Identification of a human T lymphocyte surface protein associated with the E-rosette receptor

Abstract

We describe a new monoclonal murine antibody that reacts with a 50,000-mol wt polypeptide that appears to be present on all E-rosetting cells. We conclude that this antigen is either identical to or closely associated with the E receptor because of (a) the high degree of concordance between E-rosette formation and 9.6 antigen expression, (b) the inhibition of rosette formation by preincubation of cells with 9.6 antibody, and (c) the observed failure of cells lysostripped of 9.6 antigen to form E-rosettes. This last finding suggests cocapping of 9.6 antigen and the E receptor.

Full text

Brief Definitive Report
IDENTIFICATION OF A HUMAN T LYMPHOCYTE SURFACE
PROTEIN ASSOCIATED WITH THE E-ROSETTE RECEPTOR
BY MALEK KAMOUN, PAUL J. MARTIN,* JOHN A. HANSEN,
MELISSA A. BROWN, ANTHONY W. SIADAK, ANO ROBERT C. NOWINSKI
From the Fred Hutchinson Cancer Research Center; and the Puget Sound Blood Center, Seattle, Washington
98104; and the Department of Medicine, Division of Oncology, University of Washington School of
Medicine, Seattle, Washington 98195
Human thymus-dependent lymphocytes (T cells) are defined by their ability to
form spontaneous rosettes (E-rosettes) with sheep erythrocytes (SRBC) (1-3). Re-
cently, several murine monoclonal antibodies that recognize distinct differentiation
antigens on human T cells have been described (4-11). We report here a monoclonal
antibody, designated 9.6, that identifies a 50,000-dalton surface protein that appears
to be present on all E-rosette-forming cells. Blocking and lysostripping experiments
indicate that this monoclonal antibody reacts either with the E receptor itself or with
a closely associated structure.
Materials and Methods
Cells.
Mononuclear cells from peripheral blood (PBL) or bone marrow were obtained from
normal volunteers and isolated by centrifugation over Ficoll-Hypaque (LSM; Litton Bionetics
Inc., Kensington, Md.). T cells were enumerated by rosetting with 2-aminoethylisothiouronium
bromide hydrobromide (AET)-treated SRBC (SRBC^v..r) in medium (RPMI-1640) with 12%
fetal calf serum (12). For the isolation of purified T cells, PBL were passed over nylon wool (11)
and then incubated with SRBCA~--v. Rosette-forming T cells (E +) were separated from non-T
cells (E-) by centrifugation over Ficoll-Hypaque. T cells were recovered from the rosettes by
lysis of SRBC in Tris-buffered 0.83% ammonium chloride. Normal human thymocytes were
prepared from thymus specimens obtained in the course of corrective open heart surgery from
children < 14 yr old. Long-term-cultured T cells, stimulated by alloantigen, were maintained in
medium supplemented with T cell growth factor (TCGF) (13).
A selected panel of cultured lymphoid cell lines was used for serological analysis. Included
were leukemic T cell lines (CEM, HSB2, 8402, Jurkat, Molt-4F, and KE37), the leukemic pre-
B cell line NALM-6, Epstein-Barr virus-transformed B-lymphoid cell lines (PA-3, SB, 8392,
HA, Swei), and the Burkitt's lymphoma ceil line Daudi.
Leukemic blasts from peripheral blood of patients with acute lymphocytic leukemia (ALL),
were separated over Ficoll-Hypaque, and cryopreserved. Cells were obtained either at presen-
tation or relapse when the leukocyte count was >20,000/mm s and >90% of the cells were blasts.
Subdivision of ALL into T and null cell types was based on the clinical criterion of a thymic
mass, and on determination of whether the leukemic blasts formed E-rosettes or expressed ia-
like antigen.
Immunizations, Fusion, and Screening.
