Reagents: 2-Phosphoenolpyruvic acid (crystalline tricyclohexylammonium salt), guanosine triphosphate, andpyruvate kinase (muscle) A grade, were obtained ...
SYNTHESIS OF yG ANTIBODY AND IMMUNOGLOBULIN ON POLYRIBOSOMES IN A CELL-FREE SYSTEM* BY J. P. GUSDON, JR.,t A. B. STAVITSKY, AND S. A. ARMENTROUT DEPARTMENTS OF MICROBIOLOGY AND MEDICINE, WESTERN RESERVE UNIVERSITY SCHOOL OF MEDICINE, AND UNIVERSITY HOSPITALS, CLEVELAND, OHIO
Communicated by Harland G. Wood, June 19, 1967
The aim of this study was to develop a cell-free system which synthesizes specific antibody y globulin. It was assumed that polyribosomes of antibody-producing cells would be required for this synthesis.1 2 Antibody-producing cells were induced in the lymph nodes of rabbits by injections of hemocyanin from the keyhole limpet.3 More than half the cells in lymph nodes that were removed four days after a series of injections of this antigen synthesize antibodies4 and would, therefore, be expected to contain numerous polyribosomes programed for this specific synthesis. The cell-free system finally developed consisted of polyribosomes obtained from antibody-forming cells and stabilized by a polysome-rich fraction from Novikoff hepatoma cells,5 a pH 5 supernatant fraction, and an energy-generating system from rabbit reticulocytes. The Novikoff fraction was present because it was confirmed that it preserved the integrity of the lymph node cell polyribosomes.6 Evidence is presented that this cell-free system synthesized both nonantibody and antibody 7S, yG globulin molecules. Stenzel and Rubin7 recently presented chromatographic and radioimmunoelectrophoretic evidence for the synthesis of nonantibody y globulin molecules by a cell-free, microsomal fraction from the spleens of immunized rabbits. Materials and Methods.-Antigens and antisera: The hemocyanin from keyhole limpet was obtained from Pacific Bio-Marine Supply Co., Venice, Calif. Egg albumin, human serum albumin, bovine serum albumin, rabbit -y globulin, and bovine -y globulin were obtained from Pentex, Inc., Kankakee, Ill. 2,4-Dinitrophenyl-bovine y globulin (64 moles of DNP/mole of globulin) and 2,4-dinitrophenyl-human serum albumin were prepared according to Eisen.8 Rabbit antiserum to egg albumin was prepared as described elsewhere.9 Antiserum to DNPbovine globulin was produced by repeated injections of 0.8 mg of this antigen in complete Freund's adjuvant into each hind footpad of a rabbit. Rabbit antiserum to the hemocyanin was produced by several injections of alum-precipitated antigen (2 mg) into each hind footpad. Goat antiserum to rabbit serum was prepared by multiple subcutaneous injections of this serum in complete Freund's adjuvant. Likewise, sheep antiserum was produced by injection of rabbit -y globulin which was purified by chromatography on diethylaminoethyl cellulose (DEAE cellulose). This antiserum gave only one line of precipitation upon immunoelectrophoresis with whole rabbit serum and appeared to be specific for the heavy chain of the rabbit -yG globulin molecule. Reagents: 2-Phosphoenolpyruvic acid (crystalline tricyclohexylammonium salt), guanosine triphosphate, and pyruvate kinase (muscle) A grade, were obtained from California Corp. for Biochemical Research, Inc., Los Angeles. Adenosine triphosphate (ATP), dipotassium salt, was obtained from Nutritional Biochemicals Corp., Cleveland, Ohio. Thirteen C14 amino acids were obtained from Tracerlab, Inc., Waltham, Mass.: glycine (150 mc/mmole), L-leucine (300 mc/ mmole), and L-alanine, L-arginine, L-aspartic acid, L-glutamic acid, L-histidine, L-lysine, IL phenylalanine, L-proline, L-serine, ILtyrosine, and Ivaline at 200 mc/mmole. Preparation of Novikoff postnuclear fraction: Novikoff hepatoma cells (6-7 ml) were pipetted into 6 liters of 67G medium (Grand Island Biologicals, Chagrin Falls, Ohio) and incubated at 370C. After 48 hr the log-phase cells (1-2 X 106/ml) were harvested by centrifugation and washed once with cold Earle's saline. Each ml of packed cells was suspended in 4 ml of cold hypotonic buffer (0.015 M MgCl2, 0.01 M KCl, 0.01 M Tris-HCl, pH 7.4) and allowed to swell for 10 min before rupture with 12 strokes of the pestle in a Dounce homogenizer. Nuclei and cellular debris 1189
119()
BIOCHEMISTRY: GUSDON E7' .11
PROC. N. A. S.
