| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
*
Medicine and
Surgery, University of Connecticut School of Medicine, Farmington, CT 06030
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results of the early translational studies provide a clear rationale for further studies designed to establish optimum protocols for immunization, to develop methods of patient monitoring, and to improve the efficacy of APC-based vaccine. Interestingly, most of the ongoing studies presently use multiple rounds of immunizations at fairly short intervals as the preferred method of immunization. Further, for patient monitoring, one of the popular approaches has been to look for T cells (derived from tumor nodules, vaccine sites, and circulation) capable of responding to the relevant Ag, in a suitable in vitro assay, after several rounds of specific stimulation. Indeed we and others (7, 9, 10) have been able to generate autologous tumor-reactive and peptide-specific CTL from the circulation, tumor deposits, and vaccine sites following in vitro stimulation with peptide-pulsed cultured APC. However, we also find that although in vitro stimulations of the effector cells lead to the generation of Ag-specific CTL, in vitro, repetitive stimulation with APC can at times lead to the emergence of noncytolytic CD4+ T cells exhibiting the characteristic phenotype of Th2 cells. Our studies further show that these CD4+ T cells and a cell-free supernatant factor derived from the stimulated CD4+ T cells are capable of blocking the activation of fresh lymphocytes when such lymphocytes are stimulated with lectin or by anti-CD3 Ab. A culture supernatant factor(s) from these CD4+ cells also exhibits a marked inhibitory effect on the expression of the costimulatory molecules, CD80 and CD86, by APCs. The inhibitory effect of the supernatant factor can, however, be almost totally abrogated by neutralizing the IL-10 activity in the supernatant. These observations therefore have implications on the design of APC-based tumor vaccine design.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients with metastatic melanoma and melanoma patients with no active disease but who were considered at risk of recurrence were entered into the trial with informed consent. The details of the trial design and the patient-monitoring systems have been described (7, 10). Briefly, the study consisted of two groups of patients receiving either a vaccine made of synthetic peptide-loaded APC (7) or tumor lysate-loaded APC (10). The immunization protocol consisted of four monthly intradermal immunizations with the autologous myeloid APC (grown in granulocyte-macrophage-CSF) loaded with synthetic peptide or with the tumor lysate at monthly intervals with escalating numbers of cells (5 x 105, 1 x 106, 5 x 106, and 1 x 107) with three patients in each dose level. The desired number of Ag-loaded APC were drawn in 0.1 ml of normal saline and were injected intradermally, close to a peripheral lymph node-bearing region on a rotating basis. Laboratory monitoring of immune response to the vaccine was performed with detailed phenotypic and functional analyses of lymphocytes, in circulation and infiltrating excised vaccine sites and/or tumor sites, expanded in vitro in IL-2.
Preparation of tumor lysate as a source of tumor-associated Ag
The procedure for preparation of the tumor cell lysate has been
described (10). Fresh tumor tissues were homogenized without using any
enzyme to obtain homogenates of single melanoma cells or melanoma cells
in clumps. The resulting homogenate was washed and cell concentration
was adjusted to 
106 cells/ml in PBS. A cellular lysate
was then prepared from the homogenate by six repeated freezings (at
-180°C) and thawings to room temperature. Total lysis of all the
cells was verified by trypan blue dye exclusion staining, cellular
debris was removed by centrifugation (1000 x g for 10
min), and the protein concentration of the lysate was determined.
Aliquots of the homogenates were then frozen at -80°C for loading
onto the cultured APC for vaccination or for in vitro assays.
Tissue culture
Tissue culture technique and the procedure of isolating fresh tumor cells from tissue explants have been described earlier (7, 11). Briefly, tissue cultures were performed in Iscove’s medium (Life Technologies, Gaithersburg, MD) supplemented with 10% FBS, L-arginine (0.55 mM), L-asparagine (0.24 mM), and L-glutamine (1.5 nM), henceforth described as complete medium (CM3). Fresh tumor cells were isolated by mechanical fine mincing of tumor tissues in serum-free medium.
