[CANCER RESEARCH 5l, 4219-4225. August 15. 1991]
Improved Pharmacokinetics and Tumor Localization of Immunotoxins Constructed
with the Mr 30,000 Form of Ricin A Chain
Patrick W. Trown,1 Dayton T. ReunÃ-an.Stephen F. Carroll, John B. Stoudemire,2 and Russell T. Kawahata
XOMA Corporation, Berkeley, California 94710
ABSTRACT
The two naturally occurring forms of ricin A chain, M, 33,000 and M,
30,000 ( RTAu and K I A <„)
have been purified, and their chemical com
positions, toxicities, and tissue distributions have been determined. As
reported previously, the in vitro and in vivo toxicities of K I A,,, and
RTA33 are similar. However, RTAj«,which contains less carbohydrate
with a lower mannose content than UTA,,, accumulated less in the liver
than did RIA,,. Monoconjugate immunotoxins ( />.. containing one RTA
per monoclonal antibody molecule) were constructed between RTA.W or
RIA u and the antitumor monoclonal antibody 791T/36, which recognizes
a M, 72,000 antigen on osteosarcoma and colon carcinoma cells. The two
immunotoxins had similar cytoxicities in vitro but differed substantially
in their pharmacokinetics and tissue distributions in vivo in nude mice
bearing C170 human colorectal carcinoma xenografts. The immunotoxin
derived from RIA,,, (ITi„)accumulated less in the liver than the immu
notoxin derived from RTA33 (IT33) and cleared more slowly from the
blood; the a and ßhalf-lives for ITM and IT33 were 0.50 and 20.5 versus
0.17 and 14.6 h, respectively. As a probable consequence, I I'm accumu
lated to approximately 3-fold higher levels in the C170 xenografts than
I l.u. The reduced clearance of IT-,,, by the reticuloendothelial system
thus resulted in prolonged survival in the blood and enhanced tumor
localization relative to II u.
INTRODUCTION
Immunotoxins, which are conjugates of monoclonal antibod
ies and toxic moieties such as plant or bacterial toxins, are
being increasingly investigated as a means to target and kill
tumor cells or lymphocytes specifically. The most widely stud
ied toxic moiety, both experimentally and clinically, is RTA'
(reviewed in Ref. 1), probably as a consequence of its relative
ease of preparation and high potency. This glycoprotein is a
highly specific jY-glycosidase which removes adenine 4324 from
ribosomal 28S RNA, thereby inactivating the ribosome and
inhibiting protein synthesis (2, 3).
Potent RTA immunotoxins have been prepared from anti
bodies that react with a variety of human tumor-associated
antigens (4-9) as well as with several human lymphocyte surface
antigens (10-14). Several of these immunotoxins have been
evaluated clinically (4, 12, 15), but thus far, only the trials with
an anti-pan T-lymphocyte immunotoxin (16, 17) have yielded
clear evidence of efficacy.
Several reasons have been proposed for the relatively poor
Received 2/25/91; accepted 6/10/91.
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.
' To whom requests for reprints should be addressed, at XOMA Corporation,
2910 Seventh Street, Berkeley, CA 94710.
2 Present address: Genetics Institute, 87 Cambridge Park Drive, Cambridge,
M A 02140.
3 The abbreviations used are: RTA, RTB, ricin toxin A chain and ricin toxin
B chain, respectively; RTA30, RTA33, M, 30,000 and M, 33.000 forms (A, and
A2), respectively, of RTA; IT30, IT3j. monoconjugates of RTAM and RTA33,
respectively, to monoclonal antibody 791/T36; IC!0. concentration causing 50%
inhibition of protein or DNA synthesis; FCS. fetal calf serum; M3FX, mannose
al-6 (mannose «1-3)(xylose ßl-2)mannose (¡1-4A'-acetylglucosamine (Jl-4
(fucose a 1-3) A'-acetylglucosamine; PBS, phosphate-buffered saline (10 HIMNa
phosphate-O.ISM NaCl).
efficacy noted for certain preparations of immunotoxins in vivo,
including poor access to solid tumors, antigen-negative tumor
cell mutants, antigen shedding, instability of the conjugate, and
rapid clearance from circulation (18). The latter reason may be
the most important for RTA immunotoxins, except, perhaps,
when their target is immediately accessible in circulation as it
is for the anti-lymphocyte immunotoxins.
Native RTA immunotoxins are rapidly cleared from the
bloodstream of animals (19-21), and a similar rapid clearance
has been observed in humans (16, 22, 23). The rapid clearance
of RTA immunotoxins is due primarily to the recognition of
mannose and fucose residues on the RTA by cells in the liver
and other organs of the reticuloendothelial system with recep
tors for those sugars. Thus, ricin (24, 25), RTA (26, 27), and
RTA immunotoxins (28-30) are all taken up by liver cells in
vitro or by the liver in vivo via a mechanism that is blocked by
glycoproteins and saccharides containing mannose or fucose.
Two approaches have been taken to overcome this accelerated
clearance of RTA immunotoxins: (a) chemical modification of
the RTA carbohydrate, thereby minimizing reactivity with the
reticuloendothelial cell receptors (20, 21, 27); and (b) the use
of recombinant RTA produced in bacteria and, therefore, lack
ing carbohydrate (31, 32). Immunotoxins made using these
RTAs have shown reduced liver uptake in animals and pro
longed survival in the blood (21, 27, 29, 31). For immunotoxins
made with chemically modified RTA this has translated into
improved tumor localization and efficacy (33).
In this study, we have explored an alternative approach of
utilizing a naturally occurring form of RTA that has both a
lower total carbohydrate content and a reduced amount of
mannose compared with those of native RTA. Native RTA
consists of two forms with identical protein sequences but with
different apparent molecular weights of 33,000 and 30,000,
which we have termed RTA33 and RTA30, respectively (34-38).
