Eur. J. Biochem. 107, 173-183 (1980)

A Trichodermin-Resistant Mutant of Saccharomyces cerevisiae with an Abnormal Distribution of Native Ribosomal Subunits Christopher J. CARTER, Michael CANNON, and Anlonio JIMENEZ

Department of Biochemistry, University of London King’s College, and Instituto de Bioquimica de MacromolCculas, Centro de Biologia Molecular, Universidad Autonoma de Madrid

(Received November 7, 1979)

1. A yeast mutant (CLP-8), resistant at the ribosome level to the trichothecene antibiotic trichodermin, differs from its parent in having an unusual distribution of native ribosomal subunits. Sucrose gradient analysis of cytoplasmic extracts from this mutant revealed a large excess of material sedimenting at 60 S with little or no material sedimenting at 40 S.

2. The excess 60-S material consists predominantly of functionally active 60-S ribosomal subunits, as indicated by both analysis of ribosomal RNA and studies in vitro using a poly(U)- directed protein-synthesizing system.

3. Using the poly(U) system it was found that high-salt-washed particles derived from either the excess 60-S peak or 80-S ribosomes of CLP-8 exhibited very similar levels of resistance to the antibiotic fusarenon-X, a drug closely related chemically to trichodermin. The same level of resistance to fusarenon-X was also shown by high-salt-washed 60-S ribosomal particles obtained from a further trichodermin-resistant yeast strain (TR-l), although this strain has a normal distribution of native ribosomal subunits. In addition, both CLP-8 and TR-1 are equally resistant to inhibition of protein synthesis by trichothecene antibiotics, as assayed in vivo.

4. Genetic analysis of CLP-8 indicates that the trichodermin-resistant trait can be segregated from the lesion responsible for the inbalance of native ribosomal subunits. However, the latter defect is only expressed phenotypically in cells that retain the trichodermin-resistant character.

5. CLP-8 has a further defect in that both in vivo and in vitro it fails to generate native 40-S ribosomal subunits from 80-S particles. There may be a lesion in the protein factor normally required for this process.

lsolation and analysis of antibiotic-resistant mu- tants from bacteria has greatly facilitated the study of ribosome structure and function in these organisms [l-31. The basis of resistance has often been de- termined [3 - 51 and this has provided insight into the molecular nature of certain drug receptor sites on ribosomes. Similar studies in eukaryotes have involved analysis, mainly, of antibiotic-resistant mutants of Scirchuvomyces cerevisiue some of which have ribo- somal lesions that have been genetically mapped [6- 81. However, there is little convincing evidence for the precise identification of a specifically-altered ribosomal component controlling antibiotic resis- tance in any of these mutants.

Abbreviution. Poly(U), poly(uridy1ic acid). Drfl’nition. Native 60-S* subunits, material sedimenting at

60 S observed in or isolated from extracts of Succhuromycrs cerr- visiue (strain CLP-8).

Enzyme. Creatine kinase (EC 2.7.3.2).

Drug-resistant mutants are frequently cross re- sistant to different antibiotics that block (apparently) the same ribosomal function, and the ribosomal target site(s) for these antibiotics is presumed to be in- volved intimately with the function inhibited. For example, two independent mutants (TR-1 and CLP-1) of S . cerevisiae, isolated as spontaneously resistant to trichodermin, exhibit cross resistance in vitro to many other structurally-related trichothecene antibiotics with the resistance residing in each case in the 60-S ribosomal subunit [9 - 121. Both mutants also show cross resistance, both in vivo and in vitro, to other compounds, including anisomycin, narciclasine, har- ringtonin and bruceantin [12- 151, although these compounds are very different chemically from the trichothecene gi-oup. Ncvertheless all thc inhibitors appear to block selectively the activity of the pcptidyl- transferase centre located on 60-S subunits of eukaryotic ribosomes.

174 A Yeast Mutant with an Abnormal Distribution of Ribosomal Subunits

In the present work we have studied in detail a further mutant (CLP-8) of S. cerevisiue again isolated as spontaneously resistant to trichodermin. Our work represents an attempt to identify a ribocoinal com- ponent that is altered in the trichodermin-resistant state. CLP-8 is known to exhibit cross resistance in vitro to both anisomycin and the trichothecene antibiotic nivalenol, and the resistance is associated with a property, presumed to be involved with peptidyl- transferase activity, of the larger ribosomal subunit [16]. Genetic studies on the three trichodermin- resistant yeast mutants CLP-1, CLP-8 and TR-1 have indicated that the antibiotic-resistant lesion is allelic. Furthermore both TR-1 and CLP-8 grow more slowly than the corresponding wild types [12] (and C. J. Carter, unpublished work) and are derived as single-step mutants, albeit from different parental strains. Despite these considerations, however, the actual levels of resistance shown by TR-1 and CLP-8 towards trichodermin may be different. Such a situation has already been observed in yeast mutants resistant to tubulosine where, although the mutations are again allelic, both high and low levels of re- sistance have been described [8]. Here we have determined that TR-1 and CLP-8 exhibit indistingui- shable levels of resistance to trichodermin both in vivo and in vitro.

We have been unable to identify an observable change in a ribosomal component in either of the two strains. However, CLP-8 alone possesses in its cyto- plasm a large excess of material sedimenting at 60 S. Detailed studies have been performed to characterize this material as functionally active 6 0 3 ribosomal subunits that are resistant to trichodermin and the possibility that excess 60-S ribosomal subunit pro- duction and trichodermin resistance may be linked has been studied using both biochemical and genetic analysis. Finally the ability of CLP-8 to generate native 40-S ribosomal subunits from an excess of 80-S free couples has been investigated both in vivo and in vitr-o.

MATERIALS AND METHODS

Mu t er- iu Is

~-[4,5-~H]Leucine (100 Ci/mmol) and L-[U-'~C]- Phcnylalanine (522 Ci/mol) were obtained from The Radiochemical Centre (Amersham, Bucks, Great Britain). Creatine kinase and yeast tRNA were obtained from Boehringer (Mannheim, Federal Re- public of Germany) and glusulase from Endo La- boratories Inc. (Garden City, New York, U.S.A.). Poly(U) and cycloheximide were both obtained from Sigma (London) Chemical Co. (Kingston-upon-Tha- mes, Surrey, Great Britain). The trichothecene anti-

biotics were dissolved in 50 % (v/v) dimethylsulphoxide ~ 7 1 .

