Mol Gen Genet (1990) 224:431-436 DIGG © Springer-Verlag 1990

Analysis of a eDNA encoding arginine deearboxylase from oat reveals similarity to the Escherichia cob" arginine decarboxylase and evidence of protein processing Erin Bel l* and Russell L. Malmberg

Department of Botany, University of Georgia, Athens, GA 30602, USA

Received July 16, 1990

Summary. Arginine decarboxylase is the first enzyme in one of the two pathways of putrescine synthesis in plants. We purified arginine decarboxylase from oat leaves, obtained N-terminal amino acid sequence, and then used this information to isolate a cDNA encoding oat arginine decarboxylase. Comparison of the derived amino acid sequence with that of the arginine decarbox- ylase gene from Escherichia coli reveals several regions of sequence similarity which may play a role in enzyme function. The open reading frame (ORF) in the oat eDNA encodes a 66 kDa protein, but the arginine decar- boxylase polypeptide that we purified has an apparent molecular weight of 24 kDa and is encoded in the car- boxyl-terminal region of the ORF. A portion of the eDNA encoding this region was expressed in E. coIi, and a polyclonal antibody was developed against the expressed polypeptide. The antibody detects 34 kDa and 24 kDa polypeptides on Western blots of oat leaf sam- pies. Maturation of arginine decarboxylase in oats ap- pears to include processing of a precursor protein.

Key words: Oats - Arginine decarboxylase - Polyamines - Protein processing - DFMA binding

Introduction

Polyamines are ubiquitous in living organisms; they have been implicated in several important cellular pro- cesses, including nucleic acid structure and replication, cell division, and protein synthesis (Tabor and Tabor 1984). All organisms studied synthesize the polyamine precursor putrescine from ornithine via ornithine decar- boxylase (ORNdc). Both plants and bacteria, however, have a second pathway for putrescine biosynthesis: from arginine via arginine decarboxylase (ARGdc;

* Present address." Department of Biochemistry and Biophysics, Texas A & M University, College Station, TX 77842, USA

Offprint requests to: R.L. Malmberg

E.C. 4.1.1.19) through agmatine (Tabor and Tabor 1984). In plants, each pathway apears to have specific regulated roles in growth and development (Slocum et al. 1984). For example, ORNdc is required for the growth of tobacco cell cultures (Hiatt et al. 1986) and in the early development of tomato fruit (Cohen et al. 1982). ARGdc plays a role in somatic embryogenesis in carrot (Feirer et al. 1984) and its activity is induced in the response of plants to environmental conditions such as osmotic (Flores and Galston 1982) or mineral stresses (Young and Galston 1984). These results suggest that complex regulation of polyamine biosynthesis exists in plants. We report here the results of analysis of oat (Arena sativa) ARGdc at the molecular level. This in- cludes the first isolation of the cDNA from a eukaryotic organism.

Materials and methods

Assay for ARGdc enzyme activity. Samples from protein extracts prepared as described below were diluted to 100 ~tl and mixed with 100 ~tl of reaction buffer [200 mM HEPES, pH 7.4, 1 mM EDTA, 10raM dithiothreitol (DTT), 0.02% bovine serum albumin (BSA), 1 mM pyri- doxal phosphate] and 1 nmol of 14C-arginine (0.3 Ci/ mmol). The reaction vessel was sealed with a paper filter containing 30 gl methylbenzethonium hydroxide (Sig- ma), and incubated at 25 ° C for 30 min. One hundred microlitres of 1 M KH2PO4 was added, and incubation continued for 1 h at 25 ° C. The filter was then counted in scintillation fluid to determine the amount of 14C02 trapped.

