29, f8f- 195 (1983)
Human Monocyte-Histiocyte Differentiation Antigens Identified by Monoclonal Antibodies’ MALEK KAMOUN,*** PAULJ. MARTIN,*,? LAWRENCEG. LuM,? ANDJOHN
MARSHALL E. KADIN,* A. HANSEN*,+
*Departments of Medicine and Laboratory Medicine, University of Washington, Seattle, Washington 98195, fFred Hutchinson Cancer Research Center, and #Puget Sound Blood Center, Terry at Madison, Seattle, Washington 98104 lko distinct differentiation antigens of human myelomonocytic cells are defined using murine monoclonal antibodies. The antigens recognized by antibodies 20.2 and 20.3 are expressed by all cells of the monocyte lineage in both peripheral blood and bone marrow. Cell-sorting experiments demonstrated that histiocytes and immature bone marrow cells with detectable a-naphthyl butyrate esterase activity also express both antigens. Within cells of other lineages, the antigens had distinct patterns of expression. Immature myeloid cells were 20.2 negative, but 20.3 positive; whereas mature myeloid cells were 20.2 positive, but 20.3 negative. Nucleated erythroid cells and platelets expressed only the 20.3 antigen. These results indicate that myeloid and monocytic cells share common differentiation antigens with cells of the erythroid and megakaryocytic lineages. The 20.2 and 20.3 antibodies reacted with the leukemic cells from some patients with acute nonlymphocytic leukemia (FAB, MI-MS) and with some cell lines derived from patients with nonlymphocytic leukemia, but not with blast cells from patients with lymphoid leukemia or with lymphoid leukemic cell lines. These antibodies may prove useful in studying the differentiation of bone marrow stem cells, in defining the cellular origins and classification of leukemias, and in the identification of distinct prognostic subgroups of acute nonlymphocytic leukemia.
Observations from clonal culture of bone marrow have increased our understanding of the development of hematopoietic cells and have elucidated relationships between cells of different lineages (1, 2). For example, the monocytic and myeloid lineages are thought to be closely related since a single progenitor cell can give rise to both monocytes and granulocytes. Further data in support of the concept that these two hematopoietic lineages are closely related have emerged from recent studies in which monoclonal antibodies have been used for the identification of the surface antigens expressed by monocytes and granulocytes (316). With these reagents it has been possible to identify antigens uniquely expressed by cells of each lineage or even by cells at specific stages of differentiation within a lineage. With monoclonal antibodies it has also been possible to identify ’ This investigation was supported by Grants CA 18029, CA 18221, CA 29548, and CA 31651 awarded by the National Cancer Institute, and HL 17265 awarded by the National Heart, Lung, and Blood Institute, Department of Health and Human Services. * Current address: Department of Clinical Pathology, Hospital of the University of Pennsylvania, 3400 Spruce Street, G-l, Philadelphia, Penn. 19104. 181 0090-1229183 $1.50 Copyright @ 1983 by Academic Press, Inc. AU rights of reproduction in any form reserved
other differentiation antigens which are coexpressed by cells of more than one lineage. Although such data must be interpreted with caution, parallel expression of these antigens has been taken as evidence suggesting a developmental relationship between cells of different lineages. We have identified two monoclonal antibodies designated 20.2 and 20.3 that react with distinct differentiation antigens expressed on the surface of monocytes. These antibodies react with bone marrow histiocytes and also with certain cells of other hematopoietic lineages. The 20.2 antibody reacts with mature myeloid cells, whereas the 20.3 antibody reacts with immature myeloid cells, nucleated erythroid cells, and platelets. MATERIALS
of cells. Mononuclear cells from peripheral blood (PBMC) or bone marrow were isolated by centrifugation on Ficoll-Hypaque (LSM, Litton Bionetics, Kensington, Md.). For the isolation of purified T cells, PBM were incubated with AET (2-aminoethylisothiouronium bromide hydrobromide)-treated sheep erythrocytes (SRBC). SRBC rosette-positive (E + ) cells were separated from SRBC rosette-negative (E - ) cells by centrifugation on Ficoll-Hypaque and the recovered E-t pellet was treated with Tris-buffered 0.83% NH&l to lyse erythrocytes (17). The purity of the isolated E + population was assessed by a repeated rosetting and indirect immunofluorescence with an anti-T-cell antibody 9.6 (see below). The T-cell population so obtained was 295% 9.6 positive. T cells bearing Fc-IgG receptors (T, cells) were isolated by rosetting purified E+ cells with IgG-coated bovine erythrocytes, followed by Ficoll-Hypaque gradient purification as previously described (18). The isolated T cells were more than 90% Fc-IgG-receptor positive. Monocyte-enriched adherent cells were obtained by incubation of PBMC on plastic culture dishes in the presence of 20% fetal