Growth and Harvesting of Yeast Cells

Strains of Saccharomyes cerevisiue designated A224A (a , leu-2-), CLP-8 (a, leu-2-, tcm-I-), Y166 (a, tyr-2-, his-4-) and TR-1 ( E , tyr-2-, his-4-, tcm-1 - ) were maintained on solid media [18] and subsequently were grown with shaking at 30 "C either in complete synthetic medium [19] or in YM-5 medium [18]. Cell density was measured at 500 nni in a Unicam SP600 spectrophotometer : Asoo = 0.3 corresponded to approximately 1.5 x lo7 cells/ml.

Flasks containing medium were inoculated with yeast cells from an overnight culture to an initial Asoo of about 0.2 and grown at 30'C to an A500 of 0.6-0.7. Cultures were then rapidly chilled to 2 'C by placing the flasks in acetone/solid COZ ( - 20 "C) and cells were subsequently harvested by centrifu- gation at lOOOOxg for 5min at 2°C using a MSElR centrifuge. The packed cells were then twice washed in 0.1 vol. (0.1 original culture volume) of ice-cold extraction buffer E (50 mM Tris-HC1, pH 7.4, 80 mM KCl, 12.5 mM MgC12, 1 mM dithiothreitol) and the resulting pellets stored at -70°C. The yield was approximately 2 g wet weight/l culture.

Preparation of Yeast Splzeroplusts

Cells were grown in complete synthetic medium except for haploid cells arising from tetrad analysis which were grown in rich medium. All cells were grown to an Asoo of 0.3-0.45 and spheroplasts were prepared and incubated at 30 ' C as described previ- ously [18]. When spheroplast formation was complete the suspension was then diluted as described by Udem and Warner [19] and the spheroplasts allowed to recover at 30 "C.

Use of Yeast Splzeroplust,~for- Protein S>.nthesis Stwdics in vivo

To assay for protein synthesis 10 pCi [3H]leucine (previously diluted to 50 Ci/mol with unlabelled leucine) was added to 40ml of the spheroplast suspension. After incubation at 30'C for 15 min, 5-ml samples were removed into flasks containing various concentrations of inhibitor and the incubation was continued for a further 45 min with gentle shaking throughout. Samples (200 pl) were removed at appro- priate times and precipitated, together with bovine serum albumin (100 pg/sample), by addition of 3 ml of ice-cold 5 x) trichloroacetic acid containing un- labelled leucine (50 pg/ml). They were then processed for radioactivity essentially as described elsewhere

C. J. Carter, M. Cannon, and A. Jimtnez 175

[I 7,201. Counting efficiency in a Packard liquid- scintillation counter for 3H-labelled samples was approximately 18 'x. Sucrose Grudient Anulysis of Polyribosomes and Ribosomal Subunits jrom Yeust Spheroplusts

Aliquots (10 ml) of yeast spheroplasts were in- hibited with cycloheximide (100 pg/ml final con- centration). Spheroplasts were then pelleted and lysed essentially as described previously [18] but using modified lysis butter (20 mM Tris-HCI, pH 7.4, 100 mM KCI, 30 mM MgCI2, 1 mM dithiothreitol). Lysis was achieved by successive additions of sodium deoxycholate (0.2 % final concentration) and Brij 58 (0.25 final concentration) with a 5-min incubation on ice following each addition. Lysed extracts were then analyzed on 10- 30 '%, (for polyribosome analysis) or on 15-300/, (for subunit analysis) linear sucrose gradients prepared in lysis buffer. Centrifugation was at 42000 rev./min for 45 min (polyribosomes) or 2.25 h (subunits) respectively, in a Spinco SW 50.1 rotor. Gradients were monitored in an Isco gradient fractionator, using a 20-ml syringe, by upward dis- placement with 70 % glycerol. Gradients were eluted with a chart speed of 0.5 ininjs at 1.25 ml. min. The absorbance range is indicated in each figure.

Prepurution of High-Salt- Wushed 80-S Ribosomes and Ribosomul Subunits from Yeast Spheroplusts

Yeast spheroplasts (2 1) were prepared as above except that following the recovery period different treatments were applied according to the type of ribosomal preparation required. For the preparation of native 60-S* subunits (for definition see footnote and Results) from CLP-8, cycloheximide (20 pg/ml final concentration) was added to the whole culture to prevent subsequent polyribosome 'run off inter- fering with the preparation. In contrast, when pre- paring 80-S ribosomes and derived ribosomal subunits from all strains, cycloheximide was omitted and instead solid sodium azide was added (1 mM final con- centration). After a 5-min incubation spheroplasts were cooled to 8°C and maintained at this tem- perature for a further 5 min. The combination of azide and low temperature caused total polyribosome 'run off to occur. Spheroplasts from all yeast strains were subsequently chilled, harvested and lysed in the usual way and the extracts stored overnight at - 70 'C. Thawed lysates were centrifuged at 20000 x g for 15 min, to remove membraneous debris, nuclei and mitochondria, and the resulting supernatant fractions were treated in various ways.

The 20000 x g supernatant extracts from A224A, Y 166 and TR-1 were re-centrifuged at 45000 rev./inin for 1 h at 4"C in a Spinco Ti 50 rotor to sediment

ribosomal material. Ribosome pellets were then gently resuspended, using a glass rod, in a small volume of high-salt buffer (50 mM Tris-HCI, pH 7.4, 500 mM KCI, 5 mM MgC12, 2 mM dithiothreitol) to a concen- tration of approximately 300 Azho units/ml. This procedure was used either for the preparation of derived ribosomal subunits (see below) or to prepare high-salt-washed 80-S ribosomes. For the latter prep- aration the ribosome suspension was diluted to 10 ml with buffer (50 mM Tris-HCI, pH 7.4, 500 mM KCI, 40 mM MgC12, 20 mM dithiothreitol, 25 "/, sucrose) and centrifuged at 45000 rev./min for 5 h at 4°C in a Spinco Ti 50 rotor.

