THE JOURRAL OF BIOLOGXAL Cmimsr~~ Vol. 247, No. 19, Issue of October 10, PP. 6323-6331, 1972

Printed in U.S.A.

Transport of Succinate in Escherichia coli

I. BIOCHEMICAL AND GElSETIC STUDIES OF TRAKSPORT IN WHOLE CELLS*

(Received for publication, February 23, 1972)

THEODORE C.Y. Lo,1 M. I<HALIL RAYMAN,$ AND BISHNU D. SANWAL

From the Department of Cell Biology, Xeclical Xciences Building, University of Toronto, Toronto 181, Canada

SUMMARY

A mutant of Escherichia coli unable to metabolize succinate has been used to study the nature of succinate transport in whole cells. It is demonstrated that cells accumulate suc- cinate against a concentration gradient. The uptake of succinate is competitively inhibited by fumarate and malate. The Michaelis constants for the dicarboxylic acids are all in the range of 15 to 30 pM. The accumulation of succinate is prevented by various energy poisons and sulfhydryl reagents. Inhibitors of mitochondrial dicarboxylate transport are with- out any effect on the succinate uptake system of E. coli. The influx and efflux of succinate in whole cells is temperature dependent. The influx increases rapidly between 15 and 36” and efflux between 36-50”.

The transport system is induced by succinate and re- pressed by glucose. Mutants lacking adenyl cyclase and some lacking enzyme I of phosphotransferase system acquire succinate transport activity when grown in the presence of succinate and cyclic 3’,5’-AMP. Mutants of E. coli unable to transport succinate fall into two broad categories which are referred to as dcf (dicarboxylate transport) and cf (car-

boxylate transport). The dcf mutants fail to grow on dicarboxylic acids (fumarate, malate, and succinate) but grow normally on the monocarboxylic acid, lactate. The cf

mutants do not grow on either the dicarboxylic acids or on lactate. Both classes of mutants grow well on acetate. The

dcf class of mutants can be genotypically separated into two categories. The dcf A mutants map at around 69 min, and the dcf B mutant maps at about 17 min (linked to galactose locus) on the genetic map. The cf mutants are also linked to galactose. It is inferred from these experiments that there are at least two components involved in the transport of succinate in E. coli. It is suggested that the cf gene possibly specifies a carboxylate binding protein. This suggestion is based on the evidence that osmotically shocked cells of E.

coli lose the capacity for uptake of succinate.

* This work was supported by a grant from the Medical Re- search Council of Canada.

$ Predoctoral Fellow of the Medical Research Council of Canada.

0 Postdoctoral Fellow of the i\ledical Research Council of Canada.

With the demonstrated existence in recent years of transport systems for almost every compound of metabolic interest in bacteria (review in Reference l), the pressing need seems to be the elucidation of the chemical structures and molecular mecha- nisms that are involved in the transport of metabolites through the limiting membrane. Progress in this area would demand study of transport systems which are amenable to a combined biochemical and genetic approach. The success in the elucida- tion of the mechanism of group translocation of certain sugars in Escherichiu coli cells (l-3) testifies to the suitability of such a combined approach. It is with this aim in view that we initiated studies of succinate transport in E. co&.

Compared to other transport systems, very few studies have been made on the mechanism of transport of di- and tricarboxylic acids in bacteria (4-8). Kay and Kornberg (7) described a di- carboxylic acid transport system in E. coli which is responsible for the uptake of succinate. Unfortunately, however, most of their studies were made on cells in which metabolism of succinate after uptake was not prevented by suitable genetic or inhibitor blocks. Kay and Kornberg (7, 8) did show, however, that a component determined by dct (dicarboxylate transport) gene was probably required in the transport process.

Genetic studies of various workers have indicated that the phosphotransferase system of Roseman (2) may also be some- how involved in the transport and utilization of succinate. Thus, E. co& strain MM 6 lacking enzyme I (pts I) of the phos- photransferase system is not only defective in the utilization of glucose, maltose, and some other sugars but also is unable to grow on succinate (9-11). Similar observations have been made in other E. coli strains lacking some components of the phospho- transferase system (12). It is quite possible that the defects in these cases are not directly related to transport and metabolism of succinate but mediated indirectly by cyclic-AMP,’ as has been shown for the metabolism of carbon sources other than glucose in enzyme I defective mutants (13, 14).

In order to be able to undertake and interpret cell-free studies of succinate transport at the molecular level we considered it desirable to clarify first the following questions in whole cell studies. What is the extent of involvement of phosphotrans- ferase system in succinate uptake? How many genetically de- fined components are involved in transport? What is the na-

1 The abbreviation used is: cyclic AMP, cyclic 3’,5’-AMP.

6323

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Strain designation

Hfr H Hfr wild type CB 10 sdh CB 11 sdh, frd nm 6 pts I, dct B

MP 259 CB 13 w 804

GO4 G

UC 118 CB 15 CB 16 CB 17 CB 18 CB 19 CB 20 CB 21 UC 14 CB 22 CB 23 CB 24 903

903 c

w 602

Hfr, cya cya, sdh, frd gal K, leu, Sm dct B, leu, Sm Hfr, pro, ade glt A glt A, cl” glt A, eta glt A, ct” Hfr, dct A Hfr, dct A I?fr, dct A pro, met

pro, met, ctQ pro, met, eta pro, met, eta pro, his pro, his, pts I leu, bio, gal

6324

ture of transport process (kinetics, substrate specificity, etc.) in cells in which transport and catabolism of succinate are sepa- rated?

EXPERIMENTAL PROCEDURE

l?. coli K-12 Strains-All of the strains used in the present work with their relevant genetic characteristics are listed in Table I.

,Iledia-The following media were used. Medium A was a minimal salts medium (16) with citrate omitted. It was sup- plemented with amino acids and other growth factors, wherever required, at a concentration of 25 pg per ml. Unless specified otherwise, carbon source was added at a concentration of 4 mg per ml. LB medium consisted of 1% Bacto tryptone, 0.5% yeast extract, and 1% N&I. The solution was neutralized to pH 7.0 with NaOH. LB agar was LB medium solidified with 1.1% Bacto agar. Cells were optimally induced for succinate trausport by addition of 85 mM succinate to LB medium.

