Search

We examined the effectiveness of various anti-tumour agents to natural killer (NK)-cell tumour cell lines and samples, which are generally resistant to chemotherapy, using flow cytometric terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labelling (TUNEL) assay. Although NK-YS and NK-92 were highly resistant to various anti-tumour agents, l-asparaginase induced apoptosis in these two NK-cell lines. NK-cell leukaemia/lymphoma and acute lymphoblastic leukaemia (ALL) samples were selectively sensitive to l-asparaginase and to doxorubicin (DXR) respectively. Samples of chronic NK lymphocytosis, an NK-cell disorder with an indolent clinical course, were resistant to both drugs. Our study clearly separated two major categories of NK-cell disorders and ALL according to the sensitivity to DXR and l-asparaginase. We examined asparagine synthetase levels by real-time quantitative polymerase chain reaction (RQ-PCR) and immunostaining in these samples. At least in nasal-type NK-cell lymphoma, there was a good correlation among asparagine synthetase expression, in vitro sensitivity and clinical response to l-asparaginase. In aggressive NK-cell leukaemia, although asparagine synthetase expression was high at both mRNA and protein levels, l-asparaginase induced considerable apoptosis. Furthermore, samples of each disease entity occupied a distinct area in two-dimensional plotting with asparagine synthetase mRNA level (RQ-PCR) and in vitrol-asparaginase sensitivity (TUNEL assay). We confirmed rather specific anti-tumour activity of l-asparaginase against NK-cell tumours in vitro, which provides an experimental background to the clinical use of l-asparaginase for NK-cell tumours.

Although considerable part of natural killer (NK) cell neoplasms possess EBV genome, there has been no direct evidence that EBV infects human NK cells in vitro. In this study, we demonstrated EBV entry into NK cells using a recombinant EBV, which contains enhanced green fluorescent protein (EGFP) gene in its genome (EGFP-EBV). After 48 h of exposure to EGFP-EBV, we detected EGFP signals in approximately 30% of NK-92 and NKL cells and >40% of peripheral blood NK cells from three healthy volunteers. Reverse transcription-PCR analysis of various EBV-associated genes confirmed EBV infection. In situ hybridization for EBERs and BHLFs showed that latent and lytic infections coexisted at the early phase of EBV infection in two NK cell lines. Although BHLF-positive cells in the early lytic phase were round-shaped, EBER-positive cells in latent EBV infection tended to show a bizarre shape. Flow cytometric analysis of EGFP-EBV-exposed NK cell lines showed that most of EBV-infected cells entered early apoptosis after 72 h of EBV exposure, which explains the difficulties to establish EBV-carrying NK clones. Flow cytometry and reverse transcription-PCR analysis indicated that two NK cell lines may fuse with EBV using HLA class II after binding to the virus through a distinct molecule from CD21. We established two EBV-carrying NKL clones showing latency types I and II, both of which are recognized in EBV-associated NK cell neoplasms. Because EBV-infected NKL cells showed only type I latency during the early phase of infection, the temporal profile of latent gene expression is similar to that of T cells. We first report in vitro EBV infection of human NK cells and establishment of EBV-carrying NK clones, which should contribute to elucidate the role of EBV in the development of NK cell neoplasms.

Natural killer (NK) cell neoplasms, which are derived from mature or precursor NK cells, are rare diseases and are observed predominantly in Asian countries. We analyzed the status of the Rb, p53, p15INK4B, p16INK4A and p14ARF genes in these diseases by Southern blot, polymerase chain reaction-single strand conformational polymorphism (PCR-SSCP) and western blot analysis. We used 31 NK cell neoplasms, including four cell lines derived from NK cell neoplasms, 3 myeloid / NK cell precursor acute leukemias, 4 blastic NK cell lymphoma / leukemias, 4 aggressive NK cell leukemia / lymphomas, 4 nasal NK cell lymphomas, and 12 chronic NK lymphocytosis. We found gene amplification of the p53 gene in one nasal NK cell lymphoma, and point mutations of the p53 gene in one blastic NK cell lymphoma / leukemia and one chronic NK lymphocytosis. In addition, homozygous deletions of p15, p16 and p14 genes in 5 out of 31 samples were detected; 3 were from nasal NK cell lymphoma and 2 from blastic NK cell lymphoma / leukemia. Also hemizygous deletion of the Rb gene in one blastic NK cell lymphoma was detected. Rb proteins were highly expressed in one cell line as well as two myeloid / NK cell precursor acute leukemias. In other cell lines, complete loss and an aberrant migration pattern of Rb protein expression were observed. Comparative genomic hybridization suggested that the homozygous deletions of the p15, p16 and p14 were subtle chromosomal deletions and could not be identified by standard karyotyping in some cases. Although the number of cases we analyzed was not large, alterations identified in the Rb, p53, p16, p15 and p14 genes are of significance and might be associated with tumorigenesis in NK cell neoplasms.

