Vol. 169, No. 9JOURNAL OF BACTERIOLOGY, Sept. 1987, p. 4361-43670021-9193/87/094361-07$02.00/0Copyright © 1987, American Society for Microbiology

Regulation of Fructose Uptake and Catabolism by Succinate inAzospirillum brasilense

AMIT MUKHERJEE AND SUDHAMOY GHOSH*Departmnent of Biochemistry, Bose Institute, Calcutta-700054, India

Received 8 December 1986/Accepted 1 June 1987

Fructose uptake and catabolism in AzospiriUum brasilense is dependent on three fructose-inducible enzymes

(fru-enzymes): (i) enzyme I and (ii) enzyme II of the phosphoenolpyruvate:fructose phosphotransferase systemand (iii) 1-phosphofructokinase. In minimal medium containing 3.7 mM succinate and 22 mM fructose as

sources of carbon, growth ofA. brasilense was diauxic, succinate being utilized in the first phase of growth andfructose in the second phase with a lag period between the two growth phases. None of the fru-enzymes couldbe detected in cells grown with succinate as the sole source of carbon, but they were detectable toward the endof the first phase of diauxie. All the fru-enzymes were coinduced by fructose and coordinately repressed bysuccinate. Studies on the effect of succinate on differential rates of syntheses of the fru-enzymes revealed thattheir induced syntheses in fructose minimal medium were subject to transient as well as permanent (catabolite)repression by succinate. Succinate also caused a similar pattern of transient and permanent repression of thefructose transport system in A. brasilense. However, no inducer (fructose) exclusionlike effect was observed as

there was no inhibition of fructose uptake in the presence of succinate with fructose-grown cells even when theywere fully induced for succinate uptake activity.

Azospirillum brasilense is a gram-negative, chemo-hetero-tropic, aerobic soil bacteriun that has the ability to fixatmospheric nitrogen under microaerobic conditions (30).The bacterium has a tendency to grow in association withroots of graminaceous plants, including some important cropplants such as maize, wheat, and rice, and increases plantproductivity by associative symbiosis (4, 33). During thesymbiosis, the bacteria presumably obtain their supply ofcarbon sources from the plants for growth and nitrogenfixation. These observations have led several groups toinitiate a systematic investigation of carbon metabolismn in A.brasilense (3, 6, 10, 15, 21, 34). This bacterium grows well ontricarboxylic acid cycle intermediates such as succinate ormalate as a carbon source (30). It also has the ability toutilize fructose (2, 3, 6, 15, 30), galactose (15), gluconate (15,34), and L-arabinose (21) but not glucose (30). Of thesesugars, the fructose transport system has been studied insome detail (3, 6). Fructose catabolism in A. brasilense isinitiated by a fructose-inducible phosphoenolpyruvate:fructose phosphotransferase system (fru-PTS) that mediatesthe concomitant transport and phosphorylation of the sugar(3, 6). It has also been shown that the fru-PTS of A.brasilense is composed of a soluble component (enzyme I)and a rmembrane-bound component (enzyme II), both ofwhich are necessary for the enzymatic phosphorylation offructose (3). Fructose-i-phosphate, the product of fru-PTSreactions, enters the glycolytic pathway for further catabo-lism after its conversion to fructose-1,6-bisphosphate by1-phosphofructokinase (1-PFK) (6, 15). All three enzymes,namely, enzyme I, enzyme II, and 1-PFK (collectivelyreferred to here as fru-enzymes) are induced by fructose inA. brasilense. Despite such extensive work on the carbonmetabolism of A. brasilense, there is almost a total lack ofinformation on the regulation of carbon source utilization inthis bacterium. As a first step, we focused our attention onthe regulation of fructose uptake and catabolism.

* Corresponding author.

An extremely important mode of carbon source utilizationin bacteria is manifested by the phenomenon of diauxie, firstobserved by Monod (18). Diauxic growth is shown by abacterium when a primary or preferred carbon source causessuppression of the utilization of the secondary or lesspreferred carbon source while both are present in the growthmedium. The classical example is the glucose-lactosediauxie in Escherichia coli, in which glucose is the preferredcarbon source and lactose the less preferred one. Unlikeenteric bacteria such as E. coli, aerobic bacteria generallyprefer organic acids (tricarboxylic acid cycle intermediates,acetate, etc.) to sugars. Organic acid-sugar diauxie has beenobserved in aerobic bacteria such as Rhizobium meliloti (32),Arthrobacter crystallopoietes (8), Pseudomonas spp. (1, 12,25, 31), and Azotobacter vinelandii (5). Intensive studies onglucose-lactose diauxie in E. coli by a number of investiga-tors have demonstrated that three mechanisms underlie thisphenomenon: transient repression, permanent (catabolite)repression, and inducer exclusion (13). In transient andpermanent repression, glucose affects the syntheses of theenzymes coded by the lac operon, whereas in inducerexclusion, the uptake activity of lactose (inducer) is inhibitedby glucose. However, there is a paucity of such criticalstudies in depth on diauxie in aerobic bacteria.

In this communication we show succinate-fructose diauxiein A. brasilense and present evidence of coinduction of thefru-enzymes by fructose. We also report here that in thisbacterium, succinate causes both transient and permanentrepression of induced syntheses of the fru-enzymes but failsto exert an inducer (fructose) exclusionlike effect.

