|
|
||||||||
,
,§
*
Department of Pathology,
Geriatrics Center, and
Institute of Gerontology, University of Michigan, Ann Arbor, MI 48109; and
§
Ann Arbor DVA Medical Center, Ann Arbor, MI 48109
| Abstract |
|---|
|
|
|---|
, and increased production of IL-4, IL-5, and IL-10.
This culture protocol provides a test system for exploration of factors
that regulate the conversion of naïve cells to memory cells and
the development of specific immune responses to protein Ags. The data
are consistent with models that implicate glucocorticoids as regulators
of immune response specificity. | Introduction |
|---|
|
|
|---|
Previous efforts to develop culture methods for primary sensitization in vitro have examined the effects of manipulations of cell concentration and interactions between T cells and accessory cells, but have not considered the potential effects of other physiologic agents, such as endocrine hormones, that are ordinarily produced during the course of an immune reaction and that could play a role in regulating its course. Increased glucocorticoid blood levels have been observed to occur at or near the time of peak immune responses to several Ags. This phenomenon was first described many years ago (5, 6) and was confirmed in recent publications (7, 8, 9). The association between immune responsiveness and increased glucocorticoid levels suggested that these hormones might modulate the strength of the response and in particular might contribute to its specificity (5, 10). In vitro studies using polyclonal activators have also suggested a role for corticosteroids in immune regulation, and have shown that these agents, which are well known for their immunosuppressive activity, can paradoxically promote Th2 cytokine responses by CD4+ T cells in vitro (11, 12).
Based on these earlier observations, we have conducted a series of experiments to see whether appropriately timed exposure to glucocorticoids can promote Ag-specific sensitization of naïve T cells in culture. We describe an experimental system in which CD4+ T cells are first exposed to an Ag (typically pigeon cytochrome c, PCC3) in serum-free culture, grown in IL-2-containing medium to permit clonal expansion, and then tested for Ag-dependent proliferation and cytokine production. The presence of dexamethasone (Dex) and Abs to IL-10 at the expansion stage following primary antigenic stimulation appeared to be critical for the ability of the stimulated and expanded T cells to respond to restimulation with the same Ag. The appearance of a strong, stable response of T lymphocytes to restimulation makes it possible to follow the maturation of the response after a series of restimulations with the same Ag. We report that this series of culture steps permits the routine development, from unprimed T cells, of Ag-specific cell lines that produce IL-4, IL-5, and IL-10.
| Materials and Methods |
|---|
|
|
|---|
Specific pathogen-free (BALB/cJ x C57BL/6)F1 male mice were purchased from The Jackson Laboratories (Bar Harbor, ME) and housed in a specific pathogen-free facility. Sentinel mice from this colony were checked quarterly for serologic evidence of viral infection and for parasites; all such examinations were negative throughout the course of these studies. Mice were used in experiments when 8 to 12 wk old.
Reagents
Ascites fluid was generated from hybridoma cells secreting rat IgG2a Ab to mouse CD8 (clone 53-6.72) and mouse IgM anti-mouse Thy-1.2 (clone HO-13-4) obtained from the American Type Culture Collection (Rockville, MD). FITC anti-CD4 was obtained from PharMingen (San Diego, CA). Anti-IL-10 Abs were from R&D Systems (Minneapolis, MN). Keyhole limpet hemocyanin (KLH) was purchased from Calbiochem (La Jolla, CA). PCC, mitomycin C, polyL-lysine (PLL; m.w. = 80,000), and Dex were obtained from Sigma Chemical Co. (St. Louis, MO).
Culture medium
RPMI 1640 supplemented with 10% FCS, penicillin G (100 U/ml), streptomycin (100 µg/ml), L-glutamine (2 mM), and 2-ME (5 x 10-5 M) was used for all cultures except the cultures for primary Ag stimulation, where serum-free culture medium was used to prevent interfering presentation of any serum protein Ags.
