| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Production Are Variably Coregulated in Different Human T Lymphocyte Subpopulations1
Departments of
*
Clinical Medicine and
Cellular and Developmental Biology, University of Rome ‘La Sapienza’, Rome, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. We found that, while LAG-3 was expressed only by
CD4+ T cell clones producing IFN-, most
CD8+ clones producing IL-4 but not IFN- (i.e., with a T
cytotoxic-2-like profile) were LAG-3+. The intensity of
LAG-3 expression by CD8+ clones correlated with the amount
of released IFN-, suggesting that this cytokine is not required for
expression but rather for the up-regulation of LAG-3. Flow cytometric
analyses of polyclonal T cell lines confirmed that LAG-3 could be
expressed by both CD4+ and CD8+ cells that did
not contain cytoplasmic IFN-. In these cell lines, large proportions
of CD4+ and CD8+ cells coexpressed LAG-3 and
CD30, a putative marker of Th2-like cells. Overall, our data do not
support the earlier suggestion that LAG-3 and CD30 are selective
markers of T cells with type-1 and type-2 cytokine profiles,
respectively. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, are termed Th1 cells and
mainly promote delayed-type cellular responses, while Th2 cells
producing type-2 cytokines, IL-4 and IL-5, are primarily involved in Ab
responses. Another type of cells (Th0) produces both kinds of
cytokines. Classification according to cytokine profile also
corresponds to CD8+ T cells, which are indicated as T
cytotoxic (Tc4)0, Tc1, and Tc2 cells
(1, 6, 7).
It is believed that shifts to a predominance of Th1-like or Th2-like
responses are associated with some immunologic diseases, outstandingly
allergic diseases, autoimmune disorders and AIDS, and with protective
responses to certain infections (1, 2, 3). Therefore, methods for
monitoring these types of responses might be helpful for clinical
research and are being actively pursued. In this regard, it has
recently been reported that the expression and release of CD30 appears
to be a marker for human and murine CD4+ Th2 cells (8, 9).
Conversely, it has been reported that the expression and release of the
protein encoded by the lymphocyte activation gene-3 (LAG-3) occurs
selectively in cloned CD4+ cells with a Th0/Th1 cytokine
profile (10) and is up-regulated by IL-12 through the induction of
IFN- (11).
LAG-3 is a member of the Ig superfamily which is closely related to CD4; it is expressed by activated T cells and NK cells (12). The gene encodes a transmembrane protein which, like CD4, appears to functionally interact with MHC class II molecules (13, 14, 15, 16). However, the function(s) of LAG-3 is still poorly understood. Disrupting LAG-3 in mice does not appear to determine the defects of T cell functions, but rather abolishes the NK activity that is specifically directed to certain tumor targets (17). No immune abnormalities related to defects of LAG-3 have been observed in humans, and subjects with putative autoantibodies to LAG-3 (18) do not appear to have gross alterations of T or NK cell functions (our unpublished observations). Consequently, confirming an association with a specific cytokine profile (10, 11) might help to clarify the function(s) of LAG-3.