Growth of Cells-All bacterial strains were grown under vig- orous agitation on a gyrotary shaker at 37”. Cell growth and cell concentration was determined with a Klett-Summerson calorimeter using Filter 66. Strain CB 11 was used to calibrate the readings of optical density with wet and dry weight deter- minations.

Genetic Xethods-Transductions with coliphage P1 were per- formed according to Lennox (17). The procedure used for

TABLE I

The genetic characteristics of strains used in this work

The abbreviations following the nomenclature of Demerec el a./. (15) are: cya, adenyl cyclase; sdh, succinate dehydrogenase; frd, fumarate reductase; dct A, dicarboxylate transport; dct B, dicarboxylate transport; pts I, enzyme I of phosphotransferase system; gal K, galactokinase; gll A, citrate synthase; Zeu, arg, met, pro, bio, ride refer to growth requirement for leucine, arginine, methionine, proline, biot,in, and adenine, respectively; Sm, streptomycin resistant.

Genetic markers Origin

Pasteur Institute This work This work D. Fraenkel (Pasteur Institute

collection) M. Pearson (Toronto) This work Pasteur Institute This work D. Hoar (Toronto) Laboratory stock This work This work This work This work This work This work D. Hoar (Toronto) This work This work This work M. L. Morse M. L. Morse C. Fuerst

n Probable genotype (carboxylate transport).

bacterial matings was based on the method of Curtiss et al. (18). The mating pair-s were separated by shaking in the sabre saw ap- paratus of Low and Wood (19). Mutants were selected by mutagenization with N-methyl-N’-nitro - N -nitrosoguanidine (20) followed by enrichment using penicillin (21). Hfr strains were cured of the F factor, when required, by growth in the presence of sodium dodecyl sulfate (22).

Isolation of X&ants Defective in Xuccinate Transport-Three differeut methods were used to isolate mutants defective in the transport of succinate and the other two dicarboxylic acids, fumarate and malate.

Jfethod I-This method was essentially that used by Kay and Kornberg (8). Approximately lo8 bacteria were plated without prior mutagenization on minimal medium containing 0.4y0 acetate (as the sole carbon source) and 0.05 mM 3-fluoromalate. The papillae that arose after 48 hours incubation at 37” were picked up and single colony isolations were made by restreaking on fluoromalate-containing medium.

Method II-Cells grown on LB medium were mutagenized and expressed for a few generations in Medium A containing acetate (0.4y0). The cells were then centrifuged, suspended in Medium A containing 0.4a/, succinate, and subjected to penicillin selection. Mutants were plated on acetate minimal medium and subsequently replica plated on succinate-containing medium. Colonies that did not grow on succinate were picked and purified.

Jleth,od III-This method, referred to as tritium suicide method, was based on the assumption that in a mixed population of cells, those able to transport tritiated succinate would be killed while those unable to do so would survive. Log phase cells were mutagenized and allowed to recover in minimal me- dium containing acetate. The cells were then exposed to 1.5 mCi of sodium [2, 3-3H&uccinate per ml for 90 min at 37”. The cells were then washed twice with cold Medium A, suspended in acetate minimal medium, and stored at 4”. At various time intervals (24 hours, 48 hours, etc.) samples were taken and spread on LB-succinate (0.4%) medium. After 16 hours, plates with well isolated colonies were replica plated on to LB-succinate (0.05 m&l sodium succinate) medium containing 1.5 PCi of so- dium [2, 3J4C2]succinate. These were incubated at 37” for 5 hours. -4 filter paper was then carefully pressed on top of the colonies. It was removed, dried, and exposed to a No Screen Kodak x-ray film. On the developed film wild type colonies appeared as dark circles while the succinate transport negative colonies appeared greyish. The latter type of colonies were picked from the original plate and purified. The radioauto- graphic method has in the past been used by Zwaig and Lin (23) and others (24) to identify transport deficient mutants for other metabolites.

Enzyme Assays-For enzyme assays cells were grown over- night in the required medium, washed with 0.1 M Tris-Cl buffer, pH 7.6, and sonicated. After centrifugation, the supernatant or the pellet (suspended in a small amount of Tris-Cl buffer) was used to measure enzyme activities. Succinate dehydro- genase (EC 1.3.99.1) was measured both in the supernatant and the pellet by the method of King (25). Succinic thiokinase (EC 6.2.1.5) was measured in the supernatant by the method of Gibson et al. (26). Standard procedures given in the Methods of Enzymology (27) wereused to measure fumarase (EC 4.2.1.2), malate dehydrogenase (EC 1.1.1.37), and malic enzyme (EC 1.1.1.40). P-Enolpyruvate carboxykinase (EC 4.1.1.32) and citrate synthase (EC 4.1.3.7) were measured according to Wright and Sanwal (28). Fructose diphosphatase (EC 3. 1.3.11) was assayed according to Fraenkel and Horecker (29).

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6325

Presence of fumarate reductase was tested using the method of Hauber and Singer (30). The specific activities of the various enzymes were calculated on the basis of micromoles of product formed per min.

Succinafe Metabolism in Vifro-Cells grown in LB medium plus 1% succinate were harvested at log phase. The cells (3.1 mg dry weight) were washed twice and suspended in 5 ml of 0.05 M phosphate buffer pH 7.5. Sodium [2,3-%]succinate (17.4 PCi) was added to the cells to a final concentration of 0.05 my and the suspension was incubated at room temperature for 10 min. After incubation the cells were harvested by filtration through Millipore filters. The filter was immediately trans- ferred to boiling water. After 15 min the filter and debris were removed by centrifugation and t.he supernatant was evaporated to dryness in an Evapo-mixer. The residue was redissolved in 200 ~1 of distilled water. Thirty microliters of the extract was applied to MNSOOG cellulose (polygram ccl 300) thin layer plate and subjected to chromatography using a solvent consisting of ethanol, ammonia, and water in the proportion of 80:20:10. Autoradiography was performed by exposure of the plates to Kodak x-ray film (No Screen) for 7 days.

