`
`IMMUNOGEN 2045, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`IMMUNOGEN 2045, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Papers
`
`A specific and potent immunotoxin composed
`of antibody ZME-018 and the plant
`toxin gelonin
`
`Michael G. Rosenblum, James L. Murray,* Lawrence Cheung,* Robert Rifkin, t
`Sydney Salmon, t and Richard Bartholomew+
`*Department of Clinical Immunology and Biological Therapy, M.D. Anderson Cancer Center,
`Houston, TX, USA; tDepartment of Hematology and Oncology, Arizona Cancer Center, Tucson, AZ,
`USA; and :f:HYBRITECH, Inc., Lalolla, CA, USA.
`
`Murine monoclonal antibody ZME-OI8 recognizes a 240 Kda glycoprotein present on the surface of
`most human melanoma cells and on over 80% of human biopsy specimens tested. Gelonin is a ribo(cid:173)
`some-inactivating plant toxin similar in nature and rivaling the activity of ricin A chain. ZME-OI8 was
`coupled to purified gelonin using the reagents SPDP and 2-iminothiolane. The ZME-gelonin conjugate
`was purified by S-300 Sephacryl and Blue Sepharose chromatography, removing unreacted gelonin
`and antibody, respectively. PAGE analysis showed that ZME was coupled to I, 2, or 3 gelonin mole(cid:173)
`cules. The ZME-gelonin conjugate was IfJ6-fold more active than gelonin itself in inhibiting the growth
`of log-phase human melanoma cells in culture. The immunoconjugate was not cytotoxic to antigen
`negative T-24 (human bladder carcinoma) cells. Treatment of melanoma cells with recombinant IFN-a.
`or TNF substantially augmented the cytotoxicity of the immunoconjugate while treatment with IFN--y
`had a minor effect. Using the human tumor colony assay of melanoma cells obtained from fresh
`biopsy specimens, >90% growth suppression was observed in 2 of 4 samples tested at a concentration
`of 250 nglml. In addition, 25% growth suppression was observed with a third sample tested, and no
`growth suppression was observed in I sample. Thus, clonogenic melanoma cells are sensitive in vitro
`to the cytotoxic activity of this immunotoxin at concentrations which we presume are pharmacologi(cid:173)
`cally relevant.
`
`Keywords: Human melanoma; immunotoxins; gelonin; human tumor colony assay; cytokines.
`
`Introduction
`Since the introduction of monoclonal antibody tech(cid:173)
`nology, numerous efforts have been made to exploit
`the specificity of these reagents for cancer therapy. 1- 5
`As a first step, radiolabeled monoclonal antibodies to
`tumor cell surface antigens have been utilized suc(cid:173)
`cessfully to image tumors in patients by external scin(cid:173)
`tigraphy.6·7 Extensive studies with antibodies to mela(cid:173)
`noma antigens8- 10 and to CEA11 •12 among others 13•14
`have demonstrated specific tumor localization in man
`after systemic and intraperitoneal administration. Un(cid:173)
`fortunately, the accumulation of antibody by normal
`organs (i.e., non-tumor) remains a key problem.
`Because of their unique ability to localize within
`human tumors after systemic administration, antibod(cid:173)
`ies have the potential to serve also as targeting vehi-
`
`This research was conducted, in part, by the Clayton Foundation
`for Research.
`Address reprint requests to Dr. Rosenblum at the Department of
`Clinical Immunology and Biological Therapy, M.D. Anderson
`Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA.
`Accepted for publication October 1990.
`
`cles for specific delivery of cytotoxic chemothera(cid:173)
`peutic agents, toxic peptides, biological response
`modifiers, and radionuclides. Antibody ZME-018
`(subclass IgG2A) is reactive with epitope "a" of a
`240,000 molecular weight antigen (gp240) found on the
`surface of over 80% of melanoma cell lines and fresh
`tumor samplesY Labeled with (1 11In) and adminis(cid:173)
`tered systematically to patients with melanoma, anti(cid:173)
`body ZME-018 was found to localize in 77% of soft
`tissue melanoma nodules. 16
`Several investigators have been interested in utiliz(cid:173)
`ing monoclonal antibodies as carriers of extremely ac(cid:173)
`tive protein toxins. 17- 19 In the case of these immuno(cid:173)
`toxins, the antibody serves as a vehicle for delivery of
`the toxin to the tumor. The binding of the antibody to
`the cell-surface target and pinocytosis or internaliza(cid:173)
`tion of the antibody also serves as the mechanism for
`specific intracellular entry of the toxin. 20 The first
`clinical trials with immunotoxins utilized an antibody
`against colorectal carcinoma cells and ricin A chain
`(RTA) or diphtheria toxin A chain. 21·22 Since then, a
`variety of immunotoxins have been developed utiliz(cid:173)
`ing toxins such as abrin, ricin, RTA, diphtheria toxin,
`gelonin, and pseudomonas exotoxin. 23- 26
`
`6 © 1991 Butterworth- Heinemann
`
`Mol. Biother., 1991, vol. 3, March
`
`IMMUNOGEN 2045, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`ZME-gelonin conjugate as an immunotherapeutic agent: Rosenblum eta!.