Isolation of protein. Nine-day-old oat seedlings were ex- cised, cut into short segments, and ground in a blender in 25 mM TRIS-HC1, pH 7.4, 40 gM pyridoxal phos- phate. Homogenized material was filtered through one layer of Miracloth (Calbiochem); 243 gin/1 (NH4)2SO~ was added, and the pellet discarded after centrifugation. An additional 168 gm/1 (NH4)2SO ~ was added, with

432

ARGdc activity found in the pellet following centrifuga- tion. This pellet was resuspended in and dialyzed against 25 mM TRIS-HC1, pH 7.4. The sample was further frac- tionated using a preparative isoelectric focusing appara- tus (Rotofor, BioRad). The sample was brought to 1% ampholytes (Biolyte 3-10, BioRad) and processed ac- cording to the manufacturer's recommendations. Frac- tions were assayed and active fractions pooled and run through the apparatus a second time without the addi- tion of further ampholytes. Active fractions from this run were pooled and concentrated using 430 gm/1 (NH4)2SO,~.

Binding of difluoromethylarginine (DFMA) to the protein was in 25mM TRIS-HC1, pH 7.4, 0.1 mM EDTA, 2 .5mM DTT, 70riM pyridoxal phosphate; 0.54 mg of protein and DFMA to 0.8 mM were added to the unlabelled reaction, which was incubated at 25 ° C for 1 h. Eight micrograms of protein and 0.5 gCi 3H- DFMA (32 Ci/mmol) were added to the labelled reac- tion, which was incubated at 25 ° C for 30 min. A further 22 gg of protein and unlabelled DFMA to 0.7 mM were then added, and the incubation continued for an addi- tional 30 min. Migration of the ARGdc-DFMA com- plex in SDS-PAGE (12% acrylamide) was determined using NCS solubilizer (Amersham) and liquid scintilla- tion counting to identify the gel slice containing 3H- DFMA. Protein in the slice of interest was eluted electro- phoretically in 50 mM NH4HCO3, 0.1% SDS. Further purification was achieved on a C8 reverse phase HPLC column, using a gradient of 0%-80% acetonitrile in 0.1% trifluoroacetate over 60 min at a flow rate of 200 gl/min. Purified protein was sequenced by the Uni- versity of Georgia Molecular Genetics Facility. A mix- ture of 17-mer oligonucleotides was synthesized by the same facility.

Isolation and analysis of a cDNA clone. RNA was iso- lated according to the method of Chirgwin et al. (1979), using 9-day-old oat seedlings which were frozen and pul- verized in liquid N 2. Poly(A) + RNA was isolated ac- cording to Davis et al. (1986), with SDS omitted from all buffers, cDNA was synthesized using a kit (Pharma- cia), and inserted into EcoRI-digested, dephosphorylat- ed 2gtl0 arms (Promega). The resulting library was screened with the mixed oligonucleotides using the tetra- methylammonium chloride hybridization procedure of Jacobs et al. (1988), except that the final wash was in 4 x S S C ( l x S S C is 0.15M NaC1, 15raM sodium ci- trate), 0.1% SDS. The cDNA isolated was subcloned in both orientations into the plasmid pTZ19R (USB). Deletion clones were generated using ExoIII/Mung (Stratagene), and both strands were sequenced using Se- quenase (USB).

Isolation of ARGdc-trpE fusion protein and production of antibodies. The 600 bp SalI-EcoRI fragment from pADCI was isolated and cloned into the SalI and EcoRI sites of the trpE fusion vector pATH23 (Koerner et al. 1990), with expression of the fusion protein as described by the same authors. Fusion protein was visualized on Coomassie stained SDS gels, and eluted as above. One

hundred micrograms of fusion protein was mixed with 1.5 vol. Freund's complete adjuvant and injected subcu- taneously into laying hens. A duplicate injection with incomplete adjuvant was performed 14 days later. IgY was isolated from egg yolks 21 days after initial injection according to Polson et al. (1985).

Western blots. Ten-day-old oat seedlings were processed according to the method of Cordonnier et al. (1985) in order to minimize protein degradation during harvest. The leaves were frozen in liquid N2, then pulverized and lyophilized without being allowed to thaw. Lyophi- lized powder was added to the modified SDS sample buffer, preheated to 100 ° C, at 45 mg powder/ml buffer and boiled for 5 rain. The sample was spun to pellet- insoluble material before separation in SDS gels. The resolved protein was transferred to nitrocellulose using standard Western blot methods (Evans et al. 1988). Anti- bodies were diluted in 10raM TRIS-HC1, pH 7.4, 150 mM NaC1. The primary antibody was used at 5 gg/ ml, the secondary antibodies were either rabbit anti- chicken or rabbit anti-mouse alkaline phosphate conju- gates (Sigma) and were diluted 1 : i 000.

