EP0699239A1 - Genetic stabilizing elements - Google Patents

Abstract

A stabilized gene for transforming a host plant cell, said gene comprising a gene exogenous to the plant cell, and at least one stabilizing DNA segment in a 3'- or 5'-flanking region of said gene. The gene may comprise a stabilizing DNA segment in a 3'-flanking region and a second stabilizing DNA segment in a 5'-flanking region of said gene. The invention further includes vectors and transgenic plants comprising the said stabilized gene.

Description

GENETIC STABILIZING ELEMENTS

The invention relates to the field of plant molecular biology, in particular to the technolog transferring an exogenous gene into a plant cell and conferring a stable phenotype associated said gene on a transgenic plant.

"Transgene" or "exogenous gene" are terms used in the art to denote a gene which has b transferred to a host cell or host plant from a source other than the host cell or host pl However, a transgene may herein refer to a gene normally locatable in an host plant or plant c to which is added one or more flanking regions comprising stabilizing DNA segments. Thu gene endogenous to a host plant cell or host plant may be modified with flanking regi comprising stabilizing DNA segments whereby the endogenous gene becomes an exogenous g or transgene with the meaning of the instant application. As used herein, the terms "transge and "exogenous gene" have the same meaning. Transfer of an exogenous gene to a host pl cell can be accomplished by a variety of means known in the art. Most classes of plants h been transformed and regenerated to yield adult plants expressing a phenotype associated with transgene.

The process of producing a transgenic plant typically includes exposing plant cells, which be in the form of individual cells, protoplasts or excised tissue, to DNA comprising the exogen gene, in order to introduce the exogenous gene into the host cells. Only a fraction of the c exposed to the DNA are actually transformed. The recipient cells are then cultured in vitro order to proliferate the transformants and to identify and select for those which express transgene. Frequently a selectable marker is introduced, together with the exogenous DNA, that transformants can be selected by their ability to grow under conditions that inhibit gro of nontransformed cells, or which favor growth of transformed cells. However, it is also possi in some cases to identify directly the transformed cells or callus containing them. Further st include techniques of regeneration to produce differentiated shoots, roots or embryos from whi ultimately, whole plants can be obtained. Such steps are known to the skilled man.

Primary transformants are those cells or proliferated tissue (e.g., callus colonies) which initially observable, directly or indirectly, as possessing the exogenous gene after transformation step. Most commonly, possession of the exogenous gene is observed indire as expression of a co-transformed selectable marker. In some instances the phenotype associat with expression of the exogenous gene will be observable in the primary transformant. When selectable marker is present, such as antibiotic resistance, culture in the presence of the selecti agent, the antibiotic, ensures that only cells expressing the resistance phenotype will grow. In t absence of selection, however, it has frequently been observed that descendants of prima transformants lose the phenotype associated with the transgene. For example, explants of primary transformant callus often fail to display the phenotype of the exogenous gene, in t absence of continued selection pressure. Furthermore, when whole plants are regenerated fro transformed tissue or callus, some of the regenerated plants fail to have the phenotype of t exogenous gene, and the same phenomenon is sometimes observed in the progeny of selfe transformed plants. In such cases it has not been established whether the loss of phenotype is du to loss of the exogenous gene itself, or to loss of ability to express the exogenous gene. The lo of phenotype, whatever the mechanism, results in a gradual decline in overall transformatio efficiency, i.e., the total number of transformants declines over time with respect to the numb of initial transformants. The present invention provides a means for stabilizing transforman against loss of phenotype associated with the exogenous gene, so that higher overa transformation efficiency is obtainable.

Recent studies of the structure of the eucaryotic cell nucleus and the organization of chromati within the nucleus have led to new techniques for identifying DNA components and nucle structural components that participate in organizing cellular DNA in the nucleus. Such studie have demonstrated the existence of a complex nuclear matrix which includes structur components remaining after DNAse I digestion and extraction with 2M NaCI (See Gasser, S. (1988) Architecture of Eukaryotic Genes: Symposium on Chromatin Structure of Plant Gene Frankfurt am Main, W. Germany, September 1986, XIV+ 518, P VCH Publishers; New York, p 461-471). The finding that Li-3,5-diiodosalicylate (LIS) extraction removes histone fro chromosomal DNA has made it possible to isolate nuclear scaffold by combining LIS extractio with endonuclease digestion. Such procedures leave residual DNA segments bound to the nuclea scaffold. Such DNA segments have been termed scaffold attachment regions (SAR) and matri associated regions (MAR). Such DNA segments are considered to be functionally similar i nature, independent of how they are obtained. The term MAR is used herein to refer to DN segments isolated from nuclear scaffold or nuclear matrix preparations after endonucleas treatment. MARs typically bind reversibly to nuclear matrix or scaffold preparations. Binding is saturab indicating binding to a limited number of specific sites. MARs can be of any size, however, th are generally of about 1 kb or less in size and are generally AT rich. They do not necessari share extensive sequence homology, although certain sequence motifs have been observed in so MARs. Many MARs possess a topoisomerase II cleavage site consensus sequence.

MARs are believed to function in vivo as structural attachment points linking chromosomal D to structural elements of the nucleus. Models of chromosome structure have been proposed, which chromosomal regions between two adjacent MARs form a loop of DNA between t anchor points of the MARs. It has been proposed that MAR attachment facilitates transcripti of nearby genes, by locating those genes close to nuclear pores or channels where polymeras transcription factors, substrates, etc., may concentrate. At the same time, anchorage to the nucle matrix serves to separate or isolate groups of genes on separate chromatin loops, acting boundary elements to limit the influence of nearby transcription units on one another, common called position effects.

The MARs appear to function across species boundaries, although matrix binding specificity m be diminished, for example when using animal MARs in plants. Functional association betwe MARs and DNA replication has also been implicated in studies showing that some matrix-bindi sequences of maize DNA may function as ARS (autonomous replicating sequence) elements yeast.

Phi-Van, L. et al. (1990) Mol. Cell Biol. 10:2302-2307, compared the effect on reporter ge expression of the presence or absence of MARs flanking an exogenous reporter gene. The MA sequence was isolated from the chicken lysozyme gene 5' flanking region, the host cells we fibroblasts. Both enhancement of reporter gene activity and reduction of position effec (individual variation of expression level among independent transfectants) were observed if t reporter gene construct included MARs flanking the gene. A review of boundary functio attributable to MARs was published by Eissenberg, J.C. and Elgin, S.C.R. (1991) Trends Genetics 2:335-340. The authors suggested that MARs function both as insulators wh bracketing a gene together with its enhancers, to maintain activity of the enhancers by isolati them from chromosomal position effects, and as barriers when interposed between a gene and enhancer. In higher plants, the existence of MARs has been reported by Hall, G et al. (1991) Proc. Na Acad. Sci. USA 88:9320-9324. Tobacco MARs (termed SARS therein) were isolated fro flanking regions of three root-specific genes. An "endogenous" assay for MARs was disclose based on their ability to bind to nuclear scaffold preparations. An "exogenous" assay, based ability of isolated scaffolds to bind DNA fragments containing MARs, was also disclosed. scaffold-associated DNA region located downstream of the pea plastocyanin gene was isolat and characterized by Slatter, R.E. et al. (1991) Plant Cell 1:1239-1250. The SAR was linked a downstream repeated sequence and had a sequence rich in A and T sequences, sever topoisomerase II binding sites and several ARS sequences.

Breyne, P. et al. (1992) Plant Cell 4:463-471 used a tobacco-derived SAR to analyze the effe of flanking a reporter gene in transgenic plants, similar to the experiment described by Phi- V et al. (1990). Qualitatively similar results were obtained showing an effect of a tobacco SA flanking a transgene on variance of expression among independent tobacco transformant However, no increase of average expression level was observed. The effect of reducing varian was not observed for constructs containing a mammalian β-globin SAR instead of the tobac SAR. All effects heretofore observed have related to phenomena occurring within a sing generation.

The basis of the present invention is that there exist certain stabilizing DNA segments whic when introduced together with an exogenous gene into plant host cells, serve to stabilize t exogenous gene from one generation of cells to the other, or from one generation of transgen plants to another. Stabilization has occurred when the phenotype associated with the exogeno gene (the "transgene phenotype") is retained from one generation to the next at a higher frequen than is observed in the absence of the stabilizing DNA. Since the rate of loss of the transge phenotype is most apparent by comparison with primary transformants, stabilization is al defined as a reduction of the frequency of loss of the transgene phenotype over one or mo generations when compared with the number of primary transformants.

Stabilizing DNA segments include, but are not limited to, MAR and SAR segments as identifi by art-known measures of binding with nuclear matrix or scaffold. Stabilizing DNA segmen also include certain repeated sequences, locus control regions (LCR), also known as loc activating regions (LAR) and certain other sequences such as DNAse hypersensitive regions (HR In the absence of binding assays, suitable candidate stabilizing DNA segments can be identifi by various structural attributes such as possession of one or more topoisomerase II binding sit hypersensitivity to DNAse, having high A & T content and ARS consensus sequences. stabilizing DNA segment can have one or more of these structural features, however, the absen of one or more such features would not rule out the segment as having the stabilizing functi

A functional test is described herein which exploits the property of a stabilizing DNA segm to separate adjacent genes, on a vector into independently expressed units. A stabilizing D segment interposed between two genes or transcription units oriented in tandem on a vect effectively prevents the depressed expression of the downstream gene which occurs in the absen of a stabilizing DNA segment between the genes, as disclosed herein.

Also disclosed herein are data showing that loss of transgene phenotype over a number generations is not the result of transgene loss, but rather the result of ability to express t transgene. Experiments are reported herein showing stabilization of expression over many c generations by providing the exogenous gene with flanking stabilizing DNA segments.

Accordingly the invention provides vectors for achieving stable transgene expression over ma cell generations by providing stabilizing DNA segments in one or both the 3'- and 5'-flanki regions of the transgene. The invention also includes vectors for achieving independe expression of at least two exogenous genes on the same vector by providing a stabilizing D segment interposed between the exogenous genes.

Also included as part of the invention is a method for transforming plant cells with an exogeno gene to enhance the stability of the phenotype associated with the exogenous gene, by introduci into the plant cell a vector having at least one stabilizing DNA segment present in a flanki region of an exogenous gene. In a preferred embodiment the vector will possess more than o stabilizing DNA segment such that the exogenous gene(s) lies between the stabilizing DN elements.

The invention will be further apparent from the following description taken in conjunction wi the associated Examples, Figures and Sequence Identifications. Figure 1 is a diagram of the plasmids pZO1071 and pZO1051 described in Example 2

Figure 2 is an autoradiograph of DNA after electrophoresis in a nuclear matrix bindi assay, as described in Example 9.

Figure 3 is a graph showing saturation of MAR binding to nuclear matrix as the amou of MAR DNA is increased. See Example 9.

Figure 4 is a graph showing competition of tomato MAR binding to tomato nucle matrices as amounts of competitor MAR DNA is added. Each test represents 100 ng labell tomato MAR per assay. pBR322 vector fragments

Drosophila histone MAR

Rat GDH 5' MAR

Homologous tomato MAR

Figure 5 is an autoradiograph of an electrophoretic gel showing MAR binding to nucle matrix and nuclear shell preparations in presence and absence of R. coli DNA. See Example

Figure 6 is a diagram of a T-DNA vector, pZU043A carrying the TSWV nucleoprotei gene controlled by a CaMN 35S promoter. See Example 15.

Figure 7 is an autoradiograph of a Southern blot showing the presence of TSWV DN in various transformant lines, as described in Example 15.

Figure 8 is a diagram of T-DNA vectors carrying various MAR segments, as describe in Example 17.