The ratio of RTA30 to RTA33 varies from 2:1 to 3:1. These
forms have been partially (37, 38) or completely (35) separated,
and their carbohydrate compositions and structures have been
studied. From their compositions and susceptibility to endoglycosidases and to anti-mannosidase
it was concluded that
both forms of RTA contain an oligosaccharide with the com
position (GlcNac)2(Man)3_4XylFuc, but that RTA33 also con
tains a high-mannose oligosaccharide with the composition
(GlcNac)2(Man)4-6 (20, 37).
Based upon its composition, the common oligosaccharide
was assumed to have a structure similar to that reported for
pineapple stem bromelain oligosaccharide (39). This structure
does not contain the trimannosidic core recognized with high
affinity by hepatic reticuloendothelial cells (40) and known to
be essential for binding to concanavalin A (41). In contrast, it
was believed (20, 37) that the high-mannose oligosaccharide
unique to RTA33 contains the trimannosidic core and thus,
unlike RTA30, is able to bind to concanavalin A (35, 37).
On the basis of these differences, it seemed likely that RTA30
would be less avidly bound by cells in the liver with resultant
longer survival in the blood both for itself and for immunotoxins
made with it. Consequently, we developed a procedure for large-
4219
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED
RICIN A CHAIN IMMUNOTOXINS
scale purification of RTA30 and RTA33, synthesized immunotoxins using each form, and studied their pharmacokinetics as
well as tumor and tissue distributions. Our data show that the
RTAjo immunotoxin is less localized in the liver than its
counterpart and, as a probable consequence, exhibits prolonged
survival in the blood and improved localization in tumor
xenografts.
MATERIALS
AND METHODS
RTAjo and RTA33.Ricin was extracted from macerated castor beans
with PBS, pH 7.2. The extract was clarified by centrifugación, and the
fat-free supernatant was removed by aspiration. Crude ricin was precip
itated with 60% ammonium sulfate and resolubilized with PBS.
After extensive dialysis against PBS, the clarified solution was ap
plied to a column of Sepharose 4B (Pharmacia, Piscataway, NJ). Ricin
bound weakly and was eluted with PBS as a slightly retained broad
peak. The ricin was then applied to an acid-treated Sepharose 4B
column (42), which effectively binds the toxin. Ricin A chain was
reduced from the bound toxin by application of 7% mercaptoethanol
in Tris-HCl buffer, pH 7.8, and the solution was dialyzed against PBS
to remove residual reducing agent. Contaminating whole ricin toxin
and ricin B chain were removed by passing the RTA solution through
a second acid-treated Sepharose 4B column. Residual traces of RTB
were finally removed by passage through an immunobilized anti-RTB
antibody column.
Purified RTA was adsorbed to an S-Sepharose Fast Flow column
equilibrated with 20 mM Na acetate, pH 5.5. RTA30 and RTA33 were
selectively eluted as two distinct peaks with a 10-column volume linear
gradient composed of equal volumes of 20 mM sodium acetate, pH 5.5,
and 20 mM sodium phosphate-90 mM sodium chloride, pH 7.5. Purified
RTA33 and RTA30 were concentrated via diafiltration using M, 10,000
cutoff membranes, made 10 MMin 2-mercaptoethylamine, and stored
in 10% glycerol at -20°C.The methods for analyzing the carbohydrate
composition of RTA30 and RTA33 will be described elsewhere.4
Monoclonal Antibody. The 79IT/36 murine IgG2b monoclonal an
tibody (43, 48) recognizes a M, 72,000 tumor-associated glycoprotein
found on the surface of many human tumor cells (44-46). It was
produced using an Accusyst hollow fiber bioreactor (Endotronics, Min
neapolis, MN) and purified using Protein A Avidgel (Bioprobe, Tustin,
CA). The antibody was further purified by adsorption onto a QSepharose Fast Flow column (Pharmacia, Piscataway, NJ) and elution
with 10 mM Tris-HCl, 200 mM NaCl, pH 8.O.
Preparation of ITMand IT33.79IT/36 antibody (2 mg/ml) was reacted
with a 4.5-molar excess of /V-succinimidyl-3-(2-pyridyldithio)
propionate in 0.1 M sodium phosphate-0.1 M sodium chloride, pH 7.5,
for 20 min at 20°C.The reaction mixture was diafiltered with 10
volumes of 10 mM Tris-HCl, 0.15 M NaCl, pH 8.0, to remove unconjugated 7V-succinimidyl-3-(2-pyridyldithio)propionate and then mixed
with a 5-fold molar excess of freshly reduced RTA30 or RTA33. The
conjugation reaction was incubated for 4 h at 20°Cand then 18 h at
4°C.Residual 2-thiopyridyl groups were removed by the addition of 2mercaptoethylamine.
Purification of IT»and IT33. Residual, unconjugated RTA was re
moved from the immunotoxin by diluting the reaction mixture with 20
mM Tris-HCl, pH 8.0 (5 parts reaction mixture, 3 parts Tris buffer),
and loading the diluted sample onto a Q-Sepharose Fast Flow column
equilibrated in 10 mM Tris-HCl, 0.09 M NaCl, pH 8.O. Under these
conditions free RTA flowed through the column. Immunotoxin was
subsequently eluted with 10 mM Tris-0.25 M NaCl, pH 8.0. The
immunotoxin solution was adjusted to l M(NH4)2SO4 and loaded onto
a phenyl-Sepharose column. Free antibody, mono-, and poly-RTA
immunotoxins were eluted using a decreasing linear gradient consisting
of 1.0 to 0.1 M (NH4)2SO4 in 10 mM sodium 2-(7V-morpholino)
ethanesulfonate, pH 6.O. Appropriate fractions were pooled, dialyzed
against PBS, and adjusted to approximately 1 mg/ml. Analysis of the
immunotoxins was carried out using a linear 3-7.5% sodium dodecyl
* D. Burke el al., manuscript in preparation.
sulfate-polyacrylamide gel electrophoresis followed by Coomassie blue
staining and densitometry analysis.