calf serum (FCS) in RPM1 1640 for 45 min at 37”. The plastic-adherent and -nonadherent cell fractions contained approximately 80 and 1% monocytes, respectively, as defined by morphology with Wright’s stain and positive staining for (Ynaphthyl butyrate esterase (nonspecific esterase, NSE). For isolation of B-cellenriched populations, monocyte-depleted PBMC were incubated on a nylon-wool column. Nonadherent T cells were eluted from the column with medium, and the adherent population enriched for B lymphocytes was then released by gentle physical manipulation of the column (19). Granulocytes were isolated from heparinized peripheral blood by centrifugation over a Ficoll-Hypaque density gradient. Contaminating erythrocytes were lysed by treatment with Tris-buffered 0.83% NHdCl. Nucleated bone marrow cells were isolated from bone marrow aspirates obtained from normal individuals. A buffy-coat preparation was made and red cells were lysed using Tris-buffered 0.83% NH&l solution. Cultured T cells were maintained in RPM1 medium supplemented with interleukin 2 (IL 2) (17). Cell lines. Cell lines were maintained cryopreserved until needed and then grown under standard tissue culture conditions (19). The individual cell lines studied are described in Table 4. Cells from patients with leukemias. Cells from the peripheral blood of patients with acute leukemia were separated over Ficoll-Hypaque and cryopreserved. Cells were obtained when the white blood cell count exceeded 20,000/mm3 and Zsolation
morethan80%of cellswereblasts.Thepanelwascomposed of blastcellsfrom patients with acute nonlymphocytic leukemia (ANLL) including all stagesof differentiation designated Ml through M5 according to the French-American-British (FAB) classification (20): acute myeloblastic leukemia without maturation (Ml), myeloblastic leukemia with maturation (M2), hypergranular promyelocytic leukemia (M3), acute myelomonocytic leukemia (AMML) (M4), and acute monocytic leukemia (AMoL) (M5). The classification of ANLL subtypes was based on standard morphological and histochemical criteria using Wright’s, NSE, peroxidase, and Sudan black stains (21). Also included in the panel were blast cells from patients with childhood null cell acute lymphoblastic leukemia (ALL), from patients with T-cell ALL (T-ALL), from patients with acute and chronic B-cell leukemia, and from patients with acute undifferentiation leukemia (AUL). The phenotypes of these leukemia cells were determined by Wright morphology, conventional cytochemical staining, and surface markers analysis (22). Immunization, somatic cell hybridization, and hybrid selection. Eight-weekold female BALB/c mice were immunized ip with 5 x lo6 PBMC. The cell donor was an allogeneic marrow graft recipient who had a large number of circulating Ia-positive mononuclear cells. Mice were boosted at- Day 10, and 3 days later immune spleen cells were harvested. Procedures for the fusion of immune spleen cells with the mouse myeloma cell line BALB/c MOPC2 NSF1 have been previously described (19). Initial screening of hybridoma microcultures was performed by using a complement-dependent microcytotoxicity (CDC) assay against PBMC, purified T cells, and a B-lymphoblastoid cell line (RZ). Hybrid cells from selected microcultures were serially cloned by limiting dilution using BALB/c thymocytes as feeder cells. Antibody-containing ascites fluids were obtained from pristane-primed BALB/c mice inoculated ip with 5 x lo6 hybridoma cells. Serologic analysis. Ascites fluid containing monoclonal antibody was used at a dilution of l/100, a concentration that was found to give maximum staining by indirect immunofluorescence (IF) analysis. A CDC microassay using trypan blue as an endpoint of lysis was performed as previously described (19). Indirect IF was performed using affinity-purified fluorescein-conjugated sheep F(ab’)z fragments specific for mouse Ig (19). Monoclonal antibodies 9.6 (anti-Tp50) and 10.2 (anti-Tp67) were used for detection of T cells (17, 23). Monoclonal antibody OKMl which reacts with determinants shared by monocytes and a subpopulation of lymphocytes (4) was used for comparative studies. Specificity control was provided by ascites fluid 9E8 which contains an IgG2, monoclonal antibody specific for the ~15 component of mouse leukemia virus (17). Cells were examined by phase-contrast fluorescence microscopy. In addition, quantitative fluorescence measurements were made using a fluorescence activated cell sorter (FACS II, Becton-Dickinson, Mountain View, Calif.). Fluorescence-activated cell sorting. Cells were prepared for sorting by indirect immunofluorescence staining. PBMC or marrow mononuclear cells were incubated with either antibody 20.2 or 20.3 for 30 min at 4°C and then washed three times with RPM1 1640 medium containing 5% FCS and 0.05% EDTA. The pelleted cells were resuspended in fluorescein-conjugated sheep F(ab’)z fragments specific for mouse Ig (Cappel Labs, West Chester, Pa.) and incubated for an additional
