Synthetic insecticidal crystal protein gene

FIELD OF THE INVENTION
This invention relates to the field of bacterial molecular biology and, in particular, to genetic engineering by recombinant technology for the purpose of protecting plants from insect pests. Disclosed herein are the chemical synthesis of a modified crystal protein gene from Bacillus thuringiensis var. tenebrionis (Btt), and the selective expression of this synthetic insecticidal gene. Also disclosed is the transfer of the cloned synthetic gene into a host microoorganism, rendering the organism capable of producing, at improved levels of expression, a protein having toxicity to insects. This invention facilitates the genetic engineering of bacteria and plants to attain desired expression levels of novel toxins having agronomic value.
BACKGROUND OF THE INVENTION
B. thuringiensis (Bt) is unique in its ability to produce, during the process of sporulation, proteinaceous, crystalline inclusions which are found to be highly toxic to several insect pests of agricultural importance. The crystal proteins of different Bt strains have a rather narrow host range and hence are used commercially as very selective biological insecticides. Numerous strains of Bt are toxic to lepidopteran and dipteran insects. Recently two subspecies (or varieties) of Bt have been reported to be pathogenic to coleopteran insects: var. tenebrionis (Krieg et al. (1983) Z. Angew. Entomol. 96:500-508) and var. San Diego (Herrnstadt et al. (1986) Biotechnol. 4:305-308). Both strains produce flat, rectangular crystal inclusions and have a major crystal component of 64-68 kDa (Herrnstadt et al. supra; Bernhard (1986) FEMS Microbiol. Lett. 33:261-265).
Toxin genes from several subspecies of Bt have been cloned and the recombinant clones were found to be toxic to lepidopteran and dipteran insect larvae. The two coleopteran-active toxin genes have also been isolated and expressed. Herrnstadt et al. supra cloned a 5.8 kb BamHi fragment of Bt var. san diego DNA. The protein expressed in E. coli was toxic to P. luteola (Elm leaf beetle) and had a molecular weight of approximately 83 kDa. This 83 kDa toxin product from the var. san diego gene was larger than the 64 kDa crystal protein isolated from Bt var. san diego cells, suggesting that the Bt var. san diego crystal protein may be synthesized as a larger precursor molecule that is processed by Bt var. san diego but not by E. coli prior to being formed into a crystal.
Sekar et al. (1987) Proc. Nat. Acad. Sci. USA 84:7036-7040; U.S. Patent Application 108,285, filed Oct. 13, 1987 isolated the crystal protein gene from Btt and determined the nucleotide sequence. This crystal protein gene was contained on a 5.9 kb BamHi fragment (pNSBF544). A subclone containing the 3 kb HindIII fragment from pNSBF544 was constructed. This HindIII fragment contains an open reading frame (ORF) that encodes a 644-amino acid polypeptide of approximately 73 kDa. Extracts of both subclones exhibited toxicity to larvae of Colorado potato beetle (Leptinotarsa decemlineata, a coleopteran insect). 73- and 65-kDa peptides that cross-reacted with an antiserum against the crystal protein of var. tenebrionis were produced on expression in E. coli. Sporulating var. tenebrionis cells contain an immunoreactive 73-kDa peptide that corresponds to the expected product from the ORF of pNSBP544. However, isolated crystals primarily contain a 65-kDa component. When the crystal protein gene was shortened at the N-terminal region, the dominant protein product obtained was the 65-kDa peptide. A deletion derivative, p544Pst-Met5, was enzymatically derived from the 5.9 kb BamHI fragment upon removal of forty-six amino acid residues from the N-terminus. Expression of the N-terminal deletion derivative, p544Pst-Met5, resulted in the production of, almost exclusively, the 65 kDa protein. Recently, McPherson et al. (1988) Biotechnology 6:61-66 demonstrated that the Btt gene contains two functional translational initiation codons in the same reading frame leading to the production of both the full-length protein and an N-terminal truncated form.
Chimeric toxin genes from several strains of Bt have been expressed in plants. Four modified Bt2 genes from var. berliner 1715, under the control of the 2' promoter of the Agrobacterium TR-DNA, were transferred into tobacco plants (Vaeck et al. (1987) Nature 328:33-37). Insecticidal levels of toxin were produced when truncated genes were expressed in transgenic plants. However, the steady state mRNA levels in the transgenic plants were so low that they could not be reliably detected in Northern blot analysis and hence were quantified using ribonuclease protection experiments. Bt mRNA levels in plants producing the highest level of protein corresponded to .apprxeq.0.0001% of the poly(A).sup.+ mRNA.
