JOURNAL OF BACTERIOLOGY, Feb. 1983, p. 930-936 Vol. 153, No. 20021-9193/83/020930-07$02.00/OCopyright 0 1983, American Society for Microbiology

Characterization of Two Glyceraldehyde-3-PhosphateDehydrogenase Isoenzymes from the Pentalenolactone

Producer Streptomyces arenaeKARL-HEINZ MAURER, FRIEDHELM PFEIFFER,t HARTMUT ZEHENDERJ AND DIETER

MECKE* -'Physiologisch-chemisches Institut der Universitat Tubingen, D-7400 Tubingen, Federal Republic of Germany

Received 13 September 1982/Accepted 29 November 1982

Pentalenolactone (PL) irreversibly inactivates the enzyme glyceraldehyde-3-phosphate dehydrogenase [D-glyceraldehyde-3-phosphate:NAD' oxidoreductase(phosphorylating)] (EC 1.2.1.12) and thus is a potent inhibitor of glycolysis in bothprocaryotic and eucaryotic cells. We showed that PL-producing strain Streptomy-ces arenae TU469 contains a PL-insensitive glyceraldehyde-3-phosphate dehy-drogenase under conditions of PL production. In complex media no PL produc-tion was observed, and a PL-sensitive glyceraldehyde-3-phosphatedehydrogenase, rather than the insensitive enzyme, could be detected. Theenzymes had the same substrate specificity but different catalytic and molecularproperties. The apparent Km values of the PL-insensitive and PL-sensitiveenzymes for glyceraldehyde-3-phosphate were 100 and 250 ,uM, respectively, andthe PL-sensitive enzyme was strongly inhibited by PL under conditions in whichthe PL-insensitive enzyme was not inhibited. The physical properties of the PL-insensitive enzyme suggest that the protein is an octamer, whereas the PL-sensitive enzyme, like other glyceraldehyde-3-phosphate dehydrogenases, ap-pears to be a tetramer.

The antibiotic pentalenolactone (PL) (Fig. 1)is one of the few sesquiterpene lactones pro-duced by procaryotes (4). In the fermented brothof several strains of actinomycetes (11, 16, 18),PL occurs in three different forms: chlorohy-drin, diol, and epoxide (1, 10). Several interme-diates and metabolites of PL have also beendescribed (4). PL is a potent inhibitor of theglycolytic and gluconeogenetic pathways in bothprocaryotic and eucaryotic organisms (6, 9). Invitro, an irreversible inactivation of the enzymeglyceraldehyde-3-phosphate dehydrogenase [D-glyceraldehyde-3-phosphate:NAD+ oxidore-ductase (phosphorylating); EC 1.2.1.12](GAPDH), of microbial, plant, and mammalianorigin, by 10-6 M PL has been shown (6, 9, 15).The action of the antibiotic seems to be restrict-ed to this enzyme, as the addition of noncarbo-hydrate substrates, such as pyruvate, to theculture medium restores normal growth activityto microorganisms [10].The NAD+-dependent GAPDH has been iso-

lated and characterized from a number of pro-caryotic and eucaryotic sources. The different

t Present address: Max-Planck Institut futr Psychiatrie, Abt.Neurochemie, D-8033 Martinsried, Federal Republic of Ger-many.

t Present address: Biochemisches Institut der UniversitatFreiburg, D-7800 Freiburg, Federal Republic of Germany.

enzymes are strictly homologous and show se-quence identities of 50 to 60%o. Furthermore,each of these enzymes that has so far beencharacterized has a molecular weight of 146,000and consists of four identical subunits (8).

Streptomyces arenae TU469 produces PL in aminimal medium containing glucose or mannitolas the sole carbon source (10). Despite thepresence of PL, growth or glucose utilization isnot inhibited. In a preliminary report we havedescribed a PL-insensitive GAPDH which isresponsible for this resistance of S. arenaeTU469 (20).We now report that during cultivation in com-

plex media containing peptone and yeast ex-tract, no PL was produced. In these media anisoenzyme of GAPDH was formed, exhibitingalmost the same sensitivity to inactivation by theantibiotic as the enzymes from other sources.The two isoenzymes from S. arenae TU469were isolated, and their molecular structureswere characterized. They showed similar cata-lytic properties compared with the enzymesknown so far, but they were different in respectto structural composition.

