Eur. J. Biochem. 72,491 - 500 (1977)
Subcellular Distribution of Aminoacyl-tRNA Synthetases in Various Eukaryotic Cells Michael A. USSERY, Wesley K. TANAKA, and Boyd HARDESTY Clayton Foundation Biochemical Institute, University of Texas at Austin (Received July 20/0ctober 5 , 1976)
The total amount, size distribution and binding of aminoacyl-tRNA synthetases to ribosomes in a variety of mammalian and avian cells was studied under standard conditions of sample preparation and assay. Aminoacyl-tRNA synthetases appear to exist in three general forms; ‘free’ enzyme of about 4 - 9 S, one or more ‘enzyme complexes’ of about 18 - 25 S, and in association with ribosomes. The aminoacyl-tRNA synthetase activity for many individual amino acids was surprisingly similar in cell types chosen to be diverse with respect to differentiation state, transformation, and growth rate. Total activity for all amino acids varied about 4-fold, based on a constant volume of cells. Embryonic tissues had a comparatively high proportion of total synthetase activity associated with ribosomes, whereas this value was relatively low for mouse liver. Distinctive distribution patterns with common and variable features were observed for individual enzymes. The only aminoacyl-tRNA synthetases found not to be associated in significant amounts with either 18 - 25-S enzyme complexes or ribosomes in any of the cell types examined were the enzymes for alanine, histidine, and serine. All cell types evidenced 18 - 25-S synthetase activity for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, proline, and valine, although in quite variable proportions of the total activity observed for these amino acids. For example, of the valyl-tRNA synthetase activity not associated with ribosomes, 35% and 100% were found to sediment at 18 -25 S in Friend leukemia cells and mouse liver respectively. All cells had two easily distinguishable peaks of arginyltRNA synthetase activity at 4 - 9 S and 18 - 25 S respectively; however, the relative proportion of enzyme activity in the peaks differed between cell types. Phenylalanyl-t RNA synthetase was not observed to occur in an 18 - 25-S complex in any of the cell types examined but was bound to ribosomes in variable but generally relatively high proportions. Numerous other specific differences are described. No underlying physiological or biochemical principle has been recognized to account for the specific distribution patterns observed. However, they may reflect variations in cellular architecture that may be related to regulation of protein synthesis.
Eukaryotic aminoacyl-tRNA synthetases have been examined in a number of laboratories. Bandyopadhyay and Deutscher [l] have partially purified a multienzyme complex from rat liver 105000 x g supernatant that contained aminoacyl-tRNA synthetase activities for all 18 amino acids tested. Vennegoor and Bloemendal [2] have partially purified an 18.2-S aminoacyltRNA synthetase complex from rat liver which contained activities for glutamine, isoleucine, leucine, lysine, and methionine. Som and Hardesty [3] have similarly purified a complex from rabbit reticulocytes which sediments at approximately 16 S and contains Enzymes. Aminoacyl-tRNA synthetases (EC 6.1.1); E. coli plactosidase (EC 3.2.1.23); lactate dehydrogenase (EC 1.1.1.27); catalase (EC 1.11.1.6).
activities for arginine, isoleucine, leucine, lysine, and methionine. Several workers have also reported the binding of aminoacyl-tRNA synthetases to ribosomes. Roberts and Coleman [4] found most of the phenylalanyl-tRNA synthetase activity in Ehrlich ascites cells in two forms : bound to the 60-S subunit of the ribosomes and in a complex that sediments at about 25s. Hampel and Enger studied the subcellular distribution of aminoacyl-tRNA synthetases in Chinese hamster ovary cells by differential centrifugation and found several aminoacyl-tRNA synthetase activities cosedimenting with the ribosomes [ 5 ] . Work with rabbit reticulocytes has indicated that significant aminoacyl-t RNA synthetase activities for phenylalanine, lysine, isoleucine, leucine,
492
methionine, and arginine were bound to the ribosomes 13,6,71. Thus, there is considerable evidence for the existence of an aminoacyl-tRNA synthetase complex and ribosome-associated synthetases from a number of eukaryotic organisms. These results have led to the stated conclusion or unstated implication that the observations reflect organization of these enzymes in intact cells and that this organization may play a role in regulating protein synthesis. The reported differences may reflect variations in the differentiation and physiological state of cells or tissues. However, the reporting laboratories have employed a wide range of analytical techniques thus making a direct comparison tenuous. The object of the present study is to examine the aminoacyl-tRNA synthetase distribution in various eukaryotic cells by using a standard procedure for rupture and fractionation of the cells. The results of these experiments suggest an overall qualitative similarity between very diverse cell types (normal adult, embryonic, transformed, and highly differentiated) with respect to the presence of synthetase complexes and with respect to synthetases bound to ribosomes. However, significant quantitative differences were observed for specific enzymes. These differences are considered with respect to their potential significance in the overall scheme of protein synthesis.
MATERIALS AND METHODS
Reagents Uniformly 14C-labeled amino acids and ultrapure grade sucrose were obtained from Schwarz-Mann, Inc. (Orangeburg, NY); dithioerythritol, 2-mercaptoethanol, trichloroacetic acid, 2,5-diphenyloxazole were from Sigma Chemical Company (St Louis, MO); bovine serum albumin was from Armour Pharmaceutical Company (Kanakee, IL).
Cells and Tissue The Friend leukemia cells (Friend leukemia virustransformed Murine proerythroblastic line, FSD l/clone F4) were kindly provided by Dr W. Ostertag (Max-Planck-Institut fur experimentelle Medizin, Gottingen, Germany). HeLa S 3 cells were kindly provided by Dr G. Koch (Roche Institute of Molecular Biology, Nutley, NJ). These cells were grown in suspension culture with Eagle's minimal essential medium (Grand Island Biological Co., Grand Island, NY) supplemented with non-essential amino acids and 10% fetal calf serum and antibiotics, as previously described [8]. Viable cell number was determined by exclusion of erythrosine. In all cases, unless otherwise noted, cells were harvested in logarithmic growth phase.
Eukaryotic Aminoacyl-tRNA Synthetases
CFW mice (E and R Enterprises, Roundrock, TX) were used as a source of mouse embryo and liver tissue. Rabbit reticulocytes were prepared as described previously [9]. Chicken embryos were generously provided by Dr H. R. Bose (Dept. of Microbiology, University of Texas, Austin, TX).
Dimethyl Sulfoxide Stimulation of Friend Leukemia Cells Friend leukemia cells were grown to a cell density of 1.O x lo6 cells/ml. These cells were then diluted 1 :1 with fresh medium. Dimethyl sulfoxide was added to a final concentration of 1.5% (v/v). Cells were allowed to grow in the presence of dimethyl sulfoxide for 4 days maintaining a cell density of 0.5 - 1.O x lo6 cells/ml.
Aminoacyl-tRNA Synthetase Assay Reaction mixtures for all synthetases except phenylalanyl-tRNA synthetase contained 100 mM TrisHC1 (pH 7.5), 3.0 mM MgCl,, 20 mM 2-mercaptoethanol, 2.0mM ATP, 0.1 mM I4C-labeled amino acid (10 Ci/mol), 200 pg rabbit liver tRNA, and 20 pg bovine serum albumin in a total volume of 0.2 ml. This standard aminoacylation mixture was developed to be as close as possible to the measured pH and Mg2+: ATP ratio optima for the 18 synthetases (excluding phenylalanyl-tRNA synthetase [6]) studied. Several preparations of [14C]tryptophan were tested but all gave consistently high blanks which precluded reliable measurement of tryptophanyl-tRNA synthetase activity. The reaction mixture for phenylalanyl-tRNA synthetase was similar except that it contained 100 mM Tris-HC1 (pH 8.5), 10 mM MgCI,, 100 mM KC1 and 4.0 mM ATP and the other listed reagents. For assay purposes a suitable aliquot, usually 10 p1 of enzyme fraction was added to the reaction mixture, mixed thoroughly, then incubated at 37 "C for 5 min. The reaction was stopped by addition of 5 ml of cold 5 % (w/v) trichloroacetic acid. The resulting precipitate was collected by filtration through millipore filters (0.45 pm, HAWG, Millipore Corp., Bedford, MA) and washed with a total of 15 ml of 5% (w/v) trichloroacetic acid. The filters were then dried in an oven for 5 min at 120 "C. Dried filters were put into plastic vials containing 10 ml of scintillation solution which contained 5 g of 2,5-diphenyloxazole in 1 1 of toluene. Vials were counted in a Beckman LS-150 scintillation counter with an efficiency of 85 %. Synthesis of aminoacyl-t RNA is proportional to added enzyme in the range 50 - 200 pmol product formed. A unit of enzyme activity is defined as the amount of enzyme required to form 1 nmol of aminoacyl-tRNA per min.