Reticulocyte Lysate Assay. This in vivo protein synthesis assay was
performed essentially as described by Press et al. (47), except that
[14C]leucinewas used as the radiotracer and the incubation time was 80
min.
Cytotoxicity Assays. The cytotoxicity of RTA30 and RTA33 was
determined using the human T-cell leukemia cell line HSB-2 (CCL
120.1; American Type Culture Collection, Rockville, MD). Serial di
lutions of each RTA ranging from 10 Mg/ml to 3.2 ng/ml were prepared
in leucine-free RPMI 1640 containing 10% heat-inactivated PCS, 2
mM glutamine, and 0.01 mg/ml gentamicin (LI640 PCS), and aliquots
(100 M¡)
were added to microtiter wells containing IO5 HSB-2 cells in
100 n\ L1640 FCS. After incubation at 37°Cfor 20 h, [3H]leucine (1
/¿Ci/well)was added, the cells were harvested 4 h later on glass filters,
and their radioactivity was determined by cascade gas ionization using
a Trace 96 counter (Inotech Biosystems International, Lansing, MI).
The IC50was calculated as the concentration of immunotoxin necessary
to inhibit the incorporation of radioactivity by 50% relative to control
cells incubated without immunotoxin.
The cytotoxicities of IT30 and IT33 were determined using 79IT
osteosarcoma cells that express the antigen recognized by 79IT/36
monoclonal antibody (49). Aliquots of a suspension of 79IT cells (4 x
105/ml) in RPMI 1640 containing 10% heat-inactivated fetal calf
serum, 2 mM glutamine, and 0.01 mg/ml gentamicin (1640 FCS) were
added to microtiter wells containing appropriate dilutions of the IT30
and IT33 in 1640 FCS. After incubation at 37°Cfor 42-48 h, [3H]thymidine (1 ^Ci/well) was added. Following an additional 18-h incu
bation, the supernatant was removed, trypsin-EDTA was added, and
the cells were subsequently harvested onto glass filters. The cell asso
ciated radioactivity was determined by scintillation counting. The IC50
was calculated as described above.
lodination. RTA3o, RTA33, IT30, and IT33 were radiolabeled using
lodo-Gen (l,3,4,6-tetrachloro-3a,6a-diphenylglycouril;
Sigma, St.
Louis, MO) (50). The lodo-Gen was dissolved in dichloromethane (1
mg/ml), and the volume containing 1 mg lodo-Gen/10 mg of protein
to be labeled was added to a glass vial. After evaporation of the
dichloromethane with a stream of nitrogen, a solution of RTA30, RTA33,
ITjo, or IT33 in PBS, pH 7.5, was added. Radioactive [131I]- or
[I25l]sodium iodide (New England Nuclear, Boston, MA) was then
added to achieve a level of 1 ¿¿Ci/Mg
of protein. The solutions were
incubated for 30 min at room temperature with occasional swirling.
Unreacted radiolabel was removed using a 5-ml centrifuge column
prepared with Sephadex G50 equilibrated in PBS. RTA30 and RTA33
were both labeled with 125I;in each case 99% of the radioactivity was
precipitable with trichloroacetic acid. The specific activities of RTA30
and RTA33 were 0.62 and 1.48 nCi/ßg,respectively. IT30 was labeled
with I25I,and 94% of the radioactivity was precipitable with trichloro
acetic acid. The IT33was labeled with '•"!,
and 98% of the radioactivity
was precipitable with trichloroacetic acid. The protein concentration
for each immunotoxin was determined by the bicinchoninic acid protein
assay (Pierce Biochemicals, Rockford, IL). The specific activities of the
IT3o and the IT33 were determined to be 0.57 and 0.95 ^Ci/Mg,
respectively.
Animal Studies. Two groups of 12 BALB/c mice (20-25 g) were used
for the RTA biodistribution studies. Each animal received a single i.v.
dose of '"I-labeled RTA30 or RTA33 (2.5 MCiin 150 MlPBS). At 3, 15,
45, and 90 min after injection, groups of 3 mice were bled to obtain a
serum sample and then sacrificed. The kidneys and livers were removed
and weighed, and their radioactivity was determined. These data were
used to calculate the percentage of injected dose in each organ and in
the total serum volume (assumed to be 0.045 ml/g body weight).
Female BALB/c (nu/nu) mice weighing 20-25 g were used for the
IT biodistribution studies. Each animal received a s.c. implant of 1 x
IO7cells of the C170 human colorectal carcinoma cell line in 100 ti\ of
culture media in the right flank. Three weeks after implantation, the
xenografts were well established (~200 mg), and the study was initiated.
The mice were divided into 10 groups of 3 animals each. 125I-IT30and
mI-IT33 were mixed in PBS with 0.1 % polysorbate 80 and administered
4220
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED
RICIN A CHAIN IMMUNOTOXINS
to each animal in a single i.v. dose (150 ^1 containing 7.2 ¿ig125I-IT30
and 7.2 j/g 13'I-IT33)via the tail vein. Groups of 3 mice were sacrificed
Table I Properties of RTAM and RTA,,
at 3, 15, 30, 45, 90, 180, 360, 960, 1440, and 2880 min, and the
biodistribution of each isotope was determined. At the times indicated,
the animals were weighed and anesthetized with Metofane (PitmanMoore, Inc., Washington Crossing, NJ), and a blood sample was taken
by cardiac puncture. The animals were then perfused, via the left
ventricle of the heart, with cold PBS containing heparin (1 unit/ml) to
remove most of the circulating blood pool from the tissues. Tumor,
spleen, kidneys, liver, and one hindquarter were removed and weighed.
