Arginine Decarboxylase 1s Localized in Chloroplasts - NCBI

Plant Physiol. (1 995) 109: 771-776

Arginine Decarboxylase 1s Localized in Chloroplasts’ Antonio Borrell, Francisco A. Culiaííez-Macia, Teresa Altabella, Robert T. Besford, Dante Flores, and Antonio F. Tiburcio* Laboratori de Fisiologia Vegetal, Facultat de Farmicia, Universitat de Barcelona, Diagonal 643, 08028 Barcelona, Spain (A.B., T.A., D.F., A.F.T.); Departament de Genètica Molecular, Centre d’ Investigació i Desenvolupament, Consejo Superior de lnvestigaciones Científicas, Jorge Girona 1 8-26, 08034 Barcelona, Spain (F.A.C.-M.); and Horticulture Research International, Littlehampton, West Sussex, BN 17 6LP, United Kingdom (R.T.B.) ment in such diverse phenomena is not clear. Lack of knowledge of the exact cellular and subcellular localization of polyamine biosynthetic enzymes has been one of the main obstacles in our understanding of the biological role of the ODC/ ADC/ polyamine system. In animals, cytochemical, autoradiographic, and immunocytochemical evidence suggests that ODC is localized in both the cytoplasm and the nucleus (Gilad and Gilad, 1981; Pegg et al., 1982; Persson et al., 1983; Dodds et al., 1990).To our knowledge, in plants, there is only one study on the autoradiographic localization of ODC in the nucleus and cytoplasm (Slocum, 1991). Little is known about the localization of ADC protein, despite its wide occurrence in plants (Tiburcio et al., 1990).The synthesis of Spd and Spm is carried out by addition of an aminopropyl moiety to one or both primary amino groups of Put by Spd synthase (EC 2.5.1.16) and Spm synthase (EC 2.5.1.22), respectively. Decarboxylated SAM, which acts as the aminopropyl donor, is derived from SAM via the action of SAM decarboxylase (EC 4.1.1.50)(Tiburcio et al., 1990). As for ADC, little is known about the localization of these enzymes (Slocum, 1991). It is now well documented that a variety of physiological stimuli and stress reactions affect the activity of polyamine biosynthetic enzymes in higher plants (Flores and Galston, 1982; Tiburcio et al., 1990; Slocum and Flores, 1991).However, little is known about the metabolic and molecular regulation of biosynthesis of these enzymes in plants (Slocum and Flores, 1991). For example, ADC activity in tobacco cells was fully repressed by exogenous Spd or Spm, without affecting ODC activity; however, the precise mechanism of ADC inhibition by polyamines is unknown (Hiatt et al., 1986).On the other hand, the factors determining the flux rates through the ADC versus ODC pathways in plants are still terra incognita (Slocum and Flores, 1991). In bacteria, flux through the ADC and ODC pathways appears to be determined by Orn availability, since intracellular Arg levels are always high (Tabor and Tabor, 1985). Under normal conditions, where Orn levels are not limiting, Put is synthesized primarily by ODC. If the Orn concentration drops, Arg is then utilized in Put synthesis via

Plants, unlike animals, can use either ornithine decarboxylase or arginine decarboxylase (ADC) to produce the polyamine precursor putrescine. Lack of knowledge of the exact cellular and subcellular location of these enzymes has been one of the main obstacles to our understanding of the biological role of polyamines in plants. We have generated polyclonal antibodies to oat (Avena sativa L.) ADC to study the spatial distribution and subcellular localization of ADC protein in different oat tissues. By immunoblotting and immunocytochemistry, we show that ADC is organ specific. By cell fractionation and immunoblotting, we show that ADC is localized in chloroplasts associated with the thylakoid membrane. The results also show that increased levels of ADC protein are correlated with high levels of ADC activity and putrescine in osmotically stressed oat leaves. A model of compartmentalization for the arginine pathway and putrescine biosynthesis in active photosynthetic tissues has been proposed. In the context of endosymbiote-driven metabolic evolution in plants, the location of ADC in the chloroplast compartment may have major evolutionary significance, since it explains (a) why plants can use two alternative pathways for putrescine biosynthesis and (b) why animals do not possess ADC.

