Int. J . Cancer: 18, 421-431 (1976)

PROPERTIES OF THE K562 CELL LINE, DERIVED FROM A PATIENT WITH CHRONIC MYELOID LEUKEMIA 2, Hava NEUMANN ’, Peter RALPH*, Jesper ZEUTHEN 5 , Aaron POLLIACK Eva KLEIN Hannah BEN-BASSAT and Farkas VANKY l. Department of Tumor Biology, Karolinska Insfitutet, Stockholm 60, Sweden; Chanock Centre for Virology, Hebrew University, Hadassah Medical School, Jerusalem, Israel; Department of Biophysics, Weizmann Institute of Science, Rehovoth, Israel; Sloan Kettering Institute for Cancer Research, Rye, N . Y. 10580, USA; Institute for Human Genetics, Aarhus, Denmark; and Mayer de Rothschild Hadassah University Huspital Department of Hematology, Jerusalem, Israel.

SUMMARY

The K562 cell line derived from a C M L patient in blast crisis was examined for properties of B and T lymphocytes and cell lines. KS62 lacks the B markers of immunoglobulin, Epstein-Barr virus ( E B V ) genome and associated nuclear antigen, and receptorsfor EB V. A low proportion of cells form rosettes with sheep erythrocytes, the frequency of which is considerably increased after neuraminidase treatment. Unlike B lines but like T lines, K562 cells are lysed rapidly by C’IFc receptor-positive human blood leukocytes and do not stimulate M L C reactions. On the other hand, K562 lacks T antigen, high radiosensitivity and sensitivity to growth inhibition by thymidine. The cells do not contain N-APase, an enzyme found in all lines derived from lymphoid cells and in lymphoproliferative diseases. By scanning electron microscopy, K562 cells were seen to be rounded and relatively smooth, with small numbers of short microvilli resembling undi‘j ferentiated leukemic cells. A few cells had narrow ridge-like profiles and small rufles similar to granulocytic leukemic cells. K562 is strongly positive for immunoglobulin Fc receptors and pinocytosis, but does not phagocytose or mediate antibody-dependent phagocytosis or cytolysis. Among histochrmical stains, KS62 is positive for esterase, lipid, and acid phosphatase. There seems to be no doubt that K562 is not a B cell line. While it has some T cell properties, these are not exclusive. Some of its characteristics indicate that it is probably not lymphoid. Due to its low level of differentiation, its nature cannot be stated with certainty. On the basis of the possible presence of the cellular marker of chronic myeloid leukemia, the Ph chromosome, it may be regarded as belonging to the granulocytic series of cells.

During recent years a large number of human lymphoblastoid cell lines have been established in culture. The majority of these lines represent clones of B cells carrying the EBV (Epstein-Barr virus) genome. The ease with which such cells grow under conventional culture conditions necessitates scrutinisation of the established lines derived

from tumors for properties representing the original malignant cell population (Nilsson and PontBn, 1975). The line K562 was established from pleural effusion during the blast crisis of a chronic myeloid leukemia patient (Lozzio and Lozzio, 1973). The line was reported to carry the Philadelphia chromosome, the cellular marker for this disease, and to lack immunoglobulin and EBV (Lozzio and Lozzio, 1975). On the basis of studies including reactivity with heterologous anti-leukemia sera and treatment with thymosin, K562 does not appear to be related to thymocyte precursors or to B lymphocytes (Ichiki et al., 1975). It seems to be of importance to report additional studies on known markers, as this is the only human cell line which can be regarded as being representative of malignant cells of myeloid origin. MATERIAL AND METHODS

Cell cultures K562 is a cell line established in vitro from a pleural effusion of a patient with chronic myeloid leukemia in blast crisis (Lozzie and Lozzio, 1973). Our laboratory received this line from W. Henle, Children’s Hospital, Philadelphia. The properties of the cell line were compared with those of EBVpositive B lines established from Burkitt lymphoma, Daudi (E. Klein et al., 1968), Raji (Pulvertaft, 1965), Maku (Yata and Klein, 1969), Seraphina (E. Klein et a/., 1975), Namalwa (G. Klein et al., 1972); nasopharyngeal carcinoma, LY28 (established by G . de-The); infectious mononucleosis, Kaplan (Diehl et a/., 1968); lymphosarcoma, LB2 and Hodgkin’s lymphoma, LB12 (Ben-Bassat et al., 1975), two EBV-negative lymphoma B lines, U698 (Nilsson and Sundstrom, 1974) and DG-75 (established by Ben-Bassat); lines with T cell markers, Molt-4, established from ALL (Minowada et al., 1972) and 1022, a transformed marmoset line (Falk et al., 1974) and an undifferentiated line, 1301, presumably

Received: March 18 and in revised form June 11, 1976.

422

KLEIN ET AL.

of T cell origin, a subline of CCRF-CEM (Foley et al., 1965) obtained from B. Hampar, Bethesda, through J. Becker, Jerusalem. YAC, a murine theta-positive line, was also included in the study (Kiessling et al., 1975). Light microscopy and histochemistry

Cytocentrifuge preparations and smears of the cultured cells were stained for lipid (oil red 0), sudan black, peroxidase, non-specific esterase (alpha-naphthyl acetate method), esterase activity (using AS-D chloracetate as substrate), acid phosphatase and periodic acid-Schiff (PAS). Scanning electron microscopy ( S E M )

