| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,§
*
Department of Cell Biology, Neurobiology, and Anatomy and the Program in Immunology for Aging,
Stritch School of Medicine, and Departments of
Microbiology and Immunology and
§
Medicine, Loyola University Chicago Medical Center, Maywood, IL 60153; and
¶
Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells in humans and mice
(1, 2, 3, 4). Early during embryonic development, progenitor
cells from the yolk sac and fetal liver colonize the thymic anlage
where they undergo sequential developmental stages that can be
identified by the differential expression of specific cell surface
molecules (5, 6). In humans, intrathymic progenitor cells,
identified by the
CD34+CD38low phenotype,
have recently been shown to have bipotential differentiation ability
and, under the appropriate conditions, can differentiate into
either T cells or NK cells (7, 8, 9, 10). The human T/NK
bipotential progenitors have been further characterized as
CD1a-CD5-CD3-CD4-CD8-CD7+
(9). In mice, thymic T/NK precursors are
NK1.1+CD117+CD44+CD25-
(11); a large portion of these cells express CD122 and
IL-15R (12, 13). Recently, it has been demonstrated at
the single-cell level that early thymic progenitors indeed have
the potential to differentiate into either T or NK cells (14, 15). T cell development in the thymus generally can be followed
by the expression of CD3, CD4, and CD8 surface markers. Thus, the most
immature cells are
CD3-CD4-CD8-
(triple negative, NT3) and will
differentiate into
CD3-CD4+/-
CD8-,
CD3+/-CD4+CD8+
(double positive, DP) and
CD3+CD4+ or
CD3+CD8+ mature T cells
(16, 17). In humans, expression of CD1a by thymic
precursors indicates T lineage commitment (18, 19). Although the early thymic progenitors have the capacity to differentiate into either T or NK cells, the thymic microenvironment predominantly supports T lineage commitment and differentiation, such that only a very small percentage of NK cells is present in the thymus (10, 11, 13). The development of the majority of NK cells occurs outside the thymus, in the bone marrow (11, 20, 21). Furthermore, it is established that the development of NK cells is thymus-independent because NK cells are detectable in embryonic liver as early as 6 wk of gestation before formation of the thymic anlage (22), and NK cells are present in athymic nude mice (23, 24). Thus, although it is clear that the thymus contains NK precursors, it is not known why the vast majority of T/NK progenitor cells in the thymus commit to the T lineage and only a very small percentage of T/NK cells commit to NK differentiation and maturation within the thymic microenvironment (7, 25). Whether the thymic microenvironment, particularly thymic epithelial cells, regulate the development of NK cells within the thymus or influence the preferential T lineage commitment of the T/NK bipotential progenitors has not been investigated.
It has been established that IL-2 and particularly IL-15 can induce NK differentiation from bone marrow or thymic progenitors in both humans and mice (13, 26, 27, 28, 29). Furthermore, IL-7 and stem cell factor (SCF)3 can synergize with IL-2 and IL-15 to induce NK differentiation (26, 30). In this study, we tested whether primary cultured human thymic epithelial cells (TEC) or various molecular mass fractions of their culture supernatants affect cytokine-driven differentiation of CD3-CD56+ NK cells from total human thymocytes and CD34+-enriched TN human thymocytes (CD34+ TN).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following Abs were used in this study: anti-CD1a (OKT-6), anti-CD2 (35.1), anti-CD3 (OKT-3), anti-CD4 (OKT-4), and anti-CD8 (51.1) (American Type Culture Collection, Manassas, VA); anti-CD1a-CyChrome (CyChr; HI149), anti-CD56-PE (B159), anti-CD3-CyChr (UTCHT1), anti-CD34-APC (581), and anti-CD38-FITC (HIT2) (PharMingen, San Diego, CA); anti-NKR-P1A (DX12, HP3G10, 191B8), anti-p58.1 (HP3E4, EB6), anti-p58.2 (CHL, GL183), anti-CD94 (HP3B1, XA185, HP3D9) (Leukocyte Typing VI); and goat anti-IgG (H + L)-FITC (human serum adsorbed) (Life Technologies, Gaithersburg, MD). All Abs were titrated to obtain optimal concentrations for flow cytometric analyses.