Cell preparation
Splenic single cell suspensions were isolated and mononuclear cells were enriched by centrifugation over a cushion of Lympholyte-M (Cedarlane, Hornby, Canada). T cells were then enriched by depletion of surface Ig-positive cells on petri dishes coated with anti-mouse IgM. This procedure typically yielded more than 90% pure T cells (i.e., CD3+) with fewer than 5% contaminating B cells. Collected cells were incubated on ice with rat anti-mouse CD8 mAbs (1:200) either 1) alone, or 2) together with anti-CD44 (1:200) followed by incubation with goat anti-rat IgG-coated magnetic beads (PerSeptive Diagnostics, Cambridge, MA) on ice. The bead-adherent cells were then removed using a magnetic separator. The purity of the CD44low CD4+ T (CD4 virgin) cells was 80 to 90%, when analyzed by flow cytometry.
T cell-depleted resident peritoneal cells were prepared by treating mouse peritoneal cells with anti-Thy-1.2 Abs and complement as previously described (13), except that mitomycin C (50 µg/ml added to 5 x 107 cells/ml for 30 min at 37°C) was added along with the complement treatment to diminish proliferative activity. They were used as APC mixed with a 20-fold excess of similarly treated spleen cells.
Preparation of APC monolayers for T cell stimulation
First, culture plates were coated with PLL (50 µg/ml in PBS), using 1 ml for 6-well plates (no. 3506; Costar, Cambridge, MA), or 0.2 ml for 24-well plates (no. 3524; Costar). After 60 min, plates were washed three times in PBS to remove unbound PLL. Next, APCs were resuspended at 7 x 106 cells/ml in serum-free medium, and 2.5 ml of this suspension were added to wells of 6-well plates (Costar) or 0.5 ml to wells of 24-well plates. Ag (PCC or KLH, 100 µg/ml) was added at this step if indicated by the protocol. Plates were incubated on a stationary platform for 90 min in a CO2 incubator. The wells were washed to remove unattached cells and free Ag by flushing the bottom three times with 37°C serum-free culture medium. Free Ag was removed together with nonadherent cells because of published reports (14) and our own experiments showing that the presence of free Ag in culture decreases stimulation efficiency. These APC monolayers were then covered with fresh culture medium (serum free in case of cultures for the primary antigenic stimulation) and were either used immediately (primary antigenic stimulation) or used the following day (as supporting cells for the primed T lymphocyte expansion).
Stimulation and expansion of unprimed T cells
Our method is based on the previous work of several authors (12, 15, 16), and has been described elsewhere in more detail (17). In
brief, T cells from unprimed mice are stimulated with Ag in serum-free
medium; expanded in the presence of IL-2 plus Dex and Abs to IL-10
(
IL-10); and then tested for proliferation and cytokine production
induced by the original Ag (typically PCC), an alternate nominal Ag
(typically KLH), or no Ag. In the stimulation phase, responding
CD4+ T cells (68 x 106 cells in 4
ml) were cultivated for 24 h on Ag-bearing APC monolayers in
serum-free medium, and then transferred to new culture wells containing
fresh (unpulsed) APC. In preliminary experiments (not shown) we noted
that the yield of cells at the end of the expansion phase was decreased
threefold when this transfer to fresh APCs was omitted. In the
expansion phase, the cells were cultured for an additional 4 to 5 days,
in 6- or 24-well plates, in the presence of rIL-2 (20 U/ml) with Dex
and
IL-10. Typically, each 4-ml initiation culture yielded
6 to
9 x 106 viable cells, which were divided into three
expansion cultures in 6-well plates, or into 15 expansion cultures in
24-well plates. Concentrations of Dex and
IL-10 were, respectively,
10-7 M and 4 µg/ml unless otherwise indicated in the
text. In the testing phase, expanded cells were harvested; depleted of
dead cells using a Lympholyte step gradient; and then recultured with
new APC monolayers bearing either the original Ag, a distinct Ag, or no
Ag. In most experiments these test cultures used 24-well plates
containing 3 x 106 APC pulsed with 100 µg/ml of Ag
and 3.5 x 105 harvested T cells. In experiments
designed to test a range of Ag concentrations, we used 96-well plates
containing 5 x 105 APC/well and 5 x
104 harvested T cells. Proliferation was assessed in
duplicate using 200-µl aliquots in 96-well plates on day 4 by a 12-h
pulse of [3H]thymidine. For tests of cytokine production,
50-µl aliquots of medium were harvested at intervals (13 days)
indicated in the text. IL-2 was measured by bioassay using CTLL
indicator cells, and expressed as U/ml using a standard curve. IFN-
,
IL-4, IL-5, and IL-10 were assayed using two-site ELISA assays using
standard curves with known amounts of recombinant cytokines.
