salts solution; MoAb, monoclonal antibodies; R/M-SRBC, rabbit anti-mouse ... Procurance of all specimens was approved by the Brigham and Women's. Hospital ...
PURIFICATION HEMATOPOIETIC ACUTE
AND
CHARACTERIZATION
CELLS
LYMPHOBLASTIC
THAT
EXPRESS
LEUKEMIA
OF THE
ANTIGEN
FETAL
COMMON (CALLA)*
BY PETER HOKLAND,:~ PAUL ROSENTHAL, JAMES D. GRIFFIN,§
LEE M. NADLER, HJOHN DALEY, MARIANNE HOKLAND,¶ STUART F. SCHLOSSMAN, AND JEROME RITZ** From the Division of Tumor Immunology, Sidney Farber Cancer Institute, and Department of Medicine, Harvard Medical School, Boston, Massachusetts 02l 15
Previous studies using c o n v e n t i o n a l heteroantisera a n d m o n o c l o n a l antibodies (1, 2) have characterized the expression of a c o m m o n acute lymphoblastic leukemia a n t i g e n (CALLA) 1 in various hematopoietic malignancies a n d n o r m a l tissues, C A L L A has been demonstrated to be a 100,000-dahon glycoprotein that is expressed by leukemic cells from ~80% of patients with n o n - T acute lymphoblastic leukemia (ALL), 10% of patients with T - A L L , a n d 40% of patients with chronic myelocytic leukemia (CML) in blast crisis, as well as l y m p h o m a cells from ~40% of patients with T cell lymphoblastic l y m p h o m a s a n d almost all Burkitt's l y m p h o m a s a n d n o d u l a r l y m p h o m a s (3). W i t h i n the hematopoietic system, C A L L A is also detected on a small p o p u l a t i o n of m o n o n u c l e a r cells (1-3%) in either n o r m a l or regenerating bone marrow. Fetal hematopoietic organs c o n t a i n a larger proportion of C A L L A + cells a n d m o n o n u c l e a r cell suspensions from fetal liver a n d fetal bone marrow have been found to c o n t a i n 2-10% C A L L A + cells (2-5). Although m u c h is known about the distribution of CALLA, very little is k n o w n a b o u t its functional role on the cell surface or its a p p e a r a n c e d u r i n g ontogeny. Recent studies have d e m o n s t r a t e d that C A L L A belongs to a family of 100,000-dahon cell surface proteins that are widely expressed by hematopoietic cells (6) a n d is also present in some n o n h e m a t o p o i e t i c tissues (7). In addition, b i n d i n g of j 5 m o n o c l o n a l a n t i b o d y to cell surface a n t i g e n results in the m o d u l a t i o n of the expression of C A L L A on * Supported in part by grant CA-28704 from the National Institutes of I ~cal~h. On leave from the University Department of Medicine and ttaematology, University of Aarhus. Denmark. Research Fellow of the Danish Medical Research Council. § Research Fellow of the Medical Foundation, Boston, MA and tile Joanna C. \~,~ood Foundation, Philadelphia, PA. II Research Fellowof the Medical Foundation, Boston, MA. ¶ Research Fellowof the Danish Cancer Society. ** Special Fellowof the Leukemia Society of America. 1Abbreviations u~edin thi~"paper: ALL, acute lymphoblastic leukemia: BM, bone marrow; C, complement; CALl,A, common acute lymphoblastic leukemia antigen; CFU~G/M, granulocytc/monocyte cotol~y forming unit; CML, chronic myelocyticleukemia; tyro-p,, intracytoplasmic g chain; FACS, fluorescence activated cell sorter: FCS, fetal calf serum: C,/Md"ITC, fluorescein-conjugatedgoat anti-mouse immunoglobulin; G/M-TRITC, rhodamine-col~iugatedgoat anti-mouse immunoglobulin; ItBSS, t tanks' balanced salts solution; MoAb, monoclonalantibodies; R/M-SRBC, rabbit anti-mouse Ig-coated sheep erythrocytes; SMEM + 2.5% AB, supplemented minimum cxsential medium with 2.5+;4,human AB: SRBC, sheep erytbrocytes; TdT, terminal transferase enzyme. 114
J. Exp. MEO.© The Rockefeller University Press • 0022-1007/83/01/0114/16 $1.00 Volmne 157 January 1983 114-129
HOKLAND ET AL.