30 min at 4°C. Cells were washed three times and then liltered through 40-p,m nitex screens. A fluorescence-activated cell sorter was used to separate fluorescent and nonfluorescent cells. The fluorescence intensity of nonspecifically stained cells was analyzed using cells that had been labeled only with the fIuoresceinconjugated antiserum. A “gate” of excluded cells was selected for each experiment in order to minimize the overlap between the two populations. The flow rate during cell sorting was lo3 cells per second. Cytochemical staining. Cell preparations for cytochemical staining were made by cytocentrifuge. For esterase staining, cells were incubated with ol-naphthyl butyrate (NSE) (Sigma Chemical Co., St. Louis, MO.) (21). RESULTS Development and Initial Characterization of Two Monoclonal Distinguishing Lymphocytes and Monocytes
Immune spleen cells from two BALB/c mice immunized with PBMC were fused with NSI/l cells and dispensed into 288 microwells. Culture supemates from 42 of these wells reacted with the immunizing cells in a CDC assay. The supemates from 10 wells did not react with purified autologous T cells or with an autologous B-lymphoblastoid cell line. The hybrid cells from these wells were selected for cloning. Two stable antibody-producing hybridomas, designated 20.2 and 20.3, were obtained following four cycles of cloning by limiting dilution. The immunoglobulins produced by 20.2 and 20.3 were identified as IgGzb and IgM, respectively. Inoculation of cloned hybridoma cells into pristane-primed mice resulted in the production of ascites fluids containing high titer of each antibody. Expression
of the 20.2 and 20.3 Antigens
The expression of 20.2 and 20.3 antigens by peripheral blood cells is summarized in Table 1. Neither antibody 20.2 nor 20.3 reacted with plastic-nonadherent PBMC, nylon-wool-purified T cells, or with nylon-wool-purified B cells. Both antibodies 20.2 and 20.3, however, reacted with virtually all plastic-adherent cells. This finding suggested that antibodies 20.2 and 20.3 reacted with monocytes but not with lymphoid cells. Since previous studies have indicated that monoclonal antibody OKMl identifies an antigen associated with monocytes and a subset of lymphoid cells (4), we undertook a direct comparison of antibodies OKMl, 20.2, and 20.3. The data in Fig. 1 show that antibody OKMl, but not 20.2 or 20.3, reacted weakly with a fraction of plastic-nonadherent cells. The nonadherent cells in this experiment contained less than 1% monocytes as defined by Wright’s and NSE stains. This observation confirmed that antigens 20.2 and 20.3, but not OKM 1, are restricted to adherent PBMC. Data from additive staining experiments indicated that fluorescent staining with any two combinations of 20.2, 20.3, or OKMl was always of greater intensity than staining with any one of the antibodies alone (data not shown), indicating that antibodies 20.2, 20.3, and OKMI each reacted with a distinct antigen. In further studies of other peripheral blood cells, blood granulocytes (polymorphonuclear cells) were positive with antibody 20.2, but nonreactive with antibody 20.3 (Table 1). In contrast, blood platelets were nonreactive with antibody 20.2, but positive with antibody 20.3.
REACTIVITY OFMONOCLONAL ANTIBODIES 20.2 AND~~.~WITH NORMALHUMAN PERIPHERAL BLOOD CELLSANDBONEMARROWMONONUCLEARCELLF Antibody 20.2 Cells assayed Peripheral blood cells T cells B cells Monocytes (plastic-adherent cells) Plastic-nonadherent cells Granulocytes Platelets Erythrocytes Bone marrow mononuclear cells
5 2 10 2 10 2 5 3
+++ nd nd nd
Antibody 20.3 IF 99 99
nd nd nd
a Results are expressed as the means (or range) of multiple samples. n, number of determinations. IF, percentage positive cells assessed by indirect immunofluorescence and FACS analyses. Cytotox, cell lysis assessed by quantitative complement-dependent cytotoxicity: - , no detectable lysis; + , lo-33% lysis; + + , 34-66% lysis; + + + , 67-100% lysis; nd, not done. b Bone marrow mononuclear cells prepared by centrifugation over Ficoll-Hypaque.
Testing of Fc-Receptor-Positive 20.3 Antigens
for Expression of the 20.2 and
T-Cell subpopulations expressing receptors for the Fc portion of IgG (Ty) are thought by some investigators to arise from a myeloid-monocytic precursor (24). For this reason, Ty cells were isolated from peripheral blood and tested for their reactivity with antibodies 20.2 and 20.3. The results of FACS analysis are illustrated in Fig. 2. All Ty cells were positive for the 9.6 antigen (Tp50) and approximately 40% were positive for the 10.2 antigen (Tp67). In contrast, there was no detectable expression of 20.2 or 20.3 antigens by Ty cells.
( Logo) L
FIG. 1. FACS profiles of normal human peripheral blood mononuclear cells (PBMC)after removal of adherent cells by absorption to plastic. The nonadherent cell fraction contained less than 1% monocytes as detined by Wright and NSE stains. The curves represent frequency distributions of fluorescence intensity (log amplified). Cells were stained by indirect immunofluorescence (IF) with antibodies 20.2, 20.3, OKMI, 9.6 (anti-T cell), and 9E8 (negative control).