Human TEC cultures
Normal human thymi were obtained as discard tissues through the Department of Cardiovascular Surgery and Pathology, Loyola University Medical Center from otherwise healthy children <1 year of age undergoing corrective cardiovascular procedures for congenital heart diseases. TEC cultures were initiated by an explant technique and propagated in supplemented medium as previously described (31). Contaminating thymic fibroblasts were removed by treatment with 0.02% EDTA, followed by anti-fibroblast mAb (1B10) treatment, and complement-mediated cell lysis (32). TEC cultures were monitored for contaminating macrophages and fibroblasts using Mo-1 and TE-7 mAb, respectively (American Type Culture Collection). TEC cultures used in the experiments were 85–93% positive for keratin as determined by reactivity to anti-keratin mAb NMF-116 (Dako, Carpinteria, CA).
Thymocyte cultures
Human thymocytes were gently teased from thymus tissue, isolated
by Ficoll-Hypaque centrifugation, and slowly frozen in RPMI 1640
(Mediatech, Herndon, VA) containing 20% FCS (Life Technologies), 7.5%
DMSO (Sigma, St. Louis), and 10 µg/ml gentamicin (Life Technologies).
Frozen thymocytes were thawed quickly and incubated with RPMI 1640
containing 30% FCS, 100 µg/ml DNase I (Sigma), and 10 µg/ml
gentamicin for 1 h at 37°C. Dead cells were removed by
Ficoll-Hypaque centrifugation. In some experiments, thymocytes were
used immediately after isolation from thymus tissue without freezing.
Thymocytes were cultured in RPMI 1640 supplemented with 10% FCS (Life
Technologies), 10% human pooled AB serum (Sigma), and 10 µg/ml
gentamicin. For coculturing with TEC, thymocytes were placed in insert
chambers (0.2 µm; Nunc Nalge, Naperville, IL) which were then placed
in six-well cluster dishes previously plated with TEC. Thymocytes at
3.0 x 106 cells were cultured with 5
x 105 TEC in a total volume of 3.0 ml/well. The
following cytokines were used: human rIL-2 (100 U/ml; Cetus,
Emeryville, CA), IL-7 (10 ng/ml), SCF (10 ng/ml; BioSource
International, Camarillo, CA), and IL-15 (10 ng/ml; R&D Systems,
Minneapolis, MN). Human TGF-3 was used at 5 ng/ml. The cells were
cultured with IL-2 or IL-15 alone, or with combinations of IL-2, IL-7,
and SCF or IL-15, IL-7, and SCF. Thymocytes were removed, washed,
adjusted to equal cell density, and cultured with newly plated TEC
every 5–7 days. Thymocytes were cultured up to 30 days and were
harvested at various time points and assayed for cell surface markers
by flow cytometry as described elsewhere (33).