Succinafe Transport Assay in Whole Cells-Cells grown in LB medium plus 1% succinate were harvested at log phase. The cells were washed twice and suspended in 0.05 M phosphate buffer pH 7.5. Cell concentration was adjusted to 0.64 mg dry weight per ml. In typical experiments, 5 ml of 0.10 mxr sodium [2,3- Y$]succinate were added to 5 ml of the cell suspension. Re- agents and cells were held at 23” before and after mixing. At various time intervals, aliquots of 1.5 ml were filtered through Millipore filter and washed with 5 ml of 0.05 M phosphate buffer pH 7.5. The whole procedure was completed within 10 s. The Millipore filter was then dissolved in 5 ml of Bray’s scintillation fluid (31). The radioactivity of the sample was counted in a scintillation counter. The amount of [2, 3XZ2]succinate adher- ing to the bacterial cells was measured by using cells pretreated with formaldehyde (32). This amount was subtracted from the experimental values.

Cell Shocking Procedure-Cells were osmotically shocked by the method of Heppel (33). Log phase cells were washed with 0.05 M phosphate buffer pH 7.5. The cell pellet was suspended in 80 parts of 20% sucrose containing 0.03 M Tris-HCl pH 7.2 and 0.1 mM EDTA. The cells were stirred gently for 10 min at room temperature and centrifuged. The cells were then rapidly dispersed by vigorous stirring in 80 parts of cold distilled water. The shocked cells were centrifuged again and resuspended in 0.05 M phosphate buffer.

Chemicals-Except where otherwise indicated, all chemicals were obtained from commercial sources and were of the highest available purity. [2,3-14C2]Succinic acid (23 mCi per mmole) and [2,3-aHz]succinic acid (100 mCi per mmole) were obtained from Amersham/Searle Co. Disodium fluoromalate was a gen- erous ‘gift from Dr. Paul Kent (Oxford). Pentylmalonate and n-butylmalonate were gifts from Dr. R. Williams (Toronto). 2.Phenylsuccinic acid was donated by Dr. S. I. Chavin (Dur- ham). Showdomycin was a gift from Shionogi Research Lab- oratory (Japan). Membrane filters (0.45 pm) were obtained from Schleicher and Schuell Inc.

RESULTS

Isolation of a Strain for &c&ate Uptake Studies-Essential for studies of succinate transport is the availability of a strain of E. coli which is unable to catabolize succinate to any significant extent. The major enzymes immediately involved in the catab-

olism of succinate are the two succinate dehydrogenases, one of which also functions as a fumarate reductase (34). A mutant lacking succinat’e dehydrogenase and fumarate reductase was made by a two step procedure. Hfr H was mutagenized and a mutant lacking succinate dehydrogenase (sdh) was selected by penicillin enrichment. The sdh mutants do not grow on minimal succinate medium but grow readily on fumarate minimal me- dium. From a sdh mutant a strain was selected which had lost the capacity to grow on fumarate as a sole carbon source and which was also unable to grow on glycerol plus fumarate anaero- bically (a test suggesting lack of fumarate reductase cfrd), Ref- erence 34). The sdh, frd markers (which are linked)2 were trans- ferred to the female strain W 604 (gal) by mating and a gulf, sdh, frd recombinant was selected. This strain (CB 11) had normal levels of all of the enzymes required for succinate metab- olism (for a list of enzymes assayed, see “Experimental Pro- cedure”) except succinate dehydrogenase and fumarate reduc- tase. The specific activities of the enzymes assayed in glycerol grown, stationary phase cells of a wild type organism (W 604) were 0.06 for succinate dehydrogenase and 0.008 for fumarate reductase. In strain CB 11, under identical growth conditions the specific activities were 0.0012 and 0.001, respectively.

Absence of Succinafe Catabolism in CB Ii-For studies of suc- cinate transport it was important to show that succinate taken up by CB 11 cells remains essentially unchanged. Cells in which the succinate transport system had been induced (see later) were allowed to take up [r4C]succinate for varying lengths of time and extracted in boiling water. Thin layer chromatography of ex- tracts revealed that over 95% of the radioactivity in different ex- periments was present in a spot whose RF value was the same as the reference [iQuccinate (Fig. 1). Some very faint spots re- vealed by autoradiography (Fig. 1) were probably due to slight contamination of the commercial [14C]succinate utilized in our experiments. In contrast to our results, Kay and Kornberg (7) have demonstrated that wild type as well as the sdh mutants of E. coli metabolize succinate rapidly to glutamate.

Inducibility of &c&ate Transport-Earlier studies of other workers (7,8) have demonstrated that succinate is taken up much faster by E. coli cells which have been grown on succinate as the sole carbon source; this induction of transport, however, is sensi- tive to catabolite repression by glucose. Since it is well known that succinate will induce a number of enzyme systems responsi- ble for the metabolism of succinate (26, 35, 36) and since in previous studies (7) no attempt was made to use cells in which catabolism of succinate was prevented, we decided to re-examine this question using strain CB 11. The results of this experiment are summarized in Fig. 2. It may be noted that the highest ac- cumulation of succinate is achieved in CB 11 cells grown on suc- cinate and least in cells grown on glucose. In comparison to CB 11, the apparent amount of succinate taken up in induced cells of Hfr H (wild type) or CB 10 (sdlb) is extremely high (Fig. 3). This is understandable in view of the fact that in wild type or sdh strains there is a metabolic drag on the succinate taken up through the transport system.

Since the succinate transport system is repressed by growth of cells on glucose (Fig. 2), the question arises whether this is due to catabolite repression (37). Pastan and Perlman (13, 38) have shown that most glucose sensitive enzymes require cyclic AMP and a cyclic AMP receptor protein for their induction. It is also known (39, 40) that glucose lowers the concentration of cyclic AMP in the cells by unknown mechanisms, thus repressing the

2 Lo, Rayman, and Sanwal, unpublished observations.

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6326

synthesis of inducible systems. The relationship of cyclic AMP into it via conjugation the cyo gene of the Hfr strain MP 259. to the induction of succinate transport was investigated in Results presented in Fig. 4 show that the succinate transport sys- strains MP 259 (lacking adenylate cyclase) and CB 13 (Table I). tern is induced in CB 13 only when the cells are grown in suc- The latter strain was constructed from CB 11 by introducing cinate and cyclic ANIP. Each of these compounds added alone

0 4 8 12 16 20 TIME,hn 1

FIG. 1 (left). Fate of succinate in strain CB 11. Sample 1 is the extract from cells loaded with [2,3-W&uccinate for 10 min. Sample 2 is the [2,3J%&uccinate standard. Autoradiograms were prepared as described under “Experimental Procedures.”

