(CANCER RESEARCH 51, 6236-6242. December 1, 1991|
An Immunotoxin Prepared with Blocked Ricin: A Natural Plant Toxin Adapted for
Therapeutic Use1
John M. Lambert, Victor S. Goldmacher, Albert R. Collinson, Lee M. Nadler, and Walter A. Blättler2
ImmunoGen, Cambridge, Massachusetts 02139 [J. M. L., V. S. G., A. R. C., W. A. B.J, and Division of Tumor Immunology, Dana-Farber Cancer Institute, Harvard
Medical School, Boston, Massachusetts 021 IS fL. M. N.¡
ABSTRACT
Ricin, the cytotoxic protein isolated from castor beans, is composed of
two subunits, A-chain and B-chain. Ricin intoxicates cells by binding
through its B-chain to galactose-terminated oligosaccharides found on
the surface of all eukaryotic cells and then transferring its A-chain to the
cytosol where it disrupts protein synthesis by inactivating ribosomes. In
addition to binding, the B-chain plays an important, but not yet under
stood, role in the translocation of the A-chain through a cellular mem
brane to the cytosol. Blocking the two galactose-binding sites of native
ricin by chemical modification with affinity ligands created an altered
toxin, called blocked ricin, that has at least a 3500-fold lower binding
affinity and is more than 1000-fold less cytotoxic than native ricin for
Namalwa cells (a Burkitt's lymphoma line) but that has maintained the
translocation function of the B-chain and the catalytic activity of the Achain. Conjugation of blocked ricin to monoclonal antibodies that bind to
cell surface antigens creates new cytotoxins that approach the potency of
native ricin. These cytotoxins incorporate the three essential functions of
natural toxins, i.e., binding to cells, transport through a membrane, and
catalytic inactivation of an essential cellular process; but in addition they
possess a defined cellular target specificity. Such potent immunotoxins
may play an important therapeutic role in cancer treatment. Clinical trials
with an anti-CD19-blocked ricin and an anti-CD33-blocked ricin conju
gate against B-cell cancers and acute myeloblastic leukemia have begun.
notion that replacement of the B-chain, whose only role ap
peared to be a lectin-like binding function, by a monoclonal
antibody with specificity for a selected cell-specific surface
antigen, would result in toxic agents of great specificity and
high cytotoxicity (8-12). Such immunotoxins can potentially
be of therapeutic value in diseases of clonal origin, and it is
widely believed that they could greatly advance the treatment
of cancer.
Immunotoxins made from ricin A-chain or from the homol
ogous single-chain ribosome-inactivating proteins (13,14) have
the specificity of the component antibody but are generally far
less potent than native ricin (15-18). In previous work, we have
described the preparation of blocked ricin in which the galac
tose-sélectiveoligosaccharide-binding sites of ricin are blocked
by the covalent attachment of ligands derived from glycopeptides containing triantennary TV-linked oligosaccharides (19).
We proposed that blocked ricin may be an ¡dealmolecule as a
cytotoxic agent for antibody-mediated delivery to target cells
(19). In this paper, we describe a new potent immunotoxin
made with blocked ricin and the anti-CD 19 monoclonal anti
body anti-B4 (20). The new immunotoxin combines the speci
ficity of the antibody with a potency similar to that of ricin as
assessed by direct assays of surviving fractions.
INTRODUCTION
For more than a century, the potent toxicity and cytotoxicity
(1,2) of ricin, the toxin isolated from castor beans, has been of
great interest. The finding that ricin displayed a strong antitumor effect on Ehrlich ascites tumors in mice (3) and the advent
of monoclonal antibodies (4) and their potential for the targeted
delivery of potent drugs to diseased tissue (5), such as tumors,
seemed to offer an avenue for a therapeutic application of the
toxic action of ricin.
Ricin is a cytotoxic lectin isolated from the beans of Ricinus
communis and is composed of two subunits of about equal size,
the A-chain and the B-chain (6). Ricin binds through its Bchain to galactose-terminated oligosaccharides on the surface
of cells and then transfers its A-chain to the cytosol of the cell,
where the ribosomes are catalytically inactivated. The A-chain
acts on eukaryotic ribosomes as a specific TV-glycosidase, re
moving the adenine base at position 4324 of the 28S rRNA in
the large 60S ribosomal subunit (7).
The binding of ricin to the cell surface is an essential part of
cellular intoxication, but this binding is responsible, at the same
time, for the nonselective action of the toxin, since the ricin Bchain binds to all galactose-terminated ligands, which are ubiq
uitous on all eukaryotic cell surfaces. These facts led to the
Received 5/7/91; accepted 9/24/91.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Major parts of the work described in this article were performed during the
appointment of all authors in the Division of Tumor Immunology at the DanaFarber Cancer Institute and Harvard Medical School and were supported by a
grant from ImmunoGen, Inc.
2To whom requests for reprints should be addressed, at Immunogen, Inc., 148
Sidney Street, Cambridge, MA 02139.
MATERIALS
AND METHODS
Materials. Ricin [ricin D in the nomenclature of Wei and Koh (21)]
was purchased from E-Y Laboratories (San Mateo, CA) and from
Inland Laboratories (Austin, TX). Ricin A-chain was also from Inland
Laboratories. Purified monoclonal antibody anti-B4 (20) was prepared
as described previously (15). A rabbit reticulocyte lysate system for cellfree protein synthesis, which included L-[3,4,5-3H]leucine, was obtained
from New England Nuclear (Boston, MA). Succinimidyl 3-(2-pyridyldithio)-propionate and succinimidyl 4-(A'-maleimidomethyl)-cyclohexane-1-carboxylate were from Pierce Chemical Co. (Rockford, IL). 2-(2Pyridyl-dithio)-ethanol was the generous gift of Dr. Ravi V. J. Chari
and is now available from Sigma Chemical Co. (St. Louis, MO).
Blocked Ricin. Blocked ricin was prepared from ricin by chemically
blocking the galactose-binding sites of the cytotoxic lectin with reactive
ligands made by chemical modification of glycopeptides containing
triantennary TV-linkedoligosaccharides derived from fetuin using meth
ods described elsewhere (19). The only difference in the preparation of
blocked ricin used here, from that described previously (19), is that the
sulfhydryl group introduced at the a-amino group of the glycopeptide
using 2-iminothiolane was capped as a mixed disulfide with 2-mercaptoethanol, rather than being alkylated using iodoacetamide. Instead of
adding the iodoacetamide in the final reaction to form the blocking
ligand (Ref. 19, compound 7), 2-(2-pyridyl-dithio)-ethanol was added
to the reaction in a 1.2-fold molar excess over total free sulfhydryl
groups in the mixture. The blocked ricin was purified and characterized
as described previously (19).