Fluorescence Intensity ( Loglo) FIG. 2. FACS profiles of IgG-Fc-receptor-positive T cells (T,) isolated from PBMC. The curves represent frequency distributions of fluorescence intensity (log ampliied). Cells were stained by indirect IF with antibodies 20.2, 20.3, 9.6 (anti-T cell), 10.2 (anti-T cell), and 9ES (negative control).
Characteristics of Separated 20.2 and 20.3-Positive and -Negative Peripheral Blood Mononuclear Cells
Peripheral blood mononuclear cells expressing the 20.2 and 20.3 antigens were obtained by positive selection with the FACS. The fluorescence intensity of 20.2and 20.3-stained cells before and after separation are demonstrated in Figs. 3 and 4. The efficiency of separation was high as assessed by repeat fluorescence analysis. Examination of the 20.2- and 20.3-positive cells isolated from peripheral blood indicated that both antibodies identified the same population of mononuclear cells. By Wright’s staining, most of the 20.2- and 20.3-positive cells had the characteristic appearance of monocytes. Both 20.2- and 20.3-positive cells expressed strong nonspecific esterase activity (Table 2). More than 93% of the cells
20.2 pre sort 1
/ \, : ’ 1 I i I I I
Fluorescence Intensity ! Loglo 1
FIG. 3. Separation of PBMC into 20.2+ and 20.2- fractions. The curves represent frequency distributions of fluorescence intensity before (20.2 presort) and after sorting into 20.2 + and 20.3 fractions. Cells were stained by indirect IF. The fluorescence intensity thresholds chosen for sorting are indicated by arrows.
Fluorescence Intensity ( Log,,) 4. Separation of PBMC into 20.3+ and 20.3- fractions. The curves represent frequency distributions of fluorescence intensity before (20.3 presort) and after sorting into 20.3 + and 20.3- fractions. Cells were stained by indirect IF. The fluorescence intensity thresholds chosen for sorting are indicated by arrows. FIG.
in each population expressed high-affmity Fc receptors as indicated by their ability to form rosettes with IgG-coated bovine erythrocytes. A relatively high number of cells (40-60%) in each population was phagocytically active as evidenced by erythrophagocytosis. The 20.2- and the 20.3-positive fractions also contained a small number (1.4-5.0%) of large cells approximately 20-25 pm in diameter. These cells had an abundant cytoplasm, a nuclear:cytoplasmic ratio of less than 1.O,a round regukr nucleus, and no apparent NSE activity. The exact nature of these cells has not been defined. Less than 0.4% of 20.2- and 20.3-positive cells found in peripheral blood had lymphoid characteristics and only 0.4-2.5% was identifiable as myeloid cells. In contrast to the 20.2- and 20.3-positive populations, the cells collected in the antigen-negative fractions demonstrated the morphological and cytochemical characteristics of lymphocytes and 1% or less of these cells appeared to be monocytes (Table 2). Characteristics of Separated 20.2- and 20.3-Positive and -Negative Bone Marrow Cells
Bone marrow mononuclear cells were prepared by Ficoll-Hypaque centrifugation and then stained by indirect IF (Table 1). Antibody 20.2 reacted with approximately lo-17% of the bone marrow cells, and antibody 20.3 reacted with approximately 8-13% of the bone marrow cells. For direct analysis of the 20.2 and 20.3 positive and negative populations, aliquots of bone marrow cells were prepared as buffy-coat cells or as Ficoll-Hypaque-purified mononuclear cells, stained by indirect immunofluorescence, separated by FACS into positive and negative fractions, and then analyzed by Wright’s and NSE staining. The efficiency of separation was high as assessedby repeat fluorescence analysis (Figs. 5 and 6).
TABLE 2 MORPHOL~CICAL AND CYTOCHEMICAL BLOOD MONONUCLEAR CELLS
IDENTIFICATION OF FICOLL-HYPAQUE-PURIFIED PERIPHERAL SEPARATED BY FACS FOR 20.2 AND 20.3 EXPRESSION*
Experiment 1. Expression of antigen 20.2d 2.
Expression of antigen 20.3e
Cell fraction Unfractionated Positive Negative Unfractionated 20.3 Positive 20.3 Negative 20.2 20.2
77.7 0 97.1
20.0 92.5 0.9 18.3 97.8 1.0
79.9 0.4 98.0
Unclassifiedb mononuclear cells
0.4 5.0 0
1.9 2.5 2.0
18.0 78.0 2.5 17.0
a Results are expressed as percentage positive. A total of 1000 cells were counted per slide. Experiments 1 and 2 were performed with peripheral blood drawn on two different days from a selected normal healthy donor known to have a relatively high percentage of monocytes. b Unclassified mononuclear cells were defined as large cells (approximately 25 pm in diameter) with a round nucleus and abundant cytoplasm. c A small number of polymorphonuclear cells and erythroid cells (~0.2%) were found among the Ficoll-Hypaque-purified cells. d Of the unfractionated cells, 33% were reactive with antibody 20.2. After cell sorting 10 and 35% of the total cells were recovered in the 20.2-positive and 20.2-negative fractions, respectively. e Of the unfractionated cells, 31% were reactive with antibody 20.3. After ceil sorting 9 and 39% of the total cells were recovered in the 20.3-positive and 20.3-negative fractions, respectively.