SEQ ID No. 1 shows the nucleotide sequence of MRS5 given in Table IA SEQ ID No. 2 shows the nucleotide sequence of MRS4 given in Table IB SEQ ID No. 3 shows the partial nucleotide sequence of MRS 3 given in Table 1C SEQ ID No. 4 shows the nucleotide sequence of MAR1 given in Table 5 A SEQ ID No. 5 shows the nucleotide sequence of MAR2 given in Table 5B SEQ ID No. 6 shows the nucleotide sequence of the SARLa region from a soybean sma heatshock gene (MAR3) given in Table 5C

SEQ ID No. 7 shows the nucleotide sequence of the primer 1929ECOR SEQ ID No. 8 shows the nucleotide sequence of the primer 1929ECOU

"Stabilization" is the term used herein to denote increased retention of a phenotype associate with an exogenous gene, over one or more plant cell and/or plant generations, when comparin transformants having a "stabilized exogenous gene" of the invention with those having a cont exogenous gene. The comparison is made between the number of transformants having phenotype associated with the exogenous gene ("transgene phenotype") at a given time, and number of transformants having the transgene phenotype at a later time, after correcting for to cell proliferation. Therefore, it is not the quantitative expression level which is to be measu (although that may incidentally be affected), but the rate of retention of the transgene phenoty itself in individual cell lines and/or plants descendant from primary transformants.

A stabilized exogenous gene is an exogenous gene which includes one or preferably m stabilizing DNA elements in its 3'- and 5' flanking regions. As it may be the case that more th one exogenous gene is to be introduced to the host plant, it will be recognized that stabilizi DNA elements can be provided such that all genes are flanked, and stabilized by a single pair stabilizing DNA segments, or they may be individually flanked, or one gene can be flanked wh another is left unflanked. In the latter case, the unflanked gene may be less stable than t flanked gene, which may be desired if, for example, the unflanked gene were only useful a marker for initial transformant selection, but of no value to the whole plant.

Stabilizing DNA segments include MAR, SAR, LCR or LAR, HR and the like as describ supra, repetitive elements as described infra, and other DNA segments sharing structural a functional features therewith. Some stabilizing DNA segments also exert a shielding effect wh interposed between two tandemly oriented genes on a single vector, such that the downstre gene is less affected by expression of the upstream gene than would be observed if no stabilizi DNA segment were present. A demonstration of the shielding effect of MRS elements provided herein, providing a means for recognizing a stabilizing DNA segment. Other means recognizing a stabilizing DNA segment include the various matrix binding and scaffold bindi assays known in the art. Stabilizing DNA elements can be obtained from any eukaryotic prokaryotic cell type. Preferred sources are eukaryotic cell types from animal or plant sourc and most preferred are stabilizing DNA segments compatible with the host cell, such D segments may be endogenous to the host cells. Stabilizing DNA segments display a range effectiveness and specificity. Any detectable level of stabilization is operative, thereby reduci the need to screen and evaluate large numbers of transformants to find satisfactory performe

Any gene found in or not normally found in the host plant, nor wholly obtained from the h plant, may serve as an exogenous gene. A gene of the host plant modified sufficiently to provi a distinct phenotype qualifies as an exogenous gene. For example, a gene of the host pla provided with a different promoter so that the timing, tissue specificity, expression leve inducibility or other aspect of gene control was altered such that the plant's transgene phenotyp was distinct from the wild type, would be an exogenous gene under the definition. The gene a parasite or pathogen of the host plant or from other sources such as a nonpathogen to the ho plant is also an exogenous gene.

The phenotype associated with the exogenous gene, also termed the transgene phenotype herei is any trait or characteristic conferred on the transgenic plant or host plant cells, by the exogenou gene. A phenotype can range from a measurable amount of the protein or RNA encoded by th exogenous gene, to a physical or agronomic trait. In most cases, more than one phenotype ca be detected for a given exogenous gene. For example, where the exogenous gene encodes a insecticidal protein such as the Bacillus thuringiensis toxin, the phenotypes include presence o the protein in plant tissues and resistance to certain insects. Where the exogenous gene encode an antisense RNA to a plant virus, the phenotypes include presence of the RNA in plant tissue and resistance to certain viruses. In some instances, one or more exogenous genes may b introduced in a single stabilized gene cassette, in order to provide an easily measured phenotyp linked to a difficultly measurable phenotype. An example is kanamycin resistance linked to gene for fungal resistance. Therefore, a phenotype associated with an exogenous gene include a phenotype associated with a linked exogenous gene.

Examples of exogenous genes and their associated phenotypes include, but are not limited to: a) antisense RNA to confer virus resistance or to modify expression of an endogenou gene of the host plant; b) viral coat protein and/or RNA, or other viral or plant genes to confer virus resistance c) fungus resistance, possibly conferred by a wound induced gene; d) insect resistance conferred by an insecticidal toxin or other protein; e) flower color or flower pattern conferred by genes affecting pigment production; f) yield improvement; g) drought resistance; h) self-incompatibility; i) male sterility j) delayed or accelerated maturation; k) protein production, for example, conferred by mammalian genes encodin therapeutically useful protein;

1) improved nutritional balance conferred by a seed storage protein gene modifie affect amino acid balance; m) herbicide resistance, conferred by various herbicide resistance mechanisms; n) nitrate tolerance; o) plant morphology, for example dwarf variety genes that minimize plant resou devoted to vegetative growth; p) metabolic alterations that increase or modify production of useful plant products s as sugars, starches, complex carbohydrates, oils, alkaloids, gums and the like.

Other sorts of exogenous genes useful for transfer to a plant will be recognized by those ski in the art as applicable in the present invention.

Construction of transformation vectors incorporating a stabilized exogenous gene is rea accomplished by standard techniques of DNA manipulation. A stabilizing DNA segment can inserted in the 5'-flanking region upstream of promoter sequences, or in the 3'-flanking reg which may or may not lie downstream of polyadenylation signal sequences, if present. The e distance of the stabilizing DNA segment from either end of the exogenous gene is not criti However, if the construct includes T-DNA borders of Agrobacterium tumefaciens, the stabiliz DNA segments should be inserted so as to be between the T-DNA borders, to ensure integrat of the stabilizing DNA segments. The effect of one or more stabilizing DNA elements is stabilize those genes lying in close association or proximity therewith. Therefore the prefer construction is to place the stabilizing DNA elements so that they flank only the genes wh expression is to be stabilized. Although stabilizing DNA segments do not necessarily h restriction sites at or near their ends, it is a matter of ordinary skill to modify the ends using, e ligation of oligonucleotide linkers, or primers for polymerase chain reaction incorporatin restriction site sequence, to facilitate inserting the stabilizing DNA segments at desired sites a vector. Orientation of a stabilizing DNA segment with respect to the orientation of the g to be stabilized is not a critical factor as regards the stabilization function. The stabilizing D elements flanking a given gene need not be identical, nor must they be obtained from the s source organism. The stabilizing DNA elements need not be exogenous, but can instead obtained from the host organism. Although there is superficial similarity between animal M and SAR sequences and their plant counterparts, plants are the preferred source for stabilizi DNA element, the most preferred being plants of the host plant species or closely related speci

Stabilizing DNA segments can be isolated by a variety of methods. First, it is possible to analy the 3' or 5' flanking regions of any stable gene, preferably, but not limited to, those flanki regions lying within about 500 kbp either side of the coding region. Candidates are identifia by such criteria as DNAse .hypersensitivity, topoisomerase II binding sites, ability to bind mat or scaffold preparations, and the like. An alternative method is simply to clone fragments o stable gene, then select the fragments that display such properties as DNAse hypersensitivi locus control regions, matrix binding, scaffold binding and the like. Yet another alternative is isolate plant nuclei, extract the nuclei with LIS, then treat with an endonuclease including, f example, a restriction endonuclease, extract DNA remaining bound to matrix or scaffold phenol extraction, and clone the resulting MARs. Another procedure is to isolate plant nucle matrices or scaffolds, then screen for DNA fragments preferably limited to a size approximately 0.1 - 1.0 kb capable of binding to the matrices or scaffolds. Those able to bi tightly are then cloned. As a variation of the foregoing method, an isolated locus control regi (LCR) from, e.g., chicken, is used to isolate LCR binding protein. The gene encoding an L protein is then cloned and expressed in a suitable system to produce sufficient amounts to usable to identify plant DNA segments capable of binding to the protein. In every instance whe candidate stabilizing DNA segments are cloned, their ability to stabilize a transgene through ma cell generations is testable using a marker transgene in a suitable transgenic host plant. The te host plant preferably displays a low frequency of stable expression in the absence of a stabilizi DNA segment. For example, lettuce displays a frequency of stable expression about 15% primary transformants, when not transformed with a stabilizing DNA segment.

Transformation can be carried out by any means known in the art. These include, but are n limited to, direct transfer of DNA into whole cells, tissues or protoplasts, optionally assisted chemical or physical agents to increase cell permeability to DNA, e.g., treatment wi polyethylene glycol, dextran sulfate, electroporation and ballistic implantation of DNA-coat particles. Transformation is also mediated by Agrobacterium strains, notably A. tumefaciens a A. rhizogenes, and also by various genetically engineered transformation plasmids which inclu portions of the T-DNA of the tumor-inducing plasmids of Agrobacteria. The T-DNA borders c be incorporated into other transformation constructs, to facilitate integration of a stabili exogenous gene lying between the T-DNA border elements. Other means for effecting entry DNA into cells include viral vectors and agroinfection.

DNA constructs suitable for transformation include at a minimum the exogenous gene (prom and coding sequence) to be transferred, flanked by at least one stabilizing DNA segment, but involve other elements as well. The stabilized exogenous gene can be inserted into a vec flanked by T-DNA borders, combined with a marker gene, all according to techniques know the art. The choice of construct will be influenced by the method of transformation adopted. example, if a ballistic transformation is desired, use of a vector may be superfluous, w flanking T-DNA borders (in addition to the stabilizing DNA segments) may be desired as a me of promoting genomic integration of the transgene.

Examples of vectors suitable for plant transformation include, but are not limited to:pCGN15 pART27, ρOCA18, pCVOOl, pCV002, MON200, pGV3850, pGV260, pGPTV vectors, and Mi Ti plasmids. References describing the foregoing and other vectors suitable for use in invention are also included: (Mini-Ti) Framond de, A . (May 1983) Bio/Technology, pp. 2 269; (pCV001/pCV002) Koncz, C. and Schell, J. (1986) Mol. Gen. genet. 204:383-396; Klee, et al. (1985) Bio/Technology 1:637-642; (pCGN1547) McBride, K.E. and Summerfelt, K (1990) Plant Mol. Biol. 14:269-276; Hoekema, A. et al. (1985) Plant Mol. Biol. 1:85- (pGV3850) Za bryski, P. et al. (1983) EMBO J. 2:2143-2150; (pOCA18) Olszewski, N.E. et (1988) Nucl. Acids Res. 16(22): 10765-10782; (pART 27) Gleave, A.P. (1992) Plant Mol. B 20:1203-1207; (pGV260) Deblaere, R. et al. (1985) Nucl. Acids Res. j :4777; (pGPTV vect Becker, D. et al. (1992) Plant Mol. Biol. 20:1195-1197; (pGA vectors) An, G. et al. (19 EMBO J. 4(2):277-284; (Binary vectors in general and cointegrate vectors in general in Chapt A2 and A3) Plant Mol. Biol. Manual, Gelvin and Schilperoot, Eds., Kluwer Academic Publish (1988).

Virtually all plants of agronomic or horticultural value are known to be both transformable regenerable. The techniques vary in individual detail from species to species, as is underst by those skilled in the art. The nature of applicable transformation methods to be used fo given plant species may be affected by the type of regeneration protocol that can be used in given instance. For example, where regeneration cannot be obtained from protoplasts, the met of transformation must be suitable for whole cells or tissues. Where the plant species is diffic to transform using Agrobacterium, the alternative of ballistic transformation may be preferre All such considerations are matters well-known to those of skill in the art.

