Original Manuscript | Published:

New Cell Line

A novel natural killer cell line (KHYG-1) from a patient with aggressive natural killer cell leukemia carrying a p53 point mutation

Leukemia volume 14, pages 922930 (2000) | Download Citation



We present the establishment of a natural killer (NK) leukemia cell line, designated KHYG-1, from the blood of a patient with aggressive NK leukemia, which both possessed the same p53 point mutation. The immunophenotype of the primary leukemia cells was CD2+, surface CD3, cytoplasmic CD3ε+, CD7+, CD8αα+, CD16+, CD56+, CD57+ and HLA-DR+. A new cell line (KHYG-1) was established by culturing peripheral leukemia cells with 100 units of recombinant interleukin (IL)-2. The KHYG-1 cells showed LGL morphology with a large nucleus, coarse chromatin, conspicuous nucleoli, and abundant basophilic cytoplasm with many azurophilic granules. The immunophenotype of KHYG-1 cells was CD1, CD2+, surface CD3, cytoplasmic CD3ε+, CD7+, CD8αα+, CD16, CD25, CD33+, CD34, CD56+, CD57, CD122+, CD132+, and TdT. Southern blot analysis of these cells revealed a normal germline configuration for the β, δ, and γ chains of the T cell receptor and the immunoglobulin heavy-chain genes. Moreover, the KHYG-1 cells displayed NK cell activity and IL-2-dependent proliferation in vitro, suggesting that they are of NK cell origin. Epstein–Barr virus (EBV) DNA was not detected in KHYG-1 cells by Southern blot analysis with a terminal repeat probe from an EBV genome. A point mutation in exon 7 of the p53 gene was detected in the KHYG-1 cells by PCR/SSCP analysis, and direct sequencing revealed the conversion of C to T at nucleotide 877 in codon 248. The primary leukemia cells also carried the same point mutation. Although the precise role of the p53 point mutation in leukemogenesis remains to be clarified, the establishment of an NK leukemia cell line with a p53 point mutation could be valuable in the study of leukemogenesis.


Large granular lymphocyte (LGL) leukemia can be divided into two major groups, T-LGL leukemia and natural killer (NK)-LGL leukemia.1 T-LGL leukemia presents a mostly indolent and chronic clinical course, a surface CD3 (sCD3)-positive phenotype, and rearrangement of T cell receptor (TCR) genes.123 In contrast, NK-LGL leukemia exhibits an aggressive clinical course, a lack of sCD3 but a cytoplasmic CD3ε (cCD3ε)+ phenotype, and a germline configuration of TCR genes.456789 However, variant forms of these two types of LGL leukemia have been reported.10 An aggressive variant of CD3+, CD56+ T-LGL leukemia is now considered to be the leukemic form of NK-like T cell lymphoma.1112

Recently, the origin of NK cells, especially their close relationship with the T cell lineage, has been vigorously studied, and the evidence shows that there is a common progenitor of T cells and NK cells in the human fetal thymus.13 In this connection, a case of thymic lymphoblastic lymphoma (LBL) with an NK cell precursor origin has been reported.14 Also, the relationship of NK cells to the myeloid lineage has received much attention, and CD7+ and CD56+ myeloid/NK cell precursor acute leukemia has been proposed as a distinct hematological disease entity.15 Thus, as diagnostic concepts are refined, the disease spectrum of NK cell leukemia/ lymphoma has recently been expanding.15 Detailed studies on the clinical and pathological features of such cases contribute to more precise understanding of the emerging clinical spectrum of NK-associated malignancies. Furthermore, the establishment of cell lines from these patients may be of value in studying the cell lineage and the mechanism of NK leukemogenesis.

Since p53 controls the cell cycle checkpoint responsible for maintaining the integrity of the genome, disrupting its function allows tumor cells to avoid the normal restrictions on excessive cell growth.16 Although p53 mutations occur at a lower frequency in hematological malignancies than in solid tumors,17 it has been reported that aberrant p53 function appears to permit the identification of prognostically relevant subgroups of patients with acute myeloid leukemia (AML), myelodysplastic syndrome, and aggressive non-Hodgkin's lymphoma.1819 There have also been reports of correlations between p53 mutation and progressive or relapsing disease in juvenile acute lymphocytic leukemia (ALL),20 adult T cell leukemia,21 and multiple myeloma.22 However, there has been no report of p53 mutation in LGL leukemia or of its correlation with disease progression.

We report here a case of aggressive NK cell leukemia having a CD2+, sCD3, CD7+, CD8+, CD16+, CD56+ immunophenotype with a point mutation in exon 7 of the p53 gene. We also have established an NK cell line (KHYG-1) with the same p53 point mutation by culturing peripheral leukemia cells in the presence of interleukin-2 (IL-2). We believe that this is the first human NK leukemia case from which a cell line with the aberrant p53 gene has been established.

