/. Biochem., 78, 1225-1233 (1975)
A Poly some-Membrane Binding System from Rat Liver I. Basic Characterization of the Binding System1 Masamichi TAKAGI2 and Mahlon B. HOAGLAND Worcester Foundation for Experimental Biology, 222 Maple Avenue, Shrewsbury, Mass. 01545, U.S.A. Received for publication, May 28, 1975
A simple reaction system was developed to examine the binding of polysomes to membranes of the endoplasmic reticulum and to investigate the fate of ribosomes and nascent chains during protein synthesis in vitro. The system consisted of Sephadex G-25 treated post-mitochondrial fraction prepared from rat liver (SephadexPM) as a source of membranes, and radioactive free polysomes prepared from another rat liver. The following results were obtained. 1. Nascent chains on free polysomes labeled in vivo were transferred to membranes in vitro. The process required protein synthesis. 2. This reaction occurred in two steps: a) Binding of the free polysomes to membranes in the absence of protein synthesis, b) Release of ribosomes, leaving nascent chains on the membranes, requiring protein synthesis. 3. A portion of the ribosomes found on membranes in vivo (membrane-bound ribosomes) was also released from the membranes during incubation in vitro, leaving their nascent chains on the membranes. The significance of the transfer of nascent chains from free polysomes to membranes in vitro is discussed in the light of known polysome-membrane interaction in vivo.
In many eukaryotic cells, there are two kinds of polysomes: those free in the cytoplasm and those bound to membranes of the endoplasmic reticulum. Evidence has accumulated indicating that free and membrane-bound polysomes * This work was performed with the support of a Cancer Center Support Grant («. 12708) from the National Cancer Institute of the United States to the Worcester Foundation for Experimental Biology. * Present address: Laboratory of Radiation Genetics, Faculty of Agriculture, The University of Tokyo, Tokyo, Japan. Abbreviations: DTT, dithiothreitol; HEPES, N-2hydroxymethylpiperazine-N-2-ethanesulfonic acid. Vol. 78, No. 6, 1975
synthesize different kinds of proteins. Some soluble cellular proteins have been shown to be synthesized by free polysomes (1—3) and secretory proteins by membrane-bound polysomes (/, 2, 4—6). It has also been suggested that membrane-bound polysomes are involved in the On the s y n t h e s i s o f membrane proteins (7). example Qther hand> some proteinS) for NADPH -cytOChrome c reductase [EC 1. 6. 99.1] . „ , cl , m . < 8< 9 >• C a t a l a s e fEC L 1 L L 6 ] < 10 >• a n d S e n n e dehydratase [EC 4.2.1.13] {11), have been found to be synthesized by both free and membrane-bound polysomes. A major secretory protein, serum albumin, has been detected
1225
M. TAKAGI and M.B. HOAGLANI>
1226
consistently on free polysomes (77). Furthermore, messenger RNA's for serum albumin and ferritin have been isolated from both kinds of polysomes (12). These results suggest the possibility that polysomes producing intracellular proteins on the one hand and export proteins on the other may be in dynamic equilibrium between free and membrane-bound states. If such a situation exists, it may be expected that some fraction of the free polysome population is a precursor to membranebound polysomes. To clarify this, we have investigated the binding of polysomes to the membrane in vitro and the fate of ribosomes and nascent chains after completion of the peptide chains. Polysome-membrane binding systems in vitro have been reported by several groups (13—19). Among them, the system reported by the present authors has the advantages of simplicity and directness. It was also useful in detecting certain qualitative changes in the polysome binding capacity of membranes of the endoplasmic reticulum, as described previously. The purpose of the investigation described in this paper was to characterize the binding system in detail. MATERIALS AND METHODS Materials — CsCl (optical grade), ribonuclease-free sucrose and [6-uC]orotic acid (42 mCi/mmole) were purchased from Schwarz/ mann. [4, 5-3H]Leucine was from New England Nuclear. ATP, GTP, phosphoenolpyruvate and pyruvate kinase [EC 2.7.1.40], DTT and HEPES were purchased from Sigma. Animals—Female rats, COBS.CD strain, 2—3 months old, were used after starvation for 16-20 hr. To label ribosomal RNA uniformly, each rat was injected with [14C]orotic acid intraperitoneally about 20 hr before sacrifice. To label the nascent chains on polysomes, 200 ^Ci of [3H]leucine in 0.1 ml of saline solution was injected into a rat through the portal vein 13 min before sacrifice (2). Cell Fractionation — Preparation of the post-mitochodrial fraction and free polysomes was as described previously (18). In brief, livers were homogenized in 0.25 M sucrose in
TKM (50 mM Tris-HCl, pH 7.6, 25 mM KC1, 5 mM MgCl2), then centrifuged at 15,000 Xg for 10 min. The supernatant, the post-mitochondrial fraction, was put on 1.8 M sucrose1.0 M sucrose double layers and centrifuged at 40,000 rpm for 2 hr in a SW 50.1 Spinco rotor. Free polysomes were obtained as a pellet. In some experiments, a turbid zone between the 1.8 M and 1.0 M sucrose layers was taken up, homogenized in a small volume of the incubation buffer and used as the crude membranebound fraction. Incubation In Vitro—Sephadex-treated postmitochondrialfraction (20) (Sephadex-PM) wasprepared as desribed in the previous paper (18). It was incubated in the presence of either labeled free polysomes or the crude membranebound polysome fraction. The standard reaction mixture contained 50 mM HEPES-KOH, pH 7.6, 70 mM KC1, 5 mM MgCl2, and 4 mM DTT. In some of the reaction mixtures, a. mixture of components required for protein, synthesis (referred to as "P") was added: 2 mM ATP, 0.4 mM GTP, 10 mM phosphoenolpyruvate, and 50 ftg/ml of pyruvate kinase. Centrifugal Analysis of the In Vitro Interaction of Polysomes and Nascent Chains with Membranes—The interaction of ribosomes and nascent chains of free polysomes with membranes was detected by either discontinuous, sucrose density or CsCl-sucrose gradient centrifugation. In the former method, each reaction-mixture was put on a discontinuous sucrose gradient consisting of 3 layers, 1.5 ml. each, of sucrose solution, 1.8, 1.0, and 0.25 M, all in TKM. After centrifugation for 16 hr at. 40,000 rpm, each tube was fractionated into 5 fractions. In the latter method, the reaction, mixture was first fixed with 5% formaldehyde, put on a CsCl-sucrose density gradient, centrifuged and fractionated as described previously (18). Trichloroacetic acid-insoluble radioactive material of each fraction was collected on a glass filter disc (GF/C, Whatman), solubilized in 1 ml of Protosol (New England Nuclear) and counted in toluene scintillator fluid. RESULTS Protein Synthesizing Activity and the Ef— /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. 1
fects of Inhibitors—Sephadex-PM, which was used as a source of membranes in the polysome-membrane binding system, was active in •amino acid incorporation, as described by Richardson et al. (20). The incorporation reached a plateau after incubation for 15 min at 37°. As shown in Table I, it depended on "P." ATP alone was only partially effective. Inhibitors of protein synthesis, puromycin (1 mM), cycloheximide (1 mM), and pactamycin (0.1 xnM), inhibited incorporation, while a low concentration of pactamycin (0.004 mM) had little effect. These results suggest that peptide chain elongation on pre-existing polysomes occurs in this system and that initiation of protein synthesis is negligible (21). Protein Synthesis-dependent Transfer of Nascent Peptides from Free Polysomes to Membranes—In our previous paper, it was shown that Sephadex-PM was active as a source of membranes in interacting with exogenously added free polysomes (18). As Sephadex-PM was active in protein synthesis (Table I), we wished to follow the fate of nascent peptides on free polysomes as they interacted with membranes and to assess the effect of protein synthesis in the reaction. Free polysomes whose nascent chains had been labeled with [3H]leucine in vivo were incubated with Sephadex-PM and the distribution of [3H]radioTABLE I. Effect of inhibitors on the incorporation of [3H]leucine by Sephadex-PM into the acid-insoluble fraction. Sephadex-PM (2.4 rag of protein) was incubated with 10 /iCi of [3H]leucine in a final volume of 0.5 ml. The complete reaction mixture contained " P " in the standard reaction mixture. After incubation for 30 min at 37°, 50 p\ was removed from each of the reaction tubes and trichloroacetic acidinsoluble radioactivity was counted. 1. 2. 3. 4. 5. 6. 7.