APC culture
APC used in this study were peripheral myeloid APC expanded in vitro as described earlier (7, 10). Briefly, monocyte/macrophages were isolated as adherent cells from Ficoll-Hypaque gradient-derived mononuclear cell populations. The adherent cells were then cultured in CM containing 1000 U/ml of granulocyte-macrophage-CSF (Immunex, Seattle, WA) for 14 days. The nonadherent and loosely adherent cells were harvested by vigorous washing.
Peripheral blood lymphocytes
PBL were isolated on a Ficoll-Hypaque gradient (11). All cultures and experiments were performed in CM.
In vitro expansion of infiltrating lymphocytes
Lymphocytes infiltrating tumors (TIL) and lymphocytes infiltrating the vaccine site (VIL) were expanded, ex vivo, as previously described (7, 10). Briefly, homogenates of the excised tissues were prepared by mechanical means. The minced homogenates containing the infiltrating lymphocytes were then cultured in IL-2 (50 U/ml) in the presence of the appropriate immunogen (melanoma cell lysate-loaded or peptide-loaded autologous APC). The cultures were restimulated every 7 to 10 days with the respective immunogen. Phenotypic and functional analyses were performed as sufficient numbers of lymphocytes became available. All experiments shown in this communication were performed with the ex vivo-expanded infiltrating lymphocytes that were stimulated at least three times and were maintained in culture for 3 to 6 wk or longer.
Phenotypic analysis
The immunofluorescence procedure for phenotypic analysis in cytofluorography has been described (11).
Microcytotoxicity assay
The 51Cr release microcytotoxicity assay has been described (11). Freshly prepared or cryopreserved target cells were labeled with 51Cr with good efficiency. The spontaneous release of the radioactivity from the 51Cr-labeled target cells was usually <15%.
Cytokine synthesis assay
The ex vivo-expanded VIL or TIL (103 to 2 x
103 cells/well) were stimulated against the appropriate
immunogen (autologous melanoma cells, when available, lysate-loaded
APC, etc.) in individual wells of a 48-well cluster plate (Costar,
Cambridge, MA) in 0.5 ml of medium for 24 h. The supernatants,
harvested after 24 h, were assayed for relevant type 1 or type 2
cytokine depending on the experimental design. TNF activity was assayed
in a TNF bioassay using the WEHI 164 clone 13 (12). Other cytokines
(IFN-
, IL-4 and IL-10) were assayed in appropriate ELISA assays as
per manufacturer’s (Coulter, Miami, FL) directions.
Intracellular analysis of cytokine production
The method for intracellular analysis of cytokine production by T cells has been described (13). Briefly, the effector cells (CD4+ T cells) are restimulated with the stimulator cells (APC) for 2 h (responder:stimulator ratio, 10:1) after which they are incubated with nonstimulatory doses of PMA (0.5 ng/ml) and ionomycin (4.0 ng/ml; Sigma, St. Louis, MO) for 4 h at 37°C. Brefeldin A (1 µg/ml; Sigma) is added 2 h before harvest. The cells are washed and stained with Cy-chrome-conjugated anti-CD4 (PharMingen, San Diego, CA) for 30 min on ice. They are then fixed in 4% paraformaldehyde (Sigma) for 25 min and permeabilized with PBS/BSA/saponin for 10 min. Cells were then stained with conjugated anti-cytokine mAb (IL-10) and analyzed on a FACScan, and the percentage of cytokine-producing cells within CD4+ cells are determined.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of our goals has been to establish long term lines of
autologous melanoma-reactive and Ag-specific CTL from tumor sites,
vaccine sites, and PBL from patients immunized with synthetic
peptide-pulsed or melanoma lysate-loaded APC. Our approach was to
present the relevant Ag (peptide or tumor lysate-pulsed APC) to
autologous lymphocytes and expand the CTL populations with exogenous
IL-2 through several rounds of weekly stimulation with the relevant
immunogen. In several cases, we were able to detect the generation of
CTL. Fig. 1
shows the results of
functional characterization of several in vitro-expanded VIL and TIL
populations, demonstrating their MHC class I-restricted and Ag-specific
cytolytic activity. As shown, in one of these cases (RM), the VIL
exhibited MHC class I-restricted lysis of the autologous melanoma cells
(RM-M). The VIL also recognized the MAGE-1.A1 peptide, EADPTGHSY, quite
selectively. Among the two TIL populations, TIL from RG recognized the
autologous melanoma cells only, while the TIL from JL recognized the
autologous melanoma cells (JL-M) and another allogeneic melanoma cell
line, GL-M. Of interest, both JL and GL were HLA-A2.1 positive.
