Published online before print February 14, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Immunology Group and
Department of Intramural Research Affairs, National Health Research Institutes, Taiwan, Republic of China
2Correspondence: Department of Intramural Research Affairs, National Health Research Institutes, No. 35, Keyan Road, Zhunan Town, Miaoli County 350, Taiwan, ROC. E-mail: gting{at}nhri.org.tw
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: flow cytometry • mixed lymphocyte culture • cytotoxic T lymphocyte
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
TGF-ß is known to suppress T cell functions through the induction of peripheral T cell tolerance [18 19 20 21 ]. There are two major schools of thoughts. One holds that TGF-ß acted directly on the antigen-presenting cells (APC) [22 , 23 ] or through the generation of regulatory T cells (TR) to suppress delayed-type hypersensitivity (DTH) T cell response, as represented by the inhibition by CD4+CD25+ TR on the generation of alloreactive cytotoxic T lymphocytes (CTL) [24 ]. The other holds that TGF-ß induces T helper cell type 2 (TH2)-like cytokines, such as interleukin (IL)-10, to suppress the generation of a DTH T cell response, as represented by the anterior chamber-associated immune deviation in the eye [25 ]. In both cases, the effect of TGF-ß is mediated by inducing T cell anergy through the suppression of a TH1 T cell response, the CD4+ cells. Initial reports suggested the involvement of CTL antigen-4 (CTLA-4) [26 ]. A later study clearly showed that TGF-ß mediates its suppressive effect through a pathway, which is distinct from CTLA-4 [27 ]. A major route for TGF-ß suppression appears to be targeting at attenuating the IL-12 receptor ß, which holds the key for activating TH1 cells to trigger cell-mediated immunity [28 29 30 31 ]. In doing so, TGF-ß contributes to the shift toward a TH2 response [25 , 32 ]. The effect of TGF-ß can be mediated indirectly through the production of IL-10 and other cytokines to suppress DTH T cell responses, which helped maintain an inert state at the immune-privileged sites such as the anterior chamber of eye [25 ], vitreous humor [33 ], central nervous system [34 ], testes [35 ], and adrenal cortex [36 ].
It is not clear whether TGF-ß may directly affect the CD8+ population of T cells. Recent reports by Lucas et al. [37 ] and Gorelik and Flavell [38 ] demonstrated that a transgenic mouse model, which specifically overexpresses a dominant-negative TGF-ß II receptor (DNRI) on T cells, develops CD8+ T cell lymphoprolioferative disorder [37 ] and autoimmune diseases [38 ]. These CD8+ T cells are phenotypically "naïve" but express a high level of CD44, a molecule associated with memory, suggesting that TGF-ß may regulate the expansion of CD8+ memory T cells. We adopted the use of an allogeneic mixed lymphocyte culture (MLC) model to further explore the mechanisms regarding the induction of peripheral T cell tolerance by TGF-ß and particularly, how TGF-ß affects the antigen-specific CD8+ memory T cells. Three specific issues were studied: Is the induction of anergy by TGF-ß solely responsible for the induction of peripheral T cell tolerance? Does TGF-ß induce clonal deletion of CD8+ memory T cells? Does TGF-ß restrict the clonal expansion of CD8+ memory T cells in an antigen-specific manner? To simulate the "real-time" events, which may take place in the in vivo microenvironment, we examined the effect of a sustained stimulation by TGF-ß and antigen on the activation and expansion of antigen-specific CD8+ memory cells. Two methods were adopted in this study: induction of CTL responses in MLC and T cell expansion in an adoptive transfer experiment. Flow cytometry was used extensively to analyze the changes in lymphocyte profiles during the course of immune reactions.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents
IL-2 was purchased from Chiron Corp. (Emeryville, CA). The anti-CD3 antibody was produced in hybridoma 145-2C11. Recombinant TGF-ß1 was purchased from R&D Systems (Minneapolis, MN).