FIG. 2 (center). Effect of growth on glucose and succinate on succinate uptake by strain CB 11. Cells were grown in LB me-

dium with and without additions, indicated in the figure. Stand- ard uptake assay procedure was followed.

FIG. 3 (right). Succinate uptake by various strains of E. coli. Cells were grown in LB medium containing 1% succinate. Hfr H is a wild type strain. Strain CB 10 lacks succinate dehydro- genase and CB 11 lacks both succinate dehydrogenase and fumarate reductase.

TEMPERATURE,(“C)

FIG. 4 (left). Succinate uptake by various mutants grown in t’he presence or absence of cyclic 3’,5’-AMP. Two mutants lacking adenyl cyclase (cya) are used: MP 259 (cya) and CB 13 (cya, sdh, frd). Uptake by CB 11 (cyu+, sdh, frd) is shown as control. All of the strains were grown in LB medium containing 1% sodium succinate. In addition, as shown in the figure, cyclic fiIP at the concentrations indicated was added to the growth medium of the cua strains.

FIG. 5 (center). Temperature dependence of succinate uptake and exit from strain CB 11. Uptake experiments were done as described in the text. Exit experiments were done by first pre- loading the cells at 23” for 10 min. Cells were then spun down in an Eppendorf centrifuge. The supernatant was discarded. The inner surface of t,he tube was rinsed gently with 0.05 M phosphate

I / / I / /

12.- 0 3 mM molate

buffer pH 7.5 pre-equilibrated at the required temperature. The cell pellet was then resuspended in equilibrated buffer. The whole process took about 30 s. The cell suspension was incubated at the appropriate temperature for 10 min. The cells were spun down again (15 s). Radioactivity of the supernatant was then determined.

FIG. 6 (right). Competitive inhibition of succinate uptake by dicarboxylic acids. The concentration of the inhibitor is indi- cated above t’he lines. The inhibitor was added simultaneously with succinate to the reaction mixture. Amount of succinate taken up at 23” in 1 min was determined. The velocity (V) is expressed in nanomoles X min-r X mg dry weight-r. S refers to the concentration of succinate.

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

to the growth medium has no inducing effect. Again, a? is evident for the wild type strain in Fig. 3, 1\IP 259 shows :qq)arent succinnte uptake rates much higher than CB 13. This is ver3 probably due to the simultaneous induction with t,lle transport sys:tem of euzl-mes which metabolize succinate md create a drag on the accumulated .succinate. Cyclic AYIP added at a concell- tration of 5 to 10 ~RI to uninduced cells of CB 13 during uptake esperimeilts had no effect on succinate transport.

Absence oj Involwnent of Phosphotransjernse Systenl--As was mentioned earlier, many of the pts I mutants, lacking enzyme I of the I)llo~l)llotrailafer:i~e system reported in the literature, show a eimultalleou:, lack of ability to grow on glucose, some other sugars, and succinate. The extent of involvement of pts in suc- cinate uptake ~vah investigated in strains 1\ILI 6 and 903 C. In str‘lin 1111 6, it was first established that all of the enzymes ill- volred in succinate metabolism (succinate dehydrogenase, fu- marase, malate dellydrogenase, P-enolpyruvate carboxykinase, malic enzyme, succinic thiokinase, and fructose diphosphatase) were present but the activity of the phosphotransferase system, measured by the toluenization procedure of Gachelin (41), was almost nonesistent. -1ccumulation of succinwte in MM 6 was also negligible in transport assays. If the absence of succinate transport system in 1\1JI 6 was due to pts 1 mutation it was es- pected that revertants of t,he pts locus would regain in one step the ability to grow on glucose as well as succinate. Spontane- ously occurring mutants of MM 6 were, therefore, selected on glucose minimal and succinate rninimal media. The charac- teristics of these revertants are shown in Table II. Four points are worthy of note in this table. First, MM 6 is unable to grow on the three dicarbosylic acids, malate, fumarate, and succinate. Second, revertants selected separately for growth on glucose (AI11 6-G) and succinate (1\11\1 6-S) are unable to grow on the dicarbosylic acids and glucose, respectively. Third, MM 6-G grows well with acetate or lactate as the sole carbon source which suggests that t,he functioning of tricarboxylic cycle and oxidative phosphorylation is normal in these strains. Fourth, the over-all activity of the phosphotransferase system, as measured by (Y- methyl glucoside phosphorylation. by permeabilized cells (41), is very much depressed in LIM 6 and MM 6-S, but is present at near normal levels in the revertant (NM 6-G) which can grow on glu-

TABLE II

Growth of strain iKiK 6 and its revertants

Approximately lo8 cells of MM G were spread on Medium B containing either glucose or succinate as the sole carbon source (O.JTC). The papillae arising after 48 hours incubation at 37” were purified by streaking. MM G-S (succinate revertant) XT-as recovered at a frequency of 10e6 and MM G-G (glucose revertant) at a frequency of 5 X 10d6. The symbols + and - denote pres- ence or absence of growth in 48 hours on media containing the relevant carbon source (0.4y0).

Strain

Growth on

Lac- tate

+

+

+ !

Activity of the phospho-

Malate transferase fumarate systems succinate

- 0.008 + 0.007 - 0.24

a Activity of the phosphotransferase system x-as measured in glycerol-groxx, whole cells by toluenization procedure of Gache- lin (41) and is expressed as nanomoles of or-methylglucopyranoside phosphate formed per min per pg (dry weight) cells. nology suggested by Lm (1)

6327

coce but not on the dicarboxylic acids. These results support the conclusion that pts is not involved in succinste transport and tlrat 1\1\1 6 harbors another mutation which specifies a compo- nent of succinate (and, possibly other dicarboxylic acids, see Reference 7) uptake system. This locus designated dct B (see later) can be transferred from 1111 6 by transduction to other 6. coli strains (Table IV). In one experiment PI phage was grown on NM 6 (gal+, dct B) and the phage lysate was used to transduce strain W 604 (gal I<, dcf B+). Gal+ recombinants were selected. Out of 148 gulf transductunts 21 were found to have inherited the dct B locus. Thus, the latter is linked to gal K. As expected from the location of pts on the genetic map (42), none of the gal+ recombinants inherited pts I. That the strains carrying dct B lack the capacity to transport succinate and other dicnrbosylic acids is documented later.

