley KA, McGlave P Differentiation of natural
killer (NK) cells from human primitive marrow progenitors in a
stroma-based long-term culture system: Identification of a CD34+7+
NK progenitor. Blood 83:2594, 1994
30. Toksoz D, Zsebo KM, Smith KA, Hu S, Brankow D, Suggs
SV, Martin FH, Wiliams DA: Support of human hematopoiesis in
long-term bone marrow cultures by murine stromal cells selectively
expressing the membrane-bound and secreted forms of the human
homolog of the steel gene product, stem cell factor. Proc Natl Acad
Sci USA 89:7350, 1992
31. Heinrich MC, Dooley DC, Freed AC, Band L, Hoatlin ME,
Keeble W, Peters ST, Silvey KV, Ey FS, Kabat D, Maziarz RT,
Bagby GC Jr: Constitutive expression of Steel factor gene by human
stromal cells. Blood 82:771, 1993
32. Shibuya A, Kojima H, Shibuya K, Nagayoshi K, Nagasawa
<
<
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
2640
T, Nakauchi H: Enrichment of interleukin-2-responsivenatural killer
progenitors in human bone marrow. Blood 81:1819, 1993
33. Matos ME, Schnier GS, Beecher MS, Ashman LK, Wiliams
DE, Caligiuri MA: Expression of a functional c-kit receptor on a
subset of natural killer cells. J Exp Med 178:1079, 1993
34. Carson WE, Haldar S, Baiocchi RA, Croce CM, Caligiuri
MA: The c-kit ligand suppresses apoptosis of human natural killer
cells through the upregulation of bcl-2. Proc Natl Acad Sci USA
91:7553, 1994
35. Nishikawa K, Saito S, Morii T, Hamada K,Ako H, Narita
N, Ichijo M, Kurahayashi M, Sugamura K: Accumulation of
CD16-CD56' natural killer cells with high affinity inerleukin 2
receptors in human early pregnancy decidua. Int Immunol 3:743,
1991
36. Haynes BF, Martin ME, Kay HH, Kurtzberg J: Early events
in human T cell ontogeny: Phenotypic characterization and lokaliza-
MROZEK, ANDERSON, AND CALIGIURI
tionof T cell precursors in early human fetal tissue. J Exp Med
168:1061, 1988
37. Tjonnfjord GE, Veihy OP, Steen R, Egeland T: T lymphocyte
differentiation invitro from adult human prethymic CD34' hone
marrow cells. J Exp Med 177:1531, 1993
38. Bottino C, Vitale M, Olcese L, Sivori S, Morelli L, Augugliaro R, Ciccone E, Moretta L, Moretta A: The human natural killer
cell receptor for major histocompatibility complex class I molecules.
Surface modulation of p58 molecules and their linkage to CD3 5
chain, FccRI y chain and the p56'" kinase. Eur J Immunol24:2527,
1994
39. Bancroft GJ: The role of natural killer cells in innate resistance to infection. Curr Opin lmmunol 4503, 1993
40. Heremans H, Dillen C, vanDamme J, Billiau A: Essential
role for natural killer cells in the lethal lipopolysaccharide-induced
Schwartzman-like reaction in mice. Eur J Immunol 24:l 155, 1994
From www.bloodjournal.org by guest on December 29, 2014. For personal use only.
1996 87: 2632-2640
Role of interleukin-15 in the development of human CD56+ natural
killer cells from CD34+ hematopoietic progenitor cells
E Mrozek, P Anderson and MA Caligiuri
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/7/2632.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.