Expression in Escherichia coli, Purification, and Reactivation of the

5 downloads 0 Views 4MB Size Report
Er- winia uredovora behind the lac Z promoter of pUCl8 resulting in a reading frame for the full polypeptide with additional 9 amino acids at the N terminus. This.
Vol. 267, No.

THEJOURNAL OF BIOLOGICAL CHEMISTRY

0 1992 by The American Society for Biochemistry and Molecular Biology,Inc.

Issue of October 5, PP. 19891-19895,1992

Printed in U.S.A.

Expression inEscherichia coli,Purification, and Reactivation of the Recombinant Emoiniu uredovora Phytoene Desaturase” (Received for publication, June 3, 1992)

Paul D. FraserSQ, Norihiko Misawan, Hartmut Linden$, ShigeyukiYamanoT, Kazuo Kobayashill, and Gerhard SandmannSil From the SLehrstuhl fur Physiologie und Biochemie der Pflanzen, Universitat Konstanz, P. 0. Box 5560, W-7750Konstanz, Germany and the llCentral Laboratories for Key Technology, Kirin Brewery Co., Ltd., 3, Miyahura-cho, Takasaki-shi, Gunma 370-12,Japan

A plasmid has been constructed by cloning thecomGenes encoding phytoene desaturase have now been isoplete crtI gene encoding phytoene desaturase Erfrom lated from bacteria (nonphotosynthetic and photosynthetic), winia uredovora behind the lacZ promoter of p U C l 8 fungi (see Ref. 3 for survey) as well as from cyanobacteria ( 7 ) resulting in a reading frame for the full polypeptide and higher plants (8). Comparative studies between isolated with additional9 amino acidsat the N terminus. This genes suggest that two groups of phytoene desaturases exist plasmid mediated the overexpression of phytoene de- with respectto amino acid sequence homology. However,they saturase in transformed Escherichia coli. The overex- all possess ashort highlyconserved N-terminal sequence pressed enzyme was sequestrated into inclusion bodies encoding a dinucleotide cofactor bindingsite (7). Functional requiring urea treatment forsolubilization. Purificacomplementationstudies with isolateddesaturase genes from tion to homogeneity was subsequently performed aon Rhodobacter, Erwinia, and Anacystis indicated that thedesaDEAE-cellulose column and by SDS-polyacrylamide gel electrophoresis. The purification scheme allowed turases possess a wide functional diversity in terms of their reaction products which reflect the number of desaturation the isolation of 5.3 mg of homogeneousdesaturase protein from100 ml of E. coli cell suspension. On SDS- steps performed (9). The desaturase enzyme in Erwiniu urepolyacrylamide gel electrophoresis an apparent molec- dovoru catalyzes the desaturation of phytoene to lycopene and ular mass of 56.2 kDa was determined.An antiserum is encoded by one gene. raised against phytoene desaturase cross-reacted with The purification of phytoene desaturase has been hindered by the the expressed protein and was employed to monitor thepractical difficulties associated with assayingthe enisolation steps. Upon removal of urea, desaturase ac- zyme and rapid loss of activity upon separation from its tivity was restored. The isolated desaturase catalyzedmembrane environment. As a consequence, the biochemical the conversionof 15-cis-phytoene to trans-lycopene as properties of the enzyme remain poorly understood. Typically, well as to bisdehydrolycopene. FAD was involved in properties must be assumed from crude cell-free systemsusing desaturation, whereas NAD and NADP were inhibisubstrates common to terpenoid enzymes or predicted from tory. Thisis the firsttime that a membrane-integrated gene sequences. Using a combination of molecular biological carotenogenic enzyme has been purified and finally and biochemical techniques the crtl gene has been overexobtained in an active state. pressed in E. coli and the resulting protein purified to homogeneity and subsequently reactivated. Carotenoids constitute one of the most widespread classes of pigments found in bacteria, fungi and photosynthetic organisms (1, 2). As carotenoids are isoprenoids they share a common biosynthetic pathway with related compoundsuntil the condensation of two molecules of geranylgeranyl pyrophosphate forming the first carotene, phytoene. Phytoene desaturase is the subsequent enzyme responsible for the desaturation of colorless phytoene to other colored unsaturated carotenes. Phytoene desaturase is universally a membraneintegrated enzyme amongthe carotenogenic systems reported (3). The enzmye also shows feedback regulation by subsequent carotenes (4, 5) and is a prime target site forbleaching herbicides (6). *This workwas supported by the Bundesministerium fur Forschung und Technologie and by a travel grant (to P. D. F.) from Deutscher Akademischer Austauschdienst. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 5 Present address: Dept. of Biochemistry, Royal Holloway and Bedford New College, University of London, Egham, Surrey TW20 OEX, UK. 11 To whom correspondence and reprint requests should be addressed. Tel.: 07531-883668: Fax: 07531-883042.