M. A. Ussery, W. K. Tanaka, and B. Hardesty
Preparation and Centrifugation of Lysates
Embryo tissues (15 mouse embryos, four chick embryos) or mouse liver (10 g wet weight) were minced thoroughly before further preparation. Rabbit reticulocytes, Friend leukemia cells and HeLa cells were collected by centrifugation at 4000 rev./min in a GS3 Sorvall rotor for 15 min. Cells and minced tissues were washed twice with 100 ml of a solution containing 130 mM NaC1, 5 mM KCl, 7.4 mM MgCl, and lysed by douncing 20 times with a glass douncer after addition of 1 vol. of H,O to 1 vol. of packed cells. The nuclei, mitochondria, and membranes were removed by centrifugation at 10 800 x g for 15 min. The supernatant from this centrifugation is referred to hereafter as the ‘lysate’. To determine the distribution of synthetases not bound to the ribosomes, a 1.0-ml aliquot of the resulting lysate was loaded on a 11.2-ml 10 - 30% linear sucrose density gradient containing 16 mM KH,PO,, 84 mM Na,HPO, (pH 7.5), 5 mM 2-mercaptoethanol, 1 mM dithioerythritol, and centrifuged in an SW41 rotor at 35000 rev./min for 16 h. Fractions containing 1 ml were collected with an Isco model D density gradient fractionator (Instrumentation Speialities Company, Lincoln, NB). Preparation of Ribosomal Fraction and Postribosomal Supernatant
A 2.0-ml aliquot of the lysate from the l0800xg centrifugation was loaded on a 9.5-ml cushion of 30% sucrose containing 25 mM KCl, 20 mM Tris-HC1 (pH 7.5), 5 mM 2-mercaptoethanol, 5 mM MgC1, and centrifuged in a 50Ti rotor at 50000 rev./min for 2 h. These centrifugation conditions were found to prevent the contamination of the ribosomal pellet by the larger (18 - 25-S complex) form(s) of free synthetases. A KC1 concentration of 25 mM prevented nonspecific binding of synthetases to ribosomes which was found to occur at 0 mM KCl but allowed specific binding of purified synthetase complex and phenylalanyl-tRNA synthetase to ribosomes which were removed only at KC1 concentration greater than 100 mM. The supernatant (postribosomal supernatant) was collected and the ribosomal pellet was washed with a buffer containing 5 mM 2-mercaptoethanol, 10 mM MgCl, ,20 mM Tris-HC1 (pH 7.5) and 15% glycerol and resuspended in 0.5 ml of the same buffer. Debris were removed from the resuspended ribosomes by centrifugation at 10800 x g for 15 min. The percentage of activity bound to the ribosomes was determined by dividing the activity associated with the ribosomes per ml lysate by the ribosomal activity per ml lysate plus the postribosomal activity per ml lysate and multiplying the result by 100. The concentration of ribosomes per ml of lysate differed by more than 4-fold between the extremes for the cell types tested (cf. Friend leukemia
493
cells and HeLa cells). The basis for this surprising difference is not clear but has been observed in a number of experiments. Assay of Sucrose Gradients
Samples containing 0.1 mg E. coli P-galactosidase (PL Biochemicals, Inc., Milwaukee, WI), 0.123 mg beef heart lactate dehydrogenase (Worthington Biochemical Corp., Freehold, NJ), 0.07 mg beef liver catalase (Worthington) and 0.9 mg of rabbit liver tRNA [9] were used as standards for determining sedimentation values for various portions of the sucrose gradient. P-Galactosidase was assayed by hydrolysis of o-nitrophenyl-8-galactopyranoside(Sigma Chemical Co., .St Louis, MO) [lo]; lactate dehydrogenase was assayed by oxidation of NADH (Calbiochem, La Jolla, CA) [ll], and catalase was assayed according to the method of Beers and Sizer [12]. The sedimentation of rabbit liver tRNA was determined by absorbance at 260 nm. All gradients were routinely assayed for arginyltRNA synthetase activity and, in some cases, for the activity of the various standards. Fig. 2 shows a typical distribution pattern for the two forms of arginyl-tRNA synthetase in sucrose density gradients along with the sedimentation position of the various standards. Fractions 3-6 represented the 4-9-S region of the gradient and was routinely located by the position of small ‘free’ form of arginyl-tRNA synthetase which sediments at approximately 4 S. The 18 - 25-S region of the gradient was contained in fraction 8 - 10. This region was routinely located by the position of the large ‘complex’ form of arginyl-tRNA synthetase. The percentage of the postribosomal activity sedimenting in the 18 - 25-S region was determined by dividing the total activity in fraction 8-10 per ml lysate by the activity in the postribosomal supernatant per ml lysate and multiplying the result by 100. RESULTS Lysate A ct ivities
Aminoacyl-tRNA synthetase activities in lysates of the various cells and tissues exhibit an overall similarity with some notable exceptions, as seen in Table 1. Total aminoacyl-tRNA synthetase activities of Friend leukemia cell lysates were over twice as high as the total average values for all of the cell types tested. The value for chicken embryo was one half of the total average value. Activities for some enzymes appeared to vary independently. The activity for asparagine in Friend leukemia cells was over six times larger than the average for the other cell types but about equal or below this average with respect to proline and tyrosine. Rabbit reticulocytes had relatively high levels of serine but were low in activity for
494
Eukaryotic Aminoacyl-tRNA Synthetases
Table 1. Aminoacyl-tRNA synthetase activities of lysates Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
5.0 34 7.4 35 3.5 6.8 5.1 8.6 8.8 4.1 16 39 7.8 72 0.86 26 12 4.0 5.2
2.4 26 6.0 14 3.6 2.8 3.4 4.5 6.1 1.9 5.1 20 3.2 39 5.4 24 2.6 0.80 4.0
2.2 14 4.4 11 1.2 2.0 3.0 4.3 1.6 1.8 2.6 25 2.0 25 1.4 19 2.8 0.70 1.6
Friend leukemia cell
HeLa cell
Average
units/ml lysate
7.4 20 8.7 8.0 2.9 2.7 1.3 6.3 4.8 3.0 3.4 13 4.8 64 1.8 60 13 0.90 3.5
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine ~~
Total Average
14 60 45 28 3.9 9.6 12 21 13 7.7 12 62 12 129 1.5 76 15 1.5 14
5.3 15 10 0.95 1.6 8.6 2.7 5.6 2.6 3.8 3.0 18 1.9 21 2.4 11 16 0.28 1.9
6.1 28 14 16 2.8 5.4 4.6 8.4 6.2 3.7 7.0 30 5.3 58 2.2 36 10 1.4 5.0
537.12 28
131.6 6.9
249.5 13
~
229.5 12
301.2 16
125.6 6.6
174.8 9.2
Table 2. Comparison of aminoacyl-tRNA synthetase activities of lysasates based on protein concentration and packed cell volume Protein concentration was determined by the Biuret procedure essentially as described by A. G. Gornall et a. 1131. All relative values are compared to the HeLa lysate. Total lysate activity is the total units of activity of the 19 aminoacyl-tRNA synthetase/ml lysate (from Table 1). Total specific activity is the total lysate activity divided by the protein concentration Cells
Protein concentration
Total lysate activity
Total specific activity
HeLa Chicken embryo Friend leukemia Mouse embryo Mouse liver Reticulocyte
mg/ml 22 36 39 45 110 170
U/ml (relative) 132 (1.0) 126 (1.0) 537 (4.1) 175 (1.3) 301 (2.3) 230 (1.7)
U/mg (relative) 6.0 (1.0) 3.5 (0.58) 13.8 (2.3) 3.9 (0.65) 2.7 (0.45) 1.4 (0.23)
(relative) (1.0) (1.6) (1.8) (2.0) (5.0) (7.7)
glutamine, glutamic acid, and lysine. Mouse liver was relatively rich in activity for aspartic acid, leucine, methionine, and tyrosine, but low in activity for proline. In contrast, mouse embryo tissue was unusually rich in proline activity but low in alanine, isoleucine, and threonine synthetase activities as was chicken embryo. Chicken embryo synthetases for histidine, leucine, and valine also evidenced lower than average activities. HeLa cell lysates generally were less active than average, with synthetases for aspartic acid histidine, leucine, methionine, serine, tyrosine and valine being exceptionally low. The basis for the variation between cell types is not known. In so far as was technically possible, standard conditions were maintained in preparing materials from the various sources. The average standard deviation for repetitive fractionation and assays
carried out on reticulocyte lysates for seven separate preparations was 11 %. The relative distribution of activities within successive preparations from a single cell type and between cell types has been reproducible under a rather wide range of experimental conditions used during evolution of the procedures described in this report. The synthetase activities are reported here in terms of units of activity per ml lysate. This value is directly proportional to the packed cell or tissue volume. The data were also computed as units of enzyme activity per mg protein in the lysate. The parameters required for this calculation and a comparison of the total units of activity for the 19 amino acids tested expressed per ml of lysate and per mg of protein in the lysate is given in Table2. The total activities of the lysates vary about 4-fould expressed on the basis of lysate volume but 10-fold when based on
M. A. Ussery, W. K. Tanaka, and B. Hardesty
495
Table 3. Aminoacyl-tRNA synthetase activities of Friend leukemia cell lysates Cell lysate
Synthetase activity for arginine leucine
U/ml ~ _ FSD-l/cione F 4 logarithmic phase stationary phase
_
lysine
methio- phenylnine alanine
_
81
17
80
17
110
8.5
19
119
18
96
5
58
10
61
10
75
14
107
Dimethylsulfoxidestimulated Expt 1 (80% viability) 15 Expt 2 (95% viability) 36
protein concentration. This is due to the relatively high protein concentration in mouse liver and reticulocyte lysates. Generally the specific activity (units per mg protein) decreases with increasing protein concentration in the lysate. An exception is the highly active Friend leukemia cell lysate. Variability in synthetase activity does not seem to be directly related to cell division time since Friend leukemia cells show similar synthetase activities in logarithmic growth phase or after 3 days in stationary phase (Table 3). Differences in the activities for log phase cells given here with those shown in Table 1 reflect variation in results from separate experiments. It is not clear whether this variation is due to differences in the cells or is inherent in the procedure used. Determination of the packed cell volume in preparations of these lysates is a recognized source of error. Friend leukemia cells can be stimulated to differentiate along the erythroid pathway and produce hemoglobin and Friend leukemia virus by incubation with 1.5% dimethyl sulfoxide [14]. The effect of this stimulation on synthetase activity also is shown in Table3. Cells stimulated to produce hemoglobin by dimethyl sulfoxide have somewhat lower levels of all synthetase activities examined as compared to the same cells which have not been exposed to dimethyl sulfoxide. Arginyl-tRNA synthetase activity is particularly sensitive to this treatment. Whether the lower level of activity of these enzymes is due to changes in the cellular physiology and architecture or to side effects of dimethyl sulfoxide on the aminoacyl-tRNA synthetases themselves is not known. Ribosome-Associated Synthetases The aminoacyl-tRNA synthetase activities associated with the ribosomes were determined with ribosomes that were pelleted from the cell lysate by