In the report by Vaeck et al. (1987) supra, expression of chimeric genes containing the entire coding sequence of Bt2 were compared to those containing truncated Bt2 genes. Additionally, some T-DNA constructs included a chimeric NPTII gone as a marker selectable in plants, whereas other constructs carried translational fusions between fragments of Bt2 and the NPTII gene. Insecticidal levels of toxin were produced when truncated Bt2 genes or fusion constructs were expressed in transgenic plants. Greenhouse grown plants produced .apprxeq.0.02% of the total soluble protein as the toxin, or 3 .mu.g of toxin per g. fresh leaf tissue and, even at five-fold lower levels, showed 100% mortality in six-day feeding assays. However, no significant insecticidal activity could be obtained using the intact Bt2 coding sequence, despite the fact that the same promoter was used to direct its expression. Intact Bt2 protein and RNA yields in the transgenic plant leaves were 10-50 times lower than those for the truncated Bt2 polypeptides or fusion proteins.
Barton et al. (1987) Plant Physiol. 85:1103-1109 showed expression of a Bt protein in a system containing a 35S promoter, a viral (TMV) leader sequence, the Bt HD-1 4.5 kb gene (encoding a 645 amino acid protein followed by two proline residues) and a nopaline synthase (nos) poly(A)+ sequence. Under these conditions expression was observed for Bt mRNA at levels up to 47 pg/20 .mu.g RNA and 12 ng/mg plant protein. This amount of Bt protein in plant tissue produced 100% mortality in two days. This level of expression still represents a low level of mRNA (2.5.times.10-4%) and protein (1.2.times.10-3%).
Various hybrid proteins consisting of N-terminal fragments of increasing length of the Bt2 protein fused to NPTII were produced in E. coli by Hofte et al. (1988) FEBS Lett. 2.26:364-370. Fusion proteins containing the first 607 amino acids of Bt2 exhibited insect toxicity; fusion proteins not containing this minimum N-terminal fragment were nontoxic. Appearance of NPTII activity was not dependent upon the presence of insecticidal activity; however, the conformation of the Bt2 polypeptide appeared to exert an important influence on the enzymatic activity of the fused NPTII protein. This study did suggest that the global 3-D structure of the Bt2 polypeptide is disturbed in truncated polypeptides.
A number of researchers have attempted to express plant genes in yeast (Neill et al. (1987) Gene 55:303-317; Rothstein et al. (1987) Gene 55:353-356; Coraggio et al. (1986) EMBO J. 5:459-465) and E. coli (Fuzakawa et al. (1987) FEBS Lett. 224:125-127; Vies et al. (1986) EMBO J. 5:2439-2444; Gatenby et al. (1987) Eur. J. Biochem. 168:227-231). In the case of wheat .alpha.-gliadin (Neill et al. (1987) supra), .alpha.-amylase (Rothstein et al. (1987) supra) genes, and maize zein genes (Coraggio et al. (1986) supra) in yeast, low levels of expression have been reported. Neill et al. have suggested that the low levels of expression of .alpha.-gliadin in yeast may be due in part to codon usage bias, since .alpha.-gliadin codons for Phe, Leu, Ser, Gly, Tyr and especially Glu do not correlate well with the abundant yeast isoacceptor tRNAs. In E. coli however, soybean glycinin A2 (Fuzakawa et al. (1987) supra) and wheat RuBPC SSU (Vies et al. (1986) supra; Gatenby et al. (1987) supra) are expressed adequately.
Not much is known about the makeup of tRNA populations in plants. Viotti et al. (1978) Biochim. Biophys. Acta 517:125-132 report that maize endosperm actively synthesizing zein, a storage protein rich in glutamine, leucine, and alanine, is characterized by higher levels of accepting activity for these three amino acids than are maize embryo tRNAs. This may indicate that the tRNA population of specific plant tissues may be adapted for optimum translation of highly expressed proteins such as zein. To our knowledge, no one has experimentally altered codon bias in highly expressed plant genes to determine possible effects of the protein translation in plants to check the effects on the level of expression.
