Department of Biochemistry, Oklahoma State University, Stillwater, Oklahoma 74078
Escherichia coli Crookes strain transports lipoic acid via a temperature- and energy-dependent system which is saturable. Because of lipoic acid's lipophilic nature special precautions must be taken in stopping uptake to prevent artifactual results. There is no sulfhydryl group requirement of the transport system. Both chloramphenicol and ammonium sulfide inhibited the incorporation of lipoic acid into the protein-bound form. Lipoic acid transport was sensitive to osmotic shock. Although a slight increase in the specific lipoic acid binding to certain fractions of the osmotic shock fluid was obtained, the existence of a lipoic acid-binding protein was not established. Membrane vesicles transported lipoic acid.
Lipoic acid functions as a cofactor in the oxidative decarboxylation of a-keto acids by the pyruvate and a- ketoglutarate dehydrogenase complexes (1). Activating enzymes catalyze the conversion of free lipoic acid to the protein-bound active form in both Escherichia coli and Streptococcus faecalis (2,3). This protein-bound form is linked through the x-NH2 group of lysine (4) and does not turn over in E. coli (5). Exogenously supplied lipoic acid is used preferentially by E. coli even though the organism synthesizes lipoic acid (6).
A constitutive system capable of concentrating lipoic acid 100-fold in a temperature- and energy-dependent reaction has been described in S. faecalis (7,8). Furthermore, the lipoic acid activating system in both S. faecalis 10C1 and E. coli Crookes was found entirely in the soluble fraction (9), suggesting that this system was not directly involved in the transport of lipoic acid into the cell.
This paper reports a study of lipoic acid transport in E. coli with special emphasis on the specificity and properties of the system, the effect of osmotic shock and the possible involvement of binding proteins, and interaction of lipoic acid with the cell membrane. Preliminary results of some of these studies have appeared (10,11).
Unlabeled DL - a - lipoic acid, 5- (1,2-dithiolan-3-yl) pentanoic acid, and 8-methylthioctic acid, 5- (1,2-dithiolan-3-yl) hexanoic acid, were generously supplied by Dr. D. S. Acker of du Pont. Dr. L. J. Reed of the University of Texas provided unlabeled lipoic acid and both the C7, 4-(1,2-dithiolan-3-yl) butyric acid, and C9, 6- (1,2-dithiolan-3-yl) hexanoic acid, analogs. Radioactive lipoic acid labeled with 35S was prepared as described by Leach (12). The specific activity of the lipoic acid was 46 µCi/mg.
Growth of Escherichia coli
E. coli Crookes ATCC 8739 cells were grown in M-9 minimal medium (13) supplemented with 0.2% glucose on a New Brunswick R-10 reciprocating shaker at 37 C. After overnight growth the cells were diluted 20-fold into fresh medium and incubated with shaking for 3 hr. The culture was then diluted with an equal volume of fresh medium and grown for 45 min.
Measurement of lipoic acid uptake
Cell preparations containing 1.1 to 1.3 × 109 cells/ml were incubated for 10 min in M-9 containing 0.2% glucose plus chloramphenicol (0.1 mg/ml) at 37 C. The radioactive substrate was added to a final concentration of 1 µM and the mixture was incubated with shaking for 10 min at 37 C. The uptake process was stopped by pipetting 0.5-ml samples onto 1 ml of crushed, frozen M-9. After centrifugation at 4 C and one wash with ice-cold M-9, the cell pellet was suspended in one ml of water, transferred to a scintillation vial and counted in 10 ml of Bray's scintillation cocktail (14). Samples were counted to a 1% standard counting error.
If incorporation (into protein-bound form) was being determined, the chloramphenicol was omitted and the reaction was
stopped with crushed, frozen M-9. The cell pellet was treated with 1 ml of 70% ethanol. Unless otherwise noted, data are presented from typical experiments which were repeated 2-3 times but not averaged.
Osmotic shock procedure (15)
Ten ml of cells grown on M-9 medium to the early stationary phase were added to 200 ml of fresh M-9 and incubated for 4-5 hr to obtain midlog phase cells. The cells were harvested and washed twice with 0.03 M Tris-HCl (pH 7.1). The washed cells (250 mg) were suspended in 20 ml of 20% sucrose in 0.03 M Tris with 10-4 M EDTA and the suspension slowly rotated at room temperature for 10 min. The cell suspension was then centrifuged at 104 × g for 10 min at 4 C. The supernatant fluid was removed. The cell pellet was immediately subjected to shocking by adding 20 ml of ice-cold 5 × 10-4 M MgCl2 solution. After 3 min, an equal volume of 0.05 M Tris-HCl (pH 7.1) containing 1 × 10-3 M MgCl2 was added to the cell suspension and the mixture centrifuged at 104 × g for 10 min. To count the viable cells, aliquots from the untreated and shocked cells were plated on agar plates containing 1% tryptone broth plus 0.5% NaCl.