MATERIALS AND METHODSBacterial strain and growth medium. The strain used in this

study was A. brasilense RG; it was selected earlier in ourlaboratory from a culture of A. brasilense 81 (sent to us byN. R. Krieg) on the basis of its streptomycin resistance andgood colony formation ofl a nitrogen-free minimal agar plate(16). For growth of the strain, we used minimal medium asdescribed by Okon et al. (22) containing 0.1% NH4Cl as a

4361

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

4362 MUKHERJEE AND GHOSH

nitrogen source and either 1% sodium succinate 6H20 (37mM) or 0.4% fructose (22 mM) as a carbon source, unlessotherwise mentioned. The strain was maintained on platescontaining similar minimal agar medium with succinate, butwith the addition of 0.003% yeast extract and the omission ofNH4Cl. The bacteria were grown in liquid medium withgyratory shaking at 32°C, and growth was monitored bymeasuring the optical density at 590 nm (OD590) in a Beck-man DU-6 spectrophotometer. A 1.0-ml culture with an

OD590 of 1.0 is equivalent to 3 x 108 cells or 470 ,ug (dryweight) or 230 ,ug of cell protein.

Preparation of inoculum. Cells were inoculated from platesinto fructose minimal medium (FMM) or succinate minimalmedium (SMM) containing 0.05% yeast extract, grown up toan OD59 of 0.3 to 0.6, inoculated into the same mediumwithout yeast extract, and grown overnight up to an OD5sg of1.5 to 2.5 (within the exponential phase). The cell culturewas centrifuged at room temperature, suspended in carbon-free minimal salt medium, and used as an inoculum.

Collection of cells for preparation of crude extracts. For thepreparation of crude enzyme I and enzyme II (3) of fru-PTSof A. brasilense RG, cells were grown in FMM up to an

OD590 of 1.2, chilled, harvested in a Sorvall GS-3 rotor bycentrifugation at 4°C, and washed twice with Tris hydrochlo-ride buffer (10 mM, pH 7.6). The cell pellets were stored at-20°C. In experimental samples in which the levels offru-enzymes were measured, the volume of cells harvestedwas such that the total OD590 was between 100 and 120.

Preparation and fractionation of crude extracts. All opera-tions were done between 0 and 4°C. Frozen cells werethawed and suspended in 5 ml of buffer A (10 mM potassiumphosphate [pH 7.6], 14 mM 2-mercaptoethanol, 1 mMEDTA). The cell suspension was sonicated twice for 15 seach time at 30 to 50 W in a Braunsonic model 1510 sonicatorwith an interval of 1 min to allow cooling. The sonicatedextracts were spun at 13,000 x g in a Sorvall SS-34 rotor for20 min, and the supernatant obtained was termed the crudeextract.For the preparation of enzyme I and enzyme II, crude

extracts prepared from 2 g (wet weight) of cells as describedabove were made up to 15 ml with buffer A and centrifugedin a Beckman 42.1 rotor at 120,000 x g for 2 h to separate themembrane fraction from the soluble fraction. Since themembrane fraction did not form a tight pellet, about 11 to 12ml of the supernatant was carefully withdrawn with a Pas-teur pipette and the remaining liquid was rejected to avoidcontamination of the supernatant with the membrane frac-tion. For the preparation of crude enzyme I, the clearsupernatant, termed soluble extract, was treated with am-monium sulfate to 80% saturation by gentle stirring, and theprecipitate formed was collected by centrifugation at 13,000x g for 20 min. The precipitate was dissolved in a minimumvolume of buffer A and dialyzed against the same buffer withrapid stirring for 2 h with two changes of the buffer. Thedialyzed sample (crude enzyme I) had a protein concentra-tion of 18 to 20 mg/ml. The membrane pellet obtained afterultracentrifugation was washed twice with 50 ml of buffer Aafter suspension with a hand-operated glass-Teflon homog-enizer. Each washing was followed by centrifugation at120,000 x g for 90 min. After the first wash, and onwards,the membrane formed tight pellets. The membrane pelletwas finally suspended in buffer A to a protein concentrationof 18 to 20 mg/ml (crude enzyme II). Fresh enzyme I andenzyme II were prepared for each set of experiments.Enzyme assays. (i) Enzyme I. The activity of enzyme I was

measured in crude extracts by the rate of phosphorylation of

['4C]fructose in the presence of excess enzyme II. Thereaction mixture contained (in a final volume of 20 ,ul) 50 mMpotassium phosphate buffer (pH 7.6), 5 mM MgCl2, 5 mMphosphoenolpyruvate, 2 mM ['4C]fructose (5,000 cpm/nmol,as measured on Whatman 3MM paper), an excess of enzymeII (80 to 100 ,ug of protein), and a rate-limiting amount ofenzyme I. The reaction mixture was incubated at 37°C for 30min, and the reaction was terminated by heating in a boilingwater bath for 30 s. The reaction mixture was then centri-fuged at 2,000 x g for 10 min, and 10 RI of the supernatantwas spotted on Whatman 3MM paper soaked in 0.1 Msodium acetate (pH 7.0) (electrophoresis buffer) along with 2,ul of a mixture of cold fructose and fructose-i-phosphate(0.1 M each), which served as the known markers. Afterelectrophoresis at 300 V for 90 min to separate fructose fromfructose-i-phosphate, the paper was heated in an oven at150°C until the well-separated yellow spots of fructose andfructose-i-phosphate appeared. The fructose-i-phosphateregion was cut out, and its radioactivity was counted in aBeckman LS-1800 scintillation counter with 5 ml of scintil-lation fluid (4 g of Omnifluor per liter of toluene). Under theabove assay conditions, the amount of radioactivity thatdisappeared from the fructose spot on enzyme incubationwas equal to the radioactivity that appeared in the fructose-1-phosphate spot, within an error of +5%.