MATERIALS AND METHODS

Growth of Organisms-E. coli JMlOl harboring pUCCRT-I was grown overnight in LB medium in the presence of ampicillin (0.1 mg/ ml) according to Ref. 10. Cells were harvested by centrifugation and resuspended in fresh medium. The culture was then incubated for a further 4 hat 37 “C. The C5 mutant of Phycomyces blakesleeanus was grown as described (4) andharvested after 4 days. Construction of pUCCRT-I-pUC18 (11) was digested with KpnI and SphI, treated with Klenow enzyme, and ligated with SphI linker (pGGCATGCC) creating a vector (pUC18SP) in which the ATG of the SphI site is in-frameof lacZ’. pCARl6 (12) was digested with SnaBB andEcoRV, a 1.87-kilobase SnaIB (3497)-EcoRV(5363) fragment carrying the E. uredovora crtl gene wasisolated and theninserted intothe SmaI siteof pUC18 with the same orientation as in lacZ’. This plasmid was then digested with XbaI and SphI, and treated with exonuclease 111, mung bean nuclease, Klenow enzyme, and T4DNA ligase, yielding plasmid pCRT-I which contains a 1.71-kilobase fragment carrying the intactcrtl gene. pCRT-I was digested with BarnHI and HindIII, and a 1.57-kilobase BamHI (3652)-Hind I11 fragment carrying the truncated crtl gene was isolated. This fragment was ligated with a synthetic fragment from the cohesive end for SphI containing the crtl initiation codon up to that of the BamHI site (3652) and with a SphlHindIII-digested pUC18SP treated with alkaline phosphatase creating a hybrid plasmid pUCCRT-I. Isolation and Purification of Recombinant Phytoene Desaturase from E. coli-Cells were harvested by centrifugation at 10,000X g for