Culture of CD34+-enriched TN thymocytes
CD34+-enriched TN thymocytes were isolated from freshly obtained thymocytes (not frozen) using the negative selection StemSep kit (Stem Cell Technology, Vancouver, Canada). Between 5 x 108 and 2 x 109 thymocytes were used for each enrichment separation. Freshly obtained human thymocytes were first incubated with anti-CD1a, anti-CD3 mAbs. The cells were then treated with goat anti-mouse Ab coupled to magnetic colloidal at a ratio of 80 particles/cell (BioMag; Perceptive BioSystems, Framingham, MA) for 30 min at 4°C. The CD1a+ and CD3+ cells were then removed using a strong magnetic plate. This step was repeated twice or was repeated until the cell suspension was clear of magnetic colloidal particles. The unbound cellular fraction containing predominantly the CD1a-CD3- cells was then collected and subjected to a second round of negative selection using the StemSep system. In the second step, the cells were treated with a mAb mixture containing Abs to CD2, CD3, CD14, CD16, CD19, CD24, CD56, CD66, and glycophorin A (StemSep) for 30 min on ice. A total of 2 µl of the Ab mixture was used for each 1 x 106 cells as suggested by the manufacturer. After incubation with magnetic colloid (60 µl/1.0 ml of cell suspension) for 30 min, the cells were loaded into a magnetic column and washed with 8 ml of medium according to the manufacturer’s protocol. The flow-through fraction containing lineage-negative cells enriched for CD34+ cells was collected. Purity of the isolated cell fraction was determined by flow cytometric analysis using CD34-APC, CD38-FITC, CD3-PerCP, CD56-PE, and CD1a-PE. The isolated cells (1 x 105 cells/well) were cultured in 96 U-bottom well culture dishes in RPMI 1640 plus 10% FCS and 10% human AB serum. The following combination of cytokines were included: IL-2 (100 U/ml) + IL-7 (10 ng/ml) + SCF (10 ng/ml) and IL-15 (10 ng/ml) + IL-7 + SCF. In the coculture system, TEC were plated onto transwells at 1.5 x 104cell/well. Flow cytometric analysis was performed at day 8 or 12. We also cultured CD34+-enriched TN thymocytes with cytokines in the presence of two molecular mass fractions (8,000–30,000 Da, >30,000 Da) isolated from TEC culture supernatants.
Fractionation of TEC culture supernatants
Culture supernatants from confluent cultures of human TEC were collected every 3 days. The pooled supernatants were first clarified by centrifugation and then filtered through a 0.2-µm filter. A total of 160 ml of culture supernatants was concentrated using Centricon Plus-80 centrifugal filter devices (Millipore, Bedford, MA). The supernatants were first filtered through a 30,000-Da cutoff and the filtrates were then filtered through a 8,000-Da cutoff filter. The molecular mass fractions were designated as >30,000 and 8,000–30,000 Da, respectively. The fractions were tested at various final concentrations.
NK cytotoxicity assay
Measurement of NK cell lytic activity for K562 and Daudi cells was performed as previously described (34). Briefly, 1 x 107 K562 and Daudi cells were labeled with 100 µCi of [51Cr]sodium chromate for 1 h at 37°C. Labeled target tumor cells were cultured with effector cells at various E:T ratios for 4 h. The supernatants were removed using a Skatron harvesting press (Skatron, Sterling, VA) and associated radioactivity determined with a 4/600 ME PLUS automated gamma counter. Maximum release of radioactivity was obtained by adding 0.05% Nonidet P-40 (Sigma). Lytic activity was expressed as percent cytotoxicity as calculated by the formula: percent cytotoxicity = [(experiment dpm) - (minimum dpm)]/[(maximum dpm) - (minimum DPM)] x 100. All experimental means were calculated from triplicate values.
NK-CFU colony-forming assay
NK colony assays were performed according to a method previously described by Ogawa and colleagues (35). Briefly, postnatal CD34+ TN thymocytes were cultured at 50 x 103 and 100 x 103cells in 35-mm vented lid suspension culture dishes (Nagel Nunc). The cells were suspended in 1.0 ml of Iscove’s medium (Mediatech) containing 1.2% methylcellulose (Stem Cell Technology), 10% FCS, 10% human AB serum, 10 U/ml penicillin-streptomycin, and 5 x 10-5 M 2-ME. The cells were cultured with a combination of IL-2 (100 U/ml), IL-7 (10 ng/ml), and SCF (10 ng/ml) or IL-15 (10 ng/ml), IL-7, and SCF. The cells were cultured for 20 days at 37°C in a humidified atmosphere with 5% CO2/95% air. The NK-CFU were identified by their large, diffuse morphology as previously described and confirmed by Ogawa and colleagues (35). The colonies were counted and the average number of colonies per duplicate well was determined. To confirm NK phenotype, single colonies were plugged and stained with Wright stain (New Comer Supply, Middleton, WI) to detect cytoplasmic granules.