MLC
Splenocytes were obtained from BALB/c (H-2d), C57/BL6 (H-2b), or C3H/HeN (H-2k) mice to make single-cell suspension. In most cases, the BALB/c splenocytes were used as responders (R), and 2000 R X-irradiated C57BL/6 splenocytes were used as stimulators (S) at a R/S ratio of 3:1, and they were suspended at 2 x 106 cells/ml in RPMI-1640 medium containing 5% fetal bovine serum, penicillin and streptomycin, and gentamicin. For experiments performed with 2C transgenic mice, splenocytes from these mice were used as responders, and 2000R X-irradiated BALB/c splenocytes were used as stimulators at a R/S of 10:1–3:1. After 5 days culturing, proliferative response and cell-mediated cytotoxicity were determined. For secondary response, the MLC cells were restimulated at Days 9–15 with the appropriate alloantigen at a R/S of 3:1, and proliferative response and cell-mediated cytotoxicity were determined at 2–5 days after the second challenge. In most experiments, TGF-ß was added at the onset of MLC. In long-term culture, the medium was changed on the fifth and 10th day with the addition of TGF-ß in the TGF-ß-treated MLC (MLC-TGF-ß). Except for experiments seen in Figure 1
, TGF-ß was not added after secondary challenge with the appropriate alloantigen. The cells were washed and resuspended in fresh medium at the time of second challenge. During the 5-day culture in primary MLC, the cells were suspended at 2 x 106 cells/ml. After that, the cells were resuspended at 1 x 106 cells/ml in subsequent cultures.
 View larger version (23K): [in a new window] |
Figure 1. Differential suppression by TGF-ß of primary and secondary MLC reactions. (A) Splenocytes were obtained from BALB/c (H-2d) and C57BL/6 (H-2b) mice to make single-cell suspension, and MLC {H-2d anti-H-2b MLC [MLC(d b)]} was set up according to the protocol described in Materials and Methods. For primary MLC, a parallel culture was set up to contain TGF-ß at 1 ng/ml, as indicated. For secondary response, culture medium was changed on Days 5 and 10 with fresh medium. On Day 12, a second stimulation of X-irradiated splenocytes of H-2b haplotype at a R/S of 3:1 was given to the MLC, not exposed to TGF-ß, which was added in a parallel culture, as indicated, after receiving second challenge. (B) For primary response, cell proliferation and CTL responses were determined on Day 5. (C) For secondary response, cell proliferation and CTL responses were determined on the fourth day. EL-4 (H-2b) lymphoma cells were used as the target in the cell-mediated cytotoxicity assay. Cell proliferation is expressed as cpm, and cell-mediated cytotoxicity is expressed as LU per 1 x 106 effectors. The experiments have been repeated five times.
|
Tumor targets
Three tumor lines were maintained in suspension cultures for use as targets in cell-mediated cytotoxicity assays: thymoma EL-4 (H-2b), B cell lymphoma 38C13 (H-2k), and plasmacytoma P815 (H-2d).
Antibodies and flow cytometry analysis
The following antibodies, which were conjugated with fluorescein isothiocynate or phycoerythrin, were purchased from BD PharMingen (San Diego, CA): CD3, CD4, CD25, pan natural killer, B220. Dr. John T. Kung (Institute of Molecular Biology, Academia Sinica) kindly provided the antibody for 2C cells. The stained cells were analyzed in a FACSCaliber (BD Biosciences, San Jose, CA) according to the manufacturer’s protocol. All flow cytometry studies have been repeated more than three times, and the results were reproducible.
