**2. NK cells**

The role in cytotoxicity against target cells in anti-tumor immunity is played by two populations of cells, natural killer (NK) cells and cytotoxic T lymphocytes (CTLs). NK cells have a cytotoxicity for targets which they are ready to kill without prior stimulation, whereas this readiness is absent from the killing activity of CTLs and is induced by antigen stimulation. Therefore, although the cytotoxicity of NK cells has no antigen specificity, it is thought to contribute widely to the early deletion of unhealthy cells such as virus-infected cells or transformed cells. On the other hand, it takes time to induce the differentiation of naïve CD8+ T cells to CTLs, but they can recognize target cells precisely and injure them effectively. Imai et al. examined whether differences between individuals in regard to natural immunological host defense, i.e., NK cytotoxicity, can predict the future development of cancer. They reported that medium and high cytotoxic activity of peripheral blood lymphocytes is associated with reduced cancer risk, whereas low activity is associated with increased cancer risk, and these findings suggest a role for natural immunological host defense mechanisms against cancer (Imai et al., 2000). Therefore, we examined the effect of exposure to asbestos on cytotoxicity of NK cells and alteration in cytotoxicity of NK cells in patients with malignant mesothelioma.

#### **2.1. Mechanism of cytotoxicity in NK cells**

34 Malignant Mesothelioma

The present diagnosis for malignant mesothelioma is based on X-ray and CT image analyses, as well as pathohistological analysis. However, diagnosis using these procedures is sometimes accompanied with a risk of radiation exposure or invasiveness, and results are not regarded equally among doctors because they have learned these analyses separately and it is not easy to master all of these methods. In addition, although it takes a long period to develop malignant mesothelioma after exposure to asbestos, the mean life expectancy of patients with malignant mesothelioma is short, and people exposed to asbestos need a safer analysis that can be used frequently in a year in order to detect malignant mesothelioma as early as possible. Recently, the use of products derived from mesothelioma cells for diagnosis has been proposed, including megakaryocyte potentiating factor (MPF) and mesothelin (Creaney et al., 2007; Onda et al., 2006). However, those products might not be observed in the blood that early because they will appear exactly after the development of mesothelioma, and it may take some time for those products to transfer from the pleural cavity into the blood stream. Many people exposed to asbestos worry about the development of malignant mesothelioma; however, there is little predictive information regarding the onset of mesothelioma. Therefore, there is a need to find a new parameter or method useful for the early diagnosis of malignant mesothelioma (Fig. 3). If some characteristic alteration of immune function is caused by exposure to asbestos that is also found in patients with malignant mesothelioma, which can be measured by checking lymphocytes or other cells in peripheral blood, the analysis for that alteration might

On the basis of these ideas, we started to investigate the immunological effect of exposure to asbestos and immunological alteration in patients with malignant mesothelioma. In this chapter, we show the results obtained from these studies concerning the effect of asbestos on anti-tumor immunity, focusing on NK and Th cells, and discuss the immune-suppressive effect of asbestos and the possible application of our results for the early diagnosis of

The role in cytotoxicity against target cells in anti-tumor immunity is played by two populations of cells, natural killer (NK) cells and cytotoxic T lymphocytes (CTLs). NK cells have a cytotoxicity for targets which they are ready to kill without prior stimulation, whereas this readiness is absent from the killing activity of CTLs and is induced by antigen stimulation. Therefore, although the cytotoxicity of NK cells has no antigen specificity, it is thought to contribute widely to the early deletion of unhealthy cells such as virus-infected cells or transformed cells. On the other hand, it takes time to induce the differentiation of naïve CD8+ T cells to CTLs, but they can recognize target cells precisely and injure them effectively. Imai et al. examined whether differences between individuals in regard to natural immunological host defense, i.e., NK cytotoxicity, can predict the future development of cancer. They reported that medium and high cytotoxic activity of peripheral blood lymphocytes is associated with reduced cancer risk, whereas low activity is associated with increased cancer risk, and these findings suggest a role for natural immunological host

contribute to the early detection of malignant mesothelioma.

malignant mesothelioma.