MATERIALS AND METHODSChemicals and reagents. All reagents were of analyti-

cal grade. Glyceraldehyde-3-phosphate-diethylacetal

930

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

GAPDH ISOENZYMES 931

CH3

0Po n0FIG. 1. PL (chlorohydrin form).

(cyclohexylammonium salt), NAD+,GAPDH, and the reference proteinsweight determination were obtained frMannheim Corp., Mannheim, FederGermany. DEAE-Sephacel and Blue6B were obtained from Pharmacia FUppsala, Sweden. Celite 535 was fromberg, Federal Republic of Germany. 13and Bio-Gel HTP were from Bio-RacRichmond, Calif.PL was isolated as chlorohydrin fro

fermentation broth of a 4,000-liter fernarenae TU469, carried out at the C

Biotechnologische Forschung, Braunheim, Federal Republic of Germany. Iby chromatography on Amberlite-XAIdex LH-20 and by high-pressure liqui4phy on Merck LiChrosorb RP 18. In -

the chlorohydrin form of PL was usedOrganisn. S. arenae TU469 (DSIN

kindly provided by H. Zahner, Institulogie, Tubingen, Federal Republic of (

Media and culture conditions. S.produced about 3,ug of PL per ml in sycontaining 4% mannitol, 0.25% asr(NH4)2SO4, 0.1% NaCI, 0.3% K2HPOMgSO4* 7H20, 0.04% CaC12* 2H20,* 7H20, and 0.001% ZnSO4* 7H20 at pWickerham medium (23) was used as ccand contained 0.5% peptone, 0.3% yeamalt extract, and 3% glucose at pH 7.(For enzyme production, cells were

liter fermentor (Giovanola Intensor ISwitzerland) at 30°C for 20 h at pH 7.0.was harvested by centrifugation and vmM Tris-chloride buffer (pH 8.0) coimercaptoethanol, 5 mM EDTA, and(buffer A). The cells were suspendedbuffer A and 1 volume of glass beadsmm; Braun-Melsungen, Federal Repuny). Homogenization was carried outblender at high speed for 5 min at 4°C. Twas decanted, and the glass beads weibuffer A. The pooled supernatants werc6,000 x g for 20 min to remove cell deFor some experiments S. arenae TU

in 100 ml of medium in 1-liter Erlenr30°C with shaking. Washed cells wtultrasonic disruption in a Branson Soniin two 15-s pulses at0°C.

Purification of PL-sensitive and PL-insensitiveGADPH from S. arenae TU469. Both isoenzymes wereprepared by the following procedure, in which alloperations were carried out at 4°C. To the crude

COOH extract 1 g of Celite per 100 mg of protein was added,and after 30 min of constant stirring, the suspensionwas brought to 85% saturation by the addition of solidammonium sulfate (559 g/liter) over the next 30 min.The suspension was allowed to stand for 1 h and wasthen centrifuged for 30 min at 15,000 x g. The pelletwas resuspended in buffer A saturated to 85% withammonium sulfate and poured into a column (6-cmdiameter). After settling, the enzyme was eluted with a600-ml linear gradient of 85 to 40% saturated ammoni-um sulfate. Fractions containing the enzyme weredialyzed for 24 h against three changes of 5 liters ofbuffer A. The extract was applied to a column (1 by 12

rabbit muscle cm) of DEAE-Sephacel equilibrated in buffer A. Thefor molecular enzyme was eluted with a 300-mI linear gradient of 0 to

rom Boehringer 0.5 M NaCl in buffer A and then applied to a column (2al Republic of by 150 cm) of Bio-Gel A 0.5m equilibrated in buffer A.Sepharose CL- Fractions with high activity were pooled and dialyzedine Chemicals, overnight against two changes of 2 liters of 10 mMServa, Heidel- potassium phosphate buffer (pH 6.9) containing 5 mMio-Gel A 0.5 m mercaptoethanol. The dialyzed preparation was ap-I Laboratories, plied to a column (1 by 15 cm) of Bio-Gel HTP

