ARCHIVES OF RIOCHEMISTRY ANI) BIOPHYSlCS 168, 35-39 (197-i)

Purification of Reticulocyte Ribosomes Using Polyuridylic Acid:

Sepharose Chromatography

TERRY LEE ANI) ROGER L. HEINTZ’

Department of Biochemistry and Biophyxics, Iowa State UniL~ersity, Ames, Iowa 50010

Received October 24, 1974

Deoxycholate-KC1 washed reticulocyte ribosomes are purified by affinity chromatogra- phy using Sepharose columns to which polyuridylic acid has been covalently bound. Upon passage through the column under nonenzymic conditions (high Mg”:K+ ratio) approxi- mately 10% of the ribosomes are retained. These ribosomes are then eluted with a buffer containing a high K+:Mg’+ ratio and are assayed for activity in various steps of the elongation process of protein synthesis. Approximately a 3- to 4-fold increase in the various activities compared with control ribosomes is obtained.

Treatment of eukaryotic ribosomes to study the various steps of the peptide bond forming process routinely results in only a small percentage of such ribosomes being active in such experiments (1). The re- moval of initiation. elongation, and termi- nation factors and the negation of effects of endogenous messenger RNA usually in- volve the incubation of crude ribosomes in a complete cell-free incorporation system, washing the ribosomes with a buffer solu- tion containing a very high K+:Mg2+ ratio and/or treatment with a detergent such as sodium deoxycholate. The large amount of inactive ribosomes in such a preparation makes the study of the structure of the ribosome during various stages of protein synthesis by physicochemical techniques such as analytical ultracentrifugation (21, fluorescent probing (31, reconstitution (41, or affinity labeling (5) difficult. Therefore separation of the active and inactive parti- cles is of obvious advantage in such stud- ies. The present report deals with one approach to this problem using columns of polyU2 covalently linked to sepharose to

’ Person to whom correspondence should be sent. ‘Abbreviations used: polyU, Polyuridylic acid;

GTP, Guanosine triphosphate: DTT, Dithiothreitol; EFl. Reticulocyte aminoacyl transferase I: and EFZ, Reticulocyte aminoacyl transferase II.

bind the active ribosomes and the subse- quent release of these active ribosomes to give an enriched preparation for the study of the steps of peptide bond formation.

MATERIALS AND METHODS

Deoxycholate washed reticulocyte ribosomes and the elongation factors, EFl and EF2, were prepared as described by Arlinghaus et al. (6). This ribosomal preparation by itself had little or no activity in the EFl or EF2 catalyzed steps of the elongation process (6). [‘“C ]Phenylalanyl-tRNA was prepared by the established procedure of Hoskinson and Khorana (7). As noted below all other reagents were from common commercial sources as delineated by Lee ef al. (8).

The various elongation assays, principally nonen- zymic binding of phenlalanyl-tRNA to reticulocyte ribosomes and polyphenylalanine formation, followed the procedure described by Heintz et al. (1). The nonenzymic binding reaction mixture contained: 33 mM Tris-HCl. pH 7.5; 13 mM MgCl,; 6.7 mM KCI; 1 mM dithiothreitol; 1 mM unlabeled phenylalanine, 200 kg/ml [“Cjphenylalanyl-tRNA and 1 mg/ml of the various ribosome preparations. The polyphenylala- nine synthesis assay contained: 33 mM Tris-HCl, pH 7.5; 6.7 mM MgCl,; 67 mM KCI; 1.3 mM GTP: 1 rnM unlabeled phenylalanine; 200 Kg/ml [‘“C]phenylala- nyl-tRNA; varying concentrations of the ribosome preparations; 1.3 mM DTT; and saturating amounts of either EFl and EF2 or a cruder enzyme fraction (AS70 as described by Arlinghaus et al. (6)).