These tissues or portions thereof, as well as an aliquot (100 ¿il)of
serum, were counted for 1251and 13'I using a Compugamma counter
(1282; LKB, Uppsala, Sweden) and set for dual isotope counting,
automatic decay, background, and spillover correction for both iso
topes. Averages (±SD)of the data from three mice for each group were
determined and expressed as the percentage of injected dose/g tissue.
The serum volume in each mouse was assumed to be 0.045 ml/g body
weight. Trichloroacetic acid precipitation studies demonstrated that
>90% of the radioactivity at any time point was protein associated.
Pharmacokinetics parameters were calculated from the serum radio
activity data using a two-compartmental analysis (Rstrip; MicroMath
Scientific Software, Salt Lake City, UT). A two-way analysis of variance
was also performed on the tumor localization data (SuperANOVA;
Abacus Concepts, Berkeley, CA).
RESULTS AND DISCUSSION
RTAM and RTA33. As shown in Fig. I, ion-exchange chromatography of RTA under the conditions described herein
results in the isolation of essentially homogeneous preparations
of its two components, RTA.,0 and RTA33. The properties of
those glycoproteins, summarized in Table 1, are similar to those
reported previously for partially or fully purified RTA, and
RTA2, respectively (35, 37). The carbohydrate content of RTA33
is more than double that of RTA30. The compositions of
the carbohydrates are consistent with the suggestion (20, 37)
that each form possesses a common oligosaccharide of compo
sition (GlcNac)2(Man)3XylFuc, whereas RTA3, also possesses
an additional
high-mannose
oligosaccharide.
From our
data, the composition of this oligosaccharide would be
(GlcNac)2(Man)3.5Xyl.
Recently, the structure of the common RTA oligosaccharide
was reported to be M3FX, which is linked to Asn-10 (51, 52).
However, the additional oligosaccharide at Asn-236 in a Mr
weightProtein
Molecular
(pM)Cellsynthesis ICM
(nin)LD50
cytotoxicity
(mg/kg)"Carbohydrate
w/w)Carbohydrate
content (%
composition(moles
RTA)GlcNAcManXylFucRTAjo30,0005.414.417.03.42.1
sugar/mole
(±0.21)*2.9
0.37)1.0
(±
(±0.26)0.9
(±0.18)RTAJ333,0005.925.922.57.83.9e5.61.91
" LD5o, single i.v. dose causing 50% lethality in mice.
* Mean ±SD (n = 4).
' Mean of two determinations.
32,000 form of RTA was also reported to be M3FX (52). These
results are surprising because M3FX contains the trimannosidic
core known to be required for binding to concanavalin A (41).
Thus, both RTA30 and RTA33 would be expected to bind to
concanavalin A, whereas only RTA33 does so (Refs. 35, 37; data
not shown). Studies in our laboratories4 have confirmed the
presence of M3FX on peptides containing Asn-10 and Asn236. However, additional oligosaccharides with higher mannose
content were detected in the Asn-236 peptide derived from
RTA33. These oligosaccharides could be the cause of the binding
of RTA33 to concanavalin A (Ref. 37; data not shown).
As reported previously (35), the activities of the two forms
of RTA in inhibiting the growth of cell lines and cell-free protein
synthesis were very similar. However, in contrast to previous
reports, the single i.v. dose (mg/kg) of RTA30 causing 50%
lethality in mice was lower than that of RTA33. This difference
is probably a consequence of the very different initial tissue
distributions of the two forms of RTA in mice (Fig. 2). As
predicted from its high mannose content, RTA33 was rapidly
taken up by the liver so that 52% of the radioactivity in the
injected dose of 125I-labeled protein was located therein 3 min
after injection. In contrast, only 20% of the injected radioactive
dose of RTA3o was present in the liver at 3 min, with 48%
present in the serum at that time. The lower uptake of RTA30
by the liver, possibly mediated by residual sugars (54), is accom
panied by a higher uptake by the kidney and (most likely)
excretion in the urine.
3-7.5% Gel
12.5% Gel
RTA
RTA30 RTA33
Fig. 1. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of RTAs and
ITs. Samples of RTA, RTA30. and RTA33 were
analyzed on a 12% gel. Samples of the monoclo
nal antibody 79IT/36 (Ab), the unfractionated
reaction mixture (Rxn) in the preparation of
IT30,and the purified ITMand 1T33were analyzed
on a 3-7.5% gradient gel. The gels were stained
with Coomassie blue.
Ab
Rxn
ITso
--3 RTA/Ab
-2 RTA/Ab
~1RTA/Ab
~Ab
4221
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED RICIN A CHAIN 1MMUNOTOX1NS
Biodistribution
of 125IRTA30
shown in Fig. 3. Under the conditions employed, free antibody
is eluted first, followed by peaks composed of monoconjugate,
then disconjugate, and lastly >triconjugate. The purified IT30
and IT33 monoconjugates contained <3% free RTA and <4%
free antibody. They had similar cytotoxicities against 791T cells
in vitro; the IC5oSfor IT30 and IT33 were 19.4 and 18.7 ng/ml,
respectively.
IT30 and IT33 were labeled with 125Iand I31I, respectively,
o
o
mixed, and administered i.v. to nude mice bearing s.c. C170
tumors. The time courses of the distributions of 125Iand I31I
I
15
45
Time (minutes)
Biodistribution
of 125IRTA33
50
Time (minutes)
Fig. 2. Biodistribution of radiolabeled RTA30 and RTA33. '"[-labeled RTAW
or RTA33 was injected i.v. into male BALB/c mice. Groups of 3 mice were
sacrificed at the specified time intervals; samples of serum, liver, and kidney were
obtained; and their radioactivity was determined as described in "Materials and
Methods." These data were used to calculate the percentage of the injected dose
in each tissue/organ (mean ±SD).
radioactivities to serum, tumor, liver, kidney, hindquarter, and
spleen are expressed as a percentage of injected dose/g in Fig.
