Evidence against the Folate-mediated Formylation of Formyl-accepting Methionyl Transfer Ribonucleic Acid in Streptococcus faecalis R*

SUMMARY The formylation of formyl-accepting methionyl transfer ribonucleic acid (tRNA fMet) was investigated in Sfrepto-coccus faecalis R, an organism incapable of synthesizing folic acid but whose growth does not require the vitamin under deiined conditions of culture. These studies were carried out because it was not clear how an organism with no folate could initiate protein biosynthesis. Growth of S. faecabk R on media free of folate but containing serine, methionine, thymine, adenine, and guanine was not affected by the presence of the folate antagonists, trimethoprim and aminopterin. However, growth of the organism on media lacking thymine and dependent upon the presence of either folic acid or 5formyltetrahydrofolate was severely inhibited by the analogues. It was demonstrated that extracts prepared from S. jaecalis R grown on a synthetic medium containing folic acid catalyzed the formylation of methionyl-tRNAfMet when either or formate the of the formyl


SUMMARY
The formylation of formyl-accepting methionyl transfer ribonucleic acid (tRNA fMet) was investigated in Sfreptococcus faecalis R, an organism incapable of synthesizing folic acid but whose growth does not require the vitamin under deiined conditions of culture. These studies were carried out because it was not clear how an organism with no folate could initiate protein biosynthesis. Growth of S. faecabk R on media free of folate but containing serine, methionine, thymine, adenine, and guanine was not affected by the presence of the folate antagonists, trimethoprim and aminopterin.
However, growth of the organism on media lacking thymine and dependent upon the presence of either folic acid or 5formyltetrahydrofolate was severely inhibited by the analogues.
It was demonstrated that extracts prepared from S. jaecalis R grown on a synthetic medium containing folic acid catalyzed the formylation of methionyl-tRNAfMet when either 10.formyltetrahydrofolate or formate was added as the source of the formyl group, whereas extracts prepared from cells of this organism grown on a medium free of folate but containing serine, methionine, thymine, adenine, and guanine catalyzed the formation of methionyl-tRNArMet only when lo-formyltetrahydrofolate but not when formate was used as the source of the formyl group. Extracts of these folate-free cells catalyzed the formylation of methionyl-tRNAf Met by formate if tetrahydrofolate was added to the assay mixture.
These results indicate that S. faecdis R grown in the absence of folate but in the presence of the supplements does not contain tetrahydrofolate and therefore formylated methionyl-tRNAfMet may not be involved in initiation of protein synthesis in S. faecalis R, or the formylmethionyl ester of tRNAfMet (fMet-tRNArMet) may be formed by some as yet unknown enzyme system. richiu coli B lead to extensive experimentation demonstrating its role in initiation of polypeptide chains in E. coli (2)(3)(4)(5)(6)(7)(8)(9)(10)(11). Trimethoprim ' and aminopterin, folic acid analogues, are strong inhibitors of dihydrofolate reductase in various microorganisms (12).
Consequently, the biosynthesis of tetrahydrofolate and its derivatives is presumably blocked in the presence of such inhibitors, because the reduction of the pteridine moiety of dihydrofolate or folate is an essential step in the formation of tetrahydrofolate and its subsequent formylation to lo-formyltetrahydrofolate (12,13 (27), is not capable of synthesizing folic acid (28)(29)(30). Although this organism is commonly used in the microbiological assay of folate compounds (13), it has been reported that this requirement may be entirely replaced by addition of serine, methionine, thymine, and a purine base, adenine or guanine, to the growth medium (31)(32)(33) jaecalis R (ATCC 8043) was grown in Bacto-fclic acid assay medium (38) supplemented with either folic acid, 0.001 mg per ml (final concentration), or n-serine, 0.20 mg per ml; L-methionine, 0.20 mg per ml; thymine, 0.05 mg per ml; adenine, 0.01 mg per ml; and guanine, 0.01 mg per ml. The S. juecalis R inoculum culture used in growth experiments was prepared by a series of four transfers on the Bacto-folic acid assay medium supplemented with L-serine, L-methionine, thymine, guanine, and adenine.
The inoculum was prepared from cells transferred four times on this medium. They were then collected by centrifugation and suspended in an equal volume of medium deficient in folate and thymine.
A 1 '$X0 inoculum of the above suspension was used.
