In vivo tumouricidal effects of LAD-1 monoclonal antibody on murine RL-male-1 lymphoma mediated by enhanced phagocytosis

Abstract

We have reported previously that the LAD-4 monoclonal antibody (mAb) directed against a fibronectin receptor (FNR) on RL-male-1 T lymphoma cells in BALB/c mice partially inhibited their migration to the liver. In the present study, we examined the mechanism by which another anti-FNR mAb, LAD-1, exerts its antitumourigenic effects. Administration of LAD-1 significantly prolonged survival of BALB/c mice challenged previously with RL-male-1 cells. LAD-1 enhanced phagocytosis of RL-male-1 cells by hepatic macrophages and clodronate-mediated macrophage depletion abrogated the antitumour activity of LAD-1. In vitro experiments revealed that a pan-caspase inhibitor, zVAD-fmk, did not affect the ability of LAD-1 to inhibit the proliferation of RL-male-1 cells. These data suggest that the antitumour effects of LAD-1 may be dependent on stimulation of tumour cell phagocytosis and are apoptosis-independent. Thus, LAD-1-induced phagocytosis of lymphoma cells by hepatic macrophages in mice may, at least in part, be responsible for the prolonged survival of the mice.

Introduction

Human B-cell malignancy can be treated successfully with the anti-CD20 monoclonal antibody (mAb), rituximab [1]. Antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), antibody-mediated phagocytosis and the induction of apoptosis may all be involved in the therapeutic effects of anti-CD20 mAbs [2–9]. Although several studies showed the importance of the Fcγ receptor or CDC in the therapeutic activity of anti-CD20 mAbs in a mouse model [10,11], phagocytosis of lymphoma cells in patients receiving treatment with mAbs has not been demonstrated. Because lymphomas frequently metastasize to the liver [12–15], hepatic monocytes are thought to have an important role in the clearance of lymphoma cells from this organ. The therapeutic efficacy of mAb treatment for T-cell lymphoma and its mechanism of action have not yet been defined.

We have shown previously that LAD mAb, directed against an FNR on murine T-cell lymphoma RL-male-1, possessed antitumour activity in vivo[13]. To extend these findings, here we investigated whether this mAb mediated its effects by a mechanism that was similar to that reported previously for anti-CD20 mAbs [3]. In addition, we demonstrated that LAD-1 induced the phagocytosis of lymphoma cells by mononuclear cells in the liver as early as 3 h after administration. The effect of this mAb on the phagocytic activity of tissue macrophages and its direct cytotoxicity were also addressed.

Materials and methods

Mice and murine lymphoma cell line

Male BALB/c and BALB/c nude mice, 6–8 weeks of age, were purchased from Japan SLC Inc. (Hamamatsu, Japan). T cell lymphoma cells derived from BALB/c mice, i.e. RL-male-1, were cultured and maintained in RPMI-1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% heat-inactivated fetal bovine serum and antibiotics as described previously [13]. An ascitic version of the RL-male-1 cell line was also used in this study, as reported previously [16].

Antibodies

A rat hybridoma cell line producing the anti-LFA-1 antibody M17/4 was obtained from the American Type Cell Collection (Rockville, MD, USA). LAD-1 (IgG2a, kappa) and anti-LFA-1 mAb were purified as reported previously [13]. Anti-Thy1·2 mAb (30-H12), phycoerythrin (PE)-conjugated or biotinylated hamster antimouse CD80 mAb were purchased from BD Pharmingen (San Diego, CA, USA). Biotinylated rat antimouse F4/80 mAb was from Serotec Ltd. (Oxford, UK). F(ab′)₂ fragments of LAD-1 were produced with immobilized pepsin (Pierce Biotechnology Inc., Rockford, IL, USA) according to the manufacturer's recommendations. Intact IgGs were removed by passing material through a protein G Sepharose column (Amersham Biosciences).