The second strain 903 C carrying pts I3 and unable to grow on succinate behaves differently than MM 6. Pastan and Perhnan (13) have shown that addition of cyclic AMP to some pts Z mu- tants reitores tlleir ability to be induced for a number of inducible systems. Since succinate uptake is mediated by an inducible system it seemed possible that cyclic AI\IP rnay restore succinate uptake and metabolism in strain 903 C. This was indeed found to be the case. Cells grown in 10 lllM cyclic AhIP and succinate show considerable uptake of succinate, while cells grown without cyclic ANP are unable to transport it to any significant extent.

Clzaracteristics of Succinafe Transport-Fig. 2 shows the time dependence of succinate transport. The amount of succinate accumulated is linear for at least 3 min and a steady state is reached at about 20 min. Assuming that the volume of cell water is 2.7 11 per mg dry weight (43)) it can be readily calculated that after 10 min the concentration of succinate inside the cell is 7.71 mM when the outside concentration is 0.05 rnhf. Thus, suc- cinate is taken up against a concentration gradient which implies an active process.

Another argument for the uptake of succinate as being active comes from the study of temperature dependence of uptake and exit of succinate. Many acti-ve transport systems have high temperature coefficients, since energy metabolism is itself highly temperature dependent. As Fig. 5 shows at 0” only 0.4 nmoles of succinate is taken up by cells; at 15” the cells start taking up succinate actively and reach a maximum at around 36”. There is a 44.fold increase in uptake from O-36”. This large increase of uptake and the nonlinear pattern of increase preclude the fact that succinate transport is a passive process, which has a low temperature coefficient. The decrease in succinate accumulated in the cells (Fig. 5) at higher temperatures may be both due to an inactivation of the uptake system and an increase in the rate of exit. The figure also illustrates the fact that succinate exit from preloaded cells is temperature dependent.

Michcxlis Constant for Succinate and Competitive Inhibition by Dicarboxylic Acids-To determine the K, of succinate, initial rates of uptake were measured at different concentrations of suc- cinate. L-Malate and fumarate were found to be competitive inhibitors of succinate uptake. This is illustrated in Fig. 6. We have done competition experiments with wide concentration ranges of fumarate and malate to determine the inhibition con- stants but have not presented the results here. The K, for suc- cinate is 10 to 20 FM and the Ki for both malate and fumarate are in the range of 20 to 30 PM. The competition experiments sup- port the contention of Kay and Kornberg (7) that the dicar-

3 Morse and co-workers (12) have referred to this locus as ctr I (carbohydrate transport). We have followed here the termi- -.

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6328

boxylic acids are all taken up by the same transport system in E. coli.

Inhibition of Succinate Transport-Results obtained with the use of energy poisons show (Table III) that the respiratory chain and possibly oxidative phosphorylation are required for suc- cinate transport. Inhibition by azide, cyanide, arsenate, and dinitrophenol particularly show that metabolic energy is required for the uptake process. All the sulfhydryl reagents tested show some inhibitory effects on succinate uptake. This may indicate that sulfhydryl groups are involved in succinate transport.

It has been postulated by Chappell (44) that succinate enter- ing mitochondria does so in exchange for malate. This process is carried on through the functioning of a dicarboxylate trans- porter. This transporter is specifically inhibited by some suc- ciliate analogues, such as 2-phenylsuccinate (45), n-butyl- malonate and pentylmalonate. An attempt was made to see if there are any similarities in this regard between the mitochondrial and bacterial dicarboxylate transport systems. As seen in Table III, the mitochondrial inhibitors are without any significant ef- fect on the succinate uptake system of E. coli.

E$ect of Osmotic Shocking of Cells on Succinate Transport-It is well known that EDTA-treated E. coli cells when osmotically shocked show impairment of some transport systems (see review in Reference 33). This impairment has generally been considered to be due to the loss of specific binding proteins from the peri- plasmic space between the cell wall and cell membrane. It is

TABLE III

E$ect of various inhibitors on srtccinate uptake by strain CB 11 Cells were preincubated with inhibitor for 15 min at 23”, and

uptake experiments were then performed as described under “Experimental Procedure.” Cells were allowed to take up sodium [2,3-14C2] succinate for 10 min before filtration. All the inhibitors Tvere made up in aqueous solutions.

Inhibitors Concentration

Energy poisons Sodirlm azide

Potassium cyanide

Sodium arsenate

2,4-Dinitrophenol

Sulfhydryl reagents p-Chloromercuribenzoate

XEthylmaleimide

Iodoacetate

Showdomycin

2,2’-Dithiodipyridine

Competitive inhibitors of mito- chondrial dicarbosylate transporter

P-Phenyl succinate N-Butylmalonate Pentylmalonate

10-z M 42 10-Z M 52 10-a M 37 1OW nX 03 1o-3 M 65 10-Z M 74 10-S M 28 lo-’ M 48

10-d Y 33 1o-3 M 85 lo-’ M 69 lo- M 79 1o-3 M 38 10-z Y 67 1o-5 M 21 low4 n% 56 10-4 M 24 1OV nl 59

5 x 10-4 M 0 5 x 1o-4 M 10 5 x 10-” M 1

%

presumed t’hat these binding proteins are somehow involved in the transport process. In order to find whether a binding protein was involved in succinate transport, induced wild type and vari- ous mutant cells were osmotically shocked. As shown in Fig. 7 (illustrating one of several representative experiments) shocked cells invariably show significant decreases in succinate uptake. It may be mentioned that cells shocked by the procedures used here retain their viability to the extent of 95 to 100% when com- pared with equivalent suspensions of unshocked cells. This may indicate that failure to transport succinate by shocked cells is not due to a physical damage to the cell membrane caused by the shocking procedure.

We have attempted to reconstitute the transport system by adding back concentrated shock fluid to the shocked cells. These efforts have consistently failed. Recently, we have suc- ceeded in isolating a succinate-binding protein from the shock fluid. The purification and properties of this binding protein will be the subject of a later communication.