In vitro transcription and translation. Five hundred nano- grams of pADC2 was digested with PvuI, phenol-ex- tracted, and incubated for 60 min at 37 ° C in a transcrip- tion reaction containing 40mM TRIS-HC1, pH 7.5, 6 mM MgC12, 2 mM spermidine, 10 mM NaC1, 10 mM DTT, 0.5 mM each NTP, and 20 units T7 RNA poly- merase. The reaction was then phenol-extracted and in- cubated with wheat germ extract (Amersham) according to the manufacturer's recommendations, using 35S-me- thionine as the radioactive amino acid label.

Results

Purification of ARGdc

Young oat leaves were chosen as the source of protein because of the high levels of ARGdc in this tissue. Our purification of ARGdc relied, in part, on the interaction of the protein with difluoromethylarginine (DFMA), a specific, enzyme-activated, irreversible inhibitor of ARGdc (Kallio et al. 1981 ; Bitonti et al. 1987). The gift of a small amount of 3H-DFMA permitted us to locate the ARGdc-DFMA complex following SDS-PAGE. In an initial experiment, migration of the labelled complex in SDS gels was determined by assaying radioactivity in 5 mm slices of a minigel. One slice, located between molecular weight markers of 31 kDa and 21.5 kDa, had significantly more counts than the other slices. For pre- parative isolation of the protein, a marker lane contain- ing 3H-DFMA-ARGdc was run and I mm slices in the 21.5-31 kDa range were assayed; the corresponding un- labelled slice was then excised and protein was eluted as described in the Materials and methods. The isolated protein migrated as a single band of approximately 24 kDa. However, sequence analysis indicated that more than one polypeptide was present in the sample, so fur-

433

ATG GCC AAG AAC TAC GGC MET A l a Lys Asn Tyr G l y

CTC TGC GTC AGA ATC TAT Leu Cys Va l A rg I le Ty r

AGC GCC GAT GGC ACG GGG Ser A l a Asp G l y Thr G l y

TCG CTC CAC ACC GCA TTC Ser Leu H is Thr A l a Phe

GTG AAC CAG CAC AAG GAC Val Asn G in H i s Lys Asp

AAG CCA GAG TTG CTG ATT Lys Pro GI~ ke~ Leu I le

GAC TCG GCT TAT GTC GCG Asp Ser A l a Tyr Va l A l a * * *

GAG CTG GAC ATC GTC ATC Glu Leu Asp I le Va l l i e

AAG ATA CCG GGC CAT TTT Lys l i e P ro G l y H i s Phe

GCC AAG AAG CTC AAG GCT Ala Lys Lys Leu Lys Ala

ACA GAC ATT GTC TTC AAG Thr Asp l i e Va l Phe Lys

ATG ACC ACG CTG GAC TGC Met Thr Thr Leu Asp Cys

GCG TAC GGG CTG GAG GAG A l a Tyr G l y ~eu Glu G lu

CCC GTT CTA TGC ACC GAG Pro Va l Leu C ~ Th~ GlU

CCC GAG CCG AAA GAT GAT Pro G lu Pro Lys Asp Asp

CAG CCG ACA GGC TTG TCC Gin Pro Thr G ly Leu Set

AAG CTC TCC AAG AGC GTC Lys Leu Ser Lys Ser Va l

CCC GAC TAC TGG GGC ATC Pro As~ ~ : T rp G l y ~ l e

ACG CTC GTC GAT GTC ACC Thr Leu Va l Asp Va l Thr * * *

CCG CTA GAT CCG AAG CTC Pro Leu Asp Pro Lys Leu

CAC AAC CTG TTC GGC GGC HI~ Asn ~eu Phe G l y G l y

CTC CTG GGA TCG ACA ACG Leu Leu G l y Ser Thr Thr

GAG CGG GCG AGA GAG AAT Glu A rg A l a A r g G lu Asn

GCT GAC TAC AAG CCC CCA A l a Asp Tyr Lys Pro Pro

a c a l c g g t t t t g g a c t

GAT GTG TAC CAT GTC GAG GGC TGG GGA Asp Va l Tyr H i s Va l G lu G l y T rp G l y