equilibrated with 10 mM potassium phosphate buffer.im the acidified GAPDH was eluted with a 200-ml linear gradient of 10mentation of S. to 200 mM potassium phosphate at pH 6.9. ActiveJesellschaft fiOr fractions were pooled, chromatographed on a columnkschweig-Stock- (2 by 50 cm) of Bio-Gel P6 equilibrated with buffer A,PL was purified and stored at 40C.D 2 and Sepha- GAPDH assay. GAPDH activity was measuredd chromatogra- spectrophotometrically with NAD+ as the substrate,all experiments according to the method of Furfine and Velick (7). TheI. standard assay mixture (2.5 ml) contained 50 mM Tris-4 40 734) was chloride buffer (pH 8.0), 4 mM mercaptoethanol, 4It fur Mikrobio- mM EDTA, 2 mM NAD+, and 4 mM Na2HAsO4. TheJermany. reaction was started by the addition of glyceraldehyde-arenae TU469 3-phosphate to a final concentration of 0.4 mM.nthetic medium Enzyme inactivation was measured by adding 10paragine, 0.2% nmol of PL (4 ,uM, unless otherwise stated) to the)4 3H20, 0.1% assay mixture. After 1 h of incubation at 26°C, the0.002% FeSO4 reaction was started by addition of glyceraldehyde-3-tH 6.2 and 30°C. phosphate. (PL was dissolved in 50%o methanol. AnDmplex medium equal amount was added to the reference samples.)st extract, 0.3% The increase in absorbance at 340 nm was measured,D and 30°C. and the activity was calculated from the initial reactiongrown in a 20- velocity at 260C. One unit of GAPDH was defined asb20; Monthey, the amount of the enzyme which catalyzes the reduc-.The mycelium tion of 1 ±mol of NAD+ per min.vashed with 50 D-Glyceraldehyde-3-phosphate was used for the de-ntaining 5 mM termination of the substrate specificity, whereas DL-1 ,uM NAD+ glyceraldehyde-3-phosphate was used for the routine

in 1 volume of assays. To test the reversibility of the reaction and the(diameter, 0.5 inhibition of the enzyme with iodoacetic acid, the

blic of Germa- assay was performed with 1,3-diphosphoglycerate andin a laboratory NADH as substrates, as described by Bergmeyer (3).Fhe supernatant PL determination. The concentration of PL wasre washed with determined by inhibition of 13 mU of rabbit musclee centrifuged at GAPDH after incubation for 1 h at 26°C in the assay-bris. mixture. The inhibition of the enzyme proved to be a'469 was grown function of the PL concentration. PL concentrationsneyer flasks at could be estimated by comparison with a standardere broken by curve made with pure PL.fier 12b for 30 s Protein determination was by the procedure of Low-

ry et al. (13).

VOL. 153, 1983

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

932 MAURER ET AL.

RESULTS

PL sensitvity of GAPDH activity in S. arenaeTU469 under various culture conditions. Whengrown in the synthetic medium described above,S. arenae TU469 produced PL to a concentra-tion of 10 ,uM after 50 h in submersed culture,whereas no PL was produced by this microor-ganism during 90 h of submersed culture inWickerham medium. During the productionphase of PL, no inhibition of growth or ofglucose utilization was observed. Cell-free ex-tracts of S. arenae TU469 grown under bothculture conditions were compared with respectto PL sensitivity of GAPDH activity. Althoughthe activity was inactivated by more than 90%o at0.4 FM PL after growth in Wickerham medium,it was inactivated by less than 5% at 0.4 ,M PLafter growth in the synthetic medium. Thus, thePL sensitivity of the enzyme activity was com-pletely different under the two culture condi-tions. In further experiments a PL concentrationof 4 ,uM was used to differentiate between PL-sensitive and PL-insensitive GAPDH activity.A time course of PL sensitivity and PL insen-

sitivity was followed by first growing cells for 24h in either synthetic or Wickerham medium.Cells were harvested by centrifugation, washedtwice with sterile 0.9%o NaCl, and then trans-ferred to the other medium. In one set of experi-ments, cells were transferred from synthetic

medium to Wickerham medium (Fig. 2). The PL-sensitive GAPDH activity began to appear un-der these conditions after 3 h. Only the PL-sensitive activity was found 20 h after transfer.The possibility of an infection with anotherstrain could be excluded.

