| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
UMR CNRS 7627, Hopital Pitié-Salpêtrière,
* Laboratoire de Virologie, Faculté de Médecine Cochin-Paris V; and
ESA 7087 UP6-CNRS and Laboratoire d’Immunologie et Immunopathologie de l’ Ecole Pratique des Hautes Etudes, Paris, France
Correspondence: Ali H. Dalloul, M.D., Ph.D., CERVI, Hopital Pitié-Salpêtrière, 83 Blvd. de l’Hopital, 75013, Paris, France. E-mail: dalloul{at}ccr.jussieu.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
mRNA, which is
also found in CD34+ thymocytes and in blood
CD123hi DC further linking this subset to lymphoid DC.
However, the DC generated from CD34+ thymic progenitors
under standard conditions were pT-negative. Thymic lymphoid DC
showed similar phenotype and cytokine production profile as
blood/tonsillar lymphoid DC but responded to GM-CSF, and at variance
with them produced no or little type I interferon upon infection with
viruses and did not induce a strict polarization of naive T cells into
TH2 cells. Their function in the thymus remains therefore to be
elucidated.
Key Words: thymus • blood cells • tonsillar cells
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
transcripts within
thymic progenitors [9
]. Recently we have shown that the
capacity to generate DC is restricted to the most primitive,
CD1a-, thymic CD34+ progenitors
[10
]. DC are mostly generated through CD1a+
intermediates, whereas only a few are generated through
CD14+ intermediates upon addition of macrophage
colony-stimulating factor (M-CSF) to the cultures [10
].
This further supports the lymphoid commitment of human thymic
progenitors. By contrast, CD34+ cord blood cells give rise
to DC through two independent pathways: CD1a+ and
CD14+ precursor-derived DC [11
]. For the
above reasons it is tempting to link human thymic DC to a
"lymphoid" origin inasmuch as murine DC in the thymus belong
clearly to the CD8+ lymphoid lineage [12
,
13
]. All these findings are in keeping with the fact that
thymic DC are generated in situ from lymphoid-committed
progenitors. Therefore, thymic DC extracted from tissues should be
similar to CD1a+ precursor-derived DC generated in
vitro. The interest in classifying DC as lymphoid or myeloid has
been highlighted by the finding that they dictate the maturation of
naive T cells into TH1 or TH2 cells, respectively, in mice
[14
, 15
]. The picture is somewhat different
in humans in whom at least three DC populations have been described
in vivo. Among them, the monocytic population was reported
to produce IL-12 and to induce a TH1 response, whereas a plasmacytoid
subset was reported to induce a TH2 response [16
]. DC
populations express variable levels of receptors for GM-CSF (CD116) and
for IL-3 (CD123), which are maturation and/or survival factors for
these cells. However, contrary to mice, single markers typical of one
human DC subset are still lacking. Until now, only a few studies have been devoted specifically to thymic DC, and they relied on the purification of lineage-negative HLA-DR+ cells with typical dendritic morphology and capacity to induce allogeneic mixed leukocyte culture MLR [17 ]. Herein we attempted to answer the question as to whether thymic DC belong to one (possibly lymphoid?) or to several phenotypically and functionally different DC lineages. Lin- cells from light-density cell fractions were thus assessed for the expression of CD123 and HLA-DR. We observed that 20–30% of low-density Lin- cells consisted of a discrete CD123hi HLA-DR+ population, whereas typical DC located within HLA-DRhi cells. Both populations were sorted for further experiments; their phenotype, lymphokine production pattern, and the effect of the DC on naive T-helper cells were studied.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Purification and culture of DC and naive T cells
Four types of DC were used in our experiments: (1) purified DC
from fresh thymocyte suspensions; (2) DC cultured from
CD34+CD1a- thymocytes as previously reported
[10
]; (3) DC derived from peripheral blood monocytes
(MDDC), cultured as described [18
]; (4) peripheral DC
were obtained from normal blood donors with their informed consent.