SUMMARY OF THE INVENTION
It is the overall object of the present invention to provide a means for plant protection against insect damage. The invention disclosed herein comprises a chemically synthesized gene encoding an insecticidal protein which is functionally equivalent to a native insecticidal protein of Bt. This synthetic gene is designed to be expressed in plants at a level higher than a native Bt gene. It is preferred that the synthetic gene be designed to be highly expressed in plants as defined herein. Preferably, the synthetic gene is at least approximately 85% homologous to an insecticidal protein gene of Bt.
It is a particular object of this invention to provide a synthetic structural gene coding for an insecticidal protein from Btt having, for example, the nucleotide sequences presented in FIG. 1 and spanning nucleotides 1 through 1793 or spanning nucleotide 1 through 1833 with functional equivalence.
In designing synthetic Btt genes of this invention for enhanced expression in plants, the DNA sequence of the native Btt structural gene is modified in order to contain codons preferred by highly expressed plant genes, to attain an A+T content in nucleotide base composition substantially that found in plants, and also preferably to form a plant initiation sequence, and to eliminate sequences that cause destabilization, inappropriate polyadenylation, degradation and termination of RNA and to avoid sequences that constitute secondary structure hairpins and RNA splice sites. In the synthetic genes, codons used to specify a given amino acid are selected with regard to the distribution frequency of codon usage employed in highly expressed plant genes to specify that amino acid. As is appreciated by those skilled in the art, the distribution frequency of codon usage utilized in the synthetic gene is a determinant of the level of expression. Hence, the synthetic gene is designed such that its distribution frequency of codon usage deviates, preferably, no more than 25% from that of highly expressed plant genes and, more preferably, no more than about 10%. In addition, consideration is given to the percentage G+C content of the degenerate third base (monocotyledons appear to favor G+C in this position, whereas dicotyledons do not). It is also recognized that the XCG nucleotide is the least preferred codon in dicots whereas the XTA codon is avoided in both monocots and dicots. The synthetic genes of this invention also preferably have CG and TA doublet avoidance indices as defined in the Detailed Description closely approximating those of the chosen host plant. More preferably these indices deviate from that of the host by no more than about 10-15%.
Assembly of the Bt gene of this invention is performed using standard technology known to the art. The Btt structural gene designed for enhanced expression in plants of the specific embodiment is enzymatically assembled within a DNA vector from chemically synthesized oligonucleotide duplex segments. The synthetic Bt gene is then introduced into a plant host cell and expressed by means known to the art. The insecticidal protein produced upon expression of the synthetic Bt gene in plants is functionally equivalent to a native Bt crystal protein in having toxicity to the same insects.

BRIEF DESCRIPTION OF THE FIGURES
FIG. 1 presents the nucleotide sequence for the synthetic Btt gene. Where different, the native sequence as found in p544Pst-Met5 is shown above. Changes in amino acids (underlined) occur in the synthetic sequence with alanine replacing threonine at residue 2 and leucine replacing the stop at residue 596 followed by the addition of 13-amino acids at the C-terminus.
FIG. 2 represents a simplified scheme used in the construction of the synthetic Btt gene. Segments A through M represent oligonucleotide pieces annealed and ligated together to form DNA duplexes having unique splice sites to allow specific enzymatic assembly of the DNA segments to give the desired gene.
FIG. 3 is a schematic diagram showing the assembly of oligonucleotide segments in the construction of a synthetic Btt gene. Each segment (A through M) is built from oligonucleotides of different sizes, annealed and ligated to form the desired DNA segment.

DETAILED DESCRIPTION OF THE INVENTION
The following definitions are provided in order to provide clarity as to the intent or scope of their usage in the Specification and claims.
Expression refers to the transcription and translation of a structural gene to yield the encoded protein. The synthetic Bt genes of the present invention are designed to be expressed at a higher level in plants than the corresponding native Bt genes. As will be appreciated by those skilled in the art, structural gene expression levels are affected by the regulatory DNA sequences (promoter, polyadenylation sites, enhancers, etc.) employed and by the host cell in which the structural gene is expressed. Comparisons of synthetic Bt gene expression and native Bt gene expression must be made employing analogous regulatory sequences and in the same host cell. It will also be apparent that analogous means of assessing gene expression must be employed in such comparisons.