The polyU:Sepharose columns were prepared using the cyanogen bromide activation technique of Lindberg et al. (9). The columns were freed of

35 Copyright 0 1975 by Academic Press, Inc. All rights of reproduction in any form reserved.

36 LEE AND HEINTZ

unbound polyU by washing first with a solution containing 2 M KC1 and 33 mM Tris-HCI, pH 7.5. and subsequently with the nonenzymic binding buffer. This material was stored at 4°C. To remove any polyU which may be hydrolyzed at 37”C, the temperature at which ribosomes were absorbed to the column, each matrix was washed at 37°C to further remove any free or hydrolyzed polyU prior to usage. The extent of polyU binding to the activated Sepharose was deter- mined by alkaline hydrolysis (1.0 ml 1 N NaOH per 0.1 ml matrix) overnight at room temperature of the polyU and measurement of the liberated nucleotides by the absorbance at 260 nm using an extinction coefficient of 7.8 x lo3 at pH 12. The absorbant mixture for this polyU:Sepharose column contained: 33 mM Tris+HCl, pH 7.5; 13 mM MgCl,; 6.7 mM KCl; 0.67 mM DTT; 67 pM unlabeled phenylalanine; 7.5 mg of washed reticulocyte ribosome, approximately 2 mg of phenylalanyl-tRNA, and enough polyU:Sepharose mixture for a column of 3 ml. After mixing at 37” for 10 min, this suspension was chilled (4°C) and trans- ferred to a small column (0.9 x 5 cm). After washing with at least 3 ml of the nonenzymic binding buffer to elute the inactive ribosomes, the bound ribosomes were eluted with 3-4 ml of the various release buffers as described below.

RESULTS

For the coupling of polyU to the cyano- gen bromide treated Sepharose the acti- vated Sepharose was used immediately after the activation process (9). Figure 1 shows the results of this coupling reaction in terms of polyU bound as the amount of activated Sepharose 4B is varied. Rou- tinely the coupling reaction is performed using a ratio of polyU (1 mg/ml): activated Sepharose of 3:4 (v/v). The normal yield from this procedure (after incubation for 10 min at 37°C) is 3.5 x lo3 pm01 of polyU (assuming a molecule of weight 7.6 x 104) bound per ml of matrix material.

Before using polyU:Sepharose to bind the washed reticulocyte ribosomes it is necessary to investigate the behavior of such ribosomes on a column of Sepharose itself (no bound polyu). Ribosomes are incubated with activated Sepharose at 37°C (Methods) under nonenzymic binding conditions and then this incubation mix- ture placed in a small glass column (0.9 x 5 cm). This column is then eluted with the nonenzymic binding buffer (Methods). The effluent is monitored at 260 nm to measure the ribosomes which pass through

VOLUME OF ACTIVATED SEPHAROSE

(ml. 1

FIG. I. The binding of polyuridylic acid to activated Sepharose. PolyU (1 ml of a 1 mg/ml solution in potas- sium phosphate) was added to the indicated volumes of activated Sepharose 4B (Methods). The slurry was mixed overnight at 4°C. The amount of polyU cova- lently bound to the Sepharose was determined as described in the Methods section: The results are expressed as pmoles of uridylic acid (x lo*) bound per ml of Sepharose.

the matrix. Less than 2% (0.05 mg/ml matrix) of the ribosomes of the incubation mixture remain on the column after wash- ing with nonenzymic binding buffer. Also Sepharose 4B does not interfere with the polyU directed binding of phenylalanyl- tRNA to these reticulocyte ribosomes under nonenzymic binding conditions.

When ribosomes are incubated under conditions similar to the above only using the polyU:Sepharose complex (Methods) approximately 15% of the ribosomes re- main bound to the column after washing with the nonenzymic binding buffer. Vari- ous conditions have been investigated for the release of these bound ribosomes. A buffer containing 0.15 M KCl; 33 mM Tris-HCl, pH 7.5; and 10 mM dithiothreitol results in little elution of these bound ribosomes. When the KC1 concentration in this buffer is raised to 0.3 M (Fig. 2) approximately 95% of the bound ribosomes are released. However, ribosomes released under such conditions show very little enhancement of activity compared with the original ribosomes. This result appears

SEPARATION OF ACTIVE RIBOSOMES 37

to be an inactivation of these ribosomes due to the absence of Mg2+ in the release buffer. When 5 mM MgCl, is included in the release buffer (0.3 M KCl, 33 mM TrisHCl, and 10 mM DTT) the released ribosomes showed a significant enhance- ment of activity, however, most of the ribo- somes remained bound to the column. If this MgClz concentration is lowered to 2 mM most of the ribosomes are released but again only a slight enhancement of nonen- zymic phenylalanyl-tRNA binding or poly- phenylalanine synthesis is observed. How-

ever, if MgCl, is added to the eluate containing the released ribosomes immedi- ately as they emerge from the column a significant increase in activity compared with the original ribosomes is observed (Fig. 3).