Blood was centrifuged on Ficoll-Hypaque, mononuclear cells were
depleted of cells positive for lineage markers (CD3, CD14, CD19, CD56),
stained, and finally sorted as described below for thymic DC.
Thymic tissue obtained from pediatric cardiac surgery was processed as described [10 ] with slight modifications. Briefly, thymic fragments were incubated 45 min with 2 mg/mL collagenase P (Boehringer Mannheim, Mannheim, Germany) and DNase I (Sigma). Undigested thymocyte suspension was purified by centrifugation on Ficoll-Hypaque, the interface was washed and centrifuged at 4°C on 55% Percoll (Pharmacia, Uppsala, Sweden) at 3000 rpm. Preliminary experiments had shown that it was possible to enrich CD34+ cells 8- to 10-fold on the 55% Percoll fraction (data not shown). CD34+ cells were positively selected with CD34-coated magnetic beads (Dynal, Oslo, Norway), stained with CD1a mAb, and the CD1a- fraction sorted and cultured under conditions that generate DC as described [10 ]. CD34+ cell-depleted suspensions were then resuspended on 52% Percoll and the low-density fraction was depleted with CD8-coated magnetic beads, incubated with a mixture of CD3 (UCHT1), CD14 (MY4-biotin), CD19 (B4), and CD56 (NKH1) lyophilized mAbs, which was followed by depletion with beads (3 beads/cell) coated with sheep-anti mouse Ig. The resulting population contained about 20% DC, few stromal cells and, mostly, immature thymocytes (CD4+CD8-CD3- and CD4-CD8- double-negative cells). Cells were stained with anti-HLA-DR-FITC and CD123-PE or with a mixture of FITC-conjugated CD3, CD8, CD14, CD19, PE-conjugated CD123, and biotinylated anti-HLA-DR + streptavidin-APC mAbs. Cells were sorted with the FACStarplus equipped with an argon laser emitting at 488 nm and a Helium-neon laser emitting at 630 nm.
Monocytes were obtained from Ficoll-Hypaque-separated peripheral blood mononuclear cells (PBMC) of healthy volunteers, depleted of B and T lymphocytes using M-450 pan B/CD19 and M-450 pan T/CD2 Dynabeads (Dynal) at a 1:1 bead/cell ratio. The recovered population contained >80% CD14+ cells. MDDC were obtained by 5- to 7-day culture of monocytes with GM-CSF and IL-4 (Genzyme, Cambridge, MA) [18 ].
Naive T-helper cells were purified from cord blood obtained from the Obstetrics Unit of Hopital St. Vincent-de Paul, Paris. Briefly, cells were centrifuged on Ficoll-Hypaque, the interface was washed and incubated with a mixture of CD8, CD14, CD19, CD56, and CD45RO mAbs, and thereafter submitted to two rounds of anti-mouse Ig-coated magnetic beads as above. The resulting population consisted of 98% pure CD3+CD45RA+ cells.
Cell cultures and cytokines
Cells were cultured in RPMI-1640/10% fetal calf serum (FCS)
supplemented with 2 mM glutamine, penicillin/streptomycin (all from
GIBCO Life Technologies, Gaithersburg, MD). The cytokines used were as
follows: stem cell factor (SCF), 50 ng/mL; IL-7, 20 ng/mL (Valbiotech,
Paris); IL-3, 10 ng/mL; GM-CSF, 10 ng/mL; and IL-4, 200 IU/mL (Genzyme,
Cambridge, MA). Human soluble trimeric CD40-ligand (CD40LT, 500 ng/mL)
was a gift of E. K. Thomas (Immunex, Seattle, WA).
Giemsa staining
Cells in culture medium (105/mL) were spun for 4 min
at 400 rpm. Slides (5 x 104 cells/slide) were stained
with 1x May Grünwald (RAL, Bordeaux Technopolis, France) for 3
min, then in 0.5x Giemsa for 2 min, and finally in 0.1x Giemsa (RAL)
for 20 min, washed in water, and dried.