Flow cytometric analysis
Cells were harvested and washed twice in PBS/2% BSA. Two hundred thousand cells (2 x 105) were stained and analyzed by flow cytometry using a FACScalibur (Becton Dickinson, San Jose, CA). For three-color analysis, cells were stained with each mAb for 30 min on ice and washed three times with PBS/2% BSA. If nonconjugated mAb were used, cells were first stained with these mAb followed by washing and incubation with goat anti-mouse Ig (H + L)-FITC (human serum adsorbed). Data analyses were performed with CellQuest (Becton Dickinson) or WinMDI version 2.8 (Scripps Research Institute, San Diego, CA). Appropriate isotype control Abs were used to set markers where >98% of the cells are negative.
RT-PCR
Cells were lysed in 4 M guanidine isothiocyanate and total RNA was isolated by centrifugation through a 5.7 M cesium chloride gradient as previously described (36). Alternatively, mRNA was purified using the QuickPrep Micro mRNA purification kit (Pharmacia, Piscataway, NJ). Synthesis of first-strand cDNA was performed with 3 µg of total RNA or 100 ng of mRNA using a first-strand synthesis kit (Pharmacia) according to the manufacturer’s instructions. The PCR amplification reactions contain 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 1.2 mM MgCl2, 0.125 mM dNTP, 0.25 µM primers, and 0.025 U/µl Taq (Life Technologies). For IL-7, two rounds of PCR amplification were performed: the first-round PCR was performed with 2 µl of cDNA samples for 25 cycles, and the second-round PCR was performed with 3 µl of the first-round PCR products for 30 cycles. IL-15 and GAPDH were performed with one round of 35 cycles of amplification. Each PCR cycle includes 15 s at 94°C, 20 s at 55°C, and 40 s at 72°C; the last cycle was followed by a final 7-min extension step at 72°C. PCR were performed with a GeneAmp PCR 2400 instrument (Perkin-Elmer/Cetus, Norwalk, CT). PCR products were separated in 2% agarose gel and visualized with ethidium bromide staining. Gel images were captured using a Kodak digital camera with one-dimensional image analysis software (version 2.1; Kodak, Rochester, NY). Primer sequences for human IL-7 and IL-15 were as follows: IL-7: forward, 5'-atg gcaaac aat atg aga gtg-3'; reverse, 5'-cca aat ttt att cca aca agt-3'; IL-15: forward, 5'-ttt cga aac cac att tga gaa-3'; reverse, 5'-cct cca gtt cct cac att ctt tg-3'.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because the T/NK bipotential precursors were found in the
human thymus (8, 10, 37), we determined whether the thymus
affects NK development in vivo. Using the CD1a as a marker for T
lineage commitment, we determined that >80% of total thymocytes
express CD1a (Fig. 1
B), while
only a small fraction of thymocytes are positive for the NK cell marker
CD56 (Fig. 1, C–F; Table I).
Further phenotypic analysis showed that the CD56+
thymocyte population is heterogeneous. We determined that 89–90% of
the CD56+ thymocytes also express CD1a and CD3
and only 10–11% are negative for T cell-specific markers CD1a and CD3
(Fig. 1, C–F). Thus, in postnatal human thymus, only 0.2%
(n = 7) of total thymocytes are the classical NK cells
that are
CD1a-CD3-CD56+
(Table I). Thus, the development of T/NK precursors is skewed toward
the T lineage in postnatal human thymus. The poor development of
intrathymic NK cells is not caused by the absence of cytokines, because
we showed by RT-PCR that the human TEC do express IL-15 mRNA, the
potent cytokine required for NK differentiation (Fig. 2).
