[CANCER RESEARCH 51. 1482-1487. March 1. 1991] Covalent Binding of Human a2-Macroglobulin to Deglycosylated Ricin A Chain and Its Immunotoxins1

Maria-Ana Ghetie, Jonathan W. Uhr, and Ellen S. Vitetta2 Department of Microbiology and the Cancer Immunobiology (enter, I 'niversity of Texas Southwestern Medical Center, Dallas, Texas 75235

ABSTRACT complexes which form with platelet-derived growth factor (9), lymphocyte-produced proteins (10), and interleukin 10(11). In In this report we demonstrate that human a2-macroglobulin (a2M) contrast to the bond formed between «2Mand proteases, the reacts with deglycosylated ricin A chain (dgA) and its immunotoxins to linkage between «2Mand these three proteins can be cleaved form high molecular weight complexes (molecular mass approximately 800 kDa). This interaction has a /i/2at 37°Cof5 h and reaches completion by reduction; and (c) noncovalent binding to a variety of pro at 24 h. Complexes of «2M-dgAcannot be dissociated by guanidine, teins through a reversible hydrophobic or ionic linkage (1). sodium dodecyl sulfate, or low pi I, but can be partially dissociated by In an earlier study describing the fate of antibody-ricin A reducing agents, such as 2-mercaptoethanol in the presence of sodium chain conjugates in normal rats (12), it was noted that IgG- dodecyl sulfate. This indicates that dgA or dgA-containing immunotoxins ricin A chain in serum was eluted from a sizing column as two are bound to a2M by disulfide bonds. The dgA-binding site on a2M and Chromatographie peaks with molecular masses of >800 kDa the mechanism underlying its interaction with dgA are different from and 180 kDa. In vivo administration of radiolabeled ricin A those described for proteases or methylamine. <>2Mcomplexes do not chain resulted in the rapid disappearance of radioactivity from bind to Blue-Sepharose 4B or anti-A chain-Sepharose, suggesting that the sites on dgA which bind Cibacron Blue or polyclonal anti-A chain the circulation, and the formation of a species with a high antibodies are sterically blocked or modified by interaction with ,.2M. molecular mass (13). These results, taken together with our The interaction of a2M with dgA or its immunotoxins results in a 2- to previous observation that injected immunotoxins appear in the 3-fold decrease in the activity of the dgA in both cell-free assays and serum as high molecular mass material, prompted us to inves cytotoxic assays. However 12 h after injection into mice, only 11% of tigate the interaction between ricin A chain and some of its immunotoxin was bound to «2Mbecause of the slow kinetics of the conjugates with different human plasma proteins. We found interaction versus the more rapid r1/2 of the immunotoxin in the that 90%), an IgG fraction of the «2Mmolecule is the presence of a thioester bond involv of rabbit anti-human «2M,human fibrinogen, transferrin, «,-acidgly- ing the cysteine residue and «-carboxyl group of a nearby coprotein, »-globulins(fraction IV-1). and Cibacron Blue 3GA were glutamic acid residue in each of the subunits (3, 4). There are purchased from Sigma (St. Louis. MO). The dgA, prepared and char three