WO1991009050A1 - Novel organ-specific plant promoter sequences - Google Patents

Abstract

The present invention relates to novel plant organ-specific transcriptional promoter nucleic acid sequences, which regulate the expression of glutamine synthetase isoenzymes. Specifically, promoter sequences were isolated from the nuclear gene for chloroplast GS2 glutamine synthetase and from two nuclear genes for cytosolic GS3 glutamine synthetase in the pea plant, Pisum sativum. Accordingly, the present invention provides for the nucleic acid sequences of the GS2, GS3A and GS3B promoter sequences as well as functional portions thereof. The invention further provides for promoters homologous to GS2, GS3A and GS3B, gene fusions comprising the novel glutamine synthetase promoters, and transgenic plants which comprise the promoters of the invention.

Description

NOVEL ORGAN-SPECIFIC PLANT PROMOTER SEQUENCES

1. INTRODUCTION

The present invention relates to novel plant organ-specific transcriptional promoter nucleic acid sequences which regulate the expression of glutamine synthetase isoenzymes. In particular, one promoter sequence which responds to light and two promoter sequences which are selectively active in plant vascular elements have been cloned and characterized. The promoter sequences of the invention may be used to control the expression of glutamine synthetase as well as heterologous proteins, and may

advantageously be used to render plants resistant to

herbicides or viral or pathogen infection.

2. BACKGROUND OF THE INVENTION

a. THE GLUTAMINE SYNTHETASE SYSTEM

In higher plants, many steps in nitrogen metabolism occur in multiple subcellular compartments. For example, isoenzymes for many amino acid biosynthetic enzymes arelocated in the cytosol, as well as in the mitochondria or chloroplasts. The significance of this multiplicity and compartmentalization of plant isoenzymes has yet to be fully understood. The relative function of many amino acid biosynthetic isoenzymes has been difficult to assess due to inadequate fractionation of organelle and cytoplasm

components, overlapping activity profiles, and immunological cross-reactivity (Miflin and Loa, 1982, in "Nucleic Acid and Proteins in Plants I: Structure, Biochemistry and Physiology of Proteins," eds. Boulter, D. & Parthier, B., Springer-Verlag, Berlin Heidelberg New York, pp. 5-64).

Consequently, it is unclear whether these isoenzymes carry out redundant or distinct roles in plant metabolism. The best studied example of a plant amino acid

biosynthetic enzyme shown to occur as multiple isoforms is glutamine synthetase (GS) (EC 6.3.1.2) (McNally et al., 1983, Plant Physiol. 72:22-25). Early biochemical data revealed that GS functions in the assimilation of ammoniagenerated by numerous plant processes which include seed germination (Kern and Chrispeels, 1978, Plant Physiol.

62:642-647; Winter et al., 1982, Plant Physiol. 69:41-47), photorespiration (Wallsgrove et al., 1983, Plant Cell

Environ. 6: 301-309; Wallsgrove et al., 1987, Plant Physiol.83: 155-158), nitrite reduction (Miflin, 1974, Plant Physiol. 54: 550-555), nitrogen-fixation in root nodules (Robertson et al., 1975, Aust. J. Plant Physiol. 2:265-272; Lara et al., 1983, Plants 157: 254-258), and primary ammonia

assimilation from the soil (Hirel and Gadal, 1980, Plantphysiol. 66: 619-623). An analysis of the GS genes in several species has revealed a strong correlation of

individual GS gene expression with specific aspects of plant development (Tingey et al., 1987, EMBO J. 6:1-9; Tingey et al., 1988, J. Bio. Chem. 263:9651-9657; Hirel et al., 1987,EMBO J. 6:1167-1171; Forde et al., 1989, Plant Cell 1:391- 401; Gebhardt et al., 1986, EMBO J. 5:1429-1435; Edwards and Coruzzi, 1989, Plant Cell 1:241-248). Recent sequence analysis of GS cDNAs from Pisum sativum and Phascolus vulgaris has shown that chloroplast and cytosolic GS are encoded by separate but similar nuclear genes (Tingey et al., 1987, EMBO J. 6:1-9; Tingey et al., 1988, J. Bio. Chem. 263:9651-9657; Cullimore et al., 1984, J. Mol. Appl. Genet. 2:589-599).

In pea, the single nuclear gene for chloroplast GS2 is expressed predominantly m leaves in a light-dependent fashion (Tingey et al., 1988, J. Bio. Chem. 263:9651-9657;

Edwards and Coruzzi, 1989, Plant Cell 1:241-248). The role of chloroplast GS2 in the reassimilation of photorespiratory ammonia is supported by the analysis of mutants in barley (Wallsgrove et al., 1987, Plant Physiol. 83:155-158), and is substantiated by gene expression studies in pea (Edwards and Coruzzi, 1989, Plant Cell 1:241-248). For cytosolic GS, molecular studies have revealed the presence of a number of distinct isoforms in several plant species (Tingey et al.,1988 J. Bio. Chem. 263:9651-9657; Hirel et al., 1987, EMBO J. 6 : 1167-1171; Gebhardt et al., 1986, EMBO J. 5:1429-1435; Tingey and Coruzzi, 1987, Plant Physiol. 84:366-373). In pea it has been shown that two classes of genes encode homologous but distinct cytosolic GS isoforms (Tingey, 1988, J. Bio. Chem. 263:9651-9657). One class comprises a pair of "twin" GS genes whose expression is specifically induced in two developmental contexts where large amounts of ammonia are mobilized for plant growth, during germination and nitrogen fixation. b. PLANT PROMOTER/ENHANCER SEQUENCES A number of plant promoter enhancer/sequences have been identified, including light-responsive promoter

sequences such as ribulose bisphosphate carboxylase (Coruzzi et al., 1984, EMBO J. 3: 1671-1680; Herrera-Estrella et al., 1984, Nature 310:115-120), the chlorophyll a/b binding protein (Cab) of the light-harvesting chlorophyll-protein complex (Apel et al., 1978, Eur. J. Biochem. 85:581-588;

Stiekema et al . , 1983 , Plant Physiol . 72 : 717-724 ; Thompsonet al., 1983, Plants 158: 487-500; Jones et al., 1985, EMBO J. 4 : 2411-2418) and the ST-LSl gene of potato (Stockhaus et al., 1989, Plant Cell 3.:805-814). Additional plant promoter sequences include the soybean heat shock protein hsp17.5-E or hsp17.3-B promoters (Gurley et all, 1986, Mol. Cell Biol.6:559-565); the Parasponia andersoni hemoglobin promoter

(Landsmann et al., 1988, Mol. Gen. Genet. 214:68-73); the phenylalanine ammonia-lyase promoter, which appears to be active in specific cell types which accumulate phenyl- propanoid derivatives in response to wounding and also during normal development of the xylem and flower (Bevan et al., 1989, EMBO J. 8:1899-1906); and the petunia 5- enolpyruvylshikimate-3-phosphate synthase gene promoter (Benfey and Chua, 1989, Science 244:174-181). Certain plant promoters, such as patatin, have also been shown to function in specialized organs such as tubers (Rocha-Sosa et al., 1989, EMBO J. 8:23-29).

Certain bacterial promoters have been observed to be expressed in plants, including the Rhizobium meliloti FIXD gene promoter described in U.S. Patent No. 4,782,022, issuedNovember 1, 1988, by Puhler et al., and the nopaline

synthase promoter (Ha and An, 1989, Nucleic Acids Res.

17:215-224; An et al., 1988, Plant Physiol. 88:547-552). Several promoter sequences, termed the rol A, B and C promoters, have been identified in Agrobacterium rhizogenes (Schmulling et al., 1989, Plant Cell 1:665-670; Sugaya et al., 1989, Plant Cell Physiol. 30: 649-654). The rol C promoter described by Sugaya et al. (supra), located on the bacterial Ri plasmid, has been observed to be expressed in phloem cells.

Viral genes which are expressed in plants include the patatin promoter (Rocha-Sosa et al., 1989, EMBO J. 8:23-29) and the cauliflower mosaic virus (CaMV) 35S promoter (Odell et all, 1985, Nature 313:810-812; Jensen et al., 1986, Nature 321:669-674; Jefferson et al., 1987, EMBO J. 6 :

3901-3907; and Sanders et al., 1987, Natl. Acids Res.