Preparation of Immunoconjugates. The immunoconjugate between
anti-B4 and ricin A-chain was prepared by reacting the antibody with
succinimidyl 3-(2-pyridyldithio)-propionate and then mixing the mod
ified antibody with ricin A-chain using conditions described previously
for ribosome-inactivating proteins (13). The A-chain conjugate was
then purified as described by Ghetie et al. (22). The immunoconjugate
between anti-B4 and blocked ricin was made as follows. Antibody was
6236
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS
PREPARED WITH BLOCKED RICIN
modified with succinimidyl 4-(A'-maleimidomethyl)-cyclohexane-l-carboxylate in SO HIMsodium phosphate buffer, pH 7.0, containing NaCl
(50 mM) at 30°Cfor 30 min and separated from excess reagent by gel
filtration through Sephadex G-25 in the above pH 7.0 buffer. Blocked
ricin (3 mg/ml), having ligands containing disulfide groups, was reduced
with dithiothreitol (3 HIM)in 10 mM potassium phosphate buffer, pH
6.8, containing NaCl (0.15 M), at 0°Cfor 20 h, and then separated
from excess reducing agent by gel filtration through Sephadex G-25 in
5 mM sodium acetate buffer, pH 4.7, containing NaCl (50 mM) and
EDTA (1 mM). Under these conditions, the reduction is limited to the
disulfide bond of the ligands that are covalently bound to the ricin;
there was no reduction of the disulfide bond between A-chain and Bchain as shown by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (result not shown). The modified antibody (25 mg) was mixed
with reduced blocked ricin (5 mg) at pH 7.0 and kept at 4'C for 16 h.
The anti-B4-blocked ricin conjugate was purified by ion-exchange chromatography on a column of S-Sepharose equilibrated in 50 mM sodium
acetate buffer, pH 5.0, eluting the conjugate with a gradient of NaCl
(0-400 mM), followed by affinity chromatography using a monoclonal
anti-blocked ricin antibody column, eluting the conjugate with 0.1 M
glycine-HCI buffer, pH 2.7. Finally the conjugate was submitted to gel
filtration on Sephacryl S-300 in 10 mM potassium phosphate buffer,
pH 7.4, containing NaCl (0.15 M).
Measurement of Protein Concentration. The concentrations of solu
tions of purified proteins were determined from their A..„„.
assuming
E1%,cmvalues of 14.0 for IgG (13) and 11.8 for ricin (23) and blocked
ricin (the oligosaccharide-containing ligands that block the sugar-bind
ing sites of blocked ricin have negligible absorbance at 280 nm). The
protein concentrations of purified conjugates were also estimated from
.•V,,,
values, assuming antibody-blocked ricin conjugates of molar ratio
1:1 for which an Eisicm value of 13.4 was calculated from the above
values.
Cell Free Protein Synthesis Assay. The activity of ricin A-chain in
the inhibition of protein synthesis was measured in a rabbit reticulocyte
lysate system similar to that described previously (24). The samples of
ricin or its derivatives were diluted to 2 Mg/ml of ricin A-chain in 10
mM sodium phosphate buffer, pH 7.4, containing NaCl (20 mM), bovine
serum albumin (0.2 mg/ml), and 2-mercaptoethanol (1% v/v) and
incubated at 37°Cfor 2 h to effect reduction. The test articles were
then further diluted in the same buffer containing 2-mercaptoethanol
(5 mM), and samples ( 1 n\ containing 0.4-100 ng of ricin A-chain) were
added to each assay (12 n\ total volume, containing 4 ¿il
of reticulocyte
lysate). The incorporation of [3H]leucine into protein precipitable by
trichloroacetic acid after incubating for 50 min at 30°Cwas measured
as described previously (24).
Binding Assays for Ricin and Blocked Ricin. Namalwa cells (American
Type Culture Collection CRL 1432; 2 x 10' cells/sample) were incu
bated successively for 30 min at 0°Cwith various concentrations of
ricin or blocked ricin in AB buffer [2.5% pooled human AB (Pel Freeze
Biologicals, Rogers, AR) in minimal essential medium (Cellgro, Herndon, VA) containing 10 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid buffer, pH 7.0], then with rabbit anti-ricin immunoglobulin
(Sigma Chemical Co.), and finally with fluorescein-labeled goat antirabbit immunoglobulin (Sigma Chemical Co.), with a buffer wash
between incubations. Treated cells were then fixed with 1% formalde
hyde in 10 mM potassium phosphate buffer, pH 7.2, containing 150
mM NaCl, and analyzed on a flow cytometer (FACScan; BectonDickinson, Mountain View, CA) set on linear fluorescence. The rabbit
anti-ricin immunoglobulin was shown in an enzyme-linked ¡mimmo
sorbent assay to bind ricin and blocked ricin equally well.
Competition Binding Assays. Namalwa cells (3 x IO5cells/well) were
placed in 96-well plates in 25 ¿il
of AB buffer, and 25 n\ of AB buffer
containing fluorescein-labeled ant i 1J4antibody (2-6 nM) together with
various concentrations of competing reagent were added to each well.
Positive controls lacked competing antibody while negative controls
lacked fluorescein-labeled anti-B4. The plate was incubated at 4°Cfor
set on linear fluorescence, and the binding of fluorescently labeled
antibody was expressed as a percentage of the fluorescence of the
positive control samples.
Cytotoxicity Assays. Test samples were incubated with cells at 37*C
for 24 h. The cells were then washed and placed into fresh medium for
the determination of the surviving fractions by the direct cytotoxicity
assay called growth back-extrapolation as described by Goldmacher et
al. (25) or by a clonogenic assay described by Goldmacher et al. (26)
with the following modifications. Cells were plated in medium that was
supplemented with 20% (rather than 10%) heat-inactivated fetal calf
serum, which increased the plating efficiency of the cells to such an
extent that no feeder cells were necessary. Cytotoxicity results were
compared as IC373values. When the number of lethal hits received by
a population of cells is equal to the number of cells in the population,
a Poisson distribution of lethal hits predicts that a surviving fraction of
0.37 (\/e) remains.