Anti-IFN-
, anti-IL-4, anti-IL-5, and anti IL-10 Abs
(PharMingen) were used as capture reagents, and biotinylated Abs
(PharMingen) as the detecting Abs. The biotinylated Abs were then
detected using peroxidase-conjugated streptavidin and
O-phenylenediamine (Zymed Laboratories, San Francisco, CA).
The recombinant standards were IFN-
, IL-4, and IL-2 from Genzyme;
IL-5 (PharMingen); and IL-10 (R&D Systems).
Multiple restimulations
To assess the effect of multiple stimulation events on T cell differentiation and maturation of the response, expanded cells were in some experiments cultivated on fresh Ag-pulsed monolayers and then transferred every 2 days to new cultures with fresh, pulsed APCs. After every cycle, a portion of the cells were removed for functional analysis by proliferation and/or cytokine production. Cells to be tested were transferred into three new culture wells containing, respectively, APCs with the original Ag, a distinct Ag, or no Ag. Supernatants from cultures were taken after 24 and 48 h, frozen, and then later tested for cytokine content.
| Results |
|---|
|
|
|---|
We have previously described (17) a culture system in which T
cells from unprimed mice are cultivated together with a foreign Ag
(typically PCC), IL-2, Dex, and
IL-10, and shown that with
appropriate concentrations and timed additions of these components, the
cultures develop Ag-specific responses to the Ag used for in vitro
priming, measured either as proliferation or as production of IL-2.
Figure 1
shows the results of an
experiment of this kind, in which preparations enriched for naive CD4
(i.e., CD8-, CD44low, Ig-) T
cells were used as responders. The combination of an initial 24-h
encounter with PCC in serum-free medium, together with the inclusion of
Dex and
IL-10 during a period of IL-2-driven clonal expansion, leads
to a dramatic increase in the ability of the cultured cells to
proliferate and secrete IL-2 even in the absence of restimulation with
PCC, although addition of PCC to the test cultures leads to a
significant increase in response level above that obtained in cultures
free of nominal Ag. The response after this single cycle of Ag exposure
and expansion is not Ag specific, in that similar levels of response
are seen in the test cultures whether PCC is omitted (as in the
leftmost bars of Fig. 1
) or replaced by equal concentrations
of KLH (not shown). Similar results can be obtained in cultures
initially stimulated by KLH (17) or by OVA (not shown). Although
typical experiments used preparations in which 80 to 90% of the
responder cells were CD4+ and CD44low, similar
results were obtained in a few experiments in which the responder cell
purity exceeded 98%. Nonetheless, further work would be needed to
determine whether cells with the CD44high phenotype
contribute to these responses of cells from unprimed mice.
|
IL-10,
and IL-2 for 4 days; and then tested for IL-2 production by culture for
up to 3 days with PCC at 20 µg/ml. The data show that inclusion of
PCC in the initial 24-h culture does promote responsiveness of the
cells expanded in Dex/
IL-10.