Genetic Elements Involved in Succinate Transport-Before mean- ingful attempts at chemical isolation and eventual reconstitution of succinate transport components are made we considered it worthwhile to get an indication of the number of components involved in this transport. This is best done by isolation of mutants defective in succinate transport. This approach has al- ready been used successfully, for example, in the study of sulfate (46) and histidine (47) transport. In order to obtain mutants in as many succinate transport components as possible, we have used three different methods of selection, which are discussed in detail under “Experimental Procedures.” The characteristics of these mutants (all independent isolates) including one (604 G) recovered by transduction from strain M?rI 6 (see Table I) are listed in Table IV. The mutants can be grouped into two classes phenotypically. All the fluoromalate-resistant mutants (CB 19, CB 20 and CB 21) and strain 604 G are uuable to grow 011 the dicarboxylic acids but grow on lactate (dct mutants). The rest of the mutants, obtained by tritium suicide and penicillin selec- tion method, do not grow on lactate or the dicarboxylic acids (ct mutants). Both of these two phenotypic classes of mutants grow well on acetate which shows that they do not lack any en- zymes involved in the metabolism of succinate (or other dicar- boxylic acids) and components of the respiratory chain. Each of

P -

OSMOTICALLY SHOCKED ”

0 4 8 12 16 20 TIME.(min)

FIG. 7. Effect of osmotic shock on uptake of succinate by strain CB 11. Log phase cells grown in LB plus lc/ succinate were washed and resuspended in 0.05 M phosphate buffer pH 7.5 (0.64 mg dry weight per ml). The suspension was divided into tu-o equal aliquots. One aliquot was osmotically shocked as described in the text. The shocked cells were resuspended in the original volume of buffer. Uptake experiments were then performed with both shocked and unshocked cells by standard methods.

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6329

TABLE IV

Some grow/h charac/erislics of th,e mutants of E. coli defective in

dicarbos!ylic acid transport with their probable /OCMS assignments on the chromosome

Del A refers to a gene loca.ted at approximately G9 min on the E. coli map (42). Dct B refers to a gene linked to gal I< (17 min on E. coli map). Dct A was approximately localized by inter- rupted matings. Dct H a.nd ct were localized by transduction experiments described in the text. + and - represent presence or absence of growth on solid media.

/

Strain j Mutant isolation method

CB 15 ~ Tritium (parent) suicide

CB 16 CB 17 CB 18 Hfr H Fl~loromalate

(parent) resistance CB 19 CB 20 CB 21 UC 14 Penicillin

(parent,) selection CB 22 ‘33 23

CB 24 K GO4 Transduction

(parent) from MM 6 GO4 G

‘c I

;rowth 01 wccinate, tumarate, or malate

+

- - - +

- - - +

- - - +

-

:rowth o lactate

+

-

-

-

+

+

+

+

+

-

-

-

+

+

:rowth 01 acetate

+

+

+

+

+

+

+

+

+

+

+

+

+

+

n

--

Genotype assigned

ctf

CP

Cl”

CP

dct A+

dct A dct A dct A ct+

Cl” Ct.” ct= dcl R’

dcl B

n Probable genotype.

the mut.ants listed in Table IV is a point mutation, as judged bJ the fact that each can be back mutated and such revertants ac- quire the ability simultaneously to grow on all three dicarboxylic acids and lact.ate. In addition, each can be transduced by phage P1 grown on wild type strain Hfr H and the transductants again are able to grow well on succinate, fumarate, malate, and lactate. When the mutants are grown on LB plus succinate medium (to induce the succinate transport system) and subsequently tested for their capacity for the transport of succinate, it is found that noue of them are capable of accumulating succinate to any sig-

nificant extent. Results of some representative experiments are illustrated in Fig. 8.

Assuming that one gene determines one separate component (regulatory or structural) of the transport system, we have tried to get an idea of the number of components involved by deter- mining whether the mutated locus in each of the succinate trans- port-deficient strains is the same or different. The approximate position of the mutated locus in fluoromalate resistant mutants was determined by first curing these strains of the F factor by sodium dodecyl sulfate method (22) and using these as females in interrupted matings with the multiauxotrophic Hfr strain UC 118 which donates its chromosome counterclockwise starting at about 62 to 64 min. Recombinants were selected on minimal media containing succinate as the sole carbon source (which also counterselected the Hfr due to the absence of added growth fac- tars). In all fluoromalate resistant mutants the succinate marker was transferred early (approximately 10 min). It is thus very likely that the genetic defect in these strains corresponds to the dcf gene (herein referred to as dct A) described by Kay and Kornberg (8).

0 4 8 12 16 20 TlME,(min,I

FIG. 8. Succinate uptake by various dicarboxylate transport negative mutants. The following strains were used: 1, CB 18; 2, CB 22; 3, GO4 G; 4, CB 20. Parental strains Hfr H and CB 15, and strain CB 11 are also shown as controls. Table IV gives details of the phenotypes of the various strains employed in this experi- ment.

The gene defects (ct) in strains obtained by penicillin enrich- ment (Cl3 22, CB 23, and CB 24) were mapped by transduction procedure described earlier for strain MM 6. In a typical trans- duction experiment using CB 22 (gal+, ct) as the donor and W 602 (gal, ct+) as the recipient, 858 gal+ transductants were picked up and the frequency of occurrence of succinate-negative phenotype in these was scored. A4bout 52% linkage with gal K was ob- tained. In a reciprocal transduction experiment where UC 118 (gal, CL+) was used as a donor alld CB 22 as the recipient, ct+ was used as a selected marker and gal as an unselected marker. Of the 420 ct+ colonies picked up, 191 colonies were gal K. So the linkage between ct of strain CB 22 and gal K is 45%. Strains CB 23 and CB 24 were not mapped as extensively as CB 22, but transduction results show a linkage of 40 to 50% with gal K. Cursory examination of mutants obtained by the tritium suicide technique suggests that they may be similar to CB 22 but this has to be confirmed by detailed mapping experiments.