1 15
leukemic cells a n d cell lines. J 5 m o d u l a t i o n occurs in vitro (8) a n d in vivo (9) a n d results in a reversible a n d specific loss a n d i n t e r n a l i z a t i o n o f cell surface C A L L A (10). Despite the a n a l o g o u s m o d u l a t o r y b e h a v i o r o f C A L L A a n d k n o w n m e m b r a n e receptors for various hormones, no functional role for C A L L A has yet been defined. Given the c o m p e l l i n g d a t a d e m o n s t r a t i n g that most t u m o r s o f m y e l o i d a n d l y m p h o i d systems reflect p h e n o t y p e s o f frozen stages o f differentiation c o r r e s p o n d i n g to their n o r m a l cellular c o u n t e r p a r t s , in this s t u d y we have purified n o r m a l fetal h e m a t o p o i e t i c cells that express C A L L A to further d e l i n e a t e the role o f this cell surface glycoprotein in early l y m p h o i d differentiation. Pure p o p u l a t i o n s of C A L L A + cells were o b t a i n e d b y either i m m u n e rosetting or fluorescence-activated cell sorting, a n d these cells were s u b s e q u e n t l y c h a r a c t e r i z e d using d u a l fluorescence labeling techniques for the presence o f specific c y t o p l a s m i c a n d surface markers. These d a t a suggest t h a t C A L L A + cells are c o m m i t t e d l y m p h o i d cells t h a t are in the early stages o f l y m p h o i d differentiation a n d that, furthermore, n o r m a l C A L L A + cells a p p e a r to be the n o r m a l c o u n t e r p a r t s o f C A L L A + l y m p h o b l a s t i c l e u k e m i a cells. Materials and Methods
Preparation of Mononuclear Single Cell Suspensionsfrom Fetal Tissues. Fetal tissue specimens were obtained immediately after prostaglandin- or saline- induced abortions from fetuses with no apparent abnormalities. The age of the fetus was determined by measurements of crown-rump and foot lengths. Procurance of all specimens was approved by the Brigham and Women's Hospital Committee on the Use of Human Subjects in Research, Boston, MA. Fetal bone marrow was obtained from femoral bones that were removed aseptically. Single cell suspensions were produced by flushing the intramedullary cavities with Hanks' balanced saline solution (HBSS). Fetal livers were minced into 1-cm 3 pieces and passed repeatedly in and out of a 20-cm ~ syringe until large clumps of cells were no longer visible. Subsequently, cells were passed through a cone-shaped nylon mesh to obtain single cell suspensions. Single cell suspensions from both organs were then layered onto Ficoll-Hypaque (F/H) density gradients and after centrifugation the resulting interface cells were washed twice and resuspended in supplemented minimum essential medium with 2.5% pooled human AB serum (SMEM + 2.5% AB). For preparation of fetal liver mononuclear cell suspensions, this step was repeated to maximize elimination of erythrocytes. MonoclonalAntibodies. The production and reactivities of the monoclonal antibodies (MoAb) used in this study have been described previously. A summary of the specificity of these reagents is presented in Table I. Preparation of Rabbit Anti-Mouse tg-coated Sheep E~throcytes (R/M-SRBC). As described previously (23), New Zealand white rabbits were immunized repeatedly with a mixture of MoAb including J2, J5, J13, MY7, and MYS. The immune serum was first passed through a column of human Ig coupled to Sepharose 4B (Pharmacia) and then through a column of mouse IgSepharose. Purified rabbit anti-mouse Ig was eluted from the second column with 1 M glycine buffer, pH 2.0, extensively dialyzed against 0.9% saline solution, adjusted to a concentration of 1 mg protein/ml, and stored at -20°C. Sheep erythrocytes (SRBC), washed three times in saline, were coupled with the rabbit antimouse antiserum with CrCla as follows: 0.5 ml packed SRBC were mixed with 0.5 ml rabbit anti-mouse antiserum and 0.5 mI saline, whereupon 0.5 mt of freshly prepared CrCla solution (1 mg/ml) was added dropwise under vigorous stirring. After a 6-7-min incubation at room temperature, the reaction was stopped with cold phosphate-buffered saline and the coupled SRBC were washed three times in saline and resuspended to a 10% solution in SMEM + 2.5% AB. Immune Rosette Depletion of Myeloid and E~ythroid Cells. To deplete the mononuclear cell suspensions of erythroid and myeloid precursor cells, mononuclear cells from fetal bone marrow
