| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Division of Experimental Oncology 1,*
Centro di
Riferimento Oncologico, Aviano, Italy; and the Department of
Pathology,
Belluno City Hospital,
Belluno, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells are the major effectors of cytolytic responses, and the molecular mechanisms by which they kill target cells have been recently clarified.8-10 Both classes of cytotoxic effectors induce apoptosis in target cells by a granule exocytosis pathway; these granules carry both a pore-forming protein, such as perforin,11 and serine-esterases, such as granzymes.12 Differences were found in the expression pattern of these proteins between NK and CD8+ CTLs, in that the former constitutively express both perforin and granzyme B, whereas the latter produce perforin and granzyme-B (GrB) only on activation induced by antigen recognition.8 Other granule-associated proteins are expressed by CD8+ CTLs: among them, T-cell-restricted intracellular antigen 1(TIA-1) is expressed by cytotoxic effectors independently of their activation status, thus being a reliable marker for CD8+ lymphocytes with cytotoxic potential.13,14 Antibodies recognizing these markers have recently become available, thus allowing the identification of activated cytotoxic effectors in situ.
The aim of the present study was to better define the nature of reactive T cells infiltrating primary gastric MALT lymphomas and, particularly, to shed light on the role of cytotoxic immune responses in these disorders. To this end, we analyzed the presence, activation status, and spatial distribution of cytotoxic effectors in 13 low- and 17 high-grade gastric MALT lymphomas. Moreover, possible relationships between the number of activated cytotoxic effectors and the extent of lymphoma cell apoptosis were also investigated by an in situ end-labeling method.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty formalin-fixed, paraffin-embedded surgical specimens of gastric MALT lymphoma were retrieved from the files of the Division of Pathology, Belluno City Hospital. According to the criteria of Isaacson,15 recently incorporated into the REAL classification and modified as in de Jong et al, 16 13 cases were diagnosed as low-grade and 17 as high-grade MALT lymphomas.
Immunohistochemical Determinations
Two-micron paraffin sections were cut and mounted on poly-L-lysine treated slides, deparaffined in xylene, rehydrated in graded alcohols, washed three times in phosphate-buffered saline (PBS), and microwaved for two cycles of 8 and 5 minutes at 700 W in 10 mmol/L citrate buffer, pH 6.0, or 0.1 mmol/L EGTA. After cooling to room temperature (15–30 minutes), sections were further washed and treated with 3% H2O2 in methanol for endogenous peroxidase activity block. After primary antibody incubations, samples were developed with standard streptavidin-biotin-peroxidase methods and 3',5'-diaminobenzidine (DAB) as a substrate.
Double labelings for CD8/TIA-1, GrB/CD20, GrB/CD8, and GrB/TIA-1 were performed on selected specimens. Labelings with the first and second primary antibodies were sequentially performed as described above, except for the use of streptavidin-alkaline phosphatase staining with Fast Red as a chromogen in the second reaction. The following primary antibodies were used: anti-CD3 (polyclonal, 1:1000), anti-CD8 (clone cd8/144b, 1:25), and anti-CD20 (clone L26) from DAKO (Glostrup, Denmark); anti-CD4 (clone IF6, 1:100), from Novocastra (Newcastle, UK); anti-granzyme B (Clone GrB7, 1:20) from Monosan (Leiden, The Netherlands); anti-perforin (clone KM585 P1–8, 1:1000) from Kamiya (Japan); anti-TIA-1 (1:1000) from Coulter Corp. (Hialeah, FL); and anti-CD56 (clone 123C3.D5, 1:100) from Neomarker (Fremont, CA). Intratumoral CD3+, CD4+, CD8+, TIA-1+, perforin+, GrB+ and CD56+ cells were evaluated by light microscopy with the support of an eyepiece grid. A minimum of 2000 cells (normal and neoplastic) were counted for each single determination and reported as the percentage of positive cells in total mononuclear cells. For GrB/CD20 double immunostaining, a minimum of 2000 CD20+ and the correspondent GrB+ cells were counted for each case, and expressed as a GrB+/CD20+ ratio. Evaluation was carried out on microscopic fields carefully selected in non-necrotic areas of the deeper portion of the tumor to avoid superficial, nonspecific, inflammatory infiltration.