g c g a g a t a a c t c a c c t a t c GAG CCC TAC TTT GCC GTG AAC AAG GAC GGC CAC Glu P ro Tyr Phe A l a Va l Asn Lys Asp G l y H i s

GGC CGC GAG ACG CTT CCA GGG CAG GAG G l y A rg G lu Thr Leu Pro G ly G in G lu

AAG AAA CTC CAA TTC CCC ATG ATC CTT Lys Lys Leu G in Phe P ro Met I le Leu

GCC AAC GCC ATC AAG TAC ACC CAG TAC A l a Asn A l a l i e Lys Tyr Thr Gin Tyr

GTC GTG CAG GAC ATG GTT CAC TTT GGC Va l Va l G in Asp Met Va l H i s Phe G l y

GCC ATG AGC TGC CTC ACC AAA GCC AAG A l a Met Ser Cys Leu Thr Lys A l a Lys

CTC GCC CTG GCG GCG CGC GCC ATG GGC Leu A l a Leu A l a A l a Arg A l a Met G l y

ATC GAC GTG CTC TCC GTC ATC GAG CAA GCC ACG l i e Asp Va l Leu Set Va l l i e G lu G in A l a Thr

CGC TTC CCC GAT GTG CTG AGG CAC CGC ATC AAC Arg Phe P ro Asp Va l Leu A r g H i s A rg I l e Asn

GGT TCT GTC TAC CAG GGC GTG TTC CCG GTG AAG G l y Ser Va l Tyr G in G l y Va l Phe P ro Va l ~

TAC GAC CAC AGC TAC GGG CTG GAG GCA GGC TCC Tyr Asp H is Set Ty r G l y Leu G lu A l a G ly Se~

CCT GGA GCC TAC CTG GTA TGC AAC GGC TAC AAG Pro G l y A l a Tyr Leu Va~** C ~ Asn*** Gly*** Ty~** ~

CTG AAC GTC ATC ATC GTG CTG GAG ATG GAG GAG Leu Asn Va l l i e l i e Va l Leu G lu Met G lu G lu

GAG GAG AGC AGC AAG CTC GGC GTG GAG Glu Glu Ser Ser Lys Leu G l y Va l G lu

GGC TCC ACG GCC GGC AAG CAC GGT AAG G l y Ser Thr A l a G l y Lys H i s G ly Lys

CTG AAC AAG CTG CAT TGG CTC AAG CTG Leu Asn Lys Leu H i s T rp Leu Lys Leu * * *

GCA GCT AGC GAG GCC TCT GAT ATC TAC A l a A l a Ser Glu A l a Set Asp l i e Tyr

GGC GGG GGG CTG GGA GTG GAC TAC GAC G l y G ~ G l y Leu G ly Va~ Asp I ~ ; Asp

TAC GCG TCC AGC ATC GTG CAG GCG GTG Tyr A l a Ser Ser l i e Va l G in A l a Va l

AGT GGG CGC GCC ATG GCG TCA TAC CAC Ser*** G ~ A : ~ AI~** Met A l a Ser Tyr H i s