Separation of subsets of T cells
Enriched CD4+ or CD8+ T cells were obtained by depletion of major histocompatibility complex (MHC) II+, B220+, and CD8+ or CD4+ cells. Briefly, cultured MLC cells were incubated with M5/114 (MHC class II), RA3-3A1 (B220), and 53-6.7 (CD8) or GK1.5 (CD4; TIB-120, TIB-146 TIB-105, and TIB-207, respectively, American Type Culture Collection, Manassas, VA). Subsequently, these cells were incubated with goat anti-rat immunoglobulin G microbeads (Miltenyi Biotec, Bergisch Galdbach, Germany) and separated by lymphocyte separation separation columns (Miltenyi Biotec) according to the manufacturer’s instruction. The fractions depleted of CD4+or CD8+cells are referred to as CD8+- or CD4+-enriched T cell, respectively.
Adoptive transfer of in vitro-activated T cells for measuring in vivo expansion upon antigen stimulation
The 10- to 15-day cultured MLC cells were kept in 50 U/ml IL-2 overnight. They were then labeled with Cell Tracker Green 5-chloromethylfluorescein diacetate (CMFDA; Molecular Probes, Eugene, OR) and were inoculated intravenously (i.v.) into a normal BALB/c mouse at 1–2 x 107cells/mouse. It was followed by a subcutaneous (s.c.) challenge of 1–2 x 107 allogeneic splenocytes at the right flank (H-2b or H-2k). Three days after challenge, regional lymph node (right inguinal) and contralateral node were obtained. Single-cell preparation from the lymph nodes was analyzed by flow cytometry with FACSCaliber. The experiments have been repeated three times, and the results were reproducible.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Differential suppression of primary and secondary MLC reactions by TGF-ß
We have shown that TGF-ß did induce anergy in primary CTL response, but secondary CTL response in MLC, which had not been pre-exposed to TGF-ß, was resistant to its suppressive effect [40
]. Here, we extended the study to show that although the proliferative and CTL responses in primary MLC were suppressed by TGF-ß (Fig. 1B)
, the secondary MLC response was resistant to TGF-ß treatment (Fig. 1C)
.
Effect of IL-2 on TGF-ß-induced anergy in primary and secondary MLC
Classical anergy is reversed by IL-2 [41
]. In primary response, adding IL-2 at the onset could restore the T cell responses in MLC-TGF-ß, the proliferative (Fig. 2B
) and the CTL responses (Fig. 2C)
. In the secondary response, after prolonged exposure to TGF-ß, adding IL-2 could restore the proliferative response (Fig. 2D)
, but there was only marginal restoration of the CTL response (Fig. 2E)
. Thus, the pattern of secondary CTL response in MLC-TGF-ß did not appear to be classical anergy. Yet, adding IL-2 further increased the secondary CTL response of untreated MLC (Fig. 2E)
.
 View larger version (25K): [in a new window] |
Figure 2. Effect of IL-2 on TGF-ß-induced anergy in primary and secondary MLC. (A) The primary and secondary MLCs were performed as described in Figure 1
. A parallel MLC was set up to contain TGF-ß at 1 ng/ml. Culture medium was changed on Days 5 and 10 with fresh medium and TGF-ß. At the time of secondary challenge (Day 12), the cells were washed with change of fresh medium. In this and subsequent experiments, no TGF-ß was added in cultures after receiving second challenge. This experimental design is a major difference to Figure 1
. IL-2 at 10 U/ml was added in some cultures, as indicated. Cell proliferation (B, D) and CTL response (C, E) were determined on Day 5 for primary response and on 4 days after secondary challenge, respectively. EL-4 was used as the target in the CTL assay. The experiments have been repeated three times.
|
 View larger version (22K): [in a new window] |
Figure 3. Flow cytometry analysis of the activation of CD8+ cells in primary and secondary MLC, which were performed as described in Figure 2
, except that no IL-2 was added. The T cell-enriched population (A) was gated for analyzing the profile of the CD8 population (B–D). (B) BALB/c splenocytes cultured alone for 5 days without alloantigen stimulation (control). (C) The profiles of CD8+ and CD8+CD25+ cells in MLC and MLC-TGF-ß are shown. For primary MLC, flow cytometry analysis was performed on Day 5. In secondary MLC, the 12-day cultured cells received a second challenge of H-2b splenocytes and were analyzed at 2 days after challenge. Figure numbers represent percentage of cells of a particular population. Figures in parentheses represent percentage of total CD8+ cells. The experiments have been repeated four times. (D) A pool of four experiments to show the activation of CD8+ cells in MLC and MLC-TGF-ß, as indicated by the level of increase of CD8+CD25+ cells.