Complete (37°) Complete (0°) 1-"P" 3+ATP (2mM) 1+puromycin (lmiu) 1+cycloheximide (1 mM) 1+pactamycin (0.1 mM)
8. 1+pactamycin (0.004 mM)
Vol. 78, No. 6. 1975
100% (8218 cpm) 2% 6% 21% 17% 43% 25% 91%
1227
activity between free polysomes and membrane fractions was assayed by discontinuous sucrose density gradient centrifugation (Fig. 1). Fraction No. 3 (a turbid zone between the 1.8 and 1.0 M sucrose layers) represented the main membrane fraction, and fraction No. 5 (a pellet) contained the free ribosomes (plus free 100 A
B
50
«_•_•_•_
.11
1 2 3 A 5 Number
Fig.' 1. Protein synthesis-dependent transfer of nascent chains from free polysomes to membranes (I). A rat was injected with' 200 fiCi of [3H]leucine through the portal vein. After 2.5 min, the liver was removed and free polysomes were prepared. Sephadex-PM was prepared from another rat without radioactivity. Sephadex-PM (5 mg of protein) and free polysomes (274 fig of RNA=672 cpm) were incubated in a final volume of 1 ml. The complete reaction mixture was the same as that given in the legend to Table I. After incubation at 0° or 37°, 0.6 ml of each reaction mixture was put on a discontinuous sucrose gradient as described in " MATERIALS AND METHODS." Each gradient, after centrifugation for 16 hr at 40,000 rpm, was fractionated into 5 fractions: from the top, 1) 1.5 ml, 2) 1.1 ml (the soluble protein fraction), 3) 1.2 ml (the turbid zone or main membrane fraction), 4) 1.1 ml, and 5) the pellet (the free ribosome fraction). The trichloroacetic acid-insoluble radioactivity of each fraction was counted. A) complete at 0° for 30 min, B) complete minus " P " at 37°, C) complete at 37° for 10 min, D) complete at 37° for 30 min.
1228
M. TAKAGI and M.B. HOAGLAND
polysomes when protein synthesis did not occur). Fraction No. 4 could be a mixture of both. When the incubation was carried out at 0° or at 37° without "P," i.e., where protein synthesis did not occur (cf. Table I), most of the radioactivity was recovered in the free ribosome fraction, fraction No. 5 (Fig. 1-A, 1-B). In the presence of "P" at 37°, a substantial part of the labeled nascent peptides, originally on exogenously added free polysomes, was found on membranes (Fig. 1-C, 1-D), indicating that the transfer of nascent chains on free polysomes to membranes depended on protein synthesis. In the next experiment, free polysomes were labeled doubly; the RNA was labeled uniformly with [uC]orotic acid and the nascent chains with [3H]leucine, in order to determine the location of both ribosomes and nascent chains after incubation. As shown in Fig. 2-A, neither ribosomes nor nascent chains were transferred from free polysomes to membranes
100
3
2
3
4
5
Fraction Number
Fig. 2. Transfer of nascent chains from free polysomes to membranes and release of ribosomes from the membranes. Ten //Ci of [l4C]orotic acid was injected into a rat 48 hr before sacrifice and 200 ftC\ of [3H]leucine was then injected through the portal vein 2.5 min before sacrifice. Free polysomes were prepared from the liver and incubated with SephadexPM prepared from a non-radioactive rat liver. One ml of the reaction mixture contained 4.8 mg of Sephadex-PM protein, 166 fig of RNA of free polysomes ( U C, 1448 cpm; 3H, 3720 cpm), and " P " in the standard reaction mixture. After incubation for 20 min, 0.6 ml of each reaction mixture was put on a discontinuous sucrose gradient. See the legend to Fig. 1 for other experimental conditions. A) 0°, B) 37°. Black bars, [UC]RNA radioactivity; and white bars, [3H]peptide radioactivity-
in the absence of protein synthesis. When, protein synthesis occurred, on the other hand (Fig. 2-B), a substantial amount of 3H-labeled nascent chains was transferred from free polysomes to membranes, as expected from the results shown in Fig. 1. However, most of the [MC]orotic acid-labeled ribosomes were not bound to membranes, but were free (Fig. 2-B). This result suggested that when protein synthesis occurs, free polysomes interact with membranes, leave their nascent peptides on the membranes and release ribosomes from the. membranes into a free state. This result could also have been obtained if the nascent chains on free polysomes had simply been released into the soluble protein fraction, and subsequently bound non-specifically to the membranes. This was ruled out by experiments (Fig. 3) which showed that radioactive soluble proteins incubated under conditions which promote the transfer of nascent, chains from polysomes to membranes do not bind to membranes. The transfer of nascent, chains from free polysomes to membranes when, protein synthesis is occurring may be detected more quantitatively by analysis of the reaction mixture by CsCl-sucrose density gradient centrifugation after fixing polysomes and polysome-membrane complexes with formaldehyde. This technique had been successfully used in our previous work (18) to separate free polysomes from membrane-bound polysomes. Tworeaction mixtures were incubated at 37°, one. in the presence of "P" (Fig. 4-A) and the other in the absence of "P" (Fig. 4-B). It can be seen that a substantial fraction of the radioactivity was found in the membrane fractions, (fraction No. 8—13) only when the incubation was carried out under conditions where protein synthesis took place. In subsequent experiments, this analytical method was used in place of discontinuous sucrose gradient centrifugation. Separation of the Reaction into Two Steps—The results so far described suggested that in our reaction system, exogenously added free polysomes interacted with membranes and, if protein synthesis took place, completed nascent chains were left on the membranes and ribosomes were released. If this is indeed the case,. /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM.
I
1229
100
5
-
IOO
c
D
50 -
1 1.. I
1 1.
I 2 3 4 5 2 3 4 5 Fraction Number
Fig. 3. Non-binding of soluble protein to membranes. Sephadex-PM (2.34 mg of protein) was incubated with either 60 fig of RNA of [3H]leucinelabeled free polysomes (A and B) prepared as described in the legend to Fig. 1, or 134 ftg of protein of [3H]leucine-labeled soluble protein fraction (C and D). The latter was obtained as a supernatant after centrifuging the post-mitochondrial fraction of [3H]leucine-labeled (200 ftCi injected 4.5 hr before sacrifice) rat liver for 2 hr at 40,000 rpm. The reaction mixture contained " P " and the final volume was 0.5 ml. After incubation for 30 min, transfer of nascent chains from free polysomes and binding of soluble protein to membranes were assayed by discontinuous sucrose gradient analysis as described in the legend to Fig. 1. A: Incubation with [3H]leucine-labeled free polysomes at 0°. B: The same as A, but at 37°. C: Incubation with [3H]leucinelabeled soluble protein fraction at 0°. D: The same as C, but at 37°. the reaction should be separable into two steps; (1) binding of polysomes to membranes and (2) release of ribosomes from the membranes leaving completed nascent chains on the membranes. The time course of binding of free polysomes to membranes of Sephadex-PM in the absence of protein synthesis (incubation without "P") is presented in Fig. 5. (In our previous work, ethanol was found to enhance Vol. 78, No. 6, 1975
10
15 O Fraction Number
5
10
15
Fig. 4. Protein synthesis-dependent transfer of nascent chains from free polysomes to membranes (II). The reaction systems were as described in the legend to Fig. 1. An aliquot (0.32 ml) of each of the reaction mixtures was removed after incubation for 35 min, fixed with formaldehyde and then put on a CsCl-sucrose gradint as described in " MATERIALS AND METHODS." Each gradient was centrifuged for 16 hr at 41,000 rpm, fractionated into 17 fractions and trichloroacetic acid-insoluble radioactivity was counted. A: complete at 37°. B: complete minus " P " at 37°.