Cell suspensions from three different culture flasks were centrifuged at l 0 0 g for 5 min and the cell pellet was resuspended in fresh culture medium without fetal calf serum. Next, 6-8 x lo6 cells were collected by filtration onto silver porom membranes of 0.45 pm porosity (Polliack et a/., 1974). The siiver membranes were fixed immediately for at least 24 h with 1 % glutaraldehyde (PH 7.4, 310 mOsm), rinsed twice with phosphate buffer (PH 7.4 310mOsm) and dehydrated in a graded series of alcohol and Freon 113 for 5 min each. The silver membranes were then critically point dried using Freon 13 according to Cohen et al. (1968). Portions of the membrane were attached to stubs using double-sided sticky tape and coated with a thin layer of gold palladium (approximately 30 mm) using a Hummer DC Sputtering System (Technics, Alexandria, Va. 22310, USA). The specimens were stored under vacuum until examination. A Cambridge S4 scanning electron microscope with 100 and 200pm diameter apertures was used at an accelerating voltage of 20-30 kV. Chromosome analysis

Cultures in the log phase of growth, 1 to 2 days after subculture, were treated with Colcemid (Ciba, Basel) at a concentration of 0.1 pg/ml for 2 h and harvested by centrifugation. Cells were treated with 0.56% KC1 for 8 min at room temperature, fixed three times for 30 min in methanol/acetic acid (3:l) and spread on object slides. Preparations were stained with quinacrine mustard for 5 min (Caspersson et a/., 1970) and rinsed and mounted in phosphate buffer/glycerol (1 :I). Metaphases were photographed using a Zeiss photomicroscope equipped for fluorescence microscopy with an HB 200 mercury light source, Kp 500 excitation filter, 520 barrier filter, and a 510 dichroic mirror. Photographic negatives were analyzed by projection at about 4,20Ox, and karyotypes compiled from photographic enlargements (2,400 x ).

Surface receptors

Rosette formation with sheep erythrocytes (SRBC) was performed, according to Jondal et al. (1972). Rosette formation with antibody-coated sheep erythrocytes (A-SRBC) for detection of Fc receptors was performed according to Yoshida and Andersson (1972). Rosette formation with A-SRBC is a property of cells with an Ig receptor, i.e. polymorphonuclear leukocytes, monocytes, B lymphocytes, and some T lymphocytes. Complement receptors were tested similarly using rabbit 19s anti-SRBC and fresh mouse serum. Functional tests

Phagocytosis and cytolysis of antibody-coated sheep erythrocytes and antibody-coated chicken erythrocytes were tested as described previously (Ralph and Nakoinz, 1975; Kiessling et al., 1975). Immunoglobulins and

p2 microglobulin

Viable cells were exposed to rabbit sera with specificity for human gamma and mu heavy chains, light chains and pa microglobulin (Dakopatts, Copenhagen, Denmark). After washing, antibody attachment was tested by incubation with FITCconjugated goat anti-rabbit immunoglobulin serum (Hyland Laboratories, USA). The procedure for living cells detecting surface reactivities was described previously (E. Klein et al., 1968). For intracellular immunoglobulin acetone-methanol fixed smears were stained. Binding of fluorescent Concunavalin A (Con A )

One million cells were incubated with 0.5 ml fluorescent Con A (Miles-Yeda, Rehovoth, Israel), at 10Opg/ml, for 15 min a t 37" C; the cells were washed with phosphate-buffered saline, Single cells or small aggregates (2-5 ceils) were scored for cap formation (Ben-Bassat and Goldblum, 1975). Agglutination by Concanavalin A

One-half ml of Con A at different concentrations diluted in phosphate-buffered saline was mixed with 0.5 ml cell suspension. The cell concentrations were lo6, 2 x lo6 and 5 x lo6 cells per ml in a 35-mm Petri dish. The density and size of aggregates were scored visually on a scale from - to after a 30-min incubation at 24" C with occasional shaking (Ben-Bassat and Goldblum, 1975).

+++

EBV-determined nuclear antigen (EBNA) and EBV receptor

EBNA was tested by anticomplement immunofluorescence (Reedman and Klein, 19731, using several known positive sera. The detection of cell surface receptors and sensitivity for EBV infection

CELL LINE DERIVED FROM MYELOID LEUKEMIA

423

was performed according to methods described before (Klein et al., 1972). EBV is known to infect B cells exclusively.

positive for a-napthylacetate esterases. Many cells were strongly positive for acid phosphatase.

Lymphocyte blastogenesis

Mixed cultures were set up according to Vanky et al. (1974). Incorporation of 3H-thymidine was measured during a 16-h pulse. Blastogenesis was evaluated also by mixing the cultures with sheep erythrocytes and scoring the rosettes for size.

Many K562 cells were rounded, with relatively smooth surfaces and small numbers of short stublike rnicrovilli, while a few cells showed occasional surface blebs (Fig. 1). A moderate number of cells had narrow ridge-like profiles and small ruffled membranes similar to those seen in leukemic cells of the myeloid series (Polliack et al., 1975).

Radiosensitivity

Chromosome analysis

Growing cell cultures were split into five aliquots. One was kept as a control, the others were irradiated with various doses (100-1,0oO rad). Duplicate flasks were set up and their viability was determined 1 and 2 days after incubation.