|
|
|
Because T/NK progenitors are present in the human thymus, but only
a fraction of thymocytes are
CD1a-CD3-CD56+
NK cells, we wanted to determine whether TEC affect NK cell
development. Total thymocytes were cultured with a combination of IL-15
(10 ng/ml), IL-7 (10 ng/ml), and SCF (10 ng/ml) and were analyzed for
expression of CD56 and CD3 by two-color flow cytometric analysis. The
combination of the above cytokines has been shown to optimize the
induction of NK differentiation from thymic precursors in vitro
(8). Increased percentages of
CD3-CD56+ cells were
observed with increasing culture periods (Fig. 3). In addition to the increases in the
percentage of CD3-CD56+
cells, we also observed increases in the percentage of
CD3+CD56+ cells and a
concomitant decrease in the percentage of
CD3+CD56- cells; the
percentage of CD3-CD56-
cells remained unchanged (Fig. 3). In contrast, TEC, when cultured with
thymocytes in transwells, inhibited the in vitro development of
CD3-CD56+ (Fig. 3). The
decreases in the percentage of
CD3-CD56+ cells were
associated with increases in the percentage of
CD3+CD56- T cells and
CD3-CD56- cells (Fig. 3).
TEC had no effect on the development of
CD3+CD56+ cells (Fig. 3).
Similar results were obtained when thymocytes were cultured with IL-2,
IL-7, and SCF in the presence of TEC (data not shown). To confirm that
the inhibition of
CD3-CD56+ cell development
by TEC was not due to selective TEC-induced outgrowth of other subsets,
we determined the absolute number of
CD3-CD56+ cells. Fig. 4 shows results from six independent
experiments in which cells were treated with either IL-2, IL-7, and SCF
(left panels) or IL-15, IL-7, and SCF (right
panels) in the presence or absence of TEC. We demonstrated that
the number of NK cells increased with time when thymocytes were
cultured with mixtures of cytokines in the absence of TEC. In contrast,
a dramatic reduction of NK cell numbers was observed when TEC were
present in thymocyte cultures. We observed a 6- to12-fold reduction in
the number of CD3-CD56+
cells in thymocyte cultures treated with IL-2, IL-7 and SCF, and a 3-
to 25-fold reduction in cultures treated with IL-15, IL-7, and SCF
(Fig. 4). We also determined that the absolute number of cells in the
CD3+CD56- and
CD3+CD56+ subsets were not
affected by TEC; a moderate increase in the absolute number of
CD3-CD56- cells however
were observed (data not shown).
|
|
To differentiate between mature NK expansion and differentiation
of NK progenitor cells, we next determined whether TEC inhibit NK
development from the isolated CD34+ TN
progenitors. As previously shown by others (9, 10), we
determined that the human CD34+-enriched TN
thymocytes can differentiate into functional NK cells in our culture
system. The ex vivo-enriched progenitor population was negative for CD3
and CD56 (Fig. 5A). The
CD34+-enriched TN cells were positive for CD38
and CD7, but were negative for the early T cell marker CD1a (data not
shown), characteristic of the previously described T/NK progenitors
(9, 10). After 14 days in culture with IL-2, IL-7, and
SCF, 80% of the cells differentiated into CD1a-
CD3-CD56+ NK cells (Fig. 5, B and C). However, we did not detect
CD3+, TCR+, or
TCR
+ T cells (data not shown). Further
phenotypic analysis showed that >60% of CD1a-
CD3-CD56+ cells express
CD94 and the NK cell marker NKRP1A (CD161); however, the cells do not
express the killer inhibitory receptor receptors p58.1 (HP3E4, EB6) and
p58.2 (CHL, GL 183) (Fig. 5, D–I). Functionally, the
thymic-derived CD3-CD56+
NK cells display potent cytotoxicity for both K562 and Daudi
target cells (Fig. 6). Thus, we could
generate functionally mature NK cells in vitro from the human
CD34+-enriched TN thymocytes. Similar results
were obtained with IL-15, IL-7, and SCF (data not shown). This in vitro
culture system thus provides a functional model where regulation of NK
differentiation can be investigated. We next examined whether TEC
regulate NK differentiation from the
CD34+-enriched TN thymocytes. As observed with
total thymocyte cultures, TEC also inhibited the differentiation of
CD3-CD56+ cells from the
early thymic progenitor cells; while 85% of
CD3-CD56+ cells were
detected in the absence of TEC, only 42% of
CD56+ cells were detected in the presence of TEC
(Fig. 7, top three panels). In
a separate series of experiments, a 8,000- to 30,000-Da fraction
derived from TES inhibits NK differentiation from thymic progenitors
(Fig. 7, bottom four panels). Although we noticed that the
inhibition was not as completed as in the case of total thymocytes
(Figs. 3
and 4), the consistent results from several independent
experiments strongly indicate that TEC do regulate NK cell development.