20 40 60 80 Time of Incubation (min)
Fig. 5. Time course of binding of free polysomes to membranes. Sephadex-PM (2 mg of protein) was incubated with 40 fig of RNA of uC-labeled free polysomes at 23° in the absence of " P " for 0, 10, 40, or 90 min in the presence of 1.25% ethanol, which enhances binding {18). After incubation, each reaction mixture was analyzed by CsCl-sucrose gradient centrifugation as described in the legend to Fig. 4. The distribution of the added radioactivity into the free polysome (fraction No. 1-7) and membrane (fraction No. 8-13) fractions was determined in each reaction mixture. • , free polysome fraction ; x , membrane fraction. polysome binding (18). It was used in this and subsequent experiments to maximize polysome binding in the system.) The polysome
1230
M. TAKAGI and M.B. HOAGLAND
binding reaction reached a plateau after incubation for 40 min at 23°. In the next set of experiments, free polysomes were first incubated with Sephadex-PM in the absence of "P" for 40 min at 23° to complete polysome binding. The "P" was added and incubation continued to allow completion and release of nascent chains. (The second incubation was for 10 min at 37°.) During the two successive incubations, one tube contained 1.25% ethanol, the other did not. Ethanol enhances the binding of free polysomes to membranes in the first incubation (18). Before and after the second incubation, an aliquot of each of the reaction mixtures was removed and analyzed by CsCl-sucrose gradient centrifugation. The results are shown in Fig. 6. After the first incubation (the binding reaction), a small fraction (15—20%) of [uC]ribosomes in free polysomes together with [8H]nascent chains was found to be bound to
105
A
-
50
0
B
rfl.
the membranes (Fig. 6-A). Ethanol enhanced this binding (Fig. 6-B) as expected from our previous work (18). After the second incubation (the transfer reaction), about 40% of [3H]nascent chains had been transferred from free polysomes to membranes, while less than 10% of the [uC]ribosomes in free polysomes were found in the membrane fraction (Fig. 6-C). Even in the presence of ethanol, only 10% of ["CJribosomes remained on the membranes, leaving more than 50% of [3H]nascent chains on the membranes. The time course of transfer of nascent chains from polysomes in the presence or absence of ethanol is shown in Fig. 7. These results lead us to conclude that: (1) during the first incubation (the binding reaction), nascent chains on polysomes were bound to the membranes. This binding reaction was enhanced by ethanol; (2) during the second incubation (the transfer reaction), nas-
c
D
-1
[r
in •
too
50
1-
Fig. 6. Protein synthesis-dependent transfer of nascent chains of free polysomes to membranes and the release of ribosomes from membranes. Sephadex-PM (2.8 mg of protein) and doubly labeled free polysomes (69 fig of RNA) prepared as described in the legend to Fig. 2, were incubated in the absence of " P " with or without 1.25% ethanol. After 50 min at 23°, one-half of each reaction mixture was removed (A and B) and to the other half, " P " was added. Further incubation was carried out at 37° for 10 min (C and D). These four samples were analyzed by CsCl-sucrose gradient centrifugation and fractionated as described in the legend to Fig. 4. The results are summarized showing the percentage distribution of total radioactivity in the first 7 fractions (L; lower fraction =free polysomes fraction), fractions 8 to 13 (M; middle fraction=rough membrane fraction), and the top four fractions (T; top fraction =smooth membrane and soluble fraction). A) 23° for 50 min without ethanol, B) the same as A, but with ethanol (1.25%), C) the same as A and then 37° for 10 min in the presence of " P " , D) the same as B and then 37° for 10 min in the presence of " P . " Black bars, [UC]RNA radioactivity and white bars, [3H]peptide radioactivity.
/ . Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. I
1231
cent chains of polysomes on the membranes were completed and remained on the membranes, while ribosomes (so-called "run off" ribosomes) left the membranes to enter the free fraction. Sucrose density gradient analysis showed that the polysomes were degraded to monosomes during protein synthesis in this reaction system (data not shown). Incubation of Membrance-bound Polysomes in the Reaction System—To determine the fate of nascent chains and ribosomes of indigenous membrane bound polysomes in our system dur-
ing protein synthesis, a crude membrane-bound polysome fraction, whose nascent chains and ribosomes were doubly labeled, was incubated with Sephadex-PM and the protein synthesizing system. Before and after incubation, an aliquot was removed and separated by centrifugation to analyze the distribution of nascent chains and ribosomes between the free and membrane-bound fractions. The results are presented in Table II. As the membrane-
3
Fig. 7. Time course of transfer of nascent chains from free polysomes to membranes after the binding of free polysomes to membranes in the presence or absence of ethanol. Sephadex-PM (8.4 mg of protein) and free polysomes (207 fig of RNA) prepared from rat liver labeled with [3H]leucine for 2.5 min were incubated in a final volume of 1.5 ml in the presence or absence of ethanol, as described in the legend to Fig. 6. After incubation for 50 min at 23°, " P " was added to both reaction mixtures. At 0, 3, 6, and 10 min after the addition, 0.32 ml of the reaction mixture was removed and analyzed by CsCl-sucrose gradient centrifugation. The percentage distribution of 3H-radioactivity in the free polysome and rough membrane fractions was calculated as shown in the legend to Fig. 6. O, free polysome fraction (without ethanol); A, free polysome fraction (with ethanol); • , rough membrane fraction (without ethanol); • , rough membrane fraction (with ethanol).
6 9 Time (min)
TABLE II. Release of membrane-bound ribosomes from membranes leaving nascent chains on the membranes. A rat was doubly labeled with [uC]orotic acid and [3H]leucine as described in the legend to Fig. 2. The crude membrane-bound polysome fraction (68 fg of protein) prepared from the rat was incubated with Sephadex-PM (2 mg of protein) prepared from another rat liver without radioactivity in the presence or absence of 1.25% ethanol. After incubation for 90 min at 23°, half of the mixture was removed from both the tubes and fixed with formaldehyde. To the residual half of each of the reaction tubes, " P " was added and a second incubation was carried out at 37°. After 20 min, the reaction mixtures were fixed with formaldehyde and analyzed in CsCl-sucrose gradients. The percentage distribution of 3H- and "C-radioactivity in free polysomes and rough membrane (=R-ER) fractions was calculated (see the legend to Fig. 6). % Distribution of the radioactivity 3
Conditions of incubation
1. 2. 3. 4.