Proliferation assays
For the mixed lymphocyte reaction (MLR), variable numbers of
irradiated (22 Gy) DC were seeded in 96-well round-bottomed plates with
104 allogeneic T cells in 200 µL final volume. After 5
days of culture, 1 µCi of [3H]thymidine (Amersham,
specific activity 25 Ci/mmol) was added for 15 h to the wells of
triplicate tests. Results are expressed as mean counts per minute. For
DC, sorted cells were cultured for 3 days at 105/mL in 200
µL final with IL-3, GM-CSF, or both as indicated, and thymidine was
added as above. The proliferation of thymic DC in response to cytokines
is shown as incorporation index (ratio of cpm with the cytokine/control
cpm).
Cytokine assays
The production of the following cytokines by activated DC or by
T cells was assayed by using enzyme-linked immunosorbent assay (ELISA)
kits: interferon-
(IFN-), IL-4, IL-10, IL-12, IL-13 (Diaclone,
Biosystems), and IL-6 (R & D Systems, Abingdon, UK). Supernatants of DC
cultured at 200 x 103/mL were collected 48 h
after activation with CD40LT (500 ng/mL).
CD45RA+ cord blood T cells (106/mL) were cocultured with variable amounts of DC (1:2, 1:4, and 1:8 ratios) and restimulated for 24 h with a combination of 5 µg/mL CD28 (PharMingen) and CD3 (1/400 of a UCHT1 ascite) plastic-coated mAbs before supernatants were collected. T cells were also stimulated with mAbs alone or with MDDC. MDDC were irradiated and used as T cell stimulators at a 1:4 or 1:8 ratio. Each supernatant was assayed as pure and at 1/4 dilution in duplicate. Optical densities were read at 450 nm and plotted on a standard curve according to the manufacturer’s instructions.
Assay of type I IFN
A biological assay was performed as previously described
[19
]. Stocks of herpes simplex virus type I (HSV1) were
prepared from supernatants of infected Vero cells with a titer of
2 x 107 plaque-forming units (PFU)/mL. Stocks of
Sendai virus had 108 infectious doses/mL. PBMC from healthy
donors negative for hepatitis B and C viruses, and for HTLV-1 and HIV,
were used in induction experiments the same day. Freshly isolated PBMCs
or DC were incubated at 37°C with virus (HSV-1 at 4 x
105 PFU/mL; or Sendai 5 x 105 infectious
doses/mL); for 18 h in 0.5 mL of RPMI-1640/10% FCS. Supernatants
from positive controls or from DC were serially diluted twofold in
duplicate. Madin-Darby bovine kidney (MDBK) cells that respond poorly
to human IFN-ß but are very sensitive to IFN-, were incubated for
18 h with the supernatants in 96-well plates followed by infection
with VSV. Cytopathic effects were scored under the microscope 18 h
later. End-point titers represent dilutions that resulted in the
destruction of 50% of the cells. A laboratory reference of human
IFN- that had been standardized at the National Institutes of Health
(reference: Ga 023-902-530) was included in each titration experiment.
IFN titers are expressed as international units per milliliter. A
minimum IFN titer of 2 IU/mL was detectable in the assay.
Reverse transcriptase-polymerase chain reaction (RT-PCR) assays
RNA was isolated from sorted cells with the use of
RNA+ solution (Bioprobe Systems, Montreuil, France) and
reverse transcribed using a cDNA first-strand synthesis kit (Clontech,
Palo Alto, CA). cDNA was amplified using actin primers as positive
control, according to manufacturer’s conditions (Stratagene, La Jolla,
CA) or using pT primers as described [20
]. The
primers used are as follows: sense, 5’-GTCCAGCCCTACCCACAGGTGT;
antisense, 5’-CGGGAATTCGACGTCCCTGGCTGTAGAAGCCTCTC.