distinct types of binding of molecules by «2M(5): (a) acterized as previously described (14), was purchased from Inland Laboratories (Austin, TX). Human -y-globulin (fraction II) was pur specific trapping of proteases (2). Each subunit of «2Mhas a "bait" region that is susceptible to limited proteolysis. During chased from Pentex (Miles Laboratories, Inc.) and human IgM was "entrapping" of proteases, the thiol esters are cleaved resulting purchased from Dakopats (Santa Barbara, CA). Immunotoxins con in the generation of free —SH groups (3, 4). An intramolecular taining mouse IgGl monoclonal anti-CD22 antibodies (RFB4-dgA) (15) were prepared as previously described (16). The number of free (7-glutamil) lysine bond between «2Mand the protease may be —SHgroups in «2Mwas determined by using 2 mM 5,5'-dithiobis(2- generated following the closing of the "trap" and this bond is nitrobenzoic acid) (Pierce Chemical Co., Rockville, IL) (3). resistant to reduction (6, 7); (b) covalent binding of proteins Animals. BALB/c mice (6-8 weeks old) were obtained from our containing sulfhydryl groups. One or more of the four —SH colony in the Animal Resources Center at University of Texas South groups exposed in the protein and form a di RFB4-dgA, and a2M-dgA complex) were labeled with Na I25Iusing the sulfide-bonded heterodimer. Examples of these include «2M lodo-Gen reagent (Pierce) (17). The specific activities of the proteins were approximately 1 /¡Ci/Mg. Received 9/18/90; accepted 12/20/90. In Vitro Interaction of '"I-dgA or <25l-Immunotoxins Containing dgA The costs of publication of this article were defrayed in pan by the payment with Serum or Plasma. Human, mouse, and monkey scrum or plasma of page charges. This article must therefore be hereby marked advertisement in accordance with 18 L'.S.C. Section 1734 solely to indicate this fact. (200 \i\) were mixed with radiolabeled dgA or with immunotoxins ' This work is supported by NIH Grants CA-28149 and CA-41081 and Cirant containing dgA at different ratios (30-60 ^g/ml serum or plasma) and 1-947 from the Welch Foundation. were incubated for different intervals (5 min to 24 h) at 37°Cand then 2To whom requests for reprints should be addressed, at Department of Microbiology. University of Texas Southwestern Medical C'enter. Dallas. TX subjected to analysis on SDS-PAGE and gel permeation HPLC. In Vivo Interaction of I25I-dgA or '"I-Immunotoxin with Plasma 75235. "The abbreviations used are: «2M,«¡-macroglobulin:BSA. bovine serum Proteins. l25I-dgA or 12';I-RFB4-dgA (10 ¿tgprotein containing 5-8 x albumin; dgA. deglycosylated ricin A chain; PBE. 0.05 M phosphate buffer IO6cpm) was injected i.v. (retroorbitally) into BALB/c mice and plasma containing 3 HIMdisodium EDTA. pH 7; PBS. phosphate-buffered saline; SDS- PACiK. sodium dodccyl sulfate polyacrylamide gel electrophoresis; 2-ME. 2- samples were obtained from the tail vein at intervals of 5 min to 48 h mercaptoethanol; HPLC. high pressure liquid chromatography: RFB4-dgA. thereafter. Samples were subjected to HPLC analysis and SDS-PAGE immunotoxin. followed by autoradiography. Densitometric scanning of the autoradi- 1482