The dct locus of strain MN 6, as was indicated earlier, seems to be also linked to gal K, but phenotypically it is entirely different from CB 22 (Table IV). We show in the accompanying com- munication (48) that MM B-G, as well as the fluoromalate re- sistant mutants, lack a membrane component required for the transport of succinate. The locus defective in MM 6 is desig- nated here as dct B.

DISCUSSION

It has become clear from recent studies of tricarboxylic (4-6) and dicarboxylic acid transport in various bacteria, including E. coli (7,8), that the transport systems are generally inducible and sensitive to catabolite repression by glucose (6, 37). In E. co&, thanks to the work of Perlman and Pastan (13, 38), it is now firmly established that inducible enzymes responsible for the catabolism of diverse sugars and other substrates require cyclic AMP as a part of positive control system for their induction. In the present work we have been able to establish that cyclic AMP is also involved in the induction of succinate transport system. This requirement for cyclic AMP possibly also explains the puzzling finding in the literature (9-12) that mutants lacking en-

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6330

zyme 1 of the phosphotrallsferase system (2) are unable to grow 011 succinate as the sole carbon source. For unknown reasoils, some enzyme I-defective mutants show enhanced catnbolite re- pressioil which cali be antagonized by cyclic XMP (13). In addition, we have shown that lack of succinate uptake (and toll- sequelit lack of growth 011 this carbon source) in strain AlAI 6, one of the earliest pleiotropic @s Z mutants described in the literature (91, is due to a second unli~lked mutation involved in succinate trallsl)ort (dct B). It is, thus, quite clear that the phosphotrans- fernse system is not primarily involved in any way in the suc- ciliate uptake process.

Lost transport systems described so far in bacteria (I), in- cluding the tricnrborylic acid transport in Bacillus (6), are in- volved in the concentrative uptake (active transport) of n-etabo- lites, i.e. transport ofcurd against :L concentration gradient. dome esceptions are, however, known. Glycerol, for instance, seems to be transported in E. coli by facilitated diffusion and cap- tured by ATP-dependent phosphorylation (49, 50). Group translocation systems for some sugars are also knon-n (3). Nevertheless, it is little surprising that Kay and Kornberg (7) in their study of dicarboxylic acid transport in E. coli found very little concentrative uptake of succinate in whole cells. Their studies were, however, performed with cells in which metabolism of succinate was not prevented by suitable genetic blocks. The studies presented here clearly Aow that the steady state concen- tration ratio of int,racellular to extracellular succinate can reach as high as 154 (Fig. 2). In confirmation of these results obtained with whole cells, we show in the accompanying paper (48) that membrane vesicles, devoid of enzyme systems metabolizing suc- ciliate, also accumulate succinate against a concentrat’ion gradi- ent. The inhibition of succinate uptake in whole cells by energy poisons and uncouplers of oxidatire phosphorylation (Table III) point to the conclusion that transport of succinate is an active process.

The kinetic parameters evaluated for succinate transport are Zi, 14 phi and V,,, (37”) approximately 20 pmoles X mill-l X g dry weight-‘. The latter value is smaller than is expected for most substrates which cali be utilized as single carbon sources by bacteria (51). Kay and Kornberg (7) have found an approxi- mately similar value for VX’,,,,, for all dicarboxylic acids trans- ported by their IS’. coli strains.

It seems probable from our results and those obtained by others (7, 8) that the succinate transport system is also utilized by cells for the uptake of fumarate and malate. The e\-idence rests primarily on the demonstration that fumarate and malate cause competitive inhibition of succinate and that mutants unable to accumulat’e succinate fail to transport or utilize for growth the other two dicarbosrlic acids. From competition experiments such as those presented in Fig. 6, the Zi, for fumarate and malate can be calculated to be about 20 to 30 PM, a value somewhat higher than that found for succinate.

Genetic evidence presented here suggests that in addition to a component which has been described by Kay and Kornberg (8) earlier, at least one other component is involved in dicarboxylic acid transport. The gene loci governing the synthesis of these components are referred to as dct A and dct B, the former map- ping at about 69 min (8) and the latter at around 17 min (near gal) 011 the E. coli map (42). Evidence obtained with mutants made by tritium suicide and other techniques ind’cates that a third component mapp ng near gal, and distinct from dct B, may also be involved in the uptake process.

The nature of these mutants, referred to here simply as ct, is puzzling because in addition to the lack of transport activity for

succinate these strains fail to grow on lactate, a monocarbosylic acid. It is entirely possible that lactate and the dicarboxylic acids share one common transport component (regulatory or structural), but this point remains to be inveatiguted. Trana- port systems described in the literature vary with regard to the number of components that are involved in oivo. Thus, phos- phate (52) and lactose (53) uptake systems seem to be essentially one component systems while sulfate (46) and histidine (47) transport are mediated by several components working in unison. It is still unclear how the different elements act in facilitating transport, but it is evident that progress in this area will depend upon a complete understanding in biochemical terms of the in- dividual components. Results presented in the accompanying communication (48) suggest that dct d and dct B loci are re- sponsible for the synthesis of membrane-bound elements involved .n dlcarboxy1.c acid uptake. The role of ct gene is not clear, but it may be that this locus is involved in specifying a carboxylate- binding protein. This statement, at the moment, is entirely speculative but is based on our finding that whole cells when sub- jected to osmotic shock lose the capacity for succinate uptake. In addition, we have partially purified2 from shock fluids a non- enzyme protein which binds succinate tightly.

REFERENCES

1. LIN, E. C. C. (1970) Annu. Rev. Genet. 4, 225 2. KUNDIG, W., GHOSH, S., AND KOSIDIAN, S. (1964) PTOC. Nat.

Acacl. Sci. U. S. A. 62, 1067 3. I~OSEMAN, S. (1969) J. Gen. Physiol. 64, 138s 4. VILL~~RREAL-MOGUEL, E. I., AXD ~ZUIZ-HERRER.I, J. (1969) J.

Bacleriol. 98, 552-558 5. WILKERSON, L. S., AND NAGON, R. G. (1971) Bacterial. Proc.

P216 6. WILLECKE, K., AND PARDEE, A. B. (1971) J. Biol. Chem. 246,

1032-1040 7. KAY, W. W., AXD KORKBERG, H. L. (1971) Eur. J. Biochem. 18,

274 8. KAY, W. W., AXD KORNBERG, H. L. (1969) Fed. Eur. Biochem.