Case report

A 45-year old female visited a local physician for progressive general fatigue and a bleeding tendency, and was then referred to our hospital on 25 July 1997. Her mother had died of breast cancer at the age of 68 and her father of stomach cancer at the same age. Her hemoglobin was 9.9 g/dl, hematocrit 28.7%, platelet count (PLT) 1.4 × 104/μl, and her white blood cell count (WBC) was 45500/μl with 93% lymphocytes and 7% neutrophils. Review of the peripheral blood smears revealed that the lymphocytes were abnormal with an LGL morphology, as described in the Results. Physical examination revealed apparent hepatosplenomegaly, and lymph node swelling was observed at the bilateral submandibular (1 × 1 cm), right axillary (1 × 2 cm), and bilateral inguinal sites (1 × 2 cm). No mediastinal mass was present in this case. Tests for hepatitis B virus surface antigen and antibody (Ab) to hepatitis C-type virus were negative. Anti-nuclear antibody (ANA) was positive (×160; normal <×40), but rheumatoid factor was negative. Serologic tests detected no evidence of recent infection with cytomegalovirus, syphilis, human T cell leukemia virus-1, or human immunodeficiency virus-1 or −2. The EB virus serologic examination showed the following titers: EBV capsid antigen (VCA) IgG, ×160; IgM, ×10; EB nuclear antigen (EBNA), ×20. EBV DNA was not detected in the leukemia cells by Southern blot analysis with a terminal repeat probe from the EBV genome (data not shown). Bone marrow aspiration resulted in a dry tap and a subsequent bone marrow biopsy revealed massive infiltration of abnormal lymphocytes. A lymph node biopsy also showed diffuse infiltration by the same abnormal lymphoid cells as seen in the bone marrow. These infiltrating lymphoid cells in the bone marrow and lymph nodes on paraffin sections were shown to be negative for CD20 (L26, DAKO, Glostrup, Denmark) and CD45RO (UCHL-1, DAKO). However, cytoplasmic CD3ε was immuno-histochemically detected in these cells (data not shown). The patient was then treated with CHOP regimen (cyclophos- phamide, doxorubicin, vincristine and prednisolone). The WBC count decreased in response to therapy; however, the hepatosplenomegaly increased rapidly, and the abnormal lymphocytes increased again in the peripheral blood. Severe thrombocytopenia persisted from the beginning of the clinical course. The patient died on 25 September 1997 of a respiratory hemorrhage. The autopsy revealed the diffuse invasion of leukemia cells into every organ of the body including the skin.

Materials and methods

Cell preparation and culture

Complete medium (CM) was prepared from RPMI 1640 (GIBCO, Paisley, UK), 10% heat-inactivated fetal calf serum (GIBCO), 0.29 mg/ml glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin. Peripheral blood mononuclear cells (PBMC) were isolated from heparinized peripheral blood from the patient by a standard Ficoll–Hypaque density gradient centrifugation. PBMC consisted of 99% leukemia blasts. The cells were washed twice in phosphate-buffered saline (pH 7.4) and incubated in CM containing 100 units of rIL-2 (specific activity: 106 U/mg protein; Shionogi Chemical Industries, Osaka, Japan) at 37°C in a 5% CO2 humidified atmosphere.

Surface phenotype

Surface immunotyping of the primary leukemia or cultured leukemia cells was performed by flow cytometry using a broad panel of monoclonal antibodies (MoAbs) and was confirmed by fluorescence microscopy in some cases. The cells were analyzed on a flow cytometer (Cyto ACE-150 Auto Cell Screener, Japan Spectroscopic Co., Osaka, Japan) for fluorescence intensity using fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)-conjugated antibodies or unconjugated antibodies followed by fluorescein-conjugated goat anti-mouse antibodies (Cappel, Aurora, OH, USA). The MoAbs used are listed in Table 1. Anti-myeloperoxidase (MPO) AM 7, 8, 9, and 10 antibodies were provided by Dr K Suzuki (National Institute of Health, Tokyo, Japan).23 Cytoplasmic antigens and TdT were analyzed as described elsewhere.24 For cytoplasmic CD3ε (cCD3ε) staining, the cells were fixed with 2% paraformaldehyde, permiabilized with 0.1% saponin (Sigma, St Louis, MO, USA), and stained with rabbit anti-CD3ε MoAb (DAKO) followed by FITC-conjugated goat anti-rabbit antibody (Southern Biotechnology, Birmingham, AL, USA).25

Table 1:  Phenotype analysis

Cytotoxicity assay

The NK and lymphokine-activated killer (LAK) cell activity of effector leukemia cells were measured by calcein-AM release assay using NK-sensitive K562 cells or NK-resistant Daudi or Raji cells as targets.26 Briefly, the target cells were labeled with calcein-AM for 30 min at 37°C. Then, the target and effector cells were plated in 96-well plates at the indicated effector-to-target (E/T) ratio and were incubated for 4 h at 37°C in 5% CO2 humidified air. After incubation, the supernatants were transferred to new wells and fluorescence was measured on a Wallac 1420 ARVO fluoroscan (Wallac, Turku, Finland). The percent cytotoxicity was calculated by the following formula: % lysis = (F experiment − F spontaneous)/(F maximal − F spontaneous) × 100. Assays were performed in triplicate.