19891

Recombinant E. uredovora Desaturase Phytoene

19892

5 min at 4 "C. Typically, 100 ml of culture was used and resulted in 1.2 g wet weight of cells. Pelleted cells were resuspended in freshly prepared buffer comprised of 30 mM Tris-HC1, pH 8.0, 20%, w/v, sucrose, 1 mM EDTA, and 1 mg/ml lysozyme (10 ml). The resuspended material was placed on ice for 10 min and then centrifuged at 12,000 X g for 15 min 4 "C. The supernatant (periplasmic fraction) was discarded. The cells were resuspended in 0.1 M Tris-HC1, pH 8.0, containing 1 mM EDTA and subsequently broken open by freezethawing with liquid nitrogen. This breakage procedure was repeated three times. DNase (10 pg) was added to the suspension, the suspension incubated for 30 min and then centrifuged at 12,000 X g for 10 min at 4 "C. Thesupernatant (cytoplasmicfraction) was discarded andthe pelleted membrane fraction used in subsequent inclusion body preparation. Typically 281 mg of membranes resultant from a 100-ml culture. The membrane fraction was resuspended in 50 mM Tris-HCI, p H 8.0 containing 100 mM NaCl, 1.0%, w/v, Triton X-100, and 10 mM EDTA and incubated on ice for 15 min prior to centrifugation at 12,000 X g for 15 min at 4 "C. The membranesolubilisate was discarded and the insoluble material (inclusion bodies) washed with distilled water.Typically, a 100-ml culture yielded 162 mg of inclusion bodies. This material was either stored solid at -20 "C or used immediately. Isolated inclusion bodies were solubilized with 8 M urea present in 25 mM Tris-HC1, pH 8.0, 1 mM DTT.' After a short incubation period the solution was centrifuged at 12,000 X g, 4 "C for 5 min to remove insoluble debris, thesupernatant containing solubilized phytoene desaturase was then chromatographed on DEAE-cellulose. The solubilisate was adsorbed onto a DE-52 column in running buffer consisting of 25 mM Tris-HC1, pH 8.0, containing 1 mM DTT, 1mM EDTA, and 8 M urea. After washing the adsorbed proteins were eluent with a 0-300 mM NaCl linear gradient. After development of the gradient, the column was washed with 1 M NaCI. A flow rate of 1 ml min" was used and the eluentmonitored continuously at 280 nm using a UV detector. Typically fractions of 4 ml were collected and phytoene desaturase reproducibly eluted in fraction 12 equivalent to 120 mM NaCl. Reactivation was achieved by removal of urea using gel filtration. The purified sample was loaded onto a PDlO column equilibrated in 0.4 M Tris-HC1, pH 8.0, containing 5 mM DTT. The column was developed in this buffer, and the protein was collected from the void. D T T was added to the enzyme solution from a 1 M stock solution until a final concentration of 10 mM DTT was reached, and the sample was left on ice for 30 min prior to assaying.

ison with the staining intensity of standard bovine serum albumin protein (2, 5, and 10 pg) in asimilar manner to that previously reported by Sasiak and Rilling (15). Production of Antiserum-For formation of antisera against the crtI protein, the membrane fraction from overexpressing E. coli was subjected to SDS-PAGE on a 10% polyacrylamide gel. After cutting the stained bandof the crtI protein from the slab gel and electroelution, electrophoresis was repeated on a 10% gel containing 4 M urea. This resulted in a pure protein exhibiting a single band. About 100 pg of the electroeluted protein was emulsified with Freud's adjuvant and used for immunization of a rabbit. After 3 weeks and also after another 2 weeks the rabbit was boostered each time with another 80 pg of protein. Two weeks after the last immunization the rabbit was bled and the serum collected by centrifugation of the blood at 3000 X g for 15min. Western blotting was carried out asdescribed previously (16). Enzyme Activity Assay-Phytoene desaturase activity was determined using the P. blakesleeanus C5 mutant in a coupled assay procedure as described previously (17). The assay mixture contained in 0.5 ml 0.4 M Tris-HCI, pH 8.0, buffer, 5 mM DTT, 1pmol of NAD, 1 pmol of NADP, 1 pmol of FAD, 3 pmol of ATP, 6 pmol of MgC12, 4 pmol of MnCI2, and 0.25 pCi of 3(R)-[2-"C]mevalonic acid converted to its Naf salt with 0.01 N NaOH prior to addition. To this assay mixture 200pl (20 pgof reactivated enzyme) was typically added. Incubations were terminated after 2 h with the addition of methanol (1.5 ml). This incubation mixture can be stored at -20 "C or extracted immediately. Radioactive isoprenoids were partitioned into 10% diethyl ether in light petroleum (b.p. 40-60 "C). The radiolabeled carotenes were separated by HPLC on a 3-pm Nucleosil CIR column usingethyl acetate/acetonitrile/water, 35:60:5 (v/v/v),as eluent and their radioactivitydeterminedon-line with aRamona radioactivity detector (Raytest, Straubenhardt, Germany). The absorbance of added standard carotenes was recorded with a programmable Jasco UV/visible photospectrometricdetector model 8201. Bisdehydrolycopene and cisltrans isomers of lycopene, {"carotene, phytofluene, and phytoene were isolated as reference compounds as described previously (9, 18). RESULTS

Plasmid Construction and Protein Expression-The constructed plasmid pUCCRT-Iencodes the full crtI polypeptide with a n additional 9 amino acids at itsN terminus originating from the pUC18polycloning sites as indicated.