Examples of plants suitable for use in the invention include, but are not limited to, those pla that are members of Solanaceae, Apocynaceae, Chenodiaceae, Polygonaceae, Boraginace Compositae, Rubiaceae, Scrophulariaceae, Caprifoliaceae, Leguminosae, Araccae, Morace Euphorbiaceae, Brassicaceae, Primulaceae, Violaceae, Protulaceae and Rosaceae. In additio plants of the following families are suitable for use in the invention: polypodiaceae, umbellife liliaccae, crucifereae, gramineae, geranaceae, ranunculceae, begoniaceae, labiatae, caryophyllace balsaminaceae, papilionaceae, gesneriaceae, violaceae, araliaceae, as well as plants common known as: fern, carrots, leek, asplenium, radishes, celery, onions, fennel, wheat, rye, barley, co (maize), soybean, oats, rices, geraniums, violets, windflowers, ornamental asparagus, begoni flame nettle, lark spur, carnations, gilliflowers, Busy Lizzy, lupin, crowflower, sage, bell flowe soap herbs and Panax ginseng. Specific mention is also made of the following plants: tomat melon, watermelon, pepper, lettuce, beans, brassica including rapeseed plants, cabbages, broccol cauliflowers, sunflowers, sugar beet, violas, begonia, pelargonium peltatum, pelargoniu hortorum, com (maize), sweet corn, Cyclamen and Impatiens.

Example 1: Isolation of maize repetitive sequences (MRS)

A library of Zea mays L. (A3780) genomic DNA is constructed in pTZ19R (Pharmacia) restricting genomic DNA with EcoRI plus Hindlll, separating the resulting fragments on 0.6 low-melting point agarose, excising the region containing fragments from 1 to 3 kb in siz diluting, melting and ligating to EcoRI plus Hindlll cut pTZ19R, and transforming in I c C600 cells. Unless otherwise specified, methods for cloning and preparing plasmid DNA a essentially as described (Maniatis et al. (1982) Molecular Cloning, Cold Spring Harb Laboratory). In general, all cloning is done with DNA fragments excised from low-melt agaros Individual colonies from this library are grown in individual wells of microtiter dishes. To scre this library for repetitive sequences, cells from each well are transferred to nitrocellulose filte using a dot-blot apparatus, lysed thereon using dilute NaOH, washed, then the filters are dried a baked. The filters are hybridized to maize genomic DNA labelled by nick translation. On colonies corresponding to cloned repetitive DNA give a strong signal under these condition About twenty clones are further analyzed by DNA preparation, restriction site mapping, a southern blotting. Six are selected for further testing and their inserts are named MRS1, MR MRS3, MRS4, MRS5, and MRS6. The maize inserts are cut at their Hindlll sites, made bl by treatment with T4 DNA polymerase, then EcoRI linkers are ligated using T4 DNA lig Excess linkers are removed by treatment with EcoRI. which also frees each fragment from plasmid vector. Each fragment is excised from low-melt agarose and ligated to pZO1 (Example 2), which has been cut with EcoRI and treated with calf intestinal alkaline phosphat (CAP) and excised from low-melt agarose (see Fig. 1). In general, only one orientation of e insert in pZO1071 is recovered. For cloning into EcoRI cut plus CAP treated pZO1051, e fragment is excised from its corresponding insert in pZO1071 by restriction with EcoRI. In case, the effort is made to recover inserts in each of the two possible orientations (designated and "b") for MRS3, 4, and 5. The sequence of MRS5 is given in Table IA (SEQ ID NO Table IB gives the sequence of MRS4 (SEQ ID NO:2) and Table IC gives a partial sequenc MRS3 (SEQ ID NO:3).

TABLE IA MRS 5

Eco RI

GAATTCTTATCGATACTGGAACTCAGAGCATAGGGGGGAAAGTCGATTTATGGATGGAATCAAATACGCA

A BO (9/10) Topo 11(12/15) GTATTTACAGAAAAGAGTCTTCGTTTATTGGGAAAGAATCAATATACTTTTAATGTCGAATCGGGATTCA CTAAGACAGAAATAAAGCATTGGGTCGAACTCTTCTTTGGTGTTAAGGTAGTAGCTGTGAATAGCCATCG ACTACCCGGAAAGGGTAGAAGAATGGGACCTATTCTGGGACATACAATGCATTACAGACGTATGATCATT

T Bo (9/10) ACCCTTCAACCGGGTTATTCTATTCCACTTCTAGATAGAGAACGAACTAAAGGAGAATACTTAATAATAC

Topo 11(13/15) GGCGAAACATTTATACAAAACACCTATCCCGAGCACACGCAAGGGAACCGTAGACAGGCAAGTGAAATCC AATCCACGAAATAAATTGATCCATGGACGGCACCGTTGTGGTAAAGGTCGTAATGCCAGAGGAATCATTA CCGCAAGGCATAGAGGGGGAGGTCATAAGCGCCTATACCGTAAAATCGATTTTCGACGGAATCAAAAAGA CATATCTGGTAGAATCATAACCATAGAATACGACCCTAATCGAAATGCATACATTTGTCTCATACACTAT

T Box(8/10) GGGGATGGTGAGAAGAAGATATATTTTACATCCCAGAGGGGCTATAATTGGAGATACTATTGTTTCTGGTA

Sna Bl CAAAAGTTCCTATATCAATGGGAAATGCCCTACCTTTGAGTGCGGTTTGAACTATTGATTTACGTAATTG GAAGTAACCAATTAGGTTTACGACGAAACCTAGAAATCGATCACTGATCCAATTTGACTACCTCTACGGG ATAGACCTCAACAGAAAACTGTTGAGTAACGGCAGCAAGTGATTGAGTTCAGTAGTTCCTCATAGAAAAT TATTGACTCTAGAGATATGGTAATATGGAGAAGACAAAATTGTTTGAAGCACGCACAGAACCGGAAGCGC CCCTTGTTTCAAAGAGAGGAGGACGGGTTATTCACATTTAATTTGATGGTCAGAGGCGAATTGAAAGTTA AGCAGTGGTAATTAAGACCCCCGGGTGAAAATAGGGATGTCTCCTACGTTACCCATAATATGTGGAAGTA TCGACGTAATTTCATAGAGTCATTCGATCTGAATGCTACATGAAGAACATAAGCCAGATGACGGAACGCG GAGACCTAGGATGTAGAAGATCATAACATGAGCGATTCGGCGGATTTGGATTCCTTTTCTATATATCCAC TCATGTGGTACTTCATCATACGATTCATATAAGATCCATCTGTCTAGAGATCGTCATATACATCTAGAAA GCCGTATGCTTTGGAAGAAGCTT

HIND III

60% AT Table IB MRS4 gaattctgtggaaagccgtattcgatgAAAGTCGTATGTACGGCTTGGAGGGAGATCTTTCCTATCTTTC GAGATCCaccctacaatatgGGGCCAAAAAGCCAAAAAAATAAGTGATTCGTTTTTAGCCCTTATAAAAA GAAAACGGATTCTTGAACCTCTTTCACGCTCATGTCACGTCGAGGTACTGCAGAAAAAAGAACCGCAAAA TCCGATCCAATTTTTCGTAATCGATTAGTTAACATGGTGGTTAACCGTATTATGAAAGACGGAAAAAAAT CATTGGCTTATCAAATTCTCTATCGAGCCGTGAAAAAGATTCAACAAAAGACAGAAACAAATCCACTATT G3TTTTACGTCAAGCAATACGTAGAGTAACTCCCAATATAGGAGTAAAAACAAGACGTAATAAAAAAGGA TCGACGCGGAAAGTTCCGATTGAAATAGGATCTAAACAAGGAAGAGCACTTGCCATTCGTTGGTTATTAG AAGCATCCCAAAAGCGTCCGGGTCGAAATATGGCTTTCAAATTAAGTTCCGAATTAGTAGATGCTGCCAA AGGGAGTGGGGGTGCCATACGCAAAAAGGAAGCGACTCATAGAATGGCAGAGGCAAATAGAGCTCTTGCA CATTTTCGTTAATCCATGAACAGAATCTAGGTATGTAGACACATGGATCCATACATCTCGATCGGAAAAG AATCAATAGAAGGAGAATCGGACGATATCTTTTTCGAAACAAATAAAAAGGAAAAAAAAGAGAAAACAGA AATCATGATCAACTAAGCCTCTCGGGGGCTTGCTTAAGAATAAGAAAGAGGAATCTTATGGAAATAGCAT GGAATAAGGTTTGATCCTATTCATGGGGATTCCGTAAATATCCCATTCCAAAAATCGAAACAATCGGGAC TTTTCGGAGATTGGCTGCAGTTΑCTAATTCATGATCTGGCATGTACAGAATGAAAACTTCATTCTCGATT CTACGaGAATTTTTATGAAAGCGTTTCATTTGCTTCTCTTCCATGGAAGTTTCATTTTCCCAGAATGTAT CCTAATTTTTGGCCTAATTCTTCTTCTGATGATCGATTTAACCTCTGATCAAAAAGATAGACCTTGGTTC TATTTCATCTCTTCAACAAGTTTAGTAATAAGCATAACGGCCCTATTGTTCCGATGGAGAGAAGAACCTA TAATTAGCTTTTCGGGAAATTTCCAAACGAACAATTTCAACGAAATCTTTCAATTTCTTATTTTATTATG TTCAACTTTATGTATTCCTCTATCCGTAGAGTACATTGAATGTACAGAAATGGCTATAACAGAGTTTCTG TTATTCGTATTAACAGCTACTCTAGGGGGAATGTTTTTATGTGGTGCTAACGATTTAATAACTATCTTTG TAGCTCCAGAATGTTTCAGTTTATGTTCCTACCTATTGTCTGGATATACCAAGAGAGATCTACGGTCTAA TGAGGCTACTATGAAATATTTACTCATGGGTGGGGCAAGCTCTTCTATTCTGGTTCATGGTTTCTCTTGG CTATATGGTTCATCTGGGGGGGAGATCGAGCTTCAAGAAATTGTGAATGGTCTTATCAATACACAAATGT ATAACTCCCCAGGAATTTCAATTGCGCTTATATTCATCACTGTAGGACTTGGGTTCAAGCTTU

Table IC MRS3

AAGCTTCTTCCAAAGCATACGGCTTTCTAGATGTATATGACGATCTCTAGACAGATGGATCTTATATGAA TCGTATGATGAAGTACCACATGAGTGGATATATAGAAAAGGAATCCAAATCCGCCGAATCGCTCATGTTA TGATCTTCTACATCCTAGGTCTCCGCGTTCCGTCATCTGGCTTATGTTCTTCATGTAGCATTCAGATCGA ATGACTCTATGAAATTACGTCGATACTTCCACATATTATGGGTAACGTAGGAGACATCCCTATTTTCACC CGGGGGTCTTAATTACCACTGCTTAACTTTCAATTCGCCTCTGACCATCAAATTAAATGTGAATAACCCG TCCTCCTCTCTTTGAAACAAGGGGCGCTTCCGGTTCTGTGCGTGCTTCAAACAATTTTGTCTTCTCCATA TTACCATATCTCTAGAGTCAATAATTTTCTATGAGGAACTACTGAACTCAATCACTTGCTGCCGTTACTC AACAGTTTTCTGTTGAGGTCTATCCCGTAGAGGTAGTCAAATTGGATCAGTGATCGATTTCTAGGTTTCG TCGTAAACCTAATTGGTTACTTCCAATTACGTAAATCAATAGTTCAAACCGCACTCAAAGGTAGGGCATT TCCCATTGATATAGGAACTTTTGTACCAGAAACAATAGTATCTCCAATTATAGCCCCTCTGGGATGTAAA ATATATCTCTTCTCACCATCCCCATAGTGTATGAGACAAATGTATGCATTTCGATTAGGGTCGTATTCTA TGGTTATGATTCTACCAGATATGTCTTTTTGATTCCGTCGAAAATCGATTTTACGGTATAGGCGCTTATG ACCTCCCCCTCTATGCCTTGCGGTAATGATTCCTCTGGCATTACGACCTTTACCACAACGGTGCCGTCCA TGGATCAATTTATTTCGTGGATTGGATTTCACTTGCCTGTCTACGGTTCCCTTGCGTGTGCTCGGGATAG GTGTTTTGTATAAATGTTTCGCCGTATTATTAAGTATTCTCCTTTAGTTCGTTCTCTATCTAGAAGTGGA ATAGAATAACCCGGTTGAAGGGTAATGATCATACGTCTGTAATGCATTGTATGTCCCAGAATAGGTCCCA TTCTTCTACCCTTTCCGGGTAGTCGATGGCTATTCACAGCTACTACCTTAACACCAAAGAAGAGTTCGAC CCAATGCTTTATTTCTGTCTTAGTGAATCCCGATTCGACATTAAAAGTATATTGATTCTTTCCCAATAAA CGAAGACTCTTTTCTGTAAATACTGCGTATTTGATTCCATCCATAAATCGACTTTCCCCCCTATGCTCTG AGTTCCAGTATCGATAAGAATTC