were removed by centrifugation at 2000 rpm for 10 min at 50C. The upper four fifths of the supernatant solution were removed and an equal volume of hypotonic buffer was added for re-extraction of the precipitate. The supernates were pooled and optical density was determined at 330 ma as an index of contamination with cell debris, and at 260 m1A. Ribonucleic acid was determined by subtracting the optical density at 330 from that at 260 mJA. RNA concentration was adjusted to 60 optical density units per ml with hypotonic buffer. One- to 2-ml aliquots were stored at -70'C, at which temperature the preparation retained, for 8 months, the capacity to inhibit the breakdown of polyribosomes of lymph node cells. Preparation of polyribosomal RNA for cell-free system: The polyribosomal RNA was prepared as follows: Cells were isolated from the two popliteal lymph nodes of rabbits 4 days after the last of a series of injections at 21- to 50-day intervals, of 2 mg alum-precipitated hemocyanin into the hind footpads. The cell suspension was divided into 4-6 aliquots. Two ml of freshly thawed Novikoff postnuclear fraction and 2 ml of hypotonic buffer were added to each fraction. The cells were suspended by mixing, and deoxycholate was added to 0.5% final concentration. This mixture was then layered on a sucrose density gradient (4.0 ml 0.5 M sucrose over 3 ml of 1.8 M sucrose). After centrifugation (105,000 X g for 4 hr in a 40 head in Spinco model L ultracentrifuge) the polyribosomal pellets were suspended in 0.25 M sucrose. Samples were removed for determination of RNA by spectrophotometry and of protein.'0 The polyribosomes were preserved at - 70'C for up to 15 days. Assay of labeled antibody to hemocyanin: This assay was based oii previously described methods."' Protein(s) other than antibody to hemocyanin were removed from the culture medium or concentrated cell-free supernate by coprecipitation with an egg albumin-anti-egg albumin immune precipitate prepared at equivalence." These nonspecific coprecipitins were repeated-usually twice-until the precipitates contained 100 or fewer cpm as determined in a low background Nuclear-Chicago flow counter. To the supernate which remained, carrier hemocyanin and rabbit antihemocyanin serum were added and the radioactive antibody to hemocyanin specifically precipitated and counted. Experiments in this laboratory established that the nonspecific coprecipitation removed little or no antibody to hemocyanin."2 Comparative experiments revealed similar results when labeled antibody was assayed by this coprecipitation method and by another method which depended upon elution of antibody from an antigen-antibody complex." More recent studies of the labeled antihemocyanin antibody eluted from the hemocyanin-antihemocyanin complex upon incubation at pH 2.312 disclosed that the eluate contained antibody to hemocyanin; the eluate hemagglutinated hemocyanin-conjugated erythrocytes and was coprecipitated by hemocyanin-antihemocyanin but not by egg albumin-anti-egg albumin system. However, the pH 2.3 elution procedure apparently denatured most of the antibody because only 30% of the eluted cpm were specifically taken up by the hemocyanin-antihemocyanin system.