RESULTS
Conjugate between Anti-B4 Monoclonal Antibody and Ricin
A-Chain. We first prepared an anti-B4-ricin A-chain inumino
conjugate by allowing the free sulfhydryl group of A-chain to
react with anti-B4 antibody that had been modified with succi
nimidyl 3-(2-pyridyldithio)-propionate,
thus forming a conju
gate via a disulfide bridge. The purified conjugate contained
less than 5% of free antibody or of free ricin A-chain (Fig. 1, a
and b, lanes 5) and was shown to have completely preserved
the catalytic activity of the A-chain moiety (Fig. 2a) and to
have about a 2-fold lower avidity for antigen than the free
antibody (Fig. 2c). The cytotoxicity and specificity of this
conjugate were assayed on cultured cells in vitro by measuring
the fraction of cells surviving after treatment with the immunotoxin at selected concentrations for a fixed period of time.
Surprisingly, it was found that the degrees of cytotoxicity
toward antigen-positive Namalwa cells and toward antigennegative Molt-4 cells were very similar and were comparable to
the cytotoxicity inflicted by ricin A-chain alone (Fig. 3, a and
b). We speculated that the low specific cytotoxicity could be
explained by the inability of conjugates of ricin A-chain to
transfer the A-chain efficiently into the cytoplasm of the target
cells. In contrast, ricin is an extremely potent toxin for Na
malwa cells (Fig. 3, a and c), from which it can be inferred that
native ricin contains the function of efficiently transferring the
A-chain into the cytoplasm and that the B-chain is important
for this function.
Conjugate between Anti-B4 Antibody and Blocked Ricin. We
therefore developed methods for the efficient blocking of the
lectin-binding sites of ricin (19). Covalent modification of ricin
with high-affinity ligands isolated from fetuin and purification
of the product by affinity chromatography yielded an altered
ricin species, called blocked ricin (19), that has the following
characteristics, (a) Blocked ricin is composed of a native Achain and a B-chain that is covalently modified with two affinity
ligands (major species, about 70%) or three affinity ligands
(minor species, about 30%) (Fig. \b, Lane 2). (b) The A-chain
of blocked ricin is unmodified, and its full catalytic activity is
preserved (Fig. 2a). (c) The binding affinity of blocked ricin
toward Namalwa cells is at least 104-fold lower than that of
native ricin (Fig. 2¿>),
and blocked ricin binds less well (about
102-fold) than ricin in the presence of 30 mivi lactose. Interest
ingly, at high concentrations of blocked ricin (10~5 M, 0.3 mg/
ml), lactose (at 30 mM) can further reduce the binding of
30 min. The cells were then washed once with AB buffer (175 ^1) and
3The abbreviations used are: IC37,1C»,the molar inhibitory concentration for
fixed with 1% formaldehyde in the 10 mM phosphate-buffered saline,
a toxin that leaves a surviving fraction of 0.37 (63% of cells are killed) and 0.50
pH 7.2 (200 fi\). The labeled cells were analyzed on a flow cytometer
(50% of cells are killed), respectively.
6237
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS
top -
PREPARED WITH BLOCKED RICIN
123456
1 23456
-top
-205
-116
-93
-68
205co
b
-
11693-
-45
68-
45-
-29
29Fig. 1. Analysis of blocked ricin, ricin A-chain, and their conjugates with the murine monoclonal antibody anti-B4, by sodium dodecyl sulfate-polyacrylamide gel
electrophoresis. a, a gel [5-10% (w/v) polyacrylamide gradient] run under nonreducing conditions with the following samples (1 fig of total protein in each lane):
Lane 1, ricin; Lane 2, blocked ricin; Lane 3, ricin A-chain; Lane 4, purified anti-B4-blocked ricin conjugate; Lane 5, purified anti-B4-ricin A-chain conjugate; Lane 6,
anti-B4 antibody, b, a gel [11% (w/v) polyacrylamide] run under reducing conditions with the same samples as in a ( 1.5-2.5 f<gof total protein in each lane). Ricin Achain shows two bands of apparent M, 30,000 and 32,000 under both nonreducing and reducing conditions (a and b. Lanes 3), corresponding to species of ricin Achain containing one (band at M, 30,000) or two (band at M, 32,000, the A'-chain) high-mannose yV-linked oligosaccharides (54, 55). Blocked ricin migrates as a
broad band of apparent M, 66,000-69,000 in sodium dodecyl sulfate-polyacrylamide gels under nonreducing conditions (a, Lane 2), while native ricin migrates as a
doublet of apparent M, 60,000-61,500 (a, Lane I). Analysis of blocked ricin under reducing conditions shows bands corresponding to native A-chain (M, 30,000),
A'-chain (32,000) and two other bands of apparent M, 37,500 and M, 40,000 (ft, Lane 2) that are not present in a sample of native ricin (Lane I). Native H chain (M,
32,500) is absent from blocked ricin. The new bands in blocked ricin are the result of covalent modification of B-chain with two (M, 37,500) or three (M, 40,000)
molecules of activated ligand (M, 2,422) (19). The gels were calibrated with the marker proteins carbonic anhydrase (M, 29,000). ovalbumin (M, 45,000), bovine
serum albumin (M, 68,000), phosphorylase b (M, 93,000), /3-galactosidase (M, 116,000), and myosin (M, 205,000) and were run according to the method of Laemmli
(56).
blocked ricin (Fig. 2b). This effect could not be demonstrated
with other sugars such as mannose and sucrose. Further puri
fication of blocked ricin did not change these results, which
suggests that blocked ricin at high concentrations is still capable
of interacting weakly with galactose residues, (d) Cytotoxicity
experiments with Namalwa cells (Fig. 3, a and c) showed that
blocked ricin is at least 2000-fold less cytotoxic (IC37 = 1.3 x
10~8 M) than native ricin (IC37 = 4.5 x 10~12M) and about 3fold less cytotoxic than ricin in the presence of 30 IHMlactose
(IC37 = 3.5 x 10~9 M). The cytotoxicity of blocked ricin was
still lower by about 5-fold (IC37 = 7.4 x 10~8M) in the presence
of 30 mivi lactose (Fig. 3a). Further affinity purification of
blocked ricin does not suppress the effect of lactose, which
again suggests that blocked ricin may retain some residual
galactose-binding activity of low affinity, rather like a mutant
protein (see, for example, mutant diphtheria toxins, Ref. 27).
Notably, blocked ricin is about 20-fold more cytotoxic than
ricin A-chain alone (IC37 = 3 x 10~7M), which can be explained
by the enhancement of transmembrane transport of A-chain by
the blocked B-chain after internalization by pinocytosis.