The optimal MgCl, concentration for this effect appears to be 13 mM which is the same as that of the nonenzymic binding buffer. The procedure which appears opti- mal to date utilizes a release buffer con- taining 0.3 M KCl, 2 mM MgC12, 33 mM Tris-HCl, pH 7.5, and 0.25 M sucrose with

1 WASH BUFFER RELEASE BUFFER $;;;Rj in

lo- z w

0 E

" E- E -EOO,-

I gl

-600; 1 4- _I-

-400: F +-\ 5,"

-zoo:, 2 g

I a 20 40 60 RO 100

FRACTION NUMBER

FIG. 2. The release of active ribosomes from polyU:Sepharose complex. The binding of ribosomes to the polyU:Sepharose complex, washing of the column, and subsequent elution of the active ribosomes were performed as described in the Methods section. In this experiment the release buffer contained: 0.3 M KCl, X3 mM Tris-HCl, pH 7.5, 10 mM DTT. The fraction volumes are 0.07 ml each.

FIG. 3. Stabilization of active ribosomes by sucrose and magnesium. All procedures are the same as described in Fig. 2 with the exception that the release buffer contained: 0.3 M KCl, 33 mM Tris-HCI, pH 7.5, 2 rnM MgCl,, 10 mM DTT, and 0.25 M sucrose. The Mg*+ upon collection.

concentration of the eluate was adjusted to 13 mM immediately

38 LEE AND HEINTZ

the MgCl, concentration of the eluate ad- justed to 13 mM immediately upon emer- gence from the column (Fig. 3). The en- hancement of both the phenylalanyl-tRNA binding and polyphenylalanine synthesis activities is illustrated in Figs. 4 and 5. Approximately a 3- to 4-fold increase in these activities is obtained. A maximum enhancement of activity of about 7-fold is possible. Therefore the population of puri- fied ribosomes approach 50% with respect to their ability to participate in these protein synthesis activities. With respect to Fig. 5, the polyphenylalanine synthesis activity of the control ribosomes is linear up to 50 pglassay while this is not true for the purified ribosomes. A release buffer designed to release all the ribosomes from the polyU:Sepharose complex, 10 mM EDTA, 33 mM Tris-HCl, pH 7.5, and 10 mM DTT, shows that’the release of ribo- somes under optimal conditions as de- scribed above (Fig. 3) is as effective as the drastic treatment with the chelation agent, EDTA. The EDTA eluted ribosomes are essentially inactive. All of the release buff- ers mentioned above do not appear to remove any polyU from the polyU:Sepha- rose complex.

DISCUSSION

Many details of ribosomal peptide bond formation remain to be answered. Promi- nent among such questions is the interac- tion of various ribosomal components and the specific structure of ribosomes during the various steps of the protein biosyn- thetic process (1, 2, 8). A major difficulty with such investigations is that many ribo- some preparations appear to contain only a limited proportion of active ribosomes. This observation may be due to inactiva- tion due to the preparative process or that the ribosome preparation may be heteroge- neous with respect to various of its compo- nents such as ribosomal proteins (4-11). As a result such preparations may contain completely inactive ribosomes or ribo- somes which are active in only one or two steps of the process of peptide bond forma- tion. Examples of the activity of these partially active ribosomes include the hy-

RIBOSOMES (mq/ASSAY)

FIG. 4. Purification of active ribosomes with respect to nonenzymic binding of phenylalanyl-tRNA, The active ribosomes (Fig. 3) are compared with the original high KCl, deoxycholate washed ribosomes with respect to ability to bind phenylalanyl-tRNA nonenzymically (Methods). The results are expressed as pmoles of phenylalanyl-tRNA bound per mg of ribosomes.