Example 2: Construction of pZO1051 and pZO1071

The EcoO 1091 site of pUC19 (Yanisch-Perron, C. et al. (1985) Gene 33:103-119) is conven to a Bglll site by filling in and ligation of a Bglll linker to give pZO919. The NPT II ge cassette is assembled by treating pZO919 with Bglll and CAP and ligating to it a 1.7 kb Bgi to Smal fragment, consisting of the 35S promoter (AM to Ddel, Franck A. et al. (1980) C 21:285-294), maize AdhlS intron 2 (Freeling, M. and Bennett, D.C. (1985) Ann. Rev. Gen 19:297-323), TN-5 neomycin phosphotransferase gene (Beck, E. et al. (1982) Gene 19:327-33 and a 0.24 kb blunt to Bglll fragment consisting of the NOS terminator (Bevan, M. et al. (198 Nucl. Acids Res. H:369-385). Plasmid pZO921 consists of the particular orientation of t cassette in pZO919 in which the NOS terminator is closest to the EcoRI site of the multi cloning site. The β-glucuronidase (GUS) cassette consists of the 35S promoter (Ddel to D Franck et al. ibid), maize AdhlS intron 6 (Freeling and Bennett, ibid.), GUS gene (Jefferson, R et al. (1986) PIΌC Natl. Acad. Sci. 83:8447-8451), and NOS teπninator. The GUS cassette then inserted as a 3.1 kb EcoRI to Hindlll fragment into EcoRI plus Hindlll cut pZO921 construct pZO1051 (see Figure 1).

The orientation of the multiple cloning site of pZO919 is reversed by replacing its PvuII fragm with the corresponding PvuII fragment (ca. 300 bp) from pUC18 to form pZO930. The G cassette is then inserted again as an EcoRI to Hindlll fragment to form pZO1068. Finally, NPT II cassette from pZO921 is cloned as a 1.9 kb BamHI to BglHI piece into the Bglll site pZO1068. The orientation of the NPT II cassette for which the 35 S promoter is closest to Hindlll site is named pZO1071 (see Fig. 1).

Example 3: Transient assays

Preparation and electroporation of protoplasts of maize black mexican sweet (BMS) suspensi cells and tobacco suspension cells is done essentially as described (Fromm et al. (1985) Pr Natl. Acad. Sci. USA _,82:5824-5828). Electroporated protoplasts are cultured in the dark for o to two days, then β-glucuronidase assays are done essentially as described (Jefferson, R.A. (19 Plant Mol. Biol. Rep. 5:387-405).

Table 2 shows the results of a transient assay in electroporated BMS protoplasts. The G activity of pZO1071 is set to 1.00. Results are given for each MRS cloned into the EcoRI s of either pZO1051 or pZO1071. When electroporated into BMS protoplasts, pZO1051 has o about 20% as much GUS activity as pZO1071. (Conversely, when NPT II activity is measu pZO1071 has somewhat less activity than pZO1051, data not shown.) Thus the gene which downstream in the direction of transcription shows reduced expression. The upstream gene activity unchanged from that found when another cassette is not present on the same plasm Each MRS is tested for its ability to relieve the inhibition. Whereas none of the MRS ha strong effect on GUS expression per se, as shown by the results when cloned into pZO1071, M 3, 4 and 5 each show some ability to restore the activity of the GUS cassette of pZO1051 to full level. TABLE 2

RELATIVE GUS ACTIVITY IN BMS PROTOPLASTS

INSERT SIZE 1071 1051

NONE 1.00 0.18

MRS1 1.5 1.2 0.25

MRS2 1.2 1.9 0.17

MRS3 1.5 2.1 0.72

MRS4 1.7 2.2 0.66

MRS5 1.4 2.4 0.99

MRS6 1.8 . 1.9 0.12

Example 4

In a further experiment, results from electroporating plasmids into tobacco protoplasts a obtained. The effect of the orientation of the MRS relative to the GUS cassette is also examine Table 3 shows that the GUS activity from pZO1051 is also much reduced compared to pZO107 in tobacco protoplasts, that MRS3, MRS4, and MRS5 can relieve this inhibition, and that the may be a modest preference for orientation of MRS4 and MRS3. In MRS3, the (a) orientatio occurs when the original EcoRI site is proximal to the promoter. In MRS4, the (a) orientatio occurs when the original Hindlll site is proximal to the promoter. In MRS5, the (a) orientatio occurs when the original Hindlll site is proximal to the promoter.

TABLE 3

RELATIVE GUS ACTIVITY IN TOBACCO PROTOPLASTS

INSERT IN RELATIVE

PLASMID pZO1051(OR) GUS (SE)

1071 1.00(.26)

1051 0.40(.05)

1442 MRS3 (a) 1.12(.12) 1466 MRS3 (b) 0.88(.12)

1443 MRS4 (a) l.OO(.ll)

1463 MRS4 (b) 1.50(.20)

1464 MRS5 (a) 0.83(.09)

1465 MRS5 (b) 0.84(.13)

Example 5: Activity of subclones of MRS5

To discover whether there is a region within the MRS which retains the ability to insulate th GUS gene in pZO1051, the ends of the three fragments of MRS 5, as described in the restrictio map below Table 4, are converted to EcoRI sites to allow cloning of these fragments into th EcoRI site of pZO1051. Each orientation is recovered for each fragment. The orientation correspond to those of pZO1464 and pZO1465. The results of transient assays in BMS an tobacco protoplasts shown in Table 4 clearly indicate, at least in the case of MRS5, that fragment retains only a portion of the insulating ability. The A orientation is the sam described in Example 4 for MRS5(a).

TABLE 4 ACTIVITY OF FRAGMENTS OF MRS5

RELATIVE GUS

PLASMID INSERT BMS TOBACCO

•

1071 1.0 1.0

1051 0.20 0.40

1464 MRS5-A 1.0 1.12

1465 -B 0.88

1455 HIND Ill-SNA Bl, 0.60 A 0.50 0.56

1454 B 0.34 0.64

1468 XBA I-XBA I, 0.59A 0.54 0.73

1469 B 0.31 0.49

1456 SNA BI-ECO RI, 0.75A 0.65 0.81

1457 B 0.26 0.80

Example 6: Results in stable callus cultures

Plasmids pZO1071, pZO1051, pZO1442, pZO1443, and pZO1464 are electroporated into B protoplasts, which are plated on filters, then cultured on suitable agar containing medium wi layer of feeder cells and kanamycin, 75mg/L. Extracts are prepared from about twenty kanam resistant calli for each construct. To determine that each callus is a true transformant, NP activity is confirmed by ELISA (5 Prime, 3 Prime, Inc.). GUS activity is meas spectrophotometrically and normalized to total protein.

The frequency of GUS-expressing calli is thought to be highest for those constructs contai MRS3, 4 or 5, and the levels of GUS activity found are more uniform as well.

GUS-expressing calli are maintained for several months on kanamycin containing medium periodically assayed for GUS activity. Individual calli transformed with pZO1051 or pZO10 are found to have lost GUS expression, and the fraction of such calli is found to increase ov several months. However, transformants of [one or more of] pZO1442, pZO1443, and pZO14 are found to maintain GUS expression at a significantly higher frequency.

Thus these MRS which were originally identified to have "boundary" properties in transient assa are found to also behave as stabilizing DNA segments in stable transformants.

Example 7: Stability of progeny

A similar experiment to that of Example 6 is performed, except that a regenerable maize cell li is transformed, either via electroporation or the ballistic method, depending on the cell lin Stable calli recovered after growth on selective media are transferred to suitable regenerati medium for shoot initiation, the shoots are moved to rooting medium, finally resulting in plan which are grown to maturity, characterized for GUS and NPT II expression, and out-crossed selfed. The resulting first progeny generation is also characterized for transgene expression. T expression and heritability of the transgenes continue to be followed for succeeding generation

What is found is that those families of transformants originated from plasmids containing certa MRS show improved stability of expression of the GUS gene, particularly over the generation when compared to families originating from pZO1051 or pZO1071.

Example 8: Sequencing, cloning and transient assays of matrix associated regions (MAR

Two fragments, MAR1 (maize 0.8kb AT rich region) and MAR2 (maize 1.25 kb region wi ARS3), found within a 5kb maize EcoRI fragment originally cloned by R. Berlani et al. (19 Plant Mol. Biol. ϋ:161-172) have nuclear matrix binding activity. The sequence of the MA fragment is given in Table 5A (SEQ ID NO:4) and that portion of MAR2 not previoul sequenced is given in Table 5B (SEQ ID NO:5) along with the published portion of MA named ARS3 (Berlani et al. 1988 Plant Molecular Biology U_: 173-182). Additionally t sequence from the SAR_L a region from a soybean small heat shock gene (MAR3) (soybe HSP17.6 0.4kb SARL) is shown in Table 5C (SEQ ID NO:6). (Schδffl et al. Transgenic Res. 93-100 (1993)). The fragments are sequenced by standard dideoxy methods.

Standard cloning procedures are used in the construction of plasmids for gene activity. MA is subcloned from pZMA321 as an EcoRI -Hindlll fragment into pT7T3-18U (Pharmacia) to fo pZO1927. MAR2 is subcloned from pZMA321 as a Hindlll fragment into pT7T3-18U to fo PZO1929. The Hpall end of MAR3, originally a Hpall-EcoRI fragment, had already b converted to EcoRI in pSVB20-SAR_L (Schδffl, supra). For cloning into EcoRI sites, pZO19 was cut with Hindlll, treated with T4 DNA polymerase I, then EcoRI linkers (NEB) added ligation, followed by restriction with EcoRI. The Hindlll ends of MAR2 are converted to Eco sites by PCR with pZO1029 as template using the primers "1929ECO TGAGGAATTCGCGGTCTATCCCCCGCACG, SEQ ID NO: 7, for the M13R side, a "1929ECOU", GTCGGAATTCAAGTTCCACAACTGAGACAAG, SEQ ID NO: 8, for the M1 side, followed by restriction with EcoRI. . MAR3 is directly cloned into EcoRI sites followi restriction with EcoRI. For cloning into Hindlll sites, pZO1927 and pSVB20-SARL are cut w EcoRI, treated with T4 DNA polymerase I, Hindlll linkers ligated, followed by restriction w Hindlll. MAR2 is cloned directly into Hindlll sites following restriction with Hindlll.

Plasmids pZO 107 land pZO1051 are constructed according to Example 2 and each MAR fragm is cloned into the EcoRI sites. These plasmids are contructed according to Example 2. For o orientation of each MAR at EcoRI, a second copy is also cloned into the Hindlll site to gi plasmids in which the GUS cassette is bound by a pair of MARs.