(ii) Enzyme II. Enzyme II activity was measured in thecrude extracts exactly as described for the enzyme I assay,except that the reaction mixture contained an excess ofenzyme 1 (80 to 100 ,ug of protein) and a rate-limiting amountof enzyme II.

(iii) 1-PFK. 1-PFK (EC 2.7.1.56) was assayed in crudeextracts by measuring the rate of transfer of 32P-phosphorylgroup from [_y-32P]ATP to fructose-i-phosphate resulting inthe formation of fructose-1,6-[6-32P]bisphosphate. The reac-tion mixture in a final volume of 20 pu1 contained 50 mM Trishydrochloride buffer (pH 7.6), 10mM MgC92, 1 mM fructose-1-phosphate, 2 mM [-y-32P]ATP (30,000 to 50,000 cpm/nmol,as measured after spotting on Whatman 3MM paper), and arate-limiting amount of 1-PFK. The reaction mixture wasincubated at 30°C for 12 min, and the reaction was termi-nated by heating the mixture in a boiling water bath for 30 s.The reaction mixture was centrifuged at 2,000 x g for 5 min,and 10 pu1 of the supernatant was spotted onto Whatman3MM paper along with a 2-pI solution of cold 0.1 Mfructose-1,6-bisphosphate, which served as the knownmarker. To separate fructose-1,6-[6-32P]bisphosphate from[y-32PIATP, we subjected the paper to descending chroma-tography at room temperature for about 20 h using isobutyricacid-ammonia-water (66:1:33, vol/vol/vol) as the solvent(solvent A). The paper was then air dried and heated at150°C in an oven until the yellow spot of fructose-1,6-bisphosphate appeared. The fructose-1,6-bisphosphate re-gion was cut out, and its radioactivity was counted asdescribed above. In this solvent system, ATP migratedfaster than fructose-1,6-bisphosphate.

Definition of enzyme units. One unit ofenzyme I or enzymeII was defined as the amount of enzyme that converted 1nmol of fructose to fructose-i-phosphate per min at 37°C.One unit of 1-PFK was defined as the amount of enzyme thatproduced 1 nmol of fructose-1,6-bisphosphate from fructose-1-phosphate per min at 30°C.Uptake studies. Cell cultures were centrifuged, washed

once with 50 mM sodiumn phosphate buffer (pH 7.0) (uptakebuffer), and finally resuspended in the same buffer such thatthe OD5* was between 0.4 and 0.6. All the above operationswere done at 30°C. Uptake of fructose or succinate was

J. BACTERIOL.

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

REGULATION OF FRUCTOSE UPTAKE AND CATABOLISM 4363

C)°0>,0.5-d S

0.1:

0.05

0.030 2 4 6 8 1012 1416 1B2022

Time (hours)FIG. 1. Diauxic growth of A. brasilense RG. Succinate-grown

inoculum was added to minimal medium containing 3.7 mM suc-

cinate and 22 mM fructose (O), 1.85 mM succinate and 22 mMfructose (0), or 3.7 mM succinate (A).

initiated by adding [14C]fructose (7,000 cpm/nmol) or

[14C]succinic acid (13,000 cpm/nmol) to a final concentrationof 50 or 300 RM, respectively, and incubating the mixture at300C with shaking. At defined time intervals, 50 RI of the cellsuspension was rapidly filtered through filters (type HA,0.45-pum pore size; Millipore Corp., Bedford, Mass.) andwashed twice with 5 ml of uptake buffer, and the filters were

dried and their radioactivity was counted. Under theabovementioned conditions, the uptake rate of succinate andfructose was constant up to at least 8 min.Enzyme I and 1-PFK were assayed for any particular set

of experiments within 3 h, and enzyme II was assayed within6 h of preparation of the crude extracts. This was necessarysince the half-lives of enzyme I, 1-PFK, and enzyme II were8, 8, and 15 h, respectively. Despite maintaining the standardconditions of assay, we observed that the specific activitiesof enzyme I, enzyme II, and 1-PFK showed a variation of±30% from their mean values of 71, 17, and 185 U/mg ofprotein, respectively, from batch to batch of cultures whichwere fully induced for the fru-enzymes. However, the ratioof the specific activities of the three enzymes in a particularculture at any point of time varied within ±10%. Similarvariations were also noted from batch to batch of cultureswith respect to succinate and fructose uptake activities.

Protein estimation. Protein was estimated by the method ofLowry et al. (11), using bovine serum albumin as thestandard.