Polycloning sites

the beginning of crtI

puc18 1

2

3

4

5

6

7

8

9

Thr Met Ile Thr Asn S e r Ser S e rG l y ATG ACC ATG ATT

ACG AAT TCG

10

11

1213

Met LysProThr

AGC TCG GGC ATG AAA CCA ACT

... ... ...

EcoRI

Sa cI

Polyacrylamide Electrophoresis-Using the procedure of Laemmli (13) SDS-PAGE was used to monitor purification procedures and determination of molecular mass. Protein standardsused for calibration were myosin (205 kDa), P-galactosidase (116 kDa), phosphorylase b (97.4 kDa), albumin bovine (66 kDa), and albumin egg (45 kDa). The relative mobility for each protein was determined and themolecular mass of the recombinantprotein estimated from a standard calibration curve. Determination of Protein Concentration-Protein concentrations were determined by the method of Lowry et al. (14). When appropriate protein concentrations were estimated from SDS-PAGE by comparThe abbreviations used are: DTT, dithiothreitol; PAGE, polyacrylamide gel electrophoresis; HPLC, high performance liquid chromatography.

SDS-PAGE analysis of isopropyl-1-thio-@-D-galactopyranoside-inducedcells harboring the plasmid pUCCRTI indicated the presenceof a heavily stained band corresponding to a molecular mass of 56 kDa (Fig. 1, lane 2). This band was not present to the same degree in uninduced cells and totally absent in cells devoid of pUCCRTI. The predictedmolecular mass of the native protein from the gene sequence is 55 kDa (12). The slightly higher molecular mass of our overexpressed phytoene desaturase can be attributed to the additional 9 amino acids linked to theN terminus of the crtI sequence to facilitate cloning with pUC18. As the expressed protein migrated to a position predicted from the crtI gene construct, it was identified as the E. uredovora phytoene desaturase. Den-

Recombinant E. uredovora Phytoene Desaturase 20511697.4-

664529 -

-

L

-66

”-

-

-

19893

-P-

66 -PD -45 -34

- ”

1 2 3 4 5 6 7 8

FIG. 1. SDS-polyacrylamide gel electrophoresisof fractions obtained during isolationof the recombinant phytoene desaturase. Fractions after each step in the purification were electrophoresed on a 10% polyacrylamide gel and stained for proteins. Lanes 1 and 8, molecular mass markers (kDa); lane 2, induced cells, 15 pg of protein; lane 3 , isolated membranes, 15 p g of protein; lane 4, supernatant after detergent treatment of membranes, 15 pg of protein; lune 5 , inclusion bodies, 10 pg of protein; lane 6, urea-solubilized inclusion hodies, 7 pg of protein; lune 7, purified recombinant phytoene desaturase from DEAE column, 5 pg of protein. The arrow indicates the location of the recombinant ph-ytoene desaturase during isolation.

Fraction no

FIG.2. Chromatography of urea-solubilized inclusion bodies on DEAE-cellulose. Solubilizedinclusion body material (approximately 16 mg of protein in 3 ml of buffer) was located onto the column and the bound proteins eluted with a linear NaCl gradient (0-300 mM) as indicated by the broken line. Fractions of 4 ml were typically collect.ed. The continuous line indicates the absorbance at 280 nm of eluted protein. Inset shows analysis of fractions 3, 4, 5 , and 8 by SDS-polyacrylamide gel electrophoresis (see “Materials and Methods”). The equivalent to 10 pl of each fraction was applied to the gel.