Results.-Requirements for immunoglobulin and antibody synthesis in cell-free system: The complete cell-free system contained polyribosomes obtained from rabbits immunized with hemocyanin. Controls included systems with polyribosomes from nodes of rabbits immunized with another antigen (bovine y globulin) as well as systems containing either Novikoff polyribosomes or polyribosomes from rabbits immunized with hemocyanin but not stabilized with Novikoff fraction. Finally, the energy requirements for synthesis were determined by omission of the energy-generating system from some of the reaction mixtures. The data in Table 1 show that the complete cell-free system containing polyribosomes from lymph node cells of hemocyanin-immunized animals, Novikoff fraction, pH 5 fraction, and the energy-generating system incorporated the greatest number of counts into antibody (expts. 1 and 2). When the Novikoff fraction was omitted, less incorporation into antibody occurred. This was not surprising as breakdown of polyribosomes occurred under these conditions. The system which lacked rabbit polysomes showed equivalent incorporation in the egg albumin-antiegg albumin and the hemocyanin-antihemocyanin precipitates. Omission of the
BOCHEMISTRY: GUSDON ET AL.
VOCL. 58, 1467 RADIOACTIV1TY Expt.
1
TABLE 1 PROTEIN AND ANTIBODY PRODUCED BY CELLFREE SYSTEMS PRESENCE AND ABSENCE OF SOURCES OR ENERGY
OF
System*
no.
Complete t Complete minus Novikoff fractiont
2'
3 4
1 191
Complete Complete minus Novikoff fraction Complete minus energy Complete minus rabbit polysomes Complete minus rabbit polysomes and energy Complete minus Novikoff fraction and energy Complete containing polyribosomes from bovine y globulin-immunized rabbit Complete minus pH 5 fraction
IN
Last egg albuminanti-egg albumin ppt (cpm)
Hemocyaninantihemocyanin ppt (cpm)
48 31 37 33 77 35 23 59 33
106 110 35 24 191 101 95 56 29 a31
21
22 8
20 7
j5moles
* The complete system'4 contained 100 ,gmoles of tris (hydroxymethyl) amino methane, pH 7.8; 10 magnesium acetate; 50 jmoles KCl; 6 pmoles mercaptoethanol; 0.25 /Amole each of 13 C14-amino acids (10 jc/IAM); 0.025 Wemole guanosine triphosphate; 5 jtmoles phosphenolpyruvate; 1 ;&mole K2 ATP; 20 fg of pyruvate kinase; 0.2-0.4 mg of pH 5 fraction; and approximately 5 mg polyribosomal RNA (for preparation see Methods) in 50-60 ml. The reaction mixture was incubated 1 hr at 371C; 0.003 M puromycin was added for the last 5 min to strip nascent polypeptides from the polyribosomes. The reaction mixture was centrifuged for 2 hr at 105,000 X g to remove ribosomal material, and the supernatant solution was dialyzed at 40C against 14 liters of saline which was changed twice in 24 hr. It was then concentrated by ultrafiltration to 5-10% of original volume. The egg albumin-anti-egg albumin and hemocyanin-antihemocyanin precipitates were prepared and counted as described in Methods. Background counts were subtracted. t Results of determinations in duplicate are shown.
energy-generating system resulted in about a 50 per cent decrease in incorporation into the hemocyanin-antihemocyanin-precipitated material, demonstrating both the dependence of incorporation upon energy and the presence of some sources of energy in the polyribosome preparation and/or pH 5 fraction. The cell-free system with polyribosomes from a rabbit immunized with bovine 'y globulin showed equivalent low incorporation into the egg albumin and hemocyanin precipitates (expt. 3). The latter system also showed approximately equivalent incorporation into egg albumin-anti-egg albumin and bovine y globulin-antibovine y globulin precipitates. When pH 5 fraction was omitted, the system did not incorporate amino acids into trichloroacetic acid (TCA)- precipitable protein or antibody (expt. 4). It was noted that omission of this fraction resulted in breakdown of the polyribosomes. Immunoelectrophoretic and radioimmunoelectrophoretic evidence that -y globulin and specific antibody are products of cell-free system: For immunoelectrophoretic analysis the concentrated supernate of the complete cell-free system was electrophoresed in agar."5 Enough unlabeled antihemocyanin antibody was added to the supernate in the well before electrophoresis so that visible precipitation occurred in agar when hemocyanin was added to a trough in the agar. The line of specific antigen-antibody precipitate was then excised and counted. Only 3-4 1Al of supernate-unlabeled antibody mixture was electrophoresed; therefore, the excised lines contained minimal radioactivity and had to be counted 900 minutes to attain the 97 per cent confidence level. Table 2 presents typical results. Only the line corresponding to the precipitate of protein with hemocyanin and with a mobility of yG globulin contained a significant number of counts over the control. When bovine My globulin was placed in the trough, it did not precipitate with the electrophoresed supernate from the system containing polyribosomes from hemocyanin-immunized animals. Counts equal to background were obtained upon
1192
BIOCHEMISTRY: GUSDON ET AL.