In anticipation of the need to attach blocked ricin to a
targeting antibody, the affinity ligands used for the preparation
of blocked ricin contain a disulfide group, which can be reduced
in the blocked ricin product without affecting the native disul
fide bonds of the protein (19). Approximately 1 free sulfhydryl
group/blocked ricin molecule was generated in this way. The
reduced blocked ricin was then allowed to react with anti-B4
antibody that had been modified previously with succinimidyl
4-(Ar-maleimidomethyl)-cyclohexane-l-carboxyIate
so as to
contain an average of 0.5 maleimido group/antibody molecule.
In this way, the immunoconjugate anti-B4-blocked ricin was
produced via a stable thioether link. This procedure also en
sured that the linkage between antibody and blocked ricin
involved only the B-chain of blocked ricin. The ricin A-chain,
therefore, is linked to the immunoconjugate only by its natural
native covalent and noncovalent interactions with the B-chain.
The resulting conjugate was purified by ion exchange chromatography, immunoaffinity chromatography using a mono
clonal anti-blocked ricin antibody column, and gel filtration.
The final immunotoxin preparation was analyzed by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (Fig. la,
Lane 4) and shown to be composed largely of a conjugate of
one antibody molecule linked to one blocked ricin molecule
(about 80%) with the remainder being species of higher molec
ular weight. The conjugate contained little if any free antibody
and no detectable free ricin or free blocked ricin (Fig. la, Lane
4). Electrophoresis of the conjugate under reducing conditions
(Fig. \b, Lane 4) shows bands for ricin A-chain and A'-chain
but no bands for free B-chain or free blocked B-chains, confirm
ing that the A-chain can be fully released by reduction and that
the blocked B-chain is linked in a stable manner to the antibody.
The bands that can be observed in addition to the bands for the
heavy and light chains of the antibody are mostly of higher
molecular weight than the heavy chain, indicating that the
conjugation of blocked ricin largely involves the heavy chain.
The facile release of intact A-chain from the immunotoxin was
6238
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS
IO'1
PREPARED WITH BUX KED RICIN
10
Concentration (M)
Fig. 2. Functional assays for ricin A-chain activity and for binding activity of ricin, blocked ricin, and anti-B4 immunoconjugates. For experimental details see
"Materials and Methods."/), inhibition of protein synthesis in a rabbit reticulocyte cell-free translation system by ricin (D), ricin A-chain (•),blocked ricin (O), antifvl linn A-chain (A), and anti-B4-blocked ricin (A). Samples were incubated for 50 min at 30°C,and then the incorporation of [3H]leucine into protein precipitable
by trichloroacetic acid was measured as described previously (24). Control assays (without inhibitors) show incorporation of 107,000 cpm in 50 min. B, binding of
ricin (D, •¿)
and blocked ricin (O, •¿)
to Namalwa cells. •¿.
•¿.
samples incubated in the presence of 30 mvi lactose. C, competition binding assay comparing .uni It4
ricin-A-chain conjugate (A) and ami-B4-blocked ricin conjugate (A) with native anti-B4 (•).Namalwa cells (3 x 10* cells/sample) were incubated with solutions of
fluorcscein-labeled anti-B4 antibody (2 n\i i containing various concentrations of competing reagent. Positive controls lacked competing antibody, while negative
controls lacked fluorescein-labeled anti-B4. Cell-associated fluorescence was determined on a flow cytometer and is plotted as a percentage of the fluorescence of the
positive control sample versus the concentration of competing reagent.
further confirmed by using a cell-free protein translation system
(Fig. 2a). Purified ricin A-chain and the ricin A-chains from
reduced native ricin, reduced blocked ricin, and reduced antiB4-blocked ricin were found to be equally efficient in inhibiting
protein synthesis (IC50 = 4 x 10~': M), showing that the catalytic
activity of ricin A-chain is completely preserved in the
immunotoxin.
The antibody moiety of the conjugate was tested for its
relative avidity in a competitive binding assay, in which the
inhibition of binding of fluorescein-labeled anti-B4 antibody by
free unmodified anti-B4 antibody and that by anti-B4-blocked
ricin immunoconjugate were compared (Fig. 2c). A concentra
tion of conjugate about 2-fold higher than that of native anti
body was needed to achieve a similar degree of competition.
Thus, with respect to the properties of antigen binding (Fig. 2c)
and ricin A-chain activity (Fig. 2a), the conjugate of anti-B4
with blocked ricin is virtually identical to the conjugate with
ricin A-chain. We could then approach the question of whether
the transport function of ricin B-chain (assisting the transloca
tion of the A-chain) had also been preserved.
Comparison of in Vitro Cytotoxicities of Anti-B4-blocked Ri
cin and Anti-B4-Ricin A-Chain Conjugates. The potency and
specificity of anti-B4-blocked ricin was assayed in vitro using
Namalwa cells (a B4-positive Burkitt's lymphoma line) and
Molt-4 cells (a B4-negative T-cell leukemia line) in cytotoxicity
assays that directly measure the surviving fraction of treated
cell populations. As expected, anti-B4-blocked ricin and free
unconjugated blocked ricin show very similar cytotoxicites, with
identical IC37 values of 2.5 x 10~9 M (Fig. 3A), toward the
antigen-negative Molt-4 cells. (Note that a comparison of Fig.
3, a and b, shows that Molt-4 cells are about 5-fold more
sensitive to blocked ricin than Namalwa cells, even though the
sensitivities of the two cell lines toward ricin and ricin A-chain
are very similar.) However, with B4-antigen-positive Namalwa
cells (Fig. 3a), anti-B4-blocked ricin is about 10'-fold more
cytotoxic (with an IC37 of 1.3 x 10"" M) than unconjugated
blocked ricin (with an IC.17of 1.3 x 10~8M). The immunotoxin
kills nearly all the cells in the population after a 24-h exposure,
leaving surviving fractions of treated cells of less than 10 4 at
immunotoxin concentrations high enough (10~9 M) to complex
most antigens (Fig. 3, a and c). The cytotoxicity for Namalwa
cells requires specific binding to the B4 antigen, since cell death
is blocked by an excess (10~<)M) of nonconjugated anti-B4
antibody (Fig. 3c) and is only minimally affected by lactose
(Fig. 3a).
The toxicity curve for anti-B4-blocked ricin on Namalwa
cells is remarkably similar to that for native ricin (Fig. 3c).
Both curves approach surviving fractions of less than 10~5 in a
nearly parallel manner, with the curve for anti-B4-blocked ricin
positioned relative to that for ricin at about 10-fold higher
concentrations. [The relative positions might be explained by
the dissociation constants (K,, = 1.4 x 10~'°M for the immu
noconjugate) and the number of binding sites (Namalwa cells
have approximately 4 x IO4 B4 antigens/cell) for the different
molecules.] This similarity and the very different behavior of
anti-B4-ricin A-chain indicate that blocked ricin has maintained
the inherent capability of native ricin for efficient translocation
of the A-chain into the cytoplasm.