**2. NK cells** 

The mechanism of cytotoxicity in NK cells and CTLs can be separated into two parts. The role of one part is to recognize target cells, which is followed by transduction of the stimulation signal into the cytosol, while the other part acts to kill target cells. In the killing mechanism, both NK cells and CTLs use the common molecules perforin and granzymes. Perforin- and granzyme-induced apoptosis is the main pathway used by cytotoxic lymphocytes to eliminate virus-infected or transformed cells (Trapani & Smyth, 2002). Perforin and granzymes are produced and accumulate in the cytotoxic granules of NK cells. Once NK cells are optimally stimulated, perforin and granzymes are released into the gap of the immune synapse by degranulation and act on target cells to induce apoptosis. Perforin is the protein that can disrupt the cellular membrane and create a pore in the membrane of the target cell. Granzyme is a family of structurally related serine proteases, which enters target cells through the pore made by perforin, and induces apoptosis of the target cells. The second pathway to kill targets is Fas-mediated apoptosis, induced by ligation of the Fas ligand (FasL) expressed on NK cells or CTLs with Fas on target cells. In addition to these two pathways, tumor-necrosis factor-related apoptosis-inducing ligand (TRAIL) is also known to control the growth and metastasis of tumors (Smyth et al., 2001; Takeda et al., 2001). These killing mechanisms are followed by recognition of target cells by NK cells. In contrast to T cell, which utilizes the T cell receptor (TCR) to recognize targets, NK cells utilize various kinds of receptors for target recognition. These receptors could be of either type: inhibitory or activating. In the next section, we explain the significance of the expression levels of NK cell receptors for cytolytic activity.

#### **2.2. NK cell receptors**

NK cells do not have clonal diversity like T cells, which include many repertoires, rearrangements, and somatic mutations of TCRs. However, NK cells can recognize various target cells using various kinds of receptors expressed on the cell surface of NK cells (Moretta, L. & Moretta, A., 2004; Yokoyama & Plougastel, 2003). Some NK cell receptors, the ligands of which are human leukocyte antigen (HLA) molecules, genetically differing among individuals and recognized by T cells with T cell receptor (TCR) to find abnormal cells, have the role of transducing an inhibitory signal. Those inhibitory receptors include a KIR family of receptors and heterodimer of NKG2A and CD94. The inhibitory signals derived from those receptors contribute to prohibition of cytotoxicity against normal self cells. In contrast, several other receptors transduce an activation signal after ligation with their respective ligands to induce cytotoxicity against abnormal target cells (Fig. 4). NKG2D is the best characterized activating receptor expressed on NK cells. NKG2D is a receptor belonging to the same group as NKG2A, NKG2 family, characterized by a lectin-like domain, but can transduce an activation signal unlike NKG2A. The signaling lymphocytic

activation molecule (SLAM) family is another group of activating receptors expressed on NK cells. A representative receptor of the SLAM family is 2B4, which induces cytotoxicity by stimulation with the natural ligand, CD48, or the antibody to 2B4 (Endt et al., 2007; Garni-Wagner et al., 1993; Valiante & Trinchieri, 1993). Moreover, natural cytotoxicity receptors (NCRs) make a family of receptors that includes NKp46, NKp44 and NKp30, which play a major role in the NK-mediated killing of most tumor cells (Moretta, A. et al., 2001; Sivori et al., 1999). These activating receptors transduce the stimulation signal leading to the phosphorylation of c-Jun N-terminal kinases (JNKs) and extracellular signal-regulated kinase (ERK), which cause polarization of the microtubule organizing center (MTOC) and cytotoxic granules followed by release of perforin and granzymes, producing degranulation (Chen et al., 2006; Chen et al., 2007). Thus, the various kinds of receptors expressed on NK cells control induction of cytotoxicity for target cells, and alteration in expression of these receptors is thought to affect the strength of the stimulation signal to induce cytotoxicity of NK cells. In addition, if NK cells exposed to asbestos show some characteristic alteration in expression of NK cell receptors, this alteration might be used as a possible marker of asbestos exposure-related immune alteration. Therefore, we planned to study the effect of asbestos exposure on NK cells, focusing on the expression level of NK cell receptors, as well as investigate the cytotoxicity of NK cells.