In another set of experiments, cells weretransferred from Wickerham medium to synthet-ic medium (Fig. 3). PL-insensitive activity firstappeared after 12 h, and 8 h later PL could bedetected. At the same time, the PL-sensitiveenzyme activity decreased rapidly. After 48 honly the PL-insensitive GAPDH activity wasfound.

In a control experiment in which cells weretransferred from Wickerham medium back toWickerham medium, the specific activity of thesensitive enzyme reached a maximum of 0.9U/mg after 24 h and decreased to 0.45 U/mg after48 h. Nearly the same specific activity of PL-insensitive enzyme activity was found 48 h aftercells were transferred from Wickerham mediumto synthetic medium. Cells grown in Wickerhammedium reached twice the biomass and, accord-ingly, twice the amount of GAPDH activitycompared with cells grown in synthetic medium.

After the addition ofPL (0.2 ,ug/ml) to Wicker-ham medium, there was a slow appearance ofPL-insensitive GAPDH activity which followedthe total inhibition of the PL-sensitive GAPDHactivity (Fig. 4). The cells were not strongly

c0

.Cc

100

80

60

40

20

500*400

*300 E

E200 8f

I-,

100 O

0

50 -.

0 5 10 15 20 25

Hours

FIG. 2. PL sensitivity ofGAPDH activity in cell-free extracts of S. arenae TU469 grown in synthetic mediumand transferred to Wickerham medium. PL sensitivity was measured as described in the text. The increasinginhibition corresponds to the appearance of PL-sensitive GAPDH. Inhibition at 100%0 indicates the PL-sensitiveform of the enzyme, inhibition at 0o indicates the PL-insensitive form. The specific activity was 0.45 U/mgthroughout the experiment. Symbols: A, mycelium dry weight; 0, % inhibition.

A--A

//

A/

A/j/ T-i-§-{§§

/A /~ f~"/T/I

0

1'I i 1/'~I

J. BACTERIOL.

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

GAPDH ISOENZYMES 933

7i 200-0.6- 300 E

E 100 1A,,00

50 60 80 1200

FI.3 nyeatiiyielfe exrat of a.

0-0~

20 40 60 80 100 120Hours

FIG. 3.. Enzyme activity in cell-free extracts of S.arenae TU469 grown in Wickerham medium andtransferred to synthetic medium. Enzyme activity andPL sensitivity were tested as described in the text.According to the PL sensitivity, the enzyme activitywas differentiated between the PL-insensitive and PL-sensitive forms of GAPDH. Symbols: 0, PL-insensi-tive GAPDH; 0, PL-sensitive GAPDH; 0, PL; A,mycelium dry weight.

affected by the inhibition of the PL-sensitiveGAPDH, as they were able to grow in complexmedium without catabolizing glucose.

Purification of two GAPDH isoenzymes. Toinvestigate the molecular mechanism of the de-velopment ofPL resistance, GAPDH was isolat-ed from cells grown in Wickerham medium andin synthetic medium. Although the same purifi-cation scheme was followed, differences in theelution profiles and the final specific activities ofthe enzymes purified from the two sources indi-cated the presence of two isoenzymes; i.e., thePL-sensitive enzyme eluted at 0.07 M potassiumphosphate from the Bio-Gel HTP column,whereas the PL-insensitive enzyme eluted at0.04 M. PL-insensitive GAPDH from S. arenaeTU469 was purified 50-fold, with a recovery of25% compared with the crude extract (Table 1).PL-sensitive GAPDH was purified 93-fold, witha recovery of 10% (Table 2). After purification,both forms of the enzyme showed a single bandin sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (Fig. 5). Also, the final specificactivities were in the range known for GAPDHfrom other sources (Table 3).During initial purification experiments, we

realized that Dextran-containing chromatogra-phy materials inactivated both enzymes. These

materials therefore had to be avoided. Further-more, in contrast to GAPDH from othersources, both isoenzymes did not bind to BlueSepharose CL-6B, and therefore affinity chro-matography could not be used on this material.