Based on these considerations, we conducted a series of experiments
comparing LAG-3 expression and cytokine profiles in long-term T cell
clones and in polyclonal populations of activated T lymphocytes. Our
results demonstrate that, although LAG-3 expression is indeed
associated with the production of IFN- in the case of
CD4+ long-term cultured clones, this association is not
observed with CD8+ clones or with CD4+ cells in
activated polyclonal cell lines. Furthermore, LAG-3 and CD30 can be
coexpressed by activated T cells that do not produce IL-4.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PBMCs were separated by density gradient centrifugation
(Lymphoprep, Nycomed, Oslo, Norway), washed, and resuspended in RPMI
1640 (Life Technologies, Grand Island, NY) supplemented with
antibiotics and 10% FCS (HyClone, Logan, Utah) (complete medium) at
2 x 106 cells/ml. Cells were stimulated with PHA
(1 µg/ml; Wellcome, Beckenham, U.K.) in 24-well flat-bottom plates
(Falcon, Oxnard, CA). Human rIL-2 (20 U/ml; Boehringer Mannheim,
Mannheim, Germany) was subsequently added to the cells at weekly
intervals. Long-term polyclonal cultures were maintained in
IL-2-supplemented medium and periodically (every 3–4 wk) restimulated
with PHA in the presence of -irradiated (6000 rad), allogeneic PBMCs
as feeder cells. For cloning, a portion of the cells were resuspended
in fresh complete medium supplemented with IL-2 and 2-ME (50 µM;
Sigma, St. Louis, MO) and seeded at 0.3 cells/well in U-bottom 96-well
plates (Falcon) together with 2 x 105 irradiated
allogeneic feeder cells at 3 wk after the initial PHA stimulation as
described previously (6). After 3 wk, growing clones were identified
using an inverted microscope and expanded by weekly splitting and the
addition of IL-2-supplemented medium.
A polyclonal cell line specific for the hypervariable region-1 (HVR-1) of the hepatitis C virus (HCV) envelope was obtained by stimulating PBMCs from one patient with chronic hepatitis C with the peptide ATYTTGGSAAKTAHRLASFFTVGPKQD (10 µg/ml). Starting on day 7, cultures were supplemented with 5 U/ml IL-2 and, from day 14, they were restimulated weekly with the HVR-1 peptide and irradiated autologous PBMCs.5 Several batches of cells were frozen between days 30 and 45 of culture. For the experiments presented here, cells were thawed and restimulated with Ag plus autologous PBMCs; 3 to 5 days later, the cells were surface phenotyped directly or treated further for the detection of cytoplasmic cytokines as described below.
Flow cytometry
Three-color immunophenotyping was performed according to the following procedure. Cells were washed and resuspended in cold PBS containing 2% FCS and 0.02% natrium azide (complete PBS) at 2 x 106/ml; all of the subsequent steps were completed while keeping the cells in an ice bath. A total of 100 µl of cell suspensions were mixed with 10 µl of unconjugated anti-LAG-3 (13) (clone 17B4, 10 µg/ml; obtained from Ares Serono, Geneva, Switzerland) and incubated at 4°C for 30 min. After two washes with complete PBS, cells were incubated with saturating amounts of phycoerythrin (PE)-conjugated goat anti-mouse IgG (Dakopatts, Glostrup, Denmark) in the dark at 4°C for 30 min. After two additional washes, the residual binding sites of the anti-mouse IgG Ab were blocked with 4 µl of normal mouse serum for 10 min, and the cells were subsequently counterstained with pairs of relevant mAbs or isotype-matched control mouse Ig that had been conjugated directly with FITC or with peridin chlorophyll protein (all from Becton Dickinson, Mountain View, CA). After a 30-min incubation in the dark, cells were washed twice, resuspended in 1 ml of complete PBS, and analyzed by flow cytometry.
For the detection of intracellular cytokines, T cell lines were
restimulated for 6 h with 10 ng/ml PMA and 1 µg/ml ionomycin in
the presence of either 3 µM monensin or 10 µg/ml brefeldin A (all
from Sigma). After restimulation, cells were surface-stained with
anti-LAG-3 plus PE-conjugated anti-mouse IgG or with
anti-CD30 as described above. Cells were then washed twice with
complete PBS and fixed in ice-cold PBS containing 0.4%
paraformaldehyde for 10 min. After two additional washes in cold PBS,
the cells were resuspended in 300 µl of PBS containing 0.1% saponin,
10% human AB serum, 100 µg goat IgG, and 0.01 M HEPES buffer. The
cells were spun down after 10 min, and FITC- or PE-conjugated
anti-IFN- or anti-IL-4 Abs (Becton Dickinson) that had been
diluted at 1 µg/ml in PBS containing 0.1% saponin and 0.1 M HEPES
buffer were added for 20 min at room temperature. After two additional
washes, the cells were resuspended in PBS and analyzed by flow
cytometry.