Survival curves

Mice (10 per group) were first injected intravenously (i.v.) with 2 × 10⁵ RL-male-1 cells, after which they received purified LAD-1, anti-LFA-1 or normal rat IgG (200 µg each) intraperitoneally (i.p.) once or three times weekly. In some experiments, mice were pretreated with clodronate liposomes as described below. After 24 h, animals were injected i.v. with 1 × 10⁵ lymphoma cells and then with LAD-1 or normal rat IgG (200 µg each), also i.v., at intervals of 24 h from days 1–3. All our animal studies were approved by the Local Institutional Review Board.

Immunofluorescence staining of liver cells

Liver specimens from mice that were injected i.v. with the green lipophilic dye PKH67 (Sigma, St Louis, MO, USA)-labelled RL-male-1 cells (5 × 10⁶) and either LAD-1 (200 µg) or control rat IgG (200 µg) were fixed in periodate-lysine-paraformaldehyde fixative, frozen, and sections cut (8 µm) with a cryostat as described previously [17]. The sections were blocked with phosphate buffered saline (PBS) containing 5% bovine serum albumin, and incubated for 6 h at 4°C with biotinylated anti-CD80 mAb and for 30 min at room temperature with Texas red-avidin (BD Pharmingen). A biotinylated hamster anti-KLH mAb (BD Pharmingen) was used as a negative control. The sections were examined by fluorescence microscopy (Olympus, Tokyo, Japan). Peritoneal exudate cells (PEC), obtained by rinsing the peritoneal cavity with PBS, were cultured for 3 h with PKH67-labelled RL-male-1 cells at a ratio of 1 : 1 in the presence of either LAD-1 or control M17/4 or 30-H12 IgG (10 µg each). The cells were then harvested and labelled with anti-CD80 mAb, as described above. In some experiments, cells were labelled with PE-anti-CD80 mAb and analysed with a confocal microscope as described previously [18] as well as by flow cytometry.

Flow cytometry

RL-male-1 cells (5 × 10⁵) were cultured in 1 ml medium in the presence of LAD-1, anti-LFA-1 and 30-H12 mAbs or normal rat IgG (15 µg/ml each) for 3, 24 or 48 h at 37°C. The cells were then harvested and stained with fluoroscein isothiocyanate (FITC)-labelled annexin V and propidium iodide (PI) (MBL, Nagoya, Japan) [19]. For detection of DNA fragmentation, cells (1 × 10⁶) were cultured in 1 ml of medium containing LAD-1 or control antibodies for 3, 24 or 48 h at 37°C. In some experiments, cells were pretreated with the pan-caspase inhibitor, zVAD-fmk (Peptide Institute, Osaka, Japan) for 1 h and then treated with LAD-1. Cells treated with 10 mM cyclophosphamide (Sigma) for 24 h were used as a positive control. The cells were treated with terminal deoxynucleotidyl transferase and stained with FITC-labelled dUTP using the TUNEL method (Beckman Coulter). The stained cells were then analysed using a FACS Calibur (Becton Dickinson, San Jose, CA, USA).

Preparation of liposomes with entrapped dichloromethylene-diphosphonate

Liposomal encapsulation of clodronate (dichloromethylene-diphosphonate, Sigma) was undertaken as reported elsewhere [20]. Briefly, 8 mg cholesterol was added to 86 mg egg phosphatidylcholine, and the chloroform phase was evaporated. The clodronate solution was made by dissolving 1 g of dichloromethylene-diphosphonate in 5 ml of sterile PBS. The solution was added to the liposomes and mixed thoroughly. To wash the liposomes, the mixed solution was centrifuged twice in PBS at 1 × 10⁴ g for 30 min. Thereafter, the supernatant (1 ml) was collected and diluted with 4 ml PBS, of which 0·2 ml was injected i.v. 1 day before injection of lymphoma cells.

Measurement of cell growth

Cells (2 × 10⁴) were cultured at 37°C in 100 µl of medium with different concentrations of normal rat IgG, LAD-1 or anti-LFA-1 mAb. After 24 h, growth of RL-male-1 cells was determined colourimetrically using the WST-1 tetrazolium salt reagent (TaKaRa Biomedicals, Tokyo, Japan). Substrate cleavage was monitored at 450 nm. In the assay measuring inhibition of cell growth, cells were pretreated for 1 h with different concentrations of zVAD-fmk, the cathepsin B inhibitor CA074me (Peptide Institute), the cathepsin D inhibitor pepstatin A [21,22] or with vehicle (1% dimethyl sulphoxide: DMSO), and then cultured with LAD-1 (25 µg/ml) as described above.