The clone of hybrid cells producing the antibody
described here was isolated from a previously described fusion experiment (9). Briefly, spleen
cells from BALB/c mice immunized with human peripheral blood lymphocytes were fused
with BALB/c MOPC21 NSI/1 myeloma cells. Antibody production by hybrid cells was assayed
by testing tissue culture supernates in a complement-dependent microcytotoxicity assay against
normal peripheral blood T cells, a B-lymphoid cell line, and continuously cultured T cells, all
obtained from the same donor. One culture fluid was cytotoxic for both the normal and
cultured T cells but nonreactive with the autologous B-lymphoid cell line. Cells from this
* Junior Faculty Clinical Fellow of the American Cancer Society.
J. Exp. MED. © The Rockefeller University Press • 0022-1007/81/01/0207/06 $1.00 207
Volume 153 January 1981 207-212
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981

208 KAMOUN ET AL. BRIEF DEFINITIVE REPORT
culture, designated 9.6, were serially cloned four times by limiting-dilution and then inoculated
intraperitoneally into Pristane-primed (Aldrich Chemical Co., Inc., Milwaukee, Wis.)
BALB/c mice for the production of antibody-containing ascites fluid. The immunoglobulin
produced by the 9.6 hybrid was identified as IgG2b by immunodiffusion.
Serological Analysis. Methods for complement-dependent microcytotoxicity, quantitative
cytotoxicity (using trypan blue as indicator), and indirect immunofluorescence assays have
been previously described (9, 11). A fluorescein-conjugated affinity-purified goat anti-mouse
IgG serum (Litton Bionetics Inc.) was used for indirect immunofluorescence. In some experi~
ments rhodamine-conjugated goat F(ab')2 fragments specific for mouse IgG were used (this
reagent was generously provided by Dr. Shu Man Fu, The Rockefeller University, New York).
Comparative studies were carried out using three additional monoclonal antibodies recognizing
distinct markers for human lymphocytes. Antibody 9.3 reacts with a 45,000-dahon surface
protein expressed by 70-80% of peripheral E + cells (9); antibody 10.2 reacts with a 65,000- to
67,000-dahon surface protein expressed by 85-95% of peripheral blood E + cells (11); and
antibody 7.2 reacts with a framework determinant of the human Ia bimolecular complex (9).
Blocking of E Rosette Formation and Lysostrippmg. Cells were incubated in medium that
contained monoclonal antibody for 30 min at 22°C and then washed three times. In blocking
experiments, these cells were subsequently assayed for E rosette formation. In lysostripping
experiments, treated cells were incubated with affinity-purified goat anti-mouse IgG serum
(fluorescein conjugated) for 45 rain at 37~'C, washed three times, and then tested for rosette
formation.
Immune Prectpitation of Cell Membranes. Cells of the leukemic T cell line Jurkat were washed
twice in phosphate-buffered safine, pH 7.2, and surface labeled by the 125I-lactoperoxidase
method (14) with minor modification (15). Labeled ceils were disrupted in cell lysis buffer that
contained 0.5% Nonidet P-40 (15) and 2% of the protease inhibitor aprotinin (Sigma Chemical
Co., St. Louis, Mo.). The lysate was cleared of nonsolubilized cellular structures and free 125I by
uhracentrifugation and chromatography on a Sephadex G-25 column as described elsewhere
(16).
Before use in immune precipitation analysis, the cell lysate was treated by sequential
incubation (1 h each at 4°C) with AKR normal mouse serum (1:40 dilution) and Staphylococcus
aureus, Cowan I strain (30 mg) (17). The S. aureus was removed from the reaction mixture by
centrifugation at 1,500 g for 10 min. The concentration of sodium dodecyl sulfate (SDS) in the
lysate was adjusted to 0.1% and the immune precip!tation reactions initiated by the addition
6
of 5 #1 of appropriately diluted antibody to 95-#1 (I0 cpm) aliquots of radiolabeled lysate. The
mixture was incubated for 1 h on ice and the reaction then terminated by fhe addition of 3.5
mg S. aureus for 30 min on ice. The S. aureus pellet was washed five times in buffer that contained
0.5% NP-40 and 0.1% SDS (15). Radiolabeled proteins bound to the pellet were extracted by
incubation with 50 #1 of sample electrophoresis buffer (0.062 M Tris-HCl, 2% SDS, 10%
glycerol, 5% 2-mercaptoethanol, and 0.02% bromphenol blue, pH 6.8) for 5 min at 100°C and
then analyzed by polyacrylamide gel electrophoresis (PAGE) in the presence of SDS (SDS-
PAGE) in 10% slab gels (18). Radiolabeled bands in the dried gel were identified by
radioautography (Kodak NS-2T film [Eastman Kodak Co., Rochester, N. Y.] with GAFMED
Rarex B Mid Speed intensifying screen [GAF Corp., New York] with exposure at -70°C) (19).