Chemicals. Bovine serum albumin, 2-mercaptoethanol,Tris, phosphoenolpyruvate (cyclohexyl ammonium salt),and the sodium salts of fructose-i-phosphate, fructose-1,6-bisphosphate, and ATP were purchased from Sigma Chem-ical Co. (St. Louis, Mo.). Omnifluor was purchased fromNew England Nuclear Corp. (Boston, Mass.). All otherreagents were of analytical grade and were purchased fromlocal suppliers. Tris and ammonium sulfate were recrystal-lized before use. D-[U-14C]fructose, [1,4-14C]succinic acid,and [y-32P]ATP were purchased from Bhabha Atomic Re-search Centre (Bombay, India). [-y-32P]ATP was routinelypurified by chromatography on Whatman 3MM paper em-ploying solvent A to remove impurities that interfered with

the assay. After chromatography, the paper was first dried atroom temperature and then at 40°C under vacuum for 1 h toremove completely the last traces of the solvent. The radio-activity in the ATP region was eluted with Tris hydrochlo-ride (20 mM, pH 7.6), and a cold ATP solution was added tothe eluate such that the final ATP concentration was 4 mM.

RESULTS

Diauxic growth of A. brasilense on succinate and fructose.A. brasilense can utilize both succinate and fructose as thesole source of carbon for growth with generation times of 110+ 10 min and 190 ± 10 min, respectively, at 32°C. When asuccinate-grown inoculum was added to a minimal mediumcontaining 3.7 mM succinate and 22 mM fructose as sourcesof carbon, the growth was diauxic with a doubling time of110 min in the first phase and 200 min in the second phaseand with a lag period of 30 to 40 min between the two phasesof growth (Fig. 1). When the concentration of succinate washalved to 1.85 mM without altering the fructose concentra-tion (22 mM), growth was still diauxic, but the net increasein OD590 in the first phase was reduced by approximately half(from 0.66 to 0.35) (Fig. 1). Further, when the mediumcontained only 3.7 mM succinate, growth ceased at thebeginning of the lag period without showing any secondphase of growth (Fig. 1). To confirm the sequence ofutilization of carbon sources during diauxie, we added asuccinate-grown inoculum to a minimal medium containing3.7 mM succinate and 22 mM fructose which was labeledwith either ['4C]fructose or [14C]succinate. Utilization of thelabeled carbon sources was monitored both by disappear-ance of radioactivity from the medium and by its incorpora-tion into the cells (Fig. 2). It was observed that in the diauxicgrowth ['4C]succinate was utilized first (as judged by itsdisappearance from the supernatant) and that upon its ex-haustion the cells entered the lag phase. Only about 20% ofthe radiolabeled succinate that disappeared from the mediumwas incorporated into the cells in the first growth phase (Fig.2); the remaining 80% of the radioactivity was presumablylost by the cells as 1'CO2 (10). Radiolabeled fructose wasutilized when growth was resumed in the second phase ofdiauxie (Fig. 2). However, a small amount of fructose wasalso incorporated into the cells in the first phase of growth,although its disappearance from the supernatant could not bedetected owing to the low specific activity of ["'Cifructose inthe medium. These experiments clearly established that inthe diauxic growth succinate was utilized in the first phaseand that upon its exhaustion fructose was utilized in thesecond phase.

Kinetics of induction of fru-enzymes. From the pattern offructose utilization in Fig. 2, it appeared that in the lagperiod, after the exhaustion of succinate and after the firstphase of diauxic growth, the cells synthesized the fru-enzymes to enable them to grow on fructose in the secondphase. That this was indeed the case was confirmed bymonitoring the specific activities of the fru-enzymes duringthe diauxic growth of A. brasilense (Fig. 3A). To observewhether the pattern of induced syntheses of the fru-enzymesas shown in Fig. 3A was manifested under a differentcondition of induction, exponentially growing cells in SMMwere shifted to FMM (the lag phase preceding exponentialgrowth was 90 min in this case) and induction of thefru-enzymes was monitored (Fig. 3B). The pattern of syn-theses of the fru-enzymes under this condition (Fig. 3B) waspractically similar to that shown in Fig. 3A. Both theseexperiments showed that enzyme I and 1-PFK were coordi-

VOL. 169, 1987

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

4364 MUKHERJEE AND GHOSH

Time (hours)FIG. 2. Utilization of succinate and fructose by A. brasilense RG

during diauxic growth. Succinate-grown inoculum was added to aminimal medium containing 3.7 mM succinate and 22 mM fructoseand divided into three flasks. To one flask ['4C]succinate was addedand to another flask ['4C]fructose was added such that their finalspecific activities were 78,000 and 16,000 cpm/,umol, respectively.The third flask was used to monitor growth (0). During growth['4C]fructose and [14C]succinate incorporated into cells and thatremaining in the supernatants were measured periodically as fol-lows. For cellular incorporation, cells in 200-RIl cultures werecollected on Millipore filters and washed with uptake buffer, and thefilters were then dried and their radioactivity was counted. Forincorporation in supematants, 200-,ul cultures were centrifuged at4°C, and 50 ,u1 of the supernatant was spotted on Whatman 3MMdisks which were then dried and counted. Symbols: A, [14C]suc-cinate in supernatant; A, [14C]succinate in cells; E, [14C]fructose insupernatant; O, ['4C]fructose in cells.

nately induced but that their induction lagged temporallybehind that of enzyme II. It should be noted that fru-enzymeactivities were not detectable at all in cells grown in SMM(the 0-h values in Fig. 3B). This is unlike the situation at thebeginning of the lag phase of diauxic growth when enzyme I,enzyme II, and 1-PFK activities showed 25, 50, and 22% oftheir maximum levels, respectively (Fig. 3A), which isprobably due to the small amount of fructose that entered thecell in the first phase (Fig. 2).

