Arginyl Transfer Ribonucleic Acid Protein Transferase and Endogenous Acceptor Proteins in Cultured Mammalian Cells

The presence of arginyl-tRNA protein transferase has been shown in a variety of mammalian cells in culture. Analysis by sodium dodecyl sulfate acrylamide gel electrophoresis of the available endogenous acceptor proteins (proteins that can be arginylated at their amino terminus by this enzyme) reveals that there are at least two such proteins (designated Peaks I and II) with the same relative migration during electrophoresis in a variety of tissues. Analysis of baby hamster kidney cells (BHK), polyoma-transformed BHK, and herpes simplex-infected BHK showed Peak II, but in the general region of Peak I there were two peaks. Other peaks, which varied in migration and appearance, were also observed in the various cell preparations. The incorporation of labeled arginine was inhibited by ribonuclease, canavanine, hemoglobin, and hemin but not by puromycin nor cycloheximide. The specificity of the enzyme from different species for exogenous acceptor proteins of various species is reported.


SUMMARY
The presence of arginyl-tRNA protein transferase has been shown in a variety of mammalian cells in culture. Analysis by sodium dodecyl sulfate acrylamide gel electrophoresis of the available endogenous acceptor proteins (proteins that can be arginylated at their amino terminus by this enzyme) reveals that there are at least two such proteins (designated Peaks I and II) with the same relative migration during electrophoresis in a variety of tissues. Analysis of baby hamster kidney cells (BHK), polyoma-transformed BHK, and herpes simplex-infected BHK showed Peak II, but in the general region of Peak I there were two peaks. Other peaks, which varied in migration and appearance, were also observed in the various cell preparations. The incorporation of labeled arginine was inhibited by ribonuclease, canavanine, hemoglobin, and hemin but not by puromycin nor cycloheximide.
The specificity of the enzyme from different species for exogenous acceptor proteins of various species is reported.
:2rgillyl~tRE~\ l)rotein transferase which catalyzes the transfer of arginille from arginyl-tRNX to the amino terminal of specific proteins, has been found in rat (l-3) and rabbit liver (4, 5), rat hepatoma (6), :uld sheep thyroid (7,8). Endogenous acceptor proteins occur iii all the-e tissues, ;ud in addition albumin, thyroglobill (9)) Ilence-Joneti proteins, and soybean trypsin inhibitor (10) serve as acceptor protein for arginine when a highly purified preparation 01' the transferase front rabbit liT:er is used. Tile endogenous acceptor protein (or llroteins) have llot been characterized except in a recent report by Soffer (10) wherein he offers evidence that a major endogenous acceptor protein in rabbit liver is an albumill-like protein.
This finding concurs with the facts that (a) added bovine serum albumin acts as an acceljtor, (b) albumin is a product of liver cells, (c) an acidic amill terminus is apl)arently a requirement for acceptor protein, and (d) rabbit serum albumin, an acceptor proteitl, does indeed have glutamate at the amino ternGln1.
In this report we have studied in various lines of tissue culture cells a range of cell types and animal species in which argill>-l-tRNA protein transferase is found.
The ability of the argin>-l-tRNA protein transferase from these cells to add arginine to certain exogenous proteins as well as the capacity of various compounds to inhibit the enzymic activity are presented.
These data show that the properties of the transferase in cultured cell extracts are similar to t,hose already reported.
In additiolr, we have characterized the endogenous acceptor proteins of these cell lines by sodium dodecyl sulfate polyacrylamide gel elrctrophoresis and report an identity in relative migration of some of the acceptor proteins amongst all the cell type>. The suspcnsiou 1~:~s again ceiitrilugctl :I t low speed, the pellet collected and resuspended in a \-olumc of suspension medium equal to the pellet volunic.
Tlie suspciision was frozeu and thawed three times followed by three 10-s periods of sonication with a I<ronwill 13iosonik liltrasonicator. The disrupted cells were centrifuged for 2 hours at 5' in :I Spillco 50 5893 Ti rotor at 39,000 rpm after which the supernatant fluid was recovered for further experimentation.