14: 1543-1558). Within the CaMV 35S promoter, expression conferred by domain A (-90 to +8) was found to be

particularly strong in root tissue, whereas expression conferred by domain B (-343 to -90) appeared to be strongest in the cotyledons of seeds and seedlings and in the vascular tissue of the hypocotyl (Benfey et al., 1989 EMBO J.

8:2195-2202).

3. SUMMARY OF THE INVENTION The present invention relates to novel plant organ- specific transcriptional promoter nucleic acid sequences, which regulate the expression of glutamine synthetase isoenzymes. Specifically, promoter sequences were isolated from the nuclear gene for chloroplast GS2 glutamine

synthetase and from two nuclear genes for cytosolic GS3 glutamine synthetase in the pea plant, Pisum sativum.

Accordingly, the present invention provides for the nucleic acid sequences of the GS2, GS3A and GS3B promoter sequences as well as functional portions thereof.

The invention further provides for promoters

homologous to GS2, GS3A and GS3B, gene fusions comprising the novel glutamine synthetase promoters, and transgenic plants which comprise the promoters of the invention.

Experiments which tested the activity of the GS2, GS3A and GS3B promoter sequences revealed that the GS2 promoter was inducible by light and directed high levels of

transcription in photosynthetic cells of leaves. In

addition, the GS2 promoter directs expression in

nonphotosynthetic cells of the root tip. GS3A was found to be selectively active in phloem; therefore, GS3A represents the first plant-derived, phloem-specific promoter element. In various embodiments of the invention, the GS2, GS3A, and GS3B promoter sequences may be used to control the

expression of glutamine synthetase as well as heterologous proteins in a tissue specific and/or light-inducible manner. The resulting tissue-specific expression of a desired gene product presents a wide range of potential applications for the promoter sequences of the invention, including, but not limited to, the manipulation of nutritional requirements and the induction of resistance to herbicides or pathogens. a. ABBREVIATIONS

transgenic plant: a plant which has incorporated a foreign gene into its genome transgene = transgenic sequence: a foreign gene which has been incorporated into a transgenic plant

4. DESCRIPTION OF THE FIGURES

Figure 1. GS-GUS Translational Fusions. A) pGS2^ct- GUS contains 1.5 kb of the gene for chloroplast GS2 in a translational fusion with the GUS gene of pBI101.2. B) pGS3A^cy-GUS contains 1.01 kb of the gene for cytosolic GS3A in a translational fusion with the GUS gene of pBI111. A 3' polyadenylation region from the nopaline synthase gene is present in both GS-GUS constructs and is denoted with diagonal stripes. The white areas represent the 5'

noncoding region of each GS gene. The solid black areas depict GS coding regions and the dotted area marks the GUS coding region.

Restriction sites: E = EcoRI, H = HincII, Bg = Bg1ll.

Restriction sites in parenthesis indicate original sites in plant genes which were destroyed in plasmid construction.

Figure 2. Histochemical localization of GUS activity in cross sections of mature transgenic tobacco plants.

Panels A-D represent sections from pGS^ct-GUS transformants; A) leaf cross section, B) leaf blade cross section, C) leaf midrib cross section, D) stem cross section. Panels E-H represent sections from pGS3A^cy-GUS transformants: E) leaf cross section, F) leaf midrib cross section, G) root cross section, H) stem cross section. Abbreviations: CH- chlorenchyma, CL - collenchyma, E - epidermis, LB - leaf blade, MV - midvein, PH - phloem, PP - palisade parenchyma, PT - pith parenchyma, R - root, SP - spongy parenchyma, T - trichome, V - vasculature, X - xylem.

Figure 3. Histochemical localization of GUS activity in whole mounts of 7 day-old transgenic tobacco seedlings; A) pGS2^ct-GUS transformant. B) pGS3A^cy-GUS transformant. C) control, pBI101 transformant. Abbreviations: C - cotyledon, H - hypocotyl, L - leaf, R - root, V - vasculature.

Figure 4. Ribonuclease T2 protection analysis of pGS2^ct-GUS transcripts in light- vs. dark-grown transgenic tobacco. Autoradiograph of the 162 nt fragment protected from RNAse T2 digestion in hybridizations containing 50 μg of total RNA isolated from: A) and B) two separate pGS2^ct-

GUS transformants; C) and a control, pBI101 transformant which was dark-adapted for 4 days (lanes 1 and 3), and subsequently grown in continuous white light for 24 hrs.

(lanes 2, 4, and 5).

Figure 5. Nucleotide sequence of A) GS2; B) GS3A; and C) GS3B promoter elements.

Figure 6. Promoters for A) GS2, B) GS3A, and C) GS3B subcloned into Bluescript^® SK⁺ vectors. Gene sequences are in upper case letters, vector and linker sequences are in lower case letters. Restriction sites destroyed in cloning are marked in parentheses. All restriction sites are underlined. Numbers refer to the nucleotides of GS promoter as spedcified in Figure 5.

5. DETAILED DESCRIPTION OF THE INVENTION The present invention relates to novel plant organ- specific transcriptional promoters which control the

expression of glutamine synthetase isoenzymes, and provides for (i) promoter elements such as GS2, GS3A and GS3B, which have nucleotide sequences substantially as depicted in

Figure 5, or nucleotide sequences homologous thereto; (ii) gene fusions comprising these promoter elements; (iii) methods for producing tissue-specific expression of

glutamine synthetase or heterologous proteins utilizing the novel promoters; and iv) transgenic plants comprising transgenes which include the promoter elements of the invention. The invention is based, in part, on the discovery that the promoters for chloroplast GS2 and

cytosolic GS3 of Pisum sativum confer non-overlapping, cell-specific expression patterns to the beta-glucuronidase (GUS) reporter gene (Jefferson et al., 1986, Proc. Natl. Acad. Sci. U.S.A. 83:8447-8451) in transgenic tobacco plants (See Section 6, infra). The promoter for chloroplast GS2 was observed to direct GUS expression within photosynthetic cell types (e.g. palisade parenchymal cells of the leaf blade, chlorenchymal cells of the midrib and stem, and in photosynthetic cells of tobacco cotyledons). The promoter for chloroplast GS2 appeared to retain the ability to confer light-regulated gene expression in the heterologous

transgenic tobacco system in a manner analogous to the light-regulated expression of the cognate gene for

chloroplast GS2 in pea. These expression patterns may reflect the physiological role of the chloroplast GS2 isoform in the assimilation of ammonia generated by nitrite reduction and photorespiration. In contrast, the promoter for cytosolic GS3A was found to direct expression of GUS specifically within the phloem elements in all organs of mature plants. This phloem-specific expression pattern suggests that the cytosolic GS3A isoenzyme may function to generate glutamine for intercellular nitrogen transport. In germinating seedlings, the intense expression of the

cytosolic GS3A^cy-GUS transgene in the vasculature of

cotyledons suggests a role for cytosolic GS3A in the

mobilization of nitrogen from seed storage reserves.

For purposes of clarity of disclosure, and not by way of limitation, the invention is described in the following subsections:

(i) Identification of promoter elements associated with glutamine synthetase genes;

(ii) Novel glutamine synthetase promoter elements;

(iii) Gene fusions comprising glutamine synthetase

promoter elements; and (iv) Creation of transgenic plants comprising

recombinant glutamine synthetase promoter

elements;

(v) Utility of the invention. a. IDENTIFICATION OF PROMOTER ELEMENTS ASSOCIATED

WITH GLUTAMINE SYNTHETASE GENES

According to the present invention, promoter elements associated with glutamine synthetase genes may be identified from any species of plant, bacteria, or virus using any method known in the art. For example, genomic DNA libraries may be screened for clones comprising sequences homologous to known glutamine synthetase genes or, alternatively, known glutamine synthetase promoter sequences. Thus, cDNA clones corresponding to mRNA which encodes glutamine synthetase, or oligonucleotide probes corresponding to known glutamine synthetase amino acid sequence may be used to identify homologous clones in a genomic DNA library using methods such as, for example, the method set forth in Benton and Davis (1977, Science 196:180) for bacteriophage libraries, and Grunstein and Hogness (1975, Proc. Natl. Acad. Sci.