Cultures were grown for 12 hours at 37" and the absorbance at 660 nm was measured on a Bausch and Lomb Spectronic 20 in l &mm diameter tubes. Large scale growth of S. juecuZis R yielded 2.0 g of cells from 2.5 liters of either medium. The cells were harvested by centrifugation, washed twice with 0.9% sodium chloride, and stored at -90" for less than 1 week. E. coli A19 was grown in medium containing the following: KH2P04 (anhydrous), 2.0 g per liter; K&PO4. 3Hz0, 11.0 g per liter; (NH4)2S04, 4.0 g per liter; Casamino acids, 8.0 g per liter; MgS04, 0.24 g per liter; FeCh , 0.27 mg per liter; glucose, 12.0 g per liter.
The cells were harvested by centrifugation, washed, and stored at -90". Preparation of S-100 Supernatant Fractionssupernatant fractions, referred to as S-100 fractions, containing methionyl-tRNA synthetase activity, transformylase activity, and the enzymes necessary for the formation of IO-formyltetrahydrofolate were prepared by alumina grinding as described by Nirenberg (39), modified as follows.
The standard buffer contained 30 mM 2-mercaptoethanol in order to help stabilize any endogenous tetrahydrofolate present.
After centrifugation at 30,000 X g, the supernatant solution was centrifuged at 105,000 X g for 120 min.
A portion of the upper four-fifths of the resulting S-100 solution was then dialyzed 8 hours against 120 volumes of standard buffer with one change of the dialyzing medium. The dialyzed and nondialyzed S-100 fractions were then frozen quickly, stored at -9O", and used within 2 weeks. Protein Determinations-Protein concentrations were determined by a modification of the phenol reagent method with crystalline bovine serum albumin as the reference standard. The reagents used were twice as concentrated as those described by Lowry et al. (40)  After incubation at 37" for the amount of time indicated, the reaction was stopped by the addition of cold 10% trichloracetic acid, mixed vigorously, and the precipitate was collected on Whatman glass fiber filters which had been prewashed with 1 mR;I sodium formate.
The reaction tube was washed 5 times with cold 5% trichloracetic acid and the washes were decanted onto the glass fiber filter.
The filters were then counted for radioactivity in plastic scintillation vials containing 10 ml of Bray's solution in a Nuclear-Chicago Mark I liquid scintillation counter to give a standard error of counting rate of less than 0.5%.
All samples were checked for quenching by the channel ratio method but no correction was required.

Folic Acid Requirement
for Growth of X. jaecalis R-Growth of S. jaecalis R is dependent upon the addition of folic acid or a combination of the metabolites serine, methionine, a purine, and thymine.
These metabolites are now recognized to be products Issue of October 10, 1970 C. E. Samuel, L. D'Ari, and J. C. Rabinowitx 5117 of biosynthetic reactions that require the participation of a tetrahydrofolate derivative.
Since the Bacto-folic acid assay medium is a synthetic medium that contains hydrolyzed casein, adenine, guanine, and p-aminobenzoic acid, growth of S. faecalis R on this medium was found to be entirely dependent upon the addition of folic acid or thymine. The relative rates and extents of growth of the organism in the presence of optimal quantities of folic acid or thymine, and additional serine, methionine, adenine, and guanine were compared. The inoculum used was grown through four successive transfers on folic acid-free medium with supplements in order to assure the absence of any folic acid in the inoculum. The growth rate on the medium containing folic acid was somewhat faster than on the non-folate medium, with doubling times of 34 and 48 min, respectively. The yield of cells after 24 hours growth was the same for both media.
E$ect of Folate Antagonists-Although the results of the experiment described above suggest very strongly that S. faecalis R can be grown in the absence of any folate when it is supplied with various products of folate-mediated reactions, we considered the possibility that the organism is capable of very limited synthesis of folate, and that the limited amount that is made is used exclusively for the transfer of formyl groups to Met-tRNArMet for initiation of protein synthesis. The effect of the folate antagonists trimethoprim and aminopterin on the growth of S. faecalis R under the two conditions of culture was therefore determined. Neither trimethoprim (Fig. 1) nor aminopterin (Fig. 2) inhibited growth of S. fuecalis R grown in a folate-free medium supplemented with thymine, whereas both analogues were effective inhibitors of growth in the presence of added folate or 5-formyltetrahydrofolate.