Results and Discussion
The 9.6 antigen was present on peripheral blood T cells and thymocytes, but was
not detected on mononuclear cells of peripheral blood or bone marrow after removal
ofT cells by E-rosetting (Table I). T cells purified by passage over nylon wool followed
by centrifugation of E + cells over Ficoll-Hypaque were >99% positive with 9.6
antibody. Thymocytes were also >99% positive for the 9.6 antigen. Cells from B
lymphoid lines were uniformly negative.
A high degree of concordance between E-rosette formation and expression of 9.6
antigen was observed in testing cells from a variety of sources (Table I). Similar
numbers of 9.6-positive and E + cells were found in PBL, thymus, bone marrow, and
long-term-cultured T cells. Expression of 9.6 antigen also correlated with E-rosetting
in cells of leukemic T cell lines. Only the E + leukemic T cell lines 8402, Jurkat, and
Molt-4F were 9.6 positive, each demonstrating equal numbers of E + and 9.6-positive
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981

KAMOUN ET AL. BRIEF DEFINITIVE REPORT 209
TABLE I
Comparison of the Number of E-Rosetting and Monoclonal Antibody 9.6-Positive Cells
in Normal and Cultured Lymphoid Cells*
Percent E ÷ Percent 9.6 positive
Peripheral blood lymphocytes 72
Purified T cells 99
Non-T cells <1
Thymocytes >99
Bone marrow cells 9 17
T cell depleted < 1
B-lymphoid cell lines$§ <1
Cultured T cellsll >99
Leukemic T cell lines§
GEM < 1
8402 23
Jurkat 90
HSB-2 < 1
KE37 <1
Moh-4F 17
Mofi-4F, E ÷ enriched 85
Moh-4F, E ÷ depleted 3
(9) 68 (9)
(5) 99 (5)
(3) <1 (3)
(3) >99 (3)
(3) 10 15 (4)
(2) <1 (2)
(7) <l (7)
(3) >99 (:3)
<1
20
98
<1
<1
20
93
4
* Assays for E-rosette forming cells E ÷ were performed with SRBC^m-. Testing with 9.6 antibody
was performed by indirect immunofluorescence. The number of donors tested is indicated in
parentheses. Where appropriate, results are expressed as the mean (or range) of multiple samples.
:~ NALM-6, Daudi, PA3, SB, 8392, HA and Swei.
§
Cell lines CEM and PA3 were obtained from Dr. D. Mann, National Institutes of Health,
Betbesda, Md.; HSB-2, SB, and Jurkat from Dr. W. D. Peterson, Wayne State University,
Detroit, Mich.; 8402, 8392, Molt-4F, and NALM-6 from Dr. J. Minowada, Roswell Park
Memorial Institute, Buffalo, N. Y.; KE37 from Dr. S. M. Fu; and Daudi from Dr. W. Newman,
Fred Hutchinson Cancer Research Center, Seattle, Wash. Derivation of B-lymphoid cell lines
HA and Swei has been previously described (20).
II Normal T cells stimulated with alloantigen and cultured >3 mo in TCG Factor.