The 20000 x g supernatant extracts from cyclo- heximide-treated and azide-treated CLP-8 sphero- plasts were centrifuged at 40000 rev./min for 35 min to pellet polyribosomes or 80-S ribosomal material respectively, the latter pellet being used for sub- sequent preparations of both 80-S ribosomes and de- rived subunits. The pooled supernatant fractions from both CLP-8 preparations were re-centrifuged at 45000 rev./min for 3 h, and the resulting pellet contained predominantly native 60-S" subunit ma- terial, together with some 80-S ribosomes. This pellet was used for the CLP-8 native 60-S* subunit preparations. Both 80-S and 60-S* enriched ribosomal pellets from CLP-8 were resuspended, as described above, but in a low-salt buffer (50 mM Tris-HCI, pH 7.4, 50 mM KC1, 5 mM MgC12, 2 mM dithio- threitol).

Ribosomal material (40-60 A260 units) from any or the above preparations was layered on to linear 15 - 40 'x sucrose gradients (volume 30 ml), made up in the corresponding buffer, and centrifuged at 22000 rev./min at 12°C in a Spinco SW 25.1 rotor for 15 h (low-salt conditions) or 17 h (high-salt conditions). Following centrifugation, gradients were fractionated using the Isco apparatus, essentially as described earlier, except that gradients were eluted, using a 50-ml syringe, at 2.5 ml/min using an absorbance range of 2. Fractions containing either 40-S or 60-S ribosomal subunits (high-salt gradients) and native 60-S* or 80-S ribosomal material (low-salt con- ditions) were pooled separately and at this stage total volumes and ribosomal concentrations ( A z ~ o units/ml) were determined as a guide to expected recovery. Pooled ribosomal fractions were diluted to 10 ml with the appropriate gradient buffer, the MgC12 and dithiothreitol concentrations adjusted to 40 mM and 20 m M respectively, and the KCI concentration of low-salt fractions raised to 500 mM. All fractions were then centrifuged at 45000 rev./min for 5 h at 4 'C in a Spinco Ti 50 rotor. The resulting ribosomal pellets were gently resuspended in buffer (50 mM Tris- HCI, pH 7.4, 80 mM KCI, 12.5 mM MgC12, 1 mM dithiothreitol), using a fused pasteur pipette, and adding approximately 50 pl buffer for each 10 A260

176 A Yeast Mutant with an Abnormal Distribution of Ribosomal Subunlts

units ribosomal material. Suspensions were then cleared of ribosomal aggregates by centrifugation at 20000 x g for 10 min, and ribosome concentrations in the resulting supernatants were determined by A260 measurements. Final ribosome concentrations were adjusted with buffer to give 60, 100 and 160 ,4260

units/ml from 40-S, 60-S and 80-S preparations re- spectively and these were then stored in small aliquots at - 70 "C.

Prepurcition of Partially Purified Yeast Supernatant Fraction f o r Use with the Poly( U ) System in vitro

Recently grown yeast cells (2-5 g), stored at - 70 "C, were homogenized in a mortar, previously cooled to - 15 'C, with twice their weight of purified sea sand (Merck) added in two equal portions. After each addition the mixture was ground for 2-3 min until the paste became a viscous liquid, and the paste was finally extracted with 30 ml buffer E (see above). The resulting extract was centrifuged at 10000 x g for 10 min at 4°C to remove sand and cell debris and the supernatant fraction was re-centrifuged at 20000xg for 15 min at 4°C. The latter supernatant fraction was then centrifuged at 45 000 rev./min (lOOOOOxg) for 2.5 h at 4°C to pellet all ribosomal material. The resulting supernatant fraction was further purified by adding solid (NH4)2S04 and collecting the precipitate, from a 30 - 70 saturated fraction, by centrifugation at 20000 x g for 20 min at 4 "C. The pellet, containing elongation factors EF-1 and EF-2 and arninoacyl-tRNA synthetases, was resuspended in buffer (20 mM Tris-HCI, pH 7.4,l mM dithiothreitol) to concentrate the components by approximately 15-fold. The purified fraction was dialyzed overnight against 500 vol. buffer E, followed by a second dialysis for 3 h against 500 vol. fresh buf- fer E. The resulting crude factor preparation (contain- ing about 2 mg protein/ml) was stored in small aliquots in liquid nitrogen.

Poly ( U ) - Directed Polyphenylulunine Synthesis in a Reconstituted Yeast Cell-Free System

Incubation mixtures (300 pl) contained 50 mM Tris- HC1, pH 7.4, 12.5 mM MgC12, 80 mM KCI, 4 mM creatine phosphate, 1 mM dithiothreitol, 1 mM ATP, 0.1 mM GTP, 20 pM each of 20 unlabelled amino acids minus phenylalanine, 2 pCi ['4C]phenylalanine (final specific radioactivity 100 Ci/mol)/ml, creatine kinase (40 pg/ml), yeast tRNA (1 mg/ml) and poly(U) (300 pglml). Sufficient yeast supernatant fraction (usu- ally 2-5 PI) was added to produce a maximum in- corporation activity, and either 16 A260 units/ml of high-salt-washed 80-S ribosomes or 10 and 6 A260 units/ml of 60-S and 40-S ribosomal subunits re-

spectively from appropriate sources. Reactions were initiated by adding ribosomes last and after incubation at 30 "C samples (50 pl) were removed as indicated and processed for radioactivity essentially as described earlier [17]. Counting efficiency for ''C-labelled sam- ples was 82 %.

Genetic Anulysis of Yeast Strains

Diploids were formed between CLP-8 (u) and either Y166 (a) or TR-1 (a), followed by selection on minimal plates. Random spore analysis and tetrad dissection were carried out using standard techniques [16,21]. The offspring from each haploid spore were subsequently screened both for resistance to tri- chodermin (20 pg/ml) and for the presence of a native 60-S* ribosomal subunit peak.

Preparation of a Native RihosomaCSubunit Wash from Yeast that Contains Dissociation Factor Activity

The preparation is based on that published by PEtre [22] with various modifications. Yeast sphero- plasts (4 1) were treated as for the preparation of native 60-S" ribosomal subunits from CLP-8 (see earlier section) except that spheroplasts were lysed in buffer A [23]. The native ribosomal-subunit-enriched pellet derived from the first ultracentrifugation at 45000 rev./min was resuspended in 10 mM Tris-HC1 buffer, pH 7.4, containing 10 mM MgC12, 10 mM KCI, 300 mM NH4C1, 5 mM dithiothreitol, and gently stirred on ice for 15 min. All ribosomal material was then sedimented by centrifugation at 45 000 rev./min for 3 h and the supernatant fluid was then carefully removed, mixed with solid (NH4)2S04 and the material precipitating at 35 - 75 saturation was collected by low-speed centrifugation, dissolved in 10 mM Tris-HC1 buffer, pH 7.4, containing 2 mM MgC12, 100 mM KCI, 5 mM dithiothreitol, and dialyzed overnight against two changes of the same buffer. The resulting protein solution was heated at 40'C for 30 min, clarified by low-speed centrifugation and stored at 0°C prior to use. High-salt-washed 80-S ribosomes, used to test the dissociation factor activity of this preparation, were prepared as described in an earlier section and were also resuspended in the last- mentioned buffer to a concentration of 8 A260 units/ml.