Polyamines are polycationic cellular molecules that play an essential role in cell growth and differentiation (Pegg, 1986; Auvinen et al., 1992). Put, the precursor of polyamines, is formed in animals only by decarboxylation of Orn, via ODC (EC 4.1.1.17) (Heby and Persson, 1990). In contrast, in plants and bacteria there is an alternative pathway by which Put is produced from the decarboxylation of Arg by ADC (EC 4.1.1.19)(Tabor and Tabor, 1985; Tiburcio et al., 1990). In plants, polyamines, formed by either ADC or ODC, or both, are important modulators of biological responses such as cell division, reactions to stress, and development (Galston and Kaur-Sawhney, 1990; Galston and Tiburcio, 1991). However, the precise mechanism of their involve-

* This work was supported by Comisión Interministerial de Ciencia y Tecnología BI093-130, European Union-Concerted Action Programme AIR1-CT92-569 to A.F.T., by the British-Spanish Joint Research Program “Acciones Integradas” HB-079, HB-119A (R.T.B. and A.F.T.) and by the Biotechnology and Biological Sciences Research Council (R.T.B.). * Corresponding author; e-mail afernan8farmacia.ub.es; fax 343-402-1886.

Abbreviations: ADC, arginine decarboxylase; DFMA, DL-a-difluoromethylarginine; DFMO, DL-a-difluoromethylornithine; ODC, ornithine decarboxylase; Put, putrescine; SAM, S-adenosylmethionine; Spd, spermidine; Spm, spermine. 771

Borrell et al.

772

the ADC pathway. In the plant cell, flux through these pathways may be determined by different factors, including those derived from compartmentalization of intermediates or enzymes of Put synthesis (Slocum and Flores, 1991). with the aim of understanding the molecular mechanisms regulating ADC gene expression by polyamines, we generated polyclonal antibodies to oat (Avena sativa L.) ADC to investigate changes in ADC protein levels in osmotically stressed oat leaves (Tiburcio et al., 1994a). The results suggested that Spm affects the synthesis of ADC by inhibiting the posttranslational proteolytic processing of the inactive ADC precursor with a consequent decrease in the active, processed form of ADC (Tiburcio et al., 1994a). Here we describe the use of our anti-ADC antibodies to study the spatial pattern distribution and subcellular compartmentalization of ADC protein in different oat tissues. We report evidence that ADC is localized in the chloroplast compartment. This finding opens new insights into the catabolism of Arg in photosynthetic tissues. MATERIALS A N D M E T H O D S Plant Material and Osmotic Treatment

Avena sativa L. cv Victory (Svalof International, Svalov, Sweden) plants were grown as described elsewhere (Besford et al., 1993). Plants whose leaves were used for isolating chloroplasts were kept in the dark for 12 to 18 h to allow depletion of starch and were sampled at the end of the dark period. For osmotic treatment, peeled oat leaves were dark-incubated with a phosphate buffer containing 0.6 M sorbitol for 4, 24, or 48 h (Besford et al., 1993). Subcellular Fractionation

Leaves (15 g) were cut into small pieces with scissors and homogenized in 30 mL of a buffer containing 0.35 M SUC,25 mM Hepes, 2 mM EDTA, pH 7.6. Homogenates were filtered through two layers of Miracloth (Calbiochem) and centrifuged at lOOg for 2 min at 4°C to separate nuclei and cell-wall materials (Choe and Thimann, 1975). The residue was discarded, and the supernatant was centrifuged at 40008 for 1 min at 4°C. Both supernatant and enriched chloroplast pellet fractions were analyzed by immunoblot. In another experiment, we purified intact chloroplasts from an enriched chloroplast pellet obtained as described above. This pellet was resuspended in 10 mL of 0.33 M sorbitol and 50 mM Hepes-KOH, pH 7.5, and 5 mL were layered over each of two 30-mL Percoll gradients, prepared as described by Robinson and Barnett (1988). The Percollpurified chloroplasts were lysed with 12 mL of 10 mM Tricine-KOH, pH 7.8, 4 mM MgCI,, and 1 mM PMSF. After addition of 1.8 mL of 80% ( w / v ) SUC,the mixture was layered onto a discontinuous SUCgradient of 0.98 and 0.6 M SUCand centrifuged at 90,OOOg for 2 h. Under these conditions, the thylakoid and soluble stromal fractions appeared, respectively, at the bottom and at the top of the gradient. An aliquot of each fraction was subjected to SDS-PAGE and immunoblotting.

Plant Physiol. Vol. 109, 1995

Ceneration of Antibodies

An amino acid sequence (Gly-Pro-Ser-Leu-Val-Arg-

Val-Val-Gly-Thr-Gly-Asn-Gly-Gly-Ala-Phe-Asn-Val-Glu -Ala) near the C terminus deduced from the nucleotide sequence of the oat ADC gene (Bell and Malmberg, 1990) was selected. This polypeptide was synthesized by solid-phase peptide synthesis, coupled to purified-protein-derivative of tuberculin carrier protein, and injected into rabbits to produce the corresponding polyclonal antibodies. The method is fully described elsewhere (Besford et al., 1990). To test the specificity of the antiserum, anti-ADC antibodies were covalently linked to protein A-Sepharose (Pharmacia) as described by Schneider et al. (1982). ,4fter immunoprecipitation with either preimmune serum or with anti-ADC antibodies, ADC activity was determined as described below. lmmunoblot Analysis

Proteins were extracted from plant material with a buffer containing 0.6% P-mercaptoethanol and 1% SDS in 20 mM Tris buffer, pH 8.6, boiled for 5 min, and centrifuged at 15,OOOg for 5 min. The supernatants were subjected to SDS-PAGE and immunoblotting. SDS-PAGE (15% acrylamide) was performed according to Laemmli (1970). Immunoblot analyses were carried out as described by Besford (1990). Analysis of A D C and O D C Activities

ADC and ODC activities were determined as described elsewhere (Tiburcio et al., 1986). No significant release of CO, was observed in enzyme extracts preincubated with DFMA or DFMO. Therefore, possible artifacts (Birecka and Birecki, 1993) resulting from nonspecific decarboxylation of enzyme extracts were eliminated. lmmunolocalization of A D C Protein

For immunolocalization, the immune serum obtained against the ADC protein was used. An all-purpose fixative (80% ethanol, 3.5% formaldehyde, 5% acetic acid) was used for paraffin embedding. Sections from paraffin-embedded material were blocked with 3% goat serum in PBS (10 mM phosphate, 150 mM NaC1, pH 7.4) for 30 min at 22°C' and incubated with anti-ADC immune serum (diluted 1:500) or preimmune serum (diluted 1:500). Immunoreactivity was visualized by the avidin-biotin complex (Vectastain Elite ABC Kit, Vector, Burlingame, CA) using diaminobenzidine as substrate for peroxidase. RESULTS Organ-Specific Accumulation of A D C

We generated site-directed polyclonal antibodies to oat ADC (Tiburcio et al., 1994a). The primary structure 0.f oat ADC (Bell and Malmberg, 1990) was used for the selection of an epitope by computer calculations using the Genetics Computer Group (Madison, WI) package. The choice of this epitope was based on predictions of antigenicity and

Localization of Arg Decarboxylase

the fact that this region does not contain substrate binding sites (Tiburcio et al., 1994b). Immunoprecipitation of the ADC protein present in oat leaf extracts resulted in a parallel decrease of the ADC activity of those extracts (Fig. 1). These results reflect the specificity of the antiserum used in this study. Proteins from leaves and root of oat seedlings were extracted, separated by SDS-PAGE, and analyzed by immunoblotting using our anti-ADC antibodies. Figure 2 shows that there was cross-reactivity, which is apparently leaf specific. A single, 24-kD band was recognized in leaves, whereas in roots no band was detected (Fig. 2). These results agree with the high levels of ADC activity found in leaves (5 nmol CO2 h ' mg"1 protein) compared to the negligible levels of ADC activity found in roots. Previous work has shown that oat ADC is synthesized as a preprotein of 66 kD, which is cleaved to produce a fragment of 42 kD (containing the original amino terminus) and a 24-kD polypeptide (containing the original carboxyl terminus), these two processed polypeptides being held together by disulfide bonds (Malmberg et al., 1992; Malmberg and Cellino, 1994). Thus, the 24-kD band detected in our protein gel blots, run with /3-mercaptoethanol, represents the processed form of the ADC enzyme. The spatial distribution of ADC protein in different organs of oat seedlings was confirmed by immunocytochemistry. Bright-field photographs of paraffin sections from 9-d-old seedlings incubated with anti-ADC antibodies show that ADC protein was detected in oat leaves and stems (Fig. 3, A and C) but not in roots (Fig. 3D). The ADC protein appeared evenly distributed in the chloroplasts of leaves (Fig. 3A), whereas in stems ADC was located primarily in perivascular chloroplasts (Fig. 3C). The results obtained with oat stem led us to test whether ADC could be differentially located in leaves of C4 plants containing different functional chloroplasts (Salisbury and Ross, 1991). We took advantage of the cross-reaction of the oat ADC antibody with corn (a C4 plant), in which it recognized a single band in immunoblots, with a molecular mass similar to that of oat ADC. In corn, as in oat stem, the oat ADC

75-

"-0-.,. 25-

5

10

15

20

25

Anti-ADC antiserum !/(ll Figure 1. Immunodepletion of ADC. ADC activity of oat leaf extracts remaining in the supernatant after immunoprecipitation with either preimmune serum (D) or anti-ADC antibodies (0).

773

24 »-

Figure 2. Immunoblot analysis of protein extracts from oat leaves and roots using ADC antibodies. Lane 1, Leaves from 9-d-old seedlings; lane 2, roots from 9-d-old seedlings; lane 3, molecular mass markers: 106, 80, 49, 32, 27, and 18 kD. Each lane was loaded with 30 ,u,g of protein.

antibodies reacted strongly with perivascular chloroplasts, although no reaction was seen with the mesophyll chloroplasts (data not shown).

ADC Protein Localized in Chloroplasts

To confirm the intracellular location of ADC protein, chloroplasts were purified from leaves of 9-d-old oat seedlings by differential centrifugation, and supernatant and chloroplast-enriched pellet fractions were analyzed with anti-ADC antibodies after PAGE and immunoblotting (Fig. 4). A unique band of 24 kD was detected in the pellet containing chloroplasts, whereas no band was detected in the supernatant (Fig. 4, lanes 2 and 3). In different experiments we determined the intrachloroplastic location of ADC protein. We first obtained the enriched chloroplast pellet, which was further resuspended in a Hepes-sorbitol buffer and fractionated on a discontinuous Percoll gradient. The purified chloroplasts were then lysed and fractionated on a Sue gradient. The resulting soluble stromal and thylakoid fractions were analyzed by immunoblotting (Fig. 4). The 24-kD band was detected primarily in the thylakoid fraction, whereas only a faint band was detected in the stromal fraction (Fig. 4, lanes 4 and 5). These results were further confirmed by immunocytochemistry at the electron-microscope level. Gold particlelabeled ADC antigen appeared mostly over the thylakoid membranes, whereas some label was observed on the rest of the chloroplast (data not shown). How ADC is associated with the thylakoids is not known, but studies on the ADC secondary protein structure, using the Genetics Computer Group package, have shown some putative amphiphilic-a-helix-forming regions with a high hydrophobic moment, which might be involved in membrane interactions. However, no clear transmembrane domains have been found.

Borrell et al.

774


B

r

X

#

Figure 3. Immunolocalization of ADC protein. Paraffin-embedded sections of organs from 9-d-old oat seedlings were incubated with anti-ADC (A, C, and D) or preimmune antiserum and an avidin-biotin-peroxidase detection system. Brown staining indicates the presence of peroxidase activity. A and B, Leaf sections; C, stem transverse section (magnification X 570); D, root section (magnification x 32).

Response of ADC Enzyme to Osmotic Stress

Previous work has shown that osmotic treatment of oat leaves in the dark increases the levels of ADC activity and leads to senescence (Tiburcio et al., 1994b). Hence, in this study changes in soluble ADC protein levels of oat leaves subjected to osmotic treatment were analyzed. Osmotic stress increased the levels of the 24-kD polypeptide in comparison with the 0-h control (Fig. 5). No interfering bands were detected after immunoblotting with preimmune serum (data not shown). These changes in ADC protein levels are well correlated with changes in ADC activity and Put levels (Tiburcio et al., 1994b). DISCUSSION

Although ADC is apparently ubiquitous in plants, its subcellular location has not been definitely determined. Previous work attempting the subcellular localization of ADC involved the measurement of enzyme activity (Torrigiani et al., 1986; Walker et al., 1987). However, these

studies have been questioned because of artifacts resulting from the use of 14C trapping (Birecka and Birecki, 1993; Smith, 1993). Here we show that ADC distribution is organ specific, apparently localized to the thylakoid membranes of the chloroplast. This has been demonstrated by cell fractionation and immunocytochemistry at the electron-microscope level. This indicates that a plant-specific polyamine biosynthetic pathway is located in the chloroplast, and raises the intriguing question of whether Put is synthesized in root tissue de novo or transported there. Figure 6 shows a proposed model for the Arg pathway and Put biosynthesis in active photosynthetic tissues. The primary enzymes of the Arg pathway (from glutamate to citrulline) are chloroplastic, whereas the terminal step from citrulline to Arg is carried out in the cytoplasm (reviewed by Bryan, 1990). Our study indicates that ADC is chloroplastic; therefore, transport of Arg from the cytoplasm to the chloroplast and further decarboxylation of this amino acid by ADC provides the only source of newly synthe-

Localization of Arg Decarboxylase 1

775

2 GLUTAMATE

'It"

2

Chloroplast

AG

AGP

• AGSA t

Polyamines

-24 AORN 6

Figure 4. Immunoblot analysis of ADC protein in protein extracts from different subcellular fractions. Lane 1, molecular mass markers: 106, 80, 49, 32, 27, and 18 kD; lanes 2 and 3, supernatant and chloroplast-enriched pellet fractions; lanes 4 and 5, thylakoid and stromal fractions. Each lane was loaded with 30 jxg of protein.

sized Put (and polyamines) within this organelle (Fig. 6). This hypothesis is further supported by the fact that negligible levels of ODC activity were detected in oat leaves. In contrast, in a nonchloroplastic tissue (such as oat roots) the levels of ODC activity are high (4 nmol CO2 h"1 mg"1 protein) in relation to the negligible levels found in leaves. Therefore, in the absence of ADC, the enzyme ODC may be the key enzyme of Put biosynthesis in roots. In Neurospora (a nonphotosynthetic organism), most of the primary enzymes of the Arg pathway are localized to the mitochondria, but the terminal step from citrulline to Arg is carried out in the cytoplasm (Schubert and Boland, 1990). In the absence of ADC, Arg can be catabolized by the enzyme arginase (EC 3.5.3.1) to urea and Orn (Lehninger, 1980), and this amino acid may be further metabolized to Put and polyamines. In fact, Arg in mammals is considered an essential amino acid, since it is rapidly catabolized by arginase and is usually not available for protein synthesis (Lehninger, 1980). Therefore, Orn may be the only substrate available for the synthesis of Put in mammals. This last model may also operate in roots and other nonphotosynthetic tissues.

24-

T Figure 5. Immunoblot analysis of ADC protein in osmotically stressed oat leaves. Protein extracts from osmotically treated oat leaves incubated with immune serum. Treatment with 0.6 M sorbitol: 0 time (lane 1); 4 h (lane 2); 24 h (lane 3); and 48 h (lane 4). Molecular mass markers: 106, 80, 49, 32, 27, and 18 kD (lane 5). Each lane was loaded with 50 jig of protein.

f

[ ——— * 11 ' 10 O R N - - - - - - - - - - - - - - » - PUT-«—— -«——— •«—— ARGININE

err

" 1

.

.

1

CIT ————-7————————** AS —————- ——————f- ARGININE

ASP

"

Cytosol

Figure 6. Proposed scheme for compartmentalization of Arg pathway and Put biosynthesis in actively photosynthetic tissues. Intermediates: AG, N-acetylglutamate; AGP, N-acetylglutamate 5-phosphate; AGSA, N-acetylglutamate 5-semialdehyde; AORN, N-acetylornithine; ORN, Orn; CIT, citrulline; ASP, aspartate; AS, argininosuccinate; PUT, Put. Enzymes: 1, acetyl-CoA:glutamate Nacetyltransferase (EC 2.3.1.1); 2, acetylornithine:glutamate N-acetyltransferase (EC 2.3.1.35); 3, N-acetylglutamate kinase (EC 2.7.2.8); 4. N-acetylglutamate semialdehyde oxidoreductase (EC 1.2.1.38); 5. N-acetylornithine aminotransferase (EC 2.6.1.11); 6, acetylornithine deacetylase (EC 3.5.1.16); 7, ornithine carbamoyltransferase (EC 2.1.3.3); 8, argininosuccinate synthase (EC 6.3.4.5); 9, argininosuccinate lyase (EC 4.3.2.1); 10, Arg decarboxylase; 11, Orn decarboxylase.

In the context of the endosymbiote-driven metabolic evolution in plants (Weeden, 1981), the finding that ADC is localized in the chloroplast may have a major evolutionary significance, since it explains (a) why plants can use two alternative pathways for the synthesis of the same compounds and (b) why animals, irrespective of precursor availability, do not possess ADC. Chloroplasts have the capacity to encode and synthesize only some of their proteins, and the remainder are probably all nuclear encoded (Dyer, 1984). These nuclear-encoded proteins are synthesized in the cytoplasm, generally as larger precursors containing a transit peptide, and enter chloroplasts via a posttranslational process (Keegstra et al., 1995). Since oat ADC mRNA is poly(A)+ (Bell and Malmberg, 1990), ADC protein may be nuclear encoded, raising the question of entrance to the plastid, which is currently being investigated.

ACKNOWLEDGMENTS Our thanks to Robin Rycroft for help in preparing the manuscript and Rick Walden, Maarten Chrispeels, Montse Pages, and

Carles Masgrau for their comments. Received May 9, 1995; accepted August 7, 1995. Copyright Clearance Center: 0032-0889/95/109/0771/06.

Borrell et al.

776 LITERATURE ClTED

Auvinen M, Paasinen A, Andersson LC, Holtta E (1992)Ornithine decarboxylase activity is critica1 for cell transformation. Nature 360: 355-358 Bell E, Malmberg RL (1990) Analysis of a cDNA encoding arginine decarboxylase from oat reveals similarity to the Eschevichia coli arginine decarboxylase and evidence of protein processing. Mo1 Gen Genet 224: 431436 Besford RT (1990) The greenhouse effect: acclimation of tomato plants growing in high CO,. Relative changes in Calvin cycle enzymes. J Plant Physiol 1 3 6 458463 Besford RT, Richardson C, Campos JL, Tiburcio AF (1993) Effect of polyamines on stabilization of molecular complexes in thylakoid membranes of osmotically stressed oat leaves. Planta 1 8 9 201-206 Besford RT, Thomas B, Huskisson N, Butcher G (1990) Characterization of conformers of D1 of photosystem I1 using sitedirected antibodies. Z Naturforsch 45: 621-626 Birecka H, Birecki M (1993) Polyamine biosynthesis. In PJ Lea, ed, Methods in Plant Biochemistry, Vol 8. Academic Press, New York, pp 447-467 Bryan JK (1990) Advances in the biochemistry of amino acid biosynthesis. Zn BJ Miflin, PJ Lea, eds, The Biochemistry of Plants, Intermediary Nitrogen Metabolism, Vol 16. Academic Press, New York, pp 161-195 Choe HE, Thimann KV (1975) The metabolism of oat leaves during senescence. 111. The senescence of isolated chloroplasts. Plant Physiol 55: 828-834 Dodds RA, Pitsillides AA, Frost GTB (1990) A quantitative cytochemical method for ornithine decarboxylase activity. J Histochem Cytochem 38: 123-127 Dyer TA (1984) The chloroplast genome: its nature and role in development. Zn NR Baker, J Barber, eds, Chloroplast Biogenesis. Elsevier Science Publishers, Amsterdam, pp 24-69 Flores HE, Galston AW (1982) Polyamines and plant stress: activation of putrescine by osmotic shock. Science 217: 1259-1261 Galston AW, Kaur-Sawhney R (1990) Polyamines in plants. Plant Physiol 94: 406410 Galston AW, Tiburcio AF, eds (1991) Lecture Course on Polyamines as Modulators of Plant Development, Vol 257. Fundación Juan March, Serie Universitaria, Madrid, Spain Gilad GM, Gilad VH (1981) Cytochemical localization of ornithine decarboxylase with rhodamine or biotin-labelled a-difluoromethylornithine. J Histochem Cytochem 29: 687-692 Heby O, Persson L (1990) Molecular genetics of polyamine metabolism in eukaryotic cells. Trends Biochem Sci 15: 153-158 Hiatt AC, McIndoo J, Malmberg RL (1986) Regulation of polyamine synthesis in tobacco. J Biol Chem 261: 1293-1298 Keegstra K, Bruce B, Hurley M, Li H, Peny S (1995) Targeting of proteins into chloroplasts. Physiol Plant 93: 157-162 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685 Lehninger AL (1980) Biochemistry. Editorial Worth Publishers, New York Malmberg RL, Cellino ML (1994) Arginine decarboxylase of oats is activated by enzymatic cleavage into two polypeptides. J Biol Chem 28: 2703-2706


Malmberg RL, Smith KE, Bell E, Cellino ML (1992) Arginine decarboxylase of oats is clipped from a precursor into .two polypeptides found in the soluble enzyme. Plant Physiol '100: 146-152 Pegg AE (1986)Recent advances in the biochemistry of polyamxnes in eukaryotes. Biochem J 234: 249-262 Pegg AE, Seely ]E, Zagon IS (1982) Autoradiographic identif'ication of ornithine decarboxylase in mouse kidney by means of a-[5-14C]difluor~methylornithine.Science 217: 68-70 Persson L, Rosengren E, Sundler F, Uddman R (1983) Immunocytochemical localization of ornithine decarboxylase. Methods Enzymol 94: 166-169 Robinson C, Barnett LK (1988) Isolation and analysis of chloroplasts. In C Shaw, ed, Plant Molecular Biology: A Practical Approach. IRL Press, Oxford, UK, pp 67-78 Salisbury FB, Ross CW (1991) Plant Physiology. Wadsworth I'ublishing Company, Belmont, CA Schneider C, Newman RA, Sutherland DR, Asser U, Greaves MF (1982) A one-step purification of membrane proteins u!;ing a high efficiency immunomatrix. J Biol Chem 257: 10766-10769 Schubert KR, Boland MJ (1990) The ureides. In BJ Miflin, PJ Lea, eds, The Biochemistry of Plants, Intermediary Nitrogen Metabolism, Vol 16. Academic Press, New York, pp 197-282 Slocum RD (1991) Tissue and subcellular localization of polyamines and enzymes of polyamine metabolism. In RD Slocum, HE Flores, eds, Biochemistry and Physiology of Polyamines in Plants. CRC Press, Boca Raton, FL, pp 93-103 Slocum RD, Flores HE, eds (1991)Biochemistry and Physiology of Polyamines in Plants. CRC Press, Boca Raton, FL Smith TA (1993) Amines. In PG Waterman, ed, Methods in Plant Biochemistry, Vol 8. Academic Press, New York, pp 18-49 Tabor CW, Tabor H (1985) Polyamines in microorganisms. IMicrobiol Rev 49: 81-99 Tiburcio AF, Besford RT, Borrell A (1994a) Posttranslational regulation of arginine decarboxylase synthesis by spermine in osmotically-stressed oat leaves. Biochem SOCTrans 22: 4555; Tiburcio AF, Besford RT, Cape11 T, Borrell A, Testillano PS, Risuefio, MC (199413) Mechanisms of polyamine action during senescence responses induced by osmotic stress. J Exp Bot 45: 1789-1800 Tiburcio AF, Kaur-Sawhney R, Galston AW (1990) Polyamine metabolism. In BJ Miflin, PJ Lea, eds, The Biochemistry of I?lants, Intermediary Nitrogen Metabolism, Vol 16. Academic Press, New York, pp 283-325 Tiburcio AF, Masdéu MA, Dumortier FM, Galston AW (1986) Polyamine metabolism and osmotic stress. I. Relation to protoplast viability. Plant Physiol 82: 369-374 Torrigiani P, Serafini-Fracassini D, Biondi S, Bagni N (1986) Evidence for the subcellular localization of polyamines anmd their biosynthetic enzymes in plant cells. J Plant Physiol 124: 23-29 Walker MA, Ellis BE, Chapple CCS, Dumbroff EB (1987) Subcellular localization of amines and activities of their biosynthetic enzymes in p-fluorophenylalanine resistant and wild-type tobacco cell cultures. Plant Physiol 85: 78-81 Weeden NF (1981) Genetic and biochemical implications of the endosymbiotic origin of the chloroplasts. J Mo1 Evol17: 133-139