After purification, the PL-insensitive enzymewas stable at 4°C for up to 6 months, whereasthe PL-sensitive enzyme lost activity with a half-life of 7 days in buffer A.PL sensitivity of the GAPDH isoenzymes. Table

4 lists the different PL sensitivities of the twoisoenzymes of S. arenae TU469. The purifiedisoenzymes showed the same PL sensitivities asdescribed for cell-free extracts, thus excluding aPL-inactivating mechanism. The inhibition ofseveral GAPDHs by PL is irreversible and fol-lows first-order kinetics with different rates (9,15). To measure PL sensitivity, it was thereforenecessary to standardize the percent inhibitionand the incubation time. Relative PL sensitivityis expressed as the concentration of antibioticnecessary for 90% inhibition after 1 h of incuba-tion at 26°C divided by the concentration of PLnecessary for 90% inhibition of the same activityof rabbit muscle GAPDH. The relative sensitiv-ities ofGAPDHs from various sources and from

m

0>

E

2

ua

NCw

'E8

U'0E

n=

0,

L-.o

cS

0 4 8 12 16 20 24 28Hours

FIG. 4. Enzyme activity in cell-free extracts of S.arenae TU469 grown in Wickerham medium, afterinactivation in vivo with 0.3 p.g of PL/ml. Enzymeactivity and PL sensitivity were tested as described inthe text. According to PL sensitivity, the enzymeactivity was differentiated between PL-insensitive andPL-sensitive forms of GAPDH activity. Symbols: 0,PL-insensitive GAPDH; 0, PL-sensitive GAPDH; A,mycelium dry weight in PL culture; A, mycelium dryweight in control.

VOL. 153, 1983

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

TABLE 1. Purification of PL-insensitive GAPDH from S. arenae TU469

Vol Total Total Sp act Recovery PurificationPurification step (m) activity protein (U/Mg) (% (fold)(U) (mg)

Cell extract 280 480 686 0.65 100 1Ammonium sulfate eluate 71 212 47 4.5 46 6.8DEAE-Sephacel eluate 11 314 27 12 65 18Bio-Gel A 0.5m eluate 36 170 13 13 37 19Bio-Gel HTP eluate 30 120 3.6 33 25 50

TABLE 2. Purification of PL-sensitive GAPDH from S. arenae TU469

Vol Total Total Sp act Recovery PurificationPurificationstepactivity protein (U/mg) (% (fold)Purificationstep (ml) ~(U) (mg)

Cell extract 1,000 3,340 2,780 1.2 100 1Ammonium sulfate eluate 640 717 109 6.5 22 5.4DEAE-Sephacel eluate 48 873 25 32 26 27Bio-Gel A 0.5m eluate 77 586 10 58 17.5 49Bio-Gel HTP eluate 33 362 3.3 112 11 93

S. arenae TU469 are listed in Table 3. Comparedwith PL-sensitive GAPDHs from varioussources, the PL-sensitive isoenzyme showedrelatively low sensitivity, being 16 to 100 timesless sensitive than other enzymes. The PL-insensitive isoenzyme, however, was more thanO0s times less sensitive than GAPDH from othersources (Table 3).Substrate kinetics. Both isoenzymes of

GAPDH of S. arenae TU469 require as sub-strates NAD+, arsenate or phosphate, and D-glyceraldehyde-3-phosphate for their activity.These isoenzymes are D-glyceraldehyde-3-phos-phate:NAD+ oxidoreductases (phosphorylat-ing), as NADP+ is not accepted as a substrateand both enzymes reduce 1,3-diphosphoglycer-ate with NADH. The preparations of the twoenzymes showed similar substrate kinetics forthe substrates NAD+ and glyceraldehyde-3-phosphate. The apparent Km for NAD+ was 120,uM for the PL-insensitive form and 110 ,uM forthe PL-sensitive form. For glyceraldehyde-3-phosphate an apparent Km value of 100 FM wasobtained for the PL-insensitive form, and anapparent Km value of 250 ,uM was obtained forthe PL-sensitive form. A comparison with en-zymes from other sources is given in Table 3.

Iodoacetic acid was incubated with 40 mU ofeach enzyme preparation for 5 min in the assaywhich used 1,3-diphosphoglycerate and NADHas substrates. The PL-sensitive isoenzymeshowed 50%o inhibition in the presence of 200ILM iodoacetic acid, whereas 1,200 ,uM iodoace-tic acid was necessary for 50%o inhibition of thePL-insensitive form.