Basis of Calculations

The proportions of polyribosomes, 80-S mono- mers or ribosomal subunits was determined by cutting out a tracing of the original profile, and comparing the weight of each part to that of the whole tracing. Final values quoted are an average of' four different determinations.

C. J. Carter, M. Cannon, and A. Jimenez

4 0 5

1

lf

0.30

E 5 c

a,

c e

0.15 9

C

0.x E

c

8 m a L 2 0.15 4

0

1

A2244

Sediment at ion direct ion

CLP-8 -

D

Sedimentation direction -

A224A

3

40-S 60-S8O-S

1 I I

Sedimentation direction

CLP-A -

E

40-5 60-S 80-5

0.30 E

s N c

a,

c m e 2 0.15 2

0 Sedimentation direction -

0.30

E G * c

a,

c

5 0.15 9

0

0.30

E 2J 3 " a, u c

0 L

2 0.15 9

0

TR1

C

40-5 60-S80-S

1 I 1

d Sedimentation direction

A224A+CLP-8

F ao- s

Sedimentation direction - Fig. 1. Analysis on sucro.\o fircldients (11 po1~riho.wne.s trnd rihosomcil . s h i n i / . \ / i . o i i i .spho-opltr.~l.c (1j S. cercvisiile (. \ /rcrins A224A, CLP-8 and TR-I). Spheroplasts from the various yeast strains were prepared and analyzed by sucrose gradient centrifugation, as described in Materials and Methods, to reveal polyribosomes and ribosomal subunit patterns. Polyribosome profiles from (A) extracts of A224A; (D) extracts of CLP-8. Ribosomal subunit profiles from (B) extracts of A224A; (C) extracts of TR-I; (E) extracts of CLP-8; (F) extracts of A224A and CLP-8 combined prior to co-sedimentation

Molecular weights for yeast 80-S, 60-S and 40-S ribosomal particles were taken as 4 x lo', 2.7 x 10' and 1.3 x 10' respectively [24]. A solution of 1 mg of either 80-S ribosomes or subunits/ml was assumed to have an absorbance at 260nm of 14 [12], and accordingly 1 nmol yeast ribosomes contains 56 A260

units,

RESULTS AND DISCUSSION

The majority of monoribosomes found in cellular extracts are not associated with mRNA but are present as free couples [25]. In eukaryotes, these rapidly dissociate, under high-salt conditions (5 mM Mg2+/ 500 mM K'), to produce ribosomal subunits (derived

subunits), although under low-salt conditions (5 mM Mg2+/50 mM K') the equilibrium strongly favours the existence of free couples. The small amounts of ribosomal subunits found in cell extracts under these conditions, constitute an additional special class (native subunits) that are highly active in initiation of protein synthesis.

Sucrose gradient analysis, under low-salt con- ditions, of cellular extracts prepared from sphero- plasts of Succharomyces cerevisiue, strain A224A, revealed that approximately 85% of the total ribo- somal material exists as polyribosomes. Only 2 - 3 constitutes native ribosomal subunits and the balance is made up of 80-S free couples (Fig. 1 A, B). Another yeast strain, Y 166, produced very similar results (data not shown).

178 A Yeast Mutant with an Abnormal Distribution of Ribosomal Subunits

25 ._ E

S X , 8

z15

. v) c

v

x .% 2

0 U ._ @ 5 "

5

'0 6 12 18 24 30 Incubation time (rnin) Drug concentration UM)

Fig. 2. Protrin synthesis in vitro in ~cconsriturcii~rLrsr .s~.c./cwis iind i j i ~ i i ~ i ~ i u ~ by two f,.ic./zoriic,c.c'rze ctrttihiurks of prof& synlhesis both in vivo cind in vitro. Ribosomal preparations and supernatant factors for studies in vitro and spheroplasts for studies in vivo were prepared and incubated as described in Materials and Methods. (A) Kinetics of poly(U)-directed polyphenylalanine synthesis in highly fractionated yeast systems employing 80-S ribosomes or various combinations of purified ribosomal subunits derived from CLP-8. Incorporation mixtures contained 80-S ribosomes (eo); derived 40-S particles only (-0); derived 60-S particles only (0-0); derived 60-S particles + derived 4 0 3 particles ( b w ) ; 6 0 3 particles from native 60-S* peak only (A-A); 60-S particles from native 60-S* peak + derived 40-S particles (A-A). (B) Dose response curves showing inhibition of protein synthesis in vitro [poly(U) system] and in vivo (in spheroplasts) by fusarenon-X and trichothecin respectively, utilizing material derived from both antibiotic-sensitive (A224A) and antibiotic-resistant (CLP-8 and TR-1) yeast strains. Linear rates of inhibition by fusarenon-X of protein synthesis in vitro in systems using 80-S ribosomes obtained from A224A ( L O ) ; CLP-8 (A-A); TR-1 (H). Linear rates of inhibition by trichothecin 01 incorporation of [14C]leucine into protein using spheroplasts derived from A224A (e

The spontaneous trichodermin-resistant mutants CLP-8 and TR-1, originating from A224A and Y166 respectively, arc thought to possess the same genetic lesion [lo, 12,161. However, when spheroplast ex- tracts from these mutants were subjected to sucrose gradient analysis, marked differences in native subunit patterns were apparent. Polyribosome (data not shown) and subunit patterns for TR-1 appear very similar to those observed for A224A and Y166 (Fig. 1 C). However, for CLP-8 (Fig. 1 D, E) approxi- mately 72 % only of the ribosomal material sediments as polyribosomes with the corresponding value for 60-S material being 18 %. This latter amount exceeds that found in the CLP-8 monomer peak (80 S) and is 6 - 8 times greater than that contributed by the native 60-S peak in extracts from the other strains. Furthermore, cxlracts from CLP-8 are almost totally devoid of material sedimenting at 40 S. This phenomenon can not be attributed to a poor resolution of 40-S subunits, as caused by the large excess of 60-S particles, since co-sedimentation of a mixture of CLP-8 and A224A extracts allowed clear separation of the 40-S peak (Fig. 1 F). Hereafter the 60-S material observed in or isolated from extracts of CLP-8, under the above experimental conditions, will be referred to as native 60-S* material.