In the normal developmental pathway of natural killer (NK) cells, pre-NK cells express CD161, immature NK cells express CD161 and CD56, and mature NK cells express CD161, CD56 and CD94. To identify the normal counterpart of NK cells from which neoplastic cells originate, surface antigens were analysed. Blastic NK-cell lymphoma/leukaemia lacked CD94 and CD161 but had CD56. Aggressive NK-cell leukaemia/lymphoma and nasal NK-cell lymphoma, although morphologically immature, expressed both CD56 and CD94 and strong NK activity. Cells from chronic NK lymphocytosis expressed CD56 and CD94.

P-glycoprotein (P-gp), a transmembrane efflux pump encoded by the MDR1 gene, has been found to be expressed in many normal bone marrow and peripheral blood cells. Among normal leukocytes, CD3(-)CD16(+) or CD3(-)CD56(+) lymphocytes, ie, natural killer (NK) cells, express relatively high levels of P-gp, but little is known about P-gp in abnormally expanded NK cells. In this study, we examined the expression and activity of P-gp on NK cells derived from three normal donors, six patients with indolent NK cell-lineage granular lymphocyte-proliferative disorder (NK-GLPD), three patients with aggressive NK cell tumors (one NK cell leukemia and two nasal NK cell lymphoma), and two NK cell lines. By flow cytometric analysis using the monoclonal antibody (MoAb) MRK16 and rhodamine 123 dye (Rh123), P-gp expression and the efflux of Rh123 were found in all NK samples except one NK cell line. The Rh123 efflux of NK cells was inhibited by cyclosporin A (CsA) and its analogue PSC 833, but the aggressive NK tumor cells were less inhibited than were the other NK cells. The percent inhibition of efflux in the normal NK cells, indolent NK-GLPD cells and aggressive NK cell tumors was 81.8% +/- 0. 9%, 93.4% +/- 3.1% and 36.9% +/- 11.7%, respectively, by 1 micromol/L CsA, and 80.2% +/- 3.6%, 91.7% +/- 2.6% and 32.7% +/- 10. 1%, respectively, by 1 micromol/L PSC833. In reverse transcription-polymerase chain reaction (RT-PCR) analysis, the low inhibitory effect of P-gp modulators in aggressive NK cell tumors did not correlate to the expression level of MDR1 gene, multidrug resistance-associated protein gene, or human canalicular multispecific organic anion transporter gene. This phenomenon could be related to the presence of other transporters or to unknown cellular or membrane changes. Some patients with NK cell tumors have been reported to show a highly aggressive clinical course and to be refractory to chemotherapy, and this could be related to the expression of P-gp on NK cells. Our results suggest that, although the inhibitors for P-gp have been used in combination with chemotherapy in some hematologic tumors, these inhibitors may be less effective against aggressive NK cell tumors.

In normal peripheral blood natural killer (NK) cells, the subset of CD16+ CD56+ cells is predominant, and that of CD16+ CD56- cells is rarely present. Because we have found the expansion of CD16+ CD56- NK cells in the peripheral blood of two patients with NK cell-lineage granular lymphocyte-proliferative disorders (NK-GLPD), the clinical findings and cellular characteristics of these patients were compared with those of CD16+ CD56+ NK-GLPD patients. Although CD16+ CD56- and CD16+ CD56+ NK-GLPD cells were morphologically different, clinical findings and courses, and NK activity did not differ significantly. Because strong NK activity was demonstrated in CD16+ CD56- NK-GLPD cells, the CD56 antigen, one of the adhesion molecules, did not seem to play a major role in NK cell-mediated cytotoxicity. The CD56 antigen is known to be more strongly expressed by immature NK cells than by mature NK cells. However, because interleukin 2-activated CD16+ CD56- NK-GLPD cells rapidly expressed the CD56 antigen, the degree of CD56 antigen expression did not always correlate with the maturity of NK cells.