116 CHARACTERIZATION OF FETAL HEMATOPOIETIC CELLS EXPRESSING CALLA TABLE I
Summary of Monoclonal Antibodies Used in the Purification and Characterization of Fetal Hematopoietic Cells Monoclonal antibody J5/J13
Molecular weight* 100,000
j2
26,000
I-2 (Ia antigen)
29,000 34,000 30,000
BI
Mol Mo2 MY7 MY8
94,000 155,000 55,000 160,000 --
T3 T9 (transferrin receptor)
20,000 94,000
TIO
45,000
T1 l (E rosette receptor)
55,000
ttematopoietic cellular reactivily Early lymphoid cells, common ALL cells, blast-crisis lymphoid CML, Burkitt's lymphoma, nodular lymphoma Activated T cells, some thymocytes, platelets, early lymphoid cells, ALL cells B cells, monocytes, activated T cells, early hematopoietic cells All B cells except for plasma cells, B cell tumors except for myeloma, some non-T ALL cells Monocytes, null cells, granulocytes, acute monocytic leukemia Monocytes, acute monocytic leukemia Differentiated granulocytic and monocytic cells, acute myelocytic leukemia cells Mature T cells Activated T cells, rapidly proliferating cells, early erythroid cells Activated T cells, rapidly proliferating cells, early lymphoid cells, plasma cells T cells, thymocytes
Reference 2, 32
11 12 13, 14
15, 16 I5, 16 17
18, 19 19, 20, 21 19, 21 22
* Apparent molecular weight of antigen under reducing conditions in sodium dodecyl sulfide-polyacrylamide gel electrophoresis. a n d fetal liver were i n c u b a t e d at 4°C for 30 m i n with a mixture of M o A b including: T 9 (dilution 1:2,000), M o l (1:100), Mo2 (I:100), M Y 7 (1:100), a n d M Y 8 (1:100). After three washes in S M E M + 2.5% AB, the cells were resuspended to 1.0 ml a n d mixed with 1.0 ml of the 10% R / M - S R B C solution. W h e n leukocyte cell n u m b e r s exceeded 200 X 10~, 1 3 ml of R / M - S R B C was used_ Using this method, u p to 600 X 10~ cells can be processed in one tube. After a centrifugation at 180 g for 5 min, the rosette mixture was resuspended with a Pasteur pipette until clumps were no longer visible. T h e quality of the rosettes was d e t e r m i n e d u n d e r fluorescence microscopy in a n acridine orange-stained aliquot; subsequently, the rosettes were layered on F / H a n d spun for 15 rain at 1,140 g. T h e resulting interface suspension (subsequently referred to as MoAb-rosette-negative cells) was harvested a n d washed, whereas the pellet (MoAb-rosette-positive cells) was recovered after lysis of S R B C with a m m o n i u m chloride buffer. T h e viability in each fraction was >90% a n d total recovery of cells was >60%. Purification of CALLA + Cells by Rosette Sedimentation. C A L L A + cells were purified by a second rosette sedimentation after labeling of the MoAb-rosette-negative cell fractions with a n t i - C A L L A antibodies. Pilot experiments with C A L L A + A L L lines (balm-1 a n d Laz 221) had shown that the rosetting efficiency could be increased by a d d i n g an I g M antibody, J13, which is also specific for CALLA, to J 5 d u r i n g the antibody-labeling procedure. Consequently, an equal mixture of J 5 a n d J 13 (final dilution of each, 1: 100) was used when the MoAb-roset renegative cell fraction was to be rosetted. Otherwise, the rosetting was performed as described above. Purification of CALLA+ Cells by Fluorescence-activated Cell Sorting. In some experiments, C A L L A + cells were purified from MoAb-rosette-negative fractions by fluorescence-activated Greaves, M. F., G. Hariri, R. A. Newman, D. R. Sutherland, M. A. Ritter, and J. Ritz. Selective expression of the common acute [ymphoblastic leukemia (gpl00) antigen on immature lymphoid cells and their malignant counterparts. Blood. In press..
HOKLAND ET AL.