GAA GAC GAG GCC ACC ACC GAG CAG CTG Glu Asp G lu A l a Thr Thr Glu G in Leu

ATG TCC TCC CAC GCC GTG CAC ATC AAG Met Ser Ser H i s A l a Va l H i s l i e Lys

ACC ACC GAT GCC CAC ACC ATC TAT AAC Thr Thr Asp A l a H!s Thr I le Tyr Ash

CAG CAT CTG TTC CCG ATG ATG CCG GTG Gin H i s Leu Phe Pro Mel Met Pro Va l

TGC GAC AGC GAC GGC AAA G T C G A C AAG Cys Asp $er Asp G ly Lys ~ Lys

GGT GGC TAC TAT GTG GCC GTG CTC CTG G ly G l y Tyr Tyr Va l A l a Va l Leu Leu

CCG AGT CTG GTG CGC GTA GTC GGC ACC Pro Set Leu Va l A rg Va l Va l G l y Thr

GAG GAG CTC ATC GGC ACT GTG AGT TAC Glu Glu Leu I le G l y Thr Va l Ser Tyr

AAG GTG TGG GAA ATG GTG GAG AAG CTC Lys Va l T rp Glu Met Va l G lu Lys Leu

CCA ATG GCC TAG c t g g c l g t t a a t t g c t g t c t P ro Met A l a

CCC GTC ATC GGC GTC CGC GCC AAG CTG CTC ACC Pro Va l l i e G l y Va l Ar 9 A l a Lys Leu Leu Thr

TTC GGG CTA CCG GCG GAG AAG ATT TAT GAG GTG Phe G l y Leu P ro A l a G lu Lys I le Tyr G lu Va l

CTG CAC TTC CAC GTC GGC TCC ATG ATC CCG ACC Leu Hi~ Ph~ HI] Va~ Gly Se~ Met I le Pro Thr

TGC GCC CTG GTG AAG GAG TAC GGG GTG GAG ACG Cys A l a Leu Va l Lys G lu Tyr G l y Va l G lu Thr

GGG ACC CGT TCG GGC AGC TCC GAC ATG TCC GTG GI~ Th Ar Se G l y Se Se Asp Me Ser . . . . ~ , , ~ , , ~ . . . . . ~ , , ~ . . . . . ! , . , w l

CGG CTC AAG TGC GAC TAT CAC GGC GTC CCC CAC Arg Leu Lys Cys Asp Tyr H i s G l y Va l P ro H i s

TCC ATG ATC ATC CTG GAG GCG CTC TCC GCG ATC Ser Met I le l i e Leu G lu A l a Leu Ser A l a I le

CAC GGA CGG ATC CGC GAT CTC TCT TCC AAG CTG H is G l y A rg I l e Arg Asp Leu Set Ser Lys Leu

AAG CAC GGC ATC GAG ATG TAC AAG CTG GGA AAG Lys H is G l y l i e G lu Met Tyr Lys Leu G ly Lys

TAC CAC ATG AAC CTC TCG GTC TTC TCG CTA ATG Tyr H i s Met Asn Leu Set Va l Phe $er Leu Met

AGC CGG CTA GAT GAG AAG CCG ACT CAC AAG GCC Set A rg Leu Asp G lu Lys P ro Thr H i s Lys A l a

TTC ATA CGC GAC ACC GAG ACG ATG CCG CTG CAC Phe l i e A r g Asp Thr G lu Thr Met P ro Leu H is

ACC GGC GCG TAC CAG GAG GCC CTA TCC AAC AAG Th, G l y A l a Ty r Gin G lu A l a Leu Ser Asn ~ :

GGC AAT GGC GGC GCA TTC AAC GTC GAG GCG GCC G l y Asn G l y G l y A l a Phe Asn Va l G lu A l a A l a

GAG GTG AAA CAG GAT ATC AGC AGC GTG ATT GAG Asp Va l Lys Gin Asp I le Ser Ser Va l l i e G lu

GTG GAA TCC GGA CTT CAC ACC ATG CCG TAC CTC Va l Glu Ser G l y Leu H i s Thr Met Pro Tyr Leu

g c c a a g c a c t c t a c t a c t g t a g c t t c a a c g c a t a g t g t c g l a c

~cg~cc~9~g~t~gta~actaa~a~aa~c~9c~gca~a~ccca~c~9tatc~a~ta~g~c~agagtc~aaa~g~accgaaca t t g g g t a c t g t c c t g t t g g t g t a g g t g c c a g c t a g t g t t g g a g a t g a a a