Membranes were prepared according to the lysozyme/EDTA method of Kaback (16), except that since the Crookes strain is so susceptible to the lysozyme/EDTA treatment that few viable cells remain (less than 1 in 103 by plate count), the ultracentrifugation step was omitted. The final preparation was suspended in 0.5 M potassium phosphate buffer, pH 6.6, and stored in small aliquots at - 15 C. Membrane protein was assayed by the method of Lowry, et al. (17), using bovine serum albumin as the standard protein.
All equilibrium dialysis experiments were done using Chemical Rubber Company five-chamber equilibrium dialysis cells with a volume of 1 ml for each chamber. A single thickness of dialysis tubing (Arthur Thomas Co., 48 Å pore), cut to appropriate length and treated with 1 mM EDTA, was used as the dialysis membrane. Equilibrium dialysis suspensions contained 10 mM glu-
cose and 1 mM MgSO4, unless the effect of either glucose or Mg2+ was being studied, and penicillin/streptomycin to a concentration of 114 and 0.1 µg/ml, respectively. Stirring was accomplished by means of a plastic bead inside each cell and a Chemical Rubber Company rocker motor (8 rocks/ min), and the cell was held at the desired temperature (usually 22 C) for 15 - 24 hr. Aliquots were taken from each side of the cell, one side containing membranes or protein and ligand and the other side containing ligand only. These were counted in 10 ml of Bray's solution in a Packard Tri-Carb and results were corrected by comparison to appropriately quenched standards.
Measurement of Lipoic Acid Uptake
Table I, Part A shows a comparison of various methods for stopping the uptake of lipoic acid in whole E. coli cells. Quick freezing by ejecting the cell suspension into a centrifuge tube cooled in liquid nitrogen, slow freezing by ejecting the cell suspension into a centrifuge tube in an ice bath followed by freezing in a - 15 C room, and ejecting the cell suspension onto 2 volumes of ice medium frozen and chopped into fine crystals all gave equivalent results. No correction was made for any nonspecific adsorption, but the kinetic curves suggest that any such effect is minimal.
Because of a high background adsorption of lipoic acid by membrane filters, the usual method for terminating uptake was not suitable. About 10% of the lipoic acid is retained by the filters in the absence of bacterial cells and this background retention increases linearly with increasing lipoic acid concentrations. The background was not reduced by treatment of the filters with nonradioactive lipoic acid or by exhaustively washing the membrane filters with EDTA-or glycerol-containing solutions.
Part B of Table I shows the equivalence of results obtained when separating free pool lipoic acid from that which is protein bound, by either boiling or ethanol precipitation. Previous results showed that trichloroacetic acid treatment of cells to measure protein-bound lipoic acid leads to artifacts in the distribution of lipoic acid (7). Reaction termination with ice-slushy medium and precipitation by 70% ethanol treatment were the procedures adopted for the analyses reported in this paper.
Effect of Cell Concentration
Figure 1 shows the effect of various concentrations of E. coli cells on the uptake of lipoic acid during a 5-min incubation at 20 C. The amount of radioactive lipoic acid taken up is proportional to cell concentration down to 1 mg of cells/ml.
Effect of Lipoic Acid Concentration
The saturation curves for both free pool lipoic acid and protein-bound lipoic acid are shown in Fig. 2. The protein-incorporating system is one-half saturated at a lipoic acid concentration of about 30 mM while lipoic acid transport into the pool is one-half saturated at 50 µM.
Kinetics of Lipoic Acid Uptake
The time course was determined at 20 C over a short period as seen in Fig. 3A. Maximum uptake occurred rapidly within one min and after three min the uptake decreased gradually to a plateau reached at five min. A similar time course of lipoic acid uptake was observed with S. faecalis 10C1, with the maximum uptake being obtained at 30 sec. Figure 3B shows uptake over an extended incubation period at 37 C accompanied by cell growth. The incorporation of labeled lipoic acid into the pro-
tein-bound form parallels growth for the first 2 hr of incubation. The uptake into the free pool shows the typical overshoot kinetics.