RIBOSOMES (pgm/ASSAY)

FIG. 5. Purification of active ribosomes with respect to polyphenylalanine synthesis. Active ribosomes (Fig. 3) are compared with the original ribosome preparation with respect to polyphenylalanine syn- thesis (Methods). The results are expressed as pmoles of phenylalanine incorporated into polyphenylalanine per mg of ribosomes.

tide bond formation (12) or the binding of animoacyl-tRNA without incorporation into longer peptides (13).

drolysis of GTP without concurrent pep- The experiments described above illus-

SEPARATION OF ACTIVE RIBOSOMES 39

trate one method of separating active ribo- somes from inactive ones. Affinity chroma- tography using Sepharose:polyU as a mes- senger RNA proves a useful tool in this regard. PolyU covalently bound to the gel is quite stable and efficiently serves as a messenger RNA for the nonenzymic bind- ing of phenylalanyl-tRNA to reticulocyte ribosomes. Somewhat similar results using Escherichia coli ribosomes and polynucleo- tides coupled to agar gel have been re- ported by Grosjean and his coworkers (14). Conditions for the elution of the active ribosomes from the polyU:Sepharose are extremely important and our best efforts to date result in a ribosome preparation which exhibits a 3- to 4-fold enhancement of activity using either the binding of phenylalanyl-tRNA or the synthesis of polyphenylalanine as the test assay (Figs 3-5). Such a preparation contains approxi- mately 50-60% active ribosomes as the original preparation was about 1595 active. As indicated in the very first fractions of Figs. 2 and 3, the ribosomes which are not bound to column retain some activity in the two assays. The reason for this is unclear; however, if such unbound ribo- somes are challenged with polyU: Sepha- rose again they do not bind to the column.

The magnesium ion concentration ap- pears important for the elution of the ribosomes from the column in two con- trasting ways. First there is a limitation on the Mg”+ concentration above which the ribosomes are not released from the polyU: Sepharose column. However, Mg2+ is very important for the stability of the released ribosomes. This problem has been solved to some degree by adding Mg2+ to the eluate of the column. Preliminary experiments in- dicate that the ribosomes emerge from the column as some form of the 60s and 40s

subunits. Further characterization of these particles especially with respect to the various ribosomal proteins is underway. The purified active ribosome preparation as described above should prove beneficial in studying any one of the various steps of peptide bond formation, e.g., GTP hydrol- ysis during translocation.

ACKNOWLEDGMENTS

We appreciate the excellent technical assistance of Ms. Marilyn Darkes. This investigation was sup- ported by a grant from the Heart and Lung Institute of the National Institutes of Health (HE12549).

REFERENCES

1. HEINTZ. R., SALAS, M., AND SCHWEET, R. (1968) Arch. Biochem. Biophys. 125, 488-496.

2. CIILP. W., ODOM. 0. W., ANDHARDESTY, B. (1973) Arch. Biochem. Hiophys. 155, 225-236.

3. ODOM. 0. W., HEINTZ. R., WHITE. J. M., AND

4. 5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

HARDESTY, B. (1974) Arch. Biochem. Biophys., submitted.

KURLAND. C. (19721 Annu. Reu. Biochem. 41,377. OEN, H., PELLEC.RINI. M., EILOT. D., AND CANTOR,

C. (1973) Proc. Nat. Acad. Sci. USA 70, 2799. ARLINGHALX R., HEINTZ, R., AND SCHWF,ET, R.

(19681 Meth. Enrymol. 12, 70. HOSKINSON, R., AND KHORANA. H. (1965) J. Biol.

Chem. 240, 2129. LEE, T., TSAI. P., AND HEINTZ. R. (1973) Arch.

Biochem. Biophys. 156, 463-468. LINDBERG, V., AND PERSSON, T. (19721 Eur. J.

Biochem. 31, 246. LUCAS-LENARD, J., AND LIPMANN. F. (1971) Annu.

Reu. Biochem. 40, 409. CRAVEN. G., VOUNOW. P., HARDY, S., AND KUR-

LAND, C. (1969) Biochemistry 8, 2906. NISHIXIKA. Y., AND LIPMANN. F. (19661 Proc. Nat.

Acad. Sci. lJSA 55, 212. HEINTZ, R., MCALLISTER, H., ARLINGHAUS, R., AND

SCHWEET, R. (1966) Cold Spring Harbor Symp. Quant. Biol. 31, 633.

PETRE, T., BOLLEN, A., NOKIN, P., AND GROSJEAN. H. (1972) Biochimie (Paris) 54, 823.