DISCUSSION
Our research has focused on the development of immunotoxins targeted against cancers of hematopoietic origin, in partic
ular against leukemias and lymphomas of B-cell lineage. We
selected the anti-CD 19 monoclonal antibody anti-B4 (20) for
the following reasons: (a) anti-B4 is exquisitely specific for cells
of B-cell lineage (28); (b) it is highly probable that the antigen
CD 19 is expressed on the clonogenic tumor cell, since this is a
differentiation antigen expressed at all stages during the ontog
eny of B-cells except on the final plasma cell (28); (c) the target
antigen is expressed on all neoplastic cells (29); and (d) upon
antibody binding the antigen does not disappear from the
surface of cells and is not shed into the serum of patients.
Initially, we made an immunotoxin by conjugating anti-B4
to ricin A-chain. However, this conjugate showed similar cyto
toxicity toward antigen-positive and antigen-negative cells and
6239
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS PREPARED WITH BLOCKED RICIN
1.0
0.37
0.1
1.0
.01-
IO'1
0.37
io-2
O) 1.0
'5
0.37
10'3
I o,
IO'4
CO
IO'5
.01
10'"
10"
10"
10"
10"
IO'12 IO'11
io'10io'9
10"1
Concentration of Conjugate or Toxin (M)
Fig. 3. Cytotoxicity of anti-B4-blocked ricin and anti-B4-ricin A-chain conjugates for human cell lines and comparison with the cytotoxicities of ricin, blocked
ricin, and ricin A-chain. Test samples were incubated with cells at 37°Cfor 24 h, and then the cells were washed and placed into fresh medium for the determination
of surviving fractions by the "growth back-extrapolation" assay described by Goldmacher el al. (25). A, Cytotoxicity of ricin (•),anti-B4-blocked ricin (A), anti-B4blocked ricin in the presence of 30 m,Mlactose (A), ricin in the presence of 30 mM lactose (O), blocked ricin (T), blocked ricin with 30 mM lactose (V), ricin A-chain
(D), and anti-B4-ricin A-chain (•)for the human lymphoid cell line Namalwa (American Type Culture Collection CRL 1432) which contains 4 x 10* B4 (CD19)
antigens/cell (15. 20). B, Cytotoxicity of ricin (•).blocked ricin (T), blocked ricin in the presence of 30 mM lactose (V), anti-B4-blocked ricin (A), anti-B4-blocked
ricin in the presence of 30 mM lactose (A), ricin A-chain (D), and anti-B4-ricin A-chain (•)for the human T leukemia cell line Molt-4 (American Type Culture
Collection CRL 1582). which is antigen-negative for CD19. C, Cytotoxicity curves for ricin (•,O) and anti-B4-blocked ricin (A, A) that extend the range of surviving
fractions measured to 10 '' for the human lymphoid cell line Namalwa. O, A. assays where the surviving fractions were determined by measuring plating efficiencies
as described by Goldmacher et al. (26). The Cytotoxicity of anti-B4-blocked ricin was also measured in the presence of 10~*M anti-B4 antibody (O).
was no more toxic toward these cells than ricin A-chain alone.
Ricin A-chain can be grouped in the class of proteins called
single-chain ribosome-inactivating
proteins (30-32). Previ
ously, we had argued that the degree of Cytotoxicity of immunotoxins containing such a protein correlated with the extent
of endocytosis after binding of the immunoconjugate to its
specific cell surface antigen (15) and, furthermore, that the
most important dose-limiting factor for these immunotoxins
appeared to be the low frequency with which the single-chain
toxin, such as ricin A-chain, penetrates a membrane and enters
the cytosol after endocytosis. We demonstrated that for each
cell in a population to receive, on average, one lethal hit, the
cells must maintain by endocytosis an internal concentration of
about IO4 molecules of the single-chain toxin/cell for a 24-h
period (15), even though some experiments have suggested that
a lethal hit for a ribosome-inactivating protein may be given by
as little as one molecule (or a very few) actually reaching the
cytosol (33).
In contrast, native ricin is capable of efficiently transferring
its A-chain into the cytoplasm (15). Native ricin has evidently
evolved three important functions necessary for the intoxication
of cells: binding to the cell surface receptor, transport of the
catalytic subunit into the cytoplasm of the cell, and enzymatic
catalysis of the chemical process that inactivates the ribosomes.
The ribosome-inactivating
moieties of most immunotoxins
made with ricin A-chain or with other single chain ribosomeinactivating proteins are unable to penetrate the membrane
efficiently, which suggests that this function of native ricin is
carried, at least in part, on the B-chain subunit. Indeed, the
Cytotoxicity of ricin A-chain immunotoxins is increased simply
by adding B-chain to the incubation mixture (34, 35). Experi
ments with ricin B-chain-gelonin hybrid toxins (36) further
showed that the transmembrane transport function operates
only with ricin A-chain and not with other proteins.4 We
concluded, therefore, that the preparation of specific and potent
immunotoxins might require a ricin holotoxin altered only so
as to eliminate its indiscriminate recognition of all host cells.
This conclusion led us to develop a method for blocking the
galactose-binding activity of ricin B-chain while preserving its
ability to translocate the A-chain. Early attempts to destroy the
sugar-binding sites of ricin by random chemical modifications
of functional groups have not been completely successful (3742), owing to poor yields, high random modification, and/or
incomplete blockage of both of the sugar-binding sites. lodination of tyrosine residues in ricin resulted in decreased galactose
binding and lower toxicity (37, 38). However, several tyrosine
residues were modified and the conditions of iodination are
strongly oxidizing, so that other B-chain functions could also
be compromised. Indeed, iodinated ricin B-chain was not only
impaired in galactose-binding activity but was also damaged in
other properties such as its ability to reassociate with A-chain
(39). Acetylation of tyrosine residues in ricin resulted in about
a 10-fold decrease in binding affinity and Cytotoxicity (41, 42).
Comparison of the acetylation work with our earlier work with
affinity modification, where blocking of one galactose-binding
site of ricin with a reactive galactose derivative resulted in only
about a 10-fold reduction in Cytotoxicity (43), suggests that
only one galactose-binding site of ricin was impaired by acetylating tyrosine residues. The results further suggest that the
remaining free binding site is sufficient to bind to cell surface
oligosaccharides (albeit with lower avidity) and to cause the
observed Cytotoxicity (43).