Effect of Temperature on Lipoic Acid Uptake
When the lipoic acid uptake was measured at 10 and 20 C, the amount of uptake obtained at 20 C was twice that obtained when the cells were incubated at 10 C.
Effect of Sulfhydryl Reagents and Compounds on Lipoic Acid Uptake
Many transport systems require a sulfhydryl group or groups for activity and thus are inhibited by sulfhydryl-reactive reagents. Part A of Table 2 shows that 7-amino-1,3-naphthalenedisulfonate, p-chloromercuribenzoate, and N-ethylmaleimide do not inhibit lipoic acid uptake.
Table 2, Part B shows the effect of various sulfhydryl compounds on the free pool of lipoic acid. This experiment was done in the presence of chloramphenicol, which inhibits incorporation of lipoic acid into the protein-bound form (see the following section). Both cysteine and cystine were potent inhibitors and 2-mercaptoethanol was slightly stimulatory. The inhibition requires a concentration of at least 10 times that of lipoic acid, suggesting a low specificity of inhibition. A sulfhydryl component is probably not essential for the transport of lipoic acid.
Effect of Inhibitors and Analogs on Lipoic Acid Uptake
Experiments to determine the influence of various inhibitors on the accumulation of lipoic acid into the pool and incorporation of lipoic acid into the protein-bound form were performed by incubating midlog phase cells with the inhibitor 10 min before adding radioactive lipoic acid and measurement of accumulation and incorporation after 10 min incubation. Table 3 shows these results: sodium azide, chloramphenicol, and ammonium sulfide have no effect on the accumulation. 2,4-Dinitrophenol and potassium cyanide, which inhibit energy production, reduce the accumulation of lipoic acid. As seen before, cystine reduces accumulation and also incorporation into
the protein-bound form. The most effecttive inhibitors of incorporation of lipoic acid into the protein-bound form are chloramphenicol and ammonium sulfide. This inhibition coupled with the lack of effect on accumulation makes these compounds selective reagents for differentiation of accumulation and incorporation of lipoic acid into the protein-bound form. Since the effect of 2,4-dinitrophenol and potassium cyanide was greater on accumulation than on incorporation, the level of lipoic acid accumulated in the pool is greater than that required to saturate the lipoic acid-activating system for incorporation of lipoic acid into the protein-bound form.
Table 3, Part B shows the effect of incubation of cells for 3 min with 1 mM lipoic acid analogs before addition of 10 µM lipoic acid and measurement of incorporation and accumulation. The C7 and C9 analogs of lipoic acid slightly inhibit both accumulation and incorporation. Further, oleic acid inhibits incorporation but not accumulation. These results suggest that the transport system for lipoic acid is fairly specific.
Effect of Osmotic Shock on Lipoic Acid Uptake
Treatment of E. coli cells with osmotic shock has been used to implicate binding proteins in the transport of several specific compounds. Figure 4 compares the effect of shocking of E. coli on the transport of lipoic acid and of valine. The osmotic shock procedure reduces the uptake of both substrates; about 30% of the original uptake activity remains for lipoic acid while the reduction of valine uptake was 80%. Attempts to restore lipoic acid transport by the addition of the shock fluid to the treated cells have been unsuccessful.
Measurement of the binding of lipoic acid to the concentrated protein of the osmotic shock fluid by the technique of equilibrium dialysis revealed 7 pmol of lipoic acid and 9 pmol of leucine were bound per mg of protein.
Protein fractionation experiments on the osmotic shock fluid involving ammonium sulfate precipitation, DEAE-cellulose chromatography, and gel filtration all failed to separate a protein peak with increased
lipoic acid binding activity. The highest activity achieved with lipoic acid was 10.1 pmol/mg while with thiamine-binding protein an ammonium sulfate fractionation yielded a binding activity of 25.8 pmol/mg.
The uptake of lipoic acid by membrane vesicle preparations showed a biphasic concentration dependence (Fig. 5) which is reminescent of the pattern shown for thiamine uptake (18). When the proteins and lipids were separated from these vesicles and lipoic acid binding was measured on each component separately, only nonsaturable binding was observed.
Vitamins are transported into bacterial cells by systems which are just as diverse as those systems which function in the transport of other compounds (19). The vitamin transport systems are as one would expectefficient and specific. There are advantages in studying transport systems for vitamins: 1) the substrates have a rather limited function, 2) the low concentration of vitamins present in the natural environment is reflected by high-affinity systems, 3) the metabolic functions of most of the vitamins are known, and 4) the whole spectrum of transport mechanisms can be studied with one biologically functional class of compounds.