Expression of the 20.2 Antigen by Bone Marrow Mononuclear Cells
When the FACS separation was performed using the whole bone marrow buffycoat preparation, the predominant cell types in the 20.2-positive fraction were polymorphonuclear granulocytes (approx 70%) and monocytes (approx 20%). In order to characterize more precisely the nongranulocytic 20.2-positive cells, the cell-sorting experiment was repeated using mononuclear cells prepared by Ficoll-
5. Separation of bone marrow mononuclear cells into 20.2+ and 20.2- fractions. Curves represent frequency distributions of fluorescence intensity (log amplified). Cells were stained with antibody 20.2 before sorting (20.2 presort) and after sorting (20.2+ and 20.2-). The fluorescence intensity thresholds chosen for sorting are indicated by arrows. FIG.
Fluorescence Intensity (Loglo)FIG. 6. Separation of bone marrow buffy-coat cells into 20.3+ and 20.3- fractions. Curves represent frequency distributions of fluorescence intensity (log amplified). Cells were stained with antibody 20.3 before sorting (20.3 presort) and after sorting (20.3+ and 20.3-). The fluorescence intensity thresholds chosen for sorting are indicated by arrows.
Hypaque centrifugation. The 20.2-positive fraction consisted of monocytes, histocytes, and mature myeloid cells including bands and small numbers of polymorphonuclear neutrophils (PMN), eosinophils, and basophils (Table 3, Fig. 7). Less than 1.1% of the 20.2-positive fraction was recognizable erythroid cells or immature myeloid cells (promyelocytes and myelocytes). Approximately half of the 20.2-positive cells had NSE activity. Also present in the 20.2-positive fraction was a morphologically immature type of blast cell which was approximately 20 pm in diameter, had a round and regular nucleus, a chromatin pattern similar to that of a monocyte, occasional nucleoli, a high nuclear:cytoplasmic ratio, and a basophilic cytoplasm (Fig. 7). These cells accounted for approximately 1.1% of the 20.2-positive fraction. Most were NSE positive, thus distinguishing them from immature myeloid cells found in the 20.2negative fraction. Because of these characteristics, we have referred to these cells as promonocytes. The 2O.Znegative bone marrow fraction consisted of early myeloid cells (promyelocytes, myelocytes, and metamyelocytes), nucleated erythroid cells, and lymphoid cells (Table 3). The 20.2-negative fraction contained no detectable monocytes or histiocytes, and cells of these fractions had no detectable NSE activity. In summary, expression of 20.2 antigen in bone marrow appeared to be restricted to cells of the myeloid-monocytic lineage. Furthermore, as myeloid cells mature, the expression of 20.2 antigen on these cells appears to increase, since immature myeloid cells were 20.2 negative, and mature myeloid cells were 20.2 positive. Expression
of the 20.3 Antigen
by Bone Marrow Cells (Bufy Coat)
The 20.3-positive population in the marrow contained monocytes, nucleated erythroid cells, and early myeloid cells, but not mature myeloid cells or lymphocytes (Table 3 and Fig. 6). The 20.3-positive population also contained a small number of morphologically immature-appearing cells with detectable NSE activity which as described above have monocytic features and therefore are referred
Unfractionated 20.3 Positive
2.7 5.6 2.5
Unfractionated 20.2 Positive 20.2 Negative
0.1 3.2 0
13.7 0.4 12.8
34.9 1.1 31.1
1.1 0.0 1.3
13.7 0.0 13.4
TABLE 3 OF BONE MARROW CELLS SEPARATED BY FACS
29.1 1.9 38.6
0.9 14.6 0.3
3.8 73.2 0
FOR 20.2 AND 20.3 EXPRESSIONS
2.0 52.0 0.5
u Results are expressed as percentage positive. A total of 1000 cells were counted per slide. PMN, pofymorphonuclear cells, NSE, nonspecific esterase stain; nd, not done. b BIasts are defined as undifferentiated immature cells of myeloid and monocytic lineage. Approximately 90% of the blasts in the 20.2-positive bone marrow fraction were large cells with a diameter of approximately 20 pm. with a round regular nucleus, a high nuclear:cytoplasmic ratio, a basophilic cytoplasm, and positive NSE activity. We have referred to these cells as promonocytes (see Fig. 7). c Of the unfractionated cells, 8% were reactive with antibody 20.2. After cell sorting 3 and 38% of the total cells were recovered in the 20.2-positive and 20.2-negative fractions, respectively. d Of the unfractionated cells, 13% were reactive with antibody 20.3. After cell sorting 5 and 35% of the total cells were recovered in the 20.3-positive and 20.3-negative fractions, respectively.