|
b) by IL-2 was clonal-specific for the appropriate allogeneic target (EL-4 of H-2b) and did not produce lysis against syngeneic target P815 (data not shown).
 View larger version (20K): [in a new window] |
Figure 4. Effect of TGF-ß and IL-2 on the induction of secondary CTL responses in MLC and specificity of the CTL response. The secondary MLC [MLC(db)] was performed as described in Figure 2
. (A) The Day 14 MLC-cultured cells, without (IS) or with (TGF-ß-IS) coculturing in TGF-ß, received a second challenge of H-2b splenocytes at a R/S ratio of 3:1. Cytotoxicity was performed 4 days later against EL-4 target. (B) The MLC(db)-cultured cells (IS or TGF-ß-IS) were stimulated with H-2k splenocytes at a R/S ratio of 3:1. IL-2 at 50 U/ml was added in parallel cultures. For comparison, naïve BALB/c splenocytes (NS) were stimulated in the same manner in primary H-2d anti-H-2k MLC [MLC(dk)] without adding IL-2. Cell-mediated cytotoxicity was performed on Day 5 against 38C13 lymphoma target (H-2k). (C) The Day 14 MLC(db)-cultured cells (IS or TGF-ß-IS) were washed and were then stimulated 6 days with IL-2 (50 U/ml) and CD3 (10 ng/ml). Cytotoxicity was performed against P815 plasmacytoma target (H-2d). No TGF-ß was added in these cultures after second challenge or sensitization with a third-party antigen or CD3. Figures in parentheses represent LU per 1 x 106 effecctors. The experiments for A have been repeated five times and three times each for B and C.
|
b), with or without TGF-ß, gave a similar level of CTL response to a third-party alloantigen (H-2k, Fig. 4B
). IL-2 was required for inducing a primary CTL response to the third-party antigen. It is likely a result of diminished CTL precursors for the third-party antigen or the lack of TH cells in the long-term cultures. For comparison, fresh, naïve splenocytes from BALB/c mice could mount an adequate CTL response to H-2k in MLC(dk) without exogenous IL-2. Furthermore, pre-exposure to TGF-ß did not hamper the polyclonal activation by CD3 to generate T killer cells (Fig. 4C)
. These results confirm the clonal specificity of TGF-ß-mediated suppression.
Regulation of CD8+ memory cells by IL-2 and TGF-ß is clonal-specific
Experiments were performed to further characterize the effect of TGF-ß and IL-2 on CD8+ memory cells and to determine whether they are subjected to the regulation by TGF-ß-induced CD4+ suppressor cells (TS; Fig. 5
). After removing the CD4+ cells, the CD8+ cells from MLC-TGF-ß still gave a much lower secondary CTL response, and adding IL-2 only partially restored their CTL response (Fig. 5A)
. For generating the secondary CTL response in CD8+ memory cells, the increase in untreated MLC(db) by IL-2 and the reduction in MLC(db)-TGF-ß by TGF-ß were clonal-specific for H-2b (Fig. 5A)
. TGF-ß pretreatment did not affect the CTL response to a third-party alloantigen (H-2k; Fig. 5B
). It has been shown that generation of CD4+CD25+ TS by TGF-ß was responsible for inducing anergy [24
]. Here, it is shown that the generation of a strong, secondary CTL response by CD8+ memory cells from untreated MLC was not affected by CD4+ cells from MLC-TGF-ß (Fig. 5C)
nor was it affected by adding TGF-ß (data not shown).