Preparation and electroporation of protoplasts of maize black mexican sweet (BMS) suspensi cells are decribed in Example 3. Table 6 shows the mean results of a number of transient assa in electroporated BMS protoplasts. The GUS activity of pZO1071 is set to 1.00. Results given for each MAR cloned into the EcoRI site of either pZO1071 or pZO1051. For MAR1 t A orientation occurs when the Clal site is proximal to the promoter. For MAR2, the orientation occurs when the pair of Sac II sites are proximal to the promoter. For MAR3 the orientation occurs when the EcoRV site is distal to the promoter. Results with pZO1701 indic that MARS do not significantly effect expression in the GUS cassette in either the 5 '(EcoRI) 3'(HindIII) locations. GUS activity from pZO1051 is also reduced compared to pZO1071. Ea MAR tested can partially relieve this inhibition. TABLE 5A Sequence of MAR1 (pZ01927)

Eco RI A Box (8/10)

GAATTCAGGTAATCCCGTCGGCCCAAAACCGACGGGAATTAATAGTCGGAGTTTAGTTAATTCTCGTAGG

Topo II (13/15) TAGCTGATGGGAATTAGTAATTCCCGTCGGTTTACGCGCAGTCGACATGAATTAATTAGTCAATTCCCGT

CAACCAGTTAATTTCTGTCGGACACGTCTGACCCATGGGAATTATTCGACGAGGTAATCCAAATCCACGA

Topo II (13/15) GGTCTTTGTAACAATTAATAGAGAATGCAAACTTGGACTTGATTGACATCAGCTGGGTCACGAAATCGAG

AACGGTCACATCAGTGTGCTCATGAAGAGGCTCTTTTGACGCTTTGAGGAGGTCAAAGAACTTCTGAACC

TCCCGGTGTAGTGAATCATGATGATATGGGGCCAATTTCGACTGCAAATCCACAAGCATCCGCCTAATAT ARS (10/12) /T Box (9/10) TATATATTTGTGACAAAGCAATTGCATGGTTTAGAAACATCTGAGTTTTGGCAAACCATTCACTTGTGTG

ACACCCACCCTTTTTGAAACCTTCATACATCCAATCACGTCTATCACCCATTATTGCGGCTGTGTACAAG

AT Box TAAGAAGTGTGTGTGAGACATTCATATTTCCTACACATCACACATAACATGTATGGTACATACATGTGAT

T Box (8/10) GCATAGCGGTCTGAAATGAGTGACACATAGTTTGCAAAACTAT-ATATGTAGTTGTGAAAGGGAAATAGTC

Clal TCAACATTTCCTATAATCGATTTGGGTGTTTGACGACCATAACAAACCTTGTGGACTAACCAGTTTGTCT

T Box (8/10) AGTTGATCATTCCACAGGTGCATAAGTTCATCTACAACTATTCTAAATCGACTATCCAGAATACCGTAGA

Hindlll TTATTTCGGACAGGAGAAGCTT 59% AT

TABLE 5B Sequence of MAR2 (pZ01929)

Hindlll Topo 11(13/15)

AAGCTTCCAATCTCTCCAAGTTCCACAACTGAGACAAGTGATCATTAGTGATTATAGACTTGAGAGAGAG

AGAGAGTGATCCGTGTATTATTTATCGCTCTTGTTGCTTGGCTTTTGCAATCGTGCTTTCTTCTATTCCC

ATTCTTATTCTCAAGTGACTTGTAATCAAAGAAAGAGACACCAAGTGTGGGGTGGTCCTTGTGGGGTCTA

AGTGACCCGGTTGATTAAGGAGAAAGCTCAGTCGGTCTAGGTGGCCGTTTGAGAGAGGGAAAGGGTTGAA

AGAGACCCGGTCTTTGTGACCACCTCAACGGGGACTAGGTTCTTTGGAACCGAACCTCGGTAAAACAAAT

T Box(8/10)/A Box(9/10) T Box(8/10)

CACCGTGTCATCCGCTTTATTTCTTGGTTGATTTGTTTTCGCCCTCTCTCCTAGACTTGGATTTTATTCT

AACGCTAACCCCGGCTTGAAGTGTGCTTAAAGTTTGTAAATTTCAGTTTCCGCCTATCCACCCCCCTCTA

Topo 11(13/15) GGCGACTTTCAAGTTGCAACAATGAGGTAGTAGAATCAATACTTANATAACATGACATTAATCANTTAAC Topo 11(14/15) Toppo 11(13/15)

AATATTCANAACGAATTAATAATTTGCAACAATTACTAGGTGTATAACAACACAGTCACCATCAAAATTC

AT Box TopoII(13/15)/ARS(12/12)/T Box(9/10) ATCAACTAATAACATCAAGCCACATAGTTTATATTTGCAACATAAATATAAAAATAGCAACTACAATGTA

A BOX (9/10) Topo 11(14/15)

TAAAGTCATATTAAcraCTAATAACACTATCAATTaACAAATTTAAGATAACTATAATTGCATAAAAGTTA

T Bo (8/10) Topo 11(13/15) ACTCTCGTCGGTCACAAAAAACTGACGAAAATAAATGTCTTAATATATATATTAAGCACCTCTAATTAAT

CACACTTTCATTTGTATCGCATAGGTCTAGGATTATACCTCGACGACTCGAATGACGTTGTCCCGTGGAG

<— GAGCTTGTGTAGGCCGGATGGCAGCACAGACGGTGGTCGTTGGAGGAGATTGCGAGGGCCGATGGGGGGG

SacII CGTGGTGGCCGCGGTAGTGGGCAGTCCCAACAATGTGGCAGTCACGGTTGGCGCCGGCGATGGGCGTACC SacII GCGGTGGGCAGTCCCGACGGCGTGGCAGTGNCGGTGGGCGCCGGCGACAGGCGGACCTCCAATGGCGGAC

GACGTTGGGGCGGCCAGGCGGAGGGCGGTGGCAGTGGGCGTAGGGCGACGACGGGGGACAGACCGGACCG

GCGACGACGACGTTGAAGAAAATGTTGCGGTTTGAAAATGAGCCCGTGCGGGGGATAGACCGCGCCTCAT

Hindlll AAAAGCTT 54% AT

TABLE 5C Sequence of MAR3 (SAR_L)

Hpa II ARS (11/12) A Box (9/10) ARS (10/12)

GTAACTAGCAAGTTCAGAGCATCATTTAAGTAATTAAAAGAAAAAATATTAAATATATAAATCATAAGA

Eco RV T Box (8/10) AT Box T Bo (9/10)ARS(10/12)T Box(9/10)

TGATATCAAAAAATTCATGAACAGTCTCTTCATTTTTTTTCAATAAAAATATTTTTATTTTAATTTTTTA

AAATAATATCCTCATAACATTGGTTTAACTCCCAAGTTTAAAATTTACTAGTGCTAGATAAATTCTCTAA

ARS (10/12) ARS (10/12) ARS(10/12) T Box(9/10)

GATAATGTATAGATAAAAATAAGATAAATTAGAAAATTTTTAAGGAGAGATTTTTTTTTATAAAAATTAG

AT Box ARS(10/12)T Box(9/10) Topo II (15/15) GTATATGTATTGGTTTTAGTTTACAGAGAAATATAATTTATATTTTCTTTTTGTGTAAATATTAATGAAA

T Box (9/10) Eco RI AAAATTATTCAAATTCAATCTAAATCTTAATATTTTTTTTGACAGAATCC

82% at TABLE 6

RELATIVE GUS ACTIVITY IN BMS PROTOPLASTS

PLASMID DESCRIPTION OF RELATIVE GUS (SE)

INSERT INTO pZO1071 or pZO1051

1071 GUS> NPTII> 1.00 (.08)

1941 MARIA GUS> NPTII> 1.12(.08) 1940 MAR1B GUS> NPTII> l.lθ(.lθ) 1967 MAR1B GUS> MAR1B NPTII> 0.90 (.07)

1970 MAR2A GUS> NPTII> 0.86(0.06) 1948 MAR2B GUS> NPTII> 1.12(.16) 1971 MAR2B GUS> MAR2B NPTII> 0.89(.13)

1944 MAR3A GUS> NPTII> 0.99(.18) 1945 MAR3B GUS> NPTII> 1.13(.18) 1962 MAR3B GUS> MAR3B NPTII> 0.98 (.04)

1051 <GUS <NPTII 0.20(.01)

1934 <GUS MARIA MARIA <NPTII 0.90(.06) 1937 <GUS MAR1B <NPTII 0.65(.06) 1966 MAR1B <GUS MAR1B <NPTII 0.54 (.08)

1942 <GUS MAR2B <NPTII 0.62(.05) 1968 MAR2B <GUS MAR2B <NPTII 0.59(.07)

1949 <GUS MAR3 <NPTII 0.43 (.12) 1965 MAR3B <GUS MAR3B <NPTII 0.40(.15)

Example 9: Results of stable tansformants of MARs

A series of experiments similar to that of Example 6 are performed with MARs incorporated in plasmids as described above. In one expermient, insertions of MARl in the EcoRI site pZO1051 are tested. The results are reported in Table 7, where (n) is the number transformants. There is an increase in the mean GUS levels of transformants pZO1934 an pZO1937 compared to ρZO1051. The proportion of pZO1934 and pZO1937 transformants wit GUS activity over 0.5OD/hr/mg protein is at least double compared to the pZO105 transformants. Similar results are seen for contructs with two MARs bounding the GUS casse (data not shown). TABLE 7 STABLE GUS ACTIVITY IN BMS PLASMID DESCRIPTION Mean GUS ACTIVIT (n) Fraction>0.5 OD/hr

(OD/hr/mg protein) 1071 GUS> NPTII> 0.92 (57) 0.58

1051 <GUS <NPTII 0.77 (54) 0.33

1934 <GUS MARIA MARIA <NPTII 1.26 (47) 0.74

1937 <GUS MAR1B <NPTII 1.43 (46) 0.84

Example 10; Isolation of the nuclear matrix MAR binding system

Protoplasts are isolated from 20 grams leaves, resuspended in W5 medium, Menczel et al. (198 Theor. Appl. Genet. 59:191-195, spun down (7' 80 g) and resuspended in 15 ml IB (20 mM hep pH 7.4, 0.05 mM spermine, 0.125 mM spermidine, 20 mM KCl, 1% thiodiethanol, 1 M hexyle glycol, 0.5 mM EDTA, 0.5% Triton-X-100,™ 0.2 mM PMSF, 5 mg/ml aprotinin, 10 mM E [trans-epoxy succinyl-L-leucyclamide[4 guanidino] butane]). The protoplasts are homogeniz by vortexing 20" and the resulting homogenate is centrifuged for 7' (80 g). The supernatant centrifuged (10' 300 g) and the resulting pellet (crude nuclei) is purified on a 15% perc gradient made in IB (15' 600 g). An interphase and/or a "smear" on the wall of the tube mig appear, both fractions do contain many contaminations and few nuclei. Purified nuclei in t pellet fraction are resuspended and washing in IB buffer without Triton,™ followed centrifugation (10' 400 g). Resulting pellet is washed and centrifuged again in the same buf (10' 300 g); final pellet is resuspended in 2 ml IB without Triton.™ Nucleus isolation efficien is routinely 40% as determined with DAPI staining (12 X IO⁶ nuclei).

Portions of 12 X IO⁶ nuclei are washed in 10 ml WB (3.75 mM Tris pH 7.4, 20 mM KCl, mM EDTA, 1% thiodiethanol, 0.05 mM Spermine, 0.125 mM Spermidine, 0.1% digitoni tracylol 1 mg/ml) and spun down for 10' (400 g). Washed pellets are resuspended in 100 μl and the nuclear matrix is stabilized by incubating for 20' at 42°C in a shaking water bat Histone proteins are extracted by incubating the stabilized nuclei with 10 ml LIS-HLE buffer ( mM Hepes pH 7.4, 0.1 M LiAc, 1 mM EDTA, 4 mg/ml LIS [3',5'-diiodosalicylate], 0.1 digitonin, 25 μg/ml tracylol, 1.5 mM PMSF) for 5' at room temperature. The chromosomal DN which is not bound to the skeleton of the nuclei (± 90%), looped out; after centrifugation ( 13.000 g) the extracted nuclei appear as a fluffy pellet, consisting of a stabilized nuclear skelet (the nuclear matrix or nuclear halos) associated with looped out chromosomal DNA. The matrices are washed 3 times with 12 ml DB (20 mM Tris pH 7.4, 20 mM KCl, 70 mM NaCI, mM MgCl₂, 0.05 mM Spermine, 0.125 mM Spermidine, 0.2 mM PMSF); the last matrix pell is resuspended in 9.6 ml DB containing 1200 units restriction enzyme. Looped out DNA removed by digestion for 60' at 37°C. At that stage the matrices are suitable for DNA bindi assays to select MAR sequences.