`
`The plant toxin gelonin is extracted from the seeds
`of the Gelonium multiflorum plant and exists in nature
`as a single polypeptide chain with a molecular weight
`of 29 KDa. 27 While the A chain of ricin (RTA) has
`been popular for use in immunotoxins, 28 •29 gelonin,
`one of the 60 S ribosome inactivating hemitoxins, has
`several advantages over RTA. Gelonin appears to be
`more stable to chemical and physical treatment than
`RTA. 3° Furthermore, gelonin itself does not bind to
`cells and is, therefore, non-toxic (e,Z(cept in high con(cid:173)
`centrations) and is safe to manipulate in the laborato(cid:173)
`ry. While the primary sequence of gelonin is un(cid:173)
`known, mechanistic studies indicate that it operates
`identically to that of RTA. 31,32
`We have utilized
`the antimelanoma antibody
`ZME-018 as a model delivery vehicle for the plant
`toxin gelonin. The ZME-gelonin conjugate was syn(cid:173)
`thesized, purified, and tested in a variety of in vitro
`human tumor models to determine its potential utility
`as an immunotherape'tttic agent.
`
`Materials and Methods
`Materials
`
`Antibody ZME-018 was prepared at Hybritech, Inc.
`(LaJolla, CA, USA), using salt fractionation and
`DEAE chromatography and was judged homogenous
`by SDS PAGE. 33 SPDP reagent (N-succinimidyl 3:[2-
`pyridyldithio] propionate), Sephacryl S-300 gel perme(cid:173)
`ation resin and Blue Sepharose CL-6B resin were
`purchased from Pharmacia, Piscataway, NJ, USA.
`2-Iminothiolane HCl was purchased from Pierce
`Chemical Co., Rockford, IL, USA. All materials were
`of reagent grade or higher. For estimation of the cell(cid:173)
`free protein synthesis inhibitory activity of the toxin,
`a rabbit reticulocyte translation kit was purchased
`from Bethesda Research Labs (Bethesda, MD, USA).
`
`Methods
`
`Coupling of ZME-018 to gelonin. A stock solution of
`SPDP (6 mg/ml) in dry DMF was prepared. To 1 ml of
`a PBS solution containing 1 mg of ZME-018 in a 12 x
`75 mm glass test tube, SPDP was slowly added to a
`5-fold molar excess (approx. 10 J.Ll of stock solution).
`The mixture was vortexed every 5 minutes for 30 min(cid:173)
`utes at room temperature.
`Excess unreacted SPDP was removed from the
`sample by gel filtration chromatography on a Sepha(cid:173)
`dex G-25 column (1 x 24 em) pre-equilibrated in 100
`mM sodium phosphate buffer pH 7.0 containing 0.5
`mM EDTA (Buffer A). Fractions (0.5 ml) were col(cid:173)
`lected and analyzed for protein content using the
`Bradford dye binding assay. 34 Absorbance (600 nm)
`was monitored in a 96-well plate using a Bio-TEK Mi(cid:173)
`croplate autoreader. Antibody eluted at the void vol(cid:173)
`ume (fractions 14-20) and these fractions were pooled
`and kept at 4°C.
`For these studies, gelonin toxin was extracted from
`the seeds of Gelonium multiflorum and purified to ho-
`
`mogeneity utilizing the method of Stirpe et alP One
`milligram of purified gelonin (2 mg/ml PBS) was added
`to triethanolamine hydrochloride (TEA/HCl) buffer to
`a final concentration of 60 mM TEA/HCl and adjusted
`to pH 8.0. The solution was made 1 mM EDTA. 2-
`Iminothiolane stock solution (0.5 M in 0.5 M TEA/
`HCl pH 8.0) was added to a final concentration of 1
`mM and the sample was incubated for 90 minutes at
`4°C under nitrogen gas.
`Excess 2-iminothiolane reagent was removed by gel
`filtration on a column of Sephadex G-25 (1 x 24 em)
`pre-equilibrated with 5 mM bis-tris acetate buffer pH
`5.8 containing 50 mM NaCl and 1 mM EDTA. Frac(cid:173)
`tions were analyzed for protein content in 96 well mi(cid:173)
`crotiter plates using the Bradford dye binding assay.
`Gelonin eluted at the void volume (fractions 14-20).