4. As expected from the results obtained with RTA30 and
RTA33, the serum pharmacokinetics and the tissue distributions
of the radioactivity from I-labeled IT30 and IT33 differed greatly
from one another. The serum radioactivity data were fitted to
a two-compartment pharmacokinetic model which yielded the
parameters shown in Fig. 4. Whereas the calculated ßphase
half-lives for the two ITs were similar, the a phase half-life was
nearly 3-fold longer for IT30 than for IT33, resulting in a 3-fold
increase in the area under the serum concentration versus time
curve at infinite time after injection.
At time intervals up to 6 h, the major localization of IT30
was in the serum with lesser amounts in the liver and spleen.
In contrast, the highest localization of IT33was in the liver with
lesser amounts in the serum and spleen. Thus, IT33 localized
mainly in the reticuloendothelial system organs, whereas IT30
remained longer in the serum. Of most interest was the differ
ence between localization of the two immunotoxins in the
tumors. IT30 rapidly accumulated in the tumors within the first
hour and continued to be accumulated therein for the remainder
of the study, whereas IT33 localization remained relatively con
stant. Thus, the extent of IT30 localization in tumors was
significantly greater (P = 0.0001) than for IT33.
The differential accumulation of IT30 and IT33 in tissues was
clearly evident from the tissue:serum ratios (Fig. 5). Interest
ingly, the tumonserum ratios were similar for the two immu
notoxins, suggesting that the mechanism by which they were
localized into the tumor from the blood was the same. Presum-
0.18-
IT.TOand IT33. The strikingly lower uptake of RTA30 than
RTAjj by the liver suggested that immunotoxins constructed
with RTAjo would also be less subject to hepatic clearance, with
resulting longer plasma half-lives. Consequently, immunotox
ins were constructed between RTA30 or RTA33 and the antitumor monoclonal antibody 791T/36 (42). The conjugation prod
ucts were fractionated to yield monoconjugates (IT30 and IT33,
respectively), i.e., one RTA per antibody molecule, with mini
mal contamination of RTA, unconjugated antibody, or di-,
tri-, and polyconjugates (Fig. 1). This was done in order to
simplify the interpretation of serum pharmacokinetics and tis
sue distributions of 125I-or '-"I-labeled immunotoxins, since
unconjugated monoclonal antibodies and free RTA are cleared
from the plasma much more slowly and much more quickly,
respectively, than RTA immunotoxins (28). In addition, immunotoxin polyconjugates have been reported to be cleared
from the serum of mice more rapidly than monoconjugates
(53).
Following the removal of unreacted RTA by ion-exchange
chromatography, the monoconjugate species of IT30 and IT33
were purified by Hydrophobie interaction chromatography on
phenyl-Sepharose. A representative chromatogram for IT30 is
|
I
0.08
-0.02
20
40
60
80
Fraction
Fig. 3. Fractionation of IT30 on phenyl-Sepharose. The IT30 conjugation
reaction mixture was first applied to a column of Q-Sepharose to remove un
reacted RTA».The eluted mixture of antibody and IT was then adjusted to l M
(NH4)2SO4 and loaded onto a column of phenyl-Sepharose. Following the appli
cation of a decreasing linear gradient (1.0 to 0.1 M) (NH4)2SO4, 8.5-ml fractions
were collected. Analysis of column fractions by nonreducing sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (5% gels) indicated that samples eluted
in the following order: free antibody (fractions 23-34); ITM monoconjugate
(fractions 43-51); diconjugate (fractions 54-61); triconjugate (fractions 63-68);
and > tetraconjugate (fractions 69-79).
4222
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED RICIN A CHAIN IMMUNOTOXINS
4
a
12
16
Tim«after Injection (hourt)
HI
20
z«
24
0
4
8
12
16
Tim< aller injection (houri)
24
Fig. 4. Biodistribution of radiolabeled ITM and ITJ3 in BALB/c nude (nu/nu) mice bearing C170 human coloréela!carcinoma xenografts. Tumor-bearing mice
received a single i.v. dose of a mixture of 131I-ITMand '"I-IT,,. At the specified intervals, groups of 3 mice were anesthetized, bled, and perfused with PBS. Tumor,
spleen, kidney, liver, and one hindquarter were removed, weighed, and counted for iodine radioactivity as described in "Materials and Methods." Averages of the data
(±SD)from three mice were determined and expressed as percentages of injected dose/g tissue (or ml blood).
ably this occurred in an antigen-dependent fashion via the
monoclonal antibody 79IT/36 common to each immunotoxin.
In contrast, IT.,., was localized from the blood into the liver and
spleen far more than IT30, reflecting interaction of the highmannose carbohydrate of RTA33 with mannose receptors on
cells of the reticuloendothelial system (39).
Summarizing our results, RTA30, the low-carbohydrate form
of natural RTA, accumulated less in the liver than the highmannose form RTA.u. Immunotoxins constructed with each
form of RTA had similar activities in vitro but differed dramat-
4223
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED RICIN A CHAIN IMMUNOTOXINS
H IT30
• II '
Ü 0.4
005
0.25
0.5
0.79
1.5
TIME AFTER
Fig. 5. Tumonserum ratios of radiolabeled
IT» and IT)3 in C170 tumor-bearing mice. Data,
obtained as described in Fig. 4. are expressed as
the tissue:serum radioactivity ratio derived from
the percentage of injected dose/g tissue (or ml
blood).
003
0.23
0.3
0.73
1.5
TIME AFTER
3
In conclusion, although differential stabilities of IT30 and
IT33 must also be considered, our results are fully consistent
with the notion that the improved pharmacokinetics and tumor
localization of IT30 result directly from the in vivo characteris
tics of the RTA3o moiety. Studies to determine whether this
enhanced tumor localization in turn results in higher antitumor
activity are in progress.