Exa glej : Characterization of nuclear matrix binding assays to test potential MAR DN sequences

A nuclear matrix MAR binding system, prepared from IO⁶ nuclei, is incubated overnight in D buffer at 37°C with 5 ng γ-³²P-end labelled digested plasmid carrying potential MAR sequence Binding of MAR DNA fragments will take place either at "empty" matrix binding sites or "occupied" matrix binding sites ("displacement" binding of endogenous MARs). After bindin the mixture is spun down; the pellet (containing the nuclear matrix MAR binding syste associated with bound labelled DNA MAR fragments) is washed once with DB and spun agai The two supernatant fractions (containing the unbound labelled DNA fragments) are pooled a the final pellet is resuspended in 200 μl DB. DNA is isolated from both the supernatant and t pellet fractions by a SDS/proteinase K treatment, followed by a phenol/chloroform extraction, t ether extractions and an ethanol precipitation. Both the DNAs isolated from the supernatant- a the pellet-fractions are electrophoresed on horizontal agarose gels, the gels are dried a autoradiographic exposure is performed for 1-3 days at -70°C with intensifying screens. DN fragment end-labelling, DNA extractions, electrophoresis and X-ray exposures are all perform according to Maniatis (Maniatis et al. 1989).

To determine affinity of binding a series of binding experiments are performed in which t concentration of nonspecific sonicated R. coli competitor DNA is varied. Depending on t quality of the nuclear matrix MAR binding system and the affinity of the DNA fragments to bound, different concentrations of competitor DNA are needed to result in specific binding of onl the MAR fragment Figure 2 represents an example of such binding; without E coli competit DNA the MAR- containing fragments are bound to the nuclear matrix system as well as all t other vector fragments (nonspecific condition). However, whereas the vector fragments are easil displaced into the supernatant by increasing E. coli DNA concentrations, only the binding of t MAR fragment persists under stringent competition conditions (20 μg T coli DNA is a 20,0 molar excess over the 1 kbp MAR fragment). The amount of _, coli competitor DNA wh prevents specific MAR binding, is a reference for the binding affinity of the MAR fragment.

Binding in such a system is specific as only the MAR-containing fragment is bound to the pel fraction when the proper nonspecific E. coli competitor DNA quantity is applied (see Fig. Binding in such a system is also saturable as liquid scintillation counting of a bound tomato M fragment in experiments with increasing concentrations of added end-labelled restriction fragme containing that tomato MAR [1-250 ng], show that between 50 and 100 ng fragment maximum binding level is obtained (see Fig. 3).

MAR binding sites in nuclear matrices are nonselective for a specific MAR fragment as ot MAR fragments can compete for binding effectively, consequently MAR binding is a reversible. In Figure 4 this is demonstrated for the binding of a tomato MAR; it can be compet to approximately 20% by a 6-fold excess of rat MAR. This suggests that MAR matrix interacti is conserved during evolution.

When heat stabilization during nuclear matrix isolation is omitted, the nuclear matrix skelet (internal matrix consisting of residual nucleoli and granular clusters of electron dense clust embedded in a highly branched network of thin filaments) is not stabilized, resulting in "empty" nuclear shell, consisting of only the external lamina, specific and saturable MAR D fragment binding, similar to the nuclear matrix system, is also possible with nuclear shells (s Fig. 5). The only difference is that binding at the nuclear shell binding sites is more rapi competed by nonspecific R. coli DNA, demonstrating that the number of MAR binding sites nuclear shells is lower relative to the nuclear matrix.

Example 12; Isolation of plant MAR DNA sequences from nuclear matrices

Nuclear envelopes are isolated from tomato protoplasts according to Kaufmann and Shaper (19 Exp. Cell Res. 155:477-497; Lam Bl-like molecules are isolated from those envelopes described by Aebi et al. (1986) Nature 323:560-564. Lamin Bl is coupled to inert columns, e. sepharose CL 4B, using cyanogen bromide, which are used as selection tools for potential M sequences.

Tomato nuclear chromosomal DNA is isolated according to Bernatzki, R. and Tanksley, (1986) Plant Mol. Biol. Reporter 4:37-41. Mbol digested tomato DNA is passed over a Lam Bl affinity column and specifically bound DNA fragments are eluted, cloned and furth characterized for nuclear matrix binding in the nuclear matrix MAR binding system as describ in Example 9.

Example 13; Isolation of plant MAR DNA sequences from a potential chromatin lo surrounding a stable transgene

The tomato spotted wilt virus (TSWV) (Peters, D. et al., Proceedings USDA Workshop, Beltsvill MD., Hsu and Lawson (eds.), Nat. Tech. Inf. Serv., Springfield, VA (1991); Gielen, JJ.L. et (1991) Bio/Technology 9:1363-1367) nucleocapsid gene is cloned in the plant transformati vector pBIN19 (Bevan, M. (1984) Nucl. Acids Res. 12(22):8711-8721) containing the NPT selection gene. Transformants, selected for kanamycin resistance, are analyzed for both the cop number of the TSWV transgene by Southern blotting hybridization and for expression of t TSWV nucleocapsid gene by specific ELISA assays. Transformants, containing one single cop of the TSWV nucleocapsid gene, are selfed and tested for gene expression stability in t successive generations by ELISA analysis. A genomic cosmid DNA library is constructed fro purified tomato chromosomal DNA, isolated from a stable transgenic S3 line using the restrictio endonuclease Mbol.

Colony hybridization screening with the TSWV nucleocapsid gene as probe results in th identification of clones carrying the transgene; using chromosome walking techniques (Mania 1989), 100 kbp regions upstream and 100 kpb regions downstream of the transgene are identifie and further characterized. This 200 kbp region includes part of an euchromatin ("open") loo most probably this region includes a complete euchromatin loop, as the sizes of average chromati loops are 80-90 kbp (Jackson et al. 1990). Subclones of this 200 kbp region are tested f specific nuclear matrix binding in the nuclear matrix MAR binding system, as described i Example 9. Several tomato chromosomal DNA fragments are identified as specific MAR DN sequences.

Example 14: Isolation of plant MAR DNA sequences flanking 5' and 3' of nuclear genes

A genomic cosmid DNA library is constructed (using the restriction endonuclease Mbol) fro purified tomato chromosomal DNA. Colony hybridization screening results in a clone containin the tomato plastocyanin gene with about 20 kbp flanking region from both sides. The 5' regio is subcloned and tested for specific binding in the nuclear matrix MAR binding system described in Example 9. An approximately 1 kbp 5' upstream tomato chromosomal D subfragment is identified as a specific MAR DNA sequence.

Example 15; Identification of hypersensitive chromosomal DNA regions

Chromatin is organized in topologically constrained DNA loops by the anchoring of specific M sequences to the external (shell) and internal nuclear matrix. Loops carrying potential acti genes (called dispersed euchromatin) are thought to contain regions which are accessible transcription factors, RNA polymerases and other components required for transcription, where inactive loop (regions) (called condensed heterochromatin) are inaccessible. A transgene in stable transgenic (plant) line is supposed to be integrated in an at least partly dispersed, op euchromatin loop region, which is transcriptionally active. The accessibility of such chroma loop regions carrying active (trans)genes is reflected by an increased sensitivity to nuclea digestion as compared to inactive chromatin loop regions (Weintraub and Groudine, 197 Molecular mechanisms underlying the controlling of open/dispersed or closed/ condensed lo (regions) are yet not understood. However, certain cis-acting enhancer-like DNA fragme (called LCR - locus control region (Felsenfeld, G. (1992) Nature 355:219-223) are supposed act as openoclose switches/regulators for the chromatin conformation in loop regio Differentiation is a process of tissue-specific differential gene expression by means of inactivati or closing euchromatin regions into heterochromatin regions. Hypersensitive chromosomal DN regions are defined as DNA regions showing an increased sensitivity toward nuclease digestio Such hypersensitive DNA regions contain cis-acting regulator sequences like LCR sequenc Therefore screening for DNAse I hypersensitive DNA regions is a tool for preselecting chroma conformation-regulating cis-acting elements.

Protoplasts are isolated as described in Example 8, with the modification that the cell w degradation enzymes are applied in (lower) range concentrations, resulting in permeable ce which are intermediate forms between in vivo cells and protoplasts. Cells/protoplasts are wash in W5 medium twice, followed by resuspension in nuclease buffer (0.05 M Tric-HCl, pH 7.8, mM MgCl₂, 0.01 M 2-mercaptoethanol, 10 μg/ml BSA). Cells are treated with varyi concentrations of DNAsel (IO^'3 - IO^'5 U/ml) for 10' at room temperature. DNA is isolated usi a standard phenol/chloroform method, digested with appropriate restriction enzymes and blott to Hybond N⁺. Hybridizations are carried out with probes derived from the 200 kbp genom region surrounding the stable TSWV nucleocapsid gene isolated from a stable tomato transgen line (described in Example 12). Hypersensitive regions (carrying chromatin conformati regulating cis-acting elements) are identified as those regions which show a decrease in amou of probe able to hybridize with the expected DNA restriction fragment when a series of increasi DNAse I concentrations is applied to the permeable tomato cells. Alternatively, identification HR can also be performed by analyzing additional hypersensitive related DNA subfragmen essentially as described by Forrester, W.C. et al. (1990) Genes and Development 4:1637-1649

Example 16: Transgene instability in lettuce

From a typical A grobacterium-mediated transformation experiment, sixteen independent S progeny lines are obtained that express the introduced transgene: the TSWV nucleoprotein ge cassette (a schematic representation of the T-DNA is shown in Fig. 6). Individual S 1 progen plants accumulating the TSWV nucleoprotein are maintained and self-pollinated. The resultin S2 progeny lines representing the 16 transformant lines, are again analyzed for accumulation the TSWV nucleoprotein by ELISA analysis, revealing the segregation ratio of the transgene. this time, abnormal, nonMendelian segregation ratios and "silencing" of gene expression a observed in approximately 50% of the transformant lines, indicating instability of transgen expression. Stable, "homozygous" S2 populations, in which the transgene has been "fixed" in th homozygous state are identified for only six transformant lines.

Southern blot analysis of "unstable" transformant lines that exhibit abnormal segregation rati at the expression level, demonstrates that the transgene segregates normally at the DNA lev within the S2 population. Nine individual S2 plants from two unstable S2 lines (6A-7 and 26-11 are analyzed for presence and expression of the transgene by means of Southern blot analysis an ELISA analysis, respectively. As shown in Fig. 7, a correlation between presence and expressio of the transgene cannot be observed (plant numbers that accumulate the TSWV nucleoprotein determined in the ELISA assay have been encircled; digestion with Xbal releases the complet TSWV nucleoprotein gene cassette of 1.6 kb, while digestion with Hindlll generates bord fragments correlated with the number of T-DNA copies integrated into the genome). Th transgene itself segregates "normally" at the DNA level; only two out of the nine individual analyzed do not carry the transgene, but some plants that do carry the transgene do not expres the transgene, which illustrates the instability of transgene expression. Southern blot analysis of a number of nonexpressor lines digested with Xbal demonstrates presence of the transgene, but none of these S2 lines accumulate TSWV nucleoprotein. Ag a correlation between the presence and expression of the transgene cannot be demonstrated. few of these S2 lines still accumulate the nucleoprotein gene product in the SI populat (transformant lines 11 and 21), illustrating the progressive "silencing" of transgene express over generations.