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. COVALENT BINDING OF HUMAN (.2M TO dgA ographs (using a Bio-Rad videodensitometer) was used to calculate the unbound protein was eluted with PBE and the retained material was relative amounts of l25I-dgA or I25l-immunotoxin bound to a protein eluted with 0.5 M NaCl. with a high molecular mass protein. Affinity Chromatograph) on Anti-dgA-Sepharose 4B. Purified com Identification of Protein Component(s) Reacting with dgA or Immn- plexes of 0.2 mg purified 2M. either preparative HPLC or by gel filtration on Sephacryl S-200HR. Cytotoxic Assay. The cytotoxic activity of RFB4-dgA which had been Gel Permeation HPI.C. Mixtures of radiolabeled dgA or its immu- incubated at 37°Cfor different intervals of time (1.5. and 24 h) (60 notoxin and serum, plasma, or plasma proteins were applied either to Mg/ml) with «2M(1 mg/ml) or with an irrelevant protein (BSA, 0.5 an analytical 7.5 x 600 mm TSK 3000SVV column (Spherogel; LKB, mg/ml) was compared with that of the initial immunoconjugate using Bromma, Sweden) or to a preparative 21.5 x 600 mm TSKG3000SWG Daudi cells as described previously (21). column (Ultropac; LKB) and separation was performed in PBS con Binding of Radioiodinated a;M-RFB4-dgA Complex to Daudi Cells. taining 0.1r; sodium azide at pH 7.0, at a flow rate of l ml/min The affinity constant (A'a)and the number of molecules bound per cell (n) were determined under equilibrium conditions using '25I-RFB4-dgA (Spherogel) and 3 ml/min (Ultropac). The retention times for the peaks of radioactivity (or distance measured from the starting point) were alone or complexed with «2M.The affinity constants were calculated by using the Scatchard form of the equilibrium equation (22). compared to those of standard proteins of known molecular mass Half-Life of «2M-dgAComplex in Circulation of Mice. 12liI-blood collected at intervals of 5 purified complex formed between l25I-dgA and a protein with a high min to 48 h. The total radioactivity remaining in the blood was molecular mass. determined by counting aliquots in a gamma counter and assuming a Molar Ratios of «2M-dgA.To calculate the molar ratio dgA/«2M, total blood volume of 1% of body weight (13). Acid precipitatale an excess amount of '25I-dgA (60 pg/m\) was added to 0.5 mg «2M/ml radioactivity was determined by precipitating aliquots of plasma with and the mixture was incubated overnight at 37°C.The «2M-'25I-dgA 10% trichloroacetic acid. The percentage of the injected radioactivity remaining in the circulation was determined. The half-lives (t, 2) were complex was separated from free dgA either by preparative HPLC or by gel filtration on Sephacryl S-200HR. By knowing the specific determined graphically by extrapolation to zero of the percentage of radioactivity of '25l-dgA in the purified complex and reading the ab- acid precipitatale radioactivity versus time (23). Dissociation of a2M-dgA Complexes. The «2M-dgAcomplexes were sorbance (which included both dgA and »2M).we calculated the con treated with 6 M guanidinc-HCl, glycine buffer, pH 3.O. at room centration of dgA in the complex. Absorption coefficients at 280 nm (1%) of 7.7 and 8.9 were used for dgA and «2M,respectively. By temperature for 90 min or were incubated at room temperature or subtracting the absorbance of the dgA from that corresponding to «2M- boiled for 5 min in 1% SDS with or without 2.5% 2-ME and then analyzed by SDS-PAGE and HPLC. dgA. we calculated the concentration of <\2M in the complex. Then, Treatment of dgA with lodoacetamide. In order to block the available knowing both concentrations, we calculated the molar ratio of dgA/ sulfhydryl group of dgA. dgA was reduced with 0.05 M dithiothreitol «2M. for 30 min at room temperature, chromatographed on Sephadex G-25, Sequential Immunoprecipitation. In order to definitively identify «2M and alkylated with 0.1 M iodoacetamide for 2 h at 4°C.Excess iodoa- in the complexes formed by incubating radiolabeled dgA with human plasma or serum, samples were precipitated with anti-«2M immuno- cetamide was removed by overnight dialysis against PBE. complexes (18). Rabbit anti-n2M antibodies (0.2 mg) were mixed with 250^1 of goat anti-rabbit immunoglobulin. Goat anti-human IgM (0.2 RESULTS mg) mixed with 250 M' of rabbit anti-goat antibody as well as normal Characterization of High Molecular Mass Complexes Formed rabbit IgG were used as controls. The immunocomplexes formed were by Incubating 125I-dgAor I25l-Immunotoxins with Plasma Pro incubated with the protein excluded on the Sephacryl S-200HR column (see "Gel Filtration on Sephacryl S-200HR"). The radioactivity bound teins in Vitro. Table 1 shows the results of HPLC analysis of l:5I-dgA (60 Mg/ml) incubated with humun serum at 37°Cfor to preformed immunocomplexes was determined and the percentage of binding was calculated. SDS-PAGE. Complexes of