ACKNOWLEDGMENTS
The authors would like to thank David Burke for the carbohydrate
analyses and Henry Lopez and Charlotte Chang for their excellent
technical assistance.
REFERENCES
1. Blake). D. C., and Thorpe. P. E. An overview of therapy with immunotoxins
containing ricin or its A chain. Antibody. Immunoconj. Radiopharm. /: I
15. 1988.
0.23
(HOURS)
6
INJECTION
ically in vivo. IT30 reflected the tissue distribution properties of
RTAjo and, importantly, was cleared far more slowly from the
blood than the corresponding IT3i. This resulted in higher
accumulation of IT30 in tumor xenografts. Since native RTA is
a mixture composed of both RTA30 and RTA33, we would expect
such ITs to exhibit properties intermediate between those of
IT30 and IT33. Indeed, we and others (54) have demonstrated
that IT30s exhibit prolonged pharmacokinetics relative to the
corresponding native RTA constructs.5
! P. \V. Trown et al., unpublished observations.
0.03
INJECTION
16
(HOURS)
0.3
0.73
TIME AFTER
24
0.05
025
0.5
0.75
TIME AFTER
1.5
3
6
16
INJECTION
(HOURS)
1.5
6
3
INJECTION
16
24
24
(HOURS)
Endo, Y., Mitsui. K., Motizuki. M.. and Tsurugi, K. The mechanism of
action of ricin and related toxic lectins on eukaryotic ribosomes. The site
and the characteristics of the modification in 28S ribosomal RNA caused by
the toxins. J. Biol. Chem., 262: 5908-5912. 1987.
Endo. Y.. and Tsurugi. K. RNA iV-glycosidase activity of ricin A chain.
Mechanism of action of the toxic lectin ricin on eukaryotic ribosomes. J.
Biol. Chem.. 262: 8128-8130. 1987.
Spitler. L. E., del Rio, M.. Khentigan, A., Wedel, N. L., Brophy, N. A.,
Miller, L. L.. Harkonen. W. S.. Rosendorf, L. L., Lee, H. M., Mischak, R.
P.. Kawahata, R. T„Stoudemire. J. B., Fradkin, L. B., Bautista, E. E., and
Scannon, P. J. Therapy of patients with malignant melanoma using a
monoclonal antimelanoma antibody-ricin
A chain immunotoxin.
Cancer
Res., 47: 1717-1723. 1987.
Embleton. M. J.. Byers, V. S.. Lee. H. M.. Scannon. P. J.. Blackball. N. W.,
and lialcl" in. R. W. Sensitivity and selectivity of ricin toxin A chain mono
clonal antibody 791T/36 conjugates against human tumor cell lines. Cancer
Res.. 46: 5524-5528. 1986.
Björn.M. J.. Ring. D.. and Frankel. A. Evaluation of monoclonal antibodies
for the development of breast cancer immunotoxins. Cancer Res., 45: 12141221, 1985.
Pirker, R.. Fitzgerald. J. P.. Hamilton. R. C.. Ozols, R. F., Laird, W.,
Frankel, A. E., Willingham.
M. C., and Pastan, I. Characterization
of
immunotoxins active against ovarian cancer cell lines. J. Clin. Invest., 76:
1261-1267, 1985.
Canevari. S., Orlandi. R., Ripamonti. M.. Tagliabue. E.. Aguanno. S., Miotti,
S., Menard, S.. and Colnaghi, M. 1. Ricin A chain conjugated with monoclo
nal antibodies selectively killing human carcinoma cells in vitro. J. Nati.
Cancer Inst., 75:831-839. 1985.
Griffin. T. W.. Richardson. C.. Houston. L. L.. LePage. D., Bogden. A., and
Raso. V. Antitumor activity of intraperitoneal immunotoxins in a nude mouse
model of human malignant mesothelioma. Cancer Res.. 47: 4266-4270,
1987.
10. Kernan. N. A.. Knowles, R. W., Burns, M. J., Broxmeyer, H. E., Lu, L., Lee,
H. M., Kawahata, R. T., Scannon, P. J.. and Dupont, B. Specific inhibition
of in vitro lymphocyte transformation by an anti-pan T cell (gp67) ricin A
chain immunotoxin. J. Immunol.. 133: 137-146. 1984.
Martin. P. J.. Hansen. J. A., and Vitella, E. S. A ricin A chain-containing
immunotoxin that kills human T lymphocytes in vitro. Blood. 66: 908-912,
1985.
12, Laurent, G.. Paris, J., Farcet. J. P.. Carayon, P., Blythman, H.. Casellas, P.,
4224
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
IMPROVED
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
RICIN A CHAIN IMMUNOTOXINS
Poncelet, P., and Jansen, F. K. Effects of therapy with TlOl-ricin A chain
immunotoxin in two leukemia patients. Blood, 67: 1680-1687, 1986.
Katz, F. E., Janossy, G.. Cumber, A., Ross, W.. Blacklock, H. A., Tax, W.,
and Thorpe, P. E. Elimination of T cells from human peripheral blood and
bone marrow using a cocktail of three anti-T cell immunotoxins. Br. J.
Haematol., 67:407-411, 1987.
Press, O., Vitella, E., Farr, A., Hansen, J., and Mariin, P. Evaluation of
ricin A chain immunotoxins directed against human T cells. Cell. Immunol..
102: 10-20, 1986.
Byers, V. S., Rodvien, R., Grant, K., Durrani, L. G., Hudson, K. H., Baldwin,
R. W., and Scannon, P. J. Phase I study of monoclonal antibody-ricin A
chain immunoloxin XomaZyme-791 in palienls with metastalic colon cancer.