high-speed centrifugation through a sucrose solution containing 25 mM KCI as described in Methods. The ribosomes were resuspended and the preparations assayed directly for aminoacyl-tRNA synthetase activity as described in Methods. The resulting data are presented in Table4 as the percentage of the total lysate activity associated with the ribosomal pellet. Only activities for alanine, asparagine, glycine, and serine were found consistently not be to be associated in significant amounts with ribosomes. Lysates from all cells exhibited relatively high levels of ribosomebound phenylalanyl-t RNA synthetase. Variations in ribosome-bound synthetase activity are exhibited for most of the other amino acids within a cell type and for a single amino acid measured in lysates of the different cell types. Even though the two transformed cell lines examined, HeLa and Friend leukemia cells, have very different lysate activities for arginine, asparagine, cysteine, and proline synthetases, the percentages of these synthetases that are bound to the ribosomes are significantly lower than in lysates from the other cell types tested. The embryonic tissues have a consistently high percentage of most of their aminoacyl-tRNA synthetases associated with the ribosomal pellet when compared to the other cell types tested. In contrast, mouse liver lysates have low percentages of most of the aminoacylation enzymes associated with the ribosomes. Less variation in ribosome-bound synthetase activity is evident when the data from which Table 4 was derived are presented in absolute terms, enzyme units per mg ribosome, as shown in Table5. Activities for cysteine in lysates from transformed cells, and isoleucine and serine in mouse liver are lower than the other lysates tested, while histidine and threonine activities are significantly higher in mouse embryo lysates. Rabbit reticulocyte and mouse embryo ribosomes have much more alanyl-tRNA synthetase activity than the other cell types tested. Rabbit reticulocytes had a low amount of tyrosyl-tRNA synthetase bound to their ribosomes. HeLa cell ribosomes were associated with lower synthetase activities for asparagine, aspartic acid, methionine, serine, and valine than the other cells. This finding is consistent with their lower average lysate activity, as given in Tablel. Even though the percentage of activity on the ribosomes was comparatively high in embryonic lysates (Table4), the amount of synthetase activity per mg of ribosomes for most amino acids was generally the same as in the other lysates. Transformed cells still show a reduced average synthetase activity bound to the ribosomes (0.72 unit/mg) when compared to other cell types tested (1.37 unit/mg). It may be of special physiological significance that all of the aminoacyl-tRNA synthetases consistently found in the 18- 25-S region of sucrose gradients of the postribosomal supernatant were also found in the
496
Eukaryotic Aminoacyl-tRNA Synthetases
Table 4. Aminoacyl-tRNA synthetase activity associated with ribosomes Activity is calculated as the percentage of the total activity in the lysate for the amino acid indicated Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
Friend leukemia cell
HeLa cell
Average
0.0 28 4.0 22 10 8.0 9.0 1.o 3.5 3.0 3.5 12 5.0 30 12 0.0 0.0 1.5 5.5
2.4 28 4.7 23 15 39 34 3.9 16 20 33 42 35 30 15 1.6 24 19 20
0.62 29 2.9 32 15 32 47 5.4 18 40 37 34 42 46 20 1.5 10 13 12
0.67 8.O 1.2 11 4.9 10 9.6 1.3 4.4 9.6 14 16 13 21 4.4 1.4 2.0 8.9 6.1
0.39 11 0.28 10 1.2 6.1 8.0 0.41 4.6 12 7.4 12 7.0 48 6.3 0.82 0.56 6.3 3.7
I .3 21 2.6 18 11 20 25 2.0 8.8 18 19 28 21 41 11 1.1 6.3 8.3 11
21
23
7.8
7.1
%total Alanine Arginine Asparagine Aspartic acid Cysteine Giutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine
4.0 24 2.6 12 20 27 45 0.30 6.0 23 22 53 23 68 10 1.4 1.o 0.80 18
~ _ _
~~~~
Average
19
8.3
14
Table 5. Aminoacyl-tRNA synthetase activity associated with ribosomes Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
0.12 3.3 0.24 3.0 0.34 0.64 0.62 0.068 0.86 0.58 1.9 5.9 0.84 9.3 0.25 0.37 1.1 0.15 0.86
0.01 8 2.7 0.14 2.8 0.18 0.30 0.57 0.044 0.32 0.24 0.72 5.0 0.40 6.0 0.30 0.27 0.15 0.11 0.21
Friend leukemia cell
HeLa cell
Average
units/mg ribosomes -
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phen ylalanine Proline Serine Threonine Tyrosine Valine Total Average
0.22 3.8 0.38 1.o 0.28 0.42 0.49 0.015 0.22 0.53 0.80 4.2 0.62 19.0 0.16 0.51 0.10 0.0052 0.74
0.012 0.92 0.29 0.87 0.12 0.27 0.32 0.017 0.12 0.064 0.29 2.7 0.21 12.0 0.046 0.0058 0.035 0.035 0.17
33.5 1.8
18.49 0.97
30.4 1.6
ribosomal pellets. These are the enzymes for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, proline and valine. Other experiments have demonstrated that these enzymes
20.5 1.1
0.020 1.30 0.13 0.96 0.082 0.32 0.24 0.054 0.16 0.24 0.58 2.40 0.36 5.20 0.039 0.36 0.10 0.064 0.26 12.9 0.68
0.023 1.20 0.040 0.16 0.054 0.17 0.22 0.022 0.14 0.31 0.32 2.20 0.15 8.40 0.074 0.072 0.079 0.025 0.060 14.2 0.75
0.069 2.20 0.20 1S O 0.18 0.35 0.41 0.037 0.30 0.33 0.77 3.70 0.43 10.0 0.14 0.26 0.26 0.065 0.38 21.6 1.1
are not pelleteted in the absence of ribosomes under the conditions used and that the enzymes can rebind efficiently to ribosomes from which they have been removed with 0.5 M KCl (data not shown).
M. A. Ussery, W. K. Tanaka, and B. Hardesty
491
--
I
45 7.45 11.6 S 16s 1-, . 1 1
1I P \
.. :..-+... .. .
.. .. ..
c
5 .- 12 c
m "
z 0.2
: L
c 14
Fraction number Fig. 1. Sucrose density gradient distribution of tRNA synthetasesfor threonine, phenylalaninr and Iysine. A lysate from Friend leukemia cells grown for three days in stationary phase was prepared, fractionated by sucrose gradient centrifugation and assayed for phenyland lysyl-tRNA synthetase alanyl-tRNA synthetase (O-O), (G---o) activity as described in Methods. Mouse embryo lysate was prepared as described in Methods, fractionated on a parallel gradient, and assayed for threonyl-tRNA synthetase activity (&---n). Arrows indicate position of markers used as described in Methods: rabbit liver tRNA (4 S), lactate dehydrogenase (7.4 S), catalase (1 1.6 S) and 8-galactosidase (16 S )
Sedimentation Distribution of Synthetases not Bound to Ribosomes Lysates were centrifuged on 10 - 30% linear sucrose density gradients as described in Methods. Under the conditions used ribosomes were accumulated as a pellet at the bottom of the centrifuge tubes. Lysates rather than the postribosomal superntants were used for this procedure to minimize loss of enzymic activity due to fractionation and handling. Most synthetases not bound to ribosomes sediment either as smallmolecular-weight species in the 4 - 9-S region or in the 18 - 25-S region of the gradients. Fig. 1 shows the distribution for phenylalanyl-tRNA synthetase from Friend leukemia cells and threonyl-tRNA synthetase from mouse embryo, representing a group of synthetases sedimenting in the 4 - 9-S region. Lysyl-tRNA synthetase from Friend leukemia cells represents synthetases in the 18 - 25-S region of the gradient. For the most part, synthetases in the 4 - 9-S region of the gradient appear to be sedimenting as separate entities with slightly different sedimentation values. On the other hand, synthetases in the 18 - 25-S region sediment with very similar sedimentation values. As shown in Fig. 2 the peak of various synthetase activities in the 18 - 25-S region correspond closely. Table6 shows the distribution of aminoacyl-tRNA synthetase activity sedimenting in the 18 - 25-S region of the sucrose gradient for each of the cell types tested. The numbers reported are the percentages of the postribosomal aminoacyl-tRNA synthetase activities that are present in the 18 - 25-S region of the gradient. This region of the gradient consistently contained synthetase activity for arginine, aspartic acid, glut-
OL 0
2
4
6
8
1
0
1
2
Fraction number Fig. 2. Sucrose density gradient distribufion oJ t R N A syntheiase activity .for arginine, lysine, rnethionine and Ieucine. The lysate from Friend cells described in Fig. 1 was assayed for arginyl-tRNA synthetase (.A), lysyl-tRNA synthetase (G---O), methionyland leucyl-tRNA synthetase (A----A). tRNA synthetase (-0) Protein concentration (.....) was determined by the procedure of Warburg and Christian [15]. Arrows indicate position of markers used as described in Methods: rabbit liver tRNA (4 S), lactate dehydrogenase (7.4 S), catalase (1 1.6 S) and b-galactosidase (16 S)
amine, glutamic acid, isoleucine, leucine, lysine, methionine, proline and valine. In addition, the 18 -25-S region of reticulocyte gradients also contained a large percentage of activity for asparagine, cysteine and tyrosine, while the mouse embryo gradients had additional threonyl-tRNA synthetase in the complex region. Friend leukemia cells contained a lower percentage of activity for valine in the 18 - 25-S region than the other lysates tested (HeLa lysates were not tested for sucrose gradient distribution). Of these synthetases sedimenting in the 18 - 25-S region, only arginyl-tRNA synthetase also evidenced a distinct peak in the 4-9-S region of the gradient (Fig. 2). In contrast to the transformed and embryonic cells tested which evidenced an average of 44% of their arginyl-tRNA synthetase activity in the 4 - 9-S region, only 19% of the postribosomal arginyl-tRNA synthetase activity in rabbit reticulocytes and mouse liver was present in this region. The centrifugation procedure used to separate free synthetases from those that exist as complexes was a compromise selected to give the best average separation. However, some of the free synthetases have a relatively high sedimentation coefficient and are not cleanly separated from those that exist as complexes. Phenylalanyl-tRNA synthetase is an example of this type. Even though it appears to exist as a single enzyme [7] 27% of the activity in the postribosomal supernatant was found in the 18 - 25-S region ofthe gradient. This problem of resolution may be seen clearly in Fig. 1. On this basis, caution must be exercised in interpreting the data of Tables6 and 7 in which less than 30% of the activity of the enzyme is in the 18 -25-S region as reflecting an enzyme complex.