The unusual ribosome profile observed for CLP-8 extracts could, in theory, have resulted from a mu- tation involving a structural change in the 40-S subunit such that, under the low-salt conditions used, di-

-0); CLP-8 (A -~ A); TR-1 (0-U)

merization of these subunits occurred to produce a particle with a predicted sedimentation coefficient of 65 S [26].However, sucrose gradient aniilysis of rRNA released by sodium dodecyl sulphate/EI)TA treatment of the native 60-S* particle from CLP-X revealed only one major peak corresponding in sedimentation to 25-S rRNA (results not shown). This pnrticle contains therefore no appreciable amounts of the RNA char- acteristic of either the 40-S ribosomal subunit or its dimer. The RNA content of the native 60-S* particle is, in fact, essentially equivalent to that found in derived 60-S subunits from both CLP-8 and A224A (results not shown). Furthermore, the rRNAs from the small ribosomal subunits from both strains were identical as far as could be detected by sucrose gradient analysis (results not shown).] t also appears that the protein composition of the ribosomal particles from these two strains is identical, at least in so far as analysis by two-dimensional gel electrophoresis re- vealed no differences in mobility in any of the proteins separated (data not shown).

In growing yeast cells frequent subunit exchange occurs between 80-S free couples and the native ribosomal subunit pool with no evidence of a fraction of 80-S ribosomes unable to dissociate or for accumu- lation of subunits incapable of recycling [27]. To as- certain if the native 60-s* peak of CLP-8 is functionally homogeneous and capable of such recycling we de- veloped an assay system for protein synthesis in vitro utilizing either high-salt-washed 80-S ribosomes or

C. J. Carter, M. Cannon, and A. Jimenen 179

purified ribosomal subunits, a crude factor preparation and exogenous mRNA, i.e. poly(U). This system re- sembles that employed by others [12] and its charac- teristics are illustrated in Fig.2A. It was completely dependent on poly(U) as a source of mRNA and ex- hibited little activity in the absence of the crude factor preparation or when any subunit preparation was added alone. Combinations of appropriate amounts of derived 40-S or 60-S subunits resulted in an ap- proximately 20-fold stimulation of polyphenylalanine synthesis over that observed for the subunits added separately. Furthermore about 70 % of this activity was also observed if an equivalent amount of native 60-S* material was used as the source of 60-S subunit, although both of the subunit recombinations produced somewhat less activity than an equivalent amount of high-salt-washed 80-S ribosomes. This may reflect some loss of activity resulting from the techniques of subunit preparation. Indeed the ad- ditional procedures required to prepare the native 60-S" material may explain why this preparation is somewhat less active than a corresponding amount of derived subunits. Clearly though, much of the material isolated from the native 60-S" peak from CLP-8 is functionally active, at least in vitro, although results from poly(U) systems must be interpreted with caution and ideally the subunits should also have been tested in a more naturally initiating system.

The trichodermin-resistant lesion in ribosomes froniTR-1 resides in the 60-S subunit [12] and the same mutation is thought to occur in the independently isolated mutant CLP-1 arising, like CLP-8, from A224A [10,16]. Although CLP-8 has been shown previously, using a poly(U) system [16], to be resistant at the ribosome level to both nivalenol and anisomycin the ribosome preparation used for these studies almost certainly included the native 60-S" material. Since both these antibiotics bind to normal ribosomal 60-S subunits the possibility exists that, for CLP-8, part or all of the resistance previously observed could have resulted from a sequestering of the drug by these particles which are in excess in this strain. We decided, therefore, to ascertain if 80-S ribosomes from CLP-8 are themselves truly resistant to tricho- thecene antibiotics, using an 80-S ribosome pre- paration devoid of native 60-S* material, and to determine if the level of resistance observed with CLP-8 was similar to that observed with TR-1.

Fig. 2 B illustrates inhibition by two selected tricho- thecene antibiotics of protein synthesis in various yeast strains both in vivo, using yeast spheroplasts, and in vitro, using the poly(U) system. Protein synthesis in spheroplasts derived from either A224A or Y166 was indistinguishably sensitive to the drug trichothecin whereas spheroplasts prepared from both CLP-8 and TR-1 exhibited a very similar high level of re- sistance to this antibiotic. The same trends were ob-

Table 1. Loculizution of re.ri.rtancr to fusurenon-X deternii~rcd by using reriprocd ~o~binu~ions of' purified ribosomal suhunits ruid supernutunt ,fractions derived .from both antibiotic-smsirive und untibiotic-resistunt yeasi strains to study poly(U]-directed poly- plzenylulunine synlhesis in vitro Reaction mixtures (300 PI) were prepared as described in Materials and Methods. Each incubation mixture contained 3.0 A26,, units of 60-S ribosomal subunits and 1.8 ,4260 units of 40-S ribosomal subunits. One incubation mixture (last line) contained an ad- ditional 1.0 unit of derived 60-S ribosomal subunits from A224A. 25 p1 of supernatant fraction obtained from either A224A or CLP-8 (adjusted to contain 2 mg protein/ml) were present in each reaction mixture as indicated in [he table. All reaction mixtures were incubated at 30 'C for 30 min and samples were removed at 6, 12, 18,24 and 30 min, precipitated with 5 % trichloroacetic acid and processed and counted for radioactivity as described in Materials and Methods. Time courses, similar to those shown in Fig.2.4, were then constructed and from these linear incorporation rates were determined in the presence and absence of fusarenon-X (20 pg/ml). Inhibition of incorporation of [14C]phenylalanine in the presence of the drug was then calculated

Source of Incorporation Inhibition by of phenyl- fusarenon-X

ribosomal subunits supernatant alanine by 15 pmol

40-S 60-S ribosome

~~ ~~

pmol/min "/, A224A A224A A224A 7.2 14 CLP-8 A224A A224A 6.9 73 TR-1 A224A A224A 5.6 76 A224A CLP-8 A224A 6.2 14 A224A CLP-8" A224A 5.6 16 CLP-8 CLP-8" A224A 5.3 12 A224A CLP-8" CLP-8 5.5 16 CLP-8 CLP-8 CLP-8 6.6 13 CLP-8 rii-i A224A 5.0 15 A224A A224A A224A 1.9 71

(in excess)

"60-S* material.

served using another trichothecene antibiotic, fusa- renon-X (data not shown).