For preparation of extracts from mouse kidney, the method was modified.
The kidneys were minced and then washed four to five times with phosphate-buffered saline to remove as much residual blood as possible.
This procedure rendered the fluid decanted from the minced &sue colorless to the eye. An equal volume of suspension medium was added to the washed minced tis:.iue which was homogenized by four or five strokes in an all glass homogenizer.
The suspension of disrupted cells was centrifuged at 4" for 20 min at 10,000 x g. The resulting supernatant fluid was centrifuged at 4" for 2 hours at 39,000 rpm in a Sl)inco 50 Ti rotor.
The supernatant fluid was collected. :\ssay of Arginyl-tRNA Protein Transferase and Formation of ;Irgir~yl-tRi\iA--Argin~l~tRNA protein transferase activity in l,hr 100,000 x g supernatant fluid was measured essentially as tIescribed by Dupras and de Lamirande (6). The contents of t,he incubation mixtures are indicated in the appropriate portions rnltler "Results." Radioactivit'y was measured in a scintillation nlilture with Triton X-100 as described by Patterson and Greene (11).

iciDS.dcrylanzide
Gel Analysis 0J" Endogenous Acceptor Profrji?l-One-milliliter samples of [W]arginine-labeled proteins formed by incubation of the 100,000 X g supernatant fiuid of cell extracts as described under "Results" were mixed with an equal volume of a 2% SDS solution containing sodium phosphate buffer (10 mnr) at pH 7.0 aud 2-rnercaptoethanol (200 mM). The >olutioils were mixed, heated at 100" for I min, and then dialyzed agailrst t1v-o 500-ml changes of electrode buffer described below. Sucrose and bromphenol blue dye were added t'o concentrations of 10 and 0.01 70, respectively.
The mixture \v:15 1)oured into glass cylinders (0.5 x 7.5 cm) immediat,el.y cwvrrrd with a layer of water and allowed to l>olymerize.
The coinl)osition of the electrode buffer was 0.8 g of NaZHP04, 0.6 g of SaI121'04.H~0, and 1.0 g of SDS in 1 liter of H%O. Electrol)lioresis was carried out at room temperature for approxinlatrly 2 llours at 200 volts. After completion of the electro-plioresi~ the gels were frozen at -7O", and cut into l-mm sections with a Diversified Scientific Instruments, Inc., gel slicer. The -1iches were prepared for countin, 0' by the method of Dingman :u~d Peacock (12)) which gave a recovery of 80 to 90% of the mlurts per min lmt on the gels.
The pH was then adjusted to 2 with concentrated HCl, and the unreacted tlitritroflllorobenzene was removed by three lo-ml washings with ether. .\n equal volume of 10% trichloroacetic acid was added to the aqueous phase, and the mixture was spun at 3000 x g for 5 min. The yellow precipitate was washed with 10 ml of ethanol followed by an equivalent quantity of ether, and then dried at room temperature over paraffin shavings. The dried powder was suspended in 3 ml of 6 N HCl, sealed in a glass tube under nitrogen, and digested at 105" for 12 hours. The digestion mixture was centrifuged at 3000 X g for 10 min and the supernatnnt liquid was dried i n vacua in a small Petri dish. The powder was then dissolved in 1 ml of acetone which contained 1 mg of unlabeled dinitrophenyl-L-arginine (Schwarz BioResearch Inc.) as a marker. It was necessary to add a drop of concen trated HCl to dissolve the arginyl derivative.
Descending chromatography of 0.1-ml samples of t,his solution containing 600 cpm was carried out on Whatman No. 1 paper using the butanoll-acetic acid-water (4:1:5) solvent system for 15 hours. The yellow spot was cut from the paper, placed in 10 ml of toluene 1,4-hi@(5.phenyloxazolyl)]benzene (POI'OP) and the radioactive content determined.
For BHK and L cells the rate ot incorporation diminishes aft,er about 235 hours. Similar time courses were obtained for the other cell lines t'ested. One explanation for the decrease in rate of incorporat,ion could be that nlmoat all of the available endogenous acceptor proteins hare had arginine added.