r’

sitometric comparisons with stained protein standards sugbm gested that the recombinant protein constituted about 10% of the total cellular protein. This level of expression is comparable to numerous examples of other recombinant proteins (see Refs. 19 and20 for review). Solubilization and Purification of the Recombinant Phytoene Desaturase-Fractionation of cell homogenates and subse4 5 quent separation by SDS-PAGE illustrated that the recombinant protein was localized in the particulate membrane , fraction (Fig. 1, lane 3 ) . Washing of the membranes with the nonionic detergent Triton X-100 ( l % , w/v)released about 1 2 3 4 5 10% of the recombinant protein(Fig. 1, lane4 ) . The remaining insoluble material after such detergent treatment is typiFIG.3. Western blot analysiswithan antiserum raised cally termed theinclusion bodies. It is within these inclusion against the recombinant protein. Lanes 1-4 contain the purified bodies that the recombinant phytoene desaturase was pre- protein a t concentrationsranging from 0.5,1.5, 3.0, and 6.0 pg, respectively. Lane 5 shows the blottedmolecular mass markers (kDa). dominantly (90%) sequestrated (Fig. 1, lane 5). The isolated inclusion body fraction was subjected to a variety of nonionic homogeneous protein, as judged by the single band observed and ionic detergent treatments, adjustment to an alkaline on SDS-PAGE even when overload (Fig. 2, lane 6). As exenvironment, pH 10, and exposed to the action of lipases. pected from the gene construct, the pure protein had a disNone of these approaches released the recombinant protein sociated molecular mass of 56.3 kDa, some minor bands of (datanotshown). Effectivesolubilization was, however, lower molecular mass of around 35 kDa were visualized after achieved using the denaturant urea aatconcentration of 8 M. overloading of the gel with the enriched fraction enhanced and At this concentration of urea, the recombinant protein was after the protein was electroeluted. The antiserum also recsolubilized in a nonselective manner (Fig. 1, lane 6). ognized these faint bands a t a lower molecular mass, illusUrea-solubilized inclusion body material was applied to a trating the presence of degradation products. The overall DEAE-cellulose column and eluted using the conditions de- purification procedure is summarized in Table I, indicating scribed under “Materials and Methods.” The presenceof 8 M that from a 100-ml E. coli culture, 5.3 mgof homogeneous urea in the buffer did not appear to affect chromatography protein can be obtained after 10-fold enrichment with a 30% with most of the protein binding to the anion exchanger. The recovery. profile of the eluted proteinrecorded from the absorbanceat Reactivation of Desaturase Activity-Desaturase activity 280 nm is illustrated in Fig. 2. The principal peak of protein was assayed using a coupled assay procedure (17) recently eluted a t a salt concentrationof 100 mM. Subsequent analysis developed for expressed carotenogenic enzymes (21). Crude of fractions by SDS-PAGE indicated that the recombinant extracts exhibitedpoor in vitro activity (8%conversion) which protein was responsible for this major protein peak (Fig. 2, was lost during purification as aconsequence of the urea inset). treatment required to solubilize the dense protein aggregates Purification was also monitored throughout using an anti- that exist withininclusion bodies. In order toregain activity, serum. Cross-reactivitybetween the purified protein and anti- it was necessary to remove the denaturant,allowing refolding serum is illustrated in Fig. 3. The antiserum exhibiteda good of the protein to occur. Two methods, dialysis and passage titer detecting0.5 Kg of protein after a 1:lOOO dilution. A good through a PDlO column, were employed to remove the urea specificity was also observed,as the preimmune serum exhib-(“Materials and Methods”). For optimal reactivation it was ited no cross-reactivity with the purified protein even when necessary to dilute(x 2) the sample and maintain presence the loaded at high concentrations (6-10 pg; data not shown). of D T T a tall times. After incubation of the reactivated protein with [I4C]phyThe purification protocol was reproducible and yielded a

..