Cell proliferation

Leukemic cells from the patient were cultured in a 96-well plate (2 × 105 cells/well) in CM containing varying concentrations of rIL-2 or IL-12 (specific activity: 3.57 × 106 U/mg protein; Genetics Institute, Cambridge, MA, USA) for 3 days. For proliferation assays, the colorimetric WST-1 assay was used according to the manufacturer's protocol (Takara Shuzo, Otsu, Japan). Briefly, WST-1 (10 μl of a 5 mM solution in 20 mM HEPES) was added to all wells, and the plates were incubated for 4 h. The absorbance was measured on an ELISA microplate reader (Iwaki, Osaka, Japan) at a wavelength of 405 nm and a reference wavelength of 650 nm. Also, the cell number after incubation with rIL-2 was calculated by trypan blue dye-exclusion test to check the proliferation of the cells.

Detection of cytokines

The patient's serum and the culture supernatants of leukemia cells incubated in 100 units of rIL-2 were assayed for concentrations of IFN-γ, TNF-α and IL-12. The assays were performed by enzyme-linked immunosorbent assay (with a human IFN-γ EASIA kit (BioSource Europe, Fleurus, Belgium) for IFN-γ, a TNF-α measuring kit (Nippon Kotai Institute, Tokyo, Japan) for TNF-α, and a Quantikine human IL-12 immunoassay kit (R&D Systems, Minneapolis, MN, USA) for IL-12).

Cytogenetic analysis

Karyotype analysis using conventional G-banding techniques was performed on the primary leukemia or cultured leukemia cells.

Southern blot analyses of TCR genes and EB virus genome

High molecular weight DNA was obtained by proteinase K digestion, phenol/chloroform extraction, and precipitation by ethanol. After digestion with the appropriate restriction enzymes, 5 μg of DNA was size-fractionated by electrophoresis on a 0.9% agarose gel, denatured, and transferred to nylon membranes (Biodyn B; Pall, Glen Cove, NY, USA) according to standard procedures. The filters were hybridized with specific probes labeled with 32P by random priming, washed in 0.2% standard saline citrate (SSC)/1% sodium dodecyl sulfate (SDS) for 2 h at 55°C, and autoradiographed at −80°C for 2–4 days.

DNA probes

TCR gene rearrangement analysis was performed on EcoRI, EcoRV, BamHI, or HindIII digests according to the method previously reported.27 The EBV-specific DNA sequences were detected on EcoRI and BamHI digests with tandem terminal repeated sequences of EBV genomes.7

Polymerase chain reaction/single-strand conformation polymorphism (PCR/SSCP)

PCR/SSCP was performed as previously reported.28 Genomic DNA was amplified by PCR using p53-specific primer pairs.28 Primers used for PCR were as follows: exon 5f of p53, 5′-TCT GTC TCC TTC CTC TTC CT-3′; exon 5r of p53, 5′-TCT CCA GCC CCA GCT GCT-3′; exon 6f of p53, 5′-TGA TTC CTC ACT GAT TGC TCT-3′; exon 6r of p53, 5′-GAG ACC CCA GTT GCA AAC C-3′; exon 7f of p53, 5′-TCT TGG GCC TGT GTT ATC TC-3′; exon 7r of p53; 5′-AGG GTG GCA AGT GGC TCC-3′; exon 8f of p53, 5′-GCT TCT CTT TTC CTA TCC TGA-3′; exon 8r of p53, 5′-CGC TTC TTG TCC TGC TTG C-3′. The number in each designation indicates the region of the exon of the gene subjected to examination by PCR-SSCP. The designation ‘f’ indicates a sense primer, and ‘r’ indicates an antisense primer. Forty cycles of PCR amplification were performed as follows: denaturation of 94°C for 1 min, annealing at 60°C for 1 min for exon 5, 6, and 7, and at 55°C for exon 8, extension at 72°C for 1.5 min followed by final extension at 72°C for 7 min.

Seven μl of PCR products were diluted with 10 μl of buffer consisting of 20 mM EDTA, 96% deionized formamide, and 5 mg/ml Dextran Blue 2000. Heating denature was performed at 95°C for 5 min, after which samples were placed on ice for 5 min. Then, 16 μl of this solution was applied to each lane of a 7.5% neutral polyacrylamide gel. Electrophoresis was performed at 15 mA in buffer at a temperature of 18°C (3 h for exon 5, 2.5 h for exons 6, 7 and 8). The gel was stained with 2 μg/ml ethidium bromide and visualized under ultraviolet light.

Direct sequencing

DNA fragments amplified by PCR were purified and concentrated with Microcon 100 (Amicon, Beverly, MA, USA), then sequenced with a Taq DyeDeoxy™ Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) and an automated sequencer.29 The primers used for sequencing were the same as those used for PCR-SSCP. To rule out the possibility of errors caused by Taq DNA polymerase, all mutations were confirmed twice.


Establishment of KHYG-1 cell line

The leukemia cells (1 × 105/ml) were cultured in 24-well plates (Corning, Corning, NY, USA) and half the volume of the supernatants was changed twice a week with fresh CM containing 100 units of rIL-2. After 2 weeks, the leukemia cells started to proliferate and they (1–2 × 106) were transferred to a 25-cm2 tissue culture flask (Falcon 3013). The cells proliferated continuously as free-floating single cells in suspension. The maximum cell density was 2 × 106 cells/ml. The optimal split ratio was 1:4 at 0.5 × 106/ml in CM containing 100 units of rIL-2 every 3–4 days. The established cell line was designated KHYG-1. This cell line has been growing continuously for >18 months and its doubling time is around 24–48 h. The cells can be cryopreserved in cryopreservation-medium (90% FCS, 10% DMSO), stored in liquid nitrogen, thawed again (with a viability of more than 70%) and successfully reconstituted. The data presented here were obtained from cells that had been in culture for 18 months since establishment.