Detection of Apoptosis-Associated DNA Fragmentation in Situ
The terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay was used to identify DNA fragmentation in situ and was based on the protocol derived from Gavrieli,17 as modified by Migheli.18 All reagents for the TUNEL reaction were obtained from Boehringer (Mannheim, Germany). Briefly, rehydrated sections were digested with 20 µg/ml proteinase K (Sigma) in Tris-EDTA (TE) for 20 minutes at room temperature. After washing in TE and quenching of peroxidase activity with 3% H2O2 in distilled water for 10 minutes, slides were immersed in TdT buffer (25 mmol/L Tris-HCl, 0.2 mmol/L Na-cacodylate, 2.5 mmol/L cobalt chloride, pH 6.6) for 5 minutes and then incubated for 2 hours at 37°C in TdT mixture composed of enzyme solution (TdT) and label solution (fluorescein-dUTP), 1:9 and diluted 1:4 with TdT buffer. After washing with 2x standard saline citrate and Tris-buffered saline (TBS), slides were immersed in 2% bovine serum albumin in TBS-Triton for 15 minutes at room temperature, followed by incubation with Converter peroxidase-labeled antifluorescein antiserum for 30 minutes at room temperature. After washing, peroxidase activity was visualized with a DAB color reaction and slides were counterstained with Harris' hematoxylin, dehydrated, and mounted. For all but one high-grade case, the TUNEL labeling index was determined as the percentage of positive cells in total mononuclear cell number. Double labelings for TUNEL/GrB or TUNEL/CD20 were performed on selected high-grade MALT lymphoma specimens by sequential TUNEL, with DAB as chromogen giving a brown reaction product, followed by immuno-alkaline phosphatase staining with Fast Red as substrate.
Statistical Evaluation
Results are expressed as the mean ± SD percentages of positive cells on total cell count for a given marker. A nonparametric Mann-Whitney U test was used to evaluate the differences in the distribution of positive cell count between low- and high-grade MALT lymphomas. A Spearman rank order correlation test was used for TUNEL index and CD20/GrB ratio or GrB cell count comparison.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As shown in Figure 1A
, the
number of infiltrating CD3+ T lymphocytes was similarly distributed in
low- and high-grade lymphomas, respectively (18.4 ± 6.1%
versus 22.4 ± 6.7%; P = NS).
Analogously, the number of CD4+ (7.9 ± 2.1% versus 9.6 ±
4.2%; P = NS) and CD8+ lymphocytes (10.4 ± 4.4%
versus 12.8 ± 5.1%; P = NS) was
comparable in low- and high-grade lymphomas (Figure 1B)
. Consistently,
the CD4:CD8 ratio was not significantly different between low- and
high-grade cases (0.9 ± 0.4 versus 0.8 ± 0.4;
P = NS, Figure 1C
). Immunostaining for CD56 showed an
inverse relationship between histopathological grade and the proportion
of infiltrating CD56+ NK cells. In fact, as shown in Figure 1D
,
low-grade lymphomas carried significantly higher numbers of
CD56+ NK cells than high-grade cases (2.1 ± 0.7%
versus 1.2 ± 0.6%; P = 0.001).