Effect of Asbestos on Anti-Tumor Immunity

and Immunological Alteration in Patients with Malignant Mesothelioma 37

Yodoi and was used for this experiment. Before starting the culture, the effect of exposure to chrysotile B (CB) on the survival and growth of YT-A1 cells was examined in order to avoid a high dose of exposure to CB which induces intense apoptosis and inhibition of cell growth. The apoptosis of YT-A1 was not induced upon exposure to CB at concentrations from 5 to 100 µg/ml. However, cell growth was inhibited upon exposure to CB at a concentration greater than 50 µg/ml, whereas it was not inhibited below 20 µg/ml. Therefore, the dose of CB used was defined as 5 µg/ml, representing the concentration at which YT-A1 cells were cultured with CB, named YT-CB5, and the rest of cells was cultured in the original condition of media without CB, named YT-Org. The cultures were exposed for over five months and then periodically examined for the effect of asbestos exposure on the cytotoxicity of NK cells. The cells were assayed for cytotoxicity against K562 cells and expression levels of various kinds of NK cell receptors and other molecules using flow cytometry. There were no differences in cytotoxicity or expression levels of cell-surface molecules between YT-Org and YT-CB5 for one month after exposure to CB. However, after around 5 months of such exposure, YT-CB5 showed a clear decrease in cytotoxicity compared with YT-Org. In addition, decreases in expression levels of NKG2D and 2B4, but not of NKp46, were also found in YT-CB5 (Nishimura et al., 2009b). The expression level of CD56 and CD16, NK-cell marker and low-affinity Fc receptor, respectively, altered slightly. NKG2A, which makes a suppressive receptor with CD94, did not increase. In addition, YT-CB5 showed almost a zero level of granzyme A and exhibited a significant decrease in perforin, but the degree of decrease of perforin was not large and YT-CB5 did not show a significant decrease in granzyme B. Although 2B4 is not related to cytotoxicity against K562 cells in contrast to NKG2D and NKp46, YT-CB5 also showed the decrease in cytotoxicity against P815 cells treated with antibodies to 2B4, a cytotoxicity mediated by 2B4. These results supported the supposition that the decrease in NKG2D and 2B4 might cause impairment in induction of cytotoxicity by reducing signal transduction downstream of those receptors. Therefore, the degranulation stimulated by antibodies to NKG2D and 2B4 was examined in YT-CB5 by flow cytometry, which can estimate degranulation by measuring an increase in cell-surface CD107a accompanied with secretion of cytotoxic granules. YT-CB5 showed decreases in degranulation stimulated via NKG2D and 2B4, as observed under stimulation with beadbound and plate-bound antibodies, respectively. In addition, YT-CB5 also showed a decrease in phosphorylation of ERK1/2 stimulated with K562 cells as well as antibodies to NKG2D, but not with antibodies to 2B4 (Nishimura et al., 2009a). Moreover, we examined cytotoxicity against K562 cells, expression levels of NKG2D, 2B4 and NKp46, and the phosphorylation level of ERK1/2 stimulated with their respective antibodies in peripheral blood (PB-) NK cells purified from the blood of healthy volunteers, and compared results among the volunteers. PB-NK cells with a high expression of NKG2D, derived from one individual, showed high cytotoxicity and high phosphorylation of ERK1/2, whereas both cytotoxicity and ERK phosphorylation were low in PB-NK cells with low NKG2D, derived from another individual. In a similar way, the relationship among the cell-surface level of NKp46, phosphorylation level of ERK1/2, and cytotoxicity was also investigated and confirmed in PB-NK cells of healthy volunteers. Thus, we demonstrated that exposure to

**Figure 4.** NK cell-activating receptors and the machinery of degranulation caused by stimulation with these receptors. The ligation of NK cell-activating receptors with their respective ligands induce phosphorylation of JNK and ERK, leading to polarization of the microtubule organizing center (MTOC) and granules, where perforin and granzymes are released and induce apoptosis of target cells.