Instability of transgene expression has also been observed in two transformant lines carrying bar gene from streptomyces hygroscopicus which confers resistance to phosphinothricine, active compound in a number of commercial herbicides. Herbicide treatment of SI and progeny populations revealed non-Mendelian segregation ratios and fixation of the herbic resistance trait in homozygous lines could not be accomplished.

Example 17: Transgene instability in tomato

In tomato a similar example of transgene instability is observed upon introduction of the TS nucleoprotein gene cassette. As in the case of lettuce, the first signs of "silencing" of transg expression are encountered during ELISA analyses of the progeny populations revealing abnor non-Mendelian segregation ratios. Combined ELISA and southern blot analysis of nine individ S2 progeny plants (transformant line 27-5) descending from one primary transformant result the identification of certain plant individuals that carry an inactivated transgene. Digestion w Xbal releases the complete TSWV nucleoprotein gene cassette of 1.6 kb, while digestion Hindlll generates border fragments, the number of which correlates with the number of T-D copies integrated into the genome. Apparently, the transgene itself segregates normally at DNA level; only one-third of die individuals analyzed have "lost" die transgene thro segregation, but others may harbor transgenes that have been inactivated, resulting in abnor segregation ratios at the protein level.

Detailed analysis of individual tomato plants tiiat descend from unstable transformant li provided more insight in die phenomenon of instability. Upon sampling of leaf material diffe in age from one and the same plant ELISA-positive as well as ELISA-negative samples w identified, resulting in a 'mosaic' pattern of transgene expression throughout the plant. correlation between transgene inactivation and leaf age could not be observed. In vi regeneration of tomato shoots from explants taken from 'silenced' transformants resulted in par re-activation of transgene expression, in that ELISA-positive as well as ELISA-negative sho could be identified. This result proves the possiblitity of reactivating 'silenced' transgen through tissue culture.

Example 18: Stabilization of gene expression over generations using MAR sequences

To assess the effects of MAR sequences as boundary elements on gene expression ov generations, a number of "T-DNA" constructs is transformed to tomato and lettuce, in which t NPTII selection marker and the TSWV nucleoprotein gene cassette are flanked by MA sequences isolated from rat, soybean and tomato (pMARr, pMARs and pMARt, respectivel A schematic representation of these T-DNAs is shown in Fig. 8. The pMARc construct carri boundary elements consisting of random DNA sequences (e.g., vector DNA) of about equal leng to the MAR elements, but that do not exhibit any affinity to plant nuclear matrices.

Upon Agrobacterium-mediated transformation of tomato and lettuce, primary transforman accumulating TSWV nucleocapsid protein are analyzed for their T-DNA copy number throu Southern blot analysis. Only transformants that carry a single copy of the T-DNA are maintain and self-pollinated. From the resulting SI progeny population, ten individual plants expressi die nucleoprotein transgene are once more self-pollinated to produce S2 lines. Nonsegregati S2 lines in which the transgene has been "fixed" in die homozygous state, are identified by mea of ELISA analysis for accumulation of TSWV nucleoprotein. These "homozygous" S2 lines a maintained by repeated selfing and monitored for stability of transgene expression over successi generations. In comparison to the pMARc construct, the percentage of stable pMARs and pMA transformant lines after eight generations is significandy higher, which illustrates the stabilizi effect that these MAR sequences exert on transgene expression in tomato. In case of the pMA construct the progressive transgene inactivation as observed for the pMARc vector over successi generations is almost completely prevented. To a lesser extent die same phenomenon is observ for the heterologous rat MAR element, which is not as efficient as the plant-derived MA sequences in stabilizing transgene expression, resulting in an intermediate percentage of stab transformants. In lettuce both plant derived MAR sequences perform equally well. The toma as well as the soybean MAR reduce the instability of transgene expression in lettuce exemplified by the pMARc transformant lines. Apparendy, plant-derived MAR sequences c be used in heterologous transformation systems to stabilize transgene expression over generation SEQUENCE LISTING

( 1 ) GENERAL INFORMATION :

(i) APPLICANT:

(A) NAME: Sandoz Ltd

(B) STREET: Lichtstrasse 35

(C) CITY: Basel

(D) STATE: BS

(E) COUNTRY: Switzerland

(F) POSTAL CODE (ZIP) : CH-4002

(G) TELEPHONE: 061-324-1111 (H) TELEFAX: 061-322-7532 (I) TELEX: 96 50 50 55

(A) NAME: Sandoz Patent GmbH

(B) STREET: Humboldstrasse 3

(C) CITY: Loerrach

(E) COUNTRY: Germany

(F) POSTAL CODE (ZIP) : Dτ79540

(A) NAME: Sandoz Erfindungen Verwaltungs GmbH

(B) STREET: Brunnerstrasse 59

(C) CITY: Vienna

(E ) COUNTRY : Austria

(F ) POSTAL CODE ( ZIP ) : A-1235

(ii) TITLE OF INVENTION: Genetic stabilising elements (iii) NUMBER OF SEQUENCES: 8

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #1.0, Version #1.25 (EPO)

(2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1354 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

GAATTCTTAT CGATACTGGA ACTCAGAGCA TAGGGGGGAA AGTCGATTTA TGGATGGAAT 60

CAAATACGCA GTATTTACAG AAAAGAGTCT TCGTTTATTG GGAAAGAATC AATATACTTT 120

TAATGTCGAA TCGGGATTCA CTAAGACAGA AATAAAGCAT TGGGTCGAAC TCTTCTTTGG 180

TGTTAAGGTA GTAGCTGTGA ATAGCCATCG ACTACCCGGA AAGGGTAGAA GAATGGGACC 240

TATTCTGGGA CATACAATGC ATTACAGACG TATGATCATT ACCCTTCAAC CGGGTTATTC 300 TATTCCACTT CTAGATAGAG AACGAACTAA AGGAGAATAC TTAATAATAC GGCGAAACAT 360

TTATACAAAA CACCTATCCC GAGCACACGC AAGGGAACCG TAGACAGGCA AGTGAAATCC 420

AATCCACGAA ATAAATTGAT CCATGGACGG CACCGTTGTG GTAAAGGTCG TAATGCCAGA 480

GGAATCATTA CCGCAAGGCA TAGAGGGGGA GGTCATAAGC GCCTATACCG TAAAATCGAT 540

TTTCGACGGA ATCAAAAAGA CATATCTGGT AGAATCATAA CCATAGAATA CGACCCTAAT 600

CGAAATGCAT ACATTTGTCT CATACACTAT GGGGATGGTG AGAAGAAGAT ATATTTTACA 660

TCCCAGAGGG GCTATAATTG GAGATACTAT TGTTTCTGGT ACAAAAGTTC CTATATCAAT 720

GGGAAATGCC CTACCTTTGA GTGCGGTTTG AACTATTGAT TTACGTAATT GGAAGTAACC 780

AATTAGGTTT ACGACGAAAC CTAGAAATCG ATCACTGATC CAATTTGACT ACCTCTACGG 840

GATAGACCTC AACAGAAAAC TGTTGAGTAA CGGCAGCAAG TGATTGAGTT CAGTAGTTCC 900

TCATAGAAAA TTATTGACTC TAGAGATATG GTAATATGGA GAAGACAAAA TTGTTTGAAG 960

CACGCACAGA ACCGGAAGCG CCCCTTGTTT CAAAGAGAGG AGGACGGGTT ATTCACATTT 1020

AATTTGATGG TCAGAGGCGA ATTGAAAGTT AAGCAGTGGT AATTAAGACC CCCGGGTGAA 1080

AATAGGGATG TCTCCTACGT TACCCATAAT ATGTGGAAGT ATCGACGTAA TTTCATAGAG 1140

TCATTCGATC TGAATGCTAC ATGAAGAACA TAAGCCAGAT GACGGAACGC GGAGACCTAG 1200

GATGTAGAAG ATCATAACAT GAGCGATTCG GCGGATTTGG ATTCCTTTTC TATATATCCA 1260

CTCATGTGGT ACTTCATCAT ACGATTCATA TAAGATCCAT CTGTCTAGAG ATCGTCATAT 1320

ACATCTAGAA AGCCGTATGC TTTGGAAGAA GCTT 1354 (2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1673 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

GAATTCTGTG GAAAGCCGTA TTCGATGAAA GTCGTATGTA CGGCTTGGAG GGAGATCTTT 60

CCTATCTTTC GAGATCCACC CTACAATATG GGGCCAAAAA GCCAAAAAAA TAAGTGATTC 120

GTTTTTAGCC CTTATAAAAA GAAAACGGAT TCTTGAACCT CTTTCACGCT CATGTCACGT 180

CGAGGTACTG CAGAAAAAAG AACCGCAAAA TCCGATCCAA TTTTTCGTAA TCGATTAGTT 240

AACATGGTGG TTAACCGTAT TATGAAAGAC GGAAAAAAAT CATTGGCTTA TCAAATTCTC 300

TATCGAGCCG TGAAAAAGAT TCAACAAAAG ACAGAAACAA ATCCACTATT GGTTTTACGT 360 CAAGCAATAC GTAGAGTAAC TCCCAATATA GGAGTAAAAA CAAGACGTAA TAAAAAAGGA 420

TCGACGCGGA AAGTTCCGAT TGAAATAGGA TCTAAACAAG GAAGAGCACT TGCCATTCGT 480

TGGTTATTAG AAGCATCCCA AAAGCGTCCG GGTCGAAATA TGGCTTTCAA ATTAAGTTCC 540

GAATTAGTAG ATGCTGCCAA AGGGAGTGGG GGTGCCATAC GCAAAAAGGA AGCGACTCAT 600

AGAATGGCAG AGGCAAATAG AGCTCTTGCA CATTTTCGTT AATCCATGAA CAGAATCTAG 660

GTATGTAGAC ACATGGATCC ATACATCTCG ATCGGAAAAG AATCAATAGA AGGAGAATCG 720

GACGATATCT TTTTCGAAAC AAATAAAAAG GAAAAAAAAG AGAAAACAGA AATCATGATC 780

AACTAAGCCT CTCGGGGGCT TGCTTAAGAA TAAGAAAGAG GAATCTTATG GAAATAGCAT 840

GGAATAAGGT TTGATCCTAT TCATGGGGAT TCCGTAAATA TCCCATTCCA AAAATCGAAA 900

CAATCGGGAC TTTTCGGAGA TTGGCTGCAG TTACTAATTC ATGATCTGGC ATGTACAGAA - 960

TGAAAACTTC ATTCTCGATT CTACGAGAAT TTTTATGAAA GCGTTTCATT TGCTTCTCTT 1020

CCATGGAAGT TTCATTTTCC CAGAATGTAT CCTAATTTTT GGCCTAATTC TTCTTCTGAT 1080

GATCGATTTA ACCTCTGATC AAAAAGATAG ACCTTGGTTC TATTTCATCT CTTCAACAAG 1140

TTTAGTAATA AGCATAACGG CCCTATTGTT CCGATGGAGA GAAGAACCTA TAATTAGCTT 1200

TTCGGGAAAT TTCCAAACGA ACAATTTCAA CGAAATCTTT CAATTTCTTA TTTTATTATG 1260

TTCAACTTTA TGTATTCCTC TATCCGTAGA GTACATTGAA TGTACAGAAA TGGCTATAAC 1320

AGAGTTTCTG TTATTCGTAT TAACAGCTAC TCTAGGGGGA ATGTTTTTAT GTGGTGCTAA 1380

CGATTTAATA ACTATCTTTG TAGCTCCAGA ATGTTTCAGT TTATGTTCCT ACCTATTGTC 1440

TGGATATACC AAGAGAGATC TACGGTCTAA TGAGGCTACT ATGAAATATT TACTCATGGG 1500

TGGGGCAAGC TCTTCTATTC TGGTTCATGG TTTCTCTTGG CTATATGGTT CATCTGGGGG 1560

GGAGATCGAG CTTCAAGAAA TTGTGAATGG TCTTATCAAT ACACAAATGT ATAACTCCCC 1620

AGGAATTTCA ATTGCGCTTA TATTCATCAC TGTAGGACTT GGGTTCAAGC TTU 1673 (2) INFORMATION FOR SEQ ID NO: 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1353 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3: AAGCTTCTTC CAAAGCATAC GGCTTTCTAG ATGTATATGA CGATCTCTAG ACAGATGGAT 60 CTTATATGAA TCGTATGATG AAGTACCACA TGAGTGGATA TATAGAAAAG GAATCCAAAT 120 CCGCCGAATC GCTCATGTTA TGATCTTCTA CATCCTAGGT CTCCGCGTTC CGTCATCTGG 180