IL-7 and SCF have been shown to synergize with IL-2 or IL-15 in the
induction of NK cell differentiation (26, 30). However,
IL-15 or IL-2 themselves can induce NK differentiation from thymic and
bone marrow precursors. Thus, we determined whether TEC directly
inhibit IL-15- or IL-2-driven NK cell differentiation. We showed that
TEC or the 8,000- to 30,000-Da fraction are potent inhibitors of IL-15-
or IL-2- driven NK cell development from the
CD34+ TN progenitors; the >30,000 Da showed no
inhibitory effect (Fig. 8).
Furthermore, the development of both the
CD3-CD56+ and
CD3-CD56Hi cells was
dramatically reduced by TEC or by the TEC-derived 8,000- to 30,000-Da
fraction (Fig. 8). Since TGF- has been shown to inhibit NK
development from murine bone marrow (BM) (38) and we have
shown that human TEC produce active TGF-3 (39), we
tested whether TGF-3 affects NK development from thymocytes. Fig. 9 shows that TGF-3 has no effect on
cytokine-driven NK development.
|
|
|
|
|
, TES served as a potent inhibitor of
NK-CFU when CD34+ TN thymocytes were cultured
with IL-2, IL-7, and SCF or IL-15, IL-7, and SCF; however, TES did not
affect GM-CFU formation. Thus, TEC specifically inhibit the development
of NK-CFU from thymic CD34+ TN progenitors.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells (1, 3, 40); our data
indicate that TEC also regulate NK cell development from the early
human thymic precursors.
Recently, it has been established at a single-cell level that a subset
of the TN thymocytes is bipotential and can differentiate into either T
or NK cells in vitro (14, 15). Despite this, a majority of
these cells develop into T cells, suggesting the thymic
microenvironment controls the outcome of this process.
Zú
iger-Pflüker and colleagues (11) have
demonstrated that exposure of T/NK progenitors to a thymic
microenvironment results in a predominant commitment to the T cell
lineage, whereas culture of bipotential cells with BM-derived-stroma
promotes the generation of NK cells. Plum and colleagues
(13) have suggested that the differential requirements for
stroma cell-precursor cell interaction and/or cytokines directs
differentiation toward either T or NK cells from T/NK progenitors. In
vivo studies with IL-15- and IL-15R-deficient mice indicate that
IL-15 plays an obligatory role for NK cell differentiation and
functional maturation; lack of NK cells in BM and thymus were observed
in these mice whereas the development of thymus-dependent T cells is
intact (41, 42). IL-2 has also been shown to induce NK
cell development in vitro; however, NK development proceeds normally in
IL-2-deficient mice, confirming the essential role of IL-15 in the
differentiation of NK cells in vivo (43). In addition to
IL-15 and IL-2, other cytokines such as IL-7, SCF, and Flt3 ligand can
synergize with IL-15 or IL-2 to induce NK cell development. However,
IL-7, SCF, and Flt3 ligand by themselves are not able to induce NK cell
differentiation, but mediate the IL-15 response by regulating IL-15R
expression (38, 44, 45). In contrast, IL-7 is critical for
TCR T cell development (46) and high concentrations
of IL-15 (500 ng/ml) have been shown to inhibit the development of
TCR T cells in murine fetal thymic organ culture
(13). Taken together, previous work identifies critical
cytokines for T cell and NK cell development and supports the notion
that differential requirement of cytokines affects the differentiation
of T/NK bipotential cells toward either T cell or NK cell lineages. The
in vitro data with FOTC also support the notion that differentiation of
NK cells in the thymus is regulated, and egression of mature NK cells
from the thymus is unlikely to account for the low percentage of NK in
the thymus. However, direct evidence showing the control of NK
development in the thymus is lacking.
