Cycloheximide resistance can be mediated through either ribosomal subunit

Pro. Nati. Aad. Si. USA Vol. 75, No. 7, pp. 31583162, July 1978 Biohemistry Cyloheximide resistane an be mediated through either ribosomal subunit [ Tetrahymena thermophila/poly(udireted poly(phenylalanine) synthesis/eukaryoti ribosomel CLAUDIA A. SUTTON, MANUEL ARES, JR.*, AND RICHARD L. HALLBERGt Setion of Botany, Genetis and Development, Cornell University, Ithaa, New York 14853 Communiated by Donald D. Brown, April 28, 1978 ABSTRACT Two yloheximideresistant mutants of Tetrahymena thermophila were analyzed to determine the site of their yloheximide resistane. The mutations in both strains had been previously shown to be genetially dominant and loated at separate loi (denoted ChxA and ChxB). Strains arrying these mutations were readily distinguished by the extent to whih they were resistant to the drug. The homozygous double mutant was more resistant than either single mutant. Cellfree extrats of wild type and of the three mutant strains, assayed for protein syntheti ativity by both runoff of natural mrna and poly(u)dependent phenylalanine polymerization, demonstrated that in vitro the mutants were arlmore resistant than the wild type. Further frationation of the ellfree systems into ribosomes and su ernates loalized yloheximide resistane to the ribosome for both ChxA and ChxB homozygotes. Ribosome dissoiation and pairwise subunit mixing in the in vitro system indiated that ribosome resistane was onferred by the 6S subunit from one strain whereas resistane in the other strain was mediated through the 4S subunit. This was further onfirmed by reonstrution of all four yloheximideresistane "phenotypes" by mixing ribosomal subunits from appropriate strains. This finding suggests that the mehanisms by whih these mutations onfer resistane to yloheximide are different. Cyloheximide is an antibioti that inhibits translation on eukaryoti ytoplasmi ribosomes. The primary site of inhibition has been reported as initiation (1, 2), elongation (35), and termination (6, 7). Oleinik (8) has reently shown that all energydependent steps in translation are sensitive to the drug; the relative effet of the drug on the different translation steps appears to vary depending on the yloheximide onentration. Similar onentrationdependent inhibition has been demonstrated in yeast (9). It has been suggested (7) that yloheximide may interfere with GTP binding or hydrolysis at eah of these steps, but this has yet to be shown. Rao and Grollman (1) demonstrated that the yloheximide resistane of Saharomyesfragilis (whih is naturally resistant to the drug) ould be loalized to the large subunit of the ribosome. This was aomplished by mixing ribosomal subunits of S. fragilis and S. erevislae (whih is sensitive to yloheximide) and showing that the in vitro resistane of the reonstituted ribosomes required the S. fragilis 6S subunit. The interpretation of this result by other investigators was that yloheximide resistane or sensitivity was a property of the 6S subunit, presumably beause yloheximide bound there (11). Diret evidene of yloheximide binding is laking but the results of Skogerson and Wakatama (12) and Somasunduran and Skogerson (13) with yeast ribosomes are onsistent with this interpretation. Mutations altering yloheximide sensitivity of ribosomes The osts of publiation of this artile were defrayed in part by the payment of page harges. This artile must therefore be hereby marked "advertisenmnt" in aordane with 18 U.S.C. 1734 solely to indiate this fat. have been desribed in yeast (1416), Neurospora (17), Physarum (18), and Chinese hamster ovary ell lines (19). Jimenez et al. (15) demonstrated that the yloheximide resistane of one of their mutants was attributable to an altered 6S subunit, and it was further reported that twodimensional gel eletrophoresis of this subunit revealed an altered protein. However, in a disussion of this mutation, gh2, MLaughlin (16) indiated that the moleular alteration of the 6S subunit is unknown. In this paper we desribe the preliminary haraterization of two different yloheximideresistant mutants of Tetrahymena thermophila. We show that both mutations onfer different degrees of resistane to the ribosome and that the resistane of one mutant is large subunitassoiated and that of the other is small subunitassoiated. The mehanisms by whih these two