`SPDP-modified antibody ZME was mixed with an
`equal weight of 2-iminothiolane modified gelonin. This
`proportion corresponded to a 5-fold molar excess of
`gelonin as compared to antibody. The pH of the mix(cid:173)
`ture was adjusted to 7.0 by the addition of0.5 M TEA/
`HCl buffer (pH 8.0), and the mixture was incubated
`for 20 hours at 4°C under nitrogen. Iodoacetamide (0.1
`M in H20) was added to a final concentration of 2 mM
`to block any remaining free sulfydryl groups, and in(cid:173)
`cubation was continued for an additional hour at 25°C.
`
`Purification of ZME-gelonin complexes. To remove
`low molecular weight products and non-conjugated
`gelonin, the reaction mixture was applied to a Sepha(cid:173)
`cryl S-300 column (1.6 x 31 em) previously equilibrat(cid:173)
`ed with PBS. Fractions (1.0 ml) were collected and 50
`J.Ll aliquots were analyzed for protein content using the
`Bio-Rad dye binding assay. To remove unconjugated
`ZME-018, the high molecular peak (Fractions 28-41)
`from the S-300 column was applied to an affinity chro(cid:173)
`matography column of Blue Sepharose CL-6B (1 x 24
`em) pre-equilibrated with 10 mM phosphate buffer
`(pH 7.2) containing 0.1 M NaCl. After sample loading,
`the column was washed with 50 ml of buffer to elute
`completely non-conjugated antibody. The column was
`eluted with a linear salt gradient of 0.1 to 2M NaCl in
`10 mM phosphate buffer pH 7.2. Protein content of
`the eluted fractions was determined by the dye-bind(cid:173)
`ing assay described previously. 34
`
`Cell culture methods. Human bladder carcinoma
`(T-24), human cervical carcinoma, or human metastat(cid:173)
`ic melanoma tumor cells A375M or AAB-527 were
`maintained in culture using minimal essential medium
`(MEM) supplemented with 10% heat-inactivated fetal
`bovine serum plus 100 J.LM non-essential amino-acids,
`2 mM L-glutamine, 1 mM sodium pyruvate, vitamins,
`and antibiotics. Cultured cells were screened routine(cid:173)
`ly and found free of mycoplasma infection.
`
`Cell proliferation assay. Cell lines were maintained in
`culture in complete medium at 37°C in a 5% C02-hu(cid:173)
`midified air incubator. For assays with combinations
`ofTNF, immunotoxins, riFNaA, and riFN'Y, cultures
`
`Mol. Biother., 1991, vol. 3, March
`
`7
`
`IMMUNOGEN 2045, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`ZME-oiS + ZHE-saONIN
`
`Papers
`
`1.0
`
`E' 0.8
`c
`0
`"<;)" 0.6
`.!£l
`w
`u :z: 0.4
`<C rn
`0:
`0 en
`rn 0.2
`
`<C
`
`FLOW THRU
`
`ZME-GELONIN
`
`.30
`
`.25
`
`.20
`
`.15
`
`.10
`
`.05
`
`E'
`c
`0
`"<;)"
`.!£l
`w
`u
`:z:
`<C rn
`0:
`0 en
`rn
`<C
`
`0.0
`0
`
`20
`
`60
`40
`(1 ml)
`FRACTION
`Figure 1. Purification of ZME-gelonin by S-300 gel permeation
`chromatography. High molecular weight material consisting of
`unmodified ZME and ZME-gelonin conjugate was eluted in
`fractions 28-41. Free gelonin eluted in fractions 45-65.
`
`80
`
`100
`
`were washed, detached using versene, and resuspend(cid:173)
`ed in complete medium at a density of 25 x 103 cell/
`mi. Two hundred ,_..1 aliquots were dispensed into 96-
`well microtiter plates and the cells were then allowed
`to adhere. This results in a sparsely seeded population
`of cells. Mter 24 hours, the media were replaced with
`media containing different concentrations of either
`immunotoxin, gelonin, TNF, IFN'Y, or IFNa. The
`cells were incubated for 72 hours and analyzed for rel(cid:173)
`ative cell proliferation by crystal violet staining.