Chromosome examination of the K562 line showed it to be of a near-triploid karyotype with 68 to 73 chromosomes, and a modal number of 70 chromosomes (Fig. 2). Chromosome analysis of quinacrine mustard stained metaphases identifies the karyotype as follows: 70, x x x , -13, -17, f7, +9 (pl l), small metacentric chromosome, del ( x ) (p21), del (3) (pll), del (91 (pll), del (9) (pll), t (15; 18) (q21; q23), r (22) (Fig. 3). The t (15; 18) chromosome involving the long arm of one of the three chromosomes no. 15 was found in all metaphases analyzed. The part proximal to band q21 of the long arm of chromosome no. I5 was deleted. In all metaphases of good quality analyzed it was possible to identify the small ring chromosome r (22), which most probably is derived from a deleted chromosome no. 22 constituting the Philadelphia chromosome, the cellular marker of the chronic myeloid leukemia cells from which this line was established. The karyotype reported here for the K562 myelogenous

SEM

+

Alkaline phosphatase

The cells were first extracted with 0.1 M Tris-HC1, PH 8.0; thereafter the insoluble material was extracted with 1-butanol in order to estimate the membrane-bound enzymes (Neumann et al., 1971). The efficiency of catalysing the hydrolysis of p-nitrophenyl phosphate and cysteamine-S-phosphate (pNPP and CASP) was compared. The ratio of the rate of hydrolysis of the two substrates was calThe N-phosphatase culated, R = (Vp-NPP)/(VCASP). activity, i.e. the enzyme which preferentially hydrolyses p-NPP, was expressed as percentage of the total alkaline phosphatase. This enzyme has been found in the sera of patients with lymphoid malignancies and infectious mononucleosis but not in those of patients with myeloid leukemia (Neumann et a/., 1974). RESULTS

Light microscopy

The K562 cells varied in size, ranging from 15 to 20pm in diameter. They were rounded or oval and undifferentiated in type. The nuclear to cytoplasmic ratio was high. The cytoplasm was intensely basophilic and prominent vacuoles were seen in some of the cells. Azurophilic granules were rarely encountered, while some cells showed cytoplasmic buds. The size of the nuclei was variable, most were rounded or indented, the nuclear chromatin was finely dispersed and a varying number of prominent nucleoli were seen. The cells exhibited strong pinocytosis of neutral red. Histochemistry

The following stains were negative: Sudan black, peroxidase, PAS, AS-C chloroacetate esterase. Some cells were oil red 0 positive and weekly

FIGURE 1 K562 cell showing irregular surface with ridge-like profiles and a larger, single ruffled membrane ( x 8,650).

424

KLEIN ET AL.

Lymphocyte markers

10

K562 cells were negative for surface immunoglobulin and EBV receptors. A low percentage of cells (5-973 formed rosettes with SRBC. Only 3-9% of the cells were positive for complement receptors as measured by EAC rosettes. Ninety to 95% of the cells gave rosettes with antibody-coated sheep erythrocytes, indicating the presence of Fc receptors. This interaction did not lead to phagocytosis or lysis of the antibody-coated erythrocytes, nor did the K562 cells function as effectors in the lysis of antibody-coated chicken erythrocytes.

m W

m8

a

I

a. 2 6 W I L

0 4

a i

Binding of F-Con A

2

n n

3

z

0

71 7 2 73 7 4

C H R O M O S O M E NUMBER FJGURE 2

Distribution of chromosome numbers in random metaphases of the K562 myeloid leukemia cell line (22 metaphases counted on photographic negatives).

leukemia line based on quinacrine banding differs from the karyotype reported previously (Lozzio and Lozzio, 1975; Lozzio el al., 1976) in that we find the translocation chromosome to be a t (15; 18) translocation and that we identify the Philadelphia chromosome-like marker as a ring chromosome r (22), but in other respects the karyotype corresponds exactly to that reported previously.

Pa microglobulin The reactivity with the anti$, microglobulin reagent tested by indirect fluorescence was extremely weak and, due to the presence of Fc receptors on the cells, its specificity was questionable. Thus K562 either lacks or carries low amounts of this surface moiety detected on all cell lines except the Daudi (Evrin and Nilsson, 1974). In a subsequent paper (Zeuthen et al., to be published) we shall report on the correlation between the reduced level of Bp microglobulin expression and abnormalities in the long arm of chromosome No. 15 in lymphoid cell lines.

Ninety-five to 100% of the K562 cells bound Con A in a specific way since it was completely inhibited by alpha-methyl-D-mannopyranoside.The fluorescence pattern was unlike that seen with a number of lymphoblastoid cell lines: after incubation the stain was concentrated in a round body, the rest of the cell being entirely negative. The picture seemed to indicate a prepinocytosis stage, with part of the stain being engulfed by the cell. This was in line with the observation that pinocytosis of neutral red was prominent in K562 cells. Cells incubated with F-Con A at 0" C gave continuous staining without clusters, caps or the " pre-pinocytosis " stage. Upon subsequent incubation at 37" C for 30 min these forms appeared. Agglutination by Con A

Agglutination of K562 cells by Con A at a concentration of lo6 and 2 x lo6 cells/ml was weak or absent (Table I). Five million celis/ml were agglutinated by high concentrations of Con A ( >50 ,ug/ml). This behavior distinguished the K562 line from the majority of well-characterized cell lines as these aggregated when tested under similar conditions (Table I). When the growth conditions of the K562 cells were altered by starting the cultures with higher cell density the cells were agglutinated readily by Con A. Pretreatment of the cells with trypsin (25; 2.5 or 0.25 ,ug/ml) did not alter agglutinability. Lymphocyte blastogenesis

Unlike B cell lines, K562 cells did not stimulate the DNA synthesis of allogeneic lymphocytes (Table TI). In this respect the 1301 line was similar.