Flow immunocytometry was conducted using either a Cytoron Absolute (Ortho Diagnostic Systems, Raritan, NJ) or a FACScalibur (Becton Dickinson) instrument. At least 10,000 events were acquired, and the data were analyzed with WinMDI (Joseph Trotter, Scripps Institute, La Jolla, CA) or Cell Quest (Becton Dickinson) software. Live lymphocytes were electronically gated by side and forward light scatter. The mean fluorescence intensities (MFIs) were calculated by subtracting the values obtained with the control Abs from the experimental values.
Cytokine measurements
Supernatants of T cell lines and clones were collected at
48 h after restimulation with PHA-P (1 µg/ml), frozen at
-80°C, and thawed once. The cytokine content of the supernatants was
determined using commercial ELISA kits for IFN- and IL-4
(Quantikine, R&D Systems, Minneapolis, MN). All samples were tested in
duplicate. Measurements were performed according to the manufacturer’s
instructions, and the results were interpolated from the standard
reference curve provided with each kit. Clones were classified as
nonproducers when secreted cytokines were below the lower limits of
detection (15.6 pg/ml for IFN- and 31.2 pg/ml for IL-4). The
detection of IFN- and IL-4 mRNAs by RT-PCR was performed as
described previously (6).
Statistical analyses
Statistical analyses were completed using the Mann-Whitney U test and the Spearman rank correlation coefficient. The tied p values are provided.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production is not required for LAG-3 expression by
CD8+ T cell clones
We evaluated the expression of surface LAG-3 protein in a panel of
CD4+ and CD8+ clones that had been
classified according to their cytokine profiles. This panel included
some unusual CD8+ clones, which produced IL-4 and not
IFN- (i.e., with a Tc2-like profile); all but one of these clones
were derived from HIV-infected individuals with the hyper IgE syndrome
(6). In agreement with earlier data (10), we found that the expression
of LAG-3 among CD4+ clones was restricted to those clones
releasing IFN- in supernatants (Fig. 1
). By contrast, this constraint was not
apparent among CD8+ clones, since 8 of 11 CD8+
clones with a Tc2-like profile expressed LAG-3 surface protein (Fig. 1). The absence of IFN- mRNA and the presence of IL-4 mRNA were
confirmed by RT-PCR in two representative
CD8+/LAG-3+ clones with a Tc2-like profile
(data not shown).
|
, the MFIs (Fig. 1
) were significantly lower in
Tc2-like rather than in Tc0- or Tc1-like clones
(p = 0.002 and p = 0.037,
respectively, by the Mann-Whitney U test), suggesting that
the production of IFN- could influence the level of expression of
LAG-3. A significant direct correlation between the intensity of LAG-3
expression and the amount of secreted IFN- was, indeed, confirmed in
CD8+ clones (p = 0.0005 by the
Spearman rank correlation coefficient); this correlation was only
borderline (p = 0.070) in CD4+
clones, probably owing to the relative paucity of LAG-3-expressing
clones. Finally, CD4+ clones expressed, overall,
significantly lower amounts both of surface LAG-3 and secreted IFN-
(p < 0.0001 and p = 0.0427,
respectively, by the Mann-Whitney U test) than
CD8+ clones.
Taken together, our data on T cell clones indicate that, at least in
CD8+ cells, IFN- production is not required to initiate
the biosynthesis of the LAG-3 protein; these data also suggest that the
effect of this cytokine on LAG-3 might rather be the up-regulation of
its level of expression.
Since most of the CD8+ Tc2-like clones used in this study were derived from HIV-infected subjects with hyper IgE syndrome (6), it could be argued that LAG-3 expression was promoted in these cells by viral infection. However, this possibility seems unlikely, since one CD8+/Tc2-like clone obtained from an HIV-negative subject with severe atopy was also LAG-3+, and the LAG-3+ MFIs were, overall, not significantly different in CD8+ clones derived from HIV-infected and uninfected subjects.