#### **2.3. Cytotoxicity of human NK cell line cultured with asbestos for a long period**

We initially started by making the sub-line of an NK cell line by culturing cells exposed to asbestos for a long period. The human NK cell line of YT-A1 was kindly provided by Dr. Yodoi and was used for this experiment. Before starting the culture, the effect of exposure to chrysotile B (CB) on the survival and growth of YT-A1 cells was examined in order to avoid a high dose of exposure to CB which induces intense apoptosis and inhibition of cell growth. The apoptosis of YT-A1 was not induced upon exposure to CB at concentrations from 5 to 100 µg/ml. However, cell growth was inhibited upon exposure to CB at a concentration greater than 50 µg/ml, whereas it was not inhibited below 20 µg/ml. Therefore, the dose of CB used was defined as 5 µg/ml, representing the concentration at which YT-A1 cells were cultured with CB, named YT-CB5, and the rest of cells was cultured in the original condition of media without CB, named YT-Org. The cultures were exposed for over five months and then periodically examined for the effect of asbestos exposure on the cytotoxicity of NK cells. The cells were assayed for cytotoxicity against K562 cells and expression levels of various kinds of NK cell receptors and other molecules using flow cytometry. There were no differences in cytotoxicity or expression levels of cell-surface molecules between YT-Org and YT-CB5 for one month after exposure to CB. However, after around 5 months of such exposure, YT-CB5 showed a clear decrease in cytotoxicity compared with YT-Org. In addition, decreases in expression levels of NKG2D and 2B4, but not of NKp46, were also found in YT-CB5 (Nishimura et al., 2009b). The expression level of CD56 and CD16, NK-cell marker and low-affinity Fc receptor, respectively, altered slightly. NKG2A, which makes a suppressive receptor with CD94, did not increase. In addition, YT-CB5 showed almost a zero level of granzyme A and exhibited a significant decrease in perforin, but the degree of decrease of perforin was not large and YT-CB5 did not show a significant decrease in granzyme B. Although 2B4 is not related to cytotoxicity against K562 cells in contrast to NKG2D and NKp46, YT-CB5 also showed the decrease in cytotoxicity against P815 cells treated with antibodies to 2B4, a cytotoxicity mediated by 2B4. These results supported the supposition that the decrease in NKG2D and 2B4 might cause impairment in induction of cytotoxicity by reducing signal transduction downstream of those receptors. Therefore, the degranulation stimulated by antibodies to NKG2D and 2B4 was examined in YT-CB5 by flow cytometry, which can estimate degranulation by measuring an increase in cell-surface CD107a accompanied with secretion of cytotoxic granules. YT-CB5 showed decreases in degranulation stimulated via NKG2D and 2B4, as observed under stimulation with beadbound and plate-bound antibodies, respectively. In addition, YT-CB5 also showed a decrease in phosphorylation of ERK1/2 stimulated with K562 cells as well as antibodies to NKG2D, but not with antibodies to 2B4 (Nishimura et al., 2009a). Moreover, we examined cytotoxicity against K562 cells, expression levels of NKG2D, 2B4 and NKp46, and the phosphorylation level of ERK1/2 stimulated with their respective antibodies in peripheral blood (PB-) NK cells purified from the blood of healthy volunteers, and compared results among the volunteers. PB-NK cells with a high expression of NKG2D, derived from one individual, showed high cytotoxicity and high phosphorylation of ERK1/2, whereas both cytotoxicity and ERK phosphorylation were low in PB-NK cells with low NKG2D, derived from another individual. In a similar way, the relationship among the cell-surface level of NKp46, phosphorylation level of ERK1/2, and cytotoxicity was also investigated and confirmed in PB-NK cells of healthy volunteers. Thus, we demonstrated that exposure to

36 Malignant Mesothelioma

as investigate the cytotoxicity of NK cells.