Morphology of KHYG-1 cells

The leukemia cells observed in the peripheral blood were relatively large, with a high cytoplasm-to-nucleus ratio (Figure 1a). Nuclei were irregularly-shaped. Fine azurophilic granules were seen in most of the cells by light microscopy (Figure 1a), although they were not obvious as observed in regular LGL. After establishment, KHYG-1 cells showed LGL morphology with a large nucleus, coarse chromatin, conspicuous nucleoli, and abundant basophilic cytoplasm with many azurophilic granules (Figure 1b). These azurophilic granules were negative for MPO by fluorescence microscopy (Table 1).

Figure 1
Figure 1

 (a) Leukemia cells seen in the patient's peripheral blood (May–Gründwald Giemsa-staining) (original magnification × 400). (b) Morphology of established KHYG-1 cell line under a light microscopy (magnification × 400). The KHYG-1 cells showed LGL morphology with azurophilic granules and abundant cytoplasm.

Flow cytometry

The immunophenotype of the primary leukemia cells was CD2+, cCD3ε+, CD7+, CD8α+, CD16+, CD56+, and HLA-DR+ (Table 1). Forty percent of these cells expressed the CD57 antigen. The primary leukemia cells were negative for sCD3, CD4, CD5, CD10, CD13, CD19, CD20, CD33, TCRαβ and TCRγδ (Table 1). Fas antigen (CD95) was expressed in almost all leukemia cells and Fas ligand was weakly positive (12%). KHYG-1 cells showed a similar immunophenotype (CD2+, sCD3, cCD3ε+, CD7+, CD8α+, CD56+ and HLA-DR+) to the primary leukemia cells, although they were CD16, CD33+, CD57 in contrast to the primary leukemia cells (Table 1). KHYG-1 cells were negative for CD1 (Table 1). CD122 (IL-2 receptor β-chain) and CD132 (common γ-chain) were expressed on KHYG-1 cells, but not CD25 (IL-2 receptor α-chain). CD158a (EB6) and GL183 (CD158b), two members of the p58 family expressed by discrete subsets of NK cells, were detected in about 50% of the established KHYG-1 cells (Table 1). KHYG-1 cells were negative for the Vα24 antigen, a human counterpart of mouse NK1.1 expressed on NKT cells.30

Cell proliferation

The growth of leukemia cells was maintained in the presence of 100 units of rIL-2 without feeder cells. The cell number of KHYG-1 cells (original input: 0.5 × 105/ml) increased 10 times (to 5 × 105/ml) after incubation with 100 units of rIL-2 for 7 days. This growth of KHYG-1 cells was IL-2-dependent (Figure 2). IL-12 neither stimulated the growth of this cell line (Figure 2) nor showed any additive effect on its IL-2-dependent growth (data not shown).

Figure 2
Figure 2

 IL-2-dependent proliferation of KHYG-1 cells. The proliferative response was measured by the colorimetric WST-1 assay.


The primary leukemic cells showed a low level of cytotoxicity against K562 cells (13.7% at E/T ratio of 40), which was augmented after incubation with 1000 units of rIL-2 for 24 h (24.7% at E/T ratio of 40). The KHYG-1 cells revealed not only an extremely high level of NK activity against K562 cells, but also LAK activity against NK-resistant Daudi and Raji targets, as shown in Figure 3.

Figure 3
Figure 3

 Cytotoxic activity of the established KHYG-1 cells. The cytotoxicity of KHYG-1 cells against K562, Raji, and Daudi cells is shown at E/T ratios of 12.5, 3 and 1.

Cytokine production

Human NK cells produce a variety of cytokines and hematopoietic growth-promoting factors.31 Although no significant production of IFN-γ, TNF-α, or IL-12 was detected by ELISA in the serum of the patient, a significant level of IFN-γ (2.56 ng/ml) and a low level of TNF-α (9 pg/ml) was detected in the culture supernatants of KHYG-1 cells in the presence of 100 ng/ml of IL-2. When KHYG-1 cells were stimulated with IL-12 (10 ng/ml) in addition to IL-2 (100 ng/ml), IFN-γ production increased slightly (4.06 ng/ml). However, TNF-α was not detected under those culture conditions (<5 pg/ml).


Chromosome analysis of the peripheral blood leukemia cells was not successful because no metaphase cells were obtained. However, the chromosome analysis of lymph node cells showed a complex karyotype of 47, X, −X, add(6) (q2?), −7, add(15)(p11), add(16)(p13), add (20) (q13), +3mar in one of six metaphase cells examined, and 87, XX, −X, −X, +1, −3, −5, add(6) (q2?), −7, −7, −13, −14, −15, add(15)(p11)×2, add(16)(p13)×2, add(20)(q13)×2, −21, +4mar in another one of the six metaphase cells examined. The KHYG-1 cells had a more complex karyotype [92–96(4n)], preserving −5, −7, add(6) (q2?), add(15)(p11), add(20)(q13), and −21, as shown in Figure 4.