Poly(U)-directed polyphenylalanine synthesis in yeast extracts is also particularly sensitive to in- hibition by fusarenon-X [9,12] although ribosomes from CLP-8 or TR-1 were both equally resistant to the drug in vitro (Fig. 2 B). Furthermore, using reci- procal combinations of ribosomes and supernatant fractions from sensitive and resistant strains, resistance was shown to residue in ribosomes only (results not shown). The effect of fusarenon-X (20 pg/ml) in vitro using reciprocal Combinations of purified ribosomal subunits and supernatant fractions from trichodermin- sensitive and trichodermin-resistant yeast strains, was then studied (Table 1). Trichodermin resistance clearly resides in the 60-S ribosomal subunit of both CLP-8 and TR-1. Furthermore, a similar level of resistance resides in 60-S subunits from CLP-8 re- gardless of their source (804 ribosome or native 60-S* material). Addition of an excess of sensitive 60-S subunits (corresponding to the amount of native

180 A Yeast Mutant with an Abnormal Distribution of Ribosomal Subunits

030

#

J m n 3 0.15 6

C

0.30

E 5

2

c

a,

C m

B 0.E 6

0

A

40-s 605 00-s

1 1 1

Sedimentation direction P

0

40-S 60-S 00-S

I l l

Sedimentation direction P

0

40-S 60-S80-S

I l l

Sedimentation direction - E

40-5 60-S 80-S

0' sedimentation direction P

405 60-S80-S

0.30

$ 1

0- Sedimentation direction

0.30

z 8 N

m

m u c m

- e 5 0.15 4

C Sedimentat ion direction P

Fig. 3. Anulysis on sucrose Rrudients qf y u s l rihosomui subunits f r o m splieroplust cstruc'ts of A224A or CLP-8 supplemc.nted in vitro with a measured excess of either derived 40-S or derived 60-S rihosomul subunits f rom the corresponding yeast struin. Ribosomal subunits were prepared from yeast spheroplasts and analyzed on sucrose gradients as described in Materials and Methods. Ribosome-free supernatant fractions (analyses illustrated in A and B) were obtained by centrifuging spheroplasl extracts at 45000 rev./min for 4 h in A Spinco Ti 50 rotor at 4°C. Where excess ribosomal subunits were added to extracts, addition inimediate!y followed lysis by sodium deoxycholate and Brij 58, and the sample was incubated for 10 inin at 30 ?C with gentle shaking prior to chilling and subsequent analysis on sucrose gradients under low- salt conditions (unless otherwise stated). Profiles for A224A material (A) spheroplast extract alone (-) with corresponding supernatant fraction (----) superimposed; (B) 0.08 A260 unit 40-S ribosomal subunits + 0.14 A260 unit 60-S ribosomal subunits t spheroplast super- natant fraction analyzed under low-salt (-) and high-salt (----) conditions. Subunit profiles for A224A spheroplasts extracts supplemented and incubated with (C) 0.14 Az60 unit 60-S ribosomal subunits; (D) 0.08 Az60 unit 40-S ribosomal subunits. (E) Profile of 0.3 A260 unit 40-S ribosomal subunits (----) superimposed upon the normal subunit profile (-) for extracts from strain CLP-8. (F) Ribosomal subunit profile for CLP-8 spheroplast extracts incubated with 0.3Az60 unit 40-S ribosomal subunits. Essentially identical results were obtained irrespective of the source of added ribosomal subunits (A224A or CLP-8)

60-S* material found in CLP-8 extracts) to a normal A224A preparation also failed to induce any apparent resistance to fusarenon-X and the observed resistance seems, therefore, not to result from drug sequestration by excess 60-S material. Trichodermin resistance in CLP-8 presumably results, therefore, from an (un- identified) alteration in some component(s) of the 60-S ribosomal subunit and since the same level of resistance is also observed both in vivo and in vitvo with TR-1 it seems likely that this strain carries the same lesion.

Our observations suggest that the inbalance of native subunits observed with CLP-8 must result from a second lesion which may or may not be linked gene- tically to the trichodermin-resistant trait. We therefore carried out a genetic cross between CLP-8 and Y 166 and analyzed the tetrads for both trichodermin re- sistance and presence of native 60-S* ribosomal sub- units. Seven complete tetrads were analyzed and a consistent pattern was always observed, with two strains from each tetrad being trichodermin-sensitive and two strains being trichodermin-resistant. Sur-

C. J . Carter, M. Cannon, and A. Jimenez

priaingly, however, from each tetrad only one out of four strains possessed the native 60-S* trait and this strain was always trichodermin-resistant. Since tricho- dermin-resistant strains showing normal ribosomal subunit patterns were also observed it seems that both of the lesions can segregate independently. The lesion responsible for production of native 60-S* subunits is, however, apparently only expressed in the presence of the trichodermin-resistant marker. Indeed this interpretation is supported by the results of a similar analysis involving four complete tetrads from a second genetic cross between TR-1 and CLP-8. As expected all resulting spores gave rise to cells that were tricho- dermin-resistant. However, in each set two spores gave rise to cells containing native 60-S* subunits whereas the remainder revealed a normal ribosomal subunit pattern. We deduce that the same tricho- dermin-resistant marker is present in both CLP-8 and TR-1 but the interesting possibility arises that the parent yeast strain (A224A) from which CLP-8 is derived contains a lesion that is not expressed pheno- typically unless the genotype of the cell includes a mutation that controls trichodermin resistance. In combination, the two lesions give rise to the inbalance of ribosomal subunits seen in CLP-8. It is indicated by our genetic analysis that the postulated lesion in A224A is absent in Y166 which is the parent yeast strain from which TR-1 is derived.