Statistical analysis

All statistical analyses were performed using statview software (Abacus Concepts, Berkeley, CA, USA). Survival data were analysed by the Kaplan—Meier method. Differences between groups were evaluated using one factor anova followed by Scheffé's F-test. All values are expressed as the mean ± s.d., and statistical significance was set at P < 0·05.

Results

LAD-1 treatment prolongs survival of lymphoma-bearing mice

Mice previously inoculated i.v. with RL-male-1 cells (2 × 10⁵) received one (Fig. 1a) or three (Fig. 1b) weekly i.p. injections of LAD-1 (200 µg). This treatment prolonged the survival of these animals compared to control mice that received M17/4 or normal rat IgG (P < 0·05). Mice that received three weekly injections of LAD-1 survived longer than those receiving only a single injection (Figs 1a, b) (P < 0·05). We tested whether daily administration of LAD-1 after tumour challenge improved immunotherapy (Fig. 1c). Daily injections of LAD-1 (200 µg/day) for 3 days after tumour challenge did not significantly increase median survival time, suggesting that the LAD-1 in the first injection sufficed for prolongation of survival time. We hypothesized that tissue macrophages may contribute to the action of LAD-1. To assess whether macrophages did determine the activity of LAD-1, they were depleted in vivo using chlodronate. After pretreatment with clodronate liposomes for 24 h, the numbers of F4/80⁺ macrophages in the splenic red pulp and liver were markedly reduced (data not shown). This coincided with a significant reduction in the immunotherapeutic activity of LAD-1, which was not observed using control liposomes (Fig. 1c). These data suggest that macrophages are an essential factor for the in vivo therapeutic effects of LAD-1.

Fig. 1.

Survival curves of mice given injections of different monoclonal antibodies (mAbs) following RL-male-1 cell inoculation. (a) BALB/c mice (10 mice per group) were injected intraperitoneally (i.p.) once on day 0 with either 200 µg of normal rat IgG (triangles), LAD-1 (circles), M17/4 (squares) or with saline (dashed line) 15 min after intravenous (i.v.) challenge with RL-male-1 cells (2 × 10⁵). (b) Mice (10 per group) received three weekly i.p. injections with saline or different IgGs (200 µg) after i.v. challenge with RL-male-1 cells as in (a). (c) Effect of macrophage depletion on survival of mice treated with LAD-1. Mice (n = 10) were injected i.v. with liposomal clodronate (closed circles) or liposome alone (open circles) 1 day before injection i.v. with lymphoma cells (1 × 10³). Thereafter, animals were injected i.v. with LAD-1 (200 µg) on days 1, 2 and 3. Survival of untreated mice is represented by open squares. Arrows indicate the time of antibody injection. The data were plotted using the Kaplan—Meier method and significant differences were found between LAD-1 treated and control mice in both (a) and (b), and between clodronate treated and control mice by the log-rank test (P < 0·01).

LAD-1 induces phagocytosis of RL-male-1 cells in the liver

In order to determine whether the therapeutic effect of LAD-1 was due to antibody-mediated phagocytosis, we examined tumour cell phagocytosis in the livers of BALB/c mice 3 h after i.v. injection of PKH67-labelled RL-male-1 cells (5 × 10⁶) and i.p. injection of LAD-1 or control IgG (Fig. 2). Our results showed that hepatic macrophages, banana-shaped and labelled with PE-anti-CD80, did indeed phagocytose PKH67-labelled RL-male-1 cells (Fig. 2a), whereas CD80⁺PKH67⁺ cells were rarely observed in the liver sections from mice treated with normal rat IgG (Fig. 2a). Because the number of double-positive cells in several sections may not reflect the numbers in the whole liver, we attempted to quantify the degree of phagocytosis more generally. To this end, we isolated mononuclear cells from liver homogenates by centrifugation on Ficoll-Isopaque; at the same time, splenic mononuclear cells and peripheral blood mononuclear cells were also isolated. The mononuclear cells from each of these sources were labelled with PE-anti-CD80 mAb and analysed by flow cytometry. We found that the percentage of CD80⁺PKH67⁺ cells was significantly higher (P < 0·05) in the liver of mice treated with LAD-1 than in the controls (Fig. 2b). There were few PKH67⁺ cells in splenic, and none in peripheral blood, mononuclear cells obtained from these animals (Fig. 2b).