Briefly, samples were amplified in a 25-µL reaction using supermix PCR buffer (GIBCO) and submitted to 5 mm heating at 94°C, 5 mm at 55°C, 3 mm at 72°C followed by 34 cycles of 45 min at 94°C, 1 mm at 55°C, and 3 mm at 72°C. Ten microliters of each reaction were run on a 1% agarose gel.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
As shown in Figure 3 , Lin- HLA-DRhi cells are highly enriched in DC according to their E-cadherinhi CLA+ CD4lo/+ CD40hi, and CD80+ CD83+ CD86+ phenotype, which suggested that they were mature/activated cells. They were CD2-/+ CD7- CD5-/+ as expected from in vivo-separated DC, they express little or no CD1a. They expressed myeloid markers CD33 and CD44 and they were CD11b+ CD11clo/+ but CD36-. They were homogenously CD116+ and CD123-/+ with the CD123hi fraction accounting for 5% (ranging from 3 to 28%, depending on the thymus, mean 8.5 ± 7% on 12 different samples). Finally, all cells strongly expressed CD38. The phenotype of the lin-HLA-DR+ cells was next examined.
|
|
|
CD123hi cells were cultured for 2 days in IL-3 or GM-CSF or both, and no differences were observed depending on the cytokine used (data not shown). However, when CD40LT was added to cytokines in the culture, for 48 h, cells displayed upgraded HLA-DR expression while slightly downgrading CD123, and they became CD83+, CD80+, and CD86+ (Fig. 6 ).
|
mRNA transcripts has been reported in
CD123hi tonsilar DC [22
] and more recently
in thymic committed DC precursors [20
]. In our samples
(Fig. 7
), specific amplification of pT cDNA by RT-PCR shows that it is
expressed in CD34+ thymic progenitors, higher levels being
found in the CD1a+ fraction than in the CD1a-
fraction. Most importantly, pT is present in CD123hi
cells but not on mature thymic DC (HLA-DRhi) or on myeloid
(HLA-DRhi, CD11chi, Lin-) DC from
peripheral blood. By contrast we could not detect pT mRNA in mature
DC from either the thymus or the blood. We also looked for pT in DC
generated from thymic CD34+ progenitors, cultured as
described with a combination of SCF + IL-7 + GM-CSF +
tumor necrosis factor (TNF-), which proved to be the most
efficient in generating DC [10
] and took samples at
various times (not shown). Figure 7
shows that
HLA-DRhiCD1ahi DC sorted at day 21 did not
contain pT mRNA. The implications of these findings are discussed
below, however, the results clearly link immature thymic DC to the
lymphoid lineage.
|
and to a lesser extent to IFN-ß. PBMC infected with either HSV-1 or
Sendai produced significant amounts of IFN- (Table 1
). This production was found despite the fact that only a minor
subset of PBMC is able to produce IFN, which suggests that it should be
very potent in this respect. In contrast, we clearly failed to detect
any IFN activity in supernatants from various thymic DC
(Lin- fraction, sorted CD123hi, or
HLA-DRhi) after viral infection (Table 1)
. Neither
supernatants from control MDDC or from thymus
CD34+-progenitor-derived DC contained IFN (Table 1)
. This
negative result shows that although immature thymic DC and plasmacytoid
pre DC are both CD123hi, they are functionally distinct.
|
|
We next examined the nature of the primary T cell response induced by
thymic DC and control blood DC on naive T cells. Purified
CD4+CD45RA+ umbilical cord blood T cells were
cocultured for 6 days with variable amounts of CD40LT-activated thymic
DC (T cells/DC: 2, 4, and 8). Cultured T cells were then restimulated
for 24 h on plates precoated with anti-CD3 and anti-CD28 mAbs. T
cells stimulated with antibodies alone produce low amounts of IFN-
and IL-10. T cells stimulated by thymic DC produced twice as much IL-10
(range 337 to 1425 pg/mL) than IFN- (range 92 to 918 ng/mL), no
detectable IL-4, but in two experiments out of three, greater than 100
pg/mL of IL-13 were detected (Table 2
). When we compared the effect of mature and immature thymic DC on
naive T cells, no striking difference for the induction of IL-10, IL-4,
and IL-13 was observed; however, twice as much IFN- was produced by
T cells activated with HLA-DRhi cells than with those
activated with CD123hi DC (0.1 < P <
0.05).