Because we were familiar with the enzymatic reactions for the conversion of lipoic acid to its functional form and had developed an interest in peptide transport, we undertook a study of the transport of lipoic acid. Several aspects of our research concerning lipoic acid transport in S. faecalis have been published (7-12). In this paper we have characterized the transport of lipoic acid in E. coli.
The uptake of lipoic acid by E. coli is proportional to the cell concentration (Fig. 1), is saturable with lipoic acid (Fig. 2), and is temperature-dependent. The uptake of lipoic acid is rapid (Fig. 3) and the same type of overshoot phenomenon is noted with lipoic acid in S. faecalis as with
other compounds. Becker and Lichstein (20) ascribe this commonly observed overshoot as a transinhibition of uptake (active accumulation inhibited by the intracellular pool) and the continued efflux of intracellular substrate.
Osmotic shock (Fig. 4) and 2,4-dinitrophenol (4 mM) reduced lipoic acid transport. Attempts to purify a lipoic acid-specific binding protein from the osmotic shock fluid were unsuccessful. Because of the lipophilic nature of lipoic acid, there are considerable nonspecific hydrophobic interactions. Those lipid solubility characteristics of lipoic acid required special precautions in the procedure used for stopping uptake and preventing of artifactual interactions.
Lipoic acid is accumulated by membrane vesicles (Fig. 5) of E. coli. Because of the lipophilic nature of lipoic acid and of the membrane preparations and/or vesicles, we turned our attention to thiamine transport in E. coli.
We have demonstrated the existence of a lipoic acid transport system in E. coli and defined some of its properties. Because of the nonspecific hydrophobic interactions that occur with lipoic acid, the study of vitamin transport systems is easier when other substrates are used.
This work was supported in part by NSF Grant #GB-3274, by American Cancer Society Institutional Grant #IN 91, by NIH Research Career Program Award CA-K3-6487, and Oklahoma Agricultural Experiment Station Project 1109. This is Journal Article J-4194 of the Oklahoma Agricultural Experiment Station. Drs. O. C. Dermer, R. E. Koeppe, and E. C. Nelson reviewed the manuscript and provided helpful comments.
1. L. J. REED, Acc. Chem. Res. 7: 40-46 (1974).
2. F. R. LEACH, K. T. YASUNOBU, and L. J. REED, Biochim. Biophys. Acta 18: 297 (1955).
3. L. J. REED, F. R. LEACH, and M. KOIKE, J. Biol. Chem. 232: 123-142 (1958).
4. K. DAIGO and L. J. REED, J. Am. Chem. Soc. 84: 666-671 (1962).
5. T. W. GRIFFITH and F. R. LEACH, Arch. Biochem. Biophys. 162: 215-222 (1974).
6. H. NAWA, W. T. BRADY, M. KOIKE, and L. J. REED, J. Am. Chem. Soc. 82: 896-903 (1960).
7. D. C. SANDERS and F. R. LEACH, Biochim. Biophys. Acta 82: 41-49 (1964).
8. L. E. WILSON and F. R. LEACH, Biochim. Biophys. Acta 82: 50-57 (1964).
9. Y. K. OH and F. R. LEACH, Can. J. Microbiol. 15: 183-187 (1969).
10. F. R. LEACH and Y. K. OH, Fed. Proc. 28: 870 (1969).
11. T. GRIFFITH, C. A. CARRAWAY, and F. R. LEACH, Fed. Proc. 30: 1115 (1971).
12. F. R. LEACH, Meth. Enzymol. 18A: 276-281 (1970).
13. E. H. ANDERSON, Proc. Nat. Acad. Sci. U.S.A. 32: 120-128 (1940).
14. G. A. BRAY, Anal. Biochem. 1: 279-285 (1960).
15. H. C. NEU and L. A. HEPPEL, J. Biol. Chem. 240: 3685-3692 (1965).
16. H. R. KABACK, Meth. Enzymol. 22: 99-120 (1971).
17. O. H. LOWRY, N. J. ROSEBROUGH, A. L. FARR, and R. RANDALL, J. Biol. Chem. 193: 265-275 (1951).
18. F. R. LEACH and C. A. C. CARRAWAY, Meth. Enzymol. 62: 76-91 (1979).
19. R. J. KADNER, in: B. P. ROSEN (ed.) Bacterial Transport, Dekker, N.Y., 1978, pp. 463-493.
20. J. M. BECKER and H. C. LICHSTEIN, Biochim. Biophys. Acta 282: 409-420 (1972).