498
Eukaryotic Aminoacyl-tRNA Synthetases
Table 6. Aminoacyl-tRNA synthetase activity in I8 - 25-S region of suerose gradients Values are for percentage of total activity not associated with ribosomes, calculated by dividing the total activity in fraction 8 10 of sucrose gradients by the activity in the postribosomal supernatant as described in Methods ~
Amino acid
Rabbit reticulocytes
Mouse liver
Mouse embryo
Chicken embryo
Friend leukemia cell
Average
6.0 65 12 31 30 92 87 0 4.0 96 94 53 74 6.0 84 28 27 0 89
6.0 59 24 88 4.4 59 71 0 3.0 86 86 71 79 27 71 5.8 10 4.8 35
9.4 66 20 77 21 77 76 17 7.8 90 89 80 80 14 72 13 22 13 78
46
42
49
Mouse embryo
Chicken embryo
Friend leukernia cell
Average
0.012 3.5 0.017 8.0 0.18 0.58 0.62 0.009 0.22 1.8 3.5 5.8 0.90 3.1 1.03 0.48 1.1 0.01 1 3.3
0.18 4.8 0.56 2.2 0.35 0.73 0.62 0.007 0.070 0.94 1.2 5.9 0.59 0.48 1.2 5.7 0.62 0.021 1.7
% total 22 85 41 76 49 46 47 31 5.0 70 77 90 68 11 56 20 18 41 96
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine lsoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine ~
~~
~~
56
49
~~
~
Average
100
0 43 0 100 8.6 95 98 0 6.0 99 93 96 91 13 70 2.0 45 0 71
13 78 23 91 15 92 78 46 21 99 95 92 90 15 75 7.3 9.1 20
50
~~
~
~~
Table 7. Arninoacyl-tRNA synthetase activity in 18 - 2 5 3 region of sucrose gradients Amino acid
Rabbit reticulocyte
Mouse liver
units/ml lysate ~
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine Total Average
1.8 14 4.1 7.5 0.81 0.84 0.58 2.1 0.23 1.9 3.3 7.3 2.6 3.3 1.1 11.2 2.1 0.91 5.5 71.2 3.7
0.89 32 1.6 31 0.42 5.1 3.9 1.9 1.8 4.0 12.4 34 6.9 8.4 0.37 2.1 1.2 0.86 5.0 153.8 8.1
34.2 1.8
Table7 is a representation of the amount of synthetase activity present in the 18 - 25-S region of the sucrose density gradients expressed in enzyme units per ml lysate. Thus, using the data in Tables6 and 7,
27.9 1.5
0.54 23 6.5 20 0.20 4.4 5.0 0.010 0.26 5.0 8.5 25 5.3 14 1.7 3.7 1.6 0.078 3.8 128.6 6.8
~~
0.68 15 2.6 14 0.39 2.3 2.1 0.80 0.52 2.7 5.8 16 3.3 5.9 1.08 4.6 1.3 0.37 3.9 83.3 44
it is possible to calculate the activity in the 4-9-S region of these gradients. Both mouse liver and Friend leukemia cells had more activity in the 18 - 2 5 4 region than any other cell type. This is seen in the relatively
M. A. Ussery, W. K. Tanaka, and B. Hardesty
high activities for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, and phenylalanine. DISCUSSION The primary objective of this study was to examine the subcellular distribution of aminoacyl-tRNA synthetases in different cells and tissues under as nearly identical conditions of analysis as could be maintained. This problem arose from apparent inconsistencies in reports from a number of laboratories regarding association of aminoacyl-tRNA synthetases into enzyme complexes and with ribosomes. It is unclear whether the reported differences arose from differences in analytical techniques or reflected architectural and physiological differences in intact cells. A definitive demonstration of differences within intact cells would provide considerable support for the hypothesis that protein synthesis is regulated by the intracellular organization of these enzymes. The results presented here do not conclusively answer this question. HOWever, they do demonstrate differences in lysates derived from different cell types under standard conditions. The exact proportions of the enzymes found in the three fractions examined are somewhat subject to the conditions used. The proportion of aminoacyl-tRNA synthetases associated with ribosomes is sensitive to variations in salt concentration. Most of the synthetase activity is removed from the ribosomes if they are washed through 0.3 M KCl [6]. However, the general form of the distribution patterns described here have been observed under a variety of conditions. We believe the evidence presented reflect quantitative differences in the absolute amount and distribution of aminoacyl-tRNA synthetases in the cell types examined during the course of this work. The total activities of the various lysates were surprisingly similar for all of the cells tested. This result was not expected since the cell types were chosen to represent extreme differences of differentiation, type of protein being synthesized, and length of cell generation time. Friend leukemia cells harvested under both logarithmic and stationary growth conditions both before and after being stimulated to synthesize large amounts of hemoglobin and Friend leukemia virus showed surprisingly little change in their lysate activities (Table 3). However, some striking similarities and differences were observed in the distribution and total activity for certain amino acids. Some of these are pointed out above in connection with the specific tables. The reader may find other intriguing differences without difficulty. The synthetase complex sedimenting at approximately 20s observed in the present study appears to be somewhat larger than either of the two purified complexes previously reported [2,3]. This observation
499
lends support to the proposition that certain components of the synthetase complex are less tightly bound and that in the intact cell the synthetase complex may be larger than results of work with the purified complex would suggest. The results presented here are in substantial agreement with the data on the subcellular distribution of aminoacyl-tRNA synthetases in Chinese hamster ovary cells [S] and with the composition of factor X in rat liver [2]. The major difference is the presence of 18 -25s proline activity in the systems we have examined. The aminoacyl-tRNA synthetases sedimenting in the 9-S region display a wide range of sedimentation values. In general, synthetases specific for alanine, glycine, serine, threonine and tyrosine appear to be slightly smaller than those for asparagine, aspartic acid, cysteine and phenylalanine. The sedimentation profiles for these enzymes in the soluble fraction were similar for all cells tested and suggest that these enzymes are not associated with highermolecular-weight enzyme complexes in the cell lysates. The existence of variable absolute amounts and proportions of arginyl-tRNA synthetase activity in a lowmolecular-weight form, as well as in the complex region and on ribosomes, suggests a possible functional difference among these forms. The presence of activity for methionine in every complex that has been described to date is particularly intriguing. It may reflect a role of the complex or the ribosome-bound complex in peptide initiation. We have been unable to find a satisfactory physiological or biochemical basis for the specific distribution patterns observed. The binding of the free synthetases or synthetase complex to ribosomes could provide a greater local concentration of aminoacyl-tRNA than simple diffusion from a cytoplasmic pool. This might facilitate protein synthesis. There is no information available as to the nature of the factors that control binding of an individual synthetase to an enzyme complex or to ribosomes. However, from the data presented here we conclude that such factors must exist, and suggest that they may play a role in regulating protein synthesis. Michael Ussery is a trainee supported by National Institutes of Health grant CA09182 from the National Cancer Institute. Wesley K. Tanaka was a postdoctoral fellow of the National Cancer Institute, fellowship CA0079. This work was supported in part by Grants from the National Cancer Institute, CA09182 and CA16608, and from the National Science Foundation BMS7520232. The authors would like to thank Mr J. Ybarra for technical assistance and Dr Kamales Som for advice in the early stages of this work, to Dr G. Kramer for critical discussion and to B. Anderson for preparing the typescript.
REFERENCES 1. Bandyopadhyay, A. K. & Deutscher, M . P. (1971) J. Mol. Bid. 60, 113-122. 2. Vennegoor, C. & Bloemendal, H. (1972) Eur. J. Biochem. 26, 462-473.
500
M . A. Ussery, W. K. Tanaka, and B. Hardesty: Eukaryotic Aminoacyl-tRNA Synthetases
3. Som, K. & Hardesty, B. (1975) Arch. Biochem. Biophys. 166, 507 - 512. 4. Roberts, K. &Coleman, W. H. (1972) Biochem. Biophys. Res. Commun. 46, 206-214. 5. Hampel, A. & Enger, M. D. (1973) J. Mol. Biol. 79, 285-293. 6. Irvin, J. D. & Hardesty, B. (1972) Biochemistry, 11,1915 - 1920. 7. Tanaka, W. K., Som, K. & Hardesty, B. A. (1976) Arch. Biochem. Biophys. 172, 252-260. 8. Kramer, G., Pinphanichakarn, P., Konecki, D. & Hardesty, B. (1975) Eur. J. Biochern. 53,471 -480. 9. Hardesty, B., McKeehan, W. & Culp, W. (1971) Methods Enzymol. 20 C, 316 - 330.
10. Craven, G. R., Steers, E., Jr. & Anfinsen, C . B. (1965) J . Biol. Chem. 240,2468 - 2476. 11. Pesce, A., McKay, R. H., Stolzenbach, F., Cahn, R. D. & Kaplan, M. 0. (1964) J. Biol. Chem. 239, 1753-1761. 12. Beers, R., Jr. & Sizer, I. (1952) J . Biol. Chem. 195, 133-140. 13. Gornall, A. G., Bardawill, C. J. & David, M. M. (1949) J . Biol. Chem. 177, 751 - 766. 14. Sato, T., Friend, C. & de Harven, E. (1971) Cancer Res. 31, 1402 - 1417. 15. Warburg, 0. &Christian, W. (1942) Biochem. Z. 310,384-421.