|
As a first step, we evaluated the prevalence of cytotoxic
cell precursors as indicated by TIA-1 immunostaining (Figure 2, A and B)
. High-grade lymphomas carried
a significantly higher proportion of TIA-1+ cells than low-grade cases
(11.6 ± 3.9% versus 7.8 ± 1.5%; P =
0.004, Figure 3A
). CD8/TIA-1 double
immunostaining demonstrated that cytotoxic precursors were mostly CD8+
T cells (not shown). Of note, also the effectors expressing cytolytic
activation markers were markedly more represented in high- than in
low-grade lymphomas. In fact, the absolute number of both perforin+ and
GrB+ cells detected in high-grade tumors nearly doubled those observed
in low-grade cases, respectively (8.7 ± 4.3% versus
4.0 ± 1.9%; P = 0.001 for perforin, and 8.7
± 4.5% versus 3.0 ± 1.4%, P <
0.0001 for GrB, Figures 2C, 2D, 3B, and 3C
). CD8/GrB double
immunostaining confirmed that activated cytotoxic effectors were
prevalently CD8+ T lymphocytes (Figure 2E)
. Consistently, both
perforin/CD8 (0.7 ± 0.3 versus 0.4 ± 0.2;
P = 0.004) and GrB/CD8 (0.7 ± 0.3
versus 0.3 ± 0.2, P = 0.001) ratios
were markedly increased in high-grade lymphomas (Figure 3D)
,
indicating a more pronounced activation status of CTLs in this group of
tumors.
|
|
. The
analysis confirmed that, compared to low-grade cases, high-grade
lymphomas carried a significantly higher content (about ninefold) of
activated CTLs relative to the number of CD20+
lymphoma B cells (0.081 ± 0.076 versus 0.009 ±
0.011; P < 0.0001; Figure 4A
). Of note, the large majority of GrB+
effectors was in close contact with CD20+
lymphoma cells, prevalently showing morphological features of large
transformed blast cells (Figure 2F)
.
|
Whereas in the majority of CD8+ T lymphocytes infiltrating
low-grade lymphomas GrB+ granules were dispersed throughout the
cytoplasm, a large proportion of GrB+ cells from high-grade cases
showed a polar distribution of cytotoxic granules (Figure 2F)
, a
feature that characterizes cytotoxic effectors actively involved in
killing their target cells.10,12
Moreover, activated CTLs
from high-grade lymphomas were prevalent in close proximity to
structures morphologically identifiable as apoptotic bodies. To assess
whether the presence of activated cytotoxic effectors correlated with
the apoptotic rates, 13 low- and 16 high-grade MALT lymphomas were
investigated for the prevalence and distribution pattern of apoptotic
cells by the in situ end-labeling (TUNEL) method (Figure 2G)
. As shown in Figure 4B
, markedly higher apoptotic rates were found
in high-grade compared to low-grade tumors (1.1 ± 0.9%
versus 5.3 ± 1.9%; P < 0.0001).
Apoptotic death occurred prevalently in neoplastic B cells as shown by
the cytomorphologic features of TUNEL+ cells and by TUNEL/CD20 double
immunostaining (not shown). Furthermore, both the absolute number of
GrB+ cells and the GrB+/CD20+ ratios strongly correlated with the
number of TUNEL-labeled cells (rs = 0.71,
P < 0.0001 and rs = 0.80,
P < 0.0001, respectively; Figure 4, C and D
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Interestingly, our findings are consistent with a more efficient recruitment and a stronger activation of CTLs in high-grade than in low-grade gastric MALT lymphomas. In particular, these cases showed a GrB+/CD20+ ratio about ninefold that of low-grade lymphomas, as directly evaluated by GrB/CD20 double immunostaining. Furthermore, high-grade tumors showed a greater prevalence of apoptotic death affecting neoplastic B cells than low-grade lymphomas, similar to previous findings.19,20 These differences in activation and recruitment of CTLs observed between high- and low-grade lymphomas could be related to the progressive acquisition of enhanced immunogenic properties by malignant lymphoma cells. It is tempting to speculate that this phenomenon might be related at least in part to the increasing load of genetic changes associated with the evolution to high-grade lymphomas, ie, p53, TGF-ß-RII, and c-myc alterations.21-24 Consistently, high numbers of activated cytotoxic T cells were also detected by other investigators in nodal high-grade B-cell non-Hodgkin's lymphomas.25-27
