Potential role of natural killer cells in controlling growth and infiltration of AIDS-associated primary effusion lymphoma cells

Abstract

Natural killer (NK) cells are an important component of the innate immune response against microbial infections and tumors. Direct involvement of NK cells in tumor growth and infiltration has not yet been demonstrated clearly. Primary effusion lymphoma (PEL) cells were able to produce tumors and ascites very efficiently with infiltration of cells in various organs of T-, B- and NK-cell knock-out NOD/SCID/gammac(null) (NOG) mice within 3 weeks. In contrast, PEL cells formed small tumors at inoculated sites in T- and B-cell knock-out NOD/SCID mice with NK-cells while completely failing to infiltrate into various organs. Immunosupression of NOD/SCID by treatment with an antimurine TM-beta1 antibody, which transiently abrogates NK cell activity in vivo, resulted in enhanced tumorigenicity and organ infiltration in comparison with non-treated NOD/SCID mice. Activated human NK cells inhibited tumor growth and infiltration in NOG mice. Our results suggest that NK cells play an important role in growth and infiltration of PEL cells, and activated NK cells could be a promising immunotherapeutic tool against tumor or virus-infected cells either alone or in combination with conventional therapy. The rapid and efficient engraftment of PEL cells in NOG mice also suggests that this new animal model could provide a unique opportunity to understand and investigate the mechanism of pathogenesis and malignant cell growth.

Figure 1

Successful engraftment and tumor marker of primary effusion lymphoma (PEL) cells in T, B and natural killer (NK) knock‐out NOG mice. (a) Photograph of mice inoculated with BCBL‐1 and TY‐1 cells subcutaneously in the postauricular region (left panel) and those of subcutaneously formed BCBL‐1 and TY‐1 tumor 3 weeks after inoculation of cells (right panel). (b) Subcutaneous tumor size of mice inoculated with BCBL‐1 and TY‐1 cells, shown as the mean ± s.e.m. from five mice (P = 0.0335). (c) Photograph of ascites‐bearing mice inoculated with BCBL‐1 and TY‐1 cells intraperitonealy in the abdominal region (left panel) and peritoneal cavity of mice 21 days after inoculation of BCBL‐1 (middle panel) and TY‐1 cells (right panel). Arrow head indicates the tumor in mice inoculated intraperitonealy. (d) Volume of ascites in mice inoculated with various BCBL‐1 and TY‐1 cells, shown as the mean ± s.e.m. from five mice (P < 0.0001). (e,f) Hematoxylin–eosin (HE) and immunohistochemical staining of tumor tissue of BCBL‐1 and TY‐1 cells injected mice. Upper panels represent HE staining. Immunohistochemical staining was conducted using rabbit anti‐LANA (lower panels). Left and right panels represent results with BCBL‐1 and TY‐1, respectively (magnification, ×40). Data are from (e) mice inoculated subcutaneously and (f) mice inoculated intraperitonealy.

Figure 2

Metastasis of primary effusion lymphoma (PEL) cells in various organs of T, B and natural killer (NK) knock‐out NOG mice. (a–d) Histological analysis of lung, liver and spleen of mice inoculated with BCBL‐1 and TY‐1 cells either (a,b) subcutaneously (sc) or (c,d) intraperitonealy (ip). Immunohistochemical staining was conducted using anti‐LANA. Data are from (a,c) BCBL‐1‐inoculated mice and (b,d) TY‐1‐inoculated mice. Left and right panels of all figures represent hematoxylin–eosin and immunostaining, respectively (magnification, ×40).

Figure 3

Natural killer (NK) cells in tumor growth and infiltration. BCBL‐1 cells were inoculated subcutaneously in the postauricular region or intraperitonealy in the abdominal region of T and B knock‐out NOD/SCID, TMβ1‐pretreated T and B knock‐out NOD/SCID and T, B and NK knock‐out NOG mice. (a) Photograph of mice inoculated with BCBL‐1 cells subcutaneously in the postauricular region. (b) Photograph of BCBL‐1 tumor 3 weeks formed subcutaneously after inoculation of cells. (c) Subcutaneous tumor size of mice inoculated with BCBL‐1 cells, shown as the mean ± s.e.m. from six mice. Tumor size of TMβ1‐pretreated NOD/SCID mice was significantly larger than NOD/SCID (P = 0.0065) and that of NOG mice was more significant than NOD/SCID (P = 0.0045). (d) Photograph of ascites‐bearing mice inoculated with BCBL‐1 cells intraperitonealy in the abdominal region. (e) Photograph of the peritoneal cavity of mice 3 weeks after inoculation of BCBL‐1. Left, middle and right panels represent the T and B knock‐out NOD/SCID, TMβ1‐pretreated T and B knock‐out NOD/SCID and T, B and NK knock‐out NOG mice, respectively. Arrow head indicates the tumor in mice inoculated intraperitonealy. (f) Volume of ascites in mice inoculated with BCBL‐1 cells, shown as the mean ± s.e.m. from six mice. Volume of ascites in TMβ1‐pretreated NOD/SCID mice was significantly higher than NOD/SCID (P = 0.0003) and that of NOG mice was more significant than NOD/SCID (P < 0.0001). Hematoxylin–eosin (HE) and immunohistochemical staining of (g) lung tissue and (h) tumor tissue of BCBL‐1‐injected mice. Upper panels represent HE staining. Immunohistochemical staining was conducted using rabbit anti‐LANA (lower panels). Left, middle and right panels represent results from T and B knock‐out NOD/SCID, TMβ1‐pretreated T and B knock‐out NOD/SCID and T, B and NK knock‐out NOG mice, respectively. Magnification, ×40.

Figure 4

Cytotoxic activity of activated natural killer (NK) in vitro culture cells. (a) Spontaneous cytotoxic activity of freshly isolated peripheral blood mononuclear cells (KT‐1 and KN‐2 represent samples from two donors) and NK cell line KHYG‐1 against K‐562 cells at different effector‐to‐target (E/T) ratios. (b) Cells cultured for 2 weeks against K‐562 cells at different E/T ratios. (c) Cytotoxic activity of activated NK cells (NK‐1 and NK‐2 represent samples from two donors) against primary effusion lymphoma cells in vitro at various E/T ratios.

Figure 5

Inhibition of primary effusion lymphoma (PEL) cell growth and infiltration in NOG mice. T, B and natural killer (NK) knock‐out NOG mice were injected with PEL cells (2 × 10⁶) intraperitonealy in the abdominal region. Mice were administered either RPMI‐1640 or activated NK cells (1 × 10⁷) intraperitonealy on days 4, 10 and 17 followed by observation for up to 3 weeks. (a) Photograph of ascites‐bearing control PEL mice and activated NK‐treated PEL mice. (b) Volume of ascites in control PEL mice and activated NK‐treated PEL mice. Volume of ascites in mice inoculated with BCBL‐1 cells, shown as the mean ± s.e.m. from six mice (P = 0.0003). (c) Hematoxylin–eosin (HE) and immunohistochemical staining of the lung of NOG mice 3 weeks after inoculation of PEL cells using anti‐LANA. Upper and lower panels show HE and immunohistochemical staining, respectively. Left and right panels represent the data from control mice and mice treated with activated NK cells, respectively. Magnification, ×40. The data represent six mice in each group and three healthy donors (two mice for each donor).