117
cell sorting after labeling with J5 antibody and a 1:50 dilution of fluorescein-conjugated goat anti-mouse antiserum (G/M-FITC; Tago Inc., Burlingame, CA) for 30 min at 4 C. After two washes the cells were separated into J5+ and J 5 - (CALLA+ and CALLA-) subpopulations using an EPICS V (Coulter Electronics Inc., Hialeah, FL). Post-sort viability in both fractions was >95%, and the purity of the subsets was >97% by reanalysis on the EPICS V. Phenotypic Analysis of CALLA + and CALLA- Cell Fractions. Dual fluorescence techniques were used to obtain direct evidence for the coexpression of other antigens on the CALLA+ cells. With sorted CALLA+ cells that were labeled with J5 and G/M-FITC, capping of CALLA-J5-G/M-FITC complexes was observed after an overnight incubation at 37°C. CALLA+ cells purified by J5 rosette sedimentation were labeled with G/M-FITC before overnight incubation to induce capping of J5-CALLA complexes. After overnight incubation, the CALLA+ cells were incubated with appropriate dilutions of a second monoclonal antibody followed by addition of rhodamine-conjugated goat anti-mouse antisera (G/M-TRITC, 1:40 dilution) (Sigma Chemical Co., St. Louis, MO). For detection of the B 1 antigen, biotin-labeled anti-B 1 (1:20) and rhodamine-conjugated avidin (1:80; Sigma Chemical Co., St. Louis, MO) was used. A minimmn of 200 cells were counted in a Zeiss fluorescence microscope (Carl Zeiss, Inc., New York) equipped with filter combinations that allowed for the clear distinction between fluorescein caps (J5-CALLA) and rhodamine membrane staining (second antibodies). The CALLA-negative cells, which contained no interfering antibodies, were analyzed for reactivity with specific monoclonal antibodies by indirect immunofluorescence. Cells were incubated with monoclonal antibody at 4°C for 30 min, washed twice, and then incubated with G/M-FITC for 30 min at 4°C. 10,000 cells were then analyzed for each reagent on a fluorescence-activated cell sorter (FACS-I; B-D FACS Systems, Mountain View, CA). Background fluorescence was determined by incubating identical cells with ascites fluid from a nonreactive hybridoma (J0). Detection of Intracellular Antigens. The terminal transferase enzyme (TdT) was detected by a two-layer immunofluorescence assay on cytocentrifuged, methanol-fixed cells as described by Bollum (24) using a kit from Bethesda Research Laboratories (Rockville, MD). Intracytoplasmic /z (cyto-/t) was detected on cytocentrifuged, methanol-fixed smears using directly fluorescein- or rhodamine-conjugated rabbit anti-p chain antisera (kindly provided by Dr. M. D. Cooper, University of Alabama, Birmingham, AL). Nalm- 1 cells were used as positive controls for the intracellular markers in the T d T assay and Raji cells in the cyto-# assay. Modulation In Vitro of the CALLA+ Cells. Antigenic modulation in vitro was induced in MoAb-rosette-negative cell fractions by addition of monoclonal antibody and subsequent incubation for different periods of time at 37°C. Either J5 antibody, J0, or anti-Ia was added to identical cultures. To ensure antibody excess, 0.1 mg antibody per l0 s cells was used. Cell culture media consisted of RPMI 1640 supplemented with 10% fetal calf serum (FCS). After incubation, cells were washed twice and phenotyped as described above for the C A L L A - cells. Myeloid Precursor Cell (CFU-G/M) Assays. CFU-G/M were assayed as described by Griffin et al. (23) by plating l0 s mononuclear cells/ml in Iscove's modified Dulbecco's minimal essential medium containing 20% FCS and 0.3% agar (Agar Noble; Difco, Detroit, MI), over a feeder layer of 1 × 106 peripheral blood leukocytes/ml in 0.5% agar. After 10 d of culture at 37°C in an atmosphere of 5% CO2 in humidified air, colonies (>40 cells) were enumerated using an inverted microscope. Results
Purification of CALLA + Cells from Fetal Tissues. As shown in Fig. 1 A, m o n o n u c l e a r cells o b t a i n e d from fetal b o n e m a r r o w after F / H density g r a d i e n t s e d i m e n t a t i o n primarily consisted of m a t u r i n g myeloid cells. In contrast, m o n o n u c l e a r cells o b t a i n e d from fetal liver (Fig. 1 E) after two F / H - p u r i f i c a t i o n steps consisted p r i m a r i l y of m a t u r i n g erythroid precursors. However, in both cases these cells could be removed from the m o n o n u c l e a r cell suspensions by rosette s e d i m e n t a t i o n with a m i x t u r e of M o A b (Fig. 1 B a n d F). In contrast to the MoAb-rosette-positive fractions, the M o A b -
118
C H A R A C T E R I Z A T I O N OF FETAL H E M A T O P O I E T I C CELLS EXPRESSING CALLA
HOKLAND ET AL.
119
.~X
~ >
.
o
e-
~