U.S.A. 72:3961-3965) for plasmid libraries. Retrieved clones may then be analyzed by restriction-fragment mapping and sequencing techniques according to methods well known in the art.

In specific embodiments of the invention, the

nucleotide sequences of GS2, GS3A, GS3B (see Figure 5), or portions thereof, or nucleotide sequences homologous

thereto, or portions of the promoters as found in the plasmids described in Figure 6, may be used to identify genomic clones comprising homologous promoter elements, using the standard techniques described supra. For example, homologous glutamine synthetase promoters in pea or in other species of plant, bacteria, or other organism may be

identified in this manner. b. NOVEL GLUTAMINE SYNTHETASE PROMOTER ELEMENTS

Figure 5 sets forth the nucleotide sequences of the GS2, GS3A, and GS3B promoter elements. The present

invention provides for recombinant DNA constructs which comprise nucleic acid sequences substantially as depicted in Figure 5, functional portions thereof, and nucleotide sequences homologous thereto. Functional portions of the glutamine synthetase promoters described herein refers to regions of the nucleic acid sequence which are capable of promoting transcription under a specific set of conditions, in a particular cell type, or otherwise. Nucleotide

sequences homologous to the glutamine synthetase promoters described herein refers to nucleic acid sequences which are capable of hybridizing to the nucleic acid sequences

depicted in Figure 5 in standard hybridization assays or are homologous by sequence analysis (containing a span of 10 or more nucleotides in which at least 50 percent of the

nucleotides are identical to the sequences presented

herein). Homologous nucleotide sequences refer to

nucleotide sequences including, but not limited to,

glutamine synthetase promoter elements in diverse plant species as well as genetically engineered derivatives of the promoter elements described herein. According to the latter embodiment, it may be found that altering the sequence of the promoter element may result in a change in promoter activity, such as an increase or decrease in promoter strength or a different pattern of cell or organ-specific expression. Such engineered promoter elements may be used, according to the invention, to design optimal gene fusion systems for a variety of applications. c. GENE FUSIONS COMPRISING GLUTAMINE

SYNTHETASE PROMOTER ELEMENTS

The glutamine synthetase promoter elements of the invention may be used to direct the expression of glutamine synthetase or any other protein, referred to as a

heterologous protein or as a desired protein, or the

expression of an RNA product, including, but not limited to, an "antisense" RNA. Recombinant constructs may comprise a glutamine synthetase promoter element and nucleic acid sequence encoding a desired protein. Such constructs may or may not encode amino acids of glutamine synthetase, and should be designed so that the nucleic acid encoding the desired protein is in phase with contiguous sequences such that translation will result in an amino acid sequence which correlates with the amino acid sequence of the native form of the desired protein.

In various embodiments of the invention, it may be desirable to include additional nucleic acid sequences in the glutamine synthetase promoter recombinant constructs. Such additional nucleic acid sequences include, but are not limited to, sequences encoding 5' untranslated mRNA

sequence, such as a ribosome binding site; an intron, 3' untranslated sequence, such as a polyadenylation signal; sequence encoding an initiation codon or a signal peptide (which facilitates secretion of the desired protein); and targeting peptides, such as peptides which target the desired protein to chloroplasts or to the cell nucleus. In preferred embodiments of the invention which utilize the

Agrobacterium tumefaciens system for plant transformation, the recombinant construct of the invention may comprise the left and right T-DNA border sequences.

In addition, the recombinant constructs of the

invention may include a selectable marker for propagation of the constructs. For example, if the construct is to be propagated in bacteria, it may comprise a gene for antibiotic resistance. Suitable vectors for propagating the construct would include plasmids, cosmids, and viruses, to name but a few. d. CREATION OF TRANSGENIC PLANTS

COMPRISING RECOMBINANT GLUTAMINE

SYNTHETASE PROMOTER ELEMENTS

In preferred embodiments of the invention, the

Agrobacterium tumefaciens gene transfer system may be used to introduce the recombinant constructs of the invention into plants; generally, this system may be utilized to transfer DNA into dicotyledonous plants (Bevan et al., 1982,

Ann. Rev. Genet. 16:357-384; Rogers et al., 1986,Methods

Enzymol. 118:627-641; Fraley et al., 1986, CRC Crit. Rev.

Plant Sci. 4:1-46; Hooykaas et al., 1984, Adv. Genet.

22:210-283; Nester et al., 1984, Ann. Rev. Plant Physiol.

35:387-413). To this purpose, vectors such as, but not limited to, binary Agrobacterium vectors for plant

transformation may be utilized, such as, for example, the vector described by Bevan (1984, Nucl. Acids Res. 12:8711-

8721). Xanthi may be transformed by a leaf inoculation procedure such as that described by Horsch et al. (1985,

Science 227:1229-1231).

Additional methods for introducing DNA into plants may also be utilized, particularly if the recombinant construct is to be used to create a transgenic monocotyledonous plant.

Such methods would include, but are not limited to

poly(ethylene glycol) and calcium-mediated uptake of naked

DNA (Hain et al., 1985, Mol. Gen. Genet. 199:161-168;

Paszkowski et al., 1984, EMBO J. 3 : 2717-2722 ; Potrykus etal., 1985, Mol. Gen. Genet. 199:169-177), electroporation

(Fromm et al., 1985, Proc. Natl. Acad. Sci. U.S.A. 82:5824- 5828), microinjection and cell gun. In order to identify successful transformants, it may be desirable to transform host cells with a second construct which comprises a selectable marker or reporter gene. The second construct may be introduced separately or in tandem with the construct which comprises the glutamine synthetase promoter and the sequence encoding the desired gene product. If not constructed in tandem, the second construct may also comprise a glutamine synthetase promoter, such that

expression of selectable marker or reporter gene function may serve as an indicator of promoter activity and thereby provide evidence that the gene encoding the desired gene product is actively transcribed. This may be useful when a desired protein may have activity which is difficult to identify (e.g. the desired protein influences crop yield). Selectable markers would include genes which engender antibiotic resistance (for example, kanamycin resistance) or which encode a reporter gene, including but not limited to the gene for beta-glucuronidase (Jefferson, 1987, Plant Mol. Bio. Rep. 5:387-405), neomycin phosphotransferase (NPT II), and luciferase (Ow et al., 1986, Science 234:856-859) to name but a few. Detection of reporter gene expression may then be performed using methods known in the art.

Alternatively, transformants may be tested for the presence of recombinant construct by methods which would identify foreign DNA sequences, such as the Southern blot procedure. Transcription of recombinant constructs could be detected by isolating RNA from the transformant and

screening for the expected transcript by Northern blot or RNA protection experiments (see Section 6.1.6., infra).

Likewise, translation of the desired protein could be detected by protein gel electrophoresis, Western blot techniques, immunoprecipitation or enzyme-linked

immunoaεsays. Using similar techniques, the expression of the recombinant constructs of the invention may be detected in specific plant organs or tissues by determining the presence of RNA, protein, selectable marker, or reporter gene which may serve as an indicator of transcription resulting from recombinant glutamine synthetase promoter activity.

In specific embodiments of the invention, the GS2 promoter or GS2 homologous sequences may be detectably active in tissues such as photosynthetically active cells, including the palisade and spongy parenchymal cells of the leaf blade, in collenchymal and chlorenchymal cells of the stem, in photosynthetic cotyledons, and, at low levels, in root tips. Light-enhanced promoter activity may be

detected. In a specific embodiment of the invention, tissue specific and light-enhanced activity of the GS2 promoter and its equivalents may be detected by the expression of the reporter gene beta-glucuronidase (Jefferson, 1987, Plant Mol. Biol. Rep. 5:387-405; see Section 6, infra).