Fig. 1. DNA sequence of the 2124 bp EcoRI DNA insert of the pADC2 clone. The amino acid sequence encoded is indicated below the D N A sequence. The first amino acid residue of the 24 kDa polypeptide is indicated by bold face and underlining, the amino acid sequence determined by peptide sequencing is indicated by

underlining, and the SalI site used in the construction of the fusion protein is indicated by underlining. The 9 regions of amino acid sequence homology to the Eseherichia eoli enzyme, shown in detail in Fig. 4, are indicated by triple asterisks below the appropriate amino acid segments

434

O

<

Z

Fig. 2. Northern blot analysis of poly(A) + RNA isolated from oat leaves, probed with pADC1

ther purification on a C8 HPLC column was performed, again using 3H-DFMA-ARGdc as a marker to identify the fraction of interest. Sequence analysis of this fraction yielded the amino acid sequence indicated in Fig. 1 with the amino terminus shown (the identity of the first 3 amino acids was not determined).

Isolation of a cDNA clone encoding ARGdc

A port ion of the amino acid sequence determined,- Y N Y H M N , was used to specify the composition of a mixture of 17-mer oligonucleotides with no bias in codon usage. This oligonucleotide mix was used to screen an oat cDNA library. Initially, a 1.05 kb clone (pADCI) was isolated, but hybridization of this clone to a North- ern blot of oat poly(A) + R N A indicated that the size of the m R N A was approximately 2 kb (Fig. 2). Re- screening of the library with the partial clone yielded a 2124 bp clone (pADC2), whose sequence is shown in Fig. 1. The open reading frame comprises 1820 bp, and encodes a protein of 66 kDa. This size was unexpectedly large, given that the size of the isolated protein was 24 kDa. We used pADC2 to direct an in vitro transcrip- tion-translation reaction, and determined the product size by SDS-PAGE. As shown in Fig. 3, the product migrated at approximately 62 kDa in SDS gels, much closer to the size predicted by the ORF than to the size of the protein isolated from oat extracts.

Protein sequence similarity to the E. coli ARGdc

The E. coli gene encoding ARGdc has recently been isolated by complementation and sequenced (Moore and

Fig. 3. In vitro translation products, analyzed on SDS-PAGE, of RNA synthesized by T7 RNA polymerase from the pADC2 clone. The outer lanes are controls with no RNA and RNA from the vector alone

103 16 16 A.sot~vo~KVNOHt ' -~GLE^GSKPELF'~VCNGYKD 9-/-1LLHFHVGS

' : l l l l l l 11 l l ] i l l l ' ' ' ' ' ' ' E. coli--KVNOH~GLEAGSKAEL~-VCNGYKD.~-[[HFHLGS

92

36 /"-"XGGGLGVDYDGTRSGSSDHSVAYGLEEYA 16 PHPVLCTESGRA 93

: 1 i J i 1 ~ t l l 1 i s l i l l l l I T i i l l i i t l l S

34 GGGLGVDYEGTR--OSDCSVNYGLNEYA-~-PHPTVtTESGRA~._j 130

82 HPDYWGIOHLFP 20TCDSDG 31GAYOEALSNKHNLFG,"-~ , ' I I I I I I l l ~ l : ] i l I I ' ' ' , , , ,' ,' I : ~ I I HPDAWGIDOLFP2ff TCDSDGk--yGAYOEILGNHHNLFG 77

aa Fig. 4. Regions of similarity between the oat and Escherichia coli ARGdc derived amino acid sequences. The numbers indicate the number of amino acid residues separating these regions, in the respective oat and E. coli sequences

Boyle 1990; GenBank accession number M31770). A comparison of the derived amino acid sequences of the E. coli and oat genes reveals several regions of striking similarity, which are conserved both in sequence and in spacing (Fig. 4). There are three clusters of sequence similarity located in the C-terminal half of the protein, which, based on our protein sequence data, is the region of the oat protein that binds DFMA.