23°, 90 min 1-t-ethanol 1, then plus " P " , 37°, 20 min 3+ethanol
Vol. 78, No. 6, 1975
H
Free polysomes
R-ER
Free polysomes
R-ER
9
76
43
47
12
74
46
48
12
71
63
29
11
68
61
28
1232
M. TAKAGI and M.B. HOAGLAND
bound polysome fraction used- was crude and possibly contained free cytoplasmic particles such as ribosomal subunits, a substantial portion of 14C-radioactivity of the reaction mixture, even before incubation, was found in the free ribosome fraction by analytical centrifugation. After the incubation, about one-third of membrane-bound ribosomes were released from the membranes into a free state, while nascent proteins on the membranes remained in situ during the incubation. Although release of ribosomes from the membranes was not as clear as in the case of the in znYro-constituted complex, this result suggests that the fates of nascent chains and ribosomes are similar in the in t>?7«)-constituted polysome-membrane complex and "natural" membrane-bound polysomes extracted from the cells. DISCUSSION
The general statement that free polysomes synthesize protein for intracellular use and membrane-bound polysomes synthesize protein for export no longer seems tenable. It would appear, as noted in the introduction, that each class of polysomes is capable of synthesizing both kinds of protein, although the relative amount of each kind of protein synthesized is variable (2, 8—11). Some of the observations leading to this conclusion might be artifacts of fractionation, i.e., contamination of membrane-bound polysomes by free polysomes as suggested by Cioli and Lennox (22), detachment of polysomes from membranes during isolation (23, 24), or trapping of free polysomes in membrane-rich fractions during centrifugation (25). We favor the hypothesis that free and membrane-bound polysomes are in dynamic equilibrium with each other. This equilibrium could be influenced by the character of the nascent chains on polysomes. Thus, for example, polysomes festooned with nascent albumin have a higher affinity for membranes than polysomes bearing nascent ferritin ( 1 , 2, 4—6). The equilibrium would also presumably be affected by the character of the endoplasmic reticulum. This might explain why, in hepatoma cells, nascent albumin is found mainly on free polysomes (26).
We have already reported that physiological changes induced in liver — e.g. by phenobarbital treatment or regeneration — alter the extent to which polysomes can bind to membranes in vitro (18). Furthermore, we have found that 3-methylcholanthrene treatment of a rat increases the polysome binding capacity of the membranes dramatically (27). The study of such in vitro binding reactions wilL hopefully offer clues to the regulation of the putative free polysome-bound polysome equilibrium in vivo. Membranes are readily denatured during repeated centrifugation and resuspension (28, 29) and our use of Sephadex-PM minimized this danger, thereby hopefully providing a more physiological system for the study of polysome-membrane dynamics. It also has the advantage of being active in protein synthesis. The obvious disadvantage of the system is its crudity as a mixture of several different kinds of membranes and other cellular macromolecular contaminants. Our results suggest that free polysomes not only react with membranes but they deposit their nascent chains in the membranes and then depart (Figs. 6 and 7). This stage of the reaction requires protein synthesis. Bound polysomes also participate in the same type of reaction in vitro (Table II). Recently, Borgese et al. (19) have shown that most of the nascent chains transferred from polysomes to membranes in vitro are sensitive to proteolytic enzymes. They interpreted this result as indicating that reconstituted complexes made from membranes and polysomes in their system might be adsorption artifacts and not physiological products. However, Negishi et al. (30) have reported that nascent NADPH-cytochrome c reductase found in membrane-bound polysomes was sensitive to proteolytic enzymes. On the other hand, nascent serum albumin was found to be resistant to the same treatment. Considering the fact that, in rat liver, a part of the membranebound polysome population is loosely bound while the other part is tightly bound (31), we may speculate that only those polysomes which are loosely bound to membranes are in some dynamic equilibrium with free polysomes in /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. I
1233
the cytoplasm. Our results lead us to conclude that, since protein synthesis causes nascent chains to remain bound to membranes, the binding of polysomes to membranes in our in vitro system is through nascent chains on polysomes and not through ribosomes. Other investigators have reported the same kinds of binding using high salt-puromycin treatment of reconstituted complexs (32, 33). The use of our easily prepared system may help to shed light on membrane-polysome relationships as they relate to major physiological changes, as we have reported preliminarily previously (18). Further details will be reported in a subsequent paper (27).
14. Sunshine, G.H., Williams, D.J., & Rabin, B.R. (1971) Nature New Biol. 230, 133-136 15. Ragland, W.R., Shires, T.K., & Pitot, H.C. (1971) Biochem. J. 121, 271-278 16. Shires, T.K., Narurkar, L., & Pitot, H.C. (1971) Biochem. J. 125, 67-79 17. Khawaja, J.A. (1971) Biochim. Biophys. Ada 254, 117-128 18. Takagi, M. & Hoagland, M.B. (1974) Biochem. Biophys. Res. Commun. 58, 868-875 19. Borgese, N., Mok, W., Kreibich, G., & Sabatini, D.D. (1974) / . Mol. Biol. 88, 559-580 20. Richardson, A., McGown, E., Henderson, L.M., & Swan, P.B. (1971) Biochim. Biophys. Ada 254, 468-477 21. Stewart-Blair, M.L., Yanowity, I.S., & Goldberg, I.H. (1971) Biochemistry 10, 4198-4206 22. Cioli, D. & Lennox, E.S. (1973) Biochemistry 12, 3211-3217 23. Bont, W.S., Geels, J., Huizinga, A., Mekkekhlt, K., & Emmelot, P. (1972) Biochim. Biophys. Ada 262, 514-524 24. O'Tool, K. (1974) Biochem. J. 138, 305-307 25. Vernie, I.N., Bont, W.S., & Emmelot, P. (1971) Cancer Res. 31, 2189-2195 26. Uenoyama, K. & Ono, T. (1972) Biochim. Biophys. Ada 281, 124-129 27. Takagi, M., submitted to / . Biochem. 28. Dallner, G. & Nilsson, R. (1966) / . Cell Biol. 31, 181-193 29. Glaumann, H. & Dallner, G. (1970) / . Cell Biol. 47, 34-48 30. Negishi, M., Sawamura, T., Morimoto, T., & Tashiro, Y. (1975) Biochim. Biophys. Ada 381, 215-220 31. Tanaka, T. & Ogata, K. (1972) Biochem. Biophys. Res. Commun. 49, 1069-1074 32. Adelman, M.R., Sabatini, D.D., & Blobel, G. (1973) / . Cell Biol. 56, 206-229 33. Shires, T.K., Ekren, T., Narurkar, L.M., & Pitot, H.C. (1973) Nature New Biol. 242, 198-201
REFERENCES 1. Redman, CM. (1969) / . Biol. Chem. 244, 43084315 2. Takagi, M., Tanaka, T., & Ogata, K. (1970) Biochim. Biophys. Ada 217, 148-158 3. Tanaka, T. & Ogata, K. (1971) / . Biochem. 70, 693-697 4. Redman, CM. (1968) Biochem. Biophys. Res. Commun. 31, 845-850 5. Takagi, M. & Ogata, K. (1968) Biochem. Biophys. Res. Commun. 33, 55-60 6. Takagi, M., Tanaka, T., & Ogata, K. (1969) / . Biochem. 65, 651-653 7. Dallner, G., Siekevitz, P., & Palade, G.E. (1966) / . Cell Biol. 30, 73-96 8. Ragnotti, G., Lawford, G.R., & Campbell, P.N. (1969) Biochem. J. 112, 139-147 9. Lowe, D. & Hallinan, T. (1973) Biochem. J. 138, 825-828 10. Sakamoto, T. & Higashi, T. (1973) / . Biochem. 73, 1083-1088 11. Ikehara, Y. & Pitot, H.C (1973) / . Cell Biol. 59, 28-44 12. Shafritz, D.A. (1974) / . Biol. Chem. 249, 81-88 13. James, D.W., Rabin, B.R., & Williams, D.J. (1969) Nature 224, 371-372
Vol. 78, No. 6, 1975
A Poly some-Membrane Binding System from Rat Liver I. Basic Characterization of the Binding System1 Masamichi TAKAGI2 and Mahlon B. HOAGLAND Worcester Foundation for Experimental Biology, 222 Maple Avenue, Shrewsbury, Mass. 01545, U.S.A. Received for publication, May 28, 1975
A simple reaction system was developed to examine the binding of polysomes to membranes of the endoplasmic reticulum and to investigate the fate of ribosomes and nascent chains during protein synthesis in vitro. The system consisted of Sephadex G-25 treated post-mitochondrial fraction prepared from rat liver (SephadexPM) as a source of membranes, and radioactive free polysomes prepared from another rat liver. The following results were obtained. 1. Nascent chains on free polysomes labeled in vivo were transferred to membranes in vitro. The process required protein synthesis. 2. This reaction occurred in two steps: a) Binding of the free polysomes to membranes in the absence of protein synthesis, b) Release of ribosomes, leaving nascent chains on the membranes, requiring protein synthesis. 3. A portion of the ribosomes found on membranes in vivo (membrane-bound ribosomes) was also released from the membranes during incubation in vitro, leaving their nascent chains on the membranes. The significance of the transfer of nascent chains from free polysomes to membranes in vitro is discussed in the light of known polysome-membrane interaction in vivo.