Although we clearly detected CTL activity in the cultured populations,
we were unable to keep the Ag-specific CTL in culture for a prolonged
period. In most cases, the CTL activities declined, at times
precipitously, and a CD4+ noncytolytic population emerged.
|
shows four experiments in four
separate case studies of VIL or TIL as examples of the patterns of T
cell responses observed following repeated in vitro stimulation with
peptide-pulsed or tumor lysate-pulsed autologous APC or with autologous
tumor. Fig. 2, a and b, shows one of these
patterns characterized by a decline of the CTL activities (in one VIL
and in one TIL population) in in vitro cultures with concomitant rise
of CD4+ T cells. In both cases, after 2 weeks of culture,
there were more CD8+ T cells than CD4+ cells,
which lysed the autologous melanoma cells in MHC class I-restricted
fashion (as shown in Fig. 1
). However, after several more rounds of
stimulation, the CTL activities declined and a predominantly
CD4+ population outgrew. Fig. 2c shows a
different pattern of response in which the expanding T cells derived
from a vaccine site became predominantly CD4+ from the
beginning, showing very little cytolytic activity against the
autologous melanoma cells. Fig. 2d shows the third pattern
of T cell reactivity generated in in vitro stimulation of the TIL with
autologous tumor cells. As shown, in this pattern, the CTL activity
declined with time but the decline did not correspond with a fall of
CD8+ T cells and concomitant rise of CD4+
noncytolytic T cells.
|
We argued that the outgrowth of the CD4+ cells could
have an effect on the decline of the CTL activity in these cultures.
Accordingly, we undertook a detailed functional analyses of the
CD4+ T cells emerging in these cultures. Indeed, when
functionally analyzed, some of the CD4+ T cells emerging in
culture after several rounds of stimulation showed all the
characteristics of Th2 type cells. Table I shows the cytokine synthesis profile of
the CD4+ T cells from the experiments outlined in Fig. 2, a and b. As shown, the CD4+ T cells
from both cases synthesized IL-4 and IL-10 in a class II-restricted
manner when they were restimulated with the autologous APC (Table I).
Of considerable interest, the CD4+ T cells obtained from
one of these cases (RG) recognized the autologous melanoma cells in a
class II-restricted manner (Table I). The tumor cells in this case
expressed MHC class II molecules (data not shown). Of further interest,
the CD4+ T cells did not recognize allogeneic APC or
allogeneic melanoma cells. The autologous melanoma cells RM-M or RG-M
did not synthesize any detectable IL-10, suggesting that the IL-10 in
these experiments was synthesized by the T cells. Fig. 3 shows an intracytoplasmic cytokine
analysis of the CD4+ T cells at single-cell level. As shown
in the TIL, some of the CD4+ T cells synthesized IL-10
(Fig. 3b). However, a larger number of cells synthesized
IL-10 (Fig. 3c) after stimulation with the autologous APC.