Sot. Lett. 3, 93 9. FRAENCEL, L). G., FALCOZ-KELLY, F., AXD HORECKER, B. L.

(1964) Proc. Nat. Acad. Sci. U. S. A. 52, 1207 10. ASENSIO, C., AVIGAD, G., AND HORECKLR, B. L. (1963) Arch.

Uiochem. Biophys. 103, 299 11. TANAKA, S., FRAENKEL, D. G., ASD LIN, IX. C. (1967) Biochem.

Biophys. Res. Commun. 27, 63 12. WANG, k. J., AND MORSE, M. L. (1908) J. Mol. Biol. 32,59-K 13. PASTAN, I., AND PERLMAN, R. L. (1969) J. Biol. Chem. 244,

5836-5842 14. BERI~AN, M., ZWAIG, N., AND LIN, E. C. C. (1970) Biochem.

Biophys. Res. Commun. 38, 272 15. DEMEREC, M., ADELBERG, E. A., CLARK, A. J., AND H.~RTNAN,

P. ti. (1966) Genetics 64, 61 16. I>av~s, B. D., AND MINGIOLI, E. S. (1950) J. Bacterial. 60, 17 17. LENNOX, E. S. (1955) Vi”iro2ogy 1, 190 18. CURTIS& K., III, CARO, L. G:,ALLISON, D. P., .~SD STALLIOXS,

D. R. (1969) J. Bacterial. 100, 1091-1104 19. Low, B.,‘AND’~ooD, T. H. (1985) Genet. Res. 6, 300 20. ADELBERG, E. A., MANDEL, M., AND &EN, G. C. C. (1965)

Biochem. Biophys. Res. Commun. 18, 788 21. GORINI, L., AND KAUFMBN, H. (1960) Science 131, 604 22. TOMOEDA, M., INUZUKA, &I., KUBO, N., AND N.4KAMUR.4, S.

(1968) J. Bacterial. 95, 1078-1089 23. ZWAIG, N., AND LIN, E:. C. C. (1966) Biochem. Biophys. Res.

Commun. 22, 414 24. WILSON, T. H., AND KASHKET, IZ. 12. (1969) Biochim. Biophys.

Acta 173, 501-50s 25. KING, T. E:. (1963) J. Biol. Chem. 238, 4032-4036 26. GIBSON, J., UPPER, C. I>., AXD GUNSALUS, I. C. (1967) J. Biol.

Chenz. 242, 2474-2477 27. COLOWICK, S. P., ~XD KAPLAN, N. 0. (1963) Methods Enzymol.

1, 729 28. WRIGHT, J. A., AND SANWAL, B. D. (1969, 1971) J. Biol. Chem.

244, 1833-1845; 246, 1689-1699

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6331

29. FRAENKEL, D. G., AND HORECKER, B. L. (1965) J. Bacterial. 90, 837

30. Hnunnn, J., AND SINGER, J. P. (1967) Eur. J. Biochem. 3, 107 31. ERAY, G. A. (1960) Anal. Biochem. 1, 279-285 32. KOCH, A. L. (1964) Biochim. Biophys. Acta 79, 177 33. HEPPEL. L. A. (1969) J. Gen. Phusiol. 64, 95s 34. HIISSZH,‘~. A., ~AS~INSKY, M., ~)AVIS, B. D., AND LIN, E. C.

C. (1963) J. Biol. Chem. 238, 3770-3774 35. HSIE, A. W., AND RICKENBERG, H. V. (1966) Biochem. Biophys.

R&s. Commun. 26, 676 36. SANWAL, B. D. (1970) Bacterial. Rev. 34, 20 37. M~GASANIK, B. (1961) Cold Spring Harbor Symp. Quant. Biol.

26, 249 38. PASTAN, I., AND PERLMGN, R. (1970) Science 169, 339 39. MAKMAN. R. S.. AND SUTHERLAND. E. W. (1965) J. Biol. Chem.

240, 1369-1314 ,

40. PETERKOFSKY, A., AND GAZDAR, C. (1971) Proc. Nat. Acad. Sci. U. S. A. 68, 2794

41. GACHELIN, G. (1969) Biochem. Biophys. Res. Commun. 34, 382 42. TAYLOR, A. L. (1970) Bacterial. Rev. 34, 155

43. WINKLER, H. H., AND WILSON, T. H. (1966) J. Biol. Chem. 241, 2200-2211

44. CHAPPELL, J. B. (1969) in Inhibitor Tools in Cell Research (BUTCHER, T., AND SIFX, H., eds) p. 335, Springer, New York

45. CHAVIN, S. I. (1971) Fed. Eur. Biochem. Sot. Lett. 14, 269 46. PARDEE, A. B. (1968) Science 162, 632-637 47. AMES, G. F., AND LEVER, J. (1970) Proc. Nat. Acad. Sci. U. S.

A. 66, 1096 48. RAYMAN, M. K., Lo, T. C. Y., AND SANWAI,, B. D. (1972) J.

Biol. Chem. 247, 6332-6339 49. HAYASHI, S., AND LIN, E. C. C. (1965) Biochim. Biophys. Acta

94, 479-487 50. COZARELLI, N. R., FREEDBERG, W. B., AND LIN, E. C. C.

(1968) J. Mol. Biol. 31, 371-387 51. KEPES, A. (1964) in The Cellular Function of Membrane Trans-

port (HOFFMAN, J. F., ed) p. 155, Prentice Hall, Inc., Englewood Cliffs, New Jersey

52. MEDVECZKY, N., AND ROSENBERG, H. (1970) Biochim. Biophys. Acta 211, 158-168

53. KENNEDY, E. P. (1969) J. Gen. Physiol. 64, 91s

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Theodore C. Y. Lo, M. Khalil Rayman and Bishnu D. SanwalSTUDIES OF TRANSPORT IN WHOLE CELLS

: I. BIOCHEMICAL AND GENETICEscherichia coliTransport of Succinate in

1972, 247:6323-6331.J. Biol. Chem. 

  http://www.jbc.org/content/247/19/6323Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/247/19/6323.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 20, 2018

http://ww

w.jbc.org/

Dow

nloaded from