 View larger version (21K): [in a new window] |
Figure 5. Effect of TGF-ß and IL-2 on CD8+ memory cells. This series of experiments was performed as described in Figure 4
, except a CD8+-enriched population was used. The CD8+ memory cells were separated from other populations of cells obtained in MLC(db) by a procedure that was described in Materials and Methods. (A) The Day 15 MLC(db)-CD8+ cells (I-CD8 or TGF-ß-I-CD8) were stimulated with a second challenge of H-2b splenocytes at a R/S of 3:1, and CTL response was determined 4 days later against EL-4 as a target in a cell-mediated cytotoxicity assay. IL-2 at 10 U/ml was added in parallel cultures as indicated. (B) The MLC(db) CD8+ cells (I-CD8 or TGF-ß-I-CD8) were stimulated with H-2k splenocytes at a R/S of 3:1, and CTL response was determined on Day 5 against 38C13. IL-2 at 10 U/ml was added in parallel cultures as indicated. (C) Untreated, Day 15 CD8+ memory cells from MLC(db; I-CD8) were mixed with CD4+ cells from parallel MLC(db)-TGF-ß culture (TGF-ß-CD4) at a 2:1 ratio. They received a second challenge of H-2b splenocytes at a R/S of 3:1, and CTL response was determined 4 days later against EL-4 target. No TGF-ß was added after second challenge or sensitization with a third-party antigen. Figures in parentheses represent LU per 1 x 106 effectors. The experiments have been repeated three times.
|
The effect of TGF-ß and IL-2 on cell growth and profile of T cell activation
To determine whether there is clonal deletion of the CD8+ memory cell, it is important to determine how TGF-ß affects cell survival. We found that there was no significant difference in the survival of total cells or T cells in primary and secondary MLC maintained in cultures with or without TGF-ß (Fig. 6A
), indicating that TGF-ß does not induce excessive T cell death. Then, flow cytometry analysis was performed to determine how the profile of lymphocytes was changed in the constant presence of TGF-ß and the effect of IL-2 on modifying the profile. The activated T cells were gated for analysis in this series of study (Fig. 6B)
. Here, we also examined the CD4+CD25+ population, which was found to be lower in primary MLC-TGF-ß (Fig. 6C)
but higher in the long-term culture (Fig. 6D)
. This population contains helper and TS/TR, which could not be distinguished by flow cytometry analysis. When examining the CD8+ population, it was found that the level of CD8+CD25+ T cells was low in the primary and secondary MLC-TGF-ß (Fig. 6C
and 6D)
. Adding IL-2 increases the level of CD8+CD25+ cells in all cultures. In MLC-TGF-ß, IL-2 brought the number of CD8+CD25+ cells to a comparable level (Fig. 6C
, primary MLC) or higher (Fig. 6D
, secondary MLC) than the untreated MLC without IL-2 addition. Despite restoring the proliferation of CD8+CD25+ cells by IL-2 in MLC-TGF-ß, there was no significant restoration of CTL activity (data not shown).
 View larger version (32K): [in a new window] |
Figure 6. Flow cytometry analysis of the effect of TGF-ß and IL-2 on the activation and survival of CD8 and CD4 T cells. The MLC were performed as described in Figure 2
. (A) The total viable cell count was determined by trypan blue dye exclusion. Total T cell count = total viable cell count x percent T cells (CD4++CD8+cells). The MLC cultures started at 2 x 106 cells/ml. They were subcultured on Days 5 and 12 at 1 x 106 cells/ml. XB, X-irradiated B6 splenocytes. (B) In flow cytometry analysis, the activated T cell population was gated for analysis. SSC, Side-scatter; FSC, forward-scatter. (C) Flow cytometry analysis was performed on Day 5 for primary MLC. IL-2 (50 U/ml) was added on Day 2 in parallel cultures. (D) In secondary response, 9 day-cultured MLC cells, with (TGF-ß-IS) or without (IS) coculturing in TGF-ß, were used as responders. IL-2 was added in parallel cultures as indicated. Flow cytometry analysis was performed on Day 4. The final cell count was calculated from combining the information of total viable cell count and the flow cytometry analysis. Cell number = viable cell count x percent gated cells x percent double-positive cells (CD4+CD25+ or CD8+CD25+). The experiments have been repeated three times. A representative experiment is shown in A, and a pool of three experiments is shown in C and D.