Those CD4+ cells were found to secrete IL-10 are all
CD3+ T lymphocytes (data not shown). Although the
TIL culture used in this study contains predominantly CD4+
T cells, it also contains some CD3+CD8+ T cells
and some NK-like cells (total, 25%), which might secrete IL-10
constitutively or due to the effect of PMA + ionomycin. Interestingly,
these 25% of cells do not respond to further stimulation with APC
(Fig. 3, b and c).
|
|
Suppression of T cell activation by the expanded CD4+ T cells
The CD4+ T cells obtained from these cultures affected
the activation of T cells (autologous or allogeneic) in in vitro
coculture. PBL were stimulated in culture with 2 µg/ml PHA or with
immobilized anti-CD3 Ab in the presence of these CD4+ T
cells (PBL:CD4+ cell ratio, 10:1). Fig. 4 shows a representative experiment
demonstrating that the proliferation of PBL was completely blocked in
the presence of these cells. When culture supernatants collected from
the PHA- or anti-CD3 Ab-stimulated PBL were analyzed for the
presence of IL-2, it was observed that the production of IL-2 was
significantly down-regulated in those cultures where this type of
CD4+ T cell was added (Table II).
|
|
shows an example of the
suppressive effect of the culture supernatant from the CD4+
T cell culture. As shown, the culture supernatant added to the fresh
cultures of T cells blocked T cell activation with different stimuli.
Interestingly, the supernatant factor also had a strong negative effect
on the up-regulation of the expression of the costimulatory molecules
CD80 and CD86 by the APC (data not shown). This type of inhibitory
effect of the culture supernatants on the expression of the
costimulatory molecules (CD80 and CD86) on the APC was seen in all
three of three cases studied (collective data not shown).
|
Since CD4+ T cells outgrowing in these cultures
synthesized type 2 cytokine and because IL-10 has been known to possess
considerable inhibitory activity on APC and T cells, we examined the
effect of neutralization of IL-10 on the maintenance of CTL activity
using one of our patients who was immunized with the MAGE-3.A2 peptide
(FLWGPRALV)-pulsed APC in whom we observed the rise and decline of CTL
activity with concomitant rise of CD4+ T cells in the TIL
culture (shown earlier in Fig. 2b). Fig. 5 shows the result of the experiment. The
TIL from this case were cultured in IL-2 medium in the absence or in
the continuous presence of anti-IL-4, anti-IL-10, or control
mouse Ab for 60 days. The cultures were restimulated every 10–12 days.
As shown, the CTL activity declined within 2 wk in the standard
culture. However, the neutralization of IL-10 (and not IL-4) maintained
the CTL activity up to 60 days. Similarly, when the PBL from the same
case were stimulated with the MAGE-3.A2 peptide (FLWGPRALV)-pulsed
APC in the presence of anti-IL-10 Ab, peptide-specific and
autologous melanoma-reactive CTL activity could also be generated and
maintained for 60 days (Table IV). The
neutralization of IL-10 prevented the decline of CD8+ T
cells in the continuous culture and helped maintain their continued
growth (data not shown).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this equation, suppression of T cell response that might have been
induced spontaneously (as in autoimmunity) or deliberately (as in
immunotherapy) cannot be ignored, especially when the T cell response
is induced against self Ag. The concept of T cell-mediated suppression
of cellular immune response has been controversial. However, a
functional dichotomy with CD4+ T cells into Th1 and Th2
types exhibiting opposite effects (Th2 type acting as regulatory cells
in some situations) is fairly well established (23). Indeed, in some
animal models, the ultimate nature of the immune response to certain
types of infection (i.e., protection vs infection) is often determined
by whether the animal mounts a Th1- or Th2-type response (24). Thus,
the observations reported here are noteworthy. First, the decline of
CTL response, in vitro, despite antigenic stimulation provided through
APC expressing all the essential costimulatory molecules and the
emergence of the Th2 type CD4+ T cells, suggests that the
window of opportunity to expand a CTL response to these types of self
Ags might be quite narrow. Second, the emergence of the
CD4+ Th2-type response following a CTL response might
represent a form of inherent regulatory mechanism for controlling
harmful expansion of "self-reactive" T cells. Third, the mechanism