The paucity of intrathymic NK cells may result from thymic microenvironment-mediated differential responses to cytokines present in the thymus. Thus, controlling IL-15-driven NK differentiation from thymic progenitors can regulate intrathymic NK development. It is unlikely that lack of IL-15 in the thymus is the cause for the low percentage of intrathymic NK cells because IL-15 is expressed in murine thymic stroma cells (13) and we showed here that human TEC express IL-15.
Although the CD3-CD1a-CD56+ NK cells comprise only 0.2% of total thymocytes in the thymus, dramatic increases in the percentage and the absolute number of CD3-CD56+ NK cells were obtained with thymocytes cultured with cytokines. Because we observed increases in both the percentage and absolute cell number of NK cells, the increases in NK cells were not due to selective depletion of other thymocyte subsets. Indeed, we have previously reported that TEC induce TN thymocyte proliferation (47). The inhibition was not caused by cell death of NK precursors or committed NK cells because, upon removal of TEC from the cocultures, we again detected increases in the percentage and absolute number of CD3-CD56+ cells after an additional 18 days in culture (data not shown). Furthermore, because we observed a reduction in the absolute number of NK cells, we can exclude the possibility that the inhibition was caused by the TEC-mediated selective proliferation of other T cell subsets. Although our data indicate that TEC regulate the outgrowth of pre-existing thymic cells with a NK phenotype, they do not exclude the possibility that differentiation of NK cells from thymic progenitors is also regulated by TEC.
To test whether TEC directly regulate NK differentiation, we determined
the effect of TEC on NK cell differentiation from human
CD34+CD1a- TN thymocytes.
The isolated CD34+ TN thymocytes are
CD38+ and CD7+ but are
CD1a-, characteristic of the previously reported
human T/NK thymic progenitors (8, 10); this isolated
population contains no
CD3-CD1a-CD56+
thymic NK cells. We showed that IL-15- and IL-2-driven differentiation
of NK cells was inhibited by TEC-derived soluble factors. The
inhibitory effect with the CD34+ TN thymocytes
was not as complete as in the case of total thymocytes; however, we
observed a consistent inhibition of NK differentiation in a series of
three independent experiments. Perhaps, the presence of the developing
thymocytes also contributes to the overall inhibition of NK cell
differentiation. We did not detect CD3+,
TCR+, and TCR+
T cells or CD83+ dendritic cells in our cultures.
This was not surprising because the culture conditions were not optimal
for differentiation of human T cells or dendritic cells. We confirmed
the inhibition of NK differentiation at the single-cell level by
showing that the formation of NK-CFU was inhibited by TEC-derived
soluble factors. We confirmed the nature of the colonies as NK-CFU
based on the presence of cytoplasmic dark blue granules visualized by
Wright stain (data not shown).
At the present time, we cannot rule out the possibility that TEC also regulate NK differentiation from other subsets present in the CD34+CD1a- TN population. Spits and colleagues (48) have shown that CD3-CD56+ cells can be generated from the CD34+CD1a+ precursors when the normal T cell maturation process is halted . Whether TEC can also regulate NK differentiation from BM hematopoietic stem cells remains to be determined.
The biological significance of controlling NK differentiation in the thymus is currently not known. It is possible that TEC, by inhibiting NK differentiation from T/NK bipotential progenitors, increase the pool size of precursors available for T cell differentiation and selection. Alternatively, a high number of NK cells in the thymus may alter TEC-mediated development of T cells. It is also possible that a high number of NK cells in the thymus may directly alter T cell development. It has been demonstrated that overpopulating the thymus with B cells alters thymic architecture and causes premature migration of immature thymocytes to peripheral organs (49).