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. COVALENT BINDING OF HUMAN „2MTO dgA

Table 2 del permeation IIPLC analysis of human serum after in vitro incubation KDa with '-*l-lahelctl immunoloxins containing tlgA

after of radioactivity 800 mixingdgA of8(10 with molecular mass •I PreparationIgG (h)"015122403.0rikDa*3 kDa*10068.4kDac 80 kDa'3.7 RFB4-dgA7Fab'-RFB4-dg.A*Time 1.6±2.243.0 ±2.257.0 ±4.054.1 ±4.045.9 ±9.565.0 ±9.535.0 ±0.434.0200 ±0.4100.062.330

" 37'C. * Retention time, 9-13 min. 30 ' Retention time. 14-17 min. ''Retention time. 16-18 min. 1 2 3 5 6 'Retention time. 18-22 min. 1 Mean ±SD of 3 experiments. * Average of 2 experiments.

KDa B free (Fig. \A) or conjugated (Fig. \B)\ interacts with some human proteins to yield a high molecular mass complex (=800 kDa). In order to determine which protein(s) interacts with 800 dgA, several purified serum proteins (albumin, IgM, IgG, the «-globulin fraction, «2M,

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. COVALENT BINDING OF HUMAN <>2MTO dgA

Table 3 Relative amounts of high molecular mass complexes formed in the circulation of mice given injections of either n*l-dnA or '"I-immunotoxin (A) containing tlgA ¿ 600 SOCKDa after mass800 I injection Ì 50° Co-noei« .M-ogA Preparation"dgAlgG-RFB4-dgATime(h)02.06.012.002.06.012.0MolecularkDa54.164.567.602.37.210.8200kDa10094.788.584.030kDa100.045.935.532.403.04.35.2 ° 400

Si 300

S 200

10 20 30 40 50 60 " Average of 2 experiments. Number of Tube Fig. 3. Purification of i:5l-bigh molecular mass complex [obtained by incuba tion of ii2M (0.5 mg/ml) (overnight al 37°C)with '"1-dgA (60 ng/nii)] by gel sites on dgA recognized by Cibacron Blue or anti-dgA were Illtralion on S-200 HR (.-I) or by gel permeation on a preparative HPLC column (TSKG3000SWG)(fi). unavailable after the interaction with «2M. Dissociation of the «2M-dgA complexes was attempted by treatment with 6 M guanidine hydrochloride, low pH, or 1% control (rabbit IgG-anti-rabbit ¡mmunoglobulin). No radioac SDS ±2.5% 2-ME. The dgA-«2M complexes could not be tivity (0.5ct) was bound to preformed immune complexes with dissociated by 1% SDS, 6 M guanidine hydrochloride, or at pH goat anti-human IgM and rabbit anti-goat antibodies. 3.0 (data not shown). However, following reduction with 2.5% The complexes obtained by interacting l:iI-dgA with «2M 2-ME in 1% SDS (with or without boiling), «2Mwas disso were purified either by Sepharose S-200HR or by preparative ciated into 185-kDa subunits, and dgA was not bound to these HPLC (Fig. 3) and were used to determine their molecular subunits (Fig. 4). These results suggest that the «2M-dgA mass and the molar ratios of dgA/«2M. The molecular mass complex is formed predominantly through disulfide bonds, and of the complex was =800 kDa and the molar ratio of dgA/2SI-dgA complex (since an excess of dgA was ble —SH groups, indicating that «2M molecule was in its added). The changes in electrophoretic mobility of «2Mthat "activated" form (3, 6, 7). has reacted with either trypsin or methylamine ("fast" forms of Freshly reduced and alkylated dgA does not bind to «2M, «2M)(24) were not observed with the «2M-dgA complexes and consequently, no complexes were formed (data not shown). (Fig. 4), suggesting that the mechanism of interaction of «2M This result adds further evidence to suggest that disulfide bonds with dgA is different from that involving n2M and trypsin or are directly involved in the formation of the a2M-dgA methylamine. complexes. The purified <»2M-dgAcomplex was applied to a column of The interaction of dgA with insolubili/ed «2M(«2M-Seph- Blue-Sepharose-4B and the percentage of free and bound pro arose 4B) was also tested. Insolubilized «2Mbound a larger tein was determined. An aliquot of the same complex was amount of dgA (120 ßgdgA/mg o2M) than soluble «2M, subjected to affinity chromatography on an anti-dgA-Sepha- corresponding to a dgA/a2M molar ratio of 2.9. This difference rose-4B column. The amount of bound material was calculated in the molar ratios calculated in a liquid (2.0) versus solid (2.9) by measuring radioactivity in both the void and retained frac phase might be due to some nonspecific binding of l25I-dgA to tions. In both cases, >90T¿of the «2M-dgAcomplex failed to the Sepharose matrix. In contrast to these results, 125I-«2M bind to the column (data not shown), suggesting that the binding failed to associate with insolubilized dgA (dgA-Sepharose 4B), due perhaps to the fact that the free sulfhydryl groups of dgA became inaccessible for