Lack of correlation between expression of LAG-3 or CD30 and
production of IFN- or IL-4 in polyclonal T cell lines
Our results with the T cell clones prompted us to investigate the relationship between cytokine profile and the capacity to express LAG-3 and CD30 in mitogen-stimulated or Ag-specific polyclonal T cell lines at the single-cell level.
In a preliminary kinetic study of a PHA-stimulated, IL-2-dependent cell
line (Fig. 2), LAG-3 expression by
CD4+ and CD8+ cells increased steadily
over 30 days of culture; however, more CD8+ than
CD4+ cells were LAG-3+, a finding that is
similar to our data in T cell clones and to previous observations (19).
The kinetics of expression of LAG-3 and CD25 appeared to be
independent, since CD25+ cells declined after day 14 of
culture, and LAG-3 was expressed more frequently by CD8+
than by CD4+ cells, although the latter expressed more
CD25. By contrast, LAG-3 expression was invariably associated with that
of MHC class II molecules, as all LAG-3+ cells were also
HLA-DR+ (Fig. 2 and data not shown). The fact that LAG-3
seemed to be expressed only by HLA-DR+ T cells might
reflect the postulated function of LAG-3 in modulating MHC class
II-mediated interactions (13, 14, 15, 16).
|
expressed the LAG-3 protein, we assessed the
dependence of LAG-3 expression on cytokine production at the
single-cell level. We initially examined an IL-2-dependent polyclonal
cell line derived from an individual with hyper IgE syndrome that was
secondary to HIV infection who had a predominance of CD8+ T
cells with a Tc2 cytokine profile (6); we found that a large majority
of these cells were LAG-3+, although very little IFN-
(<30 pg/ml) and abundant amounts of IL-4 (1392 pg/ml) were secreted in
the culture supernatant. This finding suggested that LAG-3 expression
could be at least partially independent of IFN- production. To
further address this issue, we examined the coexpression of
intracellular IFN- and surface LAG-3 by flow immunocytometry in
these cultured cells. As shown in Figure 3A, the large majority of
these cells were IFN-- but expressed surface LAG-3. To
rule out the possibility that the lack of association of LAG-3 with
IFN- production was due to the fact that these mostly
CD8+ cells were derived from an HIV-infected subject, we
examined an IL-2-dependent polyclonal T cell line obtained from an
HIV-negative individual with severe atopy. These cells did not secrete
detectable IFN- in the culture supernatant and were 
60%
CD4+. Three-color staining showed that, although 96% of
these CD4+ cells were negative for cytoplasmic IFN-,
nearly all expressed surface LAG-3 (Fig. 3C). The
expression of LAG-3 by a cell line polarized toward IFN- production
is shown for comparison in Figure 3D.
|
E); by contrast, secondary restimulation with PHA in
the presence of feeder cells determined the up-regulation of LAG-3 and
induced the coexpression of CD30 in a large fraction of these cells
(Fig. 3F).
Finally, we analyzed the patterns of expression of LAG-3, CD30, and
intracellular cytokines in polyclonal T cells, specific for a peptide
corresponding to the HVR-1 of HCV, that secreted significant amounts of
both IFN- and IL-4. In a representative experiment these cells, were
80% CD30+, 42% LAG-3+, and 28% CD30/LAG-3
double-positive at 5 days after restimulation with Ag. According to
three-color immunofluorescence, we found that nearly all
CD4+ cells in this population were CD30+, with
about one-third coexpressing LAG-3 (Fig. 4A); among
CD8+ cells (Fig. 4B), 27% expressed CD30
only, 21% expressed LAG-3 only, and 50% were CD30/LAG-3
double-positive. The relationship between the patterns of intracellular
cytokines and the expression of surface LAG-3 and CD30 could only be
investigated in these cells by two-color immunofluorescence, since the
treatments needed for the optimal detection of cytokines reduced the
levels of expression of surface molecules to below the threshold of
detectability using peridin chlorophyll protein-conjugated reagents.