The subunit inbalance in CLP-8 could result either from an overproduction of 60-S subunits or an under- production of 40-S ribosomal subunits and several lines of evidence do not contradict the latter possibility. Native ribosomal subunits do not normally recombine in the absence of protein synthesis. However, in Eschericliiu coli, this property is apparently de- termined only by the state of the 30-S subparticle [28] which possesses additional bound protein factors [29]. Accordingly the 50-S subparticle can be converted to a 70-S monomer in the presence of an added excess of derived 30-S subunits although the 70-S monomer does not form if the reciprocal cross is carried out (cf. [30]). A different situation does, however, exist in a mamma- lian system where both of the native ribosomal subunits are modified in such a way that prevents them combin- ing either with each other or with their corresponding derived subunits [31]. The 40-S subunits have been shown to possess additional proteins (anti-association factors) that can be removed from the particles by high-salt washing, and the derived subunit thus produced readily forms (after addition of derived 60-S particles) 80-S free couples under low-salt conditions [32]. It is envisaged that during initiation of protein synthesis the native 40-S subunit attaches to mRNA and the subsequent joining of a native 60-S subunit is accompanied by release of the anti-association factor protein(s). On completion of a nascent polypeptide, the ribosome is released as derived subunits which

Sedimentation direction -

Sedimentation direction -

181

I E 80-s

n l " Sedimentation direction -

I 40-s 60-s 80-s

0- Sedime n t at im direct ion P

Fig. 4. Antilysi.s o i i .sucrose grudieiits of polyribo.soiiirs ci t i t l rihosoimrl subunits f rom extracts of spl~eroplasts derived f rom yeast strains A224A und CLP-8 Jollowing polyribosome run ojf in vivo induced by sodium azide treatment. Spheroplasts were incubated with sodium azide (final concentration 1 mM) for 15 min before chilling and analysis as described in the Legend to Fig.l. (A) Polyribosome profile for spheroplasts of strain CLP-8. Ribosomal subunit profiles from spheroplasts of (B) A224A or (C) CLP-8 analyzed under low- salt conditions in both cases or (D) CLP-8 analyzed under high-salt conditions. Where the 80-S ribosome peak is 'cut' the absorbance of the peak above the 'cut' is off the scale indicated on the ordinate. the peak being drawn arbitrarily

may immediately re-associate and enter an 80-S free couple pool or bind anti-association factor to form native subunits capable of recycling in protein syn- thesis [31].

The situation in yeast resembles that described above since native ribosomal subunits from A224A extracts fail to combine, under low-salt conditions, not only with each other but also with the cor- responding derived subunits added in excess (Fig. 3 C, D). These two profiles should be compared with the control profiles shown as Fig.3A,B. By

182

E 0.2

3 c * I m a,

S 0.1

:: s L

0

A 80-5 40-S 60-S

sedimentation direction -

' Sedimentation direction -

0.2

8 c

a,

5 0.1 fl 8 2

0

A Yeast Mutant with an Abnormal Distribution of Ribosomal Subunits

Sedimentation direction Sedimentation direction P

E 4US 60-5 80-S

Sedimentation direction P

Fig. 5. Analysis on sui'i.o.w grudioirs o j 8&.T ribosomes rvuslietl iiiidei. /iigh-sulr wridiiions arid then imuburrd n . i r l i LI dim)ciuiiori f ucrorprepuration. High-salt-washed- 80-S ribosomes and a dissociation fHctor preparation obtained from native ribosomal subunils were prepared as described in Materials and Methods. 80-S ribosomes (0.2 AZho unit) were incubated with or without a dissociation Cactor preparation (225 p1 containing 440 pg protein/ml) at 30 "C for 30 min in a total reaction volume of 250 PI. The incubation mixture was chilled and analyzed by sucrose gradient centrifugation as described in Materials and Methods. Incubation mixtures contained 80-S ribosomes from A224A with (A) no factor added; (B) factor from A224A added; (C) factor from CLP-8 added. Incubation mixtures contained 80-S ribosomes from CLP-8 with (D) no factor added; (E) factor from CLP-8 added; (F) factor from A224A added

analogy any native 40-S subunit pool in CLP-8 would not be depleted by an overproduction of 60-S ribosomal subunits either native or derived. Addition of an excess of derived 40-S subunits to an extract from CLP-8 did, in fact, cause conversion of ap- proximately 80% of the native 60-S* material to the 80-S form (Fig. 3 E, F). Under the same experimental conditions, and as expected, derived 40-S and 60-S subunits from both A224A and CLP-8 also formed 80-S free couples (Fig. 3 B). Thus, no more than 20 7; of the CLP-8 native 60-S* peak corresponds to true native subunits and our evidence is consistent with the lesion in CLP-8 being in some way associated with the 40-S ribosomal subunit.

Although CLP-8 extracts appear to contain little or no native 40-S subunits this mutant does grow, albeit not as quickly as its parent (C. J. Carter, un- published work), and assuming that initiation of protein synthesis occurs normally, an adequate supply of these subunits must exist somewhere. However, if an underproduction of 40-S subunits does indeed occur, then the level of free active 40-S anti-as- sociation factor (in the cytoplasm) may be higher than normal. Previous studies [27] have demonstrated that frequent subunit exchange takes place iz7 vivo between 80-S free couples and native subunit pools (presumably through exchange of anti-association factor) so it should be possible to generate a normal

complement of native 40-S subunits in CLP-8 by in- creasing the amount of 80-S free couples. Accordingly, intact spheroplasts from both CLP-8 and A224A were incubated for 15 min at 30 'C in the presence of sodium azide (1 mM). For both strains, there was a progressive decrease in the size and number of poly- ribosomes with a corresponding accumulation of 80-S free couples, total polyribosome 'run off being achieved within 15 min (Fig.4A). Somewhat sur- prisingly there was no observable change in the amount of native subunits present (Fig. 4B, C), although for both strains the run off 80-S ribosomes (free couples) produced were fully dissociated under high-salt con- ditions (Fig.4D). Since the size of the native subunit pool in A224A remained essentially constant even after azide treatment (run off conditions), the amount of anti-association factor present is apparently limiting with all available protein already associated with the small pool of 40-S native subunits. The failure to generate any native 40-S ribosomal subunits in CLP-8 may result therefore from a (further) lesion affecting, in some way, the anti-association factor itself.