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. COVALENT BINDING OF HUMAN „2MTO dgA

Table 4 Half-life of human ,,2.\l. Jf>A, RFH4-uf>A. and >i2M complexed n'i KDa B or RFB4-difA Component Half-life (h)" 800 a2M* 6.2 dgA 4.8 «2M-dgA 5.2 RFB4-dgA 16.5 „2M-RI-B4-dgA 14.5 " Average of 2 experiments. * The half-life of «2Mcomplexed with trypsin was 2-4 min (28).

DISCUSSION 30 The major finding to emerge from these studies is that dgA and dgA-containing immunotoxins form stable, covalent, high molecular weight complexes with <*2M, both in vitro and in 123456789 10 vivo. The complexes consist of two molecules of dgA bound to Fig. 5. Autoradiographs of T'i SDS gels of sera from mice given injections of one molecule of «2M.The binding of dgA (or ¡mmunotoxins) '!'I-dg.\: the samples of sera were taken al different intervals after injection: 1. 2. 4. 6, and 12 h (Lanes 1-5} nonrcduccd (A} and reduced (Lanes 6-ID) (fi}. to «2Mprobably occurs by a disulfide exchange reaction as reported by other sulfhydryl-containing proteins (9-11). The binding of dgA to «2Mby disulfide bonds was demonstrated 12 h after injection of '-'I-dgA or i:5I-immunotoxins. by the ability of reducing agents to dissociate the «2M-dgA As seen in Figs. 1 and 5, an additional slow-migrating radio complex and the inability of alkylated dgA to form a high active band was always present when ':5I-dgA was incubated molecular mass complex with «2M.Since the interaction be with plasma either in vitro or in vivo. The molecular weight of tween dgA (or immunotoxins) and <>2Mtakes place through a disulfide bond, only exposed —SHgroups on the «2Mmolecule this complex suggests that it may consist of fibronectin (molec ular mass, 440 kDa). which contains two free thiol groups (25). are able to interact with dgA (or dgA-containing immunotox Scanning densitometry of the autoradiographs after SDS- ins), suggesting that only the fast form of «2M(representing PAGE showed that of the total amount of dgA or immunotoxin approximately 1% of the total <*2M)(26) can interact with dgA present in the circulation after 12 h, 57.6% of the dgA and (or immunotoxins) to form complexes in vivo. The specificity 10.8% of the immunotoxin was complexed to «2M.Since the of the reaction of dgA with u2M was demonstrated by carrying clearance studies indicate that there is little difference between out the reaction of radiolabcled dgA with «2Min the presence the half-life of complexed «2Mand free «2M(Table 4), we of an excess of cold dgA. The binding of the labeled dgA was favor the explanation that the presence of noncomplexed dgA almost completely inhibited, indicating that this interaction has or immunotoxin in plasma is due to the difference between the ligand specificity. slow kinetics of complex formation versus the rapid t\/i of I25I- Since the dgA-containing immunotoxins were bound only to dgA or]2- I-RFB4-dgA in the circulation (Table 4). The smaller those «2Mmolecules containing free —SHgroups, it is possible that the dgA molecule is attached at a site on «2Mclose to the amounts of complexes formed between «2Mand RFB4-dgA bait