Fractionation of TES showed that the inhibitory activity was
associated with the 8,000- to 30,000-Da fraction. This molecular mass
range encompasses most of the cytokines currently known to be expressed
by human TEC (33). Currently, a specific cytokine that can
inhibit IL-2- or IL-15-driven NK cell differentiation has not been
identified. Caligiuri and colleagues (38) have shown that
TNF- and TGF- inhibit NK cell development from BM precursor cells
in the presence of IL-15 and Flt-3 ligand . Our data indicate that
TGF-3, a predominant TGF- isoform in our TEC cultures
(39), does not inhibit cytokine-driven NK cell
development.
Recent emerging data at the molecular level have provided new insight
for understanding the control of NK cell development. Transcription
factors with a basic helix-loop-helix (b-HLH) motif are well
established as regulators of cellular differentiation
(50). Id, a subfamily of the b-HLH transcription family,
functions as a natural dominant negative regulator of the b-HLH
transcription factor activity (51). Mice deficient in Id2
fail to generate thymic-, spleen-, and BM-derived NK cells
(52). The NK cell developmental defect cannot be rescued
with IL-15, indicating that this is an intrinsic defect rather than
microenvironment and that expression of Id2 is required for IL15-driven
differentiation of NK cells (52). In contrast,
overexpression of Id2 in Id2-transgenic mice blocks early T cell
development (53). Similarly, forced expression of Id3,
while inhibiting TCR T cell development, promotes NK cell
development from the T/NK bipotential precursor cells
(54). From these findings, it is reasonable to speculate
that TEC regulate NK development by controlling expression and activity
of b-HLH transcription factors. Because b-HLH transcription activity is
also important for T cell development (53, 55, 56, 57),
regulation of their activities by TEC would affect both T and NK cell
development in the thymus, particularly at the T/NK bipotential
progenitor stage.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Phong T. Le, Department of Cell Biology, Neurobiology, and Anatomy, Loyola University Chicago, 2160 South First Avenue, Maywood, IL 60153.
3 Abbreviations used in this paper: SCF, stem cell factor; TEC, thymic epithelial cell; TES, thymic epithelial cell culture supernatant; BM, bone marrow; b-HLH, basic helix-loop-helix.
Received for publication September 14, 2000. Accepted for publication November 15, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
iga-Pflücker. 1997. Identification of a novel developmental stage marking lineage commitment of progenitor thymocytes. J. Exp. Med. 186:173. chain of the interleukin-2 receptor. Eur. J. Immunol. 23:3373.[Medline]
/ proteins. J. Exp. Med. 175:1055. regulate mRNA levels of interleukin 1 (IL-1), IL-1, and IL-6 in human thymic epithelial cells at a post-transcriptional level. J. Exp. Med. 174:1147.3 and express TGF- receptors. Int. Immunol. 7:1681. chain in natural T cell development. J. Exp. Med. 186:331. T cell development and early thymocyte maturation in IL-7-/- mice. J. Immunol. 157:2366.[Abstract]
but not of T cell development by overexpression of the helix-loop-helix protein Id3 in committed T cell progenitors. EMBO J. 18:2793.[Medline]
T-cell development and to rapid development of T-cell lymphomas. Mol. Cell. Biol. 17:4782.[Abstract]
thymopoiesis. J. Immunol. 163:3331.This article has been cited by other articles:
|
|
 |
|
  K. Aoyama, C. Delaney, B. Varnum-Finney, A. D. Kohn, R. T. Moon, and I. D. Bernstein The Interaction of the Wnt and Notch Pathways Modulates Natural Killer Versus T Cell Differentiation Stem Cells, October 1, 2007; 25(10): 2488 - 2497. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Without TEC; 
, with
TEC.