TABLE II
Effect of Antibody 9.6 on E-Rosette Formation
Cells treated
E-Rosette formation after incubation with
antibody*
Control Antibody 10.2 Antibody 9.6
Peripheral T cells:~ 86 92 < 1
Thymocytes 99 98 < 1
Cultured T cells 99 99 < 1
Leukemic T cell line, Jurkat 92 96 < 1
PHA blasts§ ND 94 < I
* Results expressed as the percent rosette-forming cells. Ascites fluids that contained' antibody
10.2 or 9.6 were tested at a dilution of 1 : 1,000. Rosette inhibition with antibody 9.6 persisted to
a dilution of 1:64,000. ND, not determined.
:~ Nylon wool-nonadherent peripheral blood lymphocytes.
§ Nylon wool-nonadherent peripheral blood lymphocytes were cultured for 72 h in complete
medium with 10% pooled human serum and 12 #g/ml phytohemagglutinin (PHA-P: Burrnughs
Wellcome & Co., Research Triangle Park, N. C.).
cells. Molt-4F cells, enriched for E + cells were 9.6 positive, whereas Molt-4F cells
depleted of E + cells were 9.6 negative. Cells from patients with T cell ALL whose
leukemic blasts formed E-rosettes were 9.6 positive. Leukemic blasts from patients
with null cell ALL were 9.6 negative.
The association between the E receptor and the 9.6 antigen was further investigated
by blocking experiments. Cells preincubated with antibody 9.6 did not form E-rosettes
(Table II). Incubation with either of the two other monoelonal T cell antibodies, 9.3
(data not shown) or 10.2, did not inhibit E-rosetting. Rosette inhibition by antibody
9.6 paralleled its activity in a complement-mediated cytotoxicity assay.
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981

210
KAMOUN ET AL. BRIEF DEFINITIVE REPORT
TABLE III
Lysostripping
Treated cells tested by indirect
Lysostripping E + immunofluorescence:~
antibody* 9.6 10.2
None 94 94 91
9.6 4 0 92
1(I.2 95 77 0
* Cells were incubated with monoclonal antibody diluted in medium for 0.5 h at 22°G
and washed three times. Cells were then incubated with fluorescein-conjugated goat
anti-mouse lgG serum for 45 min at 37°C and the washing was repeated. Fluorescein
staining was detected only in polar caps. Rosetting was assayed with SRBG^m,, Numbers
in the Table repre~nt the percentage of rosetting cells.
:~ To determine whether any uncapped mouse immunoglobulin remained on the cells
after lysostripping, cells were stained with rhodamine-conjugated goat F(ab')2 fragments
specific for mouse immunoglobulin. No rhodamine staining could be detected outside
the polar caps. Treated cells were also tested with 9.6 antibody and 10.2 antibody by
indirect immunofluorescence to determine whether the respective antigens had been
completely capped. The rhodamine-conjugated antiserum facilitated distinction be-
tween staining of previously formed polar caps and circumferential staining of antigens
that remained distributed on the cells. Numbers in the Table represent percent positive
cells.
Lysostripping of cells incubated with monoclonal antibody was performed to
determine whether the E receptor and the 9.6 antigen could be distinguished by co-
capping. Cells lysostripped of 9.6 antigen did not form E-rosettes, whereas cells
lysostripped of the 10.2 antigen were still capable of rosette formation (Table III).
Capping of antigen-antibody complexes appeared complete, as assessed by indirect
immunofluorescence (Table III) and quantitative cytotoxicity (data not shown). It
appeared, therefore, that the failure of cells stripped of 9.6 antigen to form E rosettes
was not a result of residual blocking antibody. When cells were lysostripped of 9.6
antigen and then incubated in fresh medium at 37°C for 18 h, they again expressed
9.6 antigen and formed E-rosettes.
Results of precipitation assays with antibody 9.6 and an 12~I-labeled lysate of the
leukemic T cell line Jurkat are presented in Fig. 1. The antibody precipitated a single
polypeptide of -50,000 mol wt. In this same assay are shown the results with
antibodies 9.3 and 10.2, which precipitated proteins of 45,000 and 67,000 daltons,
respectively. The 45,000-dalton protein precipitated by 9.3 antibody was not associ-
ated with flz-microglobulin. This was in contrast to that observed with precipitates
formed by antibodies against HLA heavy chain (W6/32) and fl2-microglobulin.