Figure 4
Figure 4

 Karyotype of the established KHYG-1 cell line obtained by G-banding technique. The KHYG-1 cells had a complex karyotype. The results of analyses of 10 cells are as follows: [92–96(4n)], +X[10], +X[10], +X[5], del(X)(q2?) x4[10] or i(X)(p10) x4[10], +2[2], −5[10], −5[10], add(6)(q2?)[10], add(6)[8], ?del(6)(q?)[9], ?del(6)[2], −7[6], add(7)(q11) x2[10], add(7)(q22)[10], add(7)[4], −8[10], −8[7], −9[10], −9[8], −10[8], add(10)(p1?)[9], add(10)[2], add(11)(q23)[10], add(11)[8], +12[3], +13[2], −13[3], −14[7], −15[6], add(15)(p11)[10], add(15)[2], −16[10], −16[8], +17[4], −18[10], −18[10], +20[10], +20[3], add(20)(q13) x2[10], add(20)[9], −21[4], +22[2], +6–17 mar. Numbers given in brackets show the number of cells carrying indicated chromosome alterations among 10 cells examined. M: marker chromosomes.

DNA analysis

Southern blot analysis revealed that leukemic cells from the peripheral blood were found to retain the germline configuration for the β and γ TCR and the immunoglobulin heavy chain genes (data not shown). In addition, established KHYG-1 cells were found to retain the germline configuration for the β, δ and γ TCR genes (Figure 5). EBV DNA was not detected in the KHYG-1 cells by Southern blot analysis using a terminal repeat probe from the EBV genome (not shown).

Figure 5
Figure 5

 Southern blot analysis of the DNA extracted from KHYG-1 cells. No rearranged bands were seen with any probes. C: germline control (human peripheral blood lymphocytes from normal subjects), P: KHYG-1 cells.

Detection of point mutation in p53 gene by PCR/SSCP analysis

Most p53 mutations occur in regions of the gene which are highly conserved through evolution, primarily exons 5–8 (codons 126–306).32 PCR analysis with primers spanning the p53 mutation-prone regions and the subsequent SSCP revealed a mobility shift in exon 7 in both primary leukemia cells and KHYG cells (not shown). When PCR products amplified by exon 7-specific primers from KHYG-1 cells and mutation-negative lung cancer cells (a negative control) were applied to the same lane of the polyacrylamide gel, four bands were clearly observed (data not shown), confirming the mobility shift in p53 exon 7 of KHYG-1 cells. These indicated the presence of mutation(s) in the leukemia cells involving p53 exon 7.

Direct sequences of the p53 gene

Direct sequencing of a PCR-amplified DNA fragment of p53 exon 7 showed the conversion of C to T at the first nucleotide (nucleotide 877) of codon 248 in both the primary leukemia cells and the established KHYG-1 cells. This point mutation results in the substitution of tryptophan for arginine. The wild-type nucleotide band at the site of the base substitution was not detected in either specimen, suggesting the absence of wild-type p53 expression. Southern blot analysis was performed to detect alterations in the p53 gene. Southern blots of EcoRI- and BglII-digested DNA hybridized with probes encompassing the entire p53 coding sequence revealed the absence of deletion or rearrangement of the p53 gene (data not shown). The germline p53 mutation of normal tissue (no invasion of leukemia cells) could not be examined in the present case because of the strong invasion of leukemic cells to the whole body. In addition, the examination of other family members for p53 status was not permitted. Therefore, we could not further determine whether this case was Li–Fraumeri syndrome, a hereditary cancer syndrome caused by the inheritance of germline alterations in the p53 gene.33


We present here a case of a 45-year-old female patient with aggressive NK cell leukemia with a p53 point mutation and the establishment of an NK leukemia cell line (KHYG-1) with the same mutation. The diagnosis of aggressive NK cell leukemia was based on the aggressive clinical course and in vitro studies of the leukemia cells, including the sCD3, CD16+, CD56+ immunophenotype, the TCR germline configuration, positive NK cytotoxicity, and IL-2-dependent proliferation.