of regulation by these CD4+ T cells, as well as their
target, is quite interesting. These CD4+ T cells suppress T
cell activation directly as evidenced by their suppressive effect on
the lectin- or TCR-driven activation of fresh T cells (Table III). This
is in line with our earlier study in which we have shown that
CD4+-regulatory cells can exert an inhibitory effect on the
T cells by down-regulating the expression of IL-2 receptor (25, 26). We
and other investigators have found that in addition to affecting T cell
activation directly, these types of CD4+ T cells can also
down-regulate the expression of costimulatory molecules on the APC
(collective data not shown (27)). Although we have not directly
addressed the issue, it is possible to argue that by down-regulating
the expression of costimulatory molecules on the APC, CD4+
T cells can render the APC ineffective as APCs and/or as stimulatory
cells. This type of CD4+ T cells, therefore, seems capable
of exerting a much broader regulatory effect, resulting in the
suppression of T cell activation. In this way, CD4+ T cells
seem to have a foolproof inhibitory effect on the generation of T
cell-mediated immune response. Fourth, it seems that this type of
CD4+ T cells can exert their regulatory function through a
cytokine that appears to be IL-10 (Table IV and Fig. 5), although
several reports have indicated that this type of suppressive effect
could in part be attributed to TGF-ß secreted by the regulatory T
cells (30). Admittedly, the relationship between the different
subsets of regulatory T cells is complex and unclear at present. In our
present observation, indeed, neutralization of IL-10 in the cultures
maintained the CTL activity for a considerable length of time (Table IV
and Fig. 5). In this context, it should be pointed out that IL-10 has
been a remarkable immunoregulatory cytokine. Although some
investigators have found IL-10 to have a "stimulatory"
effect (28, 29), the great majority of the published work attests to
its inhibitory effect on APC, T cells, and NK cells (27, 31, 32, 33, 34, 35, 36).
The physiological relevance of these types of CD4+ T cells emerging in culture is admittedly unclear. The outgrowth of CD4+ T cells in our system might have resulted from their inherent proliferative advantage in in vitro culture containing exogenous IL-2. However, an extensive literature on the negative role of CD4+ T cells in tumor immunity exists (37, 38). Ag-specific, as well as Ag nonspecific, suppressor CD4+ T cells have been described (37, 38, 42). In contrast, evidence of beneficial effect of removal of CD4+ T cells in context to tumor immune manipulations has also been described by several investigators (39, 40, 41). Thus, the emergence of this type of cells might have physiological relevance. Of interest, a profound inhibitory function of CD4+ T cells, in vitro and in vivo, has recently been documented by several different groups of investigators in two murine disease models (43, 44, 45). Although it is not possible to conclude that repeated immunization with APC-based vaccine in vivo might have a similar inhibitory effect, it is not inconceivable that repeated APC-based vaccination might facilitate the emergence of the process as a countermeasure to expansion of potentially injurious self-reactive T cell populations. This could be an Achilles heel in the currently ongoing APC-based cancer immunotherapy. Controversies will undoubtedly continue on the subject of T cell-mediated suppression of immune response. It should be pointed out, however, that in context to vaccine therapy for cancer, opportunities exist to test a number of approaches designed to widen the window of opportunity of expanding tumor-associated but self Ag-specific CTL and to circumvent this type of potentially negative role of Th2-type response. Among others, these approaches could include: a different vaccination protocol, such as quick priming, followed by periodic booster injections; the use of pharmacological inhibitors of Th2-type response; the inclusion of a Th1-type cytokine such as IL-12; or the engineering of APC to synthesize Th1-type cytokine, such as IL-12, locally.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Nitya G. Chakraborty, Department of Medicine, University of Connecticut Health Center, Farmington, CT 06030-3210. E-mail address:
3 Abbreviations used in this paper: CM, complete medium; TIL, tumor-infiltrating lymphocytes; VIL, vaccine-infiltrating lymphocytes.