TABLE 2 IMMUNOELECTROPHORETIC EVIDENCE OF THE SYNTHESIS ANTIBODY BY THE CELL-FREE SYSTEM
PRoc. N. A. S.
OF
SPECIFIC
Source of cell-free supernate
Cpm*
Complete system Complete system minus rabbit polysomes Controlt
54 31 28
A mixture containing 20% by volume of rabbit antiliemocyanin serum and 80% concentrated cell-free supernate was electrophoresed in agar. Two mg hemocyanin was then added to the trough to develop a line of precipitation with antibody. After the slides were washed and dried, the lines of precipitate were excised, dissolved in formic acid, added to 15 ml of scintillating fluid (200 ml methanol, 125 gm naphthalene, 7.5 gm of PP0, and 0.75 gm POPOP in 1 liter of p-dioxane), and counted in a Packard liquid scintillation counter. * Counted for 900 min. t A portion of agar other than the area of antigen-antibody precipitate was excised and counted (see above).
excision and counting of the y globulin region in which precipitation would have occurred, if both antigen and specific antibody had been present. These results thus provided evidence of the synthesis of a protein by the cellfree system with the mobility of -yG globulin and the specific reactivity of an antibody to hemocyanin. Radioactivity of the precipitate produced by reaction of the cell-free supernate with a sheep antirabbit y globulin serum specific for the heavy chain of yG globulin16 provided additional evidence that the system synthesized this type of globulin. The concentrated supernates from the complete cell-free system were also subjected to radioimmunoelectrophoretic analysis.17 The procedure was similar to that employed for the immunoelectrophoresis experiments with the exceptions that carrier, unlabeled antihemocyanin antibody was not added to the concentrates before electrophoresis. After electrophoresis of the supernate, hemocyanin was added to the slide troughs and then X-ray film was exposed to these slides 1-6 weeks. Antirabbit y globulin serum or antirabbit serum was placed in the troughs of other slides and films were similarly exposed. Typical results are illustrated in Figure 1. Film that was exposed 5-6 weeks to slides containing hemocyanin in the trough developed a short, distinct arc attributable to the reaction with hemocyanin of a 7G radioactive protein. Films exposed 1-2 weeks to slides containing either antirabbit y globulin or antirabbit serum in the slot developed a similarly placed arc owing to the reaction with antirabbit y globulin serum of a radioactive protein with the mobility and antigenic characteristics of a 'YG globulin. An addiA
If
C
v
FIG. 1.-Radioimmunoelectrophoretic evidence for synthesis of 7 globulin antibody and nonantibody -y globulin in cell-free system. The procedures indicated under Table 2 were employed. After electrophoresis of the concentrate, hemocyanin (2 mg/ml) was added to the trough of A, sheep antirabbit -y globulin to the trough of B, and goat antirabbit serum to the trough of C. After development of the lines of precipitation and washing of the slides, the slides were overlaid with X-ray film and exposed for varying periods.17 The line shown on slide A appeared after 6 weeks of incubation and those in slides B and C after 2 weeks of incubation.
VOL. 58, 1967
BIOCHEMISTRY: GUSDON ET AL.