region of the molecule which has already been "opened" (10.8%) as compared to those formed between «2Mand free by a previous reaction which generates thiol groups on «2M. dgA (57.6%) could be explained if «2Mbinds only to the dgA This possibility is supported by the lack of reactivity of dgA- which has either dissociated from an immunotoxin or which £v2Mcomplexes with anti-dgA antibody or with Blue-Sepha- undergoes disulfide exchange. rose. The blocking of both the Cibacron Blue-binding site and Toxicity of dgA Complexed with a2M. In a cell-free rabbit epitopes recognized by anti-dgA antibodies indicates that some reticulocyte assay, the capacity of dgA to inhibit protein syn portions of the dgA molecule are sterically blocked. A similar thesis was decreased after interaction with «2M. The toxic loss in the antigenic activity was reported for platelet-derived activity of free dgA was 3 times higher than that of dgA-«2M growth factor (9) and interleukin 1/i (11), after reaction with (average of 2 experiments). o2M. The blocking of the Cibacron Blue-binding site [the Cytotoxicity of RFB4-dgA Complexed with a2M. The 50% putative toxic site of ricin A chain (27)] on dgA after its reaction inhibitory concentration of RFB4-dgA-«2M for Daudi cells with «2Mcorrelates with the 3-fold decrease in toxicity of dgA was 3.5 ±1.2 x 10^" M. as compared to 1.6 ±0.2 x 10"" for in the reticulocyte assay following its reaction with a2M. There RFB4-dgA or RFB4-dgA incubated with BSA (average of 3 fore, it is not surprising that the cytotoxic activity of RFB4- experiments). Hence, interaction of RFB4-dgA with «2Mre dgA-(v2M on Daudi cells also decreased by 2-fold. These results duces its cytotoxic activity by about 2-fold, a value similar to could be explained by steric hindrance or direct competition by that observed for the inactivation of dgA by «2M.The decrease «2Mof the ribosome-binding site on dgA, or by the failure of in cytotoxic activity of RFB4-dgA by interaction with «2Mis dgA to be reduced from the complex within the cell so that it time dependent and is consistent with the time required for can translocate into the cytosol. In this regard, we were unable dgA (or dgA-immunotoxins) to bind to «2M. to dissociate dgA from the complex following treatment with Affinity of RFB4-dgA for Daudi Cells after Interaction with 10 niM dithiothrcitol for 90 min at 25°C.Similar behavior has a2M. The affinity constant of RFB4 antibodies or RFB4-dgA been reported for platelet-derived growth factor in which mi- for Daudi cells was not modified (A'a = 2.3 x 10s M~'; n = 28 togenic activity decreased 2-fold after reaction with «2M(9). X IO4molecules) after incubation at 37°Covernight with «2M The rate of reaction between dgA (or immunotoxins) and (K, = 2 x 10* M"': n = 25 x IO4 molecules), demonstrating (v2M has been determined /'// vitro by mixing radiolabeled dgA that the antibody-combining site of the immunotoxin is not with either plasma (or serum) or with purified <>2M;the tt/1 at blocked following interaction with <»2M. 37°Cfor binding was 5 h. This relatively slow reaction between 1486