Cancer Res., 49: 6153-6160. 1989.
Byers, V. S., Henslee, P. J., Kernan, N. A., Blazar, B. R., Gingrich, R.,
Phillips, G. L., LeMaislre, C. F., Gilliland, G.. Anlin, J. H., Marlin, P.,
Tulscha. P. J., Trown, P. W., Ackerman, S. K., O'Reilly, R. J., and Scannon,
P. J. Use of anli-pan T-lymphocyle ricin A chain immunoloxin in sleroidresistant acule grafl-vmui-hosl disease. Blood. 75: 1426-1432, 1990.
Kernan, N. A., Byers, V. S., Scannon, P. J., Mischak, R. P., Brochslein, J.,
Flomenberg, N., Duponl, B., and O'Reilly, R. J. Trealmenl of steroidresistant acule grafl-vmws-hosl disease by in vivo adminislralion of anli-T
cell ricin A chain immunoloxin. J. Am. Med. Assoc., 259: 3154-3157, 1987.
Blakey, D. C., Wawrzynczak, E. J., Wallace, P. M., and Thorpe, P. E.
Anlibody loxin conjugales: a perspeclive. Prog. Allergy, 45: 50-90, 1988.
Bourrie, B. J. P., Casellas, P., and Blylhman, H. E. Sludy of Ihe plasma
clearance of anlibody-ricin A chain immunoloxins. Eur. J. Biochem., 755:
1-10, 1986.
Blakey, D. C., Watson, G. J., Knowles, P. P., and Thorpe, P. E. Effecl of
chemical deglycosylalion of ricin A chain on Ihe in vivo fale and cyloloxic
aclivily of an immunoloxin composed of ricin A chain and anli-Thy 1.1
antibody. Cancer Res., 47: 947-952, 1987.
Fullon, R. J., Tucker, T. F., Viletla, E. S., and Uhr, J. W. Pharmacokinetics
of lumor-reaclive immunoloxins in lumor-bearing mice: effect of anlibody
valency and deglycosylalion of the ricin A chain on clearance and lumor
localizalion. Cancer Res., 48: 2618-2625, 1988.
Mia/adi. M., Wheeler, R., Haynes, A., Mischak, R., Spiller, L. E., and
LoBuglio, A. F. The effecl of immune response lo immunoloxin on alleralion
of Ihe pharmacokinelics of murine monoclonal ami-melanoma anlibodyricin A chain (Abslracl). In: International Symposium on Immunoloxins. p.
56. Durham, NC, June 1988.
Herller, A. A., Schlossman, D. M., Borowilz, M. J., Laureili. G., Jansen, F.
K., Schmidl, C., and Frankel, A. E. A phase I study of TlOl-ricin A chain
immunoloxin in refraclory chronic lymphocylic leukemia. J. Biol. Response
Modif., 7:97-113, 1988.
Thorpe. P. E., Delre, S. !.. Foxwell, B. M. J., Brown, A. N. F., Skilleter, D.
N., Wilson, G., Forresler, J. A., and Slirpe. F. Modificalion of Ihe carbohydrale in ricin wilh melaperiodale-cyanoborohydride
mixtures. Effecls on
loxicily and in vivo distribution. Eur. J. Biochem.. 147: 197-206, 1985.
Skilleter, D. N., Price, R. J., and Thorpe. P. E. Modification of Ihe carbohydrale in ricin wilh melaperiodale and cyanoborohydride mixlures: effecl
on binding, uplake and toxicily lo parenchyma! and non-parenchymal cells
of ral liver. Biochim. Biophys. Acta, 842: 12-21, 1985.
Skilleler, D. N., and Foxwell, B. M. J. Seleclive uplake of ricin A chain by
hepalic non-parenchymal cells in vivo. Importance of mannose oligosaccharides in Ihe loxin. FEBS Leti., 196: 344-348. 1986.
Blakey, D. C., and Thorpe, P. E. Effecl of chemical deglycosylalion on Ihe
in vivo fate of ricin A chain. Cancer Drug Delivery, 3: 189-196, 1986.
Bourrie, B. J. P., Casellas, P., Blylhman, H. E., and Jansen, F. K. Study of
Ihe plasma clearance of anlibody-ricin A chain immunoloxins. Evidence for
specific recognilion siles on the A chain lhal mediate rapid clearance of the
immunoloxin. Eur. J. Biochem., 155: 1-10, 1986.
Worrell, N. R.. Skilleler. D. N., Cumber, A. J., and Price, R. J. Mannose
receptor dependent uplake of a ricin A chain anlibody conjugale by rat liver
non-parenchymal cells. Biochem. Biophys. Res. Commun.. 137: 892-896.
1986.
Blakey, D. C., Skilleter. D. N., Price, R. J., and Thorpe, P. E. Uplake of
nalive and deglycosylaled ricin A chain immunoloxins by mouse liver parenchymal and non-parenchymal cells in vitro and in vivo. Biochim. Biophys.
Acia, 968: 172-178, 1988.
Masui, H., Kamralh, H., Apell, G., Houslon, L. L., and Mendelsohn, J.
Cyloloxicily againsl human lumor cells medialed by the conjugale of amiepidermal growlh factor receptor monoclonal antibody to recombinant ricin
A chain. Cancer Res., 49: 3482-3488, 1989.
Weiner, L. M., O'Dwyer, J.. Kitson, J., Comis, R. L., Frankel, A. E.. Bauer,
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
R. J., Konrad, M. S., and Groves, E. S. Phase I evaluation of an anti-breasl
carcinoma monoclonal antibody 260F9 recombinant ricin A chain inumino
conjugale. Cancer Res., 49: 4062-4067, 1989.
Fulton. R. J., Uhr, J. W., and Vilella, E. S. In vivo iherapy of Ihe BCL,
lumor: effecl of immunoloxin valency and deglycosylation of ricin A chain.