1193
tional faint line was visible in the a or 1B globulin region in some slides. Control slides employing supernate from reactions containing only Novikoff polysomes did not develop any -y globulin arcs. Nor did such arcs appear upon similar analysis with hemocyanin in the trough of supernates from systems which employed polyribosomes from bovine Sy globulin-immunized animals. The requirement for an exposure of 5-6 weeks to label the -yG hemocyanin antibody line compared to 1-2 weeks to label the -yG line not associated with hemocyanin antibody activity indicated that much more yG globulin which did not react with hemocyanin than antibody yG globulin was synthesized in the cellfree system. Chromatography of supernate from complete cell-free system: Table 3 reveals that upon chromatography of the supernates from the complete cell-free system, only fractions 2 and 3, which contain yG, 7S globulins, contained radioactivity above that of the system from which the lymph node cell polysomes were omitted. There was some suggestion of radioactivity above the control in tube 5 which contains YM, 19S macroglobulins, but upon radioautographic analysis, a line corresponding to yM globulin did not appear (see Fig. 2). Molecular size of molecules labeled in complete cell-free system: Although the previous experiments indicated that the complete cell-free system labeled molecules with the chromatographic, antigenic, electrophoretic, and antigen-combining characteristics of a -yG, 7S globulin molecule, it was conceivable that this system synthesized half-molecules19 20 containing one light and one heavy chain of this molecule which would not have been distinguished by any of the previous methods. Therefore, an experiment was done to determine the size of the molecule synthesized by the cell-free system, which combined with antigen. The products of synthesis were separated in a sucrose density gradient in the presence of an unlabeled 7S, ,yG globulin antibody to DNP-bovine globulin. The resulting fractions were then analyzed for radioactivity in the TCA precipitate, in the last egg albuminanti-egg albumin precipitate, and in the hemocyanin-antihemocyanin precipitate and for hemagglutinating 7S, yG anti-DNP antibody. Figure 2 indicates that the position of the labeled specific antihemocyanin molecule (specific precipitate) corresponded moderately well to that of the marker 7S, 'yG globulin (hemagglutination titer). TCA-precipitable counts were associated both with 7S and lighter than 7S fractions and were more numerous than those in specific antibody. Thus it TABLE 3 RADIOACTIVITY OF 7S AND 19S GLOBULIN FRACTIONS PREPARED BY DEAE-CELLULOSE CHROMATOGRAPHY FROM CELL-FREE SUPERNATES Complete system minus rabbit Complete Phosphate Buffer Used for Elution Vol system polysomes Tube no. M pH (cpm) (ml) (cpm) 180 190 6 1 0.02 6.3 285 204 12 2 0.02 6.3 224 195 15 5.4 0.01 3 189 200 4.8 5 4 0.05 220 10 177 4.8 5 0.05 The method employed was a modification of one previously described.'8 One half of concentrated cell-free supernate was dialyzed against 0.02 M, pH 6.3, sodium phosphate buffer and then layered on a 1- X 10-cm column of DEAE-cellulose already equilibrated with this buffer. Elution was done as shown in the table. Known 7S, oy globulin antibody appeared in tubes 2 and 3 and 19S globulin in tube 5. Each fraction was lyophilized, dissolved in formic acid, and added to scintillation fluid for counting in a liquid scintillation counter.
BIOCHEMISTRY: GUSDON
1194
FIG. 2.-Demonstration of labeled 78 antibody to hemocyanin, and 7S and lighter than 7S proteins among the products of the cell-free system. 0.45 ml concentrated supernate from a pulse-labeled
cr
.-* _o0*w
*
Z
I
0
'
HEMAGGLUTINATION NON SPECIFIC
TITER
PRECIPITATE
PRECIPITATE
TCA
PRECIPITATE
PROC. N. A. S.-
ET AL.
20
cell-free system plus 0.05 ml of unlabeled 7S globulin antibody
anti-DNP-bovne lineai gradient,1" and spun in a SW-39 rotor in a were layered on a
10-40%
sucrose
I o L for 16 hr.