mutations might onfer resistane to yloheximide are disussed. EXPERIMENTAL PROCEDURES Strains, Media, and Culture Conditions. The mutant strains CU333, CU334, and CU335 (Cornell University stok designations) were homozygotes, derived from T. thermophila inbred strain B. Seletion of the ChxB mutation and onstrution of strains will be desribed elsewhere (M. Ares and P. Bruns). The wildtype strain used was B1868, mating type IV. For labeling of growing ells, ultures were grown at 3 in 1% Proteose peptone (Difo)/.3% Na4EDTA (Geigy). Cellfree extrats were prepared from ultures grown at 3 in 1% Proteose peptone/.25% yeast extrat/.3% Na4EDTA, to inrease the span of exponential growth. In Vivo Labeling. Exponentially growing ells, at a density of 813 X 14 ells per ml, were added to inubation tubes ontaining 'Ao vol of yloheximide and [3H]lysine (Amersham, 83 Ci/mmol) at 1 times the final onentration desired. Total reation volumes were 22 Al. Final [3H]lysine onentration was 5,uCi/ml. Reations were inubated at 3 for 3 min, at whih time dupliate 1lO samples were spotted onto Whatman 3MM filters. Filters were preipitated in 1% trihloroaeti aid and washed in boiling 5% trihloroaeti aid, old 5% trihloroaeti aid, 7% ethanol, and ether. Dried filters were assayed for radioativity in a toluenebased fluor. Reations ontaining yloheximide were ompared to the ontrol to give perentage inorporation; inorporation at 2.5 mm yloheximide (reproduibly 2% of the total) was assumed to be due to mitohondrial protein synthesis and was subtrated from eah point. CellFree Extrats. The rude ellfree system was devised Abbreviation: Hepes, N2hydroxyethylpiperazneN'2ethafefoni aid. * Present address: Department of Biology, University of California at San Diego, Lajolla, CA 9223. t To whom reprint requests should be addressed. 3158

Biohemistry: Sutton et al. by E. Palmer and extensively modified. Exponentially growing ells (35 X 15 ells per ml) were washed and pelleted. Paked ells were resuspended in 1 vol of old 2Q mm N2hydroxyethylpiperazineN'2ethanesulfoni aid (Hepes)/. 1 M KCI/3.5 mm Mg(OA)2/.25 M surose/6 mm 2meraptoethanol, ph 7.6, and homogenized with a tightfitting Doune homogenizer (812 passes) until well broken. Cell debris was pelleted at 14, X g for 15 min. The supernate was adjusted to 1 mm ATP (neutralized)/.25 mm GTP/1 mm dithiothreitol/1 mm reatine phosphate/reatine phosphokinase (5 tg/ml) and inubated at 3 for 3 min in order to allow release of ribosomes from mrna. The reation mixture was then hilled and passed through a Sephadex G25 olumn made up in 2 mm Hepes/.1 M KCI/4 mm Mg(OA)2/6 mm 2meraptoethanol, ph 7.6. The exluded volume was used in experiments to test the protein syntheti ativities of unfrationated ellfree extrats. To further frationate the extrat, the exluded volume was entrifuged at 1, X g for 6 min at 4 to separate ribosomes and supernate. Ribosomes were resuspended in the olumn elution buffer in l/2/4 their original volume. Unfrationated extrats, supernates, and ribosomes were frozen in small portions in liquid N2 and stored at 6 until used. Ribosome Dissoiation. To prepare ribosomal subunits, ribosome aliquots from above were thawed and layered on 153% (wt/vol) surose gradients ontaining.3 M KCl (sometimes.5 M)/6 mm MgCI2 (or 1 mm)/1 mm Tris, ph 7.5/6 mm 2meraptoethanol. Gradients were spun at 2,5 rpm at 3 for 16 hr in a Spino SW 27 rotor. These salt onditions will dissoiate runoff (inative) ribosomes but not polysomal ribosomes (C. Sutton, unpublished data). Frations orresponding to the large or small subunits (derived from 8S runoff ribosomes) were pooled, dialyzed against a 1fold exess of 1 mm Hepes/5 mm KCI/2 mm Mg(OA)2 for 6 min with one hange, and pelleted at 95, X g for 2 hr. Pellets were resuspended in H2 and frozen (6) until used. Ribosomal subunits prepared in this way gave good ativity reovery upon reassoiation, being 595% as ative as undissoiated ribosomes (see legend to Table 2). Poly(U)Direted Poly(Phe) Synthesis. Reations were generally arried out in 22..gl volumes and inubated at 3 for 3 min. The following omponents were mixed on ie: 1 vol of buffered energygenerating system plus amino aids, 1 vol of poly(u) plus yloheximide, 1 vol of supernate, and 1 vol of ribosomes (or 2 vol of unfrationated supernate). Ribosomes were added bak to the supernate at a onentration equal to or less than that at whih they had been