`
`Crystal violet staining. Cells were washed 3 times with
`PBS containing calcium and magnesium fixed and
`stained with 20% (v/v) methanol containing 0.5% (w/v)
`crystal violet. Bound dye was eluted with 150 ,....I of
`Sorensen's citrate buffer (0.1 M sodium citrate, pH
`4.2-50% (v/v) ethanol) for 1 hour at room tempera(cid:173)
`ture. The absorbance was measured at 600 nm using a
`Bio-Tek microplate reader. Relative cell proliferation
`(RCP) was calculated as follows:
`
`Mean Absorbance (Drug Treated)
`OOM
`RCP = - - - - - - - - - - - - - - - - X 1
`/0
`Mean Absorbance (Non-drug Treated)
`
`[eq 1]
`
`Human tumor colony assay. Thmor biopsy specimens
`were obtained from melanoma patients during clini(cid:173)
`cally indicated biopsy procedures. Portions of tumor
`not required for standard diagnostic evaluation were
`transferred promptly to the human tumor cloning lab(cid:173)
`oratory, wherein tumor cell suspensions were pre(cid:173)
`pared aseptically. 35 Additionally, the A375P melanoma
`and the CEM leukemia cell lines from the American
`Type Culture collection (Rockville, MD, USA) were
`also studied. Testing for the effects of ZME-gelonin on
`the fresh melanoma cell suspensions and cell lines was
`assessed in the HTCA using standardized procedures
`for tumor cell plating in semi-solid medium (agarose)
`in the presence of complete medium containing 10%
`fetal calf serum. Each 0.5 ml culture plate contained
`100,000 cells from fresh tumors and 10,000 cells from
`
`80
`
`100
`
`0
`0
`
`20
`
`60
`40
`FRACTION ( 1 ml)
`Figure 2. The high-molecular weight material from S-300 chro(cid:173)
`matography was applied to a column of Blue Sepharose and
`eluted with a linear salt gradient (0-300 mM NaCI). Two protein
`peaks were demonstrated: a flow-through peak (fractions
`11-25) and a bound peak eluted with high salt (fractions
`44-80).
`
`the celllines. 36- 38 ZME-gelonin prepared as described
`above was tested by addition to the culture plates
`shortly after tumor cell plating. ZME-gelonin was
`added to triplicate plates at each of four concentra(cid:173)
`tions 0.025 ng/ml to 250 ng/ml. In addition to untreated
`control plates, unconjugated ZME-018 monoclonal
`antibody and free gelonin were tested in parallel. Cell
`lines and tumor cell cultures were incubated for an
`average of 10 days at 37°C in 5% C02 in air in a humid(cid:173)
`ified incubator. Colony formation was evaluated with
`a viability stain39 and an automated image analysis in(cid:173)
`strument optimized for colony counting.40 Percent sur(cid:173)
`vival of ZME-018-treated cultures in relation to un(cid:173)
`treated controls were determined
`in
`the same
`experiments. Dose-response curves were then plotted
`graphically.
`
`Results
`Conjugation and purification of
`ZME-gelonin immunotoxin
`Both ZME-018 and gelonin were modified by expo(cid:173)
`sure to the reagents SPDP and 2-iminothiolane, re(cid:173)
`spectively. Modification of ZME with 2 to 3 molecules
`of SPDP was accomplished by addition of a 6-fold
`molar excess of SPDP. Analysis of the number of
`SPDP molecules was accomplished spectrophotomet(cid:173)
`rically.41 The derivitization of ZME-018 and gelonin
`and the coupling were varied to provide optimal yield
`of ZME-gelonin conjugate (data not shown). Figure 1
`demonstrates separation of the ZME-gelonin conju(cid:173)
`gate from gelonin. Blue sepharose CL-6B chromatog(cid:173)
`raphy (Figure 2) was effective at separating free ZME
`from ZME-gelonin conjugate. As demonstrated in
`Figure 3, the final eluent from Blue Sepharose was
`free of unreacted gelonin and ZME-018. The final
`product contained ZME-coupled to 1, 2, and 3 gelonin
`molecules. Average gelonin content was 1.5 molecules
`per antibody molecule. Average yield of purified im-
`
`8
`
`Mol. Biother., 1991, vol. 3, March
`
`IMMUNOGEN 2045, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`ZME-gelonin conjugate as an immunotherapeutic agent: Rosenblum et at.
`
`~
`"'
`~
`
`2.0
`
`1.5
`
`w
`(._)
`:z:
`<C
`aJ 1.0
`cr::
`Cl
`(f)
`aJ
`<C
`
`0.5~
`
`0.0 '---'='-~~._,_,_[_~~~o.L_~~ ........... ~~~_._,J
`10 1
`10 2
`10 3
`10 4
`1QO
`CONCENTRATION (ug/ml)
`Figure 4. Comparative ELISA Assay of ZME (0) and ZME ge(cid:173)
`lonin (•). Various concentrations of either ZME or ZME-gelonin
`were added to antigen-positive (AAB-527) human melanoma
`cells and incubated at room temperature for 60 minutes. The
`cells were washed and a standard ELISA was performed for mu(cid:173)
`rine antibody.
`
`negative T-24 cells as estimated by ELISA assay (data
`not shown). Therefore, the apparent change in anti(cid:173)
`body affinity resulting in the coupling of ZME to ge(cid:173)
`lonin does not appear to reduce selectivity of the anti(cid:173)
`body.