|
and IL-10 in the supernatants. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
transcripts strongly suggest that they originate
from lymphoid-committed CD34+ progenitors. By contrast,
mature DC seem mostly myeloid owing to the expression of CD11c and CD33
and the lack of CD2, CD5, and CD7. Both populations show weak
expression of CD11b as expected from fresh DC at variance with blood
monocytes [26
]. Recently, several markers have been
reported that may help discriminate subsets of human DC
[26
, 27
]. These include receptors for
growth/differentiation factors of DC such as IL-3 and GM-CSF
[28
29
30
], molecules involved in adhesion to epithelial
cells such as E-cadherin and CLA [26
], and molecules of
unknown function such as CD83 [31
]. It was therefore
necessary to reassess the status of thymic DC in light of such new
markers. Until now DC have been characterized as
HLA-DR+Lin- [17
,
21
, 25
]. The authors took advantage of CD2
expression on all thymocytes, including the earliest CD34+
progenitors [7
] to deplete CD2+ cells
[17
, 25
]. Our results point to the fact
that this method may eliminate some 50% of DC, especially among the
immature subset. It is also not conceivable from our data to use CD7,
and to a lesser extent, CD5 mAbs to enrich thymocytes into DC. In
keeping with others [26
], we found no CD8 on DC; this
feature of human DC is therefore of help to eliminate most thymocytes
in combination with CD3 mAb. Finally, focusing only on the brightest
HLA-DR+ thymocytes would miss the CD123hi
subset that express intermediate levels of HLA-DR.
To which lineage do thymic DC belong? The unique solid evidence until
now that CD123hi DC belong to the lymphoid lineage is their
expression of pT mRNA at variance with mature DC. However, this does
not imply that all thymic mature DC are myeloid because it was shown
that activated DC lose pT upon maturation in vitro
[20
]. Thus mature thymic DC may be a mixture of lymphoid
and myeloid DC, the latter originating from in
situ-differentiating monocytes or from DC migrating from the
periphery. Our failure to obtain pT+ cells in
vitro from CD34+ thymic progenitors in conditions that
generated CD1a+-derived DC [10
] probably
reflects inappropriate in vitro-differentiating conditions.
Alternatively, the earliest steps of DC maturation might have been
bypassed in vitro so that pT is lost before it could be
detected, although it could not be detected in CD123+ cells
sorted as early as day 4 (not shown).
The unique expression of CD123, CD36, and other antigens makes this population distinct from MDDC and similar to the plasmacytoid/monocytoid DC already described in the periphery [16 , 23 , 28 ], albeit with a few differences. CD116 expression was reported in one article to be lower on tonsillar DC [16 ] than on our thymic DC, and GM-CSF failed to induce the maturation of blood lymphoid DC [23 ]. This does not contradict our results because thymic lymphoid DC needed GM-CSF + CD40 LT to undergo maturation. Both thymic DC had the same level of CD116 but only the immature subset proliferated slightly to GM-CSF. Overall, thymic lymphoid DC did not depend on IL-3 for survival, contrary to their peripheral counterpart. Together with the lack of production of IFN, this may reflect the influence of distinct microenvironments on these cells.