reovered (usually 46 A2M units/ml of supernate). For reassoiation of ribosomes, large and small subunits were added at an A2M ratio of 2.5:1. The final onentration of reatants was 2 mm Hepes (ph 8.), 2 mm dithiothreitol, 1 mm ATP,.25 mm GTP, 1 mm reatine phosphate, 5 Mg of reatine phosphokinase per ml, 11 mm Mg(OA)2, 125 mm KCl, 19 unlabeled amino aids at 25 MM eah, [3H]phenylalanine (New England Nulear, 22 Ci/mmol) at 12.525 uci/ml, phenylalanine to bring total phenylalanine to 5,gM, 575 gg of poly(u) per ml, and yloheximide as indiated. Dupliate Whatman 3MM filters, pretreated with 1% trihloroaeti aid and 1% Casamino Aids, were spotted with 5S,l samples at intervals throughout the ourse of the reation. Filters were proessed as above. Total trihloroaeti aidpreipitable inorporation was alulated by using zerotime ounts as bakground. Control reations without ribosomes had less than 2% of the ounts of omplete reations. Control reations were run to determine the amount of ontamination by whole ribosomes in the subunit preparations. In general, subunits had less than 5% of the ativity of whole ribosomes. No subunit preparations with >1% ativity of undissoiated subunits were used in these experiments. Pro. Natl. Aad. Si. USA 75 (1978) 3159 Maximal polymerization of phenylalanine was about 15 pmol of phenylalanine per mg of ribosomes in 3 min. The reation was omplete by this time. Only a small fration (115%) of the ribosomes partiipate in poly(phe) polymerization as judged by the perentage of ribosomes that were resistant to dissoiation in.3 M KCI (results not shown). Therefore, the average extent of polymerization was six to seven phenylalanines per ribosome. In Vitro Polyribosome Runoff. Extrats were prepared in the same way as for the poly(u) reations through preparation of the 15, X g supernate. The supernate was then passed through Sephadex G25 in the olumn elution buffer, and the turbid exlusion volume frations were pooled. Final reatant onentrations were 2 mm Hepes (ph 8.), 2 mm dithiothreitol, 1 mm ATP,.25 mm GTP, 1 mm reatine phosphate, 5 ug of reatine phosphokinase per ml, 6 mm Mg(OA)2, 15 mm KCI, 25 mm eah of 19 unlabeled amino aids, [3H]phenylalanine (New England Nulear, 22 Ci/mmol) at 25,Ci/ml, unlabeled phenylalanine to 5 MuM, and.3.5 A26 unit of supernate. Dupliate 8,Ml reations were inubated at 3 for 3 min, at whih time the total reation mix was spotted on a pretreated Whatman 3MM filter. Filters were proessed as before. RESULTS Strain Phenotypes. CU333 arries a dominant mutation (ChxA) that onfers resistane to yloheximide (2). Two mutant alleles of the ChxA lous are known (21). When first obtained, the basis of resistane in these mutants was presumed to be nonribosomal, due in part to the fat that both mutations at the ChxA lous are strongly dominant (22). Under the assumption that the basis of resistane in ChxA/ChxA lines was likely a transport defet, a homozygous ChxA strain was mutagenized and lines resistant to even higher onentrations of yloheximide were seleted, in the hope that a ribosomal mutant might be obtained. The resistant line reovered from this seletion sheme was shown to arry a seond mutation, lous designation ChxB, whih alone aused weak yloheximide resistane (unpublished data). A strain homozygous for ChxB, CU334, was onstruted. Additionally, a homozygous double mutant, CU335, was made; the resistane phenotype of this line was idential to that of the initial isolate. Table 1 lists the strain nomenlature used throughout the paper and the genotypes and phenotypes in yloheximide. CU334 is weakly yloheximide resistant, CU333 is more resistant, and the double mutant strain CU335 is more resistant than its CU333 parent. To onfirm that these yloheximide onentrations reflet inhibition of protein synthesis, we per Table 1. Cyloheximide resistanes of the various strains Genotype/ Cyloheximide dose, AM Strain phenotype Seletion Min. lethal Wild type ChxA+ ChxB+/ 2 ChxA+ ChxB+ CU334 ChxA+ ChxB/ 25 36 ChxA+ ChxB CU333 ChxA ChxB+/ 9 3 ChxACHxB+ CU335 ChxA ChxB/ 5 64 ChxA ChxB Seletion doses are yloheximide onentrations used routinely to distinguish between lones of the four phenotypes in rosses and to allow resistant lines to grow while sensitive strains die. Minimum lethal doses were defined as the lowest yloheximide onentrations that would yield no viable (transferable) ells after 3 days' exposure to the drug.