Received for publication August 3, 1998. Accepted for publication February 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory cells. J. Exp. Med. 178:1041. production and diminishes Interleukin-4 inhibition of such priming. Proc. Natl. Acad. Sci. USA 1990:10188.
This article has been cited by other articles:
|
|
 |
|
  G. Lizee, L. G. Radvanyi, W. W. Overwijk, and P. Hwu Improving Antitumor Immune Responses by Circumventing Immunoregulatory Cells and Mechanisms. Clin. Cancer Res., August 15, 2006; 12(16): 4794 - 4803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Lizee, L. G. Radvanyi, W. W. Overwijk, and P. Hwu Immunosuppression in melanoma immunotherapy: potential opportunities for intervention. Clin. Cancer Res., April 1, 2006; 12(7): 2359s - 2365s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chattopadhyay, S. Mehrotra, A. Chhabra, U. Hegde, B. Mukherji, and N. G. Chakraborty Effect of CD4+CD25+ and CD4+CD25- T Regulatory Cells on the Generation of Cytolytic T Cell Response to a Self but Human Tumor-Associated Epitope In Vitro J. Immunol., January 15, 2006; 176(2): 984 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mehrotra, A. Chhabra, S. Chattopadhyay, D. I. Dorsky, N. G. Chakraborty, and B. Mukherji Rescuing Melanoma Epitope-Specific Cytolytic T Lymphocytes from Activation-Induced Cell Death, by SP600125, an Inhibitor of JNK: Implications in Cancer Immunotherapy J. Immunol., November 15, 2004; 173(10): 6017 - 6024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Veldman, A. Hohne, D. Dieckmann, G. Schuler, and M. Hertl Type I Regulatory T Cells Specific for Desmoglein 3 Are More Frequently Detected in Healthy Individuals than in Patients with Pemphigus Vulgaris J. Immunol., May 15, 2004; 172(10): 6468 - 6475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Manjili, X.-Y. Wang, X. Chen, T. Martin, E. A. Repasky, R. Henderson, and J. R. Subjeck HSP110-HER2/neu Chaperone Complex Vaccine Induces Protective Immunity Against Spontaneous Mammary Tumors in HER-2/neu Transgenic Mice J. Immunol., October 15, 2003; 171(8): 4054 - 4061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Somasundaram, L. Jacob, R. Swoboda, L. Caputo, H. Song, S. Basak, D. Monos, D. Peritt, F. Marincola, D. Cai, et al. Inhibition of Cytolytic T Lymphocyte Proliferation by Autologous CD4+/CD25+ Regulatory T Cells in a Colorectal Carcinoma Patient Is Mediated by Transforming Growth Factor-{beta} Cancer Res., September 15, 2002; 62(18): 5267 - 5272. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. G. Chikh, S. Kong, M. B. Bally, J.-C. Meunier, and M.-P. M. Schutze-Redelmeier Efficient Delivery of Antennapedia Homeodomain Fused to CTL Epitope with Liposomes into Dendritic Cells Results in the Activation of CD8+ T Cells J. Immunol., December 1, 2001; 167(11): 6462 - 6470. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Samuel Antiviral Actions of Interferons Clin. Microbiol. Rev., October 1, 2001; 14(4): 778 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Levings, R. Sangregorio, F. Galbiati, S. Squadrone, R. de Waal Malefyt, and M.-G. Roncarolo IFN-{{alpha}} and IL-10 Induce the Differentiation of Human Type 1 T Regulatory Cells J. Immunol., May 1, 2001; 166(9): 5530 - 5539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Corcoran, S. Doyle, D. Waldron, A. Nicholson, and B. P. Mahon Impaired Gamma Interferon Responses against Parvovirus B19 by Recently Infected Children J. Virol., November 1, 2000; 74(21): 9903 - 9910. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, T4; 
, T8; •, cell-mediated cytotoxicity.