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. COVALENT BINDING OF HUMAN „2MTO dgA dgA and «2Mundoubtedly explains why both free and «2M- 8. Gonias, S. L., Reynolds, J. A., and Pizzo, S. V. Physical properties of human alpha2-macroglobulin following reaction with methylamine and trypsin. bound dgA complexes are present in the blood of animals even Biochim. Biophys. Acta. 705: 306-314, 1982. at long intervals after administration of dgA. Since the half-life 9. Huang. J. S.. Huang, S. S.. and Deuel, T. F. Specific covalent binding of platelet-derived growth factor to human plasma alpha2-macroglobulin. Proc. of dgA is short (ii/2 = 4.8 h), it is not significantly changed by Nati. Acad. Sci. USA. 81: 342-346. 1984. its reaction with «2M.Hence, the majority of the injected dgA 10. Schlesinger, C., McEntire. J.. Wallman, J., Skosey, J. L.. Hanly. W., and (or immunotoxin) molecules leave the circulation without form Teodorescu. M. Covalent binding to alpha-macroglobulins of a protein with free SH groups produced by activated B cells: blocking by n-penicillamine ing a complex with «2M.Therefore, the immunotoxins bound and gold compounds. Mol. Immunol.. 26: 255-267, 1989. to target cells in vivo should retain their cytotoxic activity. At 11. Borth. W., and Luger, T. A. Identification of alpha2-macroglobulin as a later times, immunotoxins complexed with «2Mwill bind to cytokine binding plasma protein. J. Biol. Chem., 264: 5818-5825. 1989. 12. Worrell, N. R., Cumber, A. J.. Parnell. G. D.. Ross, W. C., and Forrester, J. cells, but will have moderately reduced toxicity. In the blood of A. Fate of an antibody-ricin A chain conjugate administered to normal rats. mice collected at 12 h after injection of 125I-RFB4-dgA, 11% of Biochem. Pharmacol.. 35: 417-423. 1986. 13. Fulton. R. J.. Tucker, T. F., Vitella. E. S., and Uhr. J. W. Pharmacokinetics the immunotoxin was found complexed with «2M.Since the of tumor-reactive immunotoxins in tumor-bearing mice: effect of antibody catabolic rate of both RFB4-dgA (complexed or not with «2M) valency and deglycosylation of the ricin A chain on clearance and tumor is similar, it follows that 11% is a maximum value for the localization. Cancer Res.. 48: 2618-2625. 1988. 14. Fulton. R. J.. Blakey, D. C.. Knowles, P. P., Uhr, J. W.. Thorpe. P. E.. and proportion of RFB4-dgA leaving the circulation complexed Vitella, E. S. Purification of ricin Al, A2. and B chains and characterization with liver cells and monocytes (29, 30), it is possible that translational processing of the LDL receptor and its genetic disruption in immunotoxins complexed with «2M will be transported to familial hypercholesterolemia. Cell. 30: 715-724. 1982. 19. Laemmli. U. K. Cleavage of structural proteins during the assembly of the these tissues. head of bacteriophage T4. Nature (Lond.). 227: 680-685, 1970. 20. Press, O. W., Vitella. E. S.. Farr. A. G. Hansen. J. A., and Martin, P. J. Evaluation of ricin A-chain immunoloxins directed against human T cells. ACKNOWLEDGMENTS Cell. Immunol.. 102: 10-20. 1986. 21. Ghelie. M-A., May. R. D.. Till, M., Uhr, J. W., Ghetie. V.. Knowles. P. P., Reif. M.. Brown. À.,Wallace. P. M.. Janossy, G.. Amlot. P.. Vitella, E. S., We thank C. Baselski, K. Hunter, R. Reiber, and N. Stephens for and Thorpe. P. E. Evalualion of ricin A chain-containing immunoloxins excellent secretarial assistance, and D. Jones, T. Tucker, and L. Traban dirccled against CD19 and CD22 antigens on normal and malignanl human for technical assistance. We especially thank Dr. V. Ghetie for advice B-cells as polenlial reagenls for in vivoIherapy. Cancer Res., 48: 2610-2617, and help with the experimental procedures and their interpretation. We 1988. 