|
IL-10 to be used in the
cell expansion phase. In the absence of Dex,
IL-10 had no effect on
proliferation (top) or IL-2 production
(bottom) after cell expansion. In the absence
of
IL-10 (open symbols), Dex led to a small but significant increase
in PCC-stimulated proliferation and a modest increase in IL-2
production by expanded cells. For Dex doses between 10-8 M
and 10-7 M, the addition of
IL-10 led to synergistic
increases in both proliferation and IL-2 production. The optimal Dex
concentration varied somewhat from experiment to experiment, but was
always between 10-8 M and 10-7 M. The
IL-10 titration curve was also biphasic, with optimal Ab
concentrations typically at 4 µg/ml.
|
IL-10 on the "per cell"
responsiveness. The middle and bottom
panels of Figure 4
IL-10 at 4 µg/ml. It is clear that Dex doses up to at
least 10-7 M lead to progressive increases in responses
among those few cells that are recovered at the end of the 4 days of
expansion. Cell recovery at Dex concentrations higher than
10-7 M was typically too low to permit detailed study of
function (not shown).
|
We conducted a series of experiments to determine whether repeated
cycles of Ag exposure would alter the specificity and/or quality of the
response obtained. In these experiments, naive CD4 cells from unprimed
mice were cultured for 24 h in the presence of PCC-bearing APC,
then expanded in the presence of IL-2, Dex (10-7 M),
and
IL-10 (4 µg/ml). Aliquots of these cultures (cycle 1) were
tested for production of IL-2, IFN-
, IL-4, IL-5, and IL-10 in
responses to PCC, to the unrelated Ag KLH, or in medium that contained
no nominal Ag (but did contain 10% FBS). Other aliquots from the same
cultures were restimulated by re-exposure to fresh PCC-pulsed
monolayers in the absence of added IL-2, Dex, or
IL-10. After each
48-h interval, these cells were harvested (cycles 2, 3, etc.) and
divided into aliquots for functional testing or for further culture on
PCC-pulsed APC.
Figure 5
shows that these cycles of
repeated Ag exposure led to consistent alterations in the pattern of
cytokines produced in the PCC-stimulated test cultures. Production of
IL-2 and IFN-
declined greatly by the third and later cycles, while
production of IL-4, IL-5, and IL-10 increased in parallel.
|
IL-10.
|
| Discussion |
|---|
|
|
|---|
Several factors may contribute to the absence of Ag-specific T cell
response to new Ags under ordinary culture conditions. Virgin T cells
may well have special costimulatory requirements (23, 24, 25).
Costimulation through the CD45 Ag, for example, has been suggested to
lead to an increase in Ag-specific Th cell frequencies (26). Inhibitory
interactions between different types of T cells may also interfere with
activation of naïve T cells in primary cultures. Previously,
using a limiting dilution approach (27), we have shown that IL-10
production by memory T cells can abrogate proliferation by
naïve cells in shared cultures stimulated by plant mitogens,
and that this inhibition could be blocked by Ab to IL-10. We speculated
that T memory cells activated either cross-reactively or through
bystander effects might be able to inhibit activation of virgin T cells
in response to Ag though an IL-10-dependent mechanism, and thus
examined in the current program the effect of including anti-IL-10
Abs during the interval immediately following the initial exposure to
Ag. The speculation (5) that glucocorticoids produced during the course
of in vivo immune reactions might improve specificity by impeding the
growth of T cells with poor Ag specificity prompted us to examine the
role of Dex in the primary cultures. Our results show clearly (Figs. 1
and 3
; and 17 that optimal proliferation and IL-2 production by
expanded cells requires both Dex and
IL-10 during the expansion
phase. The dose-response curves for both Dex and
IL-10 provide only
a narrow window for positive effects. High doses of Dex, in particular,
diminish cell recovery while continuing to increase response per
recovered cell (Figs. 3
and 4
).
Survival and clonal expansion at successive phases of the sensitization
process may be modulated in part by changes in susceptibility to
apoptosis. Although virgin T cells appear to be very sensitive to the
inhibitory effects of Dex during the Ag-priming phase, they become
increasingly resistant to corticosteroid inhibition once they have
entered the proliferative phase (28, 29). Although both TCR
hyperstimulation and glucocorticoids can induce cell death
individually, previous work has shown that TCR-mediated programmed cell
death can, paradoxically, be inhibited by glucocorticoids (30, 31).