Succinate-mediated repression of fru-enzymes. Inducedsyntheses of the fru-enzymes and commencement of fruc-tose utilization after exhaustion of succinate in diauxicgrowth (Fig. 2 and 3A) suggested that succinate in some wayinterfered with the syntheses of these inducible enzymes inA. brasilense. To test how succinate affected the inducedsyntheses of the fru-enzymes, we measured the effect ofsuccinate on the differential rates of syntheses of enzyme I,enzyme II, and 1-PFK (Fig. 4). The addition of 3.7 mMsuccinate to cultures growing exponentially in FMM causedtransient repression of enzyme I, enzyme II, and 1-PFKsyntheses. The transient repression generally lasted slightlymore than one generation, after which syntheses of theseenzymes were resumed; however, permanent repressioncaused by succinate reduced the rate of syntheses of theseenzymes by approximately half at this stage compared withthose shown by the fully induced cells. In a similar experi-ment when a higher concentration of succinate (18.5 mM)was added, the pattern of repression of the fru-enzymes wasthe same as with 3.7 mM succinate (data not shown, except

I 10:CD ^ .rO-W045 1o0In

ao3 0 en0 2 4 6 8-1012 02 4 6 8

Time (hours)FIG. 3. Kinetics of induction of fru-enzymes in A. brasilense

RG. Cells were grown (0) in minimal medium containing 3.7 mMsuccinate and 22 mM fructose (A) or 22 mM fructose (B), using asuccinate-grown inoculum. Arrows indicate the points at which cellswere harvested and enzyme I (0), enzyme II (El), and 1-PFK (A)activities were assayed in their crude extracts. The 100% specificactivities of enzyme I, enzyme II, and 1-PFK in panel A were 92, 20,and 225 U/mg of protein, and in panel B they were 72, 19, and 175U/mg of protein, respectively. Activities of the enzyme at 0 h inpanel B were assayed from succinate-grown cells before shifting toFMM.

for enzyme II in Fig. 4D) and the transient repression lastedfor about one generation. (Any small difference noted in theduration of transient repression between the two experi-ments was attributed to experimental variation.) It is quiteclear that enzyme I, enzyme II, and 1-PFK were coordi-

L-ot

en

40 A enzyme I B e zymeIj 1630-Succinjte Succincate -12

20 (37 M) (3.7mM)- 4

10 -- 40 - I

.le, 120cm

° 8040

r.

C 1-PFKSuccinate /(3.7mM)

/I I.

0 0.2 0.4 0.6

D enzymeSuccinate(185mM) *M

12

84

0.8 0 0.2 0.4 0.6 0.8 1.0

a-

O)

Nccqo

OD 59oFIG. 4. Effect of succinate on the differential rates of syntheses

of fru-enzymes in A. brasilense RG. To a culture of cells growingexponentially in FMM, 3.7 mM succinate (final concentration) wasadded to a part of the culture at the point indicated by the arrows (A,B, and C). Cells were harvested at various ODs, and enzyme I,enzyme II, and 1-PFK activities were assayed in their crudeextracts. Panel D shows the result of a different but similar experi-ment in which the enzyme II activities were measured in cultures towhich 18.5 mM succinate was added instead of 3.7 mM succinate.Activities of enzymes measured from cultures growing in FMMwithout (control) and with the addition of succinate are denoted byopen and closed circles, respectively.

J. BACTERIOL.

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

REGULATION OF FRUCTOSE UPTAKE AND CATABOLISM 4365

0.,

E 1.2

3 E Succinuta_ O0.8 - (307MM)

U. U)0

0 0.2 0.4 0.6 0.8

0.D.590FIG. 5. Effect of succinate on the differential rate of synthesis of

the fructose transport system in A. brasilense RG. A culture of cellsgrowing exponentially in FMM was split at the point shown by thearrow, and to one part 3.7 mM succinate (final concentration) wasadded. Cells (1 ml) were harvested at various ODs, washed, andsuspended in the same volume of uptake buffer, and fructose uptakeactivities were measured for 5 min (uptake rate was linear in thisperiod). Fructose uptake activities measured from cultures growingin FMM without (control) and with the addition of succinate aredenoted by open and closed triangles, respectively.

nately repressed by succinate in A. brasilense cells growingwith fructose as the carbon source (Fig. 4A, B, and C). Itought to be emphasized here that the succinate-mediatedtransient and permanent repressions of the fru-enzymes inA. brasilense were not due to any reduction of total cellularprotein synthesis since the addition of succinate to culturesgrowing on FMM actually enhances the growth rate of thecells. The doubling time of cells growing in FMM wasdecreased to 110 min upon the addition of succinate (see Fig.6), which was characteristic of growth of A. brasilense inSMM.To determine how the pattern of repression of enzyme I

(Fig. 4A) and enzyme II (Fig. 4B) was reflected at the levelof uptake activity of whole cells, we also measured the effectof succinate on the differential rate of synthesis of thefructose transport system (Fig. 5). The addition of 3.7 mMsuccinate caused transient repression followed by perma-nent repression of synthesis of the fructose transport sys-tem. The transient repression persisted for one generation,and permanent repression decreased the differential rate ofsynthesis of the transport system to about 50% of that in thefully induced fructose culture. The pattern of repression ofenzyme I and enzyme II therefore correlated well with thepattern of repression of the fructose transport system bysuccinate in whole cells of A. brasilense.