The characteristics of this case, including the clinical course, resemble the aggressive NK-LGL leukemia/lymphoma reported by Imamura et al.8 However, the leukemia cells in this case expressed CD8αα, a T cell-associated marker, in contrast to Imamura's cases. Also, the KHYG-1 cells established from this patient were positive for CD33, a myeloid antigen, although the fresh leukemia cells did not express it. Therefore, it is necessary to make a careful differential diagnosis with other NK-related leukemias/lymphomas. First, we have to consider the possibility that the origin of the leukemia cells was NK-like T cells, including γδ-T cells. However, the leukemia cells of this case were negative for αβ- and γδ-TCR chains on the cell surface, and they retained the germline configuration for the β, δ and γ TCR genes (Figure 5). The leukemia cells were also negative for the Vα24 antigen which is expressed on NKT cells (Table 1).30 Therefore, it seems unlikely that this case was NK-like T cell leukemia/lymphoma of the NKT or γδ-T cell types. Second, CD7+ and CD56+ myeloid/NK cell precursor acute leukemia is another disease entity to be considered. This type of leukemia has been reported to express a CD34+, CD33+, CD7+, CD2+/−, CD56+, cCD3+ phenotype and to strongly express cytoplasmic MPO.15 Our case showed a phenotype of CD34, CD33, CD2+, CD7+, CD8+, CD56+, cCD3+ and cytoplasmic MPO (Table 1). Thus, the present case shows a mature NK phenotype and is considered to be NK cell leukemia. However, the KHYG-1 cells established from the patient were positive for CD33, a myeloid-associated antigen. This suggests that overlapping or borderline cases might exist between NK cell leukemia and myeloid/NK cell precursor acute leukemia. Third, we must also rule out blastic NK cell leukemia. It is characterized by a lymphoblastoid appearance, often with azurophilic granules, EBV negativity and a tendency to involve the skin, lymph nodes, and bone marrow.34 The leukemia cells of this type of leukemia are usually CD2, sCD3, cCD3+/−, CD5, CD7+/−, CD33, CD34, CD56+ and TdT+/−. Since some of these features were found in the present case, it is difficult to make a clear judgment. However, based on the CD2+, TdT immunophenotype of the leukemia cells, our case seems to be a differentiated form rather than a blastic form of NK cell leukemia. Lastly, thymic lymphoblastic lymphoma with a committed NK cell precursor origin can be ruled out due to the CD34, CD16+, TdT immunophenotype and the absence of a mediastinal mass in the present study.14 Taken together, we consider the present case to be an aggressive NK cell leukemia, although there are some features shared with other NK-related leukemias such as blastic NK cell leukemia or CD7+ and CD56+ myeloid/NK cell precursor acute leukemia.

Although the KHYG-1 cells retained CD3, CD56+ and CD8αα+, they showed a CD16, CD33+, CD57 phenotype whereas the original leukemia cells showed CD16+, CD33, CD57+. The induction of CD13 and/or CD33 has been reported in the culture of T-lineage acute lymphoblastoid leukemia cells.35 Conversely, the antigenic loss of CD16 might occur during the culture of NK cells in the presence of IL-2. However, it was not clear in the present study whether these phenotypic changes were caused by antigenic modulation during culture or by selective growth of the CD16, CD33+, CD57 population which has an advantage in cultures containing IL-2.

We believe that this is the first human NK leukemia case from which a cell line with a p53 point mutation has been established. The p53 gene, which maps to 17p13, is known as a tumor suppressor gene.36 Most non-functioning p53 proteins result from missense mutations of one p53 allele coupled with the deletion of the second one.32 It has been also reported that the mutated p53 may inhibit wild-type p53 and act in a dominant negative fashion even in a heterozygous state.2037 Since the wild-type p53 expression was not detected, p53 function is thought to have been lost or greatly impaired in the present case.

The importance of p53 mutations has been discussed in other hematopoietic cell lines.3839 In a report on the establishment of an NK cell line from a p53 knockout mouse, the authors suggested that genetic intervention of p53 may be a useful strategy for the establishment of cell lines.40 In the present case, it was easy to establish the cell line by incubating leukemia cells in IL-2 containing medium without any other cytokines or feeder cells. Thus, it is possible that the p53 mutation might have played a role in the establishment of the NK cell line, although it should be further examined whether the p53 mutation alone is a prerequisite for the establishment of the NK cell line.

Ross et al recently reported that apoptosis is induced in normal NK cells or NK cell line cells (NK-92)41 cultured with both IL-2 and IL-12 in vitro, and that it is mediated by the high level of IFN-γ and TNF-α produced by these NK cells.42 However, no apoptosis-associated decrease in cell number was observed when KHYG-1 cells were stimulated with both IL-2 (100 ng/ml) and IL-12 (10 ng/ml) for 3 days (data not shown). The most plausible explanation for this observation is that the KHYG-1 cells were able to resist apoptosis because of p53 dysfunction associated with the point mutation.

The important issue is how the aberrant p53 function is involved in the oncogenesis and progression of NK-LGL leukemia. Since the p53-related disruption of the apoptoic pathway may lead to longer survival of lymphocytes, the accumulation of additional genetic lesions is likely to occur.16 In some human malignancies, the activation of oncogene(s) and inactivation of tumor suppressor gene(s) are considered to play a role in tumorigenesis.43 In lymphoid leukemia, it has been suggested that the altered p53 gene and the activated K-ras gene cooperate in leukemogenesis.44 We thus examined alterations in the N-ras, K-ras, retinoblastoma (RB) and interferon responsive factor (IRF) genes. However, no alterations were detected in these genes (data not shown).

Most p53 mutations in leukemia and lymphoma have been reported to occur at CpG dinucleotides, and more than half of the transition mutations are at CpG hot spots (codons 175, 245, 248, 273 and 282), as seen in other solid cancers such as colon and lung cancers.32 In particular, codons 248, 273 and 282 in the p53 protein contact with the target gene of p53. Therefore, mutations in these codons are now regarded as contact mutations which might acquire stronger carcinogenic ability.454647 In the present case, a point mutation was found at one of the CpG hot spots, codon 248 in exon 7 which is located in the L3-loop of the p53 protein. The establishment of an NK leukemia cell line with a contact mutation may serve as a good model for further study of p53-associated leukemogenesis.

In conclusion, we established an NK leukemia cell line (KHYG-1) from the blood of a patient with aggressive NK leukemia both of which possessed the same p53 point mutation. Several other NK leukemia cell lines have been described.4148495051 However, NK leukemia cell lines having a p53 point mutation have not been described. This cell line may provide a suitable model to study the mechanism of NK cell leukemogenesis.