c‘-66

I

‘+I

Recombinant E. uredovora Desaturase Phytoene

19894

TABLE I Purification of the recombinant E. uredovora phytoene desaturase Step

Volume

Cells Membranes Inclusion bodies Solubilized inclusion bodies DEAE fraction

100 10 10 10 8

Total

Phytoene desaturase proteinb

Relative phytoene desaturase content

Specific activity

76

dprnlpg protein

Recovery

%

180 100 78 50 53

17.7 12 10 8.3 5.3

'

desaturase, it was necessary to remove all endogenous cofactors from theC5extract by passage through a desalting column. Using acombination of different nucleotide cofactors as shown in TableI1 it was evident thatFAD wasthe essential electronacceptorduringthedesaturation of phytoeneto lycopene. The pyridinenucleotide cofactors NAD and NADP both had a similar inhibitory effect on desaturation. In addition, the position of isomerization within the sequence was observed occurring at phytoene, which like the desaturation reaction exhibited a dependence on FAD.

10 12 13 17 100

12

3,723' (273-fold) Protein determined using the method of Lowry et ai. (14). *Estimated from the intensity of staining on SDS-PAGE using a bovine serum albumin calibration (2, 5, 10 fig). ' After reactivation.

100 68 56 47 30

a

'L754L25&350hm

DISCUSSION

This article reports the isolation of phytoene desaturase from E. uredovora after overexpression in E. coli. Expression 10 20 30 was mediated by the construct pUCCRT-I which enabled Retention time ( m i d direct expression of the protein to a level constituting about FIG. 4. HPLC separation of radiolabeled carotenes formed 10% of the total cellular protein. The recombinant protein b y the reactivated phytoene desaturase after isolation. The was the principal component of the insoluble inclusionbodies. upper trace shows the position of authentic marker carotenes required for identification. 1, bisdehydrolycopene; 2, all-trans-lycopene; 3, all- Such sequestration is typical of many expressed systems (19, the formation of inclusion trans-["carotene; 4 , all-trans-phytofluene; 5 , 15-cis-phytofluene. 20, 22). The precisereasonfor bodies is not known with any certainty, although it hasbeen Lower trace illustrates the corresponding radiolabeled carotenes. suggested that they ariseas a consequence of protein exceeding its solubility in the cell (23). Unfortunately, their formatoene produced in situ, the reaction products formed were tion is associated with insolubility, inactivity, and denatured separated from thesubstrate by HPLCandradioactivity protein.Suchcharacteristics were alsoobserved withthe detected on-line as shown in Fig. 4. The top trace indicates system of expression used in our work. The only effective the position of expected caroteneproductsintheHPLC solubilizing agent being the denaturant urea. profile. Controlexperimentswithoutaddedprotein(not Despite the denaturing conditions, a rapid andreproducible shown) exhibit the peaks a t 32 min of the substrate 154sfor the purificationof the recombinant method was developed phytoene together with two conformations of prenyl alcohol at 4 min and squalene at 18 min. After incubationradioactive phytoene desaturaseto homogeneity. The major advantage of labeled all-trans-lycopene (15 min), all-trans-{-carotene (21 this overexpressingsystem was the high quantity of pure min), and all-trans-phytofluene were found (Fig. 4, bottom protein obtained from a relatively small amount of biological of pure protein achieved from trace). Under optimal conditions small amounts of bisdehy- starting material. The amounts this protocol are ideal for both polyclonal and monoclonal drolycopene with six additional double bonds appeared occawill make crystallizationa promising sionally. Lycopene with four double bonds additional to phy- antibody production and toene was typically the principal desaturase product consti- possibility, especially as reactivation of the enzyme has been achieved. The conditionsrequired for reactivation are similar tuting 60% of the total reaction products. The desaturase intermediates phytofluene (one additional double bond) and to those determined for otherenzymes, e.g. fumarase (24). It is difficult to quantify thedegree of functional refolding {-carotene (two additional double bonds) were also detected and constituted 27 and 12% of the total desaturase products achieved during reactivation, but judging by the purification formed, respectively. Due to the possibility of reactivating the fold calculated from activity comparedwith that obtained for purified enzyme,a 273-fold purification can be calculated relative desaturase content (Table I), it would appear to be substantial.Thiscorrelated with recentcomplementation from thespecific activities of phytoene desaturase in the crude membrane fractionwhich was 12 dpm/pg of protein and3,273, studies whereby the level of carotenes produced were lower than that expectedfor an overexpressedsystem (9). It is respectively, in the reactivatedpurified enzyme (Table I). exThe geometric isomers formed were elucidated by co-chro- feasible that the only functionallytranslatedprotein matography with authentic standard identified by their rela- pressed is the relatively low percentage released from membranes by nonionic detergent, with the remaining protein tivepolaritiesandspectralproperties. All thedesaturase residing within inclusion bodies and being inactive. products showed a trans configuration incontrasttothe The relative amounts (lycopene as the major product) and substrate phytoenewhich was supplied in a 15-cis form. Cofactor Requirements for Desaturation-In order to deter- geometric isomers (all-trans for each product and intermedimine the nucleotide cofactors required for the E. uredouora ate) formed by the recombinant phytoene desaturase(Fig. 4) I