In further embodiments of the invention, the GS3A or GS3B promoters or homologous sequences may be detectably active in vascular tissues such as leaves, stems, and roots of the mature plant, and in the cotyledons and roots of developing seedlings. In a specific embodiment of the invention, tissue specific activity of the GS3 promoters and their equivalents may be detected by the expression of the reporter gene beta-glucuronidase (Jefferson, 1987, Plant Mol. Biol. Rep. 5:387-405; see Section 6, infra). e. UTILITY OF THE INVENTION

The present invention may be utilized to direct the expression of glutamine synthetase or heterologous proteins using novel plant organ specific promoter elements.

According to specific embodiments of the invention, the GS2,

GS3A, or GS3B promoter elements, functional portions

thereof, or sequences homologous thereto , may be used to direct the expression of glutamine synthetase or

heterologous gene products via recombinant nucleic acid constructs.

In particular embodiments of the invention, the novel promoter elements may be used to direct the expression of glutamine synthetase. Such embodiments may be useful in the engineering of plants which are genetically deficient in endogenous synthesis of glutamine synthetase or which may benefit from over production of glutamine synthetase, and may be used to introduce an isoenzyme of glutamine

synthetase into a cellular compartment normally occupied by an alternate isoenzyme. Furthermore, engineered forms of the promoters of the invention which result in greater activity or altered tissue distribution of promoter

function, may be used to alter expression patterns of glutamine synthetase. Manipulation of glutamine synthetase production may be advantageously used to confer herbicide resistance (see infra).

According to particular embodiments of the invention, the GS2 promoter and its functional or homologous

equivalents may be used to provide tissue specific and light enhanced expression of desired proteins or gene products (e.g., antisense RNA). The GS2 promoter is selectively active in photosynthetic tissues. The tissue selectivity of the GS2-like promoters may be utilized to express desired proteins or gene products in photosynthetic tissues. The gene products may impact on the physiology of the plant (i.e., alter the size, growth rate, or density of

photosynthetic tissues) or may introduce a molecule which does not naturally occur in the tissue, such as an insect repellant or fungus-retardant agent. The fact that GS2 promoter activity is enhanced by light provides the

opportunity to control the onset, duration, and termination of expression of a desired gene product; this capability may be especially useful when a critical period of development of a plant or plant product exists and exposure to a gene product (e.g. a growth horme expressed under the control of a GS2 promoter) may optimally be used to alter the

properties of the plant or plant product.

According to further particular embodiments of the invention, the GS3 promoters and their functional or

homologous equivalents may be used to provide phloem

specific expression of desired proteins or gene products. Phloem specific expression of desired gene products may be used, for example, to alter plant metabolism; since plant nutrients (carbon and nitrogen-containing compounds) are transmitted via the phloem, expression of foreign genes specifically within the phloem could be used to increase the efficiency of nutrient uptake. In further embodiments of the invention, phloem-specific expression could be utilized in the mass production of foreign proteins (including, for example, lymphokines or antibody molecules) which could be recovered from the phloem exudate by "bleeding".

GS2 and GS3 promoter elements may be useful in

imparting resistance to viral diseases to transgenic plants. Although viral infection in plants is not completely

understood, it is believed that viruses move through plants either by short distance cell to cell spread (through plasmodesmata) or by dissemination over longer distances via the plant vascular system. It has been suggested that if a virus has a specific relationship to a plant tissue, it is most commonly associated with phloem tissue; both phloemspecific as well as nonrestricted viruses have been

identified in the phloem (Esau, 1969, in "The Phloem,"

Gebruder Borntraeger, Berlin, pp. 252-262). A role for viral coat protein in preventing long-distance transport of virus has been observed in a variety of systems. Plants transformed to express tobacco mosaic virus protein were found to be resistant to tobacco mosaic virus infection;

interestingly, efficient movement of virus through plant stem was found to be prevented by grafting a section of plant expressing viral coat protein into the movement path of the virus (Baulcombe and Hull, 1989, Nature 341:189).

Tuner et al. (1987, EMBO J. 6:H81-1181) found that tobacco and tomato plants which expressed chimeric alfalfa mosaic virus (AlMV) coat protein were significantly delayed in symptom development after exposure to infectious AlMV, and some escaped infection altogether. Hemenway et al., (1989, in "Discoveries in Antisense Nucleic Acids," Brakel, ed., Gulf Publication Co., Houston, TX, pp. 165-174) demonstrated protection against virus infection in transgenic plants expressing the viral coat protein or corresponding antisense RNA from tobacco mosaic virus, cucumber mosaic virus, and potato virus X. In specific embodiments of the invention, the GS2 and/or GS3 promoter elements may be used to express viral coat protein or the corresponding antisense mRNA in viral target tissues. Since viral spread appears to occur, at least in part, via the phloem, in a preferred embodiment a GS3 promoter element may be used to direct phloem-specific expression of a viral coat protein or its corresponding mRNA.

Furthermore, non-viral pathogens including

mycoplasma-like organisms (MLO's) are also transmitted by phloem. MLO's cause a severe plant disease called "yellows" which devastates many citrus crops. GS3 phloem specific promoter may be used to express protein or nucleic acid which negatively affects MLO expression.

In still further embodiments of the invention, GS2 and GS3 promoter elements may be used to develop plants which are resistant to herbicides. Like viruses, many herbicides are transported through plants via phloem tissue.

Furthermore, many of the newer, highly potent herbicides inhibit plant growth by interfering with the biosynthesis of essential amino acids rather than by inactivating a

component of the photosynthetic apparatus (Shah et al., 198, in "Temporal and Spatial Regulation of Plant Genes", Verma and Goldberg, eds., Springer-Verlag, NY); as a result, these new herbicides have a broad spectrum of activity which discriminates poorly between weeds and crops. Several of these herbicides are directed at glutamine synthetase activity and/or are directed toward enzyme expressed in chloroplasts.

For example, glyphosate (N-[phosphonomethyl]glycine) is a broad spectrum nonselective herbicide which inhibits 5-enolpyruvylshikimate-3-phosphate (EPSP) synthase, an enzyme normally localized in the chloroplast; overproduction of unaltered or altered EPSP synthase, targeted to

chloroplasts via a transit peptide, appeared to inhibit glyphosate toxicity (Shah et al., 1986, Science 233:478-481; Steinrucken and Amrhein, 1980, Biochem. Biophys. Res.

Commun. 94:1207; Rubin et al., 1984, Plant Physiol. 75:839).

Phosphinothricin, an herbicide derived from a

Streptomyces tripeptide antibiotic is structurally similar to glutamine and glutamate, and is a competitive inhibitor of glutamine synthetase (La Rossa and Falco, 1984, Trends in Biotechnology 2:158-161). Inhibition of phosphinothricin causes rapid accumulation of ammonia which is toxic to the plant (Tachibana et al., 1986, J. Pest. Sci. 11:33-37). A mutant of alfalfa tissue which overproduces glutamine synthetase has been observed to be resistant to the effects of the herbicide (Donn et al., 1984, J. Mol. Appl. Genet. 2:621-635). Similarly, a gene that encodes the detoxifying enzyme phosphinothricin acetyl transferase has been cloned; when expressed in calli, under the control of the CamV 35S promoter, the calli are resistant to the herbicide.

Sulfonylurea herbicides inhibit the activity of acetolactate synthase (ALS), a nuclear-encoded chloroplast localized enzyme (Chaleff and Ray, 1984, Scilence 223:1148- 1151; Jones et al., 1985, Plant Physiol. 77:S293). Mutations of the ALS gene which have resulted in resistance to sulfonylurea herbicides have been reported (Yadav et al., 1986, Proc. Natl. Acad. Sci. U.S.A. 83:4418-4422).

Resistance to atrazine (Cheung et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85:391-395), glyphosate (Comai et al., 1985, Nature 317:741-744), and sulfonylurea herbicides (Haughn et al., 1988, Mol. Gen. Genet. 211:266-271) have been achieved by the introduction of foreign genes encoding modified insensitive target proteins. Alternatively,

resistance to phosphinotricin (De Block et al., 1987, EMBO J. 6:2513-2518) and bromoxynil (Stalker et al., 1988,

Science 242:419-423) has been achieved by the expression of detoxifying enzymes. (See Streber and Willmitzer, 1989, Bio/Technology 7 : 811-815 for review).