M. A. Ussery and B. A. Hardesty, Clayton Foundation Biochemical Institute, Department of Chemistry, University of Texas at Austin, Austin, Texas, U.S.A. 78712 W. K. Tanaka, Department of Chemistry, University of Wisconsin at Eau Claire, Eau Claire, Wisconsin, U.S.A. 54701
Subcellular Distribution of Aminoacyl-tRNA Synthetases in Various Eukaryotic Cells Michael A. USSERY, Wesley K. TANAKA, and Boyd HARDESTY Clayton Foundation Biochemical Institute, University of Texas at Austin (Received July 20/0ctober 5 , 1976)
The total amount, size distribution and binding of aminoacyl-tRNA synthetases to ribosomes in a variety of mammalian and avian cells was studied under standard conditions of sample preparation and assay. Aminoacyl-tRNA synthetases appear to exist in three general forms; ‘free’ enzyme of about 4 - 9 S, one or more ‘enzyme complexes’ of about 18 - 25 S, and in association with ribosomes. The aminoacyl-tRNA synthetase activity for many individual amino acids was surprisingly similar in cell types chosen to be diverse with respect to differentiation state, transformation, and growth rate. Total activity for all amino acids varied about 4-fold, based on a constant volume of cells. Embryonic tissues had a comparatively high proportion of total synthetase activity associated with ribosomes, whereas this value was relatively low for mouse liver. Distinctive distribution patterns with common and variable features were observed for individual enzymes. The only aminoacyl-tRNA synthetases found not to be associated in significant amounts with either 18 - 25-S enzyme complexes or ribosomes in any of the cell types examined were the enzymes for alanine, histidine, and serine. All cell types evidenced 18 - 25-S synthetase activity for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, proline, and valine, although in quite variable proportions of the total activity observed for these amino acids. For example, of the valyl-tRNA synthetase activity not associated with ribosomes, 35% and 100% were found to sediment at 18 -25 S in Friend leukemia cells and mouse liver respectively. All cells had two easily distinguishable peaks of arginyltRNA synthetase activity at 4 - 9 S and 18 - 25 S respectively; however, the relative proportion of enzyme activity in the peaks differed between cell types. Phenylalanyl-t RNA synthetase was not observed to occur in an 18 - 25-S complex in any of the cell types examined but was bound to ribosomes in variable but generally relatively high proportions. Numerous other specific differences are described. No underlying physiological or biochemical principle has been recognized to account for the specific distribution patterns observed. However, they may reflect variations in cellular architecture that may be related to regulation of protein synthesis.
Eukaryotic aminoacyl-tRNA synthetases have been examined in a number of laboratories. Bandyopadhyay and Deutscher [l] have partially purified a multienzyme complex from rat liver 105000 x g supernatant that contained aminoacyl-tRNA synthetase activities for all 18 amino acids tested. Vennegoor and Bloemendal [2] have partially purified an 18.2-S aminoacyltRNA synthetase complex from rat liver which contained activities for glutamine, isoleucine, leucine, lysine, and methionine. Som and Hardesty [3] have similarly purified a complex from rabbit reticulocytes which sediments at approximately 16 S and contains Enzymes. Aminoacyl-tRNA synthetases (EC 6.1.1); E. coli plactosidase (EC 3.2.1.23); lactate dehydrogenase (EC 1.1.1.27); catalase (EC 1.11.1.6).
activities for arginine, isoleucine, leucine, lysine, and methionine. Several workers have also reported the binding of aminoacyl-tRNA synthetases to ribosomes. Roberts and Coleman [4] found most of the phenylalanyl-tRNA synthetase activity in Ehrlich ascites cells in two forms : bound to the 60-S subunit of the ribosomes and in a complex that sediments at about 25s. Hampel and Enger studied the subcellular distribution of aminoacyl-tRNA synthetases in Chinese hamster ovary cells by differential centrifugation and found several aminoacyl-tRNA synthetase activities cosedimenting with the ribosomes [ 5 ] . Work with rabbit reticulocytes has indicated that significant aminoacyl-t RNA synthetase activities for phenylalanine, lysine, isoleucine, leucine,
492
methionine, and arginine were bound to the ribosomes 13,6,71. Thus, there is considerable evidence for the existence of an aminoacyl-tRNA synthetase complex and ribosome-associated synthetases from a number of eukaryotic organisms. These results have led to the stated conclusion or unstated implication that the observations reflect organization of these enzymes in intact cells and that this organization may play a role in regulating protein synthesis. The reported differences may reflect variations in the differentiation and physiological state of cells or tissues. However, the reporting laboratories have employed a wide range of analytical techniques thus making a direct comparison tenuous. The object of the present study is to examine the aminoacyl-tRNA synthetase distribution in various eukaryotic cells by using a standard procedure for rupture and fractionation of the cells. The results of these experiments suggest an overall qualitative similarity between very diverse cell types (normal adult, embryonic, transformed, and highly differentiated) with respect to the presence of synthetase complexes and with respect to synthetases bound to ribosomes. However, significant quantitative differences were observed for specific enzymes. These differences are considered with respect to their potential significance in the overall scheme of protein synthesis.
MATERIALS AND METHODS
Reagents Uniformly 14C-labeled amino acids and ultrapure grade sucrose were obtained from Schwarz-Mann, Inc. (Orangeburg, NY); dithioerythritol, 2-mercaptoethanol, trichloroacetic acid, 2,5-diphenyloxazole were from Sigma Chemical Company (St Louis, MO); bovine serum albumin was from Armour Pharmaceutical Company (Kanakee, IL).
Cells and Tissue The Friend leukemia cells (Friend leukemia virustransformed Murine proerythroblastic line, FSD l/clone F4) were kindly provided by Dr W. Ostertag (Max-Planck-Institut fur experimentelle Medizin, Gottingen, Germany). HeLa S 3 cells were kindly provided by Dr G. Koch (Roche Institute of Molecular Biology, Nutley, NJ). These cells were grown in suspension culture with Eagle's minimal essential medium (Grand Island Biological Co., Grand Island, NY) supplemented with non-essential amino acids and 10% fetal calf serum and antibiotics, as previously described [8]. Viable cell number was determined by exclusion of erythrosine. In all cases, unless otherwise noted, cells were harvested in logarithmic growth phase.
Eukaryotic Aminoacyl-tRNA Synthetases
CFW mice (E and R Enterprises, Roundrock, TX) were used as a source of mouse embryo and liver tissue. Rabbit reticulocytes were prepared as described previously [9]. Chicken embryos were generously provided by Dr H. R. Bose (Dept. of Microbiology, University of Texas, Austin, TX).
Dimethyl Sulfoxide Stimulation of Friend Leukemia Cells Friend leukemia cells were grown to a cell density of 1.O x lo6 cells/ml. These cells were then diluted 1 :1 with fresh medium. Dimethyl sulfoxide was added to a final concentration of 1.5% (v/v). Cells were allowed to grow in the presence of dimethyl sulfoxide for 4 days maintaining a cell density of 0.5 - 1.O x lo6 cells/ml.
Aminoacyl-tRNA Synthetase Assay Reaction mixtures for all synthetases except phenylalanyl-tRNA synthetase contained 100 mM TrisHC1 (pH 7.5), 3.0 mM MgCl,, 20 mM 2-mercaptoethanol, 2.0mM ATP, 0.1 mM I4C-labeled amino acid (10 Ci/mol), 200 pg rabbit liver tRNA, and 20 pg bovine serum albumin in a total volume of 0.2 ml. This standard aminoacylation mixture was developed to be as close as possible to the measured pH and Mg2+: ATP ratio optima for the 18 synthetases (excluding phenylalanyl-tRNA synthetase [6]) studied. Several preparations of [14C]tryptophan were tested but all gave consistently high blanks which precluded reliable measurement of tryptophanyl-tRNA synthetase activity. The reaction mixture for phenylalanyl-tRNA synthetase was similar except that it contained 100 mM Tris-HC1 (pH 8.5), 10 mM MgCI,, 100 mM KC1 and 4.0 mM ATP and the other listed reagents. For assay purposes a suitable aliquot, usually 10 p1 of enzyme fraction was added to the reaction mixture, mixed thoroughly, then incubated at 37 "C for 5 min. The reaction was stopped by addition of 5 ml of cold 5 % (w/v) trichloroacetic acid. The resulting precipitate was collected by filtration through millipore filters (0.45 pm, HAWG, Millipore Corp., Bedford, MA) and washed with a total of 15 ml of 5% (w/v) trichloroacetic acid. The filters were then dried in an oven for 5 min at 120 "C. Dried filters were put into plastic vials containing 10 ml of scintillation solution which contained 5 g of 2,5-diphenyloxazole in 1 1 of toluene. Vials were counted in a Beckman LS-150 scintillation counter with an efficiency of 85 %. Synthesis of aminoacyl-t RNA is proportional to added enzyme in the range 50 - 200 pmol product formed. A unit of enzyme activity is defined as the amount of enzyme required to form 1 nmol of aminoacyl-tRNA per min.