FJCURE 3 b

Representative karyotype of thc K562 myeloid leukemia cell line. The karyotype is near-triploid 70, XXX, 13, -17, -17, -1 9 (pll), I small metacentric chromosome, del (X) (p21), del (3) (pll), del (9) ( p l l ) , del (9) (pl I ) . t(l5; 18) (q21; q23), r(22). The characteristic marker chromosomes are identified by arrows. Note the small r(22) chromosome probably derived from a deleted chromosome No. 22 constituting the Philadelphia chromosome, the marker of chronic myeloid leukemia cells from which the line was derived. ~

CELL LINE DERIVED FROM MYELOID LEUKEMIA

425

Herpes virus ateles transformed marmoset lymphocyte

Acute leukemia

1022

1301

4 5 6

3

1 2

Lymphocyte donor

128 122 896 866 376 324

Control (identical lymphocytes) cpm

Lymphosarcoma

DG-75

{ l-cell EBV-negative undefined

B cell B cell B cell B cell B cell B cell B cell B cell EBV-nega t ive B cell EBV-negative

Propertles 0

TABLE I1

1 .

+ + + + + f ++

5

7-

-

+

+++

++ +

T

+ ++

10

~

+++ +++ ++++ ++ ++ ++ +++ ++

50

Concentration of Con A &ml

122 118 1,130 1,065 511 266

cpm

K562 cpm

162 157 1,525 1,104 531 277

RI

1.o 1.o 1.3 1.2 1.4 0.8

MOLT-4

cell lines

1.3 1.3 1.7 1.3 1.4 0.9

RI

325 2,084 4,08 1 5,988 12,275

550

cpm

Daudi

4.3 2.7 2.3 4.7 15.9 37.9

RI

INCORPORATION OF 8H-THYMIDINE BY LYMPHOCYTES CO-CULTIVATED WITH CELL LINES

Lymphosarcoma

U 698

K562

Kaplan Daudi Raji Maku Ly-28 LB-2 LB-10

Origin

Blast crisis of chronic myeloid leukemia Infectious mononucleosis Burkitt lymphoma Burkitt lymphoma Burkitt lymphoma Nasopharyngeal carcinoma Lymphosarcoma Hodgkin’s disease

Line

-~

AGGLUTINABILITY OF LYMPHOBLASTOID CELL LINES BY CON A

TABLE I

f

cpm

12,854 19,302

-

5,373 18,023 248,835

++

f++

++++ ++++ ++++ +++ +++ +++ ++++

100

PHA

42.0 147.7 278.3 34.2 59.6

RI

-

++

+T+-

+-++ +++-

+?++ +-+

+-++-++

I+T+f

2.50

P

P

z

PN 6

427

CELL LINE DERIVED FROM MYELOID LEUKEMIA

As expected, the Molt-4, a T-cell line, did not stimulate. In the same experiments confrontation withthe Daudi line induced increased 3H-thymidine incorporation. The lack of blast transformation was confirmed by inspection of the E rosette-forming cells at varying intervals after initiation of the mixed cultures. Radiosensitivity

The effect of irradiation, as assayed by viability testing, is presented in Table 111. The Molt4 cells were highly sensitive. YAC cells, a murine T cell leukemia, were somewhat less sensitive. The Raji culture-a B cell line-was inhibited in its cell multiplication, but the viability of the culture was unaffected and the cells increased in size indicating a continuation of protein synthesis. The same was true for the 1301 and K562 cell lines which showed good viability and cell size increase even after the high doses.

Alkaline phosphatase activity

Alkaline phosphatase activity was detected in all cell lines tested. When two substrates were used, however, differences were found. The lymphoblastoid lines hydrolysed with greater efficiency p-nitrophenylphosphate, thus meeting the criterion for the presence of N-APase. This was not the case with the K562 line which did not differentiate between the two substrates, indicating thus the lack of N-APase (Table IV). DISCUSSION

Human lymphoblastoid cells can be easily characterized for B or T properties if they have immunoglobulin on the cell surface, or if they form rosettes with sheep erythrocytes (Jondal et al., 1973). However, some lines have very low amounts of immunoglobulins, detectable only by very sensitive

TABLE Ill RADIOSENSITIVITY OF CELL LINES MEASURED BY CELL SURVIVAL

Cell line

Percentage of dead cells dose of irradiation

Days post irradiation

Raji

1

YAC MOLT-4 1301 K562

Raji

2

YAC MOLT-4 1301 K562

-

100 rad

250 rad

500 rad

1,000 rad

2 3 15 8 14 2 11 13 9 13

4 3 37 21 29 0 15 95 13 16

3 23 85 20 28 0 54 100 6 26

5 33 98 20 16 1 85 100 18 15

4 23 98 9 26 23 89 100 26 25

TABLE IV ALKALINE PHOSPHATASE IN EXTRACTS OF HUMAN LYMPHOBLASTOID CELL LINES

Line

K562

Daudi Raji Maku Seraphina Namalwa Kaplan LY28 1301 1

Specific activity* p-NPP

Specific activity* CASP

'

% N-APase'

Tris

Butanol

Tris

Butanol

Tris

Butanol

Tris

Butanol

59.0 23.0 64.0 20.0 28.0 5.1 14.0 16.6

32.0 12.0 36.0 0 14.0 12.8 10.0 1.2 19.5

50.0 1.1 0.5 0.8 3.0 0.3 0 0.4 -

31.0 0.2 1.5 0 1.o 0.2 0 0.2 0.7

1.2 22.0 128.0

1.03 6.0 23.8

9.3 27.0

14.0 61.0

44.0

6.0 26.5

0 90 98 95 85 95 100 90

0 80 96 90 98 100 75 93

-

25.0

-

-

Several cell cultures of each cell line were tested. Representative results are given.-* Specific activity is exprcssed as my mole substrate

hydrolysed per minute per loEcells.--l

% N-APase

=

):-

(1

100. considering the R value for normal APase 1.6 (Neumann er al. 1974).