Cancer Res., 48: 2626-2631, 1988.
Olsnes, S., and Pihl. A. Differenl biological properlies of the two constiluenl
peplide chains of ricin, a loxic prolein inhibiling prolein synlhesis. Biochemislry, 12: 3121-3126, 1973.
Vidal, H., Casellas, P., Gros, P., and Jansen, F. K. Sludies on components
of immunoloxins: purificalion of ricin and ils subunils and influence of
unreacled anybodies. Int. J. Cancer, 36: 705-711, 1985.
Ramakrishnan, S.. Eagle. M. R., and Houslon, L. L. Radioimmunoassay of
ricin and A and B chains applied lo samples of ricin A chain prepared by
chromalofocusing and by DEAE Bio-gel A chromalography. Biochim. Bio
phys. Acia, 719: 341-348, 1982.
Foxwell, B. M. J., Donovan, T. A., Thorpe, P. E., and Wilson, G. The
removal of carbohydrales 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
Vilella, E. S. Purificalion of ricin Ai, A2. and B chains and characterization
of Iheir loxicily. J. Biol. Chem., 261: 5314-5319, 1986.
Ishihara, H., Takahashi, N., Oguri, S., and Tejima, S. Complele slructure of
Ihe carbohydrale moiety of slem bromelain. An application of Ihe almond
glycopeptidase for slruclural studies of glycopeplides. J. Biol. Chem., 254:
10715-10719,1979.
Maynard, Y., and Baenziger, J. U. Oligosaccharide specific endocylosis by
isolaled ral hepalic reliculoendolhelial cells. J. Biol. Chem., 256:8063-8068,
1981.
Baenziger, J. U., and Fiele, D. Slruclural delerminanls of concanavalin A
specificily for oligosaccharides. J. Biol. Chem., 254: 2400-2407, 1979.
Schluler, S. F., and Ey, P. L. The inleraction of lactose-specific leclins wilh
acid-lrealed and laclose-subsliluled Sepharose. J. Immunol. Melhods, 66:
89-97, 1984.
Emblelon, M. J., Gunn, B., Byers, V. S., and Baldwin, R. W. Amilumor
reactions of monoclonal antibody againsl a human osleogenic sarcoma cell
line. Br. J. Cancer, 43: 582-587, 1981.
Price, R. J., Campbell, D. G., and Baldwin, R. W. Idenlificalion of an antihuman osteogenic sarcoma monoclonal-antibody-defined anligen on milogen-slimulaled peripheral blood mononuclear cells. Scand. J. Immunol., IN:
411-420, 1983.
Price, R. J., Campbell, D. G., Robins, R. A., Blecher, T. E., and Baldwin, R.
W. Characterislics of Ihe cell surface anligen, p72, associaled with a variely
of human lumors and milogen-slimulaled T-lymphoblasts. FEBS Leti., 17:
31-35, 1984.
Campbell, D. G., Price, M. R.. and Baldwin, R. W. Analysis of a human
osleogenic sarcoma anligen and ils expression of various human lumor cell
lines. Int. J. Cancer, 34: 31-37. 1984.
Press, O. W., Vilelta, E. S., and Marlin, P. J. A simplified microassay for
inhibition of protein synlhesis in reliculocyle lysales by immunoloxins.
Immunol. Lell., 74:37-41, 1986.
Robins, R. A., Laxlon, R. R., Garnell, M., Price, M. R., and Baldwin, R. W.
Measuremenl of lumor reaclive anlibody and anlibody conjugale by compe
tition, quantilated by flow cytofluorimetry. J. Immunol. Melhods, 90: 165172, 1986.
Smilh, H. S., Owens, R. B.. Militer, A. J., Nelson-Rees, W. A., and Johnston,
J. O. Biology of human cells in tissue culture. I. Characlerizalion of cells
derived from osleogenic sarcoma. Int. J. Cancer, 17: 219-234, 1976.
Fraker, P. J., and Speck, J. C. Prolein and cell membrane iodinalions with a
sparingly soluble chloroamide, l,3,4,6-lelrachloro-3a,6a-diphenylglycouril.
Biochem. Biophys. Res. Commun., SO: 849-857, 1978.
Kimura. Y., Hase, S., Kobayashi, Y., Kyogoku, Y., Ikenaka, T., and Funalsu,
G. Slruclures of sugar chains of ricin D. J. Biochem., 103: 944-949, 1988.
Kimura, Y., Kusuoku, H., Tada, M., Takagi, S., and Funalsu, G. Slruclural
analyses of sugar chains from ricin A-chain varianl. Agrie. Biol. Chem., 54:
157-162, 1990.
Sloudemire, J. B., Kawahata, R. T., Carroll, S. F., Reardan, D. T., and
Trown, P. W. Improved tumor localizalion of immunoloxin conslrucled wilh
Ihe 30 kD form of ricin A chain. In: Third International Conference on
Monoclonal Anlibody Immunoconjugales for Cancer, p. 55. San Diego, CA,
February 1988.
Wawrzynczak, E. J.. Cumber, A. J., Henry, R. V., and Parnell, G. D.
Comparalive biochemical, cytoloxic and pharmacokinelic properties of im
munoloxins made wilh nalive ricin A chain, ricin A, chain and recombinanl
ricin A chain. Int. J. Cancer, 47: 130-135. 1991.
4225
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.
Improved Pharmacokinetics and Tumor Localization of
Immunotoxins Constructed with the Mr 30,000 Form of Ricin A
Chain
Patrick W. Trown, Dayton T. Reardan, Stephen F. Carroll, et al.
Cancer Res 1991;51:4219-4225.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/51/16/4219
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 pubs@aacr.org.
To request permission to re-use all or part of this article, contact the AACR Publications
Department at permissions@aacr.org.
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1991 American Association for Cancer Research.