Fig. 2.

Phagocytosis of RL-male-1 cells in the liver. (a) Phagocytosis of RL-male-1 cells by macrophages was examined by immunofluorescence after intravenous (i.v.) injection of PKH-labelled RL-male-1 cells (5 × 10⁶) in BALB/c mice. Mice received an intraperitoneal (i.p.) injection of either 200 µg of normal rat IgG or LAD-1 15 min after injection of RL-male-1 cells. Each section was immunostained with Texas red (TR)-avidin and biotinylated anti-CD80 monoclonal antibodies (mAbs). A typical photomicrograph (× 20 objective) of a representative liver section from a mouse treated with LAD-1 is shown. The arrow indicates macrophages that have phagocytosed RL-male-1 cells. Scale bar, 100 µm. (b) Mice received IgG and PKH67-labelled RL-male-1 cells as in (a), and after 3 h their liver, spleen and peripheral blood mononuclear cells were isolated and labelled with PE-anti-CD80 mAb and analysed by flow cytometry. The bar chart indicates the percentages of double-positive cells (CD80⁺PKH67⁺). Data represent mean ± s.d. (n = 5). A significant difference (*P < 0·01) was observed in the percentage of double-labelled cells in the liver and spleen between the LAD-1 and rat IgG treatment groups.

Phagocytosis of RL-male-1 cells by peritoneal macrophages in vitro

We next determined whether peritoneal macrophages were capable of phagocytosing RL-male-1 cells in the presence of LAD-1. PEC were cocultured with PKH67-labelled RL-male-1 cells in the presence of either LAD-1 or control IgGs for 3 h, after which they were labelled with PE-anti-CD80 mAb and then analysed (Fig. 3a). Our data showed that peritoneal macrophages phagocytosed PKH67-labelled RL-male-1 cells in the presence of LAD-1 but not control antibodies (Fig. 3a). Because the binding level of LAD-1 compared with the control is important for subsequent phagocytotic activity, we evaluated this parameter by flow cytometry. The mean fluorescence intensities of control mAbs were comparable to LAD-1, suggesting that LAD-1, M17/4 and 30-H12 bound to similar degrees to the surface of RL-male-1 cells (Fig. 3b). As shown in Fig. 3c, approximately 16% of CD80⁺ macrophages phagocytosed RL-male-1 lymphoma cells in the presence of LAD-1 whereas less than 6% did so in the presence of M17/4 or 30-H20 mAb (P < 0·01, LAD-1 versus control antibodies).

Fig. 3.

Phagocytosis of RL-male-1 cells by PEC. Cells were co-cultured with PKH67-labelled RL-male-1 cells (green) in the presence of LAD-1 or control rat IgG for 3 h, after which they were labelled with anti-CD80 monoclonal antibodies (mAbs) as in Fig. 2a (red). (a) Phagocytosis of RL-male-1 cells by macrophages was examined by immunofluorescence. Typical photomicrographs (40 × objective) of these cells are shown. Arrows indicate double-positive cells. A confocal image of the double-positive cell is shown in the right lower photograph. Scale bars, 10 µm. (b) Surface levels of molecules recognized by each antibody on RL-male-1 cells was measured by flow cytometry. Control (dotted line): secondary antibody only. The numbers in the histograms indicate the mean fluorescence intensity. One representative experiment out of three is shown. (c) The phagocytosis was analysed by flow cytometry. The bar chart indicates the percentages of double-positive cells as in Fig. 2b. Data represent mean ± s.d. (n = 3). A significant difference (**P < 0·01) was observed in the percentage of double-labelled cells between the LAD-1 and controls.