That such is t,lie case, has been substantiated by the experimrnt~ depicted in Fig. 1 is not due to exhaustion of the supply of arginine or ,1Tl'.
Second, Fig. 1C shows that after a similar 120-min incubation, the addition of bovine serum albumin, an exogenous acceptor protein (9)) results in a prompt and marked stimulation of the incorporation of [14C]argirrine into hot acid-insoluble product, thus indicat'ing that the other components of the assa? mixture are active. The lnost feasible interpret,ation of these experiments is t'hat, t'he incorporation of labeled arginine essen tially stops due t'o eshautitioii of available eridogenous acceptor protein.
Moreover, the plateau level observed should represent a reasonable estimate of the amount of available acceptor protein in any particular preparation, provided arginine and AiTP are l)reseiit in excess.
Opfi?izal Assay Co&i/ions-The optimal condit'ions for measuring arginyl-tRyA prot'ein transferase activity were determined for BHK and L cells when an ATP-generating system, arginine, and uncharged tRT\;A rvere used with the 100,000 X g superilatant fluid. The ol~tirnum for such a crude preparat,ion reflects, The ol)timal KC1 concentratiorl for L cell extracts was 35 mat and for BHK, 65 rnht. For both cell types, N&l was equally effective alld at the same concentrations as KU.
At 100 and 150 mM mercaptoethai~ol there was maximal transferase activity in L and I3HK cell extracts, respectively.
The pli versus transferase activity relationship for both BHK and L cells in the range tested (pH 7.4 to 9.6, 100 m&f Tris-HCl, 30-min incubation) showed a broad peak from about 7.6 to 8.6. Above pH 8.6, activity decreased gradually so that at pH 9.6 the activity was 507; that of the maximum.
Little an effect on arginyl-tRNX synthetase activity as well as increasing lability of arginyl-tRNX at the higher pH (6).

Specificity of Arginyl-tRNA Protein
Transjerase for drginine-- The specificity of the enzyme for transferring only arginine WI:: tested to determine whether it was the same as reported for the transferase from excised tissue. First, as may be seen in Fig. ID in a preparation of trarrsferase from WHK cells in which there is marked incorporation of labeled arginine a similar incubation with [Wjleucine shows no incorporation of radioactivity into the hot acid-insoluble fraction. Second, a mixture of several 14C-labeled amino acids (New England Nuclear, 14Cmlabeled, Lamino acid mixture) incubated with the BHK preparation also results in incorporation of radioactivity (Fig. 113). ;iddition of excess unlabeled arginine alone to this incubation almost corn pletely prevents measurable incorporation of radioactivity ( Addition of either unlabeled leucine or phenylalanine did not similarly diminish incorporation of radioactivity.

SDS-Acrylamide Gel Patterns oj" Endogenous Acceptor
Protein-The SDS-acrylamide gel patterns of ['4C]argilline-l:tbeled acceptor protein incorporated via the usual incubation are shower for several of the cell lines (Fig. 2). Of note, is the great sim ilarity of patter11 in the extract.s from the various species rep resented by the different cell lines. In all the cell extracts analyzed, a major peak is evident between Slices 15 to 20 (11) of the electrophoretograms.
The protein ill this peak has a molecular weight of approximately 65,000 based upon the migration of bovine serum albumin under the same conditions. In by guest on January 20, 2018 http://www.jbc.org/ I Ilr gels of the extracts from the various cells examilled, one or t,wo lesser peaks are also observable in gel Slices 9 to 13 and are :issumedly of greater than 65,000 molecular weight. Two peaks in Slices 9 to 13 (la and 16) were always seen with gels of incllbat ions of extracts from BHK, herpes-infected, and polyomat,r:Lllsfol n!ed (I'yH.7) IjHK.
In extracts from other cells only ctne Ileak (I) was aln-ays clearly discernible.