expression of antigen 20.3d
1. Bone marrow mononuclear expression of antigen 20.2’ 2. Bone marrow buffy-coat
MORPHOLOGICAL AND CYTOCHEMICAL
F 5 5
FIG. 7. Wright’s stain of 20.2+ bone marrow mononuclear cells showing a histiocyte (arrowhead) and an immature mononuclear cell (large arrow) that we have referred to as a promonocyte. x 1000.
to as promonocytes. Virtually no mature myeloid cells (metamyelocytes and PMN) were observed. The 20.3-negative fraction contained relatively mature myeloid cells (myelocytes, metamyelocytes, and PMN) and lymphoid cells, but no detectable monocytes or histiocytes. Less than 1% of 20.3-negative marrow cells were NSE positive. Thus, 20.3 antigen differs from 20.2 antigen in that it its expressed by early immature myeloid cells and by nucleated erythroid cells. In contrast to expression of the 20.2 antigen, expression of 20.3 antigen appeared to decrease with myeloid maturation. Expression of the 20.2 and 20.3 Antigens by Cultured Lymphoid and Myeloid Cells Neither antibody 20.2 nor 20.3 reacted with a cultured B-lymphoblastoid cell line from a normal donor, with the Burkitt’s cell line Daudi, with the pre-B-cell line NALM-6, with three leukemic T-cell lines, or with continuously cultured normal T cells (Table 4). One histiocytic cell line, MH-1, was partially lysed by both antibodies 20.2 and 20.3 in the presence of complement, but no binding of either antibody was detected by indirect IF. The erythroblastoid CML cell line K562 and the promyelocytic cell line HL-60 were nonreactive with both 20.2 and 20.3. One cell line, HEL, derived from a patient with erythroleukemia, was strongly positive for antigen 20.3 but negative for 20.2. Expression
of the 20.2 and 20.3 Antigens
by Cells from Patients
Acute nonlymphocytic leukemia (ANLL). Leukemic cells from patients with ANLL, representing monocyte-myeloid lineages, expressed 20.2 and 20.3 anti-
TABLE 4 REACTIVITY
20.2 AND 20.3 WITH CULTURED CELL LINES”
Antibody 20.2 Cell-line assayed B-Lymphoblastoid cell lines (n = 3)b T-ALL cell lines (n = 3)c Cultured normal T cells (n = 1) Nonlymphoid cell lines MH-1, histocyti& K562, erythroblastic CML HL-60, promyelocytic HEL, erythroleukemiad
(1Results are expressed as the means (or range) of multiple samples. n, number of determinations. IF, percentage positive cells assessed by indirect immunofluorescence and FACS analysis. Cytotox, cell lysis assessed by quantitative complement-dependent cytotoxicity: - , no detectable lysis; + , lo-33% lysis; + + , 3466% lysis; + + + , 67-100% lysis; nd, not done. b B-Lymphoblastoid cell lines tested included: (i) a B-lymphoblastoid cell line obtained from a normal donor by EB virus transformation (donor RZ); (ii) the cell-line Daudi obtained from a patient with Burkitt’s lymphoma; and (iii) Nalmd, a leukemic pre-B-cell line. c Leukemic T-cell lines tested included: HSB-2, Molt-4F, and Jurkat (JM). d MH-1 is a cell line obtained from a patient with malignant histiocytosis (see Kadin et al. (27)). HEL is a cell line obtained from a patient with acute erythroleukemia (see Martin and Papayannoupoulou (28)).