4J. M. Lambert, unpublished observations.
6240
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS PREPARED WITH BLOCKED RICIN
Another approach to decreasing the affinity of ricin for
binding galactose residues on cell surface oligosaccharide moie
ties has been to conjugate native ricin to monoclonal antibodies,
followed by fractionation of the conjugates by galactose affinity
chromatography. Species are obtained that contain ricin that
has impaired binding to galactose, apparently due to steric
hindrance by the antibody cross-linked to it (44-47). However,
these conjugates are still quite toxic to antigen-negative cells in
the absence of galactose (45, 47). The two galactose-binding
sites of ricin are 75 A apart on opposite ends of the B-chain
(48), and it is therefore likely that only one of these sites is
sterically hindered by covalent linking of one ricin molecule to
one antibody molecule vìa
a short cross-linking agent.
Our approach to blocking both of the galactose-binding sites
of ricin was based on the notion that to truly block these binding
sites, they should be targeted by a ligand of high affinity (a
natural ligand for ricin) capable of covalent reaction specifically
with the galactose-binding sites. Also, since it is not known
what structural features of the B-chain are important for other
B-chain functions, especially that of enabling efficient translo
cation of the A-chain across a membrane, we wished to target
modification of the B-chain solely to the galactose-binding sites,
thereby avoiding random chemical modification which may
impair several functions. We developed a method for blocking
the galactose-binding sites of native ricin by covalently linking
to them naturally occurring oligosaccharide-containing ligands
which bind with high affinity. The preparation, purification,
and properties of blocked ricin have been described previously
(19), and its characteristics are summarized under "Results."
Covalent linking of at least two complex oligosaccharide moie
ties results in blocking both sugar-binding sites of ricin (19).
However, our results indicate that blocked ricin does have a
residual ability to bind with low affinity to cell-surface galactose
residues. Indeed, it is possible that the remaining weak galactose
binding is responsible for the effectiveness of immunotoxins
containing blocked ricin, a point that is being further
investigated.
Blocked ricin, when covalently linked to anti-B4 via a stable
noncleavable link generated by reaction of a thiol group on a
blocking ligand of the blocked ricin with a maleimido group
introduced into the antibody, forms a very potent immunotoxin.
This is in marked contrast to the lack of cytotoxicity exhibited
by anti-B4-ricin A-chain. Another antibody to CD 19, HD37,
has also been conjugated to ricin A-chain (49-51), and these
preparations did show some target-directed cytotoxicity. Since
there is evidence to suggest that the cytotoxicity exhibited by
immunotoxins made with ricin A-chain (52) or gelonin (15) is
determined by the extent of their internalization, we presume
that HD37 can induce some internalization of CD19. The
extent of internalization of anti-B4 has been shown to be very
low (15).
The anti-CD 19 ricin A-chain conjugates are, however, much
less cytotoxic toward CD19-positive cell lines than is anti-B4blocked ricin. In direct assays of surviving fraction of tumor
cell lines with 24-h exposure to the toxin (comparable to the
cytotoxicity assays used in our work), the IC37 of the HD37ricin A-chain was only about 3 x 10"' M (taken from Ref. 49,
Fig. 1), and the lowest surviving fraction measured was 10~2at
1 X 10~7M immunotoxin. Indirect assays of surviving fraction
(assessed by measuring the incorporation of radiolabeled leucine relative to control cultures) show an IC37 of about 10~'°M
(50, 51). However, the apparent surviving fraction at 1 x 10~8
M immunotoxin
was only about 0.1 to 0.2, although this is
really at the limit of detection of the surviving fraction in assays
of this type. A conjugate of another anti-CD 19, B43, to pokeweed antiviral protein (53) gave results similar to those obtained
for the HD37-ricin A-chain conjugates. In direct assays of the
surviving fraction using a B-ALL cell line, the B43 conjugate
gave a surviving fraction of only 0.1 at about 1 x 10~8 M
immunotoxin (53).
We conclude that conjugates between antibody and blocked
ricin are highly cytotoxic agents because they retain the three
essential functions that are present in natural cytotoxins: the
monoclonal antibody binds to a cell surface receptor, the
blocked ricin B-chain translocates the A-chain into the cytosol,
and the ricin A-chain catalytically inactivates ribosomes. In
contrast to natural toxins, however, the binding specificity of
the conjugate toxins can be altered at will, which permits their
adaption for human therapeutic use. Anti-B4-blocked ricin
(anti-CD 19) and an analogous anti-My9-blocked ricin (antiCD33) conjugate have already entered clinical trials.
ACKNOWLEDGMENTS
The authors are grateful for superb technical assistance from Gail
Mclntyre, Michael Gauthier, David Zullo, Vikram Rao, Rita Sleeves,
Faith Barnard, Bernadette Keane, Beth Levine, Peter Mikkonen, Sherrilyn Cook, and Jean Stewart.
REFERENCES
1. Ehrlich, P. Experimentelle Untersuchungen überImmunität.I. Ueber Ricin.
Dtsch. Med. Wochenschr., 17: 976-979, 1891.
2. Flexner, S. The histological changes produced by ricin and abrin intoxica
tions. J. Exp. Med., 2: 197-216, 1897.
3. Lin, J-Y., Tserng, K-Y., Chen, C-C., Lin, L-T., and Tung, T-C. Abrin and
ricin: new anti-tumor substances. Nature (Lond.), 227: 292-293, 1970.
4. Köhler,G., and Milstein, C. Continuous cultures of fused cells secreting
antibody of predefined specificity. Nature (Lond.), 256:495-497, 1975.
5. Möller,G. (ed.). Antibody Carriers of Drugs and Toxins in Tumor Therapy.
Immunological Reviews, Vol. 62. Copenhagen: Munksgaard, 1982.
6. Olsnes, S., and Phil, A. Toxic lectins and related proteins. In: P. Cohen and
S. Van Heyningen (eds.), Molecular Action of Toxins and Viruses, pp. 55105. Amsterdam: Elsevier. 1982.
7. Endo, Y., Mitsui, K., Motizuki, M., and Tsurugi, K. The mechanism of
action of ricin and related toxic lectins on eukaryotic ribosomes. J. Biol.
Chem., 262:5908-5912, 1987.
8. Blythman, H. E., Casellas. P., Gros, P., Gros, O., Jansen, F. K., Paolucci,
F., Pam, B., and Vidal, H. Immunotoxins: hybrid molecules of monoclonal
antibodies and a toxin subunit specifically kill tumour cells. Nature (Lond.),
290: 145-146, 1981.