22. Trucco, M., and De Pétris.S.Determination of equilibrium binding param thank Dr. P. Thorpe, Imperial Cancer Research Fund, London, for eters of monoclonal antibodies specific for cell surface antigens. In: I. Lef- discussing unpublished findings which are similar to the results reported koviis. and B. Pernis (eds.). Itnmunological Melhods. Vol. 2. pp. 1-26. New in this paper. York: Academic Press. 1981. 23. Becken. W. L.. Volwiler. N.. Goldsworlhy. P. D.. Garby. L. E.. Reynolds. W. E., Slogsdill. R.. and Stemler. R. S. I13'-Albumin calabolism and dislri- bution in normal males. J. Clin. Invest., 40: 1312-1332. 1962. REFERENCES 24. Barrett, A. J.. Brown. M. A., and Sayers, C. A. The electrophoretically "slow" and "fast" forms of alpha2-macroglobulin molecule. Biochem. J., ¡SI:401- 1. Slarkey, P. M., and Barrett. A. J. In: A. J. Barrett (ed.), Proteinases in 418. 1979. Mammalian Cells and Tissues, pp. 663-696. New York: North-Holland. 25. Vartio. T. Disulfide-bonded polymerization of plasma fibronectin in the 1977. presence of metal ions. J. Biol. Chem.. 261: 9433-9437. 1986. 2. Barrett. A. J.. and Starkey, P. M. The interaction of alpha2-macroglobulin 26. Banks. R. E.. Evans. S. W.. Van Leuven, F., Alexander. D.. McMahon, M. with proteinases. Biochem. J.. 133: 709-724. 1973. J., and Whichcr, J. T. Measurement of the "fast" or complexed form of 3. Sottrup-Jensen. L.. Petersen. T. F... and Magnusson, S. A thiol-cster in alpha2 macroglobulin in biological fluids using a sandwich enzyme immu- alpha;-macroglobulin cleaved during proteinase complex formation. FEBS noassay. J. Immunol. Methods. 126: 13-20, 1990. Lett.. 121: 275-279. 1980. 27. Appukuttan. P. S.. and Bachhawat. B. K. Separalion of polypeplide chains 4. Roberts. R. C. Alpha2-macroglobulin. J. Med. (Weslbury), 16: 129-224. of ricin and the interaction of the A chain with Cibacron Blue F3GA. 1985. Biochim. Biophys. Acta. S80: 10-14, 1979. 5. Barrett. A. J. Alpha2-macroglobulin. Methods Enzymol., 80:737-754. 1981. 28. Imber, M. J.. and Pizzo, S. V. Clearance and binding of two eleclrophoretic 6. Sottrup-Jensen. L., Petersen. T. E.. and Magnusson. S. Trypsin-induced "fasi" forms of human alpha2-macroglobulin. J. Biol. Chem.. 256: 8134- activation of the thiol esters in alpha,-macroglobulin generates a short-lived intermediate ("nascent" alpha2M) that can react rapidly to incorporate not 8144. 1981. 29. Gliemann. J.. Davidsen. O.. and Moeslrup, S. K. Characlerizalion. size only methylamine or putrescine but also proteins lacking proteinase activity. eslimation and solubilizalion of iv-macroglobulin complex receptors in liver FEBS Lett., 128: 123-126, 1981. membranes. Biochim. Biophys. Acta. 980: 326-332, 1989. 7. Sottrup-Jensen. L.. Petersen, T. E.. and Magnusson. S. Mechanism of protein 30. Petersen. C. M., Ejlersen. E.. Hansen. P. W.. and Gliemann, J. Binding of complex formation with alpha2-macroglobulin. FEBS Lett.. 128: 127-132. alpha-2-macroglobulin trypsin complex lo human monocytes in culture. 1981. Scand. J. Clin. Lab Invest.. 47: 55-61. 1987.

1487

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research. Covalent Binding of Human α2-Macroglobulin to Deglycosylated Ricin A Chain and Its Immunotoxins

Maria-Ana Ghetie, Jonathan W. Uhr and Ellen S. Vitetta

Cancer Res 1991;51:1482-1487.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/51/5/1482

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/51/5/1482. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1991 American Association for Cancer Research.