For the same reason, the proportions of LAG-3+ and
CD30+ cells as well as the corresponding MFIs were much
lower in the samples treated for cytokine analyses than in the
untreated cells used for the analyses of surface molecules (Fig. 4, C and D and Fig. 4, A and
B, respectively). Nevertheless, we were able to determine
that about one-half of the LAG-3+ cells were
IFN-+, while very few were IL-4+ (Fig. 4, C and D); 45% of the CD30+
cells contained IFN- (Fig. 4E). Although
IL-4-producing cells (8% of the total) were primarily contained in
the CD30+ population, the large majority of
CD30+ cells were negative for intracellular IL-4 (Fig. 4F). This cell line contained 74% CD4+
cells and 26% CD8+ cells as well as 35% type 1-like,
3% type 2-like, and 5% type 0-like cells (data not shown). Thus, it
can be argued according to the overall data that the majority of
LAG-3+ cells (necessarily including those coexpressing
CD30) were type 1-like rather than type 0- or type 2-like among the
cells in which intracellular cytokines could be detected. However,
these data should be interpreted with some caution because of both the
suboptimal detection of surface molecules and the presence of a
relatively large number of cells in which intracellular cytokines were
undetectable.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) (10). This situation is a mirror image of the restricted
expression of CD30, another lymphocyte activation molecule, to T cell
clones with a Th2 profile (8). An important practical implication of
these findings is the evaluation of the "Th1/Th2 balance" in
pathologic conditions characterized by polarization toward either Th1-
or Th2-like responses simply by measuring LAG-3- or CD30-expressing
cells in tissues or in other compartments or by measuring the amounts
of the soluble forms of these molecules in body fluids (2, 3, 10, 20).
Our findings in CD4+ T cell clones are in agreement with
those of Annunziato et al. (10), as LAG-3 expression was observed in
30% of clones with a Th0/Th1 cytokine profile but was not observed
in clones with a Th2 profile. By contrast, IFN- production did not
appear to be required for LAG-3 expression by CD8+ T cell
clones, since nearly all clones with a Tc2 cytokine profile that did
not produce IFN- were LAG-3+. These findings were
reinforced by studies with polyclonal T cell lines derived from
patients with highly polarized T cell responses showing that both
Tc2-like CD8+ cells and Th2-like CD4+ cells can
express LAG-3. Our data clearly demonstrated that the expression of
LAG-3 does not require endogenous or exogenous IFN-; however, we
also observed that the intensity of LAG-3 expression by
CD8+ clones correlated directly with the amount of IFN-
released in supernatants, a finding that is consistent with previous
observations in CD4+ T cell clones (10). Taken together,
our data suggest that, at least in CD8+ T cells, the
expression of the LAG-3 protein is primed by factor(s) other than
IFN- and, once initiated, is up-regulated by the latter cytokine. A
complex regulation of LAG-3 expression by cytokines is consistent with
recent functional studies (11) showing that, although exogenous IL-12
enhanced and neutralizing anti-IFN- Abs markedly suppressed it,
the addition of exogenous IL-4 was apparently capable, at least after
secondary stimulation, of up-regulating LAG-3 without increasing
IFN- production.
Another issue addressed by our study was the proposed paradigm (3, 10)
of a dependency of expression of LAG-3 and CD30 on type-1 and type-2
cytokines, respectively. Although a role for IL-4 in the modulation of
CD30 expression was recently confirmed in a murine model (9), the
original statement (8) that this surface molecule is expressed only by
human T cells producing type-2 cytokines was subsequently challenged
(21). Our present findings demonstrate that LAG-3 and CD30 do not
strictly require IFN- or IL-4, respectively, and that they can be
coexpressed irrespectively on the produced cytokines. Analyses of
Ag-specific polyclonal T cells confirmed that LAG-3 and CD30 are
coexpressed by large proportions of CD4+ and
CD8+ cells after secondary stimulation. Under these
conditions, most CD4+ cells were CD30+, and
nearly all LAG-3+ cells coexpressed CD30, while a discrete
proportion of LAG-3 single-positive cells could be observed only in the
CD8+ subset. The expression of CD30 was, as reported
previously (21), largely independent on intracellular IL-4, and
LAG-3/CD30 coexpression was primarily seen in cells with a type 1-like
cytokine profile.