|
 View larger version (12K): [in a new window] |
Figure 7. Effect of TGF-ß on the activation of transgenic 2C cells. The MLC cultures were set up with splenocytes from transgenic 2C mice as described in Figure 2
, and second challenge was given on Day 12. Open bars denote MLC without TGF-ß, and solid bars denote MLC containing TGF-ß at 1 ng/ml. (A) Survival of 2C cells. (B) Lymphocyte profile was determined by flow cytometry as described in Figures 3
and 6
. The activation of 2C cells, as indicated by the increase of the 2C+CD25+ population, is shown. The experiments have been repeated four times.
|
Restriction of clonal expansion of the CD8+ memory cells by TGF-ß-adoptive transfer of in vitro-activated T cells for determination of in vivo T cell expansion
To further differentiate the distinction between facilitating cell death and restricting cell expansion and to avoid the limitation by the in vitro environment, we performed the in vivo adoptive transfer experiments. The 12-day MLC-cultured cells, without or with TGF-ß treatment, were labeled with a cell tracker (CMFDA) and were transferred adoptively by i.v. inoculation into the syngeneic BALB/c mice. They were then challenged s.c. with the appropriate alloantigen or third-party alloantigen. Flow cytometry analysis was performed at 3 days after challenge (Fig. 8
). There was a noticeable increase of proliferation of the labeled CD8+ cells in the regional node after adoptive transfer of the MLC(db)-cultured cells upon challenge with the appropriate antigen (H-2b) but not the third-party antigen (H-2k; Fig. 8A
, left panel). There was no detectable CD8+ cell proliferation after adoptive transfer of MLC-TGF-ß-cultured cells after antigen challenge. Cells obtained from the contralateral lymph node failed to show any increase of proliferation of the labeled T cells (data not shown). Likewise, after adoptive transfer of the MLC(dk)-cultured cells, proliferation of CD8+ cells was only seen upon the challenge of H-2k but not H-2b splenocytes (Fig. 8A
, right panel). Figure 8B
shows the results from a pool of three experiments. Again, these findings underscore the clonal specificity in the expansion of CD8+ memory cells and its restriction by TGF-ß in the in vivo host. These results provide direct evidence to show that TGF-ß induces clonal restriction of the expansion of CD8+ memory cells.