Recently, the guideline for the characterization and description of new cell lines has been published.52 The salient features of KHYG-1 are summarized in Table 2 according to this guideline.

Table 2:  Synopsis of data on new NK leukemia cell line (KHYG-1)


  1. 1.

    . Clonal diseases of large granular lymphocytes Blood 1993 82: 1–14

  2. 2.

    , . Large granular lymphocyte leukemia Medicine 1987 66: 397–405

  3. 3.

    , , , , , , , , , , , , , , . The lympho-proliferative disease of granular lymphocytes Cancer 1987 60: 2971–2978

  4. 4.

    , , , . Aggressive natural killer cell leukemia in an adult and establishment of an NK cell line Blood 1986 67: 925–930

  5. 5.

    , , , , , , , , , . Malignant clonal expansion of large granular lymphocytes with a Leu-11+, Leu-7 surface phenotype: in vitro responsiveness of malignant cells to recombinant human interleukin 2 Blood 1986 68: 1065–1073

  6. 6.

    , , , , , , , . Leukemia of non-T lineage natural killer cells Blood 1988 72: 1701–1707

  7. 7.

    , , , , , , , . CD3-negative lymphoproliferative disease of granular lymphocytes containing Epstein–Barr viral DNA J Clin Invest 1989 84: 51–55

  8. 8.

    , , , , , , , , , , . Aggressive natural killer cell leukemia/lymphoma: report of four cases and review of the literature Br J Haematol 1990 75: 49–59

  9. 9.

    , , , , , , , . Epstein-Barr viral DNA in acute large granular lymphocytes (natural killer) leukemic cells Blood 1992 79: 2116–2123

  10. 10.

    . Classification of natural killer (NK) cell and NK-like T-cell malignancies Blood 1996 87: 1207–1210

  11. 11.

    , , , , , , . CD3+, CD56+ aggressive variant of large granular lymphocyte leukemia Blood 1994 84: 2315–2321

  12. 12.

    , , , , , , . Natural killer-like T-cell lymphomas: aggressive lymphomas of T-large granular lymphocytes Blood 1996 87: 1474–1483

  13. 13.

    , , , , . Identification of a common T/natural killer cell progenitor in human fetal thymus J Exp Med 1994 180: 569–576

  14. 14.

    , , , , , , , , , . Thymic lymphoblastic lymphoma of committed natural killer cell precursor origin Cancer 1996 77: 2592–2603

  15. 15.

    , , , , , , , , , , , , . CD7+ and CD56+ myeloid/natural killer cell precursor acute leukemia: a distinct hematolymphoid disease entity Blood 1997 90: 2417–2428

  16. 16.

    . Defects in a cell cycle checkpoint may be responsible for the genomic instability of cancer cells Cell 1992 71: 543–546

  17. 17.

    , . The clinical significance of mutations of the p53 tumor suppressor gene in haematological malignancies Br J Haematol 1997 98: 502–511

  18. 18.

    , , , , , , . Myelodysplastic syndromes and acute myeloid leukemia with 17p deletion. An entity characterized by specific dysgranulopoiesis and a high incidence of p53 mutations Leukemia 1995 9: 370–381

  19. 19.

    , , , , , , , . Mutations of the p53 gene as a prognostic factor in aggressive B-cell lymphoma New Engl J Med 1997 337: 529–534

  20. 20.

    , , , , , , , , , , , , . Analysis of p53 mutations in a large series of lymphoid hematologic malignancies of childhood Blood 1993 82: 3163–3169

  21. 21.

    , , , , , , . Mutations of the p53 gene in adult T-cell leukemia Blood 1992 79: 477–480

  22. 22.

    , , , , , . p53 gene mutations in multiple myeloma are associated with advanced forms of malignancy Blood 1993 81: 128–135

  23. 23.

    , , , , , , . Preparation and characterization of monoclonal antibodies against human myeloperoxidase Arch Biochem Biophys 1989 273: 189–196

  24. 24.

    , , , , . Flow cytometric analysis of terminal deoxynucleotidyl transferase Am J Clin Pathol 1991 95: 376–380

  25. 25.

    , , , , . Expression of functional molecules by human CD3-decidual granular leucocyte clones Immunology 1996 87: 609–615

  26. 26.

    , , , , . CARE-LASS (calcein-release-assay), an improved fluorescence-based test system to measure cytotoxic T lymphocyte activity J Immunol Meth 1994 172: 227–239

  27. 27.

    , , , , , , , , , . Molecular analysis of cutaneous B- and T-cell lymphomas Blood 1995 86: 3160–3172

  28. 28.

    , , , , , . Mutations in exon 7 and 8 of p53 as poor prognostic factors in patients with non-small cell lung cancer Oncogene 1998 15: 2469–2477

  29. 29.

    , , , , , , , . Mutations of the Btk gene in 12 unrelated families with X-linked agammaglobulinemia in Japan Hum Genet 1996 97: 424–430

  30. 30.

    , . TCR Vα24 and Vβ11 coexpression defines a human NKT cell analogue containing a unique Th0 subpopulation J Immunol 1997 159: 5862–5870

  31. 31.

    , . Biology and clinical relevance of human natural killer cells Blood 1990 76: 2421–2438

  32. 32.