M. A. Ussery, W. K. Tanaka, and B. Hardesty
Preparation and Centrifugation of Lysates
Embryo tissues (15 mouse embryos, four chick embryos) or mouse liver (10 g wet weight) were minced thoroughly before further preparation. Rabbit reticulocytes, Friend leukemia cells and HeLa cells were collected by centrifugation at 4000 rev./min in a GS3 Sorvall rotor for 15 min. Cells and minced tissues were washed twice with 100 ml of a solution containing 130 mM NaC1, 5 mM KCl, 7.4 mM MgCl, and lysed by douncing 20 times with a glass douncer after addition of 1 vol. of H,O to 1 vol. of packed cells. The nuclei, mitochondria, and membranes were removed by centrifugation at 10 800 x g for 15 min. The supernatant from this centrifugation is referred to hereafter as the ‘lysate’. To determine the distribution of synthetases not bound to the ribosomes, a 1.0-ml aliquot of the resulting lysate was loaded on a 11.2-ml 10 - 30% linear sucrose density gradient containing 16 mM KH,PO,, 84 mM Na,HPO, (pH 7.5), 5 mM 2-mercaptoethanol, 1 mM dithioerythritol, and centrifuged in an SW41 rotor at 35000 rev./min for 16 h. Fractions containing 1 ml were collected with an Isco model D density gradient fractionator (Instrumentation Speialities Company, Lincoln, NB). Preparation of Ribosomal Fraction and Postribosomal Supernatant
A 2.0-ml aliquot of the lysate from the l0800xg centrifugation was loaded on a 9.5-ml cushion of 30% sucrose containing 25 mM KCl, 20 mM Tris-HC1 (pH 7.5), 5 mM 2-mercaptoethanol, 5 mM MgC1, and centrifuged in a 50Ti rotor at 50000 rev./min for 2 h. These centrifugation conditions were found to prevent the contamination of the ribosomal pellet by the larger (18 - 25-S complex) form(s) of free synthetases. A KC1 concentration of 25 mM prevented nonspecific binding of synthetases to ribosomes which was found to occur at 0 mM KCl but allowed specific binding of purified synthetase complex and phenylalanyl-tRNA synthetase to ribosomes which were removed only at KC1 concentration greater than 100 mM. The supernatant (postribosomal supernatant) was collected and the ribosomal pellet was washed with a buffer containing 5 mM 2-mercaptoethanol, 10 mM MgCl, ,20 mM Tris-HC1 (pH 7.5) and 15% glycerol and resuspended in 0.5 ml of the same buffer. Debris were removed from the resuspended ribosomes by centrifugation at 10800 x g for 15 min. The percentage of activity bound to the ribosomes was determined by dividing the activity associated with the ribosomes per ml lysate by the ribosomal activity per ml lysate plus the postribosomal activity per ml lysate and multiplying the result by 100. The concentration of ribosomes per ml of lysate differed by more than 4-fold between the extremes for the cell types tested (cf. Friend leukemia
493
cells and HeLa cells). The basis for this surprising difference is not clear but has been observed in a number of experiments. Assay of Sucrose Gradients
Samples containing 0.1 mg E. coli P-galactosidase (PL Biochemicals, Inc., Milwaukee, WI), 0.123 mg beef heart lactate dehydrogenase (Worthington Biochemical Corp., Freehold, NJ), 0.07 mg beef liver catalase (Worthington) and 0.9 mg of rabbit liver tRNA [9] were used as standards for determining sedimentation values for various portions of the sucrose gradient. P-Galactosidase was assayed by hydrolysis of o-nitrophenyl-8-galactopyranoside(Sigma Chemical Co., .St Louis, MO) [lo]; lactate dehydrogenase was assayed by oxidation of NADH (Calbiochem, La Jolla, CA) [ll], and catalase was assayed according to the method of Beers and Sizer [12]. The sedimentation of rabbit liver tRNA was determined by absorbance at 260 nm. All gradients were routinely assayed for arginyltRNA synthetase activity and, in some cases, for the activity of the various standards. Fig. 2 shows a typical distribution pattern for the two forms of arginyl-tRNA synthetase in sucrose density gradients along with the sedimentation position of the various standards. Fractions 3-6 represented the 4-9-S region of the gradient and was routinely located by the position of small ‘free’ form of arginyl-tRNA synthetase which sediments at approximately 4 S. The 18 - 25-S region of the gradient was contained in fraction 8 - 10. This region was routinely located by the position of the large ‘complex’ form of arginyl-tRNA synthetase. The percentage of the postribosomal activity sedimenting in the 18 - 25-S region was determined by dividing the total activity in fraction 8-10 per ml lysate by the activity in the postribosomal supernatant per ml lysate and multiplying the result by 100. RESULTS Lysate A ct ivities
Aminoacyl-tRNA synthetase activities in lysates of the various cells and tissues exhibit an overall similarity with some notable exceptions, as seen in Table 1. Total aminoacyl-tRNA synthetase activities of Friend leukemia cell lysates were over twice as high as the total average values for all of the cell types tested. The value for chicken embryo was one half of the total average value. Activities for some enzymes appeared to vary independently. The activity for asparagine in Friend leukemia cells was over six times larger than the average for the other cell types but about equal or below this average with respect to proline and tyrosine. Rabbit reticulocytes had relatively high levels of serine but were low in activity for
494
Eukaryotic Aminoacyl-tRNA Synthetases
Table 1. Aminoacyl-tRNA synthetase activities of lysates Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
5.0 34 7.4 35 3.5 6.8 5.1 8.6 8.8 4.1 16 39 7.8 72 0.86 26 12 4.0 5.2
2.4 26 6.0 14 3.6 2.8 3.4 4.5 6.1 1.9 5.1 20 3.2 39 5.4 24 2.6 0.80 4.0
2.2 14 4.4 11 1.2 2.0 3.0 4.3 1.6 1.8 2.6 25 2.0 25 1.4 19 2.8 0.70 1.6
Friend leukemia cell
HeLa cell
Average
units/ml lysate
7.4 20 8.7 8.0 2.9 2.7 1.3 6.3 4.8 3.0 3.4 13 4.8 64 1.8 60 13 0.90 3.5
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine ~~
Total Average
14 60 45 28 3.9 9.6 12 21 13 7.7 12 62 12 129 1.5 76 15 1.5 14
5.3 15 10 0.95 1.6 8.6 2.7 5.6 2.6 3.8 3.0 18 1.9 21 2.4 11 16 0.28 1.9
6.1 28 14 16 2.8 5.4 4.6 8.4 6.2 3.7 7.0 30 5.3 58 2.2 36 10 1.4 5.0
537.12 28
131.6 6.9
249.5 13
~
229.5 12
301.2 16
125.6 6.6
174.8 9.2
Table 2. Comparison of aminoacyl-tRNA synthetase activities of lysasates based on protein concentration and packed cell volume Protein concentration was determined by the Biuret procedure essentially as described by A. G. Gornall et a. 1131. All relative values are compared to the HeLa lysate. Total lysate activity is the total units of activity of the 19 aminoacyl-tRNA synthetase/ml lysate (from Table 1). Total specific activity is the total lysate activity divided by the protein concentration Cells
Protein concentration
Total lysate activity
Total specific activity
HeLa Chicken embryo Friend leukemia Mouse embryo Mouse liver Reticulocyte
mg/ml 22 36 39 45 110 170
U/ml (relative) 132 (1.0) 126 (1.0) 537 (4.1) 175 (1.3) 301 (2.3) 230 (1.7)
U/mg (relative) 6.0 (1.0) 3.5 (0.58) 13.8 (2.3) 3.9 (0.65) 2.7 (0.45) 1.4 (0.23)
(relative) (1.0) (1.6) (1.8) (2.0) (5.0) (7.7)
glutamine, glutamic acid, and lysine. Mouse liver was relatively rich in activity for aspartic acid, leucine, methionine, and tyrosine, but low in activity for proline. In contrast, mouse embryo tissue was unusually rich in proline activity but low in alanine, isoleucine, and threonine synthetase activities as was chicken embryo. Chicken embryo synthetases for histidine, leucine, and valine also evidenced lower than average activities. HeLa cell lysates generally were less active than average, with synthetases for aspartic acid histidine, leucine, methionine, serine, tyrosine and valine being exceptionally low. The basis for the variation between cell types is not known. In so far as was technically possible, standard conditions were maintained in preparing materials from the various sources. The average standard deviation for repetitive fractionation and assays
carried out on reticulocyte lysates for seven separate preparations was 11 %. The relative distribution of activities within successive preparations from a single cell type and between cell types has been reproducible under a rather wide range of experimental conditions used during evolution of the procedures described in this report. The synthetase activities are reported here in terms of units of activity per ml lysate. This value is directly proportional to the packed cell or tissue volume. The data were also computed as units of enzyme activity per mg protein in the lysate. The parameters required for this calculation and a comparison of the total units of activity for the 19 amino acids tested expressed per ml of lysate and per mg of protein in the lysate is given in Table2. The total activities of the lysates vary about 4-fould expressed on the basis of lysate volume but 10-fold when based on
M. A. Ussery, W. K. Tanaka, and B. Hardesty
495
Table 3. Aminoacyl-tRNA synthetase activities of Friend leukemia cell lysates Cell lysate
Synthetase activity for arginine leucine
U/ml ~ _ FSD-l/cione F 4 logarithmic phase stationary phase
_
lysine
methio- phenylnine alanine
_
81
17
80
17
110
8.5
19
119
18
96
5
58
10
61
10
75
14
107
Dimethylsulfoxidestimulated Expt 1 (80% viability) 15 Expt 2 (95% viability) 36
protein concentration. This is due to the relatively high protein concentration in mouse liver and reticulocyte lysates. Generally the specific activity (units per mg protein) decreases with increasing protein concentration in the lysate. An exception is the highly active Friend leukemia cell lysate. Variability in synthetase activity does not seem to be directly related to cell division time since Friend leukemia cells show similar synthetase activities in logarithmic growth phase or after 3 days in stationary phase (Table 3). Differences in the activities for log phase cells given here with those shown in Table 1 reflect variation in results from separate experiments. It is not clear whether this variation is due to differences in the cells or is inherent in the procedure used. Determination of the packed cell volume in preparations of these lysates is a recognized source of error. Friend leukemia cells can be stimulated to differentiate along the erythroid pathway and produce hemoglobin and Friend leukemia virus by incubation with 1.5% dimethyl sulfoxide [14]. The effect of this stimulation on synthetase activity also is shown in Table3. Cells stimulated to produce hemoglobin by dimethyl sulfoxide have somewhat lower levels of all synthetase activities examined as compared to the same cells which have not been exposed to dimethyl sulfoxide. Arginyl-tRNA synthetase activity is particularly sensitive to this treatment. Whether the lower level of activity of these enzymes is due to changes in the cellular physiology and architecture or to side effects of dimethyl sulfoxide on the aminoacyl-tRNA synthetases themselves is not known. Ribosome-Associated Synthetases The aminoacyl-tRNA synthetase activities associated with the ribosomes were determined with ribosomes that were pelleted from the cell lysate by