LAD-1 induces PI incorporation by RL-male-1 cells in vitro

To determine whether LAD-1 induces apoptotic cell death, RL-male-1 cells were cultured in the presence of an optimal concentration of LAD-1 or control antibodies. The percentage of cells undergoing apoptosis was then determined by annexin V binding and PI uptake. Three hours after the addition of LAD-1, the percentage of apoptotic cells (i.e. annexin V⁺ and PI⁻) increased from the control value of 0·5% to 10%, whereas no such increase occurred after the addition of M17/4 mAb (Fig. 4a). Similarly, cells with plasma membrane damage, i.e. PI⁺ and PI⁺/annexin V⁺, increased from 13 to 40% after LAD-1 treatment, whereas no significant increase occurred after the addition of M17/4 (13 versus 17) (Fig. 4a). The percentage of LAD-1-treated cells with plasma membrane damage increased to 62% at 24 h and then decreased to 48% at 48 h, but did not change with control rat IgG (28% and 29%, respectively). No significant increase compared to the control was observed with M17/4 (Fig. 4a).

Fig. 4.

Effect of LAD-1 on annexin V binding, propidium iodide (PI) uptake and DNA fragmentation in RL-male-1 cells. (a) RL-male-1 cells (5 × 10⁵) were cultured for 3 h (upper panel), 24 h (middle panel) or 48 h (bottom panel) with or without the indicated monoclonal antibodies (mAbs) (15 µg/ml), after which they were washed and stained with fluoroscein isothiocyanate (FITC)-annexin V and PI. The numbers at the right of the dot plots indicate the percentage of cells in the upper right and lower right regions. (b) RL-male-1 cells (5 × 10⁵) were cultured in the presence of LAD-1 (15 µg/ml) (solid lines) for the indicated times alone or in the presence of the pan-caspase inhibitor zVAD-fmk (100 µ M). The dotted lines indicate the rat IgG control. Cells cultured with cyclophosphamide (10 mM) for 24 h were used as a positive control. The cell populations are divided into M1 and M2 regions. The percentages in the M2 population are shown in each histogram. The cells, which were fixed in paraformaldehyde, were treated with terminal deoxynucleotidyl transferase and stained with FITC-dUTP (TUNEL assay). One representative experiment of three is shown.

We also determined whether LAD-1 induced DNA fragmentation using the TUNEL method. The percentage of TUNEL-positive cells was 12, 5 and 14% after LAD-1 treatment for 3, 24 and 48 h, respectively (Fig. 4b). Use of a pan-caspase inhibitor, zVAD-fmk (100 µ M), did not reverse this minimal DNA fragmentation after treating with LAD-1, whereas cyclophosphamide as a positive control induced DNA fragmentation in 54% of RL-male-1 cells at 24 h (Fig. 4b).

LAD-1 inhibits proliferation of RL-male-1 cells in vitro

In light of the above data suggesting a cytopathic effect of LAD-1 on RL-male-1 cells, we investigated whether LAD-1 inhibited their proliferation in vitro. The F(ab′)₂ fragment as well as whole LAD-1 antibody was found to inhibit proliferation of RL-male-1 cells in a dose-dependent manner, while M17/4 mAb and normal rat IgG had no significant effects (Fig. 5a). We next examined whether cathepsin B, an execution protease in tumour cell apoptosis, might have been responsible for the LAD-1-induced growth inhibition. Pretreatment with CA074me, a specific inhibitor of cathepsin B, reversed the LAD-1-induced growth inhibition as measured after 48 h in a dose-dependent manner, whereas pretreatment with zVAD-fmk or pepstatin A, a specific inhibitor of cathepsin D, did not (Fig. 5b).

Fig. 5.