In many eukaryotic cells, there are two kinds of polysomes: those free in the cytoplasm and those bound to membranes of the endoplasmic reticulum. Evidence has accumulated indicating that free and membrane-bound polysomes * This work was performed with the support of a Cancer Center Support Grant («. 12708) from the National Cancer Institute of the United States to the Worcester Foundation for Experimental Biology. * Present address: Laboratory of Radiation Genetics, Faculty of Agriculture, The University of Tokyo, Tokyo, Japan. Abbreviations: DTT, dithiothreitol; HEPES, N-2hydroxymethylpiperazine-N-2-ethanesulfonic acid. Vol. 78, No. 6, 1975
synthesize different kinds of proteins. Some soluble cellular proteins have been shown to be synthesized by free polysomes (1—3) and secretory proteins by membrane-bound polysomes (/, 2, 4—6). It has also been suggested that membrane-bound polysomes are involved in the On the s y n t h e s i s o f membrane proteins (7). example Qther hand> some proteinS) for NADPH -cytOChrome c reductase [EC 1. 6. 99.1] . „ , cl , m . < 8< 9 >• C a t a l a s e fEC L 1 L L 6 ] < 10 >• a n d S e n n e dehydratase [EC 4.2.1.13] {11), have been found to be synthesized by both free and membrane-bound polysomes. A major secretory protein, serum albumin, has been detected
1225
M. TAKAGI and M.B. HOAGLANI>
1226
consistently on free polysomes (77). Furthermore, messenger RNA's for serum albumin and ferritin have been isolated from both kinds of polysomes (12). These results suggest the possibility that polysomes producing intracellular proteins on the one hand and export proteins on the other may be in dynamic equilibrium between free and membrane-bound states. If such a situation exists, it may be expected that some fraction of the free polysome population is a precursor to membranebound polysomes. To clarify this, we have investigated the binding of polysomes to the membrane in vitro and the fate of ribosomes and nascent chains after completion of the peptide chains. Polysome-membrane binding systems in vitro have been reported by several groups (13—19). Among them, the system reported by the present authors has the advantages of simplicity and directness. It was also useful in detecting certain qualitative changes in the polysome binding capacity of membranes of the endoplasmic reticulum, as described previously. The purpose of the investigation described in this paper was to characterize the binding system in detail. MATERIALS AND METHODS Materials — CsCl (optical grade), ribonuclease-free sucrose and [6-uC]orotic acid (42 mCi/mmole) were purchased from Schwarz/ mann. [4, 5-3H]Leucine was from New England Nuclear. ATP, GTP, phosphoenolpyruvate and pyruvate kinase [EC 2.7.1.40], DTT and HEPES were purchased from Sigma. Animals—Female rats, COBS.CD strain, 2—3 months old, were used after starvation for 16-20 hr. To label ribosomal RNA uniformly, each rat was injected with [14C]orotic acid intraperitoneally about 20 hr before sacrifice. To label the nascent chains on polysomes, 200 ^Ci of [3H]leucine in 0.1 ml of saline solution was injected into a rat through the portal vein 13 min before sacrifice (2). Cell Fractionation — Preparation of the post-mitochodrial fraction and free polysomes was as described previously (18). In brief, livers were homogenized in 0.25 M sucrose in
TKM (50 mM Tris-HCl, pH 7.6, 25 mM KC1, 5 mM MgCl2), then centrifuged at 15,000 Xg for 10 min. The supernatant, the post-mitochondrial fraction, was put on 1.8 M sucrose1.0 M sucrose double layers and centrifuged at 40,000 rpm for 2 hr in a SW 50.1 Spinco rotor. Free polysomes were obtained as a pellet. In some experiments, a turbid zone between the 1.8 M and 1.0 M sucrose layers was taken up, homogenized in a small volume of the incubation buffer and used as the crude membranebound fraction. Incubation In Vitro—Sephadex-treated postmitochondrialfraction (20) (Sephadex-PM) wasprepared as desribed in the previous paper (18). It was incubated in the presence of either labeled free polysomes or the crude membranebound polysome fraction. The standard reaction mixture contained 50 mM HEPES-KOH, pH 7.6, 70 mM KC1, 5 mM MgCl2, and 4 mM DTT. In some of the reaction mixtures, a. mixture of components required for protein, synthesis (referred to as "P") was added: 2 mM ATP, 0.4 mM GTP, 10 mM phosphoenolpyruvate, and 50 ftg/ml of pyruvate kinase. Centrifugal Analysis of the In Vitro Interaction of Polysomes and Nascent Chains with Membranes—The interaction of ribosomes and nascent chains of free polysomes with membranes was detected by either discontinuous, sucrose density or CsCl-sucrose gradient centrifugation. In the former method, each reaction-mixture was put on a discontinuous sucrose gradient consisting of 3 layers, 1.5 ml. each, of sucrose solution, 1.8, 1.0, and 0.25 M, all in TKM. After centrifugation for 16 hr at. 40,000 rpm, each tube was fractionated into 5 fractions. In the latter method, the reaction, mixture was first fixed with 5% formaldehyde, put on a CsCl-sucrose density gradient, centrifuged and fractionated as described previously (18). Trichloroacetic acid-insoluble radioactive material of each fraction was collected on a glass filter disc (GF/C, Whatman), solubilized in 1 ml of Protosol (New England Nuclear) and counted in toluene scintillator fluid. RESULTS Protein Synthesizing Activity and the Ef— /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. 1
fects of Inhibitors—Sephadex-PM, which was used as a source of membranes in the polysome-membrane binding system, was active in •amino acid incorporation, as described by Richardson et al. (20). The incorporation reached a plateau after incubation for 15 min at 37°. As shown in Table I, it depended on "P." ATP alone was only partially effective. Inhibitors of protein synthesis, puromycin (1 mM), cycloheximide (1 mM), and pactamycin (0.1 xnM), inhibited incorporation, while a low concentration of pactamycin (0.004 mM) had little effect. These results suggest that peptide chain elongation on pre-existing polysomes occurs in this system and that initiation of protein synthesis is negligible (21). Protein Synthesis-dependent Transfer of Nascent Peptides from Free Polysomes to Membranes—In our previous paper, it was shown that Sephadex-PM was active as a source of membranes in interacting with exogenously added free polysomes (18). As Sephadex-PM was active in protein synthesis (Table I), we wished to follow the fate of nascent peptides on free polysomes as they interacted with membranes and to assess the effect of protein synthesis in the reaction. Free polysomes whose nascent chains had been labeled with [3H]leucine in vivo were incubated with Sephadex-PM and the distribution of [3H]radioTABLE I. Effect of inhibitors on the incorporation of [3H]leucine by Sephadex-PM into the acid-insoluble fraction. Sephadex-PM (2.4 rag of protein) was incubated with 10 /iCi of [3H]leucine in a final volume of 0.5 ml. The complete reaction mixture contained " P " in the standard reaction mixture. After incubation for 30 min at 37°, 50 p\ was removed from each of the reaction tubes and trichloroacetic acidinsoluble radioactivity was counted. 1. 2. 3. 4. 5. 6. 7.