 View larger version (23K): [in a new window] |
Figure 8. Adoptive transfer of MLC-cultured cells for determining the clonal specificity of the in vivo expansion of CD8+ cells upon secondary antigen challenge. The MLCs were performed as described in Figure 2
, except that no IL-2 was added. The 12 day-cultured cells from MLC(db; left panels) or MLC(dk; right panels), with or without coculturing in TGF-ß, were incubated overnight in 50 U/ml IL-2. On the following day, the cultured cells were labeled with Cell Tracker Green (CMFDA) and were then given i.v. to normal BALB/c mice at 1–2 x 107 cells per mouse and followed by a second s.c. challenge of the appropriate alloantigen [1–2x107 of C57BL/6 (H-2b) or C3H/HeN (H-2k) splenocytes] at the right flank. (A) At 3 days after challenge, regional lymph node (right inguinal) and contralateral node (left inguinal) were obtained for flow cytometry analysis, and a representative experiment is shown. (B) A pool of three experiments to show the expansion of CD8+ memory cells upon second antigen challenge.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We undertook the effort of searching other possible means, which may be used by TGF-ß to mediate antigen-specific immunosuppression, particularly its effect on the CD8+ memory cells. Two major pathways lead to peripheral T cell tolerance: anergy and clonal deletion. Our initial experiments supported the notion that TGF-ß induces T cell anergy: It suppressed the induction of T cell proliferation and CTL response in primary MLC (Fig. 1) and suppressed the expression of costimulatory factors on dendritic cells [40 ]. However, anergy does not seem to satisfactorily explain the entire phenomenon. IL-2 reverses classical T cell anergy, which was defined by a proliferative response [41 ]. The restoration by IL-2 of the MLC-proliferative responses suppressed by TGF-ß (Fig. 2B and 2D) was consistent with this notion. However, adding IL-2 could only restore the primary and not the secondary CTL response (Fig. 2C and 2D) [40 ]. Although the CD8+ memory cells, which were pre-exposed to TGF-ß during sensitization, show a sustained state of unresponsiveness (Figs. 1 2 4 5 6 8) , the untreated CD8+ memory cells are resistant to the suppression by TGF-ß or TGF-ß-induced TS to mount a secondary CTL response (Figs. 1C and 5C) , thus indicating that TGF-ß cannot induce anergy in these memory cells.
We further carried out experiments to explore the issue of clonal deletion versus restriction of clonal expansion, and the emphasis was on the CD8+ memory cells. The information gathered from the subsequent experiments lent support to these possibilities. The increase of secondary CTL response by IL-2 and the suppression of it by a sustained antigen and TGF-ß stimulation are clonal-specific and are directed at the CTL clones against the immunizing antigen (Figs. 4A and 5A) ; TGF-ß does not affect the CTL response to a third-party antigen (Figs. 4B and 5B) or polyclonal activation of CD8+ killer T cells (Fig. 4C) ; and flow cytometry analysis showed that the activated CD8+ cells could be reduced selectively by TGF-ß (Fig. 3C and 3D) , whereas the survival of total cells or T cells in long-term cultures was not affected by TGF-ß (Figs. 6A and 7A) . As we failed to detect any increase of apoptosis in the CD8+ memory cells derived from MLC-TGF-ß before and after secondary antigen challenge, the possibility of inducing clonal deletion by TGF-ß has to be ruled out.
We then focus on whether TGF-ß restricts the clonal expansion of CD8+ memory cells. Lucas et al. [37 , 52 ] showed that in the TGF-ß DNR II transgene model, absence of TGF-ß regulation leads to excessive proliferation of CD8+CD44hi cells, which resulted in lymphoproliferative disorders [37 ] and leukemia [52 ], suggesting that TGF-ß may down-regulate the expansion of CD8+ memory cells. In the absence of direct evidence for inducing apoptosis, our findings are consistent with restriction of the expansion of antigen-specific CD8+ memory cells by TGF-ß (Figs. 3 4 5 6 7) . The result obtained by adoptive transfer experiments provides, probably for the first time, the direct evidence to confirm this notion to be the fact (Fig. 8) . Upon antigen re-stimulation, although the CD8+ memory cells, which had not been exposed to TGF-ß, showed rapid expansion, the TGF-ß-treated CD8+ memory cells were prevented from entering the cell cycle to proliferate and thus were prevented from expansion. The restriction is clonal-specific.