428

KLEIN ET AL.

methods. Other exclusive markers for B cells are related to EBV: cell-surface receptor for the virus, the presence of viral genome and the virally determined nuclear antigen (Jondal and Klein, 1973 ; G . Klein, 1975). Only a few B cell lines exist which are devoid of EBV genome, but all except one of these carry receptors for EBV and can be superinfected (G. Klein et al., 1974a, b; G. Klein, 1975). The exceptional B line is DG75 which lacks EBNA as well as EBV surface receptors (unpublished experiments). A few myeloma cell lines have been described in which intracellularly localized immunoglobulin is demonstrable (Nilsson and PontCn, 1975). Four lines are known which are derived from histiocytic lymphoma (Epstein and Kaplan, 1974; Sundstrom and Nilsson, 1976). The problem arises with characterization of lines which cannot be classified as B cells, but also lack the capacity to rosette spontaneously with sheep erythrocytes. These lines can still be T-derived as rosette formation often declines or is lost during prolonged propagation. Other properties of T cell lines include a T cell surface antigen detected with specific heterologous sera, growth inhibition by 1 0 - 4 ~ thymidine, toxic effect of low amounts of phytohemagglutinin (Ralph, 1975) and relatively high radiosensitivity (Han et al., 1974). These latter markers persist in T cell lines even if they have lost sheep erythrocyte rosetting capacity (unpublished). Terminal deoxynucleotidyl transferase, an enzyme found in thymus but not in bone marrow, can also be used as a marker for T cell lines (McCaffrey et a/., 1975). Another characteristic which differentiates B and T cell lines is their behavior in mixed lymphocyte cultures. While B cells are good stimulators, T lines lack this capacity (Pauly et al., 1975). In addition, certain T cell lines were found to be quickly killed by a subpopulation of normal blood leukocytes; B cells are also sensitive to this effect though with different kinetics (Jondal and Pross, 1975). Alkaline phosphatase activity with unique substrate specificity was discovered in tissues of mice with lymphatic leukemia (Neumann et al., 1971). This enzyme, denoted N-APase, was found also in the sera of patients with lymphoproliferative disorders, such as acute and chronic lymphatic leukemia and infectious mononucleosis (Neumann el a/., 1974), but was not detected in lymphocytes, even after mitogen stimulation, or in sera or cells of A M L or CML patients (Neumann el a/., 1974, 1975). N-APase was also absent from cells of mice with myeloid leukemia or reticulum cell sarcoma, in spite of considerably elevated alkaline phosphatase activity (Neumann et al., 1971). The characteristics of the K562 line are summarized in Table V.

With regard to the lymphocyte parameters, the K562 cell line lacks properties of B lines, such as surface immunoglobulin, EBV-associated nuclear antigen, EBV genome and receptors for EBV. It also lacks the T antigen, sensitivity to growth inhibition by thymidine, and the high sensitivity to radiation. In contrast to B cell lines but in common with T lines, K562 cells are Iysed rapidly by normal peripheral blood leukocytes (spontaneous killing) and d o not stimulate MLC reactions. A few cells rosette with sheep erythrocytes. After neuraminidase treatment the frequency of rosettes is considerably elevated (Galili and Klein, unpublished experiments). The surface architecture of K562 cells as seen by SEM was similar to that of undifferentiated leukemic cells, while other cells exhibited surface ridges and small ruffles like those seen in granulocytic leukemic cells (Polliack et al., 1975). The K562 line lacks N-APase, which is present in CLL and in all lymphoblastoid lines tested. This finding correlates well with the absence of N-APase from AML sera or blasts collected from the blood of these patients. Also, K562 cells agglutinate weakly with Con A, a property which is shared with AML blasts (Ben-Bassat, unpublished experiments), and conforms to origin from myelogenous leukemia of mice (Inbar et al., 1973).

TABLE V CHARACTERISTICS OF THE KS62 LINE

SRBC receptor

Reactivity with anti-T serum Surface Ig (intracellular also) Fc receptor C3 receptor EBV receptor, genome, EBNA Blastogen. in MLC Surface & figlobulin Con A agglutinability Sensitivity to cytotoxic effect of human lymphocytes Sensitivity to PHA Growth inhibition by thymidine Radiosensitivity Phagocytosis Pinocytosis K cell function Alkaline phosphatase N-APase Acid phosphatase Lysozyme production PAS Peroxidase Sudan black B a- naphthylacetate esterase Oil red 0 AS-D chloracetate esterase

Low, increased after neuraminidase treatment

Negative Negative Positive Negative Negative Negative Negative Weak High Negative Negative Low Low Positive Negative Positive Negative Positive Negative Negative Negative Negative Positive Positive Negative