CTTATGTTCT TCATGTAGCA TTCAGATCGA ATGACTCTAT GAAATTACGT CGATACTTCC 240

ACATATTATG GGTAACGTAG GAGACATCCC TATTTTCACC CGGGGGTCTT AATTACCACT 300

GCTTAACTTT CAATTCGCCT CTGACCATCA AATTAAATGT GAATAACCCG TCCTCCTCTC 360

TTTGAAACAA GGGGCGCTTC CGGTTCTGTG CGTGCTTCAA ACAATTTTGT CTTCTCCATA 420

TTACCATATC TCTAGAGTCA ATAATTTTCT ATGAGGAACT ACTGAACTCA ATCACTTGCT 480

GCCGTTACTC AACAGTTTTC TGTTGAGGTC TATCCCGTAG AGGTAGTCAA ATTGGATCAG 540

TGATCGATTT CTAGGTTTCG TCGTAAACCT AATTGGTTAC TTCCAATTAC GTAAATCAAT 600

AGTTCAAACC GCACTCAAAG GTAGGGCATT TCCCATTGAT ATAGGAACTT TTGTACCAGA 660

AACAATAGTA TCTCCAATTA TAGCCCCTCT GGGATGTAAA ATATATCTCT TCTCACCATC 720

CCCATAGTGT ATGAGACAAA TGTATGCATT TCGATTAGGG TCGTATTCTA TGGTTATGAT 780

TCTACCAGAT ATGTCTTTTT GATTCCGTCG AAAATCGATT TTACGGTATA GGCGCTTATG 840

ACCTCCCCCT CTATGCCTTG CGGTAATGAT TCCTCTGGCA TTACGACCTT TACCACAACG 900

GTGCCGTCCA TGGATCAATT TATTTCGTGG ATTGGATTTC ACTTGCCTGT CTACGGTTCC 960

CTTGCGTGTG CTCGGGATAG GTGTTTTGTA TAAATGTTTC GCCGTATTAT TAAGTATTCT 1020

CCTTTAGTTC GTTCTCTATC TAGAAGTGGA ATAGAATAAC CCGGTTGAAG GGTAATGATC 1080

ATACGTCTGT AATGCATTGT ATGTCCCAGA ATAGGTCCCA TTCTTCTACC CTTTCCGGGT 1140

AGTCGATGGC TATTCACAGC TACTACCTTA ACACCAAAGA AGAGTTCGAC CCAATGCTTT 1200

ATTTCTGTCT TAGTGAATCC CGATTCGACA TTAAAAGTAT ATTGATTCTT TCCCAATAAA 1260

CGAAGACTCT TTTCTGTAAA TACTGCGTAT TTGATTCCAT CCATAAATCG ACTTTCCCCC 1320

CTATGCTCTG AGTTCCAGTA TCGATAAGAA TTC 1353 (2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 862 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

GAATTCAGGT AATCCCGTCG GCCCAAAACC GACGGGAATT AATAGTCGGA GTTTAGTTAA 60

TTCTCGTAGG TAGCTGATGG GAATTAGTAA TTCCCGTCGG TTTACGCGCA GTCGACATGA 120

ATTAATTAGT CAATTCCCGT CAACCAGTTA ATTTCTGTCG GACACGTCTG ACCCATGGGA 180 ATTATTCGAC GAGGTAATCC AAATCCACGA GGTCTTTGTA ACAATTAATA GAGAATGCAA 240

ACTTGGACTT GATTGACATC AGCTGGGTCA CGAAATCGAG AACGGTCACA TCAGTGTGCT 300

CATGAAGAGG CTCTTTTGAC GCTTTGAGGA GGTCAAAGAA CTTCTGAACC TCCCGGTGTA 360

GTGAATCATG ATGATATGGG GCCAATTTCG ACTGCAAATC CACAAGCATC CGCCTAATAT 420

TATATATTTG TGACAAAGCA ATTGCATGGT TTAGAAACAT CTGAGTTTTG GCAAACCATT 480

CACTTGTGTG ACACCCACCC TTTTTGAAAC CTTCATACAT CCAATCACGT CTATCACCCA 540

TTATTGCGGC TGTGTACAAG TAAGAAGTGT GTGTGAGACA TTCATATTTC CTACACATCA 600

CACATAACAT GTATGGTACA TACATGTGAT GCATAGCGGT CTGAAATGAG TGACACATAG 660

TTTGCAAAAC TATATATGTA GTTGTGAAAG GGAAATAGTC TCAACATTTC CTATAATCGA 720

TTTGGGTGTT TGACGACCAT AACAAACCTT GTGGACTAAC CAGTTTGTCT AGTTGATCAT 780

TCCACAGGTG CATAAGTTCA TCTACAACTA TTCTAAATCG ACTATCCAGA ATACCGTAGA 840

TTATTTCGGA CAGGAGAAGC TT 862 (2) INFORMATION FOR SEQ ID NO: 5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1268 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5: AAGCTTCCAA TCTCTCCAAG TTCCACAACT GAGACAAGTG ATCATTAGTG ATTATAGACT 60

TGAGAGAGAG AGAGAGTGAT CCGTGTATTA TTTATCGCTC TTGTTGCTTG GCTTTTGCAA 120

TCGTGCTTTC TTCTATTCCC ATTCTTATTC TCAAGTGACT TGTAATCAAA GAAAGAGACA 180

CCAAGTGTGG GGTGGTCCTT GTGGGGTCTA AGTGACCCGG TTGATTAAGG AGAAAGCTCA 240

GTCGGTCTAG GTGGCCGTTT GAGAGAGGGA AAGGGTTGAA AGAGACCCGG TCTTTGTGAC 300

CACCTCAACG GGGACTAGGT TCTTTGGAAC CGAACCTCGG TAAAACAAAT CACCGTGTCA 360

TCCGCTTTAT TTCTTGGTTG ATTTGTTTTC GCCCTCTCTC CTAGACTTGG ATTTTATTCT 420

AACGCTAACC CCGGCTTGAA GTGTGCTTAA AGTTTGTAAA TTTCAGTTTC CGCCTATCCA 480

CCCCCCTCTA GGCGACTTTC AAGTTGCAAC AATGAGGTAG TAGAATCAAT ACTTANATAA 540

CATGACATTA ATCANTTAAC AATATTCANA ACGAATTAAT AATTTGCAAC AATTACTAGG 600

TGTATAACAA CACAGTCACC ATCAAAATTC ATCAACTAAT AACATCAAGC CACATAGTTT 660

ATATTTGCAA CATAAATATA AAAATAGCAA CTACAATGTA TAAAGTCATA TTAAGACTAA 720 TAACACTATC AATTAACAAA TTTAAGATAA CTATAATTGC ATAAAAGTTA ACTCTCGTCG 780

GTCACAAAAA ACTGACGAAA ATAAATGTCT TAATATATAT ATTAAGCACC TCTAATTAAT 840

CACACTTTCA TTTGTATCGC ATAGGTCTAG GATTATACCT CGACGACTCG AATGACGTTG 900

TCCCGTGGAG GAGCTTGTGT AGGCCGGATG GCAGCACAGA CGGTGGTCGT TGGAGGAGAT 960 GCGAGGGCC GATGGGGGGG CGTGGTGGCC GCGGTAGTGG GCAGTCCCAA CAATGTGGCA 1020

GTCACGGTTG GCGCCGGCGA TGGGCGTACC GCGGTGGGCA GTCCCGACGG CGTGGCAGTG 1080

NCGGTGGGCG CCGGCGACAG GCGGACCTCC AATGGCGGAC GACGTTGGGG CGGCCAGGCG 1140

GAGGGCGGTG GCAGTGGGCG TAGGGCGACG ACGGGGGACA GACCGGACCG GCGACGACGA 1200

CGTTGAAGAA AATGTTGCGG TTTGAAAATG AGCCCGTGCG GGGGATAGAC CGCGCCTCAT 1260

AAAAGCTT 1268 (2) INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 399 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Zea mays

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

GTAACTAGCA AGTTCAGAGC ATCATTTAAG TAATTAAAAG AAAAAATATT AAATATATAA 60

ATCATAAGAT GATATCAAAA AATTCATGAA CAGTCTCTTC ATTTTTTTTC AATAAAAATA 120

TTTTTATTTT AATTTTTTAA AATAATATCC TCATAACATT GGTTTAACTC CCAAGTTTAA 180

AATTTACTAG TGCTAGATAA ATTCTCTAAG ATAATGTATA GATAAAAATA AGATAAATTA 240

GAAAATTTTT AAGGAGAGAT TTTTTTTTAT AAAAATTAGG TATATGTATT GGTTTTAGTT 300

TACAGAGAAA TATAATTTAT ATTTTCTTTT TGTGTAAATA TTAATGAAAA AAATTATTCA 360

AATTCAATCT AAATCTTAAT ATTTTTTTTG ACAGAATCC 399 (2) INFORMATION FOR SEQ ID NO: 7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 29 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iii) ANTI-SENSE: NO (vi) ORIGINAL SOURCE:

(A) ORGANISM: synthetic

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7: TGAGGAATTC GCGGTCTATC CCCCGCACG 29

(2) INFORMATION FOR SEQ ID NO: 8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 31 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iii) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: synthetic

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8: GTCGGAATTC AAGTTCCACA ACTGAGACAA G 31

Claims

1. A stabilized gene for transforming a host plant cell, said gene comprising a ge exogenous to the plant cell, and at least one stabilizing DNA segment in a 3'- or 5'-flanki region of said gene.

2. A stabilized gene according to claim 1 comprising a first stabilizing DNA segment in 3'-flanking region and a second stabilizing DNA segment in a 5'-flanking region of said gen

3. A plant transformation vector, said vector comprising a gene exogenous to the plant ce and at least one stabilizing DNA segment in a 3'- or 5 '-flanking region of said gene.

4. A vector according to claim 3 comprising a stabilizing DNA segment in a 3 '-flanki region of said gene and a second stabilizing DNA segment in a 5 '-flanking region of said gen

5. A vector according to either of claim 3 or 4, wherein said exogenous gene compris more than one coding region, each coding region having its own promoter.

6. A vector according to claim 5, comprising at least two stabilizing DNA segments, a fir stabilizing DNA segment being situated 5' to both coding regions, a second stabilizing DN segment being situated between said coding regions.

7. A vector according to claim 5, comprising at least two stabilizing DNA segments, a fir stabilizing DNA segment being situated 5' to both coding regions, a second stabilizing DN segment being situated 3' to both coding regions.

8. A method for transforming a plant cell comprising introducing into said cell a stabilize exogenous gene comprising a coding region, a control region at the 5 '-end of said coding regio and at least one stabilizing DNA segment situated in a 3'- or 5'-flanking region of said gene.

9. A method according to claim 8, wherein the stabilized exogenous gene comprises a fir stabilizing DNA segment situated in a 3'-flanking region of said gene and a second stabilizin DNA segment situated in a 5 '-flanking region of said gene.

10. A method according to claim 8, wherein the exogenous gene comprises more than o coding region.

11. A method according to claim 8, wherein the stabilizing DNA segment is selected fro the group consisting of a plant matrix associated region, a plant scaffold attachment region or maize repetitive sequence.

12. A method for securing independent expression of two transformed exogenous genes a plant, comprising inserting between said genes a stabilizing DNA segment.

13. The method of claim 12, wherein the stabilizing DNA element is a lettuce, tomat soybean or maize repetitive sequence.