NK cell-activating

receptors

NKp46

NKG2D

activation molecule (SLAM) family is another group of activating receptors expressed on NK cells. A representative receptor of the SLAM family is 2B4, which induces cytotoxicity by stimulation with the natural ligand, CD48, or the antibody to 2B4 (Endt et al., 2007; Garni-Wagner et al., 1993; Valiante & Trinchieri, 1993). Moreover, natural cytotoxicity receptors (NCRs) make a family of receptors that includes NKp46, NKp44 and NKp30, which play a major role in the NK-mediated killing of most tumor cells (Moretta, A. et al., 2001; Sivori et al., 1999). These activating receptors transduce the stimulation signal leading to the phosphorylation of c-Jun N-terminal kinases (JNKs) and extracellular signal-regulated kinase (ERK), which cause polarization of the microtubule organizing center (MTOC) and cytotoxic granules followed by release of perforin and granzymes, producing degranulation (Chen et al., 2006; Chen et al., 2007). Thus, the various kinds of receptors expressed on NK cells control induction of cytotoxicity for target cells, and alteration in expression of these receptors is thought to affect the strength of the stimulation signal to induce cytotoxicity of NK cells. In addition, if NK cells exposed to asbestos show some characteristic alteration in expression of NK cell receptors, this alteration might be used as a possible marker of asbestos exposure-related immune alteration. Therefore, we planned to study the effect of asbestos exposure on NK cells, focusing on the expression level of NK cell receptors, as well

**Figure 4.** NK cell-activating receptors and the machinery of degranulation caused by stimulation with these receptors. The ligation of NK cell-activating receptors with their respective ligands induce phosphorylation of JNK and ERK, leading to polarization of the microtubule organizing center (MTOC)

2B4

Phosphorylation of JNK and ERK

Polarization of MTOC and granules

> Perforin and granzymes

Degranulation

Cytotoxic granules

**2.3. Cytotoxicity of human NK cell line cultured with asbestos for a long period** 

We initially started by making the sub-line of an NK cell line by culturing cells exposed to asbestos for a long period. The human NK cell line of YT-A1 was kindly provided by Dr.

and granules, where perforin and granzymes are released and induce apoptosis of target cells.

asbestos caused an impairment in cytotoxicity of NK cells with decreases in NK cellactivating receptors. The decrease in NKG2D caused the low level of signal transduction, followed by a decrease in degranulation, in the asbestos-exposed subline of cells.

Effect of Asbestos on Anti-Tumor Immunity

and Immunological Alteration in Patients with Malignant Mesothelioma 39

Macrophage NK

Cytokine-mediated control of functions

Chemokine-mediated

DC

with CB, whereas the expression of NKp46 decreased in those NK cells, resembling NK cells of patients with malignant mesothelioma (Nishimura et al., 2009b). To determine whether such restrictive alteration in expression of NK cell-activating receptors is caused by exposure to other mineral fibers, PBMCs were cultured with glass wool and then examined for the expression level of NKp46 on NK cells. Glass wool is a kind of man-made mineral fiber and is used as a representative substitute for asbestos. However, unlike asbestos, exposure to glass wool did not cause a decrease in expression of NKp46. These results indicate that exposure to asbestos causes the characteristic abnormality in human primary NK cells, resulting in a decrease in NKp46 but not in NKG2D. As described above, this characteristic is similar to that of PB-NK cells in patients with malignant mesothelioma, suggesting the possible relationship between asbestos exposure-related suppression of NK

What is the function of CD4+ T cells in regard to an appropriate immune response? CD4+ T cells, Th, contribute to various kinds of responses in innate and acquired immunity, including activation of NK cells, macrophages and dendritic cells (DC), as well as induction of antibody production and CTL development (Monney et al., 2002; Parker, 1993; Smith et al., 2004; Vivier et al., 2008). The production of various cytokines, including IL-2, IL-4, IL-5, IL-10, IL-17, IFN-γ, and TNF-α, by Th cells is one of the reasons for the multiple

**Figure 5.** Cytokine-mediated control of immune functions and chemokine-mediated control of cell migration. The various kinds of cytokines produced by Th cells allow them to contribute widely to functions of immune competent cells (top). Concentration gradient of chemokines and expression of

control of migration Th cells

CTL Th cells Cytokines

Concentration gradient of chemokine

chemokine receptors play a key role in appropriate cell migration (bottom).

cell function and development of malignant mesothelioma.

contributions of those cells to immune function (Fig. 5).