9. Vitella, E. S., and Uhr, J. W. Immunotoxins. Annu. Rev. Immunol., 3:
197-212,1985.
10. Frankel, A. E., Houston, L. L., and Isseil, B. F. Prospects for immunotoxin
therapy in cancer. Annu. Rev. Med., 37: 125-142, 1986.
11. Blakey, D. C., and Thorpe, P. E. An overview of therapy with immunotoxins
containing ricin or its A-chain. Antibody, Immunoconj. Radiopharm., /: 116, 1988.
12. Frankel, A. E. (ed.). Immunotoxins. Boston: Khmer Academic Publishers,
1988.
13. Lambert, J. M., Senter, P. D., Yau-Young, A., Blättler,W. A., and Gold
macher, V. S. Purified immunotoxins that are reactive with human lymphoid
cells. Monoclonal antibodies conjugated to the ribosome-inactivating pro
teins gelonin and the pokeweed antiviral proteins. J. Biol. Chem., 260:
12035-12041, 1985.
14. Lambert, J. M., Blättler,W. A., Mclntyre, G. D., Goldmacher, V. S., and
Scott, C. F., Jr. Immunotoxins containing single chain ribosome-inactivating
proteins. In: A. E. Frankel (ed.), Immunotoxins, pp. 175-209. Boston:
Kluwer Academic Publishers, 1988.
15. Goldmacher, V. S., Scott, C. F., Jr., Lambert, J. M., Mclntyre, G. D.,
Blättler,W. A., Collinson, A. R., Stewart, J. K., Chong, L. D., Cook, S.,
Slayter, H. S., Beaumont, E., and Watkins, S. Cytoloxicity of gelonin and
its conjugates with antibodies is determined by the extent of their endocytosis.
J. Cell. Physiol., 141: 222-234, 1989.
16. Thorpe, P. E. Antibody carriers of cytotoxic agents in cancer therapy: a
review. In: A. Finchera, G. Doria, F. Dammacco, and A. Bargellesi (eds.).
Monoclonal Antibodies '84: Biological and Clinical Applications, pp. 475512. Milan, Italy: Editrice Kurtis s. r. I., 1985.
17. Blättler,W. A., Lambert, J. M., and Goldmacher, V. S. Realizing the full
6241
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMMUNOTOXINS
PREPARED WITH BLOCKED RICIN
potential of immunotoxins. Cancer Cells (Cold Spring Harbor), /: 50-55,
1989.
18. Vitella, E. S., and Uhr, J. W. The potential use of immunotoxins in trans
plantation, cancer therapy, and immunoregulation. Transplantation (Balti
more). 37: 535-538, 1984.
19. Lambert. J. M.. Mclntyre, G., Gauthier, M. N., Zullo, D., Rao. V., Sleeves,
R. M.. Goldmacher, V. S., and Blatller, W. A. The galaclose-binding siles of
the cytotoxic lectin ricin can be chemically blocked in high yield with reactive
ligands prepared by chemical modificalion of glycopeptides containing triantennary A'-linked oligosaccharides. Biochemistry, 30: 3234-3247, 1991.
20. Nadler, L. M., Anderson, K. C.. Marti, G., Bales, M. P., Park, E., Daley, J.
F., and Schlossman, S. F. B4, a human B lymphocyle-associated anligen
expressed on normal, milogen aclivated, and malignant B lymphocytes. J.
Immunol., 131: 244-250, 1983.
21. Wei, C. H., and Koh, C. K. Crystallographic characterization of a principal
non-toxic leclin from seeds of Ricinus communis. J. Mol. Biol., 123: 707711, 1978.
22. Ghelie, V., Ghelie, M-A., Uhr, J. W., and Vitella, E. S. Large scale preparalion of immunoloxins conslrucled with ihe Fab' fragment of IgGi murine
monoclonal anlibodies and chemically deglycosylaled ricin A-chain. J. Im
munol. Melhods, 112: 267-277, 1988.
23. Olsnes, S. Ricin and Ricinus agglulinin. loxic leclins from caslor bean.
Melhods Enzymol., 50: 330-335, 1978.
24. Blattler, W. A., Kuenzi, B. S., Lambert, J. M., and Senler, P. D. New
helerobifunclional prolein cross-linking reageni thai forms an acid-labile
link. Biochemislry, 24:1517-1524, 1985.
25. Goldmacher. V. S., Anderson, J.. Blattlcr, W. A., Lamben, J. M.. and Senler.
P. D. Antibody-complemenl-medialed cytotoxicity is enhanced by ribosomeinaclivaling proleins. J. Immunol.. 135: 3648-3651, 1985.
26. Goldmacher. V. S., Tinnel. N. L., and Nelson, B. C. Evidence that pinocytosis
in lymphoid cells has a low capacily. J. Cell Biol.. 102: 1312-1319, 1986.
27. Greenfield, L., Johnson, V. G.. and Youle. R. J. Mulalions in diphtheria
loxin separate binding from enlry and amplify immunoloxin selectivity.
Science (Washinglon DC), 238: 536-539. 1987.
28. Hokland. P., Rilz, J., Schlossman, S. F., and Nadler, L. M. Orderly expres
sion of B cell anligens during Ihe in vitro differenlialion of nonmalignanl
human pre-B cells. J. Immunol., 135: 1746-1751, 1985.
29. Freedman, A. S., and Nadler, L. M. Cell surface markers in hemalologic
malignancies. Semin. Oncol.. 14: 193-212, 1987.
30. Barbieri. L.. and Slirpe. F. Ribosome-inaclivaiing proleins from plants:
properties and possible uses. Cancer Surv., /: 489-520, 1982.
31. Jimenez, A., and Vazquez, D. Plani and fungal proteins and glycoprotein
toxins inhibiling eukaryole prolein synthesis. Annu. Rev. Microbio!., 39:
649-672, 1985.
32. Endo, Y., Tsurugi. K., and Lambert, J. M. The site of action of six different
ribosome-inactivating proteins from plants on eukaryolic ribosomes: Ihe
RNA N-glycosidase aclivily of ihe proleins. Biochem. Biophys. Res. Com
mun., 150: 1032-1036. 1988.
33. Eiklid, K., Olsnes, S., and Pihl, A. Enlry of lelhal doses of abrin, ricin and
modeccin inlo Ihe cylosol of HeLa cells. Exp. Cell Res., 126:321 -326. 1980.
34. Vitella, E. S., Cushley, W., and Uhr. J. W. Synergy of ricin A-chainconlaining immunoloxins and ricin B-chain-conlaining immunotoxins in in
vitro killing of neoplastic human B cells. Proc. Nail. Acad. Sci. USA. SO:
6332-6335, 1985.