316 Biohemistry: Sutton et al. 1k 8 4 fi (a6a 84 En 16~ *N A..1.1.1 1 1 Cyloheximide, pm "" 1 1 FIG. 1. Effets of varying onentrations of yloheximide on [3H]lysine inorporation in exponentially growing ells. *, Wild type; A, CU334;, CU333; &, CU335. formed an in vivo doseresponse study with exponentially growing ells. Fig. 1 shows the results of one suh experiment. Although there was not a perfet orrespondene between minimum lethal dose and maximal inhibition of inorporation in vivo, the results are in relative agreement. In both instanes, CU335 was about 2 times more resistant than CU334 and about 1.5 times more resistant than CU333. In Vitro Cyloheximide Sensitivity. As an assay for protein synthesis in vitro, poly(u)direted poly(phe) synthesis was measured in ellfree extrats with varying onentrations of yloheximide. By this assay we found (Fig. 2) that all three mutant strains were more resistant to yloheximide inhibition than was the wild type. Furthermore, CU335 extrats were more resistant than either CUM or CU334 extrats. Although the orders of magnitude of in vitro resistane were not the same as the in vvo resistanes (disussed below) the results were reproduible (three experiments) and suggested that dereased uptake was not the ause of resistane. To identify the resistant ell omponent we frationated the rude extrat. Ribosomes were pelleted from eah of the ellfree extrats, and yloheximide sensitivity of wildtype ribosomes was tested in eah of the four ribosomefree supernates. In addition, ribosomes from the four strains were assayed for yloheximide resistane in wildtype supernate. These mixing experiments should (i) onfirm the ytoplasmi loation of the resistane and (ii) loalize the resistane of the extrats to either the ribosome fration or to the postribosomal supernate. When wildtype ribosomes were tested in the four ribosomefree extrats, all four ombinations showed equivalent dose responses 'ru CL 8 6 4 21[ a Cyloheximide, jpm FIG. 2. Effets of varying onentrations of yloheximide on poly(u)direted poly([3h]phe) synthesis in unfrationated ellfree extrats. *, Wild type; A, CU334;, CU333; A, CU335. Pro. Natl. Aad. Si. USA 75 (1978) (Fig. 3 left). However, the reiproal mixing experiment demonstrated that the resistane seen in the rude ellfree extrats (Fig. 3 right) ould be generated by ribosomes from the three mutant strains. Comparison of Figs. 2 and 3 right shows that the doseresponse urves are nearly superimposable; thus, both quantitatively and qualitatively, the ribosomes possess the yloheximide resistane demonstrable in the poly(u) system. Validity of the Assay System. Whereas the in vivo yloheximide responses of the four strains were maximally separated by two orders of magnitude of yloheximide onentration (Fig. 1), the maximal resistane in vitro was found only at 1 times the onentration of yloheximide neessary to inhibit wildtype extrats (Figs. 2 and 3 right). However, a poly(u) system annot assay effets on any step in protein synthesis exept elongation (23, 24). The measurement of in vivo inhibition is the result of inhibition of all steps in protein synthesis. If elongation in wildtype ells is less sensitive to yloheximide than is either initiation or termination, as has been shown in Chinese hamster ovary ells (8), then the in vitro results need not agree quantitatively with the in vivo measurements. Furthermore, the poly(u)direted poly(phe) inorporations are not arried out under optimal onditions for natural mrna translation. To assay the inhibition of elongation under more "natural" onditions, we measured inhibition of inorporation in a ellfree polyribosome runoff system. With this system as our estimate of inhibition of elongation, the mutant extrats were more resistant than the wildtype extrat (Fig. 4). The resistane of CU335 was 1 times greater than that of the wild type; the values of CU333 and CU334 again fell betweefi'those for the wild type and the double mutant (results not shown). The atual levels of resistane in the runoff system