Although fewer in number than in high-grade cases, a proportion of T cells phenotypically identifiable as cytotoxic precursors (TIA-1+) was also detected in the majority of low-grade lymphomas. In these cases, however, only a small number of T lymphocytes with cytotoxic potential expressed GrB or perforin, indicating a globally lower level of activation in vivo. It is likely that the peculiar microenvironment characterizing this subset of cases may contribute to this phenomenon. In fact, reactive T cell subpopulations infiltrating low-grade gastric MALT lymphomas were shown mainly to provide survival- and/or growth-promoting signals to the expanding B-cell clones.5,7 In particular, previous in vitro findings suggested a likely crucial role for local Th2 signaling in the early phases of MALT lymphomagenesis,6 a setting that is probably not compatible with the induction of strong cytotoxic responses. The possible role of locally enhanced production of IL-10 and other Th2-type cytokines by H. pylori,28,29 also found in gastric MALT lymphomas,30 may be relevant in this respect. Moreover, the lower apoptotic indices detected in low-grade cases may be related to the peculiar pathobiological interactions presumably occurring between clonal B cells and reactive T lymphocytes, favoring the rescue of low-grade lymphoma cells from programmed cell death. In fact, it has recently been reported that signaling through CD40 was able to promote the survival of B lymphocytes from low-grade MALT lymphomas in vitro.6 Besides promoting the proliferation and differentiation of clonal B cells, this microenvironmental setting may also inhibit cytotoxic immune responses.31 Accordingly, Th1 responses and CD8+ CTL activity were strongly inhibited in mice with persistent gastric H. pylori infection.32 Moreover, loss of H. pylori dependence for growth occurring during the evolution to high-grade histology may be associated with microenvironmental changes (ie, a shift toward a Th1 response) that may allow the occurrence of CTL immune responses.
The significantly different distribution of CD56+ NK cells observed between low- and high-grade MALT lymphomas is intriguing. The increased content of these cells in low-grade cases may be the consequence of microenvironmental stimuli, such as IL-10, that are able to enhance the proliferation of NK cells. It is worth considering, however, that recent evidence indicates these cells may have immunoregulatory functions, including the ability to modulate both humoral and cell-mediated immune responses.33,34 In particular, the possibility that NK cells may contribute to limit the extent of autoreactive responses elicited by H. pylori in low-grade lymphomas constitutes an attractive hypothesis.34 Whether the increased content of NK cells detected in low-grade MALT lymphomas has a role in the pathogenesis of these disorders remains, however, to be elucidated.
Although the extent of cytotoxic responses specifically directed against lymphoma cells was not directly investigated in the present study, our findings have implications of potential clinical relevance. In fact, they are consistent with the possibility that most patients with high-grade gastric MALT lymphomas may mount antitumor cytotoxic immune responses, which, however, cannot eliminate lymphoma cells completely. Therefore, therapeutic strategies including treatments aimed at potentiating host cytotoxic immune responses could be effective in the clinical management of these lymphomas. With respect to low-grade lymphomas, the limited extent of cytotoxic responses observed in the majority of these cases is probably related to the peculiar pathobiological setting that characterizes these disorders. On these grounds, patients with low-grade lymphomas may probably benefit from treatment modalities that, besides removing the triggering antigenic stimulus (ie, H. pylori eradication), can directly inhibit the pathogenically relevant interactions occurring between clonal B cells and reactive T lymphocytes.
|
Footnotes |
|---|
Supported in part by Associazione Italiana per la Ricerca sul Cancro and Ricerca Sanitaria Finalizzata Regione Veneto.
Accepted for publication June 17, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
secretion. Gut 1997, 40:739-744This article has been cited by other articles:
 |
|
 |
|
  T. Ishikawa, T. Fujita, Y. Suzuki, S. Okabe, Y. Yuasa, T. Iwai, and Y. Kawakami Tumor-specific Immunological Recognition of Frameshift-mutated Peptides in Colon Cancer with Microsatellite Instability Cancer Res., September 1, 2003; 63(17): 5564 - 5572. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Morgner, S. Miehlke, W. Fischbach, W. Schmitt, H. Muller-Hermelink, A. Greiner, C. Thiede, J. Schetelig, A. Neubauer, M. Stolte, et al. Complete Remission of Primary High-Grade B-Cell Gastric Lymphoma After Cure of Helicobacter pylori Infection J. Clin. Oncol., April 1, 2001; 19(7): 2041 - 2048. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