Molecular weight. The molecular weights ofthe native enzymes were estimated by chroma-tography of 0.5 mg of enzyme on a Bio-Gel A

0.5m column (2 by 150 cm). The referenceproteins were catalase, aldolase, fumarase, andrabbit muscle GAPDH. PL-sensitive GAPDHfrom S. arenae TU469 eluted before rabbit mus-

A BC D

FIG. 5. Sodium dodecyl sulfate-polyacrylamide gelelectrpphoresis of the purified enzymes. An 11 to 15%gradient gel was run according to the method ofLaemmli (12). Samples were prepared by incubation at100°C for 3 min in the presence of 1% sodium dodecylsulfate and 1% mercaptoethanol. Lanes: A, referenceproteins bovine serum albumin, ovalbumin, trypsino-gen, lactoglobulin, and lysozyme; B, PL-insensitiveGAPDH from S. arenae TU469; C, PL-sensitiveGAPDH from S. arenae TU469; D, GAPDH fromrabbit muscle.

934 MAURER ET AL. J. BACTERIOL.

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

GAPDH ISOENZYMES 935

TABLE 3. Comparison of NAD+-dependent GAPDHs from different sourcesaApparent Km (RM) Relative

Source Mol wt Mol wt No. of for: Sp act" sensitivity tosubunit subunits (U/mg) inactivationNAD+ GAPd by PLC

Rabbit (7) 146,000 36,500 4 13 90 164 1Spinach (21) 145,000 37,000 4 NDe ND 80 5.7 (15)Yeast (22) 146,000 36,500 4 44 160 100 3.2 (9)E. coli (2) 144,000 35,000 4 166 142 40 1.4 (9)S. arenae TU469PL sensitive 180,000 43,000 4 110 250 112 100PL insensitive 290,000 37,000 8 120 100 33 105a Numbers within parentheses are reference numbers.b Specific activity of purified enzyme.c Expressed as PL concentration required for 90% inhibition divided by that required for 90% inhibition of

rabbit muscle enzyme.d GAP, Glyceraldehyde-3-phosphate.' ND, Not determined.

cle GAPDH but after fumarase. The PL-insensi-tive enzyme eluted before all reference proteinsused. By comparison with the elution volumesof the reference proteins, the molecular weightsof the isoenzymes were calculated to be 280,000for the PL-insensitive form and 180,000 for thePL-sensitive form.Upon gel electrophoresis in the presence of

sodium dodecyl sulfate and mercaptoethanol,the PL-insensitive enzyme showed one subunitwith an apparent molecular weight of 37,000,whereas the PL-sensitive enzyme gave one sub-unit with an apparent molecular weight of 43,000(Fig. 5). Thus, the PL-sensitive GAPDH proba-bly consists offour subunits, in contrast to eightsubunits in the PL-insensitive form.

DISCUSSIONS. arenae TU469 produces PL, an antibiotic

shown to inactivate GADPH from varioussources (9, 15). The production ofPL is stronglydependent on culture conditions. Under condi-tions of PL production, the organism strictly

TABLE 4. PL sensitivity of purified GAPDH fromrabbit muscle and from S. arenae TU469

Inhibition (%) of 50 mU of GAPDH activityafrom:

PL concn S. arenae TU469(M) Rabbit (PL

sensitive) PL PLinsensitive sensitive

4 x 10-6 100 <2 1004 x 10-7 100 <2 924 x 10-8 100 <2 704 x 10-9 90 <2 33

a Inhibition was measured after 1 h of incubation at26°C; 0.5 mg of each PL-sensitive enzyme and 1.5 mgof the PL insensitive enzyme were used. Values givenare means of triplicate determinations.

depends on carbohydrate utilization. Therefore,the producer has to be resistant to the antibiotic.