high-speed centrifugation through a sucrose solution containing 25 mM KCI as described in Methods. The ribosomes were resuspended and the preparations assayed directly for aminoacyl-tRNA synthetase activity as described in Methods. The resulting data are presented in Table4 as the percentage of the total lysate activity associated with the ribosomal pellet. Only activities for alanine, asparagine, glycine, and serine were found consistently not be to be associated in significant amounts with ribosomes. Lysates from all cells exhibited relatively high levels of ribosomebound phenylalanyl-t RNA synthetase. Variations in ribosome-bound synthetase activity are exhibited for most of the other amino acids within a cell type and for a single amino acid measured in lysates of the different cell types. Even though the two transformed cell lines examined, HeLa and Friend leukemia cells, have very different lysate activities for arginine, asparagine, cysteine, and proline synthetases, the percentages of these synthetases that are bound to the ribosomes are significantly lower than in lysates from the other cell types tested. The embryonic tissues have a consistently high percentage of most of their aminoacyl-tRNA synthetases associated with the ribosomal pellet when compared to the other cell types tested. In contrast, mouse liver lysates have low percentages of most of the aminoacylation enzymes associated with the ribosomes. Less variation in ribosome-bound synthetase activity is evident when the data from which Table 4 was derived are presented in absolute terms, enzyme units per mg ribosome, as shown in Table5. Activities for cysteine in lysates from transformed cells, and isoleucine and serine in mouse liver are lower than the other lysates tested, while histidine and threonine activities are significantly higher in mouse embryo lysates. Rabbit reticulocyte and mouse embryo ribosomes have much more alanyl-tRNA synthetase activity than the other cell types tested. Rabbit reticulocytes had a low amount of tyrosyl-tRNA synthetase bound to their ribosomes. HeLa cell ribosomes were associated with lower synthetase activities for asparagine, aspartic acid, methionine, serine, and valine than the other cells. This finding is consistent with their lower average lysate activity, as given in Tablel. Even though the percentage of activity on the ribosomes was comparatively high in embryonic lysates (Table4), the amount of synthetase activity per mg of ribosomes for most amino acids was generally the same as in the other lysates. Transformed cells still show a reduced average synthetase activity bound to the ribosomes (0.72 unit/mg) when compared to other cell types tested (1.37 unit/mg). It may be of special physiological significance that all of the aminoacyl-tRNA synthetases consistently found in the 18- 25-S region of sucrose gradients of the postribosomal supernatant were also found in the
496
Eukaryotic Aminoacyl-tRNA Synthetases
Table 4. Aminoacyl-tRNA synthetase activity associated with ribosomes Activity is calculated as the percentage of the total activity in the lysate for the amino acid indicated Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
Friend leukemia cell
HeLa cell
Average
0.0 28 4.0 22 10 8.0 9.0 1.o 3.5 3.0 3.5 12 5.0 30 12 0.0 0.0 1.5 5.5
2.4 28 4.7 23 15 39 34 3.9 16 20 33 42 35 30 15 1.6 24 19 20
0.62 29 2.9 32 15 32 47 5.4 18 40 37 34 42 46 20 1.5 10 13 12
0.67 8.O 1.2 11 4.9 10 9.6 1.3 4.4 9.6 14 16 13 21 4.4 1.4 2.0 8.9 6.1
0.39 11 0.28 10 1.2 6.1 8.0 0.41 4.6 12 7.4 12 7.0 48 6.3 0.82 0.56 6.3 3.7
I .3 21 2.6 18 11 20 25 2.0 8.8 18 19 28 21 41 11 1.1 6.3 8.3 11
21
23
7.8
7.1
%total Alanine Arginine Asparagine Aspartic acid Cysteine Giutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine
4.0 24 2.6 12 20 27 45 0.30 6.0 23 22 53 23 68 10 1.4 1.o 0.80 18
~ _ _
~~~~
Average
19
8.3
14
Table 5. Aminoacyl-tRNA synthetase activity associated with ribosomes Amino acid
Rabbit reticulocyte
Mouse liver
Mouse embryo
Chicken embryo
0.12 3.3 0.24 3.0 0.34 0.64 0.62 0.068 0.86 0.58 1.9 5.9 0.84 9.3 0.25 0.37 1.1 0.15 0.86
0.01 8 2.7 0.14 2.8 0.18 0.30 0.57 0.044 0.32 0.24 0.72 5.0 0.40 6.0 0.30 0.27 0.15 0.11 0.21
Friend leukemia cell
HeLa cell
Average
units/mg ribosomes -
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phen ylalanine Proline Serine Threonine Tyrosine Valine Total Average
0.22 3.8 0.38 1.o 0.28 0.42 0.49 0.015 0.22 0.53 0.80 4.2 0.62 19.0 0.16 0.51 0.10 0.0052 0.74
0.012 0.92 0.29 0.87 0.12 0.27 0.32 0.017 0.12 0.064 0.29 2.7 0.21 12.0 0.046 0.0058 0.035 0.035 0.17
33.5 1.8
18.49 0.97
30.4 1.6
ribosomal pellets. These are the enzymes for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, proline and valine. Other experiments have demonstrated that these enzymes
20.5 1.1
0.020 1.30 0.13 0.96 0.082 0.32 0.24 0.054 0.16 0.24 0.58 2.40 0.36 5.20 0.039 0.36 0.10 0.064 0.26 12.9 0.68
0.023 1.20 0.040 0.16 0.054 0.17 0.22 0.022 0.14 0.31 0.32 2.20 0.15 8.40 0.074 0.072 0.079 0.025 0.060 14.2 0.75
0.069 2.20 0.20 1S O 0.18 0.35 0.41 0.037 0.30 0.33 0.77 3.70 0.43 10.0 0.14 0.26 0.26 0.065 0.38 21.6 1.1
are not pelleteted in the absence of ribosomes under the conditions used and that the enzymes can rebind efficiently to ribosomes from which they have been removed with 0.5 M KCl (data not shown).
M. A. Ussery, W. K. Tanaka, and B. Hardesty
491
--
I
45 7.45 11.6 S 16s 1-, . 1 1
1I P \
.. :..-+... .. .
.. .. ..
c
5 .- 12 c
m "
z 0.2
: L
c 14
Fraction number Fig. 1. Sucrose density gradient distribution of tRNA synthetasesfor threonine, phenylalaninr and Iysine. A lysate from Friend leukemia cells grown for three days in stationary phase was prepared, fractionated by sucrose gradient centrifugation and assayed for phenyland lysyl-tRNA synthetase alanyl-tRNA synthetase (O-O), (G---o) activity as described in Methods. Mouse embryo lysate was prepared as described in Methods, fractionated on a parallel gradient, and assayed for threonyl-tRNA synthetase activity (&---n). Arrows indicate position of markers used as described in Methods: rabbit liver tRNA (4 S), lactate dehydrogenase (7.4 S), catalase (1 1.6 S) and 8-galactosidase (16 S )
Sedimentation Distribution of Synthetases not Bound to Ribosomes Lysates were centrifuged on 10 - 30% linear sucrose density gradients as described in Methods. Under the conditions used ribosomes were accumulated as a pellet at the bottom of the centrifuge tubes. Lysates rather than the postribosomal superntants were used for this procedure to minimize loss of enzymic activity due to fractionation and handling. Most synthetases not bound to ribosomes sediment either as smallmolecular-weight species in the 4 - 9-S region or in the 18 - 25-S region of the gradients. Fig. 1 shows the distribution for phenylalanyl-tRNA synthetase from Friend leukemia cells and threonyl-tRNA synthetase from mouse embryo, representing a group of synthetases sedimenting in the 4 - 9-S region. Lysyl-tRNA synthetase from Friend leukemia cells represents synthetases in the 18 - 25-S region of the gradient. For the most part, synthetases in the 4 - 9-S region of the gradient appear to be sedimenting as separate entities with slightly different sedimentation values. On the other hand, synthetases in the 18 - 25-S region sediment with very similar sedimentation values. As shown in Fig. 2 the peak of various synthetase activities in the 18 - 25-S region correspond closely. Table6 shows the distribution of aminoacyl-tRNA synthetase activity sedimenting in the 18 - 25-S region of the sucrose gradient for each of the cell types tested. The numbers reported are the percentages of the postribosomal aminoacyl-tRNA synthetase activities that are present in the 18 - 25-S region of the gradient. This region of the gradient consistently contained synthetase activity for arginine, aspartic acid, glut-
OL 0
2
4
6
8
1
0
1
2
Fraction number Fig. 2. Sucrose density gradient distribufion oJ t R N A syntheiase activity .for arginine, lysine, rnethionine and Ieucine. The lysate from Friend cells described in Fig. 1 was assayed for arginyl-tRNA synthetase (.A), lysyl-tRNA synthetase (G---O), methionyland leucyl-tRNA synthetase (A----A). tRNA synthetase (-0) Protein concentration (.....) was determined by the procedure of Warburg and Christian [15]. Arrows indicate position of markers used as described in Methods: rabbit liver tRNA (4 S), lactate dehydrogenase (7.4 S), catalase (1 1.6 S) and b-galactosidase (16 S)
amine, glutamic acid, isoleucine, leucine, lysine, methionine, proline and valine. In addition, the 18 -25-S region of reticulocyte gradients also contained a large percentage of activity for asparagine, cysteine and tyrosine, while the mouse embryo gradients had additional threonyl-tRNA synthetase in the complex region. Friend leukemia cells contained a lower percentage of activity for valine in the 18 - 25-S region than the other lysates tested (HeLa lysates were not tested for sucrose gradient distribution). Of these synthetases sedimenting in the 18 - 25-S region, only arginyl-tRNA synthetase also evidenced a distinct peak in the 4-9-S region of the gradient (Fig. 2). In contrast to the transformed and embryonic cells tested which evidenced an average of 44% of their arginyl-tRNA synthetase activity in the 4 - 9-S region, only 19% of the postribosomal arginyl-tRNA synthetase activity in rabbit reticulocytes and mouse liver was present in this region. The centrifugation procedure used to separate free synthetases from those that exist as complexes was a compromise selected to give the best average separation. However, some of the free synthetases have a relatively high sedimentation coefficient and are not cleanly separated from those that exist as complexes. Phenylalanyl-tRNA synthetase is an example of this type. Even though it appears to exist as a single enzyme [7] 27% of the activity in the postribosomal supernatant was found in the 18 - 25-S region ofthe gradient. This problem of resolution may be seen clearly in Fig. 1. On this basis, caution must be exercised in interpreting the data of Tables6 and 7 in which less than 30% of the activity of the enzyme is in the 18 -25-S region as reflecting an enzyme complex.