Complete (37°) Complete (0°) 1-"P" 3+ATP (2mM) 1+puromycin (lmiu) 1+cycloheximide (1 mM) 1+pactamycin (0.1 mM)
8. 1+pactamycin (0.004 mM)
Vol. 78, No. 6. 1975
100% (8218 cpm) 2% 6% 21% 17% 43% 25% 91%
1227
activity between free polysomes and membrane fractions was assayed by discontinuous sucrose density gradient centrifugation (Fig. 1). Fraction No. 3 (a turbid zone between the 1.8 and 1.0 M sucrose layers) represented the main membrane fraction, and fraction No. 5 (a pellet) contained the free ribosomes (plus free 100 A
B
50
«_•_•_•_
.11
1 2 3 A 5 Number
Fig.' 1. Protein synthesis-dependent transfer of nascent chains from free polysomes to membranes (I). A rat was injected with' 200 fiCi of [3H]leucine through the portal vein. After 2.5 min, the liver was removed and free polysomes were prepared. Sephadex-PM was prepared from another rat without radioactivity. Sephadex-PM (5 mg of protein) and free polysomes (274 fig of RNA=672 cpm) were incubated in a final volume of 1 ml. The complete reaction mixture was the same as that given in the legend to Table I. After incubation at 0° or 37°, 0.6 ml of each reaction mixture was put on a discontinuous sucrose gradient as described in " MATERIALS AND METHODS." Each gradient, after centrifugation for 16 hr at 40,000 rpm, was fractionated into 5 fractions: from the top, 1) 1.5 ml, 2) 1.1 ml (the soluble protein fraction), 3) 1.2 ml (the turbid zone or main membrane fraction), 4) 1.1 ml, and 5) the pellet (the free ribosome fraction). The trichloroacetic acid-insoluble radioactivity of each fraction was counted. A) complete at 0° for 30 min, B) complete minus " P " at 37°, C) complete at 37° for 10 min, D) complete at 37° for 30 min.
1228
M. TAKAGI and M.B. HOAGLAND
polysomes when protein synthesis did not occur). Fraction No. 4 could be a mixture of both. When the incubation was carried out at 0° or at 37° without "P," i.e., where protein synthesis did not occur (cf. Table I), most of the radioactivity was recovered in the free ribosome fraction, fraction No. 5 (Fig. 1-A, 1-B). In the presence of "P" at 37°, a substantial part of the labeled nascent peptides, originally on exogenously added free polysomes, was found on membranes (Fig. 1-C, 1-D), indicating that the transfer of nascent chains on free polysomes to membranes depended on protein synthesis. In the next experiment, free polysomes were labeled doubly; the RNA was labeled uniformly with [uC]orotic acid and the nascent chains with [3H]leucine, in order to determine the location of both ribosomes and nascent chains after incubation. As shown in Fig. 2-A, neither ribosomes nor nascent chains were transferred from free polysomes to membranes
100
3
2
3
4
5
Fraction Number
Fig. 2. Transfer of nascent chains from free polysomes to membranes and release of ribosomes from the membranes. Ten //Ci of [l4C]orotic acid was injected into a rat 48 hr before sacrifice and 200 ftC\ of [3H]leucine was then injected through the portal vein 2.5 min before sacrifice. Free polysomes were prepared from the liver and incubated with SephadexPM prepared from a non-radioactive rat liver. One ml of the reaction mixture contained 4.8 mg of Sephadex-PM protein, 166 fig of RNA of free polysomes ( U C, 1448 cpm; 3H, 3720 cpm), and " P " in the standard reaction mixture. After incubation for 20 min, 0.6 ml of each reaction mixture was put on a discontinuous sucrose gradient. See the legend to Fig. 1 for other experimental conditions. A) 0°, B) 37°. Black bars, [UC]RNA radioactivity; and white bars, [3H]peptide radioactivity-
in the absence of protein synthesis. When, protein synthesis occurred, on the other hand (Fig. 2-B), a substantial amount of 3H-labeled nascent chains was transferred from free polysomes to membranes, as expected from the results shown in Fig. 1. However, most of the [MC]orotic acid-labeled ribosomes were not bound to membranes, but were free (Fig. 2-B). This result suggested that when protein synthesis occurs, free polysomes interact with membranes, leave their nascent peptides on the membranes and release ribosomes from the. membranes into a free state. This result could also have been obtained if the nascent chains on free polysomes had simply been released into the soluble protein fraction, and subsequently bound non-specifically to the membranes. This was ruled out by experiments (Fig. 3) which showed that radioactive soluble proteins incubated under conditions which promote the transfer of nascent, chains from polysomes to membranes do not bind to membranes. The transfer of nascent, chains from free polysomes to membranes when, protein synthesis is occurring may be detected more quantitatively by analysis of the reaction mixture by CsCl-sucrose density gradient centrifugation after fixing polysomes and polysome-membrane complexes with formaldehyde. This technique had been successfully used in our previous work (18) to separate free polysomes from membrane-bound polysomes. Tworeaction mixtures were incubated at 37°, one. in the presence of "P" (Fig. 4-A) and the other in the absence of "P" (Fig. 4-B). It can be seen that a substantial fraction of the radioactivity was found in the membrane fractions, (fraction No. 8—13) only when the incubation was carried out under conditions where protein synthesis took place. In subsequent experiments, this analytical method was used in place of discontinuous sucrose gradient centrifugation. Separation of the Reaction into Two Steps—The results so far described suggested that in our reaction system, exogenously added free polysomes interacted with membranes and, if protein synthesis took place, completed nascent chains were left on the membranes and ribosomes were released. If this is indeed the case,. /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM.
I
1229
100
5
-
IOO
c
D
50 -
1 1.. I
1 1.
I 2 3 4 5 2 3 4 5 Fraction Number
Fig. 3. Non-binding of soluble protein to membranes. Sephadex-PM (2.34 mg of protein) was incubated with either 60 fig of RNA of [3H]leucinelabeled free polysomes (A and B) prepared as described in the legend to Fig. 1, or 134 ftg of protein of [3H]leucine-labeled soluble protein fraction (C and D). The latter was obtained as a supernatant after centrifuging the post-mitochondrial fraction of [3H]leucine-labeled (200 ftCi injected 4.5 hr before sacrifice) rat liver for 2 hr at 40,000 rpm. The reaction mixture contained " P " and the final volume was 0.5 ml. After incubation for 30 min, transfer of nascent chains from free polysomes and binding of soluble protein to membranes were assayed by discontinuous sucrose gradient analysis as described in the legend to Fig. 1. A: Incubation with [3H]leucine-labeled free polysomes at 0°. B: The same as A, but at 37°. C: Incubation with [3H]leucinelabeled soluble protein fraction at 0°. D: The same as C, but at 37°. the reaction should be separable into two steps; (1) binding of polysomes to membranes and (2) release of ribosomes from the membranes leaving completed nascent chains on the membranes. The time course of binding of free polysomes to membranes of Sephadex-PM in the absence of protein synthesis (incubation without "P") is presented in Fig. 5. (In our previous work, ethanol was found to enhance Vol. 78, No. 6, 1975
10
15 O Fraction Number
5
10
15
Fig. 4. Protein synthesis-dependent transfer of nascent chains from free polysomes to membranes (II). The reaction systems were as described in the legend to Fig. 1. An aliquot (0.32 ml) of each of the reaction mixtures was removed after incubation for 35 min, fixed with formaldehyde and then put on a CsCl-sucrose gradint as described in " MATERIALS AND METHODS." Each gradient was centrifuged for 16 hr at 41,000 rpm, fractionated into 17 fractions and trichloroacetic acid-insoluble radioactivity was counted. A: complete at 37°. B: complete minus " P " at 37°.