Inability of succinate to cause an inducer exclusionlikeeffect. Although our experiments demonstrated that suc-cinate mediates transient and permanent repression of theinduced syntheses of enzyme I and enzyme II (Fig. 4A andB) as well as of the fructose transport system (Fig. 5), noconclusion could be drawn from the experiments as towhether succinate also mediated inducer (fructose) exclu-sion. To prove inducer exclusion, it was essential to dem-onstrate inhibition of fructose uptake in the presence ofsuccinate while succinate uptake occurred at the same time.Hence, it was necessary to know whether the succinatetransport system was fully induced or not under such con-

ditions. We observed a sixfold increase in the succinateuptake rate in cells growing in FMM within 40 min of theaddition of succinate, with no further increase thereafter(Fig. 6, inset). Therefore, to test succinate-mediated inducerexclusion in A. brasilense, cells growing exponentially inFMM (having the basal level of succinate uptake activity)were collected at an OD59 of 0.18 and the fructose uptakerates were measured (for 6 min) in the absence and presenceof 5 mM succinate. The rates were 17.2 and 19.6 nmol min-mg of protein-1, respectively. To the same culture of cells,3.7 mM succinate was added at an OD59 of 0.18, and cellswere collected after 80 min to allow full induction of thesuccinate transport system. The fructose uptake rates in theabsence and presence of 5 mM succinate in this case were11.1 and 11.5 nmol min-' mg of protein-', respectively.Therefore, 5 mM succinate did not inhibit the fructoseuptake activity in either type of cell. The lower values offructose uptake activity in the latter case were due to the factthat the fructose transport system was in a state of transient

4 6 8 10 12Time(hours)

FIG. 6. Effect of succinate on growth and uptake of succinate inA. brasilense growing in FMM. A culture of cells growing exponen-tially in FMM (0) was split at the point shown by the arrow (0 min),and to one part 3.7 mM succinate (final concentration) was added,and the cells were grown further (0). (Inset) Cells were collected at0 min (A) and at 40 min (O) and 80 min (0) after the addition ofsuccinate (as shown by the arrows), and succinate uptake rates weremeasured.

VOL. 169, 1987

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

4366 MUKHERJEE AND GHOSH

repression when the exponentially growing cells were col-lected (Fig. 5). It is perhaps relevant to mention here that incells fully induced for succinate and fructose transport, thesubstrate concentrations necessary to attain maximum ratesof succinate and fructose uptake were observed to be 300and 25 ,uM, respectively. When inducer exclusion was testedunder conditions as described above, except that a sub-saturating concentration of [14C]fructose (10 ,uM) was used,5 mM succinate did not inhibit fructose uptake (data notshown).

DISCUSSION

In this study, an attempt was made to delineate some ofthe biochemical processes involved in succinate-fructosediauxie in A. brasilense. The use of radiolabeled carbonsources in the minimal growth medium demonstrated thatsuccinate was utilized in the first phase and that fructose wasutilized in the second phase of growth. Since we found thatgrowth on 4 mM L-malate and 22 mM fructose was alsodiauxic (unpublished observation), it seems that in A.brasilense tricarboxylic acid cycle intermediates are pre-ferred sources of carbon over fructose. Preference for suc-cinate or malate over carbohydrates, therefore, seems to bea general feature of aerobic bacteria (8, 9, 32). The charac-teristic feature of diauxic growth is that the utilization of thefirst carbon source is by a pathway that is constitutive, whilethat of the second carbon source is by an inducible pathway(18). The kinetics of induction of the fru-enzymes indeedshowed that the enzymes were undetectable in the beginningof the first phase; their syntheses began late in the first phaseand attained maximal levels only during the second phase.On the other hand, the enzymes of the tricarboxylic acidcycle of A. brasilense which are essential for succinatemetabolism are known to be constitutive (6, 15). In addition,two of our observations also support the hypothesis thatsuccinate utilization is constitutive in A. brasilense: (i)immediate stimulation of growth by addition of succinate toa culture growing exponentially in FMM, and (ii) resumptionof exponential growth with a generation time of 110 minwithout any discernible lag upon shifting exponentiallygrowing cells from fructose to SMM (unpublished observa-tion). Based on the latter observation, succinate utilizationin Pseudomonas aeruginosa was also considered to beconstitutive (31). What is apparently incompatible with ourobservation of constitutive succinate utilization in A.brasilense was the finding that there was a sixfold inductionof the succinate transport system upon the addition ofsuccinate to exponentially growing cells in FMM. A logicalconclusion of these facts would be that the basal level ofsuccinate uptake (as found in fructose-grown cells) is suffi-cient to meet the carbon source requirement ofA. brasilenseto grow at the maximal rate with a doubling time of 110 minas in SMM. It is perhaps relevant to point out at this juncturethat in bacteria and yeasts, some fluctuations in enzymelevels whose syntheses are considered constitutive havebeen known to occur depending on the carbon source ofgrowth (14, 24, 26).The addition of succinate to cells growing exponentially in