**3. CD4+ T cells** 

## **2.4. Low cytotoxicity and low NKp46 level in NK cells of mesothelioma patients**

The result of the experiment using the human NK cell line described above suggested that inhaled asbestos might cause impairment in cytotoxicity with altered expression of NK cellactivating receptors, and that PB-NK cells in patients with malignant mesothelioma might show a similar impairment in cytotoxicity to YT-CB cells. Therefore, peripheral blood mononuclear cells (PBMCs) prepared from the blood of patients with mesothelioma were examined for cytotoxicity against K562 cells and expression levels of NKG2D, 2B4, and NKp46 in NK cells. The cytotoxicity of PBMCs against K562 in the tree different ratios of effector and targets was measured for each individual, and expressed as the percentage of specific lysis. To evaluate cytotoxicity per given number of NK cells, The number of NK cells in the cytotoxic reaction was calculated from the percentage of CD3- CD56+ NK cells in PBMCs and the number of PBMCs dispensed to each well of the culture plate for cytotoxic reaction. A linear regression line with the percentage of specific lysis and number of NK cells, in which the formula is *[percentage of specific lysis] = A [number of NK cells] + B,* was determined by using the three sets of percentage of specific lysis and number of NK cells for each individual. Finally, the percentages of specific lysis per 5000 NK cells from each individual were calculated from these regression lines and compared with those of healthy volunteers. Mesothelioma patients showed significantly lower cytotoxicity than healthy volunteers (Nishimura et al., 2009b). However, unlike YT-CB5, NK cells in patients with mesothelioma did not show a decrease in expression level of NKG2D or 2B4, whereas a decrease in NKp46 was observed in those NK cells. Thus, although NK cells in the peripheral blood of patients with malignant mesothelioma were not of the same character as YT-CB5, they also showed alteration in the expression of one of the NK-cell activating receptors, albeit a different one namely NKp46, with low expression of NKp46, and low cytotoxicity.

## **2.5. Low cytotoxicity of NK cells with low NKp46 in the culture of PBMCs with asbestos**

As described above, NK cells in patients with malignant mesothelioma showed low cytotoxicity with low expression of NKp46, which was not found in YT-CB5. Therefore, to examine whether asbestos exposure causes such a decrease in NKp46 on NK cells, we performed a different experiment in which PBMCs from healthy volunteers were cultured in IL-2-supplemented media with chrysotile B at 5 µg/ml, and after 7 days cells were harvested and examined for expression levels of NKG2D, 2B4, and NKp46 in CD3- CD56+ NK cells by flow cytometry. IL-2 is a representative cytokine that induces proliferation and activation of NK cells and is commonly used to culture these cells. The results showed no decreases in cell-surface expression of NKG2D and 2B4 in NK cells derived from the culture with CB, whereas the expression of NKp46 decreased in those NK cells, resembling NK cells of patients with malignant mesothelioma (Nishimura et al., 2009b). To determine whether such restrictive alteration in expression of NK cell-activating receptors is caused by exposure to other mineral fibers, PBMCs were cultured with glass wool and then examined for the expression level of NKp46 on NK cells. Glass wool is a kind of man-made mineral fiber and is used as a representative substitute for asbestos. However, unlike asbestos, exposure to glass wool did not cause a decrease in expression of NKp46. These results indicate that exposure to asbestos causes the characteristic abnormality in human primary NK cells, resulting in a decrease in NKp46 but not in NKG2D. As described above, this characteristic is similar to that of PB-NK cells in patients with malignant mesothelioma, suggesting the possible relationship between asbestos exposure-related suppression of NK cell function and development of malignant mesothelioma.