In various embodiments of the invention, the GS2 and/or GS3 promoters may be used to achieve herbicide resistance in the herbicide target tissues. Thus, GS3 promoters may be used to achieve glutamine synthetase overproduction or the expression of inhibitory substrate or detoxifying enzyme in the phloem, the avenue of transport for most herbicides. Similarly, G2S promoters may be used to direct the expression of glutamine synthetase, inhibitory substrate or detoxifying enzyme in photosynthetic tissue; in specific embodiments, the glutamine synthetase, inhibitory substrate or detoxifying enzyme may be targeted to the chloroplasts via a transit peptide when herbicides are toxic to a chloroplast enzyme . The use of the GS2 and GS3

promoters of the invention focuses anti-herbicide activity to the tissue compartment most affected by herbicide instead of altering the physiology of the entire plant. For

example, overproduction of glutamine synthetase throughout a plant may result in aberrancies of nitrogen metabolism, whereas overproduction in one tissue compartment would not.

Furthermore, it may be possible to inhibit herbicide action using a mutant form of glutamine synthetase or an isoenzyme not normally found in a particular compartment, thereby minimizing the interference with endogenous isozenyme activity.

6. EXAMPLE: IDENTIFICATION OF MESOPHYLL-SPECIFIC

AND PHLOEM-SPECIFIC PROMOTER ELEMENTS

a. MATERIALS AND METHODS

i. ISOLATION OF GLUTAMINE SYNTHETASE GENOMIC CLONES

Genomic clones encoding chloroplast or cytosolic GS of pea were isolated from a genomic library of Pisum sativum cv. "Sparkle" (Rogers Brothers Seed Co., Twin Falls, ID) constructed in "Lambda Dash" (Stratagene, La Jolla, CA).

Complete sequence analysis of each genomic clone revealed that the genomic clone for chloroplast GS2 (GS2^ct)

corresponds to the GS185 cDNA (Tingey et al., 1988, J. Bio.

Chem. 263:9651-9657), while the genomic clone for cytosolic

GS (GS3A^cy) corresponds to the GS341 cDNA (Tingey et al.,

1988, J. Bio. Chem. 263:9651-9657). ii. CONSTRUCTION OF PLASMIDS AND

TRANSFORMATION OF AGROBACTERIUM

A 1.5 kb EcoRI-HincII fragment of the promoter region of the pea nuclear gene for chloroplast GS2 (GS2^ct) was inserted into the polylinker of pBI101.2 (Jefferson, 1987,

Plant Mol. Bio. Rep. 5:387-405) (Clontech, Palo Alto, CA) to create the plasmid, pGS2^ct-GUS (Fig. 1A). In addition to the promoter region of GS2, pGSct-GUS contains approximately

65 nt of the 5' untranslated leader of the GS2 mRNA and encodes 53 amino acids of the chloroplast transit peptide

(Tingey et al., 1988, J. Bio. Chem. 263:9651-9657). The

GS3A^cy-GUS fusion was constructed by inserting a 1.01 kb DNA fragment encompassing nucleotide position -903 to a Bg1ll site at position + 107 of the GS3A^Cy gene into the BamHI site upstream of the GUS gene in pBI101 (Jefferson, 1987, Plant Mol. Bio. Rep. 5:387-405). The GS3A^cy-GUS fusion gene was released from the plasmid as an Xbal-EcoRI fragment which was subsequently cloned into pMON505 (Horsch and Klee, 1986, Proc. Natl. Acad. Sci. USA. 83:4428-4432) to create the plasmid, pGS3A^cy-GUS (Fig. 1B). pGS3A^cy-GUS contains 88 nt of the 5' untranslated leader of GS3A mRNA and encodes six amino acids of the cytosolic GS protein (Tingey et al., 1988, J. Bio. Chem. 263:9651-9657). iii. TRANSFORMATION AND GROWTH OF

TRANSGENIC TOBACCO PLANTS

Binary vectors containing the GS-GUS constructs were transferred into the disarmed Agrobacterium stain LBA4404 by triparental mating as previously described (Bevan, 1984, Nucleic Acids Res. 12:8711-8721). Nicotiana tabacum cv. SR1 or Nicotiana tabacum cv. Xanthi was transformed by a leaf inoculation procedure (Horsch et al., 1985, Science

227:1229-1231). Regenerated shoots were selected for growth on medium containing kanamycin (200 μg/ml). Primary

transformants were maintained in sterile culture and were also grown to maturity in soil. F1 seeds were sterilized in 10% sodium hypochlorite and germinated on MS medium

containing 3% sucrose, 100 μg/ml kanamycin, and 500 μg/ml carbenicillin. Seedlings were grown in culture for several days at 26°C in continuous white light. iv. DETERMINATION OF BETA-GLUCURMIDASE EXPRESSION

GUS enzyme assays and histochemical staining of mature plants were performed as previously described (Jefferson, 1987, Plant Mol. Bio. Rep. 5:387-405; Jefferson et al., 1987, EMBO J. 6:3901-3907). The whole mount histochemical staining of seedlings was performed as previously described (Benfey et al., 1989, EMBO J. 8:2195-2202). After

incubation with the GUS substrate, 5-bromo-4-chloro-3- indolyl-B-D-glucuronic acid (Clontech, Palo Alto, CA), cross sections of mature plant organs and whole seedlings were cleared for chlorophyll by incubation with a solution of 5% formaldehyde, 5% acetic acid, and 20% ethanol for 10 minutes followed by 2 minute incubations with, respectively, 50% and 100% ethanol. Photomicrographs were taken with a Nikon Optiphot microscope using phase contrast optics. v. PLANT GROWTH CONDITIONS FOR

LIGHT INDUCTION EXPERIMENTS

Transgenic plants containing the GS2^ct-GUS fusion gene were germinated and grown in soil in continuous white light for 4 weeks. The plants were transferred to black Lucite boxes within a dark environmental chamber for 4 days.

Several leaves of each plant were collected in the dark and immediately frozen in liquid nitrogen. The plants were subsequently transferred to continuous white light for 24 hours and leaves were collected and frozen for RNA

extraction. vi. ISOLATION OF RNA AND RIBONUCLEASE

PROTECTION ASSAY

RNA was extracted from leaves of dark-adapted and light-grown transgenic tobacco plants using guanidine thiocyanate as a protein denaturant (Chirgwin et al., 1979, Biochem. 18:5294-5304). The DNA vector used (pJE1005) contained a 1.5 kb EcoRI-HincII fragment of the nuclear gene for chloroplast GS2 (GS2^ct) in the plasmid pTZ18U (US

Biochemical, Cleveland, OH). A DNA template encompassing the 5' end of GS2ct was generated by Hindlll digestion of pJE1005. The radioactive, antisense RNA probe for the RNAse protection assay was generated in vitro using T7 RNA

polymerase (Melton, 1984, Nucleic Acid Res. 12:7035-7056).

50 ug of total RNA from transgenic tobacco plants was hybridized to an excess of the antisense RNA probe overnight in 80% formamide, 60 mM Pipes pH 6.4, 400 mM NaCl and 1 mM EDTA at 60°C. RNAse T2 digestions were performed in a volume of 390 μl containing 50 mM NaOAc pH 5.0, 100 mM NaCl, 2 mM EDTA, and 60 units/ml of RNAse T2 (Bethesda Research Labs.) (Costa et al., 1989, EMBO J. 8 : 23-29 ) . Digestion products were separated on an 8% acrylamide/7M urea gel and exposed to X-Ray film at -80°C. b. RESULTS

i. CONSTRUCTION OF GS-GUS REPORTER GENE FUSIONS

AND QUANTIFICATION OF BETA-GLUCURONIDASE

ACTIVITY IN TRANSGENIC PLANTS

Genomic clones encoding chloroplast or cytosolic GS of pea were isolated by hybridization to the corresponding cDNAs, pGS185 (Tingey et al., 1988, J. Bio. Chem. 263:9651-

9657) and pGS341 (Tingey et al., 1988, J. Bio. Chem.