    , , , . p53 mutations in human cancers Science 1991 253: 49–53

  33. 33.

    , , , , . Germ-line transmission of a mutated p53 gene in a cancer-prone family with Li–Fraumeni syndrome Nature 1990 348: 747–749

  34. 34.

    , , , , , , , , , , , , , , , . Clinicopathologic study of CD56 (NCAM)-positive angiocentric lymphoma occurring in sites other than the upper and lower respiratory tract Am J Surg Pathol 1995 19: 284–296

  35. 35.

    , , , , , , , , , . Lineage determination of CD7+ CD5− CD2− and CD7+ CD5+ CD2− lymphoblasts: studies on phenotype, genotype, and gene expression of myeloperoxidase, CD3ε, and CD3δ Am J Hematol 1994 45: 310–320

  36. 36.

    , , . p53 mutations. Gains or losses? J Cell Biochem 1991 45: 22–29

  37. 37.

    , , , , , . Oncogenic forms of p53 inhibit p53-regulated gene expression Science 1992 256: 827–830

  38. 38.

    , , , , , , , , . Frequent mutations in the p53 gene in human myeloid leukemia cell lines Blood 1992 79: 2378–2383

  39. 39.

    , , , , , . P53 alterations in human leukemia–lymphoma cell lines: in vitro artifact or prerequisite for cell immortalization? Leukemia 2000 14: 198–206

  40. 40.

    , , . Ly-49-independent natural killer (NK) cell specificity revealed by NK cell clones derived from p53-deficient mice J Exp Med 1995 181: 1785–1795

  41. 41.

    , , . Characterization of a human cell line (NK-92) with phenotypical and functional characteristics of activated natural killer cells Leukemia 1994 8: 652–658

  42. 42.

    , . Cytokine-induced apoptosis of human natural killer cells identifies a novel mechanism to regulate the innate immune response Blood 1997 89: 910–918

  43. 43.

    . Identifying tumor suppressor genes in human colorectal cancer Science 1990 247: 12–13

  44. 44.

    , , , , , , , . Mutations of the p53 gene in lymphoid leukemia Blood 1991 77: 1153–1156

  45. 45.

    , , . Specific DNA binding by different classes of human p53 mutants Oncogene 1995 11: 763–770

  46. 46.

    , , , , , , , . TP53 DNA contact mutations are selectively associated with allelic loss and have a strong clinical impact in head and neck cancer Oncogene 1998 16: 1671–1679

  47. 47.

    , , , , , . Overexpression of bax associated with mutations in the loop-sheet-helix motif of p53 Am J Pathol 1999 155: 955–965

  48. 48.

    , , , , , , , , , , , . TCGF (IL-2)-receptor inducing factors: regulation of IL-2 receptors on a natural killer-like cell line (YT cells) J Immunol 1985 134: 1623–1630

  49. 49.

    , , , , , . Characterization of a cell line, NKL, derived from an agressive human natural killer cell leukemia Exp Hematol 1996 24: 406–415

  50. 50.

    , , , , , , , , , , , , . Characterization of a novel human natural killer-cell line (NK-YS) established from natural killer cell lymphoma/leukemia associated with Epstein–Barr virus infection Blood 1998 91: 1374–1383

  51. 51.

    , , , , , , , , , . Establishment of an IL-2-dependent cell line derived from ‘nasal-type NK/T-cell lymphoma of CD2+, sCD3, CD3ɛ+, CD56+ phenotype and associated with the Epstein–Barr virus Br J Haematol 1998 103: 669–677

  52. 52.

    , . Guidelines for the characterization and publication of human malignant hematopoietic cell lines Leukemia 1999 13: 835–842

Download references


We wish to thank Drs S Yamamori and H Matsumoto (Mitsubishi Kagaku Bio-Clinical Labs, Tokyo) for DNA sequencing; and Drs S Inoue and O Yoneda (Osaka Dental University) for proliferation assays. We thank the laboratory staff of SRL, Inc. for providing technical support. We also thank Drs E Tatsumi (Kobe University, Kobe, Japan), K Kita (Mie Medical School, Mie, Japan) and K Oshimi (Juntendo University School of Medicine, Tokyo, Japan) for their valuable comments.

Author information


  1. Department of Clinical Immunology and Hematology, Tazuke-Kofukai Medical Research Institute, Kitano Hospital, Osaka, Japan

    • M Yagita
    • , R Tabata
    • , Y Konaka
    •  & K Takatsuki
  2. Department of Thoracic Surgery, Tazuke-Kofukai Medical Research Institute, Kitano Hospital, Osaka, Japan

    • CL Huang
    •  & M Miyake
  3. Department of Internal Medicine, Osaka Dental University, Osaka, Japan

    • H Umehara
  4. Fujisaki Cell Center, Hayashibara Biochemical Labs, Inc., Okayama, Japan

    • Y Matsuo


  1. Search for M Yagita in:

  2. Search for CL Huang in:

  3. Search for H Umehara in:

  4. Search for Y Matsuo in:

  5. Search for R Tabata in:

  6. Search for M Miyake in:

  7. Search for Y Konaka in:

  8. Search for K Takatsuki in:

Corresponding author

Correspondence to M Yagita.

About this article

Publication history






Further reading