gens to a variable extent. The results summarized in Table 5 show that the 20.3 antigen was detected on the blast cells from six of six patients with acute monocytic (AMoL) or acute myelomonocytic (AMML) leukemia. The 20.2 antigen, however, was detected on the blast cells of only three of these patients. The trend of these data suggests that most leukemic cells with features of monocytic diiferentiation would be 20.3 positive. Antigens 20.2 and 20.3 were also detected on the blast cells of some patients with acute granulocytic leukemia (AGL). Acute undifferentiated leukemia (AUL) and acute and chronic lymphocytic leukemia. Antigens 20.2 and 20.3 were not detected on blast cells from two patients with AUL, nine patients with null-cell ALL, four patients with T-cell ALL, or on the malignant cells from 10 patients with acute or chronic B-cell lymphocytic leukemia representing various stages of lymphoid differentiation (data not shown). DISCUSSION In this report we have described two murine monoclonal antibodies, 20.2 and 20.3, that recognize distinct differentiation antigens expressed by cells of the myeloid-monocytic series. Antibody 20.2 identified an antigen that is expressed by all monocytes, histiocytes, and polymorphonuclear granulocytes (including neutrophils, eosinophils, and basophils). Antibody 20.3 identified an antigen that is expressed by all monocytes, histiocytes, immature myeloid cells (myeloblasts, promyelocytes, and myelocytes, but not metamyelocytes, bands, or polys), nucleated erythroid cells, and platelets. The expression of the 20.2 and 20.3 antigens
TABLE 5 REACTIVITY OF ANTIBODY 20.2 AND 20.3 WITH TUMOR CELLS FROM PATIENTS WITH ACUTE MYELOID AND [email protected] LEUKEMIAS~ Cells assayed AMolC W-679 W-1365 w-1545 w-410 AMMLC W-1321 W-1358 AGL’ W-456 w-1535 W-616 W-567
MS MS Unclassifiedd MS
+ + + +
Ml MI M2 M3
+ + +
+ + -
B Reactivity was assessed by indirect immunofluorescence, cytoxicity, or both. + , Indicates that at least 50% cells were stained or lysed; - , indicates no significant reactivity detected. b FAB, French-American-British classification. c AMoL, acute monocytic leukemia; AMML, acute myelomonocytic leukemia. Leukemic cells were NSE+ d Case W-1545 had features of a monoblastic histocytic leukemia. All leukemic cells were NSE+ and peroxidase+ . Morphologically this leukemia could not be classified according to FAB criterion. p AGL, acute granulocytic leukemia. Leukemic cells were NSE-negative, peroxidase-positive.
by all blood and bone marrow monocytes was demonstrated by positive and negative selection experiments. Monocytes were identified by the following criteria: adherence to plastic, characteristic morphology by Wright’s stain, NSE activity, immune adherence via Fc receptors, and phagocytic activity. Although the 20.2 and 20.3 antigens were found to be coexpressed by all bone marrow monocytes and histiocytes, they had distinct patterns of expression on myeloid cells, nucleated erythroid cells, and platelets. A diagram illustrating the expression of 20.2 and 20.3 antigens by hematopoietic cells is shown in Fig. 8. Among bone marrow blast cells, the 20.2 antigen was found mostly on cells of the monocyte lineage. These cells were characterized as having a round and regular nucleus, a chromatin pattern similar to that of monocytes, a basophilic cytoplasm, and NSE activity. We have, therefore, provisionally referred to these cells as promonocytes. Thus the 20.2 antibody appears to be useful in distinguishing immature cells of the monocytic lineage from those of the myeloid lineage. In contrast, the 20.3 antibody reacts with the earliest recognizable cells of both monocytic and myeloid lineages. Although the ability of cells to form spontaneous rosettes with sheep erythrocytes (E +) has generally been regarded as a marker for thymus-dependent lymphocytes, recent controversy has arisen as to whether all E+ cells are T cells. Reinherz ef al. (24) and Kay and Horwitz (25) have suggested that E+ cells which express an Fc receptor for IgG (Ty) may be progeny of myelomonocyte precursors and not true T cells. These investigators found that antibody OKMl ,
KAMOUN ANTIGEN Mmyles/
FIG. 8. A diagrammatic representation illustrating the expression of 20.2 and 20.3 antigens on morphologically identifiable immature and mature bone marrow and peripheral blood cells. Immature cells include promonocytes, myeloblasts, promyelocytes, myelocytes metamyelocytes, and nudeated erythroid cells. Mature cells include monocytes, histiocytes, bands, polymorphonuclear neutrophik, eosinophils, and basophils, platelets, and erythrocytes.
which detects a monocyte-associated antigen, also reacts with 6 to 18% of T-y cells. Fox et al. (26), however, have demonstrated that Ty cells also express at least two distinct T-cell antigens in addition to the E-receptor-associated (Tp50) antigen defined by monoclonal antibody 9.6 (15). Evaluation of the expression of 20.2 and 20.3 antigens on established cell lines and cryopreserved leukemic cells confirmed the distribution of these antigens found on PBMC and bone marrow cells, i.e., neither antigen is expressed on lymphoid cell lines or on cells from patients with lymphocytic or undifferentiated leukemia. In contrast these antigens were found on blasts from some patients whose leukemic cells had morphologic and cytochemical features of either monocytic or myeloid differentiation. Several other antigens whose expression is restricted to human monocytes or shared among bone marrow cells have been recently described. Mac120 (5), Mo2 (8), Mo3 (8), and DSD6 (11) are expressed exclusively on cells of the monocyte lineage and are thus distinct from 20.2 and 20.3 with respect to cellular distribution. Mo4 (8) is expressed on monocytes and platelets among peripheral blood cells. Its distribution on bone marrow cells was not reported. GKMl (4) and Mol (3) are shared by monocytes, granulocytes, and a small fraction of lymphoid cells. Direct comparative analysis of cell distribution of GKMl, 20.2, and 20.3 has indicated that these antigens are unrelated. 36D3 is a 200,000-Da antigen expressed by peripheral blood monocytes and weakly expressed by myeloid cells (6). Likewise IGlO (9), MY3, MY4, MY7, and MY8 (12) are antigens shared to a variable degree by monocytes and granulocytes. The possible relationship of these determinants to 20.2 antigen awaits further comparative studies. Bernstein er al. (9) have described a murine monoclonal antibody, 5F1, defining a determinant expressed by peripheral blood monocytes and platelets. Preliminary evidence from comparative studies of antibodies 5Fl and 20.3 indicates that they block each other in competitive inhibition assays (personal communication, I. D.