20 40 60 80 Time of Incubation (min)
Fig. 5. Time course of binding of free polysomes to membranes. Sephadex-PM (2 mg of protein) was incubated with 40 fig of RNA of uC-labeled free polysomes at 23° in the absence of " P " for 0, 10, 40, or 90 min in the presence of 1.25% ethanol, which enhances binding {18). After incubation, each reaction mixture was analyzed by CsCl-sucrose gradient centrifugation as described in the legend to Fig. 4. The distribution of the added radioactivity into the free polysome (fraction No. 1-7) and membrane (fraction No. 8-13) fractions was determined in each reaction mixture. • , free polysome fraction ; x , membrane fraction. polysome binding (18). It was used in this and subsequent experiments to maximize polysome binding in the system.) The polysome
1230
M. TAKAGI and M.B. HOAGLAND
binding reaction reached a plateau after incubation for 40 min at 23°. In the next set of experiments, free polysomes were first incubated with Sephadex-PM in the absence of "P" for 40 min at 23° to complete polysome binding. The "P" was added and incubation continued to allow completion and release of nascent chains. (The second incubation was for 10 min at 37°.) During the two successive incubations, one tube contained 1.25% ethanol, the other did not. Ethanol enhances the binding of free polysomes to membranes in the first incubation (18). Before and after the second incubation, an aliquot of each of the reaction mixtures was removed and analyzed by CsCl-sucrose gradient centrifugation. The results are shown in Fig. 6. After the first incubation (the binding reaction), a small fraction (15—20%) of [uC]ribosomes in free polysomes together with [8H]nascent chains was found to be bound to
105
A
-
50
0
B
rfl.
the membranes (Fig. 6-A). Ethanol enhanced this binding (Fig. 6-B) as expected from our previous work (18). After the second incubation (the transfer reaction), about 40% of [3H]nascent chains had been transferred from free polysomes to membranes, while less than 10% of the [uC]ribosomes in free polysomes were found in the membrane fraction (Fig. 6-C). Even in the presence of ethanol, only 10% of ["CJribosomes remained on the membranes, leaving more than 50% of [3H]nascent chains on the membranes. The time course of transfer of nascent chains from polysomes in the presence or absence of ethanol is shown in Fig. 7. These results lead us to conclude that: (1) during the first incubation (the binding reaction), nascent chains on polysomes were bound to the membranes. This binding reaction was enhanced by ethanol; (2) during the second incubation (the transfer reaction), nas-
c
D
-1
[r
in •
too
50
1-
Fig. 6. Protein synthesis-dependent transfer of nascent chains of free polysomes to membranes and the release of ribosomes from membranes. Sephadex-PM (2.8 mg of protein) and doubly labeled free polysomes (69 fig of RNA) prepared as described in the legend to Fig. 2, were incubated in the absence of " P " with or without 1.25% ethanol. After 50 min at 23°, one-half of each reaction mixture was removed (A and B) and to the other half, " P " was added. Further incubation was carried out at 37° for 10 min (C and D). These four samples were analyzed by CsCl-sucrose gradient centrifugation and fractionated as described in the legend to Fig. 4. The results are summarized showing the percentage distribution of total radioactivity in the first 7 fractions (L; lower fraction =free polysomes fraction), fractions 8 to 13 (M; middle fraction=rough membrane fraction), and the top four fractions (T; top fraction =smooth membrane and soluble fraction). A) 23° for 50 min without ethanol, B) the same as A, but with ethanol (1.25%), C) the same as A and then 37° for 10 min in the presence of " P " , D) the same as B and then 37° for 10 min in the presence of " P . " Black bars, [UC]RNA radioactivity and white bars, [3H]peptide radioactivity.
/ . Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. I
1231
cent chains of polysomes on the membranes were completed and remained on the membranes, while ribosomes (so-called "run off" ribosomes) left the membranes to enter the free fraction. Sucrose density gradient analysis showed that the polysomes were degraded to monosomes during protein synthesis in this reaction system (data not shown). Incubation of Membrance-bound Polysomes in the Reaction System—To determine the fate of nascent chains and ribosomes of indigenous membrane bound polysomes in our system dur-
ing protein synthesis, a crude membrane-bound polysome fraction, whose nascent chains and ribosomes were doubly labeled, was incubated with Sephadex-PM and the protein synthesizing system. Before and after incubation, an aliquot was removed and separated by centrifugation to analyze the distribution of nascent chains and ribosomes between the free and membrane-bound fractions. The results are presented in Table II. As the membrane-
3
Fig. 7. Time course of transfer of nascent chains from free polysomes to membranes after the binding of free polysomes to membranes in the presence or absence of ethanol. Sephadex-PM (8.4 mg of protein) and free polysomes (207 fig of RNA) prepared from rat liver labeled with [3H]leucine for 2.5 min were incubated in a final volume of 1.5 ml in the presence or absence of ethanol, as described in the legend to Fig. 6. After incubation for 50 min at 23°, " P " was added to both reaction mixtures. At 0, 3, 6, and 10 min after the addition, 0.32 ml of the reaction mixture was removed and analyzed by CsCl-sucrose gradient centrifugation. The percentage distribution of 3H-radioactivity in the free polysome and rough membrane fractions was calculated as shown in the legend to Fig. 6. O, free polysome fraction (without ethanol); A, free polysome fraction (with ethanol); • , rough membrane fraction (without ethanol); • , rough membrane fraction (with ethanol).
6 9 Time (min)
TABLE II. Release of membrane-bound ribosomes from membranes leaving nascent chains on the membranes. A rat was doubly labeled with [uC]orotic acid and [3H]leucine as described in the legend to Fig. 2. The crude membrane-bound polysome fraction (68 fg of protein) prepared from the rat was incubated with Sephadex-PM (2 mg of protein) prepared from another rat liver without radioactivity in the presence or absence of 1.25% ethanol. After incubation for 90 min at 23°, half of the mixture was removed from both the tubes and fixed with formaldehyde. To the residual half of each of the reaction tubes, " P " was added and a second incubation was carried out at 37°. After 20 min, the reaction mixtures were fixed with formaldehyde and analyzed in CsCl-sucrose gradients. The percentage distribution of 3H- and "C-radioactivity in free polysomes and rough membrane (=R-ER) fractions was calculated (see the legend to Fig. 6). % Distribution of the radioactivity 3
Conditions of incubation
1. 2. 3. 4.