`
`Cytotoxicity of ZME-gelonin
`Cytotoxicity studies of the ZME-gelonin conjugate
`were performed on antigen-positive cells after contin(cid:173)
`uous (72-hour) exposure to the immunotoxin or native
`gelonin. As shown in Figure 5, 50% cell death by the
`immunotoxin occurred at a concentration of approxi(cid:173)
`mately 0.02 f.Lg/ml. A similar effect (i.e., 50% cell
`death) was observed for native gelonin at a concentra(cid:173)
`tion of 5 f.Lg/ml. These concentrations are equated to a
`dose of 0.1 nM for ZME-gelonin and a dose of 100 nM
`of free gelonin to achieve reduction of the number of
`viable cells to 50% of control value.
`The activity of both ZME-gelonin immunotoxin and
`free gelonin were determined using a cell-free transla(cid:173)
`tion assay system. 42 One unit of activity in this assay
`was defined as the amount of protein required to pro(cid:173)
`vide 50% inhibition of protein synthesis compared to
`untreated controls. Utilizing this assay, the specific
`activity of both the native gelonin and the ZME(cid:173)
`gelonin conjugate were determined to be 2 x 108 f.L/mg
`and 8.2 x 105 f.L/mg, respectively. Target cells then
`were treated with various concentrations of ZME-ge(cid:173)
`lonin and gelonin alone. As shown in Figure 6, a 50%
`inhibitory concentration was obtained using 50 units/
`ml of ZME-gelonin conjugate while 1 x 107 units/ml
`of the free gelonin were required to achieve the same
`effect.
`The effect of ZME-gelonin was determined against
`antigen-negative T-24 cells in log-phase culture. Ge(cid:173)
`lonin alone produced 50% cytotoxicity in these cells at
`a concentration of 10 f.Lg/ml; similar to that found on
`AAB-527 melanoma cells (data not shown). However,
`the ZME-gelonin immunotoxin was not active against
`
`Figure 3. Silver stained PAGE analysis of ZME-gelonin conju(cid:173)
`gation and purification procedure.
`LANE 1-ZME-018 antibody standard.
`LANE 2-Native gelonin.
`LANE 3-Unpurified ZME-gelonin reaction mixture showing
`the presence of unreacted ZME antibody and unreacted gelon(cid:173)
`in. Also shown are high molecular weight conjugate bands cor(cid:173)
`responding to ZME + 1 gelonin molecule, ZME + 2 gelonin
`molecules and ZME + 3 gelonin molecules.
`LANE 4-High molecular weight peak from S-300 chroma(cid:173)
`tography demonstrating removal of low molecular weight com(cid:173)
`ponents including most of the free gelonin.
`LANE 5--Fiow-through fraction from application of the con(cid:173)
`jugate to a Blue Sepharose column. All of the unreacted anti(cid:173)
`body eluted from the column in the low-salt wash.
`LANE 6--Final purification product eluted from the Blue
`Sepharose column. As shown, there was no remaining free an(cid:173)
`tibody and only small amounts of unreacted gel-onin in the elut(cid:173)
`ed fraction.
`
`munotoxin was approximately 20% of expected maxi(cid:173)
`mal value.
`
`Binding of ZME-gelonin to cells in culture
`The binding of ZME-gelonin immunotoxin to antigen
`positive (AAB-527 cells) or antigen negative (T-24
`cells) was tested by ELISA assay. As shown in Figure
`4, both native ZME and the ZME-gelonin conjugate
`bound well to target cells after 60-minute exposure.
`Surprisingly, the ZME-gelonin conjugate bound target
`cells better than did the native antibody. This increase
`was not due to modification of the antibody by SPDP
`since SPDP-modified ZME behaved identically to that
`of native ZME. The increase was also not due to bind(cid:173)
`ing of target cells to the gelonin portion of the mole(cid:173)
`cule since pre-treatment of target cells with native
`gelonin had no effect on either antibody or immuno(cid:173)
`toxin binding. Thus, these results suggest that cou(cid:173)
`pling of ZME to gelonin may affect antibody affinity
`for the gp240 antigen.
`Neither ZME nor ZME-gelonin bound to antigen
`
`Mol. Biother., 1991, val. 3, March
`
`9
`
`IMMUNOGEN 2045, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Papers
`
`100
`
`75
`
`50
`
`25
`
`__j
`
`C) a:
`1-:z:
`8
`6
`
`o GELONIN
`e ZME-GELONIN
`
`o~~~~~~~~~~~~~~~~
`
`10- 3
`
`10- 2
`
`10 1
`10°
`10- 1
`(nM)
`CONCENTRATION
`Figure 5. Cytotoxicity of ZME-gelonin and free gel on in on log(cid:173)
`phase AAB-527 cells. Cells were plated for 24 hours and various
`doses of either gelonin or ZME-gelonin were added in complete
`media and incubated for 72 hours. Cell viability was determined
`by crystal violet staining.