Control reactions performed with normal mouse serum and monoclonal antibody 7.2
against human Ia antigen failed to show precipitation with the cell lysate. In
additional precipitation assays with 12~I-labeled lysates of normal T cells (data not
shown), the 9.6 antibody demonstrated essentially similar results to those found with
the Jurkat cell line, although the antigen precipitated from normal T cells showed a
slightly more disperse pattern on the gel. These results suggested that the antigen
precipitated by 9.6 may have variable glycosylation patterns on T cells from different
sources. Alternatively, the different pattern on gels might reflect mild proteolytic
degradation of the antigen that varied from one T cell type to another.
Summary
We describe a new monoclonal murine antibody that reacts with a 50,000-mol wt
polypeptide that appears to be present on all E-rosetting cells. We conclude that this
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981

KAMOUN ET AL. BRIEF DEFINITIVE REPORT 211
FtG. 1. Radioimmune precipitation assays with monoclonal antibodies and a lzSI-surface-labeled
lysate of the leukemic T cell line Jurkat. Aliquots of radiolabeled cell lysate were tested in immune
precipitation assays with: (1) normal mouse serum (1:50 dilution); (2) monoclonal 7.2 antibody
from ascites fluid (1:500) which recognizes a monomorphic determinant of human la antigens (9);
(3) rabbit anti-human/32-microglobulin serum (Dakopatts, Copenhagen, Denmark; 1:50 dilution);
(4) monoclonal antibody W6/32 (Allied Chemical Corp., Morristown, N.J.; 1:100 dilution) which
recognizes a monomorphic determinant of the HLA-A,B,C heavy chain (21); (5) monoclonal 9.3
antibody from ascites fluid (1:500 dilution) which recognizes a 45,000-dahon protein specific for T
cells (9); (6) monoclonal 10.2 antibody from ascites fluid (1:500 dilution) which recognizes a 65,000-
67,000 dalton protein specific for T ceils (11); and (7) monoclonal 9.6 antibody from ascites fluid
(1:500 dilution). Immune precipitates were collected on S. aureus, eluted in electrophoresis buffer
that contained 2-mercaptoethanol, and analyzed by SDS-PAGE. Radioautography of the dried gel
was enhanced for 2 d on x-ray intensifying screens.
antigen is either identical to or closely associated with the E receptor because of (a)
the high degree of concordance between E-rosette formation and 9.6 antigen expres-
sion, (b) the inhibition of rosette formation by preincubation of cells with 9.6 antibody,
and (c) the observed failure of cells lysostripped of 9.6 antigen to form E-rosettes. This
last finding suggests cocapping of 9.6 antigen and the E receptor.
Received for publication 1 August 1980 and in revised form 3 November 1980.
References
1. Jondal, M., G. Holm, and H. Wigzell. 1972. Surface markers on human T and B
lymphocytes. I. A large population of lymphocyte* forming nonimmune rosettes with sheep
red blood cells.J. Exp. Med. 136:207.
2. Forland, S. S. 1972. Binding of sheep erythrocytes to human lymphocytes. A probable
marker of T lymphocytes. Scand. J. Immunol. 1:269.
3. Hoffman, T., and H. G. Kunkel. 1976. The E rosette test. In In Vitro Methods in Cell-
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981

212 KAMOUN ET AL. BRIEF DEFINITIVE REPORT
Mediated Tumor Immunity. B. R. Bloom and J. R. David, editors. Academic Press, Inc.,
New York. 71.
4. McMichael, A. J.,J. R. Pilch, G. Galfre, D. Y. Mason, J. W. Fabre, and C. Milstein. 1979.
A human thymocyte antigen defined by a hybrid myeloma monoc]onal antibody. Eur. J.
Imraunol. 9:205.