I

I

Recombinant E. uredovora Phytoene Desaturase

19895

TABLEI1 Formation of desaturation products from14C-labeled 15-cis-phytoene in Distribution Addition of cofactors

in

incorporation Radioactivity

cis-Phytoene tram-Phytoene

Lycopene" cis-Phytoene tram-Phytoene Total

Lycopene" %

dpm

FAD NAD NADP

2,856 44,867 50,617

3,383 9,447 3,850

12,373 7,405 7,128

18,617 66,659 61,595

15 75 82.2

18 14.2 6.3

66 11 11.6

NAD/NADP FAD/NADP FAD/NAD FAD/NAD/NADP

28,874 10,621 16,792

2,364 19,733 16,136

6,195 7,766 12,246

37,433 38,120 39,174

77 27.8 28

6.3 51.7 41.2

16.5 20.3 31.3

4,544

12,975

6,165

23,684

19.2

55

26

No cofactors

34,447 38.896

0

5,824 0

40,271 0

86

0

14

C5 alone

0

" Including bisdehydrolycopene.

are consistant with those found accumulating inE. coli con- suggest that FAD is a dissociable moiety and not covalently taining the Erwinia crtl gene complementated with the Er- bound to theenzyme. By using a combination of molecular biological and biowinina genefor phytoene formation (9). The formation of bisdehydrolycopene suggests that desaturation in Erwinia is chemical techniques, the isolation of the Erwinia phytoene large quantities of not that specific with regard to its general in vivo terminal desaturasehas beenachievedyielding reactivated enzyme. Although the authenticity of expressed product. Presumably the cyclase is in close proximity to the desaturase in vivo, and transfer of lycopene is very efficient. proteins is often questioned, the recombinant desaturase proDespite thesupply of the 15-cis-phytoene isomer as substrate vides an ideal opportunity toelucidate its properties, enabling to therecombinant desaturase,all subsequent products areof valuable comparison between otherphytoenedesaturases, especially those of the cyanobacterial/higher plant type (9). a n all-trans configuration, suggesting that isomerization ocREFERENCES curs initially to the phytoene molecule, supporting the view that it isa dynamic effect between the enzyme, phytoene, and 1. Goodwin, T.W. (1983) Biochem. SOC.Tram. 11,473-483 2. Bramley, P. M. (1985) Adu. Lipid Res. 2 1 , 243-279 perhaps the bound cofactor (1).Further evidence supporting 3. Sandmann, G. (1991) Physiol. Plant. 8 3 , 186-193 4. Bramley, P. M., and Davies, B. H. (1975) Phytochemist 14,463-469 t h e initial isomerization of phytoene prior to desaturation was 5. Sandmann, G., and Kowalczyk, S. (1989) Biochem. Biopyys. Res. Commun. obtained during the determination of cofactorswhere all1 6 3 , 916-921 6. Sandmann, G., and Boger, P. (1989) in Target Sites o Herbicide Action trans-phytoene was apparent as an intermediate and accu(Boger, P., and Sandmann,G., eds) pp. 25-44, CRC #res, Boca Raton, FL mulated when desaturation was inhibited by the pyridine D., Pecker, and