23°, 90 min 1-t-ethanol 1, then plus " P " , 37°, 20 min 3+ethanol
Vol. 78, No. 6, 1975
H
Free polysomes
R-ER
Free polysomes
R-ER
9
76
43
47
12
74
46
48
12
71
63
29
11
68
61
28
1232
M. TAKAGI and M.B. HOAGLAND
bound polysome fraction used- was crude and possibly contained free cytoplasmic particles such as ribosomal subunits, a substantial portion of 14C-radioactivity of the reaction mixture, even before incubation, was found in the free ribosome fraction by analytical centrifugation. After the incubation, about one-third of membrane-bound ribosomes were released from the membranes into a free state, while nascent proteins on the membranes remained in situ during the incubation. Although release of ribosomes from the membranes was not as clear as in the case of the in znYro-constituted complex, this result suggests that the fates of nascent chains and ribosomes are similar in the in t>?7«)-constituted polysome-membrane complex and "natural" membrane-bound polysomes extracted from the cells. DISCUSSION
The general statement that free polysomes synthesize protein for intracellular use and membrane-bound polysomes synthesize protein for export no longer seems tenable. It would appear, as noted in the introduction, that each class of polysomes is capable of synthesizing both kinds of protein, although the relative amount of each kind of protein synthesized is variable (2, 8—11). Some of the observations leading to this conclusion might be artifacts of fractionation, i.e., contamination of membrane-bound polysomes by free polysomes as suggested by Cioli and Lennox (22), detachment of polysomes from membranes during isolation (23, 24), or trapping of free polysomes in membrane-rich fractions during centrifugation (25). We favor the hypothesis that free and membrane-bound polysomes are in dynamic equilibrium with each other. This equilibrium could be influenced by the character of the nascent chains on polysomes. Thus, for example, polysomes festooned with nascent albumin have a higher affinity for membranes than polysomes bearing nascent ferritin ( 1 , 2, 4—6). The equilibrium would also presumably be affected by the character of the endoplasmic reticulum. This might explain why, in hepatoma cells, nascent albumin is found mainly on free polysomes (26).
We have already reported that physiological changes induced in liver — e.g. by phenobarbital treatment or regeneration — alter the extent to which polysomes can bind to membranes in vitro (18). Furthermore, we have found that 3-methylcholanthrene treatment of a rat increases the polysome binding capacity of the membranes dramatically (27). The study of such in vitro binding reactions wilL hopefully offer clues to the regulation of the putative free polysome-bound polysome equilibrium in vivo. Membranes are readily denatured during repeated centrifugation and resuspension (28, 29) and our use of Sephadex-PM minimized this danger, thereby hopefully providing a more physiological system for the study of polysome-membrane dynamics. It also has the advantage of being active in protein synthesis. The obvious disadvantage of the system is its crudity as a mixture of several different kinds of membranes and other cellular macromolecular contaminants. Our results suggest that free polysomes not only react with membranes but they deposit their nascent chains in the membranes and then depart (Figs. 6 and 7). This stage of the reaction requires protein synthesis. Bound polysomes also participate in the same type of reaction in vitro (Table II). Recently, Borgese et al. (19) have shown that most of the nascent chains transferred from polysomes to membranes in vitro are sensitive to proteolytic enzymes. They interpreted this result as indicating that reconstituted complexes made from membranes and polysomes in their system might be adsorption artifacts and not physiological products. However, Negishi et al. (30) have reported that nascent NADPH-cytochrome c reductase found in membrane-bound polysomes was sensitive to proteolytic enzymes. On the other hand, nascent serum albumin was found to be resistant to the same treatment. Considering the fact that, in rat liver, a part of the membranebound polysome population is loosely bound while the other part is tightly bound (31), we may speculate that only those polysomes which are loosely bound to membranes are in some dynamic equilibrium with free polysomes in /. Biochem.
A POLYSOME-MEMBRANE BINDING SYSTEM. I
1233
the cytoplasm. Our results lead us to conclude that, since protein synthesis causes nascent chains to remain bound to membranes, the binding of polysomes to membranes in our in vitro system is through nascent chains on polysomes and not through ribosomes. Other investigators have reported the same kinds of binding using high salt-puromycin treatment of reconstituted complexs (32, 33). The use of our easily prepared system may help to shed light on membrane-polysome relationships as they relate to major physiological changes, as we have reported preliminarily previously (18). Further details will be reported in a subsequent paper (27).
14. Sunshine, G.H., Williams, D.J., & Rabin, B.R. (1971) Nature New Biol. 230, 133-136 15. Ragland, W.R., Shires, T.K., & Pitot, H.C. (1971) Biochem. J. 121, 271-278 16. Shires, T.K., Narurkar, L., & Pitot, H.C. (1971) Biochem. J. 125, 67-79 17. Khawaja, J.A. (1971) Biochim. Biophys. Ada 254, 117-128 18. Takagi, M. & Hoagland, M.B. (1974) Biochem. Biophys. Res. Commun. 58, 868-875 19. Borgese, N., Mok, W., Kreibich, G., & Sabatini, D.D. (1974) / . Mol. Biol. 88, 559-580 20. Richardson, A., McGown, E., Henderson, L.M., & Swan, P.B. (1971) Biochim. Biophys. Ada 254, 468-477 21. Stewart-Blair, M.L., Yanowity, I.S., & Goldberg, I.H. (1971) Biochemistry 10, 4198-4206 22. Cioli, D. & Lennox, E.S. (1973) Biochemistry 12, 3211-3217 23. Bont, W.S., Geels, J., Huizinga, A., Mekkekhlt, K., & Emmelot, P. (1972) Biochim. Biophys. Ada 262, 514-524 24. O'Tool, K. (1974) Biochem. J. 138, 305-307 25. Vernie, I.N., Bont, W.S., & Emmelot, P. (1971) Cancer Res. 31, 2189-2195 26. Uenoyama, K. & Ono, T. (1972) Biochim. Biophys. Ada 281, 124-129 27. Takagi, M., submitted to / . Biochem. 28. Dallner, G. & Nilsson, R. (1966) / . Cell Biol. 31, 181-193 29. Glaumann, H. & Dallner, G. (1970) / . Cell Biol. 47, 34-48 30. Negishi, M., Sawamura, T., Morimoto, T., & Tashiro, Y. (1975) Biochim. Biophys. Ada 381, 215-220 31. Tanaka, T. & Ogata, K. (1972) Biochem. Biophys. Res. Commun. 49, 1069-1074 32. Adelman, M.R., Sabatini, D.D., & Blobel, G. (1973) / . Cell Biol. 56, 206-229 33. Shires, T.K., Ekren, T., Narurkar, L.M., & Pitot, H.C. (1973) Nature New Biol. 242, 198-201
REFERENCES 1. Redman, CM. (1969) / . Biol. Chem. 244, 43084315 2. Takagi, M., Tanaka, T., & Ogata, K. (1970) Biochim. Biophys. Ada 217, 148-158 3. Tanaka, T. & Ogata, K. (1971) / . Biochem. 70, 693-697 4. Redman, CM. (1968) Biochem. Biophys. Res. Commun. 31, 845-850 5. Takagi, M. & Ogata, K. (1968) Biochem. Biophys. Res. Commun. 33, 55-60 6. Takagi, M., Tanaka, T., & Ogata, K. (1969) / . Biochem. 65, 651-653 7. Dallner, G., Siekevitz, P., & Palade, G.E. (1966) / . Cell Biol. 30, 73-96 8. Ragnotti, G., Lawford, G.R., & Campbell, P.N. (1969) Biochem. J. 112, 139-147 9. Lowe, D. & Hallinan, T. (1973) Biochem. J. 138, 825-828 10. Sakamoto, T. & Higashi, T. (1973) / . Biochem. 73, 1083-1088 11. Ikehara, Y. & Pitot, H.C (1973) / . Cell Biol. 59, 28-44 12. Shafritz, D.A. (1974) / . Biol. Chem. 249, 81-88 13. James, D.W., Rabin, B.R., & Williams, D.J. (1969) Nature 224, 371-372
Vol. 78, No. 6, 1975