CELL LINE DERIVED FROM MYELOID LEUKEMIA

The characteristic movement of the Con-A receptor on the K562 cell surface resulting in accumulation of Con A in a well-circumscribed small area is in line with its low agglutinability. It has been reported that the two properties, " capping" and agglutinability, correlate in such a way that cells which readily cap are usually only weakly agglutinated (Ben-Bassat and Goldblum, 1975). K562 has thus many properties in common with myelogeneous leukemia cells. Some of the K562 cells had small amounts of non-specific esterases and lipid (oil red 0 positive) while many cells had acid phosphatase. Although the presence of these enzymes suggests a non-lymphoid origin of the K562 cell, these are not absolute discriminating markers as they can also be found in lymphoid cells. Recently acid phosphatase has been demonstrated in T-type acute lymphoblastic leukemic cells and in T-type chronic lymphocytic leukemic cells (Catovsky et al., 1975), while it was absent in cases of null type ALL and in most B-derived leukemic lymphocytes. However, acid phosphatase has also been described in non-lymphoid leukemias, being distributed in a more diffuse pattern (Flandrin and Bernard, 1975). In spite of the presence of Fc receptors, K562 does not phagocytose or lyse antibody-coated cells. Production of lysozyme and myeloid colony stimulating factor was not detected (Ralph et al., to be published) although these would be expected of myeloid or monocytoid cells. Since the cell line was received through an intermediate laboratory, its identity with the original line could be questioned. However, the close correspondence in the karyotypes of the cell line used in the present studies and in that published previously (Lozzio and Lozzio, 1975; Lozzio et al., 1976) confirms the identity of our line with the original K562 line. BZ p-globulin has been demonstrated on the cell surface of all human cell lines tested, except Daudi (Evrin and Nilsson, 1974). Bz p-globulin

429

has been shown to be a subunit of purified HL-A preparations and correspondingly in a " co-capping " experiment the two moieties moved in association on the cell surface. Daudi lacks P2 p-globulin and does not express HL-A. It reacts with certain human sera, probably due to the presence of anti-Ia activity. The reactivity pattern of K562 is similar, i.e., it has no definable HL-A components (Brautbar and Terasaki to be published). Another ceIl line received by us under the code 1301, derived from a child with acute leukemia, has not been fully tested. Presently, it does not form rosettes with sheep erythrocytes (West and Herberman, 1974) and does not react with an anti-human T cell serum (Galili and E. Klein, unpublished experiment). Except for the relatively low radiosensitivity, the properties of 1301, i.e. lack of surface immunoglobulin, EBNA, EBV genome and receptors, and stimulation in MLC, correspond to T characteristics. The presence of N-APase classifies it as a lymphoid line. In view of the difficulties in the morphological and cytochemical characterization of certain poorlydifferentiated leukemias and lymphomas, the new developments concerning lymphocyte markers are considered to provide important contribu.ions to the panel of diagnostic criteria. Our findings indicate that the studies on alkaline phosphatase may provide an additional marker to differentiate cells of lymphoid origin. ACKNOWLEDGEMENTS

The work upon which this work is based was performed pursuant to Contract NO1-CB-64023 and Contract 1-CP-3342 with the Division of Cancer Biology and Diagnosis, National Cancer Institute, Department of Health, Education and Welfare, and the Swedish Cancer Society, US National Science Foundation Grant BMS 75-19734 and grants from the Behr-Lemsdorf and Leukemia Research Foundations.

PROPRIETES DE LA LIGNEE K562, QUI PROVIENT D'UN SUJET ATTEINT DE LEUCEMIE MYELOTDE CHRONIQUE La lignie K562, qui provient d'un sujet atteint de CML en crise blastique, a it6 itudi& du point de vue des propriitis des lymphocytes B et T et d'autres lignies cellulaires. Elk ne possdde ni les marqueurs B de I'immunoglobuline, ni le ginome du virus d'Epstein-Barr (EBV), ni l'antigdne nucliaire qui h i est associi, ni les ricepteurs de I'EB V . Unefaible proportion des cellulesforment avec les irythrocytes de mouton des rosettes dont la friqueirce s'accroit considerablement aprds traitement a la neuraminidase. A la difkrence des lignies B mais comme les ligntes T, les cellules K562 sont lysies rapidement par les leucocytes du sang humain portant des rkcepteurs C'IFc et ne stimulent pas les rtactions en MLC. Par contre, elks nepossbdentpas l'antigbne T e t ne sont pas fortement radiosensibles; leur croissance n'est pas inhibke par la thymidine; elles ne contiennentpas de N-APase, une enzyme prisente dans toutes les lignies provenant de cellules lymphoides et dans les maladies Iymphoproliferatives.Au microscope dectronique Ci balayage, on constate que les cellules $362 sont arrondies,

430

KLEIN ET AL.

relativement lisses, avec quelqiies courtes microvillosites qui les font ressembler it des cellules Eeuckmiques indifkrencikes. Quelques cellules ont des protubkrances en forme d’aretes ktroites et de petites rides comme les cellules leucemiques granulocytaires. Pour les rkcepteurs Fc de I’immunoglobuline et la pinocytose, la lignke K562 est fortement positive. Elk ne phagocyte pas et ne dkclenche pas de phagocytose ou de cytolyse dkpendant de I’anticorps. Dans les tests de coloration histochimique, on a constatk que la lignke K562 est positive pour l’estkrase, les lipides et la phosphatase acide. I1 ne s’agit incontestablement pas d’une lignie de cellules B. Si elk a quelques-unes des propriktks des cellules T, e l k a aussi d’autres caractkristiques, dont certaines indiquent qu’elle n’est probablement pas lymphoide. Du fait de son faible niveau de diffkrenciation, on ne peut pas dkterminer sa nature avec certitude. En raison de la prksence possible du marqueur cellulaire de la leuckmie mykloide chronique, le chromosome Ph, on peut considkrer qu’elle appartient a la skrie granulocytaire.