Effect of LAD-1 on the in vitro proliferation of RL-male-1 cells. (a) Cells (2 × 10⁴) cultured in the presence of the indicated doses of rat IgG (squares), LAD-1 (circles), F(ab′)₂ LAD-1 (closed circles) or anti-LFA-1 monoclonal antibodies (mAbs) (M17/4) (triangles) at 37°C for 48 h. Viable cells were quantified using the WST assay. (b) Cells were preincubated for 1 h with increasing concentrations of zVAD-fmk (closed circles), a specific inhibitor of cathepsin B, CA074me (closed squares), a specific inhibitor of cathepsin D, pepstatin A (closed triangles) or vehicle alone (1% dimethylsulphoxide (DMSO), at dose 0) after which they were cultured in the presence of LAD-1 (25 µg/ml) for 48 h. Viable cells were quantified as in (a). The value of the untreated control was taken to be 100%. Data, in triplicate samples, are expressed as the mean ± s.d. *Significant difference (P < 0·01) was observed in cathepsin B-treated and pepstatin A-treated samples. Similar results were obtained in three independent experiments.

Discussion

It is shown here that mAb LAD-1 has a therapeutic effect in mice inoculated with RL-male-1 lymphoma cells. In contrast, the anti-LFA-1 mAb, M17/4, of the same IgG isotype as LAD-1, had no effect on survival. We found that LAD-1 induced phagocytosis of RL-male-1 cells by hepatic but not by peripheral blood mononuclear cells within 3 h, and that clodronate-mediated macrophage depletion abrogated the therapeutic effects of LAD-1 in vivo. The fact that phagocytosis occurred within 3 h suggested that LAD-1 might have directly induced apoptosis of the RL-male-1 cells because cells with phosphatidylserine (PS) externalization are known to be recognized rapidly by phagocytes and ingested while still intact [23]. Our results showed that LAD-1 induced both an apoptotic cell population (FITC-annexin V⁺/PI⁻) and a necrotic cell population (PI⁺ and PI⁺/FITC-annexin V⁺) 3 h after treatment. However, LAD-1 induced only minimal DNA fragmentation in RL-male-1 cells 3 h after treatment. Furthermore, the percentage of FITC-annexin V⁺/PI⁻ cells induced by LAD-1 did not parallel the percentage of cells showing DNA fragmentation 48 h after treatment. Thus, we could not establish that the FITC-annexin V⁺/PI⁻ cell population in RL-male-1 cells was indeed apoptotic. Cell clusters induced by LAD-1 may result in ambiguous higher annexin V staining as suggested previously [3], and other molecules as well as PS can also be targets for phagocytes [24,25]. Neither necrotic cells nor apoptotic cells were observed in the liver of lymphoma-bearing mice 3 h after injection of LAD-1 (unpublished data), suggesting that lymphoma cells that have died by a non-apoptotic mechanism may be cleared quickly by hepatic macrophages. There have been reports of a similar type of cell death that was induced by mAbs and found to be complement-independent, lacking DNA fragmentation, and accompanied by cell membrane damage [26–29]. Three of these mAbs, directed against different antigens, induced the non-apoptotic cell death of normal T cells and Jurkat T cells, suggesting that this type of cell death may be a biologically relevant vis-à-vis the target of phagocytosis by mononuclear cells in the liver.

Another mechanism by which LAD-1 may exert its therapeutic effects is direct cytotoxicity to RL-male-1 cells. LAD-1 inhibited RL-male-1 cell proliferation in vitro, but we did not detect CDC against these cells in the presence of LAD-1 as assessed by the PI exclusion test, nor did we detect ADCC using splenic effector cells in a 4-h ⁵¹Cr-release assay (unpublished data). These findings suggest that neither CDC nor ADCC play an important role in mediating the effects of LAD-1 in vivo, in contrast to the findings with the anti-CD20 mAb, rituximab [2–11].

Taken together, our data suggest that anti-T lymphoma LAD mAbs enhance survival, at least in part, by up-regulating phagocytosis in organs which are targets for metastasis. Thus, LADs and other similar anti-FNR mAbs, when available as humanized mAbs [30], may represent a new therapeutic modality for T cell lymphoma.