Bernstein and R. G. Andrews). Further studies to identify the antigen detected
by thesetwo antibodies are still in progress. In this report we have defined two distinct myelomonocytic differentiation antigens 20.2 and 20.3 that are coexpressed by all cells of the monocyte lineage in peripheral blood and bone marrow including bone marrow histiocytes. The 20.2 antigen is also expressed by mature granulocytes whereas the 20.3 antigen is expressed by immature myeloid cells, nucleated erythroid cells, and platelets. Antibodies 20.2 and 20.3 should prove useful in studying the differentiation of bone marrow stem cells and in defining the cellular origins of acute nonlymphocytic leukemia. ACKNOWLEDGMENTS We are grateful to Agnes Chang, Gary Longton, Mira Zalokar, Yvonne Betson, Irene Pierovich, and Rosa Mae McDonald for excellent technical assistance, and to Pauline Marsden for help in preparation of this manuscript.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
23. 24. 25. 26. 27. 28.
Van Furth, R., and Cohn, Z. A., J. Exp. Med. 128, 415, 1%8. Moore, M. A. S., Williams, N., and Metcalf, D., J. Cell Physiol. 79, 283, 1972. Todd, R. F., Nadler, L. M., and Schlossman, S. E, J. Zmmunol. 126, 1435, 1981. Breard, J., Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman, S. E, J. Zmmunol. 124, 1943, 1980. Raff, H. V., Picker, L. J., and Stobo, S. D., J. Exp. Med. 152, 581, 1980. Ugolini, V., Nunez, G., Smith, R. G., Stastny, P., and Capra, J. D., Proc. Nar. Acad. Sci. USA 77, 6764, 1980. Hogg, N., Slusarenko, M., Cohen, J., and Reiser, J., Cell 24, 875, 1981. Todd, R. F., and Schlossman, S. F., Blood 59, 775, 1982. Bernstein, I. D., Andrews, R. G., Cohen, S. F., and McMaster, B. E., J. Zmmunol. 128, 876, 1982. Lebien, T. W., and Kersey J. H., J. Zmmunol. 125, 2208, 1980. Linker-Israeli, M., Billing, R. J., Fon, K. A., and Terasaki, F! I., J. Zmmunol. 127, 2473, 1981. Griffin, J. D., Ritz, J., Nadler, L. M., and Schlossman, S. F., J. Clin. Invest. 68, 932, 1981. Springer, T., Galfre, G., Secher, D. S., and Milstein, C., Eur. J. Zmmunol. 9, 301, 1979. Austyn, J. M., and Gordon, S., Eur. J. Zmmunol. 11, 805, 1981. Hirsch, S., Austyn, J. M., and Gordon, S., J. Exp. Med. 154, 713, 1981. Katz, H. R., LeBlanc, P A., and Russell, S. W., J. Reticuloendothel. Sot. 30, 439, 1981. Kamoun, M., Martin, P. J., Hansen, J. A., Brown, M. A., Siadak, A. W., and Nowinski, R. C., J. Exp. Med. 153, 207, 1981. Lum, L. G., Muchmore, A. V., O’Connor, N., Strober, W., and Blaese, R. M., J. Zmmunol. 123, 714, 1979. Hansen, J. A., Martin, P. J., and Nowinski, R. C., Immunogenetics 10, 247, 1980. Bennett, J. M., Catovsky, D., Daniel, M. T., Flandrin, G., Gauton, D. A. G., Graunick, H. R., and Sultan C, Brit. J. Haematol. 33, 451, 1976. Yam, L. T., Li, C. Y., and Crosby, W. H., Amer. J. Clin. Pathol. 55, 283, 1971. Greaves, M. F., Prog. Hematol. 9, 255, 1975. Martin, P. J., Hansen, J. A., Nowinski, R. C., and Brown, M. A., Immunogenetics 11,429, 1980. Reinherz, E. L., Moretta, L., Roper, M., Breard, J. M., Mingari, M. C., Cooper, M. D., and Schlossman, S. E, J. Exp. Med. 151, 969, 1980. Kay, H. D., and Horwitz, D. A., J. Clin. Invest. 66, 847, 1980. Fox, R. L., Thompson, L. F., and Huddlestone, J. R., J. Zmmunol. 126, 2062, 1981. Kadin, M. E., Holt, L., and Najfeld, V., Blood 54(Suppl. I), 173a, 1979. Martin, P. J., and Papayannopoulou, T., Science 216, 1233, 1982.
Received November 10, 1982; accepted with revisions February 23, 1983