Although lymphoid DC have important functions in peripheral lymphoid organs [23 , 24 ], their role in inducing central tolerance within the thymus could not be inferred from their phenotype and their cytokine production profile. The lack of IL-10 production and the expression of E-cadherin would link thymic lymphoid DC to the CD1a+ differentiation pathway [32 ]. Of note, although myeloid DC in mice are DC2 cells inducing a TH2 response [15 ], the opposite was reported in humans in whom monocytic DC induce a TH1 response [16 ]. By contrast the only described human DC2 lineage is the plasmacytoid one, which induced a strict TH2 response in one report [16 ] but not in another [23 ]. Our thymic population was clearly unable to induce the polarization of naive cord blood T helper cells. In this respect, thymic DC either generated in vitro or extracted from tissue would have intermediate properties between DC1 and DC2. Therefore cytokine production may not be relevant with the deletion of self-reactive thymocytes. Finally, the relevance of IL-3 and GM-CSF on DC survival is suggested from their production in situ by thymic epithelial cells [33 , 34 ]. Together with CD40L+ thymocytes, these cytokines should induce the terminal maturation of DC into cells expressing costimulatory receptors and high amounts of HLA-DR in order to perform closer interactions with thymocytes [35 ] and ultimately lead to apoptosis of self-reactive cells [2 ]. This is supported by the fact that we could only detect apoptotic thymocytes in association with mature DC but not immature ones. It is also likely that DC-thymocyte interactions lead to the apoptosis of DC as observed in lymph nodes [36 ]. DC homeostasis would then be maintained through a continuous maturation from CD34+ progenitors in the thymus.
|
ACKNOWLEDGEMENTS |
|---|
|
FOOTNOTES |
|---|
Received February 28, 2000; revised July 2, 2000; accepted July 5, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
rearrangement and partial (DJ) recombination of of the TCR ß locus in CD34+CD7+ precursors from cord blood: evidence for T lineage committment J. Immunol. 157,4120-4127
-TCRß) gene expression during human thymic development J. Exp. Med. 184,519-527 J. Exp. Med. 184,695-706- dendritic cells but not of lymphoid-related CD8+ dendritic cells Immunity 9,839-847[Medline]
+ and CD8- subclasses of dendritic cells direct the development of distinct T helper cells in vivo J. Exp. Med. 189,587-592 by glycoprotein D of herpes simplex virus: a possible role of chemokine receptors Virology 251,317-326[Medline]
mRNA in a commmitted dendritic cell precursor in the human thymus Blood 94,2647-2657 cooperate in the generation of dendritic langerhans cells Nature 360,258-261[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Varas, C. Hernandez-Lopez, J. Valencia, S. Mattavelli, V. G. Martinez, L. Hidalgo, C. Gutierrez-Frias, A. G. Zapata, R. Sacedon, and A. Vicente Survival and function of human thymic dendritic cells are dependent on autocrine Hedgehog signaling J. Leukoc. Biol., June 1, 2008; 83(6): 1476 - 1483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Koelle, J. Huang, M. T. Hensel, and C. L. McClurkan Innate Immune Responses to Herpes Simplex Virus Type 2 Influence Skin Homing Molecule Expression by Memory CD4+ Lymphocytes J. Virol., March 15, 2006; 80(6): 2863 - 2872. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. O'Keeffe, H. Hochrein, D. Vremec, B. Scott, P. Hertzog, L. Tatarczuch, and K. Shortman Dendritic cell precursor populations of mouse blood: identification of the murine homologues of human blood plasmacytoid pre-DC2 and CD11c+ DC1 precursors Blood, February 15, 2003; 101(4): 1453 - 1459. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Comeau, A.-R. Van der Vuurst de Vries, C. R. Maliszewski, and L. Galibert CD123bright Plasmacytoid Predendritic Cells: Progenitors Undergoing Cell Fate Conversion? J. Immunol., July 1, 2002; 169(1): 75 - 83. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Weijer, C. H. Uittenbogaart, A. Voordouw, F. Couwenberg, J. Seppen, B. Blom, F. A. Vyth-Dreese, and H. Spits Intrathymic and extrathymic development of human plasmacytoid dendritic cell precursors in vivo Blood, April 15, 2002; 99(8): 2752 - 2759. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