It is known that many antibiotic-producingstreptomycetes shift from a state of antibioticsensitivity, during which no antibiotic is pro-duced, to a state of antibiotic resistance duringthe production phase (14). This report nowshows that in S. arenae TU469 a PL-sensitiveGAPDH was present when no PL was pro-duced, whereas a distinct, PL-insensitive isoen-zyme was expressed during PL production.Thus, it was possible to distinguish betweenthese two phases by determination of the isoen-zymes of GAPDH (Fig. 2 and 3).The resistance of a bacterium to its antibiotic

is often inducible; for example, chloramphenicolresistance could be induced by the addition ofchloramphenicol under conditions where noantibiotic was produced in Streptomyces vene-zuelae (14). This might hold true as well for S.arenae, in which synthesis of PL-insensitiveenzyme after addition of PL to Wickerham me-dium was observed (Fig. 4). In such a case thelow activity of the PL-insensitive GAPDH mightbe sufficient to maintain growth in complexmedium. After supplementation with PL, thegrowth rate of S. arenae TU469 was not affect-ed.The two isoenzymes, purified to apparent

homogeneity, had similar substrate specifity andsubstrate kinetics, and both were susceptible tosulfhydryl reagents. Because of their substratespecifity, both isoenzymes are D-glyceralde-hyde-3-phosphate:NAD+ oxidoreductases(phosphorylating) (EC 1.2.1.12). The two isoen-zymes had differences in both subunit composi-tion and total molecular weight. The PL-sensi-tive form had a molecular weight of 180,000 andconsisted offour subunits, each with a molecularweight of 43,000. The molecular weight of thePL-insensitive GAPDH was 290,000; this en-

VOL. 153, 1983

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

936 MAURER ET AL.

zyme was composed of eight subunits, each witha molecular weight of 37,000. Because GAPDHhas not been purified so far from other Strepto-myces strains, these values can only be com-pared with enzyme preparations from othersources (Table 3). The enzymes from pig, rabbit,lobster, yeast, and spinach and from the pro-caryotes Escherichia coli, Bacillus stearother-mophilus, and Thermus aquaticus have beenpurified, and the amino acid sequences havebeen established (8). The sequences are strictlyhomologous and show identities of 50 to 60%6.The NAD+-specific enzymes from all of thesesources seem to be composed of four identicalpolypeptide chains, with a resulting molecularweight of about 146,000. Possibly the uniquestructure of the octameric isoenzyme in S.arenae TU469 is responsible for its PL insensi-tivity. In Streptococcus mutans an NAD+-spe-cific GAPDH with a molecular weight of 260,000has been reported (5). It is similar to the PL-insensitive isoenzyme in this respect, but thenumber and size of subunits was not deter-mined.

In spinach chloroplasts and the alga Scenedes-mus obliquus, two forms of an NAD+/NADP+-dependent GAPDH (EC 1.2.1.13) have beendescribed which have different substrate specifi-ties for NAD+ and NADP+. It seems that thereis a metabolite-dependent interconversion of thehigh-molecular-weight GAPDH, primarily ac-tive with NAD+, to a low-molecular-weightform which has a substrate specifity primarilyfor NADP+ (17, 19). It might be speculated thatsome sort of interconversion takes place in S.arenae TU469, e.g., by partial proteolysis, totransform the PL-sensitive enzyme to the insen-sitive enzyme. On the other hand, differentgenes might code for the two isoenzymes. Fur-ther research is required to clarify the relationsbetween the two forms of GAPDH in S. arenaeTU469. Peptide mapping might show whetherthe PL-insensitive enzyme arises from the sensi-tive enzyme by limited proteolysis.

ACKNOWLEDGMENTSWe thank J. Lehmann from the Gesellschaft fOr Biotechno-

logische Forschung for help in fermentation of S. arenae andD. Graham for carefully reading the manuscript.

This work was supported by the Deutsche Forschungsge-meinschaft (Sonderforschungsbereich 76) and by the Fondsder Chemischen Industrie.

LITERATURE CITED1. Alawa, S., H. Akutsu, T. Satomi, S. Kawabata, and K.

Sasaki. 1978. AA-57, a new antibiotic related to pentaleno-lactone. J. Antibiot. 31:729-731.

2. Alesao, G., and J. Josse. 1971. Glyceraldehyde phosphatedehydrogenase, phosphoglycerate kinase and phospho-

glyceromutase of Escherichia coli. J. Biol. Chem.246:4319-4333.