263:9651-9657), respectively. The genomic clone pGS2^ct corresponds to the single nuclear gene for chloroplast GS2

(Tingey et al., 1988, J. Bio. Chem. 263:9651-9657). The genomic clone pGS3A^cy corresponds to a gene for cytosolic GS (GS341) which encodes the predominant mRNA for cytosolic GS in a number of organs examined. Promoter elements from the gene for chloroplast GS2 and cytosolic GS3A were subcloned in translational fusions to the GUS reporter gene of

pBI101.2 or pBI101, respectively, as described in Materials and Methods (Jefferson, 1987, Plant Mol. Bio. Rep. 5:387-

405) to create pGS2^ct-GUS and pGS3A^cy-GUS (Fig. 1). These chimeric genes were introduced into Nicotiana tabacum via

Agrobacterium mediated plant transformation (Bevan, 1984, Nucleic Acids Res. 12:8711-8721).

GUS enzyme activity was measured in soluble protein extracts of leaves of 19 individual transgenic plants by a fluorimetric assay (Jefferson, 1987, Plant Mol. Bio. Rep. 5:387-405; Jefferson et al., 1987, EMBO J. 6:3901-3907)

(Table I). GUS activity in leaves of primary transgenic plants containing the pGS2^ct-GUS construct averaged 46,984 pmol MU/mg protein/min, while GUS activity detected in the leaves of transgenic plants containing the pGS3A^cy-GUS chimeric construct was approximately 17-fold lower. The amount of GUS activity produced in transgenic plants containing pGS2^ct-GUS is comparable to that reported for other "strong" promoters such as that for the cauliflower mosaic virus 35S protein (Benfey et al., 1989, EMBO J.

8:2195-2202) and patatin (Rocha-Sosa et al., 1989, EMBO J.

8:23-29).

Southern blot experiments using DNA from transgenic tobacco containing pGS2^ct-GUS or pGS3A^cy-GUS revealed that each transformed plant contained 1 or 2 copies of the transgene. The variation in the amount of GUS activity between individual transgenic plants was 7-fold for the pGS2^ct-GUS plants, and 25-fold for the pGS3A^cy-GUS plants (Table I). This degree of variation in GUS expression among individual transgenic plants is similar to that reported by others (Benfey et al., 1989, EMBO J. 8:2195-2202; Rocha-Sosa et al., 1989, EMBO J. 8:23-29), and is most likely the result of differences in positional insertion in the tobacco genome, or differences in the developmental stages of each plant used in this analysis. ii. THE PROMOTER FOR CHLOROPLAST GS2

DIRECTS GUS EXPRESSION SPECIFICALLY

IN PHOTOSYNTHETIC CELL TYPES

In situ GUS assays were performed on sections of leaves, stems, and roots of mature transgenic tobacco plants

(Fig. 2) and on whole tobacco seedlings (Fig. 3). These assays revealed that pGS2^ct-GUS or pGS3A^cy-GUS confer nonoverlapping patterns of cell- and organ-specific expression on the GUS reporter gene in transgenic plants.

In situ staining of sections of the pGS2ct-GUS

transgenic plants reveals that the promoter for chloroplast

GS2 directs high-level GUS expression in leaves,

specifically in the parenchymal cells of the leaf blade

(Fig. 2A). The most intense GUS staining occurs in the palisade parenchymal cells of the leaf blade which are specialized for photosynthesis and contain a large number of chloroplasts (Fig. 2B). In a cross section of the leaf midrib pGS2^ct-GUS activity is detected only in two

photosynthetic cell layers (collencyma and chlorenchyma), while the adjacent epidermal cell layer comprised of non- photosynthetic cells, shows no GUS expression (Fig. 2C). There is no GUS expression in the central vascular bundle of the midvem m the pGS2ct-GUS plants (Fig. 2A). In cross sections of stem, GUS activity is detected in the

photosynthetic chlorencymal cells (Fig. 2D), while there is no GUS staining in the pith parenchymal, vascular,

epidermal, or trichome cells of the stem (Fig. 2D). In roots, pGS2^ct-GUS is expressed at low levels m root tips where GS in plasmids functions in ammonia assimilation from the soil (Miflin, 1974, Plant Physiol. 54:550-555). iii. THE PROMOTER FOR A CYTOSOLIC GS GENE DIRECTS

GUS EXPRESSION EXCLUSIVELY IN PHLOEM

Analysis of the pGS3^cy-GUS transgenic plants reveals that the promoter for cytosolic GS directs expression of GUS specifically within the vascular elements of leaves, stems, and roots of mature plants (Fig. 2E-2H). In leaves of pGS3^cy-GUS transgenic tobacco, histochemical staining for GUS occurs exclusively in the vasculature, in a punctate pattern indicative of phloem-specific expression (Fig. 2E and 2F). In roots, the triarc staining pattern observed for pGS3^cy-GUS is also indicative of phloem-specific expression (Fig. 2G). This punctuate pattern of GUS expression is also observed in a stem cross section where the internal phloem stains intensely (Fig. 2H). IV. EXPRESSION OF THE GS-GUS FUSIONS IN

GERMINATING TRANSGENIC TOBACCO SEEDLINGS

To examine the organ and cell-specific expression of pGS2^ct-GUS and pGS3^cy-GUS during plant development, GUS enzyme activity was detected in situ in whole amounts of germinating tobacco seedlings (Benfey et al., 1989, EMBO J.

8:2195-2202). This analysis reveals a striking contrast between the expression patterns conferred by the promoters for chloroplast GS2 and cytosolic GS3A (Fig. 3). In

transgenic tobacco seedlings containing pGS2^ct-GUS, intense GUS staining is seen throughout the cotyledons, which are photosynthetic in tobacco (Avery, 1932, Am. J. Bot. 20:309- 327) (Fig. 3A). In pea cotyledons, which are non- photosynthetic (Lovell (1977) in The Physiology of the Garden Pea, eds. Sutcliffe, J. & Pate, J. S. (Academic Press, London), pp. 265-290), there is low level expression of the mRNA for chloroplast GS2. Therefore, expression of chloroplast GS2 correlates with photosynthetic capacity rather than strict organ-type. In these same pGS2^ct-GUS seedlings, GUS activity is not detected in the hypocotyl (Fig. 3A), and is present at very low levels in the root tips (Fig. 3A).

For cytosolic GS, pGS3A^cy-GUS constructs are expressed exclusively in vasculature of developing transgenic

seedlings (Fig. 3B). This vasculature-specific staining pattern is most intense in the cotyledons and is also evident in the hypocotyl and root (Fig. 3B). The emerging leaves of pGS3A^cy-GUS transgenic seedlings do not contain detectable levels of GUS (Fig. 3B). The absence of GUS expression in these young leaves is consistent with the apparent lack of vascularization of leaves in these young seedlings (Pato et al., 1970, Protoplasma 71:313-334). As the seedlings mature, and the leaves become vascularized, GUS activity is detected in the vasculature of pGS3A^cy-GUS transgenic seedlings. Control, F1 seedlings derived from plants transformed with a "promoter-less" GUS construct (pBI101) show no detectable GUS activity in histochemical assays. (Fig. 3C).

V. THE CHLOROPLAST GS2 PROMOTER CONFERS LIGHT- REGULATED EXPRESSION ON THE GUS REPORTER GENE

Previous results have demonstrated that white light induces the accumulation of the mRNA for chloroplast GS2 in mature pea plants and in etiolated seedlings (Edwards and

Coruzzi, 1989, Plant Cell 1:241-248). To determine whether the promoter for chloroplast GS2 is responsible for the light-induced accumulation of the mRNA for chloroplast GS2, the amount of GS2ct-GUS RNA from transgenic plants grown m the light or dark was measured in a ribonuclease protection assay (Fig. 4). In two separate transgenic plants assayed, the amount of RNA corresponding to the chloroplast GS2-GUS chimeric RNA drops to undetectable levels when the mature light-grown plants are placed in the dark for 4 days (Fig.