Processing of the pADC2 gene product

The ORF shown in Fig. I contains the amino acid se- quence determined from analysis of isolated ARGdc, but the sequence is internal and contained in the C- terminal half of the putative protein. The predicted mo- lecular weight of the protein beginning with the internal amino end that we identified by sequencing is approxi- mately 20.5 kDa, which is close to the size of 24 kDa predicted from migration of the A R G d c - D F M A com- plex in SDS gels (for clarity, we will refer to this protein

435

o

Fig. 5. Western blot analysis of protein extracted from oat leaves, probed with chicken anti-trpE-ARGdc (IgY). The arrows indicate the 34 kDa and 24 kDa forms of ARGdc. The controls include: preimmune chicken IgY; rabbit anti-trpE serum; and no primary antibody

as 24 kDa). This, plus the fact that there is no methio- nine at the internal amino end, suggests that the isolated A R G d c - D F M A complex contained a fragment of the full-length protein. Protein processing or proteolysis must have occurred prior to the fnal stages of purifica- tion. Since the 24 kDa band was the only one identified when gels were assayed for radioactivity, either this pro- cessing event is very thorough or D F M A binds preferen- tially to the processed protein. D F M A binding is an enzyme catalyzed process (Kallio et al. 1981; Bitonti et al. 1987), and preferential binding to the processed form of ARGdc might mean that proteolysis is required for maximal enzyme activity. Alternatively, it may be that the full-length protein contains structures which are sensitive to cleavage under the isolation conditions used. An antibody raised against a trpE-ARGdc fusion pro- tein was used to examine this possibility. Cordonnier et al. (1985) have described a technique for the detection on Western blots of the intact form of a highly labile protein, phytochrome, from oat seedlings. Protein iso- lated in this manner was solubilized, separated by SDS- PAGE, and transferred to nitrocellulose. Two major bands react with the antibody (Fig. 5), one at 24 kDa and one at 34 kDa; the 24 kDa band of the Western blot migrates at precisely the s a m e e f as does the DFMA-binding polypeptide. The presence of these two bands in this type of leaf extract suggests that protein

processing of ARGdc is an in vivo physiological event and not just an artifact of isolation.

D i s c u s s i o n

Two features of the arginine decarboxylase protein in oats are noteworthy. First, there is the extensive amino acid sequence similarity to the E. coli enzyme, with 9 dis- tinct regions scattered throughout the full 66 kDa ORF. Do these regions have functional significance? The three regions closest to the C-termius are within the 24 kDa fragment we identified by D F M A binding. Inhibition of ARGdc by D F M A can be slowed by the presence of the true substrate, arginine, suggesting that inhibition involves the active site of the enzyme (Kallio et al. 1985; Bitonti et al. 1987). The conservation of protein se- quence in this carboxyl region may thus be related to a function associated with the active site.

Secondly, we present evidence for protein processing of ARGdc in vivo, yielding both the 24 kDa DFMA- binding polypeptide, and another, unexpected, 34 kDa polypeptide. While pADC2 encodes a 66 kDa protein, and a form near this molecular weight can be detected after in vitro transcription-translation, Western blot analysis indicates that in oat leaves two much smaller forms of the protein predominate. Since the antibody was raised to a fusion protein containing the carboxyl end of the 24 kDa form, the 34 kDa and 24 kDa forms of ARGdc presumably share a common carboxyl end epitope. They may represent alternative forms of pro- cessing of oat leaf ARGdc.

We have no direct evidence that processing precedes D F M A binding, rather than vice versa, but the abun- dance of the 24 kDa polypeptide makes it likely that the former is true. We have not yet been able to isolate the native ARGdc, so we do not known the size and number of polypeptides in the enzyme; however, the sequence similarity between the E. coli and oat ARGdc in the N-terminal half of the protein suggests that this region may also be a part of the mature enzyme. A similar example exists in mammalian polyamine synthe- sis, where the mature s-adenosylmethionine decarboxy- lase has been shown to consist of a major polypeptide and a minor peptide, both of which are derived from a common precursor or translation product (Pajunen et al. 1988). There is a precedent for protein processing being a regulator of physiological processes (Neurath 1989); perhaps the processing we detect has such a role in the regulation of ARGdc activity in oats.