498
Eukaryotic Aminoacyl-tRNA Synthetases
Table 6. Aminoacyl-tRNA synthetase activity in I8 - 25-S region of suerose gradients Values are for percentage of total activity not associated with ribosomes, calculated by dividing the total activity in fraction 8 10 of sucrose gradients by the activity in the postribosomal supernatant as described in Methods ~
Amino acid
Rabbit reticulocytes
Mouse liver
Mouse embryo
Chicken embryo
Friend leukemia cell
Average
6.0 65 12 31 30 92 87 0 4.0 96 94 53 74 6.0 84 28 27 0 89
6.0 59 24 88 4.4 59 71 0 3.0 86 86 71 79 27 71 5.8 10 4.8 35
9.4 66 20 77 21 77 76 17 7.8 90 89 80 80 14 72 13 22 13 78
46
42
49
Mouse embryo
Chicken embryo
Friend leukernia cell
Average
0.012 3.5 0.017 8.0 0.18 0.58 0.62 0.009 0.22 1.8 3.5 5.8 0.90 3.1 1.03 0.48 1.1 0.01 1 3.3
0.18 4.8 0.56 2.2 0.35 0.73 0.62 0.007 0.070 0.94 1.2 5.9 0.59 0.48 1.2 5.7 0.62 0.021 1.7
% total 22 85 41 76 49 46 47 31 5.0 70 77 90 68 11 56 20 18 41 96
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine lsoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine ~
~~
~~
56
49
~~
~
Average
100
0 43 0 100 8.6 95 98 0 6.0 99 93 96 91 13 70 2.0 45 0 71
13 78 23 91 15 92 78 46 21 99 95 92 90 15 75 7.3 9.1 20
50
~~
~
~~
Table 7. Arninoacyl-tRNA synthetase activity in 18 - 2 5 3 region of sucrose gradients Amino acid
Rabbit reticulocyte
Mouse liver
units/ml lysate ~
Alanine Arginine Asparagine Aspartic acid Cysteine Glutamine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tyrosine Valine Total Average
1.8 14 4.1 7.5 0.81 0.84 0.58 2.1 0.23 1.9 3.3 7.3 2.6 3.3 1.1 11.2 2.1 0.91 5.5 71.2 3.7
0.89 32 1.6 31 0.42 5.1 3.9 1.9 1.8 4.0 12.4 34 6.9 8.4 0.37 2.1 1.2 0.86 5.0 153.8 8.1
34.2 1.8
Table7 is a representation of the amount of synthetase activity present in the 18 - 25-S region of the sucrose density gradients expressed in enzyme units per ml lysate. Thus, using the data in Tables6 and 7,
27.9 1.5
0.54 23 6.5 20 0.20 4.4 5.0 0.010 0.26 5.0 8.5 25 5.3 14 1.7 3.7 1.6 0.078 3.8 128.6 6.8
~~
0.68 15 2.6 14 0.39 2.3 2.1 0.80 0.52 2.7 5.8 16 3.3 5.9 1.08 4.6 1.3 0.37 3.9 83.3 44
it is possible to calculate the activity in the 4-9-S region of these gradients. Both mouse liver and Friend leukemia cells had more activity in the 18 - 2 5 4 region than any other cell type. This is seen in the relatively
M. A. Ussery, W. K. Tanaka, and B. Hardesty
high activities for arginine, aspartic acid, glutamine, glutamic acid, isoleucine, leucine, lysine, methionine, and phenylalanine. DISCUSSION The primary objective of this study was to examine the subcellular distribution of aminoacyl-tRNA synthetases in different cells and tissues under as nearly identical conditions of analysis as could be maintained. This problem arose from apparent inconsistencies in reports from a number of laboratories regarding association of aminoacyl-tRNA synthetases into enzyme complexes and with ribosomes. It is unclear whether the reported differences arose from differences in analytical techniques or reflected architectural and physiological differences in intact cells. A definitive demonstration of differences within intact cells would provide considerable support for the hypothesis that protein synthesis is regulated by the intracellular organization of these enzymes. The results presented here do not conclusively answer this question. HOWever, they do demonstrate differences in lysates derived from different cell types under standard conditions. The exact proportions of the enzymes found in the three fractions examined are somewhat subject to the conditions used. The proportion of aminoacyl-tRNA synthetases associated with ribosomes is sensitive to variations in salt concentration. Most of the synthetase activity is removed from the ribosomes if they are washed through 0.3 M KCl [6]. However, the general form of the distribution patterns described here have been observed under a variety of conditions. We believe the evidence presented reflect quantitative differences in the absolute amount and distribution of aminoacyl-tRNA synthetases in the cell types examined during the course of this work. The total activities of the various lysates were surprisingly similar for all of the cells tested. This result was not expected since the cell types were chosen to represent extreme differences of differentiation, type of protein being synthesized, and length of cell generation time. Friend leukemia cells harvested under both logarithmic and stationary growth conditions both before and after being stimulated to synthesize large amounts of hemoglobin and Friend leukemia virus showed surprisingly little change in their lysate activities (Table 3). However, some striking similarities and differences were observed in the distribution and total activity for certain amino acids. Some of these are pointed out above in connection with the specific tables. The reader may find other intriguing differences without difficulty. The synthetase complex sedimenting at approximately 20s observed in the present study appears to be somewhat larger than either of the two purified complexes previously reported [2,3]. This observation
499
lends support to the proposition that certain components of the synthetase complex are less tightly bound and that in the intact cell the synthetase complex may be larger than results of work with the purified complex would suggest. The results presented here are in substantial agreement with the data on the subcellular distribution of aminoacyl-tRNA synthetases in Chinese hamster ovary cells [S] and with the composition of factor X in rat liver [2]. The major difference is the presence of 18 -25s proline activity in the systems we have examined. The aminoacyl-tRNA synthetases sedimenting in the 9-S region display a wide range of sedimentation values. In general, synthetases specific for alanine, glycine, serine, threonine and tyrosine appear to be slightly smaller than those for asparagine, aspartic acid, cysteine and phenylalanine. The sedimentation profiles for these enzymes in the soluble fraction were similar for all cells tested and suggest that these enzymes are not associated with highermolecular-weight enzyme complexes in the cell lysates. The existence of variable absolute amounts and proportions of arginyl-tRNA synthetase activity in a lowmolecular-weight form, as well as in the complex region and on ribosomes, suggests a possible functional difference among these forms. The presence of activity for methionine in every complex that has been described to date is particularly intriguing. It may reflect a role of the complex or the ribosome-bound complex in peptide initiation. We have been unable to find a satisfactory physiological or biochemical basis for the specific distribution patterns observed. The binding of the free synthetases or synthetase complex to ribosomes could provide a greater local concentration of aminoacyl-tRNA than simple diffusion from a cytoplasmic pool. This might facilitate protein synthesis. There is no information available as to the nature of the factors that control binding of an individual synthetase to an enzyme complex or to ribosomes. However, from the data presented here we conclude that such factors must exist, and suggest that they may play a role in regulating protein synthesis. Michael Ussery is a trainee supported by National Institutes of Health grant CA09182 from the National Cancer Institute. Wesley K. Tanaka was a postdoctoral fellow of the National Cancer Institute, fellowship CA0079. This work was supported in part by Grants from the National Cancer Institute, CA09182 and CA16608, and from the National Science Foundation BMS7520232. The authors would like to thank Mr J. Ybarra for technical assistance and Dr Kamales Som for advice in the early stages of this work, to Dr G. Kramer for critical discussion and to B. Anderson for preparing the typescript.
REFERENCES 1. Bandyopadhyay, A. K. & Deutscher, M . P. (1971) J. Mol. Bid. 60, 113-122. 2. Vennegoor, C. & Bloemendal, H. (1972) Eur. J. Biochem. 26, 462-473.
500
M . A. Ussery, W. K. Tanaka, and B. Hardesty: Eukaryotic Aminoacyl-tRNA Synthetases
3. Som, K. & Hardesty, B. (1975) Arch. Biochem. Biophys. 166, 507 - 512. 4. Roberts, K. &Coleman, W. H. (1972) Biochem. Biophys. Res. Commun. 46, 206-214. 5. Hampel, A. & Enger, M. D. (1973) J. Mol. Biol. 79, 285-293. 6. Irvin, J. D. & Hardesty, B. (1972) Biochemistry, 11,1915 - 1920. 7. Tanaka, W. K., Som, K. & Hardesty, B. A. (1976) Arch. Biochem. Biophys. 172, 252-260. 8. Kramer, G., Pinphanichakarn, P., Konecki, D. & Hardesty, B. (1975) Eur. J. Biochern. 53,471 -480. 9. Hardesty, B., McKeehan, W. & Culp, W. (1971) Methods Enzymol. 20 C, 316 - 330.
10. Craven, G. R., Steers, E., Jr. & Anfinsen, C . B. (1965) J . Biol. Chem. 240,2468 - 2476. 11. Pesce, A., McKay, R. H., Stolzenbach, F., Cahn, R. D. & Kaplan, M. 0. (1964) J. Biol. Chem. 239, 1753-1761. 12. Beers, R., Jr. & Sizer, I. (1952) J . Biol. Chem. 195, 133-140. 13. Gornall, A. G., Bardawill, C. J. & David, M. M. (1949) J . Biol. Chem. 177, 751 - 766. 14. Sato, T., Friend, C. & de Harven, E. (1971) Cancer Res. 31, 1402 - 1417. 15. Warburg, 0. &Christian, W. (1942) Biochem. Z. 310,384-421.
M. A. Ussery and B. A. Hardesty, Clayton Foundation Biochemical Institute, Department of Chemistry, University of Texas at Austin, Austin, Texas, U.S.A. 78712 W. K. Tanaka, Department of Chemistry, University of Wisconsin at Eau Claire, Eau Claire, Wisconsin, U.S.A. 54701