were approximately 5 times greater than in the poly(u) system. Although natural initiation does not our in the latter system (24), the ribosomes must attah to the poly(u) in some manner that may be sensitive to yloheximide. This ould affet the absolute levels of resistane. However, beause the relative differenes were the same, we onluded that both in vitro systems measured similar or idential steps of protein synthesis. Even though we annot aount for the relative differenes seen when omparing the in vitro and in vivo dose responses, our data show that the ribosomes from the four strains differ in their yloheximide sensitivity at least one step in the protein syntheti pathway. Ribosomal Subunit Loalization of Resistane. Further analysis of the mutants was done with the poly(u) assay system to determine whih ribosomal subunit was required to onfer yloheximide resistane to the ribosome. Ribosomes from CU333, CU334, CU335, and wildtype extrats were dissoiated and reassoiated in all pairwise ombinations and tested for ativity at 1 AM yloheximide, a onentration that should disriminate among all possible ribosomal phenotypes (see Fig. 2). Tables 24 give the results of several experiments. First, yloheximide resistanes of homologous reassoiations were ompared to the resistane of the undissoiated ribosomes from whih they were derived. Table 2 indiates that (i) dissoiation and reassoiation does not ause a large loss in overall protein syntheti ativity of the ribosome and (ii) the in vitro yloheximide resistane is retained (in fat it is slightly enhaned). Thus, at 1.tM yloheximide, the four "phenotypes" are readily distinguishable from one another. The yloheximide sensitivity of hybrid ribosomes omposed of one wildtype and one mutant subunit learly suggested (Table 3) that resistane in CU333 was mediated through the large subunit and resistane in CU334 was onferred by the small subunit. This finding predits that ribosomes with in vitro phenotypes of either single mutant may be reonstruted from double mu

Biohemistry: Sutton et al. Pro. Natl. Aad. Si. USA 75 (1978) 3161 1t 1 a 8 I 4 6.E 4 2~~~~~~~~~~~~ a"'6 A 2. l._.1.25.5 1 2.5 5 1 25 5 1 Cyloheximide,uMM 2.1.25.5 1 2.5 5 1 25 5 1 25 5 Cyloheximide, JAM FIG. 3. Effets of varying onentrations of yloheximide on poly(u)direted poly([3h]phe) synthesis in reonstruted ellfree extrats. (Left) Wildtype ribosomes ombined with ribosomefree extrats of wild type (), CU334 (A), CU333 (), and CU335 (A) ells. The inorporation values taken as 1% (in pm/od ribosomes) in the various reations were: wildtype, 23,829; CU334, 23,39; CU333, 18,446; CU335, 28,754. (Right) Wildtype ribosomefree extrats ombined with ribosomes of wild type (), CU334 (A), CU333 (), and CU335 (A) ells. The inorporation values taken as 1% (in pm/od ribosomes) in the various reations were: wild type, 23,829; CU334, 24,451; CU333, 24,54; CU335, 24,492. tantwildtype subunit ombinations and that phenotypially double mutant and wildtype ribosomes may be formed by using CU333 and CU334 subunit ombinations. These hybrid ombinations, their predited levels of resistane, and the atual resistanes found are shown in Table 4. Exept for the third ombination set (CU334 + CU335 subunits), whih gave higher resistanes than predited (although relative resistanes are still onsistent), there is good agreement between atual and predited resistanes. We onlude that CU334 resistane in vitro requires the CU334 4S subunit, CU333 ribosomal resistane is onferred by its 6S subunit; and CU335 in vitro resistane is due to ombined effets of the two subunits. DISCUSSION Two yloheximideresistane mutants of T. thermfophila have been shown to have resistant ribosomes in vitro. Resistane of CU333 ribosomes requires the presene of the 6S subunit. Given the urrent information on the probable yloheximide binding site (or sites) (12, 13) and the ribosomal subunit loation of other yloheximideresistant mutants (1, 15), this result is not surprising. CU334 resistane, however, is onferred by the 45 subunit. If