FMM caused both transient and permanent repression of theinduced syntheses of enzyme I, enzyme II, and 1-PFK. Thismode of repression of the fru-enzymes by succinate isreminiscent of the repression of P-galactosidase by glucosein E. coli (19, 23). Unlike the case of P. aeruginosa (28, 29),neither the duration of transient repression nor the intensityof permanent repression was altered in A. brasilense by

increasing the concentration of added succinate by evenfivefold. Ucker and Signer (32) also observed succinate-mediated transient and catabolite repression of ,-galacto-sidase synthesis in R. meliloti. We wish to point out here thatin our large-scale preparations the specific activity of thephosphotransferase system in the crude extracts of A.brasilense grown with fructose as the only carbon sourcewas in the range of 18 to 22 nmol min-1 mg of protein-'(unpublished observation), which is much higher than wereported in our previous study (5 nmol min-' mg ofprotein-1), as the cells were then grown on a mixture ofsuccinate and fructose (3). Undoubtedly, the reduced activ-ity observed was due to the permanent repression effect ofsuccinate on fru-PTS. With Pseudomonas doudoroffii, Bau-mann and Baumann (1) also observed that the specificactivities of fru-PTS and 1-PFK were about 3.5-fold higherwhen cells were grown on fructose than when they weregrown on a mixture of succinate and fructose.

In a previous report, succinate was shown to repress aswell as inhibit the glucose permease activity in Arthrobactercrystallopoietes (8). Midgley and Dawes (17) also reportedthe repression and inhibition of glucose permease by suc-cinate in P. aeruginosa, but they mentioned that the phe-nomenon of inhibition was not reproducible. Mukadda et al.(20), on the other hand, observed that the glucose permeaseactivity in P. aeruginosa was repressed but not inhibited bysuccinate; however, they themselves pointed out that theirresults must be interpreted with caution since the succinatetransport system itself was inducible in the bacterium. Itshould be borne in mind, though, that in both Arthrobactercrystallopoietes and P. aeruginosa, uptake of glucose ismediated by an active transport process. However, in P.duodoroffli, in which fructose is transported via PTS, suc-cinate was reported to cause 36% inhibition of fructoseuptake (1). To demonstrate unequivocally whether inducerexclusion occurred in A. brasilense, we chose fructose-grown cells which were fully induced for succinate uptakeand also those cells which were not induced by succinate.With neither type of cells could succinate inhibit the uptakeof fructose. This is unlike the case of enteric bacteria, inwhich inducer exclusion plays a very important role indiauxic growth (7, 24, 27).

It is therefore evident from our studies that the sole modeof succinate regulation of fructose uptake and catabolism inA. brasilense is by repression (transient and permanent) ofthe induced syntheses of the fru-enzymes. Nonetheless, thebiochemical basis of coinduction of the fru-enzymes byfructose and their coordinate repression by succinate re-mains unclear.

ACKNOWLEDGMENTSWe thank Pushpita Maulik and Anand Bachhawat for their helpful

suggestions.This work was funded by grant F19(15)/80-FCII from the Indian

Council for Agricultural Research, New Delhi.

LITERATURE CITED1. Baumann, P., and L. Baumann. 1975. Catabolism of D-fructose

and D-ribose by Pseudomonas doudoroffli. I. Physiologicalstudies and mutant analysis. Arch. Microbiol. 105:225-240.

2. Das, A., and A. K. Mishra. 1983. Utilization of fructose byAzospirillum brasilense. Can. J. Microbiol. 29:1213-1217.

3. DuttaGupta, K., and S. Ghosh. 1984. Identification of aphosphoenolpyruvate:fructose 1-phosphotransferase system inAzospirillum brasilense. J. Bacteriol. 160:1204-1206. (Erratum,163:410, 1985.)

4. Elmerich, C. 1984. Molecular biology and ecology of diazo-

J. BACTERIOL.

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

REGULATION OF FRUCTOSE UPTAKE AND CATABOLISM 4367

trophs associated with non-leguminous plants. Bio/Technology2:967-978.

5. George, S. E., C. J. Costenbader, and T. Melton. 1985. Diauxicgrowth in Azotobacter vinelandii. J. Bacteriol. 164:866-871.

6. Goebel, E. M., and N. R. Krieg. 1984. Fructose catabolism inAzospirillum brasilense and Azospirillum lipoferum. J. Bacte-riol. 159:86-92.

7. Kornberg, H. L., P. D. Watts, and K. Brown. 1980. Mechanismsof inducer exclusion by glucose. FEBS Lett. 117(Suppl.):K28-K36.

8. Krulwich, T. A., and J. C. Ensign. 1969. Alteration of glucosemetabolism of Arthrobacter crystallopoietes by compoundswhich induce sphere to rod morphogenesis. J. Bacteriol. 97:526-534.

9. Lessie, T. G., and P. V. Phibbs, Jr. 1984. Alternative pathwaysof carbohydrate utilization in pseudomonads. Annu. Rev. Mi-crobiol. 38:359-387.

10. Loh, W. H. T., C. I. Randles, W. R. Sharp, and R. H. Miller.1984. Intermediary carbon metabolism of Azospirillum bra-silense. J. Bacteriol. 158:264-268.

11. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenoi reagent. J,Biol. Chem. 193:267-275.

12. Lynch, W. H., and M. Franklin. 1978. Effect of temperature ondiauxic growth with glucose and organic acids in Pseudomonasfluorescens. Arch. Microbiol. 118:133-140.

13. Magasanik, B. 1970. Glucose effects: inducer exclusion andrepression, p. 189-219. In J. R. Beckwith and D. Zipser (ed.),The lactose operon. Cold Spring Harbor Laboratory, ColdSpring Harbor, N.Y.

14. Maitra, P. K., and Z. Lobo. 1971. Kinetic study of glycolyticenzyme synthesis in yeast. J. Biol. Chem. 246:475-488.

15. Martinez-Drets, G., M. D. Gallo, C. Burpee, and R. H. Burris.1984. Catabolism of carbohydrates and organic acids andexpression of nitrogenase by azospirilla. J. Bacteriol. 159:80-85.

16. Maulik, P., and S. Ghosh. 1986. NADPH/NADH-dependentcold-labile glutamate dehydrogenase in Azospirillum brasilense.Purification and properties. Eur. J. Biochem. 155:595-602.

17. Midgley, M., and E. A. Dawes. 1973. The regulation of transportof glucose and methyl a-glucoside in Pseudomonas aeruginosa.Biochem. J. 132:141-154.

18. Monod, J. 1947. The phenomenon of enzymatic adaptation andits bearings on problems of genetics and cellular differentiation.Growth Symp. 11:223-289.

19. Moses, V., and C. Prevost. 1966. Catabolite repression of3-galactosidase synthesis in Escherichia coli. Biochem. J.

100:336-353.20. Mukadda, A. J., G. L. Long, and A. H. Romano. 1973. The

uptake of 2-deoxy-D-glucose by Pseudomonas aeruginosa andits regulation. Biochem. J. 132:155-162.

21. Novick, N. J., and M. E. Tyler. 1982. L-Arabinose metabolism inAzospirillum brasilense. J. Bacteriol. 149:364-367.

22. Okon, Y., S. L. Albrecht, and R. H. Burris. 1976. Factorsaffecting growth and nitrogen fixation of Spirillum lipoferum. J.Bacteriol. 127:1248-1254.

23. Perlman, R. L., B. de Crombrugghe, and I. Pastan. 1969. CyclicAMP regulates catabolite and transient repression in E. coli.Nature (London) 223:810-812.

24. Postma, P. W., and J. W. Lengeler. 1985. Phosphoenol-pyruvate:carbohydrate phosphotransferase system of bacteria.Microbiol. Rev. 49:232-269.

25. Roehl, R. A., and P. V. Phibbs, Jr. 1982. Characterization andgenetic mapping of fructose phosphotransferase mutations inPseudomonas aeruginosa. J. Bacteriol. 149:897-905.

26. Saier, M. H., Jr. 1977. Bacterial phosphoenolpyruvate:sugarphosphotransferase systems: structural, functional, and evolu-tionary interrelationships. Bacteriol. Rev. 41:856-871.

27. Saier, M. H., Jr., and S. Roseman. 1976. Sugar transport.Inducer exclusion and regulation of the melibiose, maltose,glycerol, and lactose transport systems by the phospho-enolpyruvate:sugar phosphotransferase system. J. Biol. Chem.251:6606-6615.

28. Siegel, L. S., P. B. Hylemon, and P. V. Phibbs, Jr. 1977. Cyclicadenosine 3',5'-monophosphate levels and activities of adenyl-ate cyclase and cyclic adenosine 3',5'-monophosphate phospho-diesterase in Pseudomonas and Bacteroides. J. Bacteriol.129:87-96.

29. Smyth, P. F., and P. H. Clarke. 1975. Catabolite repression ofPseudomonas aeruginosa amidase: the effect of carbon sourceon amidase synthesis. J. Gen. Microbiol. 90:81-90.

30. Tarrand, J. J., N. R. Krieg, and J. Dobereiner. 1978. Ataxonomic study of the Spirillom lipoferum group, with descrip-tions of a new genus, Azospirillum gen. nov., and two species,Azospirillum lipofemum (Beijerinck) comb. nov. and Azospiril-lum brasilense sp. nov. Can. J. Microbiol. 24:967-980.

31. Tiwari, N. P., and J. J. R. Campbell. 1969. Enzymatic control ofthe metabolic activity of Pseudomonas aeruginosa grown inglucose or succinate media. Biochim. Biophys. Acta 192:395-401.

32. Ucker, D. S., and E. R. Signer. 1978. Catabolite-repression-likephenomenon in Rhizobium meliloti. J. Bacteriol. 136:1197-1200.

33. Von Bullow, J. W. F., and J. Dobereiner. 1975. Potential fornitrogen fixation in maize genotypes in Brazil. Proc. Natl. Acad.Sci. USA 72:2383-2393.

34. Westby, C. A., D. S. Cutshall, and G. V. Vigil. 1983. Metabolismof various carbon sources by Azospirillum brasilense. J. Bacte-riol. 156:1369-1372.

VOL. 169, 1987

on October 15, 2020 by guest

http://jb.asm.org/

Dow

nloaded from