4, lanes 1 and 3). When the dark-adapted plants are

returned to white light for 24 hours, the GS2^ct-GUS mRNA accumulates approximately 8-fold (Fig. 4, lanes 2 and 4).

In control plants transformed with pBI101 and grown in continuous white light, no cross-hybridization of the RNA probe with the endogenous tobacco GS mRNA is observed (Fig.

4, lane 5). The white-light induction of the steady-state levels of GS2^ct-GUS mRNA in transgenic plants demonstrates that cis-acting elements involved in the light regulation ofthe pea GS 2^ct gene are contained within a 1.5 kb promoter fragment. c. DISCUSSION

Historically it has been difficult to access the relative functions of chloroplast and cytosolic GS due to their similarities in physical properties, as well as their immunological cross reactivity. Here, the ability to localize gene expression at the single-cell level reveals that GS isoforms function in different cell types which have distinct nitrogen metabolic needs. These molecular studies have addressed previously unanswered questions concerning the cell-specific location of glutamine synthesis during plant development.

Here we have demonstrated that the promoters from the nuclear genes for chloroplast GS2 and cystolic GS3A of pea confer unique, cell-specific patterns of expression on a GUS reporter gene in transgenic tobacco plants. The promoter for chloroplast GS2 directs GUS gene expression

predominantly within photosynthetically active cells, the palisade and spongy parenchymal cells of the leaf blade, in collenchymal and chlorenchymal cells of the stem, and in photosynthetic tobacco cotyledons. In contrast, the

promoter for cytosolic GS3A confers vasculature-specific GUS expression in leaves, stems, and roots of the mature plant, and in the cotyledons and roots of developing seedlings. These non-overlapping patterns of GUS expression signify that the chloroplast GS2 and cytosolic GS3A isoforms perform separate functions in plant nitrogen metabolism.

The activity of the promoter for chloroplast. GS2 predominantly in photosynthetic cell types is consistent with previously reported findings that chloroplast GS2 functions in the reassimilation of photorespiratory ammonia (Wallsgrove, et al., 1987, Plant Physiol. 83:155-158;

Edwards et al., 1989, Plant Cell 1:241-248), and the

assimilation of reduced nitrite in plasmids (Miflin, B.J., 1974, Plant Physiol. 54:550-555). Previous analysis of photorespiratory mutants revealed that plants which lacked chloroplast GS2 were inviable when grown under

photorespiratory conditions even though they contained normal levels of cytosolic GS (Wallsgrove et al., 1987, Plant Physiol. 83:155-158). It has also been shown that mRNA for the chloroplast GS2 accumulates preferentially in plants grown under photorespiratory conditions whereas the levels of cytosolic GS mRNAs are unaltered (Edwards et al.,

1989, Plant Cell 1:241-248). The results presented here indicate that the genes for chloroplast GS2 (GS2ct) and cystolic GS (GS3A^cy) are expressed in distinct cell types. Therefore, in the previous analysis of photorespiratory mutants, the inability of cytosolic GS to compensate for the loss of the chloroplast GS2 activity in photosynthetic cells of mutant plants (Wallsgrove, et al., 1987, Plant Physiol.

83:155-158) may be explained by the fact that cytosolic GS and chloroplast GS2 are expressed in separate cell types, as demonstrated here. The expression of chloroplast GS2 and cytosolic GS in separate cell types may also explain why cytosolic GS gene expression is unaffected by the generation of photorespiratory ammonia (Edwards et al., 1989, Plant Cell 1:241-248). It will be interesting to determine whether expression of a GS isoform within the cytoplasm of photosynthetic cell types can functionally replace

chloroplast GS2 in the reassimilation of photorespiratory ammonia.

The unforeseen finding of this transgenic analysis was the confinement of cytosolic GS3A gene expression

exclusively to the vascular elements. This result has elucidated the role of this cytosolic GS isoform in plant development. While glutamine serves as a major compound for intercellular nitrogen transport in higher plants, and is found in high levels in both the xylem and phloem saps (35), its source of synthesis was heretofore unknown. From the transgenic data presented here, it is apparent that at least one cytosolic GS isoform is expressed exclusively in the phloem elements and most likely functions to generate glutamine for intercellular nitrogen transport. The high- level expression of the gene for cytosolic GS3A in the vasculature is particularly intense in the cotyledons of germinating seedlings where glutamine serves to transport nitrogen from seed storage reserves to the developing plant. These findings in transgenic tobacco correlate well with the abundant accumulation of mRNA corresponding to this gene for cytosolic GS in germinating pea cotyledons and in nitrogen- fixing nodules (Tingey et. al., 1987, EMBO J. 6:1-9), two contexts where large amounts of glutamine are synthesized for nitrogen transport (Lea et al., 1983, in Recent Advances in Phytochemistry: Mobilization of Reserves in Germination, eds. Nozzolillo, C., Lea, P.J. & Loewus, F.A. (Plenum Press,

NY, pp. 77-109); Pate et al., 1969, Planta 85:11-34). Since expression of pea cytosolic GS3A in tobacco cotyledons is confined to the vasculature, it will be of interest to determine whether induced expression of this cytosolic GS isoform in pea nodules correlates with the vascularization of this organ. Recently, it has been shown that promoters for two cytosolic GS genes of Phasocolus vulgaris can direct expression in transgenic Lotus corniculatus nodules, and that one of these promoters is active in vascular and cortical cells of the nodule (Forde et. al., 1989, Plant Cell 1:391-401).

The quantification of GUS activity detected in whole leaf extracts of plants transformed with either pGS2^cl-GUS of pGS3A^cy-GUS revealed that plants containing the

chloroplast GS2cl-GUS transgene contained, on average, 17 times more GUS activity than plants containing the GS3A^cy- GUS construct. However, because the expression of each of the GS-GUS constructs is confined to distinct leaf cell types which comprise different fractions of the total leaf cell population, the relative amount of GUS activity in whole leaf extracts cannot be regarded as a measure of absolute promoter strength.

The light-induced accumulation of the transgenic

GS2ct-GUS mRNA reveals that the promoter for GS2^cl contains a cis-acting DNA element involved in light regulation.

Since previous experiments have demonstrated that

phytochrome is partially responsible for the white-light induction of the mRNA for chloroplast GS2 (Tingey et al.,

1988, J. Bio. Chem. 263:9651-9657; Edwards et al., 1989, Plant Cell 1:241-248), studies of the promoter for GS2^ct should contribute to the understanding of phytochrome- mediated gene expression. It is noteworthy that plasmid GS2 is also present in non-photosynthetic cell types such as etiolated leaves (Hirel, 1982, Planta 155:17-23) and roots

(Miflin, 1974, Plant Physiol. 54:550-555). This is

corroborated by the low-level of GUS expression in roots of pGS2^ct-GUS transgenic plants. Therefore, the regulation of expression of the GS2^ct gene is likely to differ from that of other light-regulated genes which function solely in photosynthesis (e.g. ribulose 1,5 bis-phosphate carboxylase, chlorophyll a/b binding protein).

The unique expression patterns conferred upon the GUS reporter gene by the promoters for chloroplast GS2 and cytosolic GS3A and the light-regulated induction of pGS2^ct-

GUS RNA levels are most likely due to the transcriptional regulation of these transgenes. However, because the GS-GUS fusions contain the 5' noncoding leader of the GS mRNAs and a small portion of the GS coding regions, it is possible that post-transcriptional events (e.g. RNA stability, translational regulation, and subcellular

compartmenalization) also contribute to the observed

differences in transgene expression. Future experiments directed at characterizing the specific cis-acting

regulatory regions of the GS genes will distinguish between these possibilities.