Acknowledgements. We thank Katherine E. Smith for technical as- sistance. We are grateful to Alan Bitonti and Peter McCann of the Merrell-Dow Research Institute for sharing DFMA with us, to John Wunderlich of the University of Georgia Molecular Genet- ics Facility for his help with protein purification, and peptide se- quencing, to Robert Moore and Stephen Boyle of Virginia Tech for sharing their sequence information with us, to Mark Compton of the University of Georgia for help with the chicken immuniza- tions, and to Kathy Spindler for the rabbit anti-trpE antibody. This research was supported by National Science Foundation grant DMB-87-15799, by U.S. Department of Agriculture competitive research grant GAM-89-01056, and by a University of Georgia Biotechnology Fellowship.

436

References

Bitonti A J, Casara P J, McCann PP, Bey P (1987) Catalytic irrevers- ible inhibition of bacterial and plant arginine decarboxylase activities by novel substrate and product analogues. Biochem J 242: 69-74

Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:5294-5299

Cohen E, Arad S, Heimer YM, Mizrahi Y (1982) Participation of ornithine decarboxylase in early stages of tomato fruit devel- opment. Plant Physiol 70 : 540-543

Cordonnier MM, Greppin H, Pratt LH (1985) Monoclonal anti- bodies with differing affinities to the red-absorbing and far red- absorbing forms of phytochrome. Biochemistry 24:3246-3253

Davis LG, Dibner MD, Battey JF (1986) Basic methods in molecu- lar biology. Elsevier, New York

Evans PT, Holaway BL, Malmberg RL (1988) Biochemical differ- entiation in the tobacco flower probed with monoclonal anti- bodies. Planta 175:259-269

Feirer RP, Mignon G, Litvay JD (1984) Arginine decarboxylase and polyamines are required for embryogenesis in the wild car- rot. Science 223 : 1433-1435

Flores HE, Galston AW (1982) Polyamines and plant stress: acti- vation of putrescine biosynthesis by osmotic shock. Science 217:1259-1261

Hiatt AC, McIndoo J, Malmberg RL (1986) Regulation of polya- mine synthesis in tobacco. J Biol Chem 261 : 1293-1298

Jacobs KA, Rudersdorf R, Neill SD, Dougherty JP, Brown EL, Fritsch EF (1988) The thermal stability of oligonucleotide dup- lexes is sequence independent in tetraalkylammonium salt solu-

tions: application to identifying recombinant DNA clones. Nucleic Acids Res 16 : 4637-4650

Kallio A, McCann PP, Bey P (1981) DL-c~(difluormethylarginine): a potent enzyme activated irreversible inhibitor of bacterial ar- ginine decarboxylases. Biochemistry 20: 3163-3166

Koerner TJ, Hill JE, Myers AM, Tzagoloff A (1990) High expres- sion vectors with multiple cloning sites for construction of trpE- fusion genes: pATH vectors. Methods Enzymol, in press

Moore RC, Boyle SM (1990) Nucleotide sequence and analysis of the speA gene encoding biosynthetic arginine decarboxylase in Escherichia coll. J Bac 172 : 4631-4640

Neurath H (1989) Proteolytic processing and phyiological regula- tion. Trends Biochem Sci 14:268-271

Pajunen A, Crozat A, Janne OA, Ihalainen R, Laitinen PH, Stanley B, Madhubala R, Pegg AE (1988) Structure and regulation of mammalian s-adenosylmethionine decarboxylase. J Biol Chem 263:17040-17049

Polson A, Coetzer T, Kruger J, Maltzahn E von, Merwe KJ van der (1985) Improvements in the isolation of IgY from the yolks of eggs laid by immunized hens. Immunol Invest 14:323-327

Slocum RD, Kaur-Sawhney R, Galston AW (1984) The physiology and biochemistry of polyamines in plants. Arch Biochem Bio- phys 235:283-303

Tabor CW, Tabor H (1984) Polyamines. Annu Rev Biochem 53 : 749-790

Young ND, Galston AW (1984) Physiological and control of argi- nine decarboxylase in K-deficient oat shoots. Plant Physiol 76:331 335

C o m m u n i c a t e d by E. M e ye row i t z