REFERENCES

BEN-BASSAT,H., and GOLDBLUM,N., Concanavalin A receptors on the surface membrane of lymphocytes from patients with Hodgkin’s disease and other malignant lymphomas. Proc. nut. Acad. Sci. (Wash.), 72, 1046-1049 (1975). BEN-BASSAT, H., GOLDBLUM, T., MITRANI,S . , BENTWICH, Z., and GOLDBLUM, N., Concanavalin A receptors and other cell surface and antigenic characteristics of continuous lymphoblastoid cell lines derived from patients with Hodgkin’s disease and other malignant lymphomas. Frog. med. Yirol., 21, 177-187 (1976). CASPERSSON, T., ZECH,L., JOHANSSON, C., and MODEST,E. J., Identification of the human chromosomes by DNA reacting fluorescing agents. Chromosoma, 30, 215-227 (1970). CATOVSKY,D., T-cell origin of acid phosphatase-positive lymphoblasts. Lancer, 2, 327-328 (1975). COHEN,A. L., MARLOW,D. P., and GARNER, G. E., A rapid critical point method using fluorocarbons (“ Freons ”) as intermediate and transitional fluids, J , Microscopie, 7 , 331-342 (1968). DIEHL,V., HENLE,G., HENLE,W., and KOHN,G., Demonstration of a herpes group virus in cultures of peripheral lymphocytes from patients with infectious mononucleosis. 1. Yirol., 2, 663-669 (1968). EPSTEIN,A. L., and KAPLAN,H. S . , Biology of the human malignant lymphomas. I. Establishment in continuous cell culture and heterotransplantation of diffuse histiocytic lymphomas. Cancer, 34, 1851-1872 (1974). EVRIN,P. E., and NILSSON,K., p 2 microglobulin production in virro by human hematopoietic mesenchymal and epithelial cells. J. Immunol., 112, 137-144 (1974). FALK,L., WRIGHT,J., WOLFE,L., and DEINHARDT, F., Herpesvirus ateles. Transformation in vitro of marmoset splenic lymphocytes. Int. J. Cancer, 14, 244-251 (1974). FLANDRIN, G., and BERNARD, J., Cytological classification of acute leukemias. Blood C e h , 1, 7-15 (1975). FOLEY, G . E., LAZARUS,H., FARBER,S., UZMAN,B. G., BOONE,9. A., and MCCARTHY,E., Continuous culture of human lyrnphoblasts from peripheral blood of a child with acute leukemia. Cancer, 18, 522-529 (1965). HAN, T., PAULY,J. L., and MINOWADA, J., I n vitro preferential effect of irradiation on T lymphoid cell line. Clin. exp. Immunol., 17,455-462 (1974). ICHIKI,A. T., BAMBERGER, E. G., LOZZIO,C. B., LOZZIO,B. B., and GIRARDI, A. J., Further characterization of the chronic myelogeneous leukemic (CML) cell line K-562, with the Philadelphia chromosome. IRCS med. Sci. (Suppl.), 3, 12 (1975). INBAR,M., BEN-BASSAT, H., FIBACH, E., and SACHS,L., Mobility of carbohydrate-containing structures on the surface membrane and the normal differentiation of myeloid leukemic cells to macropbages and granulocytes. Proc. nut. Acad. Sci. (Wash.), 70, 2577-2581 (1973). JONDAL,M.,HOLM,G., and WIGZELL,H., Surface markers on human T and B lymphocytes. I. A large population of

lymphocytes forming nonimmune rosettes with sheep red blood cells. J. exp. Med., 136, 207-215 (1972). JONDAL,M., and KLEIN,G., Surface markers o n human B and T lymphocytes. 11. Presence of Epstein-Barr virus (EBV) receptors on B lymphocytes. J. exp. Med., 138, 1365-1378 (1973). JONDAL,M., and PROSS,H., Surface markers o n human B and T lymphocytes. VI. Cytotoxicity against cell lines as a functional marker for lymphocyte populations. In?. J . Cancer, 15, 596-605 (1975). JONDAL, M., WICZELL,H., and AUTI,F., Human lymphocyte subpopulations : classification according to surface markers and/or functional characteristics. Transplant. Rev., 16, 163-195 (1973). KIESSLING,R., KLEIN, E., PRoss, H., and WICZELL,H., Natural ” killer cells in the mouse. 11 Cytotoxic cells with specificity for mouse Moloney leukemic cells. Characteristics of the killer cell. Europ. J. Immunol., 5, 117-121 (1975). KLEIN,E., KLEIN,G., NADKARNI, J. S., NADKARNI, J. J., WIGZELL, H., and CLIFFORD, P., Surface IgM kappa specificity on a Burkitt lymphoma cell in vivo and in derived culture lines. Cancer Res., 28, 1300-1310 (1968). KLEIN,E., NILSSON, K., and YEFENOFF, E., An established Burkitt’s lymphoma line with cell membrane IgG. Clin. Immunol. Immunopath., 3, 575-583 (1975). KLEIN,G., Studies on the Epstein-Barr virus genome and the EBV-determined nuclear antigen in human malignant diseases. Cold Spr. Harb. Symp. quant. Biol., 39, 783-790 (1975). KLEIN, G., DOMBOS,L., and GOTHOSKAR, B., Sensitivity of Epstein-Barr virus (EBV) producer and non-producer human lymphoblastoid cell lines to superinfection with EB-virus. Int. J. Cancer, 10, 44-57 (1972). KLEIN, G., GIOVANELLA, B., WESTMAN,A,, STEHLIN,J., and MUMFORD,D., An EBV genome negative cell line established from an American Burkitt lymphoma; receptor characteristics, EBV-infectability and permanent conversion into EBV positive sublines by in vitro superinfection. Intervirology, 5 , 319-334 (1976). KLEIN, G., LINDAHL,T., JONDAL, M., LEIBOLD,W., MENEZES, J., NILSSON,K., and SUNDSTROM, C., Continuous lymphoid cell lines with B-cell characteristics that lack the Epstein-Barr genome, derived from three human lymphomas. Proc. nut. Acad. Sci. (Wash.), 71, 3283-3286 (1 9744. KLEIN, G., SUGDEN,B., LEIBOLD,W., and MENEZES,J., Infection of EBV genome negative and positive human lymphoblastoid cell lines with biologically different preparations of EBV. Intervirology, 3,232-244 (1974b). Lozzro, C. B., and Lozzro, B. B., Cytotoxicity of a factor isolated from human spleen. J. nab. Cancer Inut., 50, 535-538 (1973). LOZZIO,C. B., and LOZZIO,B. B., Human chronic myelogenous leukemia cell line with positive Philadelphia chromosome. Blood, 45, 321-334 (1975).