Our data indicate that, at least after secondary stimulation, LAG-3 and CD30 are expressed by T cells as nonselective activation markers. Thus, from a practical standpoint, the possibility of monitoring the magnitude of Th1- or Th2-like immune responses in vivo simply by measuring the production of LAG-3 and CD30 does not seem to be realistic. Nevertheless, the measurement of these molecules on cells or in biologic fluids might be useful for assessing immune activation in patients with specific conditions. In particular, the high level of LAG-3 expression in patients with multiple sclerosis (10) or Crohn’s disease (22), which are both considered Th1-mediated immune disorders, and the high level of in vivo CD30 expression in a variety of human diseases with a predominant activation of Th2-like cells (20, 23, 24), warrant further investigation.
|
Footnotes |
|---|
2 Current address: Department of Immunodermatology, Istituto Dermopatico dell’Immacolata, Via dei Monti di Creta 104, 00167 Rome, Italy.
3 Address correspondence and reprint requests to Dr. M. Fiorilli, Department of Clinical Medicine, University of Rome ‘La Sapienza’, Viale dell’Università 37, 00185 Rome, Italy. E-mail address:
4 Abbreviations used in this paper: Tc, T cytotoxic; LAG-3, lymphocyte activation gene-3; HVR-1, hypervariable region-1; HCV, hepatitis C virus; PE, phycoerythrin; MFI, mean fluorescence intensity.
5 P. Del Porto, G. Puntoriero, A. Nicosia, and E. Piccolella, Cross-reactive T cell response to the hypervariable region-1 in HCV-infected patients, manuscript in preparation.
Received for publication August 11, 1997. Accepted for publication March 2, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Immunol. 158:2090.[Abstract]
production. FASEB J. 10:769.[Abstract]
on CD30 and lymphocyte activation gene-3 (LAG-3) expression by activated naive T cells. Eur. J. Immunol. 27:2239.[Medline]
This article has been cited by other articles:
|
|
 |
|
  N. Li, C. J. Workman, S. M. Martin, and D. A. A. Vignali Biochemical Analysis of the Regulatory T Cell Protein Lymphocyte Activation Gene-3 (LAG-3; CD223) J. Immunol., December 1, 2004; 173(11): 6806 - 6812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Giovannetti, F. Mazzetta, E. Caprini, A. Aiuti, M. Marziali, M. Pierdominici, A. Cossarizza, L. Chessa, E. Scala, I. Quinti, et al. Skewed T-cell receptor repertoire, decreased thymic output, and predominance of terminally differentiated T cells in ataxia telangiectasia Blood, December 1, 2002; 100(12): 4082 - 4089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-i. Kashiwakura, N. Suzuki, H. Nagafuchi, M. Takeno, Y. Takeba, Y. Shimoyama, and T. Sakane Txk, a Nonreceptor Tyrosine Kinase of the Tec Family, Is Expressed in T Helper Type 1 Cells and Regulates Interferon {gamma} Production in Human T Lymphocytes J. Exp. Med., October 18, 1999; 190(8): 1147 - 1154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Yamaguchi, J. de Vries, and H. Yssel Differentiation of human single-positive fetal thymocytes in vitro into IL-4- and/or IFN-{gamma}-producing CD4+ and CD8+ T cells Int. Immunol., April 1, 1999; 11(4): 593 - 603. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) individuals.
Clones are classified as type 0, 1, or 2 according to their cytokine
secretion profiles. The LAG-3 MFIs are calculated by subtracting those
obtained with the negative control Ab.