Aminopterin was active as an inhibitor at much lower concentrations than was trimethoprim, and was much more effective in the presence of folate than it was in the presence of 5-formyltetrahydrofolate.
These results provide additional evidence that S. faecalis R can grow at a relatively normal rate in the absence of added folic acid and that the cell does not contain any functional folic acid. Formation of the radioactive product was dependent upon the presence of both Lmethionine and tRNA in the standard assay as described under "Experimental Procedure." No labeled product was formed when the S-100 fraction was omitted or boiled before use in the standard assay.
The S-100 fraction prepared from X. fuecalis R grown in me-  Fig. 3. The reaction was not inhibited by [14C]formate but was slightly inhibited by tetrahydrofolate (Fig. 3). Similar re- s&s m-ere obtained with S-100 fractions of X. faecalis R grown in medium lacking folic acid but supplemented with L-serine, L-methionine, thymine, adenine, and guanine (Fig. 4). This reaction was likewise not inhibited by [Wlformate but was slightly inhibited by tetrahydrofolate (Fig. 4) Under these conditions, the S-100 preparation from S. faecalti R grown with folate catalyzed the formation of fMet-tRNAfMet from the [Wlformate (Fig. 6). The reaction was dependent upon the amount of S-100 preparation used but exhibited autocatalytic behavior at low enzyme levels (Fig. 7). The reaction was linear with respect to enzyme concentration over a narrow range of enzyme concentrations.
The S-100 preparation obtained from X. jaecalis R grown in the absence of folate in a medium supplemented with L-serine, L-methionine, adenine, guanine, and thymine was not capable of catalyzing the transformylation reaction with [14C]formate under comparable conditions (Figs. 6 and 7). However, upon addition  (Fig. 8).
The formylation of Met-tRNArMet by S-100 preparation of X. faecalis R cells grown on medium containing folic acid was not inhibited, but was rather slightly stimulated by the addition of an equal amount of S-100 from cells grown on medium lacking folic acid (Table I), thus establishing that this S-100 fraction does not contain an inhibitor that interferes with fMet-tRNA formation by the S-100 fraction prepared from cells grown without folate.
Control experiments carried out with S-100 preparations of E.
coli A19 demonstrated that transformylation occurred in the absence of added tetrahydrofolate with this extract when ['"Clformate was supplied as the formyl donor. DISCUSSION Previous work has shown that the folic acid requirement of S. faecalis R may be replaced by supplementing the growth medium with a mixture of substances that are now recognized to be products of biosynthetic reactions in which folic acid functions as a cofactor.
Microbiological assay of cells grown in the absence of folic acid, but with these supplements, failed to detect the presence of any folate (28,32), suggesting that S. jaecalis R cannot synthesize folic acid de novo. In experiments reported here, it is shown that the growth rate of the organism on the folate-free medium is somewhat less than on a medium containing folic acid, but that comparable cell yields are obtained on both media. Therefore, in view of the role of lo-formyltetrahydrofolate in the Met-tRNAfMet transformylation reaction to yield fMet-tRNAyMet, and the role of this material in the initiation of polypeptide chain biosynthesis in bacteria, it seemed as if S. fuecalis R might provide an interesting exception to the generally accepted mechanism for initiation of protein biosynthesis in bacteria.
Evidence for the complete absence of folate in X. faecalis R cells has been further substantiated by demonstrating that cells grown in the absence of folate by the addition of various metabolic supplements are resistant to the folic acid analogues trimethoprim and aminopterin. However, these analogues inhibit The possibility that initiation of protein biosynthesis in prokaryotes may not in all cases necessarily involve formylated Met-tRNArMet also deserves consideration in view of observations concerning this process in eukaryotes. It has not been possible to demonstrate the presence of formylated aminoacyl-tRNA in eukaryotic cytoplasmic sources such as rat liver (23), guinea pig liver (43), rabbit reticulocytes (25), chicken reticulocytes (44), or HeLa cells (24).
It has been suggested that Met-tRNAfMet, whether formylated or not, may possess a unique structural conformation which facilitates initiation of protein biosynthesis (5,45,46). The results presented in this paper indicate that such a mechanism involving a particular conformation of methionyl-tRNArMet that allows selective binding to the initiation site on the 30 S ribosomal subunit might also be operative under certain conditions in S. juecalis R, a 70 S ribosomal system.