Hirschberg,J. (1991) Plant Mol. Biol. 1 6 , 967nucleotides used. The presence of one protein on SDS-PAGE 7. Chamovitz, 974 G. E. Viitanen, P. V. Pecker I., Chamovitz, D., Hirschberg, J., having a molecular mass of 56 kDA indicates that this protein 8. Bartley, and Scolnik: P. A. (1991) Pr& Natl. h a d . Sci. U. S. A. 88,6532-6536 is solely responsiblefor desaturationinErwinia,andno 9. Linden, H., Misawa, N., Chamovitz, D., Pecker, Hirschberg, J., and Sandmann, G. (1991) 2. Naturforsch. 46c, 1045-1051 separate isomerase exists. 10. Maniatis, T.,Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloni Laboratory Manual,Cold Spring Harbor Laboratory, Cold Spring Ha%$ The requirement for FAD indicates that the Erwinia phyx1v toene desaturase reactionis a classical dehydrogenation proc- 11. Yanish-Perron, C., Vieira, J., and Messing, J. (1985) Gene (Amst.) 33,103119 ess. This is the first direct proof with an isolated phytoene 12. Misawa, N., Naka awa, M , Kobayashi, K., Yamano, S., Izawa, Y., Nakadesaturase, in contrast to our results with membranes from mura, K., and Arashima, K. (1990) J. Bacteriol. 172,6704-6712 K. (1970) Nature 227,680-685 13. Laemmli, the cyanobacterium Anacystis where phytoene desaturation 14. Lowry, 0.U. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 showed a dependence on NAD or NADPH (5). As demonK., and Rilling, H. C. (1988) Arch. Biochem. Biophys. 2 6 0 , 622strated recently, theAnacystis phytoene desaturase gene and 15. Sasiak, 627 its product arecompletely different to the onefrom Erwinia 16. Schmidt. A,. Sandmann. G.. Armstrone. G. A,. Hearst.. J. E... and Boeer. P. ( 1 9 8 9 j ~ uJ~. .Biocheh. i84,375-3178 (9). Depending onbiochemical criteria, a bacterial/fungal and 17. Sandmann, G., and Bramle , P M (1985) Planta 164,259-263 18. Ernst, S., and Sandmann, d (1988')Arch. Microbiol. 150,590-594 a cyanobacterial/higher plant phytoene desaturase type could 19. Marston, F. A. 0. (1986) Biochem. J. 2 4 0 , 1-12 be discriminated. Obviously, both types need different cofac- 20. Marston, F. A. O., and Hartley, D. L. (1990) Methods Enzymol. 1 8 2 , 2669 7 Ators ashydrogen acceptors. The inhibitionof phytoene desat- 21. Fraser, P. D., and Sandmann, G. (1992) Biochem. Biophys. Res. Commun. 185.9-15 uration carried out by the Erwinia enzymes by the pyridine 22. Schoemaker, J. M., Brasnett, A. H., and Marston, F. A. 0. (1985) EMBO nucleotides NAD and NADPH probably results from their -.I . A-,775-7RO . .- . K. E., Berg, T. F., Strickland, T. W., Fenton, D. M., Boone, T. C., competition withFADfor thecofactor-binding site. Such 23. Langley, and Wypych, J. (1987) Eur. J. Biochem. 163,313-321 findings as well as the exclusion of FAD during reactivation 24. Kelly, S. M., and Price, N. C. (1991) Biochem. J. 2 7 5 , 745-749 1 .

1

I.

"