CELL LINE DERIVED FROM MYELOID LEUKEMIA

LOZZIO,B. B., MACHADO, E. A., LOZZIO,C. B., and LAIR,S., Hereditary asplenic-athymic mice transplantation of human myelogenous leukemic cells. J. exp. Med., 143, 225-231 (1976). MCCAFFREY, R. P., HARRISON, T. A., PARKMAN, R., and BALTIMORE,D., Terminal deoxynucleotidyl transferase activity in human leukemic cells and in normal human thyrnocytes. N. Engl. . I Med., . 292, 775-780 (1975). J., OHNUMA, T., and MOORE,G. E., RosetteMINOWADA, forming human lymphoid cell lines. I. Establishment and evidence for origin of thymus-derived lymphocytes. J . nay. Cancer Inst., 49, 891-895 (1972). NEUMANN, H., KLEIN,E., HAUCK-GRANOTH, R., YACHNIN, s., and BEN-BASSAT, H., Comparative study of alkaline phosphatase activity in lymphocytes, mitogen-induced blasts, lymphoblastoid cell lines, acute myeloid leukemia and chronic lymphatic leukemia cells. Proc. nat. Acad. Sci. (Wash.), 73, 1432-1436 (1976). H., MORAN,E. M., RUSSELL,R. M., and NEUMANN, ROSENBERC, I., Distinct alkaline phosphatase in serum of patients with lymphatic leukemia and infectious mononucleosis. Science, 186, 151-153 (1974). R., and NEUMANN, H., WILSON,K. J., HAUCK-GRANOTH, HARAN-GHERA, N., A comparative biochemical study of alkaline phosphatases in normal and leukemic mice. Cancer Res., 31, 1695-1701 (1971). NILSSON,K., and PONTBN,J., Classification and biological nature of established human hematopoietic cell lines. Int. J. Cancer, 15, 321-341 (1975). NILSSON,K., and SUNDSTROM, C., Establishment and characteristics of two unique cell lines from patients with lymphosarcoma. Inr. J. Cancer, 13, 808-823 (1974). J., HAN,T., and MOORE,G. E., PAULY,J. L., MINOWADA, Disparity of mixed lymphocyte reactivity to cultured cells of human T and B lymphoid lines. J. nat. Cancer Inst., 54, 557-562 (1975). POLLIACK, A., LAMPEN, N., and DE HARVEN, E., Scanning electron microscopy of lymphocytes of known B and T derivation. In: Proceedings of the 7th Annual SEM Symposium (TTRI, Chicago, Illinois), p. 673-682 (1974).

43 1

POLLIACK,A., MCKENZIE,S., GEE, T., DE HARVEN,E., LAMPEN, N., and CLARKSON, B. D., A scanning electron microscopy study of 34 cases of acute granulocytic, myelomonocytic, monoblastic and histiocytic leukemia. Amer. J. Med., 59, 308-315 (1975). PULVERTAFT, R. J. V., A study of malignant tumors in Nigeria by short term tissue culture. J. din. Path., 18, 261-273 (1965). RALPH,P., Differential toxicity of Con-A and PHA on murine and human hematopoietic cell lines. In: H. Bittinger and H. P. Schnebli (ed.), Concanavalin A as a tool, John Wiley and Sons, London (1975). RALPH,P., and NAKOINZ,I., Phagocytosis and cytolysis by a macrophage tumor and its cloned cell line. Nature (Lond.), 257, 393-394 (1975). B. M., and KLEIN,G., Cellular localization of an REEDMAN, Epstein-Barr virus (EBV)-associated complement fixing antigen in producer and non-producer lymphoblastoid cell lines. Int. J . Cancer, 11, 499-520 (1973). SUNDSTROM, Ch., and NILSSON,K., Establishment and characterization of a human histiocytic lymphoma cell line, Int. J. Cancer, 17, 565-577 (1976). VANKY,F., KLEIN,E., STJERNSWARD, J., and NILSONNE, U., Cellular immunity against tumor-associated antigens in humans: lymphocyte stimulation and skin reaction. Int. J. Cancer, 14, 777-788 (1974). WEST,W., and HERBERMAN, R. B., A human lymphoid cell line with receptors for both sheep red blood cells and complement. Cell. Immunol., 14, 139-145 (1974). YATA,J., and KLEIN,G., Some factors affecting membrane immunofluorescence reactivity of Burkitt lymphoma tissue culture cell lines. Int. J . Cancer, 4, 767-775 (1969). YOSHIDA, T., and ANDERSON,B., Evidence for a receptor recognizing antigen complexed immunoglobulin on the surface of activated mouse thymus lymphocytes. Scand. 1. Immunol., 1, 401-408 (1972). ZEUTHEN, J., FRIEDRICH, U., ROSBN,A., and KLEIN,E., Structural abnormalities in chromosomes No. 15 in lymphoid cell lines with abnormal Bz microglobulin expression. Manuscript in preparation.