`
`100
`
`80
`
`60
`
`40
`
`20
`
`.......
`0
`_,_,
`'-
`c
`0
`LJ
`.......
`0
`
`~
`
`•••
`•,.
`'
`\
`•
`•
`'
`\
`•
`' •
`• \ •
`
`\
`
`0-o-o-o-o
`\
`0.0
`
`0
`
`0
`
`\
`\
`\
`
`0
`
`o~~~~~~~~~~~=L~=L~~
`
`10°
`
`10 1
`
`10 7
`
`108
`
`10 4
`10 3
`10 2
`10 6
`105
`Concentration (u/ml)
`Figure 6. Cytotoxicity of ZME-gelonin and free gelonin on log(cid:173)
`phase AAB-527 cells. As in Figure 6, log-pha~e AAB-527 cells
`were treated with various doses of gelonin or ZME-gelonin. The
`biological activity of gelonin or ZME-gelonin was determined by
`a cell-free translation inhibition assay. One unit activity was de(cid:173)
`fined as the amount of toxin required to inhibit translation activ(cid:173)
`ity by 50%.
`
`100
`
`80
`
`60
`
`40
`
`20
`
`__j
`
`C) a:
`I -
`:z:
`C)
`LJ
`
`l.J_
`C)
`
`~
`
`<)
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0 0-0
`
`8 TARGET
`
`(AAB-527)
`
`<>NON-TARGET
`
`IT -24)
`
`0
`10- 2
`
`10 1
`10°
`10- 1
`CONCENTRATION (ug/ml)
`Figure 7. Cytotoxicity of ZME-gelonin on antigen positive mel(cid:173)
`anoma cells (AAB-527) and antigen negative T-24 cells in cul(cid:173)
`ture. As in Figure 6, log-phase AAB-527 or T-24 cells were treat(cid:173)
`ed for 72 hours with various concentrations of ZME-gelonin
`immunotoxin. Cell viability was then determined by crystal vio(cid:173)
`let staining.
`
`100
`
`80
`
`60
`
`40
`
`>-
`I -
`>--<
`LJ
`>--<
`><
`0
`I -
`C)
`I -
`>-
`LJ
`
`l.J_
`C)
`
`~ 20
`
`o 15A8
`e ZME-018
`
`0
`10-1
`
`10 3
`
`10 2
`10 1
`10°
`CONCENTRATION (ug/ml)
`Influence of free antibody on ZME-gelonin cytotoxicF
`Figure 8.
`ty. A fixed concentration of ZME-gelonin was mixed with in(cid:173)
`creasing concentrations of ZME-018 or irrelevant antibody
`15A8. The mixture was then added to log-phase AAB-527 mela(cid:173)
`noma cells for 72 hours. As shown, increasing amounts of
`ZME-018 antibody prevented ZME-gelonin cytotoxicity. This ef(cid:173)
`fect was specific since irrelevant 15A8 antibody had no effect.
`
`these cells even at the highest concentration tested
`(Figure 7).
`If the cytotoxic effect of ZME-gelonin is mediated
`through recognition of a specific cell-surface antigen,
`addition of free ZME antibody should reduce cytotox(cid:173)
`icity of the immunotoxin. To demonstrate this, a fixed
`immunotoxin designed
`to
`dose of ZME-gelonin
`achieve 80% cytotoxicity was added to log-phase mel(cid:173)
`anoma cells in culture. Simultaneously, various doses
`of ZME or irrelevant antibody were added. As shown
`in Figure 8, high concentrations of ZME-018 antibody
`substantially reduced the cytotoxicity of ZME-gelonin
`conjugate, while addition of irrelevant antibody had
`no effect.
`
`Modulation of ZME-gelonin cytotoxicity with
`IFNa, IFNy, and TNF
`To determine the effects of treatment with various bio(cid:173)
`logical response modifiers on immunotoxin cytotoxici-
`
`ty, log-phase melanoma cells were treated for 24 hours
`with fixed doses of IFNaA (200 f.L/ml), IFN-y (20,000
`f.L/ml), or rTNF-a (20,000 f.L/ml). These doses were de(cid:173)
`termined previously to have minimal (~20%) cytotox(cid:173)
`ic effect against these cells. The cells then were
`washed and treated for 72 hours with various doses of
`ZME-gelonin. As shown in Figure 9, treatment with
`riFN-y resulted in a 2-fold increase in sensitivity to the
`immunotoxin. However, pre-treatment with both
`riFNaA and TNF both resulted in a 100-fold increase
`in sensitivity to the immunotoxin.