there is not a yloheximide binding site on the small ribosomal subunit, then another means of onferring in vitro resistane must be invoked. We have evidene that CU334 ribosomes are onformationally altered relative to wild 14 Cyloheximide, MM FIG. 4. Effets of varying onentrations of yloheximide on [3H]phenylalanine inorporation into polypeptides synthesized in vitro by polyribosome runoff. Polyribosomes were from wildtype () and CU335 (A) ells. type (25). Assuming that the CU334 4S ribosomal subunit is hanged in struture, one an imagine that the affinity of yloheximide for the ribosome is also different. This ould our either by hanging the onformation of the entire ribosome or by masking the yloheximide binding site or both. Is the in vio resistane of the two strains due entirely to the demonstrated in vitro ribosomal resistane? Genetially, eah mutant behaves as if it were the result of a mutation at a single lous; typial Mendelian ratios are found in rosses with these strains (unpublished data). Although pleiotropi mutations that affet ribosomal resistane to antibiotis and other physiologial funtions at the ell membrane exist in bateria (26), they are unommon. It seems unlikely that we have pleiotropi mutations like the above. However, until yloheximide uptake and internal onentration an be measured in our strains we annot rule out the possibility of the exsistene of losely linked mutations that alter yloheximide transport (or toxiity) in intat ells. However, assuming that the ribosomal resistane deteted Table 2. In vitro assays for yloheximide resistane: Maintenane of resistane in reassoiated subunits* Ativity in 1,gM yloheximide, % of ontrol Strain Undissoiated Reassoiated phenotypet ribosomes subunitst ++ 15.4 ± 7.4 22. ± 6.4 (n =7) (n =1) +B 31.2 ± 6.7 36. ± 7.9 (n =6) (n =9) A+ 33. ± 18 48.3 ± 7.4 (n =3) (n =3) AB 53.2 7.7 69.1±8.5 (n =5) (n =3) * Undissoiated runoff ribosomes and purified ribosomal subunits were used in poly(u)direted poly(phe) synthesis reations with (1 AM) and without (ontrol) yloheximide. Typial ontrol reations with undissoiated ribosomes gave 812 pmol of phenylalanine inorporation per mg of ribosomes in a 3min inubation. All values are means + SD obtained from independent ribosome isolations. t ++, wild type; +B, CU34; A+, CU33; AB, CU335. * In 2 diret omparisons of reassoiated subunits with the preparations of undissoiated ribosomes from whih they ame, we reovered 71.3 ± 11.2% of the protein syntheti ativity in the reassoiated subunits relative to the undissoiated ribosomes in reations without yloheximide; the range was 579%.

3162 Biohemistry: Sutton et al. Table 3. In vitro assays for yloheximide resistane: Loalization of yloheximide resistane by subunit mixing Subunit soure* Ativity in (strain phenotype) 1,uM yloheximide, Large Small % of ontrol +B ++ 24.2 8. (n =9) ++ +B 41.2 + 6.5 (n =6) A+ ++ 59. 2.8 (n = 2) ++ A+ 33.5 +.5 (n = 2) * As in Table 2 footnote t. in vitro is responsible for resistane in whole ells, learly our assay method is not optimal. CU334 and CU333 have different phenotypes in vivo; their behavior in vitro is similar in terms of relative yloheximide resistane. Beause yloheximide is not primarily an inhibitor of elongation in other organisms (see ref. 8), it is likely that the two mutants respond differently to the drug at either initiation or peptide hain release. Unfortunately, we urrently have no assay system for these steps. Prior to these experiments, the ChxA lous was onsidered to be an unlikely andidate for a ribosomal funtion beause mutations at the lous are dominant. Ribosomal drug resistane in prokaryotes is reessive (27). Reessive ribosomal drug resistane has also been shown for Chinese hamster ovary ell lines (28) and for yeast (29). Interestingly, the yloheximideresistant ribosomal mutations desribed in yeast (15), Neurospora (17), and Physarum (18) do not behave as true reessives, either in yeast diploids or in heterokaryons of Neurospora and of Physarum. A trivial explanation for semidominane in the latter Table 4. In vitro assays for yloheximide resistane: Reonstrution of ribosome phenotypes by subunit mixing Subunit soure* Ativity in 1MM yloheximide, (strain phenotype) % of ontrol Large Small Preditedt Found +B A+ 22 28. 