This possibility was investigated using a pre- paration containing dissociation factor activity pre- pared from the ammonium chloride Hash of a fraction from yeast extracts that was enriched in native sub- units [22]. Incubation of such a preparation from A224A with high-salt-washed 80-S ribosomes resulted

C. J . Carter, M. Cannon, and A. Jimenez 183

in the conversion of approximately 60% of these par- ticles into subunits over a 30-min period (Fig. 5A, B). This level was considerably reduced if the experiment was repeated using material derived entirely from CLP-8 (Fig. 5 D, E); using reciprocal combinations of factor preparations and ribosomes from both strains (Fig. 5 C, F) the lowered dissociation factor activity was shown to reside in the CLP-8 factor pre- paration. It seems possible that the activity we observe corresponds to that of the anti-association factor referred to above although we do not know whether the factor is present in reduced amounts in CLP-8 or has a relatively lower activity than that prepared from the wild-type strain.

In summary, therefore, we have shown that the trichodermin-resistant yeast strain CLP-8 has an inbalance of ribosomal subunits possibly resulting from a relative deficiency of 40-S native subunits in the cytoplasm. The antibiotic-resistant trait is not directly linked to the inbalance but the latter is only expressed in the presence of the trichodermin-resistant lesion which itself is associated with a property of the 6 0 3 ribosomal subunit. The strain also has a reduced ability to generate native 40-S ribosomal subunits from 80-S free couples although we do not know if this property is linked directly to the other two lesions observed. Further studies on these phenomena may well provide additional valuable information on the mechanism and control of ribosome function in eukaryotic cells.

The financial support of both the Medical Research Council of Great Britain and the Wellcome Trust is gratefully acknowledged.

REFERENCES 1. Davies, J. E. & Nomura, M. (1972) Annu. Rev. Genet. 6,

2. Jaskunas, S. R., Nomura, M. & Davies, J . E. (1974) in Ribo.~onic.s (Nomura, M.. Tissieres, A. & Lengyel, P., eds) pp. 333 - 368, Cold Spring Harbor Laboratory, New York.

3. Stomer, G. & Wittmann, €I. G . (1977) in Molecular Mechanisms of Protein Biosyntliesis (Wcissbach, H. & Pestka, S., eds) pp. 117-202, Academic Press, New York.

4. Helscr, T. L., Davies, J . E. & Dahlberg, J. E. (1972) Nut. New Bid . 235. 6 - 9.

203 - 234.

5. Lai, C.-J., Weisblum, B., Fahnestock, S. R. & Nomura, M . (1973) J . Mol. Biol. 74, 67-72.

6. Jimenez, A., Littlewood, B. & Davies, J. E. (1972) in Molecular Mechanisms of Antibiotic Action on Protein Biosyntkesis and Membranes (Muiioz, E., Garcia-Ferrandiz, F. & Vazquez, D., eds) pp. 292-306, Elsevier, Amsterdam.

7. Grant, P. G., Sinchez, L. & Jimenez, A. (1974) J . Bucteriol. 120, 1308-1314.

8. Sinchez, L., Vbzquez, D . & Jimenez, A. (1977) Mol. Gen. Genet. 156, 319-326.

9. Schindler, D . (1974) Nuture (Lond.) 24Y, 38-41. 10. Schindler, D., Grant, P. & Davies, J . (1974) Nuture (Lond.)

11. Jimenez, A. & Vazquez, D. (1975) Eur. J . Biochem. 54, 483-

12. Jimenez, A., Sanchez, L. & Vazquez, D. (1975) Biochim. Bio-

13. Jimenez, A., Sanchez, L. & Vazquez, D. (1975) FEBS Lett. 55,

14. Fresno, M., Jiminez, A. & Vizquez, D. (1977) Eur. J. Bio-

15. Fresno, M., Gonzilez, A. & Vlizquez, D. (1978) Biocliim.

16. Grant, P., Schindler, D . & Davies, J . E. (1976) Genetics, 83,

17. Carter, C. J . & Cannon, M. (1978) Eur. J . Biochem. 84, 103-

18. Cannon, M. & Jimenez, A. (1974) Biochem. J . 142, 457-463. 19. Udem,S.A. & Warner, J . R. (1972) J . Mol. Biol. 65, 227-242. 20. Cannon, M., Davies, J . E. & Jimenez, A. (1973) FEBS L.ett. 32,

21. Mortimer, R. K. & Hawthorne, D. C. (1966) Gmrtic.s, 53,165-

22. P&tre, J . (1970) Eur. J. Bioc/zem. 14, 399-405. 23. Udem, S. A. &Warner, J . R. (1973) J. Bid . Chem. 248, 1412-

24. Petermann, M. L. (1964) The Physical and Chemicul Properties

25. Falvey, A. K. & Staehelin, T. (1970) J. Mol. Biol. 53, 1-19. 26. Van der Zeijst, B. A. M., Kool, A. J . & Bloemers, H. P. J.

27. Kaempfer, R. (1969) Proc. Nut1 Acud. Sci. U . S . A . 68, 2458-

28. Parenti-Rosina, R., Eisenstadt, A. & Eisenstadt, J . M. (1969)

29. Subramanian, A. R. &Davis, B. D. (1970) Nature (Lond.J 228,

30. Noll, M. & Noll, H. (1972) Nut. New Biol. 238, 225-228. 31. Henshaw, E. C., Guiney, D. G. & Hirsch, C. A. (1973) J . Bid.

32. Ayuso-Parilla, M., Henshaw, E. C. & Hirsch, C. A. (1973) J .

248, 535 - 536.

492.

phys. Actu, 383,427-434.

53 - 56.

cheni. 72, 323 - 330.

Biophys. Actu, 518, 104- 112.

667-673.

111 .

277 - 280.

173.

1416.

uf Ribosomes, Elsevier, Amsterdam.

(1972) Eur. J . Biochem. 30, 15-25.

2462.

Nature (Lond.) 221, 363 - 365.

1273-1275.

Clzem. 248,4367 - 4376.

Biol. Chem. 248, 4386-4393.

C. J. Carter and M. Cannon, Department of Biochemistry, University of London King’s College, Strand, London, Great Britain WC2R 2LS

A. Jimi-nez, lnstituto de Bioquimica de Macromoleculas, Centro de Biologia Molecular, Consejo Superior de Investjgaciones Cientificas y Univcrsidad Autbnoma de Madrid, Facultad de Ciencias, Universidad Autonoma de Madrid, Instituto de Biologia del Desarrolo, Canto Blanco, Madrid-34, Spain