5. Reinherz, E. L., P. C. Kung, G. Goldstein, and S. F. Schlossman. 1979. A monoclonal
antibody with selective reactivity with functionally mature human thymocytes and all
peripheral human T cells. J. Immunol, 123:1312.
6. Kung, P. C., G. Goldstein, E. L. Reinherz, and S. F. Schlossman. 1979. Monoclonal
antibodies defining distinctive human T cell surface antigens. Science (Wash. D. C.). 206:
347.
7. Levy, R., J. Dilley, R. I. Fox, and R. Warnke. 1979. A human thymus-leukemia antigen
defined by hybridoma monoclonal antibodies. Proc. Natl. Acad. Sci. ~Z S. A. 76:6552.
8. Reinherz, E. L., P. C. Kung, G. Goldstein, R. H. Levey, and S. F. Schlossman. 1980.
Discrete stages of human intrathymic differentiation: analysis of normal thymocytes and
leukemic lymphoblasts ofT-cell lineage. Proc. Natl. Acad. Sci. U. S. A. 77:1588.
9. Hansen, J. A., P. J. Martin, and R. C. Nowinski. 1980. Monoclonal antibodies identifying
a novel T-cell antigen and la antigens of human lymphocytes, lmraunogenetics. 10:247.
10. Wang, C. Y., R. A. Good, P. Ammirati, G. Dymbort, and R. L. Evans. 1980. Identification
of a p69,71 complex expressed on human T cells sharing determinants with B-type chronic
lymphocytic leukemic cells.]. Exp. Med. 151:1539.
11. Martin, P. J., J. A. Hansen, R. C. Nowinski, and M. A. Brown. A new human T-cell
differentiation antigen: unexpected expression on chronic lymphocytic leukemia cells.
[mmunogenetics. In press.
12. Kaplan, M. E., and C. Clark. 1974. An improved rosetting assay for detection of human T
lymphocytes. J. Immunol. Methods. 5:131.
13. Ruscetti, F. W., D. A. Morgan, and R. C. Gallo. 1977. Functional and morphologic
characterization of human T cells continuously grown in vitro. J. lmmunol, 199:131.
14. Vitetta, E. S., S. Baur, andJ. W. Uhr. 1971. Cell surface immunoglobulin. II. Isolation and
characterization ofimmunoglobulin from mouse splenic lymphocytes.J. Exp. Med. 134:242.
15. Ledbetter, J. A., J. W. Goding, T. T. Tsu, and L. A. Herzenberg. 1979. A new mouse
lymphoid alloantigen (Lgpl00) recognized by a monoclonal rat antibody, lmmunogenetics.
8:347.
16. Siadak, A. W., and R. C. Nowinski. Thy-2: a murine thymocyte-brain alloantigen controlled
by a gene linked to the major histocompatibility complex, lmmunogenetics. In press.
17. Kessler, S. W. 1975. Rapid isolation ofamigens from cells with a staphylococcal protein A-
absorbent: parameters of the interaction of antigen-antibody complexes with protein A.J.
Immunol. 115:1617.
18. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature (Lond.). 227:680.
19. Swanstrom, R., and P. R. Shank. 1978. X-ray intensifying screens greatly enhance the
detection by autoradiography of the radioactive isotopes 32p and 12'~I. Anal. Biochem. 86:184.
20. Hansen, J. A., S. M. Fu, P. Antonelli, M. Kamoun, J. N. Hurley, R. J. Winchester, B.
Dupont, and H. G. Kunkel. 1979. B-lymphoid cell lines derived from HLA-D homozygous
donors, lmmunogenetics. 8:51.
21. Barnstable, C. J., W. F. Bodmer, G. Brown, G. Galfre, C. Milstein, A. F. Williams, and A.
Ziegler. 1978. Production of monoclonal antibodies to group A erythrocytes, HLA and
other human cell surface antigens--new tools for genetic analysis. Cell, 14:9.
on July 8, 2015jem.rupress.orgDownloaded from
Published January 1, 1981