`
`Human tumor colony assay (HTCA) of
`ZME-gelonin immunotoxin
`An alternative method of assessing in vitro cytotoxici(cid:173)
`ty against human cells in culture is the soft-agar
`HTCA. 39 Various doses of ZME-gelonin immunotoxin
`were added to an antigen positive (A-375 melanoma)
`
`10
`
`Mol. Biother., 1991, vol. 3, March
`
`IMMUNOGEN 2045, pg. 6
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`ZME-gelonin conjugate as an immunotherapeutic agent: Rosenblum eta/.
`
`·- '
`
`-.. __
`-.
`---._ ' '
`•--.. ___
`
`1000
`
`100
`
`10
`
`_j
`<C
`>
`>-<
`>
`a:
`::::J
`(f)
`
`1-:z:
`w
`u
`a:
`w
`0...
`
`,--<
`0
`c_
`_,__,
`c:
`0
`u
`
`"---
`0
`
`;,-E!
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`10-2
`
`10 1
`10-1
`100
`Concentration (ug/ml)
`Figure 9. Effect of IFN-a, IFN--y, and TNF on ZME-gelonin cyto(cid:173)
`toxicity. Log-phase AAB-527 cells were treated with various
`doses of ZME-gelonin alone (-•), or ZME-gelonin in the pres(cid:173)
`ence of fixed concentrations of IFN--y (<>), IFN-a (6), or TNF
`(•· · ·•). Doses of IN F-a, IFN--y, or TNF alone demonstrated mini(cid:173)
`mal cytotoxicity (<20%) against this cell line.
`
`10 2
`
`and antigen negative CEM cell lines. Survival of colo(cid:173)
`nies was assessed 72 hours after addition. As shown in
`Figure 10, doses of immunotoxin between 0.25 and
`2500 ng/ml resulted in almost complete suppression of
`colony survival of the antigen-positive line. Unconju(cid:173)
`gated ZME-018 and free gelonin alone or combined
`together were not cytotoxic. There was no effect
`against the antigen-negative line (CEM) even at the
`highest concentration of immunotoxin tested.
`The effect of ZME-gelonin against four different
`fresh biopsy specimens is shown in Figure 11. Eighty
`to 90% reduction in survival of melanoma colony
`forming cells was found in two specimens at the high(cid:173)
`est immunotoxin dose tested (250 ng/ml). Only a mod(cid:173)
`est (25%) reduction in colony number was noted with
`a third specimen. Growth enhancement·was noted in
`the fourth sample at the highest immunotoxin dose. In
`addition, growth enhancement was observed in one
`specimen at low doses, while higher doses produced
`substantial cytotoxicity. As in the cell line experi(cid:173)
`ments, addition of unconjugated ZME-018 and free
`gelonin were not cytotoxic at the doses tested. The
`surface antigen expression of gp240 and the internal(cid:173)
`ization rates for ZME-gelonin conjugate were not ex(cid:173)
`amined in these four lines and are currently under in(cid:173)
`vestigation.
`
`Discussion
`
`The necessity for precisely targeting cancer therapy is
`critical since adequate tumor response is dependent
`upon delivery and maintenance of intratumor thera(cid:173)
`peutic concentrations of drugs. Site-directed therapy
`has become a goal of several investigators utilizing
`monoclonal antibodies as specific carriers of thera(cid:173)
`peutic agents.
`The cytotoxic agents frequently utilized for anti(cid:173)
`body conjugates primarily fall into three classes of
`agents: toxins, radionuclides, and chemotherapeutic
`agents. Antibody conjugates with each of these types
`
`1L-~~~~~--~~~~~~~~~~
`
`.01
`
`.1
`
`100
`10
`CONCENTRATION (ng/ml)
`Figure 10. Effect of ZME-gelonin on antigen positive (A-375 e)
`and antigen negative (CEM D) cells in a human tumor stem cell
`assay. There was no effect of ZME-gelonin on antigen negative
`cells. In contrast, there was only 2% survival of antigen positive
`cells exposed to 20 ng/ml of immunotoxin.
`
`1000 10000
`
`10110r
`
`100
`
`10
`
`_j
`<C
`>
`>-<
`> a:
`
`::::J
`(f)
`
`1-:z:
`w
`u
`a:
`w
`0...
`
`1~~~L-~~~~~~~~~~~~
`
`100
`
`1000
`
`.01
`
`.1
`
`10
`CONCENTRATION (ng/ml)
`Figure 11. Cytotoxic effect of ZME-gelonin on stem cell surviv(cid:173)
`al of different lines obtained from fresh biopsy specimens of 4
`different patients. As shown by the various symbols, two line
`exhibited marked growth inhibitory effects with the ZME-gelon(cid:173)
`in conjugate (•,•). Approximately 25% growth inhibition was
`obtained with one line (6) and slight growth enhancement was
`observed with another line tested (A).