The number and 1Gtion of the peaks of radioactivity beyond the major peak are somewhat variable from preparation to preparation from even the same cells although a peak was almost always seen in the Slice 30 to 40 range. Treatment of the incubation mixture after incubation and prior to gel analysis with 50 pg per ml of pallcreatic ribonucleaxe for 30 min did not noticeably change the gel l):ltterns thus indicating that the observed peaks are not argitlyl-tRN;1.
2,4-Dinitrophenol analysis indicated that after correcting for 2,4-tlinitrophenol quenching 80 to 90% of the [Y']arginine associated with the endogenous acceptor proteins is at the amino termilial.
'I'hr SDS-acrylamide gel pattern of incubations of extracts front l)olyoma-transformed (PyH3 cells) and herpes simplexinfected BHK cells \vere examined for differences (Fig. 3, top and boltoln left). No striking difference was found in gel pattern in incubated extracts from these cells.
To show that the gel patterns observed are not an artifact of, or lleculiar to, tissue culture-grown cells, freshly excised mouse kithley, processed as described under "Experimental Procedure," The results are partly summarized ill Table I where the R, values of the I, Ia, Ib, and II peaks are c*ompared in electrophoretograms of the different cell preparations.
Note the a!most perfect agreement for the RF values of Peaks I and II from the various cell extracts and the close agreement for the Peaks I:I and Ib in the gels of incubations of extra& from the BHK al 1t1 related cells.
hkogenous Accepfor Pro!etns-As rnelltioned above, bovilie serum albumin exhibits acceptor activity. Such activity is S~IOJVJI for UHK cells in Fig. 4. This is in agreement with Soffer's work with rabbit, liver (9) as is the apparent accel)tor activit>-of bovine thyroglobulin (Fig. 4). The fact that the activity has bee11 shower in a variety of cells from different species allowed the investigation of species specificity of the enzyme activity.
X0 such specificity was fou~id. Preparations from cells of various species were tested with proteins (all albumins except for imnnmoglobulin G (IgG)) for acceljtor activity.
However, as may be see11 (Fig. 5), there are differences amongst the various combinations of enzymes and acceptor protein.
For example, mouse albumin acted as a much better acceptor protein for the transferase from human I-IeLn cells than mouse L cells, and bovine albumill-stimulated arginirre incorl)oration by human HeLa cell extracts much better than leaving an argilline resitlue, althollgh Thompson (15) has relwrted that, the amino-termit~al amino acid of human albumill is aspartate and therefore has the requisite acidic amino-terminal amino acid (9). It is 1)ossible that the source, purity, or mode of preparation of the human albumin is a factor also.
The relatively high concentration of canavanine required for such inhibition can be accounted for solely or in lwt~ by the reports that hi& concentrations of canavanine are needed to inhibit binding of arginine to tRNA by both bacterial (16) and rat liver (17) preparations of purified arginyl-tRK.2 .S~IIthetase.
Allende and Xllende (17) reljorted that with rat live, arginyl-tRNA synthetase lOO-fold more cannvanine than arginine (1 mM versus 0.01 rn*r) was needed for 50% inhibition of arginine binding. This is in excellent agreement with the present finding that loo-fold more cnnavanine than arginine (1.5 mx versus 0.015 mM) caused about 50 c/o inhibit,iorl of incorporation into hot acid-insoluble product (Table II). Soffer and Mendelsohn (7) reported that the enzyme from sheep thyroid is not inhibited by puromycin. Similarly, the arginyl-tRNA protein trausferase from UHK is inhibited neither by 2 X 10W4 M purornycin nor by 30 pg per ml of cyclohesimidr (Table II).
Soffer (9) has reported t'hat hemoglobin does not have accel)tol activity.
We also find this to be the case with the cell extracts esanlirled in this paper and, moreover, find that bovine hemoglobin appears to inhibit the transfer of arginine to other 1x0teins which do have acceptor activity (Fig. 6A). Hemin 11:~ :I similar effect (Fig. 6~1). Increasillg tile ar@lille (2oncelrtlxtioll or adding additional xceptor protein (bovine serum albumin) does not overcome the inhibition by llemoglobin (Table I I I