Removal of memory T cells from human PBMC makes the residual population
more resistant to Dex (32). One effect of Dex that might interfere with
expansion of Ag-stimulated naïve cells, i.e., its ability to
block IL-2 production (33), was mitigated in our culture system by the
addition of exogenous IL-2. The ability of IL-2 to rescue Ag-specific T
cells from apoptosis may also contribute to positive selection during
the expansion phase, because T cells that do not get antigenic
stimulation cannot be rescued by IL-2 from apoptosis (34). The use of
Dex and
IL-10 during the interval immediately following Ag exposure
permits the rapid expansion of the Ag-primed cells, but does not by
itself achieve antigenic specificity. Normal in vivo immune responses
are also often accompanied by bystander proliferation of cells that
have been activated nonspecifically (35). The maturation of
naïve CD4 cells into effector cells able to generate Th2
cytokines has in other laboratories also depended upon addition of
growth factors (16, 36) or corticosteroids (12, 37), or on repetitive
in vitro restimulation (38). In our protocol, the combination of
Dex-supported clonal expansion followed by several cycles of Ag
exposure leads routinely to Ag-specific T cell lines. The observation
that these lines gradually acquire specialization for secretion of
IL-4, IL-5, and IL-10 rather than IL-2 and IFN-
is consistent with
other reports that glucocorticoids promote Th2 cell maturation
(11, 29, 39).
Further exploitation of this experimental system may provide insights
into the factors that promote the activation, growth, and
differentiation of Ag-responsive naïve T cells. Studies of
factors, such as IL-12, that could channel the differentiation process
toward the development of IL-2 and IFN-
-secreting effector cells
will be of particular interest. The data presented here are consistent
with models that implicate endocrine factors not merely as inhibitory
modulators of response strength, but as required participants in early
activation events. This experimental system presents opportunities for
the investigation of other neuroendocrine influences that might
modulate immune responses in intact organisms. It will also be of
interest to determine whether analogous approaches allow the routine
production of Ag-specific T cell lines from human blood.
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Richard A. Miller, University of Michigan, Room 5316 CCGCB, Box 0940, 1500 East Medical Center Drive, Ann Arbor, MI 48109. E-mail address: ![]()
3 Abbreviations used in this paper: PCC, pigeon cytochrome c; KLH, keyhole limpet hemocyanin; PLL, polyL-lysine; Dex, dexamethasone;
IL-10, Abs to IL-10. ![]()
Received for publication February 10, 1997. Accepted for publication November 24, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. GALON, D. FRANCHIMONT, N. HIROI, G. FREY, A. BOETTNER, M. EHRHART-BORNSTEIN, J. J. O'SHEA, G. P. CHROUSOS, and S. R. BORNSTEIN Gene profiling reveals unknown enhancing and suppressive actions of glucocorticoids on immune cells FASEB J, January 1, 2002; 16(1): 61 - 71. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Van Laethem, E. Baus, L. A. Smyth, F. Andris, F. Bex, J. Urbain, D. Kioussis, and O. Leo Glucocorticoids Attenuate T Cell Receptor Signaling J. Exp. Med., March 26, 2001; 193(7): 803 - 814. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Milner, S. C. Kent, T. A. Ashley, S. B. Wilson, J. L. Strominger, and D. A. Hafler Differential Responses of Invariant V{alpha}24J{alpha}Q T Cells and MHC Class II-Restricted CD4+ T Cells to Dexamethasone J. Immunol., September 1, 1999; 163(5): 2522 - 2529. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Eisenbraun and R. A. Miller mdr1a-Encoded P-Glycoprotein Is Not Required for Peripheral T Cell Proliferation, Cytokine Release, or Cytotoxic Effector Function in Mice J. Immunol., September 1, 1999; 163(5): 2621 - 2627. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |