THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1987 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 262, No. 24, Issue of August 25, pp. 11726-11730 1987 Printed in i r . . ~ . ~ .

Differential Expression of C4 Pathway Genes in Mesophyll and Bundle Sheath Cells of Greening Maize Leaves*

(Received for publication, July 30, 1986)

Jenq-Yunn Sheen$ and Lawrence Bogorad From the Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138

Pyruvate orthophosphate dikinase, phosphoenolpy- ruvate carboxylase, and NADP-malate dehydrogenase function in a series of reactions for fixing Con in mes- ophyll cells and NADP-malic enzyme (ME) catalyzes the production of CO, and NADPH in bundle sheath cells of maize which is a NADP-ME type C4 plant. Northern blot analyses with cDNA clones for pyruvate orthophosphate dikinase and phosphoenolpyruvate carboxylase and in vitro translation-immunoprecipi- tation experiments with antiserum to NADP-malate dehydrogenase showed that pools of transcripts of these three genes grow and shrink coordinately in mesophyll cells but not in bundle sheath cells upon illumination of dark-grown maize seedlings. Western blot analyses indicated that the protein levels of phos- phoenolpyruvate carboxylase and pyruvate orthophos- phate dikinase are low in dark-grown maize seedlings and increase progressively following light-induced transient accumulation of their mRNAs in mesophyll cells. These proteins continue to accumulate and pla- teau in late-greening and green leaves in spite of a rapid drop in the sizes of their mRNA pools. Surpris- ingly, relatively large amounts of NADP-malate de- hydrogenase are present in mesophyll cells of etiolated leaves despite the low level of the corresponding mRNA. No phosphoenolpyruvate carboxylase or NADP-malate dehydrogenase were detected in bundle sheath cells. On the other hand, the M E gene responds to light induction at both the transcriptional and trans- lational levels only in bundle sheath cells. Moreover, the steady-state level of ME mRNA stays high in late- greening and green leaves in contrast to the rapid decline of mRNA levels of three other Cq pathway genes in mesophyll cells. In addition, low levels of both the mRNA and protein encoded by the PPDK gene were detected in bundle sheath cells. These levels were not influenced by light as distinguished from the patterns observed in mesophyll cells.

In general, C4 plants exhibit some very interesting features such as two distinct photosynthetic cell types often associated with dimorphic chloroplasts, high photosynthetic rates and growth rates, low photorespiration rates, and markedly re- duced rates of water loss. These features are all interrelated and result from modified metabolic processes for photosyn-

* This research was supported in part by a research grant from the National Institute of General Medical Sciences and was also sup- ported in part by the Maria Moors Cabot Foundation for Botanical Research of Harvard University. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact.

4 Supported by a graduate fellowship from Harvard University.

thesis (1-3). Maize is a NADP-malic enzyme type C, plant, its two sequential COz fixation pathways are separated, one in mesophyll and the other in bundle sheath cells, and are outlined as follows. In bundle sheath cells:

pyruvate orthophosphate dikinase pyruvate + ATP + Pi b

phosphoenolpyruvate + AMP + PPi phosphoenolpyruvate +

coz phosphoenolpyruvate carboxylase + oxalacetate

oxalacetate + NADPH NADP-malate dehydrogenase

malate + NADP+

Malate is transferred from mesophyll cells to bundle sheath cells. In mesophyll cells:

malate + NADP+ NADP-malic enzyme +

C02 + NADPH + pyruvate

ribulose-1,5-bisphosphate carboxylase COz + RuBP 4

2(3-phosphoglycerate)

We are interested in elucidating the molecular mechanisms by which these C, enzymes are differentially expressed in bundle sheath and mesophyll cells of mature leaves (1-8) and the relationships between the expression programs of C4 path- way genes and the plastid differentiation and maturation processes in these two types of cells.

As an initial step toward studying mechanisms for the regulation of expression of C, pathway genes, we have made and identified cDNA clones of PPDK, PEPC, and M E genes. By RNA blotting, with cDNA clones of PPDK, PEPC, and ME genes as well as protein blotting and immunoprecipita- tion with antisera to pyruvate orthophosphate dikinase, phos- phoenolpyruvate carboxylase, NADP-malate dehydrogenase, and ME, we have determined the expression programs for mRNA and protein of each of these C, pathway genes in bundle sheath and mesophyll cells during photoregulated greening of etiolated maize leaves. The levels of transcripts of all four genes were induced by light in a cell-specific fashion in bundle sheath and mesophyll cells of illuminated dark- grown maize seedlings. With the exception of NADP-malate dehydrogenase protein, which accumulated in the dark, the protein levels of pyruvate orthophosphate dikinase, phos- phoenolpyruvate carboxylase, and M E all rose coordinately and cell-specifically upon illumination and remained stable in completely developed mature green leaves. The accumula- tion of proteins followed increases in mRNA pools and thus

11726

C4 Genes in Bundle Sheath and MeL,pnyll Cells of Maize 11727

probably resulted from rises in mRNA levels. The expression programs of these four C4 pathway genes were not always tightly coordinated with one another.

MATERIALS AND METHODS

Plant Materials and Growth-Maize seedlings (Zea mys: FR9'"' X FR37; Illinois Foundation Seeds) were grown in soaked Vermiculite at 28 "C in a dark room for 7 days and greened for the periods indicated under fluorescent lamps of 1400 lux at 25 "C. Maize seed- lings were also grown in the greenhouse with 16-h photoperiods for 10 days.

RNA Extraction and Isolation of Polyadenylated RNA-Total leaf RNA was extracted as described (10). Polyadenylated RNA was isolated by dissolving 50 mg of total RNA in 25 ml of 0.5 M LiCl, 1 mM EDTA, 20 mM Tris-HCI, pH 7.5, 0.1% SDS.' To the RNA solution was added 0.5 g of oligo(dT)-cellulose (Type 3 from Collab- orative Research Inc.) equilibrated with the same buffer. The mixture was incubated at room temperature with a rocking motion for a t least 6 h. After hybridization, the mixture was washed three times with equal volumes of the same buffer and then three times with equal volumes of the same buffer except that the concentration of LiCl was reduced to 0.1 M. The slurry of oligo(dT)-cellulose containing poly- adenylated RNA was loaded onto a 5-ml disposable column (Isolab, Inc.) from which polyadenylated RNA was eluted with 10 mM Tris- HCl, pH 7.5, 1 mM EDTA, 0.1% SDS (11).

Sucrose Gradient Fractionation of Polyadenylated RNA-Twelve ml of 5-30% sucrose gradients were made in 10 mM Tris-HC1, pH 7.5,l mM EGTA, 0.5% SDS. One hundred to 300 pg of polyadenylated mRNA was dissolved in 100 pl of sterile double-distilled water and heated at 90 "C for 1 min before loading. Gradients were centrifuged in an SW40 rotor a t 28,000 rpm for 24 h at 20 "C (12). Fractions of 400 p1 were collected from the bottom of the gradient and the RNA was precipitated by the addition of 1 ml of ethanol containing 0.15 M sodium acetate.

In Vitro Translation of Polyadenylated RNA-No more than 0.5 pg of polyadenylated RNA or 5 pg of total RNA was translated in 30 p1 of a rabbit reticulocyte lysate in vitro translation mixture (Bethesda Research Laboratories) with [%S]methionine added as described (10).

SDS-Polyacrylamide Gel Electrophoresis and Fluorography-The Laemmli (13) buffer system was used for polyacrylamide gel electro- phoresis as described. To achieve the maximal resolution without the use of gradient gels, we typically ran a 12.5% polyacrylamide gel with a low ratio of acry1amide:bisacrylamide (30:0.4) at a constant current of 20 mA for 3 h a t room temperature. Gels for fluorography were fixed with 50% methanol and 10% acetic acid for 20 min before being treated with Enlightning (New England Nuclear) for 20 min. Gels were dried and exposed to Kodak x-ray films at -80 "C.

Methods for cDNA Cloning-One hundred ng of large polyadenyl- ated RNA that had been enriched by sucrose gradient fractionation was incubated at 37 "C for 1 h in a reaction volume of 10 p1 containing buffer and other components as described (14) to synthesize the first- strand cDNA. The second-strand cDNA was synthesized in the same tube by diluting the first-strand synthesis reaction directly with 90 pl of second-strand synthesis mixture as described (14) at 14 "C for 1 h and then transferred to room temperature and incubated for another hour. Products were purified by phenol extraction and precipitated with ethanol before the tailing reaction was carried out. The pUC13 plasmid DNA was linearized with Sac1 and tailed with dCTP. The double-stranded cDNA was tailed with dGTP. The tailing reactions were carried out with BRL terminal transferase in the buffer supplied by the manufacturer. The enzyme concentration was 1 unit/pl, and the reaction was incubated at room temperature for 1 h for cDNA tailing and 25 min for plasmid tailing at a DNA concentration of 1 mg/ml. The enzyme was then inactivated by heating for 10 min at 65 "C. Annealing was done in 10 mM Tris-HC1, pH 7.5,l mM EDTA, 150 mM NaCl at a DNA concentration of 0.5 pg/ml for 1 h at 58 "C (14). MC1061 competent cells were used as hosts.

Hybrid Selection and Immunoprecipitation-Typically, the plasmid DNA of each clone was prepared from 3 ml of bacterial culture by the alkaline mini-lysate method (15). One hundred ng of plasmid DNA of candidate clones, selected by the colony hybridization method of Maas (16) with enriched large size polyadenylated mRNA, was

' The abbreviations used are: SDS, sodium dodecyl sulfate; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid ME, NADP-malic enzyme.

digested with BamHI and then extracted with phenol and precipitated with ethanol. DNA was denatured in 20 pl of 0.5 M NaOH at room temperature for 20 min and boiled for 1 min and then kept on ice. One hundred pl of 1 M NH, acetate was added to neutralize the DNA (17) before spotting onto a 9-mm2 nitrocellulose square by vacuum

with 0.5 M NaCl before baking at 80 "C in a vacuum oven for 2 h. filtration. The filters were air-dried at room temperature and washed

The hybridization and mRNA elution were done as described (18). A published protocol was followed for immunoprecipitation (19) with the modification of separating IgGsorb (The Enzyme Center, Inc.) from unreacted proteins by spinning through an 0.8-ml 30% sucrose cushion in a microfuge for 2 min. Preparation of the antiserum to phosphoenolpyruvate carboxylase is described elsewhere.' Antisera to pyruvate orthophosphate dikinase, ME, and NADP-malate dehy- drogenase were kindly provided by H. Roy, Rensselaer Polytechnic Institute, P. Collins, Ohio State University (20), and B. B. Buchanan, University of California, Berkely (8, 21), respectively.

Northern Blot Analysis-Thirty pg of total RNAs isolated from bundle sheath and mesophyll cells of etiolated, greening (7, 12, 24, 48, 72, and 96 h), and green second leaves of maize seedlings were fractionated and blotted as described previously (10). Only the fully differentiated middle sections (4-12 cm) of the second leaves were used to isolate bundle sheath and mesophyll cells (10, 22-24). Meso- phyll cells were prepared generally using conditions as described previously (10). Bundle sheath cells were then isolated from digested leaf preparations after mesophyll had been removed. The probes used for analyses were cDNA clones for pyruvate orthophosphate dikinase (clone 1-9), phosphoenolpyruvate carboxylase (clone 4-9), and ME (clone 4-15) with inserts of approximately 550, 1000, and 450 base pairs, respectively. The hybridization and washing conditions have been described elsewhere (10).

Protein Extraction and Western Blot Analysis-Samples of bundle sheath and mesophyll cells were ground to fine powders in Microfuge tubes with a metal spatula in the presence of liquid N2. One ml of cell lysis buffer containing 100 mM Tris-HC1, pH 7.5,2% SDS (w/v), 5% 0-mercaptoethnol (v/v), 6% (w/v) sucrose and 1 mM EDTA was added and the samples were boiled for 5 min. The clear supernatant of each sample obtained after brief spinning was aliquoted into microfuge tubes and stored at -80 "C. For Western blotting, protein samples, normalized with respect to one another by their RNA content, were fractionated by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose (0.1 pm) as described (25) in the presence of 0.1% SDS. The blots were blocked with 30% calf serum for 1 h at room temperature before applying antisera made against pyruvate orthophosphate dikinase, phosphoenolpyruvate carboxyl- ase, NADP-malate dehydrogenase, and ME. The bound antibodies were revealed by horseradish peroxidase conjugated goat anti-rabbit second antibody and the color development reagent 4-chloro-1-naph- tho1 following a method provided by the manufacturer (Bio-Rad).

RESULTS AND DISCUSSION

An Improved Method for Sucrose Gradient Fractionation of Large Size Polyadenylated RNAs-Two methods have gener- ally been used to enrich for mRNAs of certain sizes for cDNA cloning. Electrophoresis and recovery from low-melting aga- rose/methyl mercury/hydroxide gels is toxic, tedious, and gives very low yields, especially of mRNAs encoding polypep- tides larger than 60 kDa (26,27). On the other hand, although the conventional sucrose gradient fractionation method gives very good yields of all sizes of mRNAs, fractions containing large mRNAs are generally heavily contaminated with aggre- gated small sized mRNAs (12,20,31). The conditions we used for sucrose gradient centrifugation (described under "Mate- rials and Methods") eliminated the aggregation problem and facilitated the production and cloning of cDNAs for phos- phoenolpyruvate carboxylase (100 kDa) (7,28,29,30) and the precursors of pyruvate orthophosphate dikinase (110 kDa) (29,31, 32) and M E (66 kDa) (20) (Fig. 1).

Identification of cDNA Clones Encoding Pyruvate Ortho- phosphate Dikinase, Phosphoenolpyruvate Carboxylase, and ME-Clones of maize cDNAs for pyruvate orthophosphate

J.-Y. Sheen and L. Bogorad, manuscript in preparation.

11728 C4 Genes in Bundle Sheath and Mesophyll Cells of Maize 1 2 3 4 5 6 7 8 9 1011 12 Mr-10”

=92

ME - -66

-45 -a-

-14

FIG. 1. Improved sucrose g rad ien t f rac t iona t ion of po ly- adenylated RNA and in vitro translation. RNAs collected from the first 12 fractions were each dissolved in 20 pI of double-distilled H,O. One pl of each RNA sample was translated in vitro, as described under “Materials and Methods,” and the [‘sS]methionine-labeled products were fractionated electrophoretically on an SDS-polyacryl- amide (12.5% acrylamide) gel. The fluorography was done as de- scribed under “Materials and Methods.” The arrows indicate the locations of phosphoenolpyruvate carboxylase (PEPC) and precur- sors for pyruvate orthophosphate dikinase (PPDK) , ME, and NADP- malate dehydrogenase ( M D H ) .

A 0

a b C d HS-I V T PPDK PEPC ME

a b A+a b a b

FIG. 2. Identification of cDNA clones by hybrid selection, in vitro t ranslat ion, and immunoprecipi ta t ion. The products of hybrid selection and in vitro translation (HS-IVT) were examined by electrophoresis on an SDS-polyacrylamide gel. Autoradiograms are shown. A: HS-IVT, products of in vitro translation of poly(A+) RNA hybrid selected by: lane a, pUC13 DNA (control); lane b, PPDK cDNA; lane c, PEPC cDNA; lane d, M E cDNA. B: lane A+, in vitro translation products of unfractionated poly(A+) RNA of maize leaves. Lune a, the products of in vitro translation of unfractionated poly(A+) RNA of maize leaves precipitated with antibodies to pyruvate ortho- phosphate dikinase (PPDK), phosphoenolpyruvate carboxylase (PEPC), and ME, respectively, and solubilized with SDS. Lune b, immunoprecipitates of HS-IVT products of pyruvate orthophosphate dikinase, phosphoenolpyruvate carboxylase, and M E cDNA clones dissolved with SDS.

dikinase, phosphoenolpyruvate carboxylase, and M E were identified by in vitro translation of mRNAs hybrid-selected with each of the three cDNAs and immunoprecipitation of translation products with monospecific antisera. The high backgrounds observed in hybrid selection and in vitro trans- lation experiments (Fig. 2) were due to endogenous mRNA in the translation mixture and nonspecific mRNA selection, as shown in the control experiments without exogenous mRNA and with mRNA selected by pUC13 DNA without any insert. Both pyruvate orthophosphate dikinase and M E were trans- lated in vitro as precursors 8-10 kDa bigger than the in vivo proteins (20, 31, 32).

Mesophyll Cell-specific Expression of PPDK and PEPC Genes-By Northern Blot analysis, we found that the levels of transcripts encoded by PPDK and PEPC genes increased over 50-fold (determined by densitometer scanning) in mes- ophyll cells after 24 h of illumination of dark-grown seedlings and then declined to a low steady-state level in 96-h greening and green leaves. The mRNA for phosphoenolpyruvate car- boxylase was not detectable in bundle sheath cells; however, small amounts of pyruvate orthophophate dikinase tran-

b B

-1 -1

Y 0 6 ? ? = 4 8 7 2 m G O 6 122448RRIG h

;FCC 3.3kb

8s 6 5

‘POI 3.4 h b

IE 2 2 k b

FIG. 3. Nor the rn blot analysis to s tudy cell-specific expres- sion o f PPDK, PEPC, and ME genes. RNA samples (30 pg) isolated from bundle sheath cells ( E ) and mesophyll cells of etiolated (O), greening (6 , 12,24, 48, 72, and 96 h), and green ( C ) maize leaves were fractionated on a formaldehyde-agarose gel (1% agarose) and blotted to nitrocellulose. a, a photograph of ethidium bromide-stained RNA gel. b, autoradiograms of Northern blots probed with nick- translated cDNA clones of pyruvate orthophosphate dikinase (PPDK), phosphoenolpyruvate carboxylase (PEPC), and ME, re- spectively. The 23S* is a broken product of 23 S rRNA.

scripts were found in bundle sheath cells throughout the greening time course without any fluctuation (Fig. 3).

The accumulation patterns of pyruvate orthophosphate dikinase and phosphoenolpyruvate carboxylase in mesophyl! cells of illuminated dark-grown maize seedlings were similar (Fig. 5). Mesophyll cells of etiolated leaves contain about 10% as much of each protein as mesophyll cells of green leaves. The levels of pyruvate orthophosphate dikinase and phos- phoenolpyruvate carboxylase increased progressively in mes- ophyll cells after illumination and reached a plateau by 72 h of greening despite declines in their corresponding mRNAs. Bundle sheath cells had no detectable phosphoenolpyruvate carboxylase but contained about as much pyruvate ortho- phosphate dikinase as mesophyll cells of etiolated leaves (Fig. 5).

The protein levels of PPDK and PEPC genes are induced about 10-20-fold in mesophyll cells upon illumination of dark- grown maize seedlings. The maximum in accumulation of pyruvate orthophosphate dikinase and phosphoenolpyruvate carboxylase transcripts preceded, by 24 h, the maximum for accumulation of corresponding proteins and declined to a low steady-state level, whereas the protein levels stayed high. Thus, we have confirmed that the increased activity of phos- phoenolpyruvate carboxylase in greening maize seedlings (33) is related to the synthesis and accumulation of its mRNA and protein as described before (34-36); and, furthermore, we have demonstrated that those events occur only in mesophyll cells with the result that phosphoenolpyruvate carboxylase is com- partmentalized in mesophyll cells of green mature leaves (1- 5, 7, 29). It has been suggested that different forms of phos- phoenolpyruvate carboxylase are present in etiolated and greening leaves of C4 plant (37) and in different tissues of C3, C4, and crassulacean acid metabolism plants (38-41), and it has been shown recently that different PEPC genes are ex- pressed in maize leaves and roots (29, 30). We have found several PEPC genes on maize genomic blots, using the cDNAs described here as probes (29, 30).3 The possibility that there may be a C4 pathway specific PEPC gene that is expressed in mesophyll cells of green leaves needs further investigation.

J.-Y. Sheen, unpublished data.

C4 Genes in Bundle Sheath and Mesophyll Cells of Maize 11729

Transcripts of the PPDK gene in maize seedlings have been shown to increase upon illumination (29, 31). We have dem- onstrated here that the light inducibility is mesophyll cell- specific and similar to changes in phosphoenolpyruvate car- boxylase mRNA. Low levels of the mRNA and protein en- coded by the PPDK gene were clearly detected in bundle sheath cells. Since several PPDK genes are encoded in the maize genome (29); it is possible that different PPDK genes are expressed in bundle sheath and mesophyll cells or that the same gene is expressed a t different levels in the two cell types. Furthermore, the pyruvate orthophosphate dikinase detected in bundle sheath cells is probably different from the pyruvate orthophosphate dikinase found in seeds (32) and roots (29) since the former was synthesized as a 110-kDa precursor (corresponding to 3.3-kilobase pair transcripts) in- stead of the 94-kDa pyruvate orthophosphate dikinase (cor- responding to 3.0-kilobase pair transcripts) found in seeds and roots which is not synthesized as a precursor in vitro (29, 32). The PPDK gene expressed in bundle sheath cells may well correspond to the gene expressed in the leaves of Cs plants (9).

Mesophyll Cell-specific Expression of MDH Gene-By in vitro translation and immunoprecipitation analysis, we have also shown that the accumulation of translatable mRNA encoding NADP-malate dehydrogenase was induced by light in mesophyll cells. However, the effect of light on the accu- mulation of NADP-malate dehydrogenase mRNA seemed less dramatic than on the accumulation of phosphoenolpyruvate carboxylase mRNAs (Fig. 4). In contrast to pyruvate ortho- phosphate dikinase and phosphoenolpyruvate carboxylase, mesophyll cells of etiolated leaves have amounts of NADP- malate dehydrogenase close to those in 72-h greening leaves. No NADP-malate dehydrogenase was detected in bundle sheath cells (Fig. 5). Therefore, the growth of the NADP-

0 6 12244872966 0 6 12 24487296G h MrxlO-‘ B M

Ls-

ss-

F -66

I MDH .45

-3 1

-2 1

-14

FIG. 4. The mRNAs of MDH, PEPC, and ribulose-1.5-bi- phosphate carboxylase large (LS) and small (SS) subunit genes in bundle sheath and mesophyll cells of developing maize leaves assayed by in vitro translation and immunopre- cipitation. Five pg of total RNAs isolated from bundle sheath ( R ) and mesophyll ( M ) cells of etiolated (O), greening (12, 24,48, 72, and 96 h), and green (G) maize leaves were translated in oitro in 30 pl reaction mixtures. Antibodies to NADP-malate dehydrogenase ( M D H ) , phosphoenolpyruvate carboxylase (PEPC), ribulose-1,5-bi- phosphate carboxylase, large subunit, and small subunit (LS, S S ) were used to precipitate the [lsSS]methionine-labeled radioactive prod- ucts which were separated by gel electrophoresis. A fluorogram is shown. A 45-kDa polypeptide is a prominent endogenous product of the rabbit reticulocyte translation system we used; the band at about 45 kDa seen in both bundle sheath ( B ) and mesphyll ( M ) cell panels is probably this material occluded in the immunoprecipitate.

B 0 6 12XU)RB(IG 0 6 1 2 X U ) R B g G h

M

“=PC -pR*(

YI

-yHI

FIG. 5. Western blot analysis of cell-specific expression of PPDK, PEPC, ME, and MDH genes. Proteins were analyzed from %o-aliquots of the same set of bundle sheath ( R ) and mesophyll ( M ) cell samples as used for Northern blot analyses shown in Fig. 3; the standardization of all protein samples followed that for RNA samples proportionally. Antibodies to phosphoenolpyruvate carboxylase (PEPC), pyruvate orthophosphate dikinase (PPDK), ME, and NADP-malate dehydrogenase ( M D H ) were used to estimate the

described under “Materials and Methods.” relative amounts of a protein across the developmental program as

malate dehydrogenase transcript pool in mesophyll cells was induced by light during the first 6 h of illumination of dark- grown maize seedlings; however, the accumulation of NADP- malate dehydrogenase in mesophyll cells is not influenced by light up to 72 h of illumination. It has been shown that the activity of NADP-malate dehydrogenase is induced by light in greening maize seedlings as well as phosphoenolpyruvate carboxylase and ME (33). In the case of NADP-malate de- hydrogenase, unlike phosphoenolpyruvate carboxylase as dis- cussed above, the increased activity observed by Kobayashi (33) might have been due to the light activation of preexisting enzyme (42).

Bundle Sheath Cell-specific Expression of the M E Gene- The accumulation of transcripts of ME genes was stimulated by light in bundle sheath cells but quantitatively less than pyruvate orthophosphate dikinase and phosphoenolpyruvate carboxylase transcripts. Transcripts of t he ME gene were clearly detectable in bundle sheath cells of etiolated leaves. They increased about 5-fold (determined by densitometer scanning) after 24 h of illumination and declined only slightly after peaking at 24 h. The pool remained rather large in late- greening (72 and 96 h) and green leaves. No mRNA for M E was detected in mesophyll cells (Fig. 3). A low level of M E was present in bundle sheath cells of etiolated leaves and the ME level increased upon illumination (Fig. 5). No M E was detected in mesophyll cells. Unlike the rapid decline in pyru- vate orthophosphate dikinase and phosphoenolpyruvate car- boxylase mRNAs after 24-h greening and the low steady-state level of both mRNAs in green leaves, the level of ME mRNA decreased only slightly after 24-h greening and remained high in green leaves. This high level of ME transcripts suggests that ME may be less stable than pyruvate orthophosphate dikinase, phosphoenolpyruvate carboxylase, and NADP-ma- late dehydrogenase, as higher levels of steady-state mRNA are perhaps needed to maintain a high level of ME.

It has been shown that the amounts of pyruvate orthophos- phate dikinase, phosphoenolpyruvate carboxylase, and M E increase following the appearance of Kranz anatomy in a series of progressively more mature sections of leaves in light- grown maize seedlings and suggested that cell differentiation may play an important role in regulating the appearance of pyruvate orthophosphate dikinase, phosphoenolpyruvate car- boxylase, and ME (22-24). We have shown that the morpho- genesis of bundle sheath and mesophyll cells is completed in etiolated leaves (10) and only determines the potential for the PPDK, PEPC, and ME genes to be induced by light in mesophyll and bundle sheath cells. The full differential expression of PPDK, PEPC, and M E genes in mesophyll and

11730 C, Genes in Bundle Sheath and Mesophyll Cells of Maize

bundle sheath cells is a light-dependent event and expression of these genes parallels the development of etioplasts into dimorphic chloroplasts in bundle sheath and mesophyll cells during greening rather than the morphogenesis of the two cell types.

In summary, we have prepared cDNA clones of three genes encoding the C., pathway enzymes, pyruvate orthophosphate dikinase, phosphoenolpyruvate carboxylase, and M E and de- termined the expression patterns of these genes in bundle sheath and mesophyll cells of etiolated, greening, and green maize leaves. For NADP-malate dehydrogenase, we have used antibodies to follow pools of translatable mRNA and levels of proteins in vivo. We conclude that there is a tight temporal correlation between protein and mRNA accumulation induced by light for PPDK, PEPC, and ME genes except that mRNA accumulation is transient for pyruvate orthophosphate diki- nase and phosphoenolpyruvate carboxylase, whereas the amount of the protein remains about constant after accumu- lation. The proteins appear to be relatively stable; sizes of the mRNA pools change. The changes are cell-type specific and result in the localization of pyruvate orthophosphate dikinase and phosphoenolpyruvate carboxylase in mesophyll cells and ME in bundle sheath cells. In other words, the primary regulation of the C4 pathway-specific PPDK, PEPC, and ME genes appears to be mainly via light modulation of transcript pools. In contrast, the mesophyll cell-specific localization of NADP-malate dehydrogenase is only slightly influenced by light inasmuch as NADP-malate dehydrogenase is already present in mesophyll cells of etiolated leaves. However, a slight transient increase in NADP-malate dehydrogenase mRNA was observed in mesophyll cells, although little change in protein level occurred upon greening.

Acknowledgments-We are grateful to Dr. H. Roy, Dr. P. Collins, and Dr. B. B. Buchanan for providing antisera to pyruvate ortho- phosphate dikinase, ME, and NADP-malate dehydrogenase, respec- tively.

REFERENCES

1. Edwards, G. E., and Huber, S. C. (1981) in The Biochemistry of Plants: Photosynthesis (Hatch, M. D., and Broadman, N. K., eds) Vol. 8, pp. 237-281, Academic Press, Orlando, FL

2. Hatch, M. D. (1976) in Plant Biochemistry (Bonner, J., and Varner, J. E., eds) 3rd Ed., pp. 797-844, Academic Press, Orlando, FL

3. Hatch, M. D. (1978) Curr. Top. Cell. Regul. 1 4 , 1-27 4. Slack, C. R., Hatch, M. D., and Goodchild, D. J. (1969) Biochem.

5. Perrot-Rechenmann, C., Vidal, J., Brulfert, J., Burlet, A., and Gadal, P. (1982) Planta (Berl.) 1 5 5 , 24-30

6. Perrot-Rechenmann, C., Jacquot, J. P., Gadal, P., Weeden, N. F., Cseke, C., and Buchanan, B. B. (1983) Plant Sci. Lett. 3 0 ,

7. Broglie, R., Coruzzi, G., Keith, B., and Chua, N.-H. (1984) Plant

8. Aoyagi, K., and Nakamoto, H. (1985) Plant Physwl. (Bethesda)

9. Aoyagi, K., and Bassham, J. A. (1986) Plant Physiol. (Bethesdu)

J. 1 14,489-498

219-226

Mol. Biol. 3, 431-444

78,661-664

80,334-340

10. Sheen, J.-Y., and Bogorad, L. (1985) Plant Physiol. (Bethesda) 79. 1072-1076

11. Aviv,'H., andLeder, P. (1972) Proc. Natl. Acad. Sci. U. S. A. 69,

12. McCarty, K. S., Jr., Vollmer, R. T., and McCarty, K. S. (1974)

13. Laemmli, U. K. (1970) Nature 227,680-685 14. Gubler, U., and Hoffman, B. J. (1983) Gene (Amst.) 25,263-269 15. Birnboim, H. C., and Doly, J. (1979) Nucleic Acids Res. 7,1513-

16. Maas, R. (1983) Plasmid 10 , 296-298 17. Kafatos, F. C., Jones, C. W . , and Efstratiadis, A. (1979) Nucleic

Acids Res. 7 , 1541-1552 18. Parnes, J. R., Velan, B., Felsenfeld, A., Ramanathan, L., Ferrini,

Sci. U. s. A. 7 8 , 2253-2257 U., Appella, E., and Seidman, J. G. (1981) Proc. Natl. Acad.

19. Franssen, H., Goldbach, R., Broekhuijsen, M., Moerman, M., and van Kammen, A. (1982) J. Virol. 41,8-17

20. Collins, P. D., and Hague, D. R. (1983) J. BioL Chern. 258,4012- 4018

21. Jacquot, J.-P. P., Buchanan, B. B., Martin, F., and Vidal, J. (1981) Plant Physwl. (Bethesda) 68,300-304

22. Aoyagi, K., and Bassham, J. A. (1986) Plant Physwl. (Bethesda)

23. Mayfield, S. P., and Taylor, W . C. (1984) Planta (Berl.) 161 ,

24. Martineau, B., and Taylor, W . C. (1985) Plant Physiol. (Bethesda)

25. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76,4350-4354

26. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

27. Miller, J. S., Paterson, B. M., Ricciardi, R. P., Cohen, L., and Roberta, B. E. (1983) Methods Enzymol. 101,650-674

28. Izui, K., Ishijima, S., Yamaguchi, Y., Katagiri, F., Murata, T., Shigesada, K., Sugiyama, T., and Katsuki, H. (1986) Nucleic Acids Res. 14 , 1615-1628

29. Hudspeth, R. L., Glackin, C. A., Bonner, J., and Grula, J. W. (1986) Proc. NatL Acad. Sci. U. S. A. 8 3 , 2884-2888

30. Harpster, M. H., and Taylor, W . C. (1986) J. Biol. Chem. 261,

31. Hague, D. R., Uhler, M., and Collins, P. D. (1983) Nucleic Acids

32. Aoyagi, K., and Bassham, J. A. (1984) Plant Physwl. (Bethesda)

33. Kobayashi, H., Asami, S., and Akazawa, T. (1980) Plant Physwl.

34. Hague, D. R., and Sims, T. L. (1980) Plant Physiol. (Bethesda)

35. Sims, T. L., and Hague, D. R. (1981) J. Biol. Chern. 256,8252- 8255

36. Nelson, T., Harpster, M. H., Mayfield, S. P., and Taylor, W. C. (1984) J. CeU Biol. 98,558-564

37. Gadal, P., Vidal, J., Bouville, J., and Godbillon, G. (1981) in Photosynthesis (Akoyunoglou, C., ed) Vol. 4, pp. 81-90, Balaban

38. Ting, I. P., and Osmond, C. B. (1973) Plant Physwl. (Bethesda) International Science Service, Philadelphia

39. Ting, I. P., and Osmond, C. B. (1973) Plant Physwl. (Bethesda)

40. Uedan, K., and Sugiyama, T. (1976) Plant Physwl. (Bethesda)

41. Mukeji, S. K. (1977) Arch. Biochem. Biophys. 182,343-351 42. Edwards, G. E., Nakamoto, H., Burnell, J. N., and Hatch, M. D.

1408-1412

Anal. Biochem. 61,165-183

1523

80,322-333

481-486

78,399-404

6132-6136

Res. 11,4853-4865

76.278-280

(Bethesda) 65, 198-203

66,505-509

61,439-447

61,448-453

57,906-910

(1985) Annu. Rev. Plant Physwl. 36,255-286