`
`of agents offer substantial promise as therapeutic
`agents as well as unique problems. 43 •44 Immunoconju(cid:173)
`gates containing plant toxins offer a unique advantage
`to other types of antibody conjugates because:
`
`1) Doses of immunotoxins required for antitumor ac(cid:173)
`tivity are, in general, much lower than that re(cid:173)
`quired for antibody-drug conjugates.
`2) The conjugation of toxins to antibodies does not
`appear to affect antibody affinity.
`
`Previous studies have described a number of anti(cid:173)
`body-toxin conjugates containing gelonin. 45- 51 Re(cid:173)
`cently, Ozawa et al. 50 constructed a gelonin immuno(cid:173)
`toxin comprised of antibody B4G7 which binds to the
`cellular receptor for epidermal growth factor (EGF).
`This B4G7-gelonin conjugate was highly cytotoxic for
`EGF receptor expressing cells but was non-cytotoxic
`
`Mol. Biother., 1991, vol. 3, March
`
`11
`
`IMMUNOGEN 2045, pg. 7
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Papers
`
`for receptor-deficient cells. Sivam et al. 51 have made
`a conjugate of the antimelanoma antibody 9.2.2.7 with
`gelonin and compared in vitro and in vivo cytotoxic
`activity with a 9.2.2.7 conjugate of abrin and ricin A
`chain. These studies demonstrate that gelonin conju(cid:173)
`gates demonstrate substantial cytotoxic effects selec(cid:173)
`tively against antigen-positive cells in vitro. In vivo
`experiments demonstrated that gelonin conjugates are
`not toxic up to 2 mg total antibody dose/mouse and
`that multiple I. V. administration of gelonin immuno(cid:173)
`toxin significantly retarded the growth of an estab(cid:173)
`lished S.C. human tumor xenograft in nude mice.
`Compared to conjugates with abrin and ricin, gelonin
`conjugates appeared to have similar potency, better
`selectivity and tumor localization with more signifi(cid:173)
`cant in vivo therapeutic properties.
`In the current study, we have demonstrated that the
`ZME-gelonin immunoconjugate appears in vitro to be
`a highly cytotoxic reagent. Moreover, the cytotoxicity
`of the immunotoxin appears to be specific for antigen(cid:173)
`bearing cells.
`Substantial augmentation of ZME-gelonin cytotox(cid:173)
`ic effect was observed with pre-treatment of riFNaA
`and rTNF but not with riFN--y. While previous stud(cid:173)
`ies by our group52 and by others53- 57 have demonstrat(cid:173)
`ed that IFNa and IFN--y can up-regulate some mela(cid:173)
`noma surface antigens such as P-97, there was little
`effect of these biological response modifiers on the
`high molecular weight antigen (gp240) recognized by
`ZME. Therefore, the mechanism ofTNF-a and IFN-a
`induced augmentation of ZME-gelonin activity is not
`clear but could involve changes in the antibody inter(cid:173)
`nalization rate, changes in the cellular processing of
`the immunotoxin, or a modulation of any one of sever(cid:173)
`al interferon-mediated enzymes. 57
`The evaluation of ZME-gelonin using the HTCA is
`of potential clinical interest. In other studies, the
`HTCA has been used as a predictive model for clinical
`response of a given tumor to a given potential thera(cid:173)
`peutic agent. 37•58 The results indicate that ZME-gelon(cid:173)
`in was active against 3 of 4 fresh human tumor lines
`tested and against all antigen positive melanoma cell
`lines. These findings are important since most human
`tumors are composed of heterogeneous populations of
`malignant cells with varying antigen expression. Anti(cid:173)
`genic modulation by tumor cells may result in the de(cid:173)
`velopment of resistance to ZME-gelonin, and such re(cid:173)
`sistance might be predicted utilizing the HTCA. Phase
`I clinical evaluation and in vivo pharmacokinetics will
`be required to determine if the doses of ZME-gelonin
`required for cytotoxic effect against fresh human
`tumors in culture will be achievable without unaccept(cid:173)
`able toxicity.
`While site-directed therapy with immunotoxins,
`specifically ZME-gelonin, appears to be feasible, at
`least in vitro, there still remain substantial practical
`barriers to the optimal delivery of immunotoxins and
`maintenance of therapeutic concentrations of drug at
`the tumor sites. Further studies with immunotoxins
`will explore in vivo behavior of ZME-gelonin and
`
`evaluation of this reagent as a potential and practical
`tool for clinical use.
`
`Reference