12.5 (n = 3) A+ +B 69 6.3 2.3 (n = 3) ++ AB 36 4 AB ++ 48 5 +B AB 36 59 AB +B 69 79 A+ AB 69 64 AB A+ 48 47 * As in Table 2 footnote t. t Assuming the CU334 small subunit and the CUM large subunit to be the mediators of yloheximide resistane, one an predit the "phenotype" of a ribosome reonstruted from two different ribosomes. The level of resistane found in Table 2 for reassoiated subunits in 1,uM yloheximide was taken as the "phenotype" expeted+ = 22%; CU334 = 36%; CU3M3 = 48%; CU335 = 69%. Pro. Natl. Aad. Si. USA 75 (1978) two organisms involves unequal nulear ratios. However, a possible explanation for the semidominane of yloheximide resistane may be the site of ation of the drug. If inhibition ours prior to the formation of the mrnaribosome initiation omplex, then sensitive ribosomes will be seletively exluded from polyribosomes. Thus, resistant ribosomes will preferentially translate mrna, and heterozygotes will still have funtioning protein syntheti mahinery. However, until ribosomal yloheximideresistant mutants that have lear alterations in ribosomal strutural proteins are found, this hypothesis annot be tested. We thank Ed Palmer for developing the in vitro system for Tetrahymena and Peter Bruns for the use of his Tetrahymena failities. This work was supported by U.S. Publi Health Servie Grant HD6448 to R.L.H.; C.A.S. is a predotoral genetis trainee supported by U.S. Publi Health Servie Grant GM135. 1. Baliga, B. S., Pronzuk, A. W. & Munro, H. N. (1969) J. Biol. Chem. 244, 4484489. 2. Obrig, T. C., Culp, W. J., MKeehan, W. L. & Hardesty, B. (1971) J. Biol. Chem. 246, 174181. 3. Stanners, C. P. (1966) Biohem. Biophys. Res. Commun. 24, 758764. 4. Lodish, H. F. (1971) J. Biol. Chem. 246,71317138. 5. Kisilevsky, R. (1972) Biohim. Biophys. Ata 272, 463472. 6. Godhaux, W., Adamson, S. D. & Herbert, E. (1967) J. Mol. Biol. 27,5772. 7. Rajalakshmi, S., Liang, H., Sarma, D. S. R., Kisilevsky, R. & Farber, E. (1971) Biohem. Biophys. Res. Commun. 42, 259 265. 8. Oleinik, N. L. (1977) Arh. Biohem. Blophys. 182, 17118. 9. Cooper, T. G. & Bossinger, J. (1976) J. Biol. Chem. 251, 7278 728. 1. Rao, S. S. & Grollman, A. P. (1967) Biohem. Biophys. Res. Commun. 29, 69674. 11. Warner, J. R. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), pp. 461488. 12. Skogerson, L. & Wakatama, E. (1976) Pro. Natl. Aed. Si. USA 73,7376. 13. Somasunduran, U. & Skogerson, L. (1976) Biohemistry 15, 4764764. 14. Cooper, D., Banthorpe, D. V. & Wilkie, D. (1967) J. Mol. Biol. 26,347349. 15. Jimenez, A., Littlewood, B. & Davies, J. (1972) in Moleular Mehanisms of Antibioti Ation on Protein Synthesis and Membranes, eds. Munoz, E., GariaFerrandiz, F. & Vasquez, D. (Elsevier, New York), pp. 29236. 16. MLaughlin, C. S. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), pp. 815827. 17. Pongratz, M. & Klingmuller, W. (1973) Mol. Gen. Genet. 124, 359363. 18. Haugli, F. B. & Dove, W. F. (1972) Mol. Gen. Genet. 118,97 17. 19. Pohe, H., Junghahn, I., Giessler, E. & Bielka, H. (1975) Mol. Gen. Genet. 138, 173177. 2. Byrne, B. C. & Bruns, P. J. (1974) Genetis 77, s. 78 (abstr.). 21. Bleyman, L. K. & Bruns, P. J. (1977) Genetis 87,275284. 22. Roberts, C. T. & Orias, E. (1973) Exp. Cell Res. 81, 312316. 23. Crystal, R. G., Elson, N. A. & Anderson, W. F. (1974) in Methods in Enzymology, eds. Moldave, K. A. & Grossman, L. (Aademi, New York), Vol. 3, pp. 11123. 24. Falvey, A. K. & Staehelin, T. (197) J. Mol. Biol. 53, 119. 25. Hallberg, R. L. & Sutton, C. A. (1978) J. Supramol. Strut. Suppl. 2, 334 (abstr). 26. Mizuno, T., Yamada, H., Yamagata, H. & Mizushima, S. (1976) J. Bateriol. 125,52453. 27. Lederberg, J. (1951) J. Bateriol. 61,54955. 28. Gupta, R. S. & Siminovith, L. (1977) Cell 1, 6166. 29. Grant, P., Sanhez, L. & Jimenez, A. (1974) J. Bateriol. 12, 1381314.