In addition to elucidating the individual roles of the GS isoforms in plant nitrogen metabolism, the transgenic studies presented here describe plant promoters which may be used to direct cell-specific expression of foreign genes in plants. In particular, a promoter which confers specific expression of foreign genes in phloem cells has potential application in generating resistance to viral pathogens transmitted within the phloem (Schneider, 1965, 11:163-221). Finally, since glutamine synthetase is the target of several herbicides (Kishore, et al., 1988, 57:627-663), the

expression studies presented here indicate that it may be necessary to express herbicide resistant forms of GS in both photosynthetic and vascular cell types in order to confer resistance to GS inhibitors. Nucleotide analysis of the GS genes in P. sativum has shown that chloroplast and cytosolic GS are derived from duplications of a single ancestral gene followed by

specialization of each locus for distinct expression (Tingey et al., 1988, J. Bio. Chem. 263:9651-9657; Coruzzi et al., 1989, in The Molecular Basis of Plant Development, Alan R. Liss, Inc., pp. 223-232). Future studies directed at the identification of the necessary cis-acting promoter regions of the nuclear genes for chloroplast and cytosolic GS should uncover DNA elements which have evolved to confer distinct spacial and temporal patterns of expression to these genes.

7. DEPOSIT OF MICROORGANISMS

The following microorganisms have been deposited with the Agricultural Research Culture Collection, Northern

Regional Research Center (NRRL) and have been assigned the following accessin numbers:

Microorganism Plasmid Accession No.

Escherichia coli XL1 pGS2ct-1/1583 NRRL-18575

Escherichia coli XL1 pGS3Acy-1/1941 MRL-18576

Escherichia coli XL1 pGS3Acy-931/1941 NRRL-18577

Excherichia coli XL1 pGS3Bcy-1/1248 NRRL-18578

The present invention is not to be limited in scope by the specific embodiments described herein,

plndeed, various modifications of the invention in addition to those described herein will become apparent to those skilled in the art from the foregoing description and accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

Various publications are cited herein, the disclosures of which are incorporated by reference in their entireties.

Claims

WHAT IS CLAIMED IS:

1. A promoter element comprising a nucleotide sequence substantially as depicted in Figure 5A, or a functional portion thereof.

2. A promoter element which is homologous to the promoter element of claim 1.

3. The promoter element of claim 1 which is expressed selectively in photosynthetically active cells.

4. The promoter element of claim 2 which is expressed selectively in photosynthetically active cells.

5. The promoter element of claim 1 which is inducible by light.

6. The promoter element of claim 2 which is inducible by light.

7. A promoter element comprising a nucleotide

sequence substantially as depicted in Figure 5B, or a functional portion thereof.

8. A promoter element which is homologous to the promoter element of claim 2.

9. A promoter element comprising a nucleotide

sequence substantially as depicted in Figure 5C, or a functional portion thereof.

10. A promoter element which is homologous to the promoter element of claim 9.

11. The promoter element of claim 7 which is

expressed selectively in plant vascular elements.

12. The promoter element of claim 11 which is

expressed selectively in phloem.

13. The promoter element of claim 8 which is

expressed selectively in plant vascular elements.

14. The promoter element of claim 13 which is

expressed selectively in phloem.

15. The promoter element of claim 9 which is

expressed selectively in plant vascular elements.

16. The promoter element of claim 15 which is

expressed selectively in phloem.

17. The promoter element of claim 10 which is

expressed selectively in plant vascular elements.

18. The promoter element of claim 17 which is

expressed selectively in phloem.

19. A gene fusion comprising:

(i) a promoter element comprising a nucleotide sequence substantially as depicted in Figure 5A, or a functional portion thereof; and (ii) a second nucleotide sequence which encodes a desired gene product.

20. A gene fusion comprising:

(i) a promoter element which is homologous to the nucleotide sequence depicted in Figure 5A, or a portion thereof; and (ii) a second nucleotide sequence which encodes a desired gene product.

21. The gene fusion of claim 19 which is expressed selectively in photosynthetic cells.

22. The gene fusion of claim 20 which is expressed selectively in photosynthetic cells.

23. The gene fusion of claim 19, the expression of which is inducible by light.

24. The gene fusion of claim 20, the expression of which is inducible by light.

25. A gene fusion comprising

(i) a promoter element comrpising a nucleotide sequence substantially as depicted in Figure 5B, or a functional portion thereof; and (ii) a second nucleotide sequence which encodes a desired gene product.

26. A gene fusion comprising

(i) a promoter element which is homologous to the nucleotide sequence depicted in Figure 5B, or a portion thereof; and

(ii) a second nucleotide sequence which encodes a desired gene product.

27. A gene fusion comprising:

(i) a promoter element comprising a nucleotide sequence substantially as depicted in Figure 5C, or a functional portion thereof; and

(ii) a second nucleotide sequence which encodes a desired gene product.

28. A gene fusion comprising:

(i) a promoter element which is homologous to the nucleotide sequence depicted in Figure 5C, or a portion thereof; and

(ii) a second nucleotide sequence which encodes a desired gene product.

29. The gene fusion of claim 25 which is expressed selectively in plant vascular elements.

30. The gene fusion of claim 29 which is expressed selectively in phloem.

31. The gene fusion of claim 26 which is expressed selectively in plant vascular elements.

32. The gene fusion of claim 31 which is expressed selectively in phloem.

33. The gene fusion of claim 27 which is expressed selectively in plant vascular elements.

34. The gene fusion of claim 33 which is expressed selectively in phloem.

35. The gene fusion of claim 28 which is expressed selectively in plant vascular elements.

36. The gene fusion of claim 35 which is expressed selectively in phloem.

37. A transgenic plant which comprises a transgene which includes a promoter element comprising a nucleotide sequence substantially as depicted in Figure 5A, or a functional portion thereof.

38. A transgenic plant which comprises a transgene which includes a promoter element which is homologous to the nucleotide sequence depicted in Figure 5A, or a portion thereof.

39. A transgenic plant which comprises a transgene which includes a promoter element comprising a nucleotide sequence substantially as depicted in Figure 5B, or a functional portion thereof.

40. A transgenic plant which comprises a transgene which includes a promoter element which is homologous to the nucleotide sequence depicted in Figure 5B, or a portion thereof.

41. A transgenic plant which comprises a transgene which includes promoter element comprising a nucleotide sequence substantially as depicted in Figure 5C, or a functional portion thereof.

42. A transgenic plant which comprises a transgene which includes promoter element which is homologous to the nucleotide sequence depicted in Figure 5C, or a portion thereof.

43. The transgenic plant of claim 37 in which the transgene renders the plant resistant to viral infection.

44. The transgenic plant of claim 43 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

45. The transgenic plant of claim 43 in which the transgene comprises a nucleotide sequence which encodes an antisense RNA that corresponds to the mRNA which encodes viral coat protein.

46. The transgenic plant of claim 38 in which the transgene renders the plant resistant to viral infection.

47. The transgenic plant of claim 46 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

48. The transgenic plant of claim 39 in which the transgene renders the plant resistant to viral infection.

49. The transgenic plant of claim 48 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

50. The transgenic plant of claim 40 in which the transgene renders the plant resistant to viral infection.

51. The transgenic plant of claim 50 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

52. The transgenic plant of claim 41 in which the transgene renders the plant resistant to viral infection.

53. The transgenic plant of claim 52 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

54. The transgenic plant of claim 42 in which the transgene renders the plant resistant to viral infection.

55. The transgenic plant of claim 54 in which the transgene comprises a nucleotide sequence which encodes a viral coat protein.

56. The transgenic plant of claim 37 in which the transgene alters the plant's nutritional requirements.

57. The transgenic plant of claim 38 in which the transgene alters the plant's nutritional requirements.

58. The transgenic plant of claim 39 in which the transgene alters the plant's nutritional requirements.

59. The transgenic plant of claim 40 in which the transgene alters the plant's nutritional requirements.

60. The transgenic plant of claim 41 in which the transgene alters the plant's nutritional requirements.

61. The transgenic plant of claim 42 in which the transgene alters the plant's nutritional requirements.

62. The transgenic plant of claim 37 in which the transgene renders the plant resistant to herbicides.

63. The transgenic plant of claim 38 in which the transgene renders the plant resistant to herbicides.

64. The transgenic plant of claim 39 in which the transgene renders the plant resistant to herbicides.

65. The transgenic plant of claim 40 in which the transgene renders the plant resistant to herbicides.

66. The transgenic plant of claim 41 in which the transgene renders the plant resistant to herbicides.

67. The transgenic plant of claim 42 in which the transgene renders the plant resistant to herbicides.