35. Mclnlosh. D. P.. Edwards. D. C.. Cumber. A. J.. Parnell, G. D., Dean, C.
J., Ross, W. C. J., and Forrester, J. A. Ricin B-chain converts a non-cytoloxic
anlibody-ricin A-chain conjugate into a polenl and specific cylotoxic agent.
FEBS Leti., 164: 17-20, 1983.
36. Goldmacher, V. S., Anderson, J., Schullz. M-L.. Blà lller, W. A., and Lam
bert, J. M. Somatic cell mutanls resistanl to ricin, diphlheria loxin, and lo
immunotoxins. J. Biol. Chem., 262: 3205-3209, 1987.
37. Shimoda. T., and Funalsu. G. Binding of laclóseand galaclose lo nalive and
iodinaled ricin D. Agrie. Biol. Chem., 49: 2125-2130. 1985.
38. Mise, T., Shimoda, T., and Funatsu, G. Identification of lyrosyl residue
present in Ihe high-affmily saccharide-binding sile of ricin D. Agrie. Biol.
Chem., 50: 151-155, 1986.
39. Vilella, E. Synergy between immunotoxins prepared wilh nalive ricin A
chains and chemically-modified ricin B chains. J. Immunol., 136: 18801887, 1986.
40. Yamasaki. N.. Kakiuchi, S., and Funatsu, G. Significance of carboxyl group
present in the high affinity saccharide-binding site of ricin D. Agrie. Biol.
Chem., 52:2123-2124, 1988.
41. Sandvig, K.. Olsnes, S., and Pihl, A. Chemical modificalions of Ihe toxic
lectins abrin and ricin. Eur. J. Biochem., 84: 323-331. 1978.
42. Youle, R. J., Murray, G. J., and Neville, D. M., Jr. Studies on the galaclosebinding sile of ricin and Ihe hybrid loxin Man6P-ricin. Cell, 23: 551-559,
1981.
43. Moroney, S. E., D'Alarcao, L. J., Goldmacher, V. S., Lambert, J. M., and
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
Blättler,W. A. Modification of the binding site(s) of lectins by an affinity
column carrying an activated galactose-lerminaled ligand. Biochemislry, 26:
8390-8398,1987.
Thorpe, P. E.. Ross, W. C. J., Brown, A. N. F., Myers, C. D., Cumber, A.
J.. Foxwell. B. M. J., and Forresler, J. T. Blockade of Ihe galactose-binding
siles of ricin by its linkage to antibody: specific cytotoxic effecls of ihe
conjugales. Eur. J. Biochem.. 140: 63-71. 1984.
Callel, L., Delprino, L., Brusa, P., Dosio, F., Comoglio, P. M., and Pral, M.
Comparison of blocked and non-blocked ricin-antibody immunoloxins
against human gastric carcinoma and colorectal adenocarcinoma cell lines.
Cancer Immunol. Immunother., 27: 233-240, 1988.
Pietersz, G. A.. Kanellos. J., and McKenzie, 1. F. C. Novel synthesis and in
vitro characterization of disulfide-linked ricin-monoclonal antibody conju
gates devoid of galactose binding aclivily. Cancer Res., 48:4469-4476,1988.
Wawrzynczak, E. J., Walson, G. J., Cumber, A. J., Henry, R. V., Parnell,
G. D., Rieber, E. P., and Thorpe, P. E. Blocked and non-blocked ricin
immunotoxins against the CD4 antigen exhibil higher cyloloxic potency lhan
a ricin A chain immunoloxin potenlialed wilh ricin B chain or wilh a ricin
B chain immunoloxin. Cancer Immunol. Immunolher.. 32: 289-295, 1991.
Rutenber, E.. Ready. M., and Robertus, J. D. Structure and evolution of ricin
B-chain. Nalure (Lond.), 326: 624-626, 1987.
Kiesel, S., Pezzullo, A., Haas, R., Moldenhauer, G.. and Dörken,B. Funclional evaluation of CD19- and CD22-negative varianls of B-lymphoid cells.
Immunology, 64: 445-450, 1988.
Ghelie, M-A.. May, R. D., Till. M., Uhr, J. W., Ghelie, V., Knowles, P. P.,
Reif. M., Brown, A., Wallace, P. M., Janossy, G., Amlol, P., Vilella, E. S.,
and Thorpe, P. E. Evalualion of ricin A-chain-conlaining immunoloxins
direcled againsl CD 19 and CD22 antigens on normal and malignant human
B-cells as polenlial reagenls for in vivo Iherapy. Cancer Res., 48: 2610-2617,
1988.
May, R. D., Vilella, E. S., Moldenhauer, G., and Dörken,B. Seleclive killing
of normal and neoplaslic human B cells with anli-CD19- and anli-CD22ricin A-chain immunotoxins. Cancer Drug Delivery, 3: 261-272, 1986.
Preijers, F. W. M. B., Tax, W. J. M., De Wille, T.. Janssen, A., Heijden, H.
V. D.. Vidal. H., Wessels, J. M. C, and Capel, P. J. A. Relalionship belween
intcrnalization and cytoloxicily of ricin A-chain immunotoxins. Br. J. Haemalol., 70:289-294, 1988.
Uckun, F. M., Ramakrishnan. S., and Houslon, L. L. Immunoloxin-medialed
fi iininal nHIof clonogenic tumor cells in the presence of human bone marrow.
J. Immunol.. 134: 2010-2016, 1985.
Foxwell, B. M. J., Donovan. T. A., Thorpe, P. E., and Wilson, G. The
removal of carbohydrates from ricin with endoglycosidases H, F and D and
a-mannosidase. Biochim. Biophys. Acia, 840: 193-203, 1985.
Fullon, R. J., Blakey, D. C, Knowles, P. P., Uhr, J. W., Thorpe, P. E., and
Vilelta, E. S. Purification of ricin A!, A2, and B chains and characlerizalion
of their loxicily. J. Biol. Chem., 261: 5314-5319, 1986.
Laemmli, U. K. Cleavage of slruclural proleins during the assembly of the
head of bacteriophage T4. Nalure (Lond.), 227: 680-685, 1970.
6242
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
An Immunotoxin Prepared with Blocked Ricin: A Natural Plant
Toxin Adapted for Therapeutic Use
John M. Lambert, Victor S. Goldmacher, Albert R. Collinson, et al.
Cancer Res 1991;51:6236-6242.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/51/23_Part_1/6236
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.