It should also be pointed out that the effect of TGF-ß on the negative regulation of the expansion of CD8+ memory cells appears to be different from that of naïve CD8+ cells. We showed that naïve CD8+ cells would be anergized when they encounter the appropriate antigen in the presence of TGF-ß (Fig. 1B) [40 ]. Restriction of their expansion may also play a role [51 ], depending on the strength of the antigen [50 ]. However, in the absence of the appropriate antigen, merely TGF-ß exposure does not have a permanent effect on their expansion. The TGF-ß-pre-exposed, naïve CD8+ cells retain the ability to respond properly when they encounter the appropriate antigen for the first time in the absence of TGF-ß. This is shown by the induction of proper CTL response to the stimulation of a third-party antigen in TGF-ß-pre-exposed CD8+ cells (Figs. 4B and 5B) . This finding is in sharp contrast to the effect of TGF-ß on CD8+ memory cells. After converting the naïve CD8+ cells into memory cells by antigen activation in the absence of TGF-ß, they become resistant to the suppression by TGF-ß or the TGF-ß-induced TS(Figs. 1C and 5C) . However, if naïve CD8+ cells are activated in the presence of TGF-ß, then they are not only anergized, but the memory cells they have become are converted permanently into an inert state, which makes them unable to enter the cell cycle to proliferate (Fig. 8) and thus prevents them from expansion when they re-encounter the appropriate antigen. It has been shown that the frequency in antigen-specific CD8+ memory T cells was similar whether or not TGF-ß was present [53 ]. We also found that naïve CD8+ cells could be activated in the presence of TGF-ß in primary sensitization, as shown by a reduced but still sizable CD8+CD25+ population in primary MLC-TGF-ß (Figs. 3 6 7) . It is when they become memory cells that they fail to expand upon antigen restimulation (Figs. 3 6 7 8) . Hence, the restriction of the CD8+ cell expansion by TGF-ß is temporal for the naïve cells, but it is permanent for the memory cells.
However, there are other possibilities that may account for the reduced CTL activity in the TGF-ß-treated CD8+ memory cells, i.e., attenuation of the effector function. A recent report showed that the acquisition and expression of effector function of tumor-specific human memory CD8+ T cells are attenuated by TGF-ß [53 ]. We also found that TGF-ß induced defective perforin production in CD8+ killer cells, which impaired their cytolyic function [40 ]. These mechanisms may explain the dilemma in our observation that despite the reversal of the TGF-ß-induced reduction of CD8+CD25+ cells by IL-2 and the fact that they were raised to a level that was comparable with or higher than the MLC without TGF-ß (Fig. 6C and 6D) , IL-2 could only restore the proliferative response (Fig. 2B and 2D) but not the secondary CTL response (Figs. 2E 4A , and 5A) . In addition, the magnitude of increase was not proportional to the increase of CD8+CD25+ cells (Fig. 6D) , indicating defective CTL.
It appears that through multistep maneuvers, TGF-ß modulates the immune responses at different stages of activation and differentiation. Inducing anergy at the early stage of immune response is the first step used by TGF-ß to induce peripheral T cell tolerance. To sustain the tolerant state, TGF-ß converts the CD8+ memory cells into a permanent inert state, which restricts them from expansion upon antigen restimulation. For the CD8+ memory cells that escape its influence, TGF-ß attenuates their effector function, making them unable to attack the targets. Thus, TGF-ß production by tumors poses a serious challenge to the host for mounting an effective defense or achieving an effective cancer immunotherapy. Conversely, using TGF-ß may help to combat autoimmune diseases. Thus, TGF-ß may prove to be an important therapeutic target for treating cancer and autoimmune disease.
Received August 23, 2005; revised December 15, 2005; accepted January 1, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
} have opposing effects, while TGF-{ß} positively regulates its own production J. Immunol. 160,5719-5728 to elaborate TGF-ß-based suppression J. Immunol. 174,7625-7632This article has been cited by other articles:
 |
|
 |
|
  S.-J. Liu, J.-P. Tsai, C.-R. Shen, Y.-P. Sher, C.-L. Hsieh, Y.-C. Yeh, A.-H. Chou, S.-R. Chang, K.-N. Hsiao, F.-W. Yu, et al. Induction of a distinct CD8 Tnc17 subset by transforming growth factor-{beta} and interleukin-6 J. Leukoc. Biol., August 1, 2007; 82(2): 354 - 360. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