3. Bergmeyer, H. U. 1974. Methods of enzymatic analysis.Academic Press, Inc., New York.

4. Cae, D. E. 1981. Biosynthesis of sesquiterpenes, p. 285.In J. W. Porter and S. C. Spurgeon (ed.), Biosynthesis ofisoprenoid compounds, vol. 1. John Wiley & Sons, Inc.,New York.

5. Crow, V. L., and C. L. Wlttenberger. 1979. Separationand properties of NAD+- and NADP+-dependent glycer-aldehyde-3-phosphate dehydrogenases from Streptococ-cus mutans. J. Biol. Chem. 254:1134-1142.

6. Duenko, M., H. Balla, and D. Mecke. 1982. Specificinactivation of glucose metabolism from eukaryotic cellsby pentalenolactone. Biochim. Biophys. Acta 714:344-350.

7. Furfine, C. S., and S. F. Velck. 1965. The acyl-enzymeintermediate and the kinetic mechanism of the glyceralde-hyde-3-phosphate dehydrogenase reaction. J. Biol. Chem.240:844-855.

8. Harris, J. I., and M. Waters. 1976. Glyceraldehyde-3-phosphate dehydrogenases, p. 1-49. In P. D. Boyer (ed.),The enzymes, vol. 13. Academic Press, Inc., New York.

9. H n, S., J. Neeff, U. Heer, and D. Mecke. 1978.Arenaemycin (pentalenolactone): a specific inhibitor ofglycolysis. FEBS Lett. 93:339-342.

10. KedSculerlden, W., J. Lemle, R. Nyfeler, and H.Zihner. 1972. Stoffwechselprodukte von Mikroorganis-men. 105. Mitteilung: Arenaemycin E, D und C. Arch.Mikrobiol. 84:301-316.

11. Koe, B. K., B. A. Sobla, and W. D. Celmer. 1957. PA-132,a new antibiotic. I. Isolation and chemical properties.Antibiot. Annu. 195647:672-675.

12. Laennil, U. K. 1970. Cleavage of structural proteinsduring the assembly of the head of bacteriophage T4.Nature (London) 227:680-685.

13. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folin phenolreagent. J. Biol. Chem. 193:265-275.

14. Mailk, V. S. 1979. Regulation of chorismate-derived anti-biotic production. Adv. Appl. Microbiol. 25:75-93.

15. Mann, K., and D. Mecke. 1979. Inhibition of spinachglyceraldehyde-3-phosphate dehydrogenases by pentalen-olactone. Nature (London) 282:535-536.

16. MarIn, D. G., G. Slomp, S. Mlisak, D. J. Duchamp, andC. G. Chlgester. 1970. The structure and absolute configu-ration of pentalenolactone. Tetrahedron Lett. 56:4901-4903.

17. O'Brien, M. J., and R. Powls. 1976. Algal glyceraldehyde-3-phosphate dehydrogenase. Eur. J. Biochem. 63:155-161.

18. Okazakl, T., R. Enokita, A. Torikata, M. Inukal, M.Takeuchi, S. Takahashi, and M. Aral. 1979. Studies onactinomycetes producing pentalenolactone and its relatedcompounds. Annu. Rep. Sankyo Res. Lab. 31:94-103.

19. Pawltzd, K., and E. Latzko. 1974. Partial separation andinterconversion of NADH- and NADPH-linked activitiesof purified glyceraldehyde-3-phosphate dehydrogenasefrom spinach chloroplasts. FEBS Lett. 42:285-288.

20. Pfeiffer, F., and D. Mecke. 1980. Purification of pentaleno-lactone-insensitive glyceraldehyde-3-phosphate dehy-drogenase from the producer strain Streptomyces arenae.Hoppe Seyler's Z. Physiol. Chem. 361:314.

21. Pupilo, P., and R. Fagglan. 1979. Subunit structure ofthree glyceraldehyde-3-phosphate dehydrogenases ofsome flowering plants. Arch. Biochem. Biophys. 194:581-592.

22. Velck, S. F., and S. UdenflHend. 1953. The compositionand amino acid end groups of glyceraldehyde-3-phosphatedehydrogenase. J. Biol. Chem. 203:575-582.

23. Wkkerham, L. J. 1951. U.S. Dep. Agric. Tech. Bull.no. 1029.

J. BACTERIOL.

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from