Promoter refers to the nucleotide sequences at the 5' end of a structural gene which direct the initiation of transcription. Promoter sequences are necessary, but not always sufficient, to drive the expression of a downstream gene. In prokaryotes, the promoter drives transcription by providing binding sites to RNA polymerases and other initiation and activation factors. Usually promoters drive transcription preferentially in the downstream direction, although promotional activity can be demonstrated (at a reduced level of expression) when the gone is placed upstream of the promoter. The level of transcription is regulated by promoter sequences. Thus, in the construction of heterologous promoter/structural gene combinations, the structural gene is placed under the regulatory control of a promoter such that the expression of the gone is controlled by promoter sequences. The promoter is positioned preferentially upstream to the structural gene and at a distance from the transcription start site that approximates the distance between the promoter and the gene it controls in its natural setting. As is known in the art, some variation in this distance can be tolerated without loss of promoter function.
A gene refers to the entire DNA portion involved in the synthesis of a protein. A gene embodies the structural or coding portion which begins at the 5' end from the translational start codon (usually ATG) and extends to the stop (TAG, TGA or TAA) codon at the 3' end. It also contains a promoter region, usually located 5' or upstream to the structural gene, which initiates and regulates the expression of a structural gene. Also included in a gene are the 3' end and poly(A)+ addition sequences.
Structural gene is that portion of a gene comprising a DNA segment encoding a protein, polypeptide or a portion thereof, and excluding the 5' sequence which drives the initiation of transcription. The structural gene may be one which is normally found in the cell or one which is not normally found in the cellular location wherein it is introduced, in which case it is termed a heterologous gene. A heterologous gene may be derived in whole or in part from any source know to the art, including a bacterial genome or episome, eukaryotic, nuclear or plasmid DNA, cDNA, viral DNA or chemically synthesized DNA. A structural gene may contain one or more modifications in either the coding or the untranslated regions which could affect the biological activity or the chemical structure of the expression product, the rate of expression or the manner of expression control. Such modifications include, but are not limited to, mutations, insertions, deletions and substitutions of one or more nucleotides. The structural gene may constitute an uninterrupted coding sequence or it may include one or more introns, bounded by the appropriate splice junctions. The structural gene may be a composite of segments derived from a plurality of sources, naturally occurring or synthetic. The structural gene may also encode a fusion protein.
Synthetic gene refers to a DNA sequence of a structural gene that is chemically synthesized in its entirety or for the greater part of the coding region. As exemplified herein, oligonucleotide building blocks are synthesized using procedures known to those skilled in the art and are ligated and annealed to form gene segments which are then enzymatically assembled to construct the entire gene. As is recognized by those skilled in the art, functionally and structurally equivalent genes to the synthetic genes described herein may be prepared by site-specific mutagenesis or other related methods used in the art.
Transforming refers to stably introducing a DNA segment carrying a functional gene into an organism that did not previously contain that gene.
Plant tissue includes differentiated and undifferentiated tissues of plants, including but not limited to, roots, shoots, leaves, pollen, seeds, tumor tissue and various forms of cells in culture, such as single cells, protoplasts, embryos and callus tissue. The plant tissue may be in planta or in organ, tissue or cell culture.
Plant cell as used herein includes plant cells in planta and plant cells and protoplasts in culture.
Homology refers to identity or near identity of nucleotide or amino acid sequences. As is understood in the art, nucleotide mismatches can occur at the third or wobble base in the codon without causing amino acid substitutions in the final polypeptide sequence. Also, minor nucleotide modifications (e.g., substitutions, insertions or deletions) in certain regions of the gene sequence can be tolerated and considered insignificant whenever such modifications result in changes in amino acid sequence that do not alter functionality of the final product. It has been shown that chemically synthesized copies of whole, or parts of, gene sequences can replace the corresponding regions in the natural gene without loss of gene function. Homologs of specific DNA sequences may be identified by those skilled in the art using the test of cross-hybridization of nucleic acids under conditions of stringency as is well understood in the art (as described in Hames and Higgens (eds.) (1985) Nucleic Acid Hybridization, IRL Press, Oxford, UK). Extent of homology is often measured in terms of percentage of identity between the sequences compared.
Functionally equivalent refers to identity or near identity of function. A synthetic gene product which is toxic to at least one of the same insect species as a natural Bt protein is considered functionally equivalent thereto. As exemplified herein, both natural and synthetic Btt genes encode 65 kDa> insecticidal proteins having essentially identical amino acid sequences and having toxicity to coleopteran insects. The synthetic Bt genes of the present invention are not considered to be functionally equivalent to native Bt genes, since they are expressible at a higher level in plants than native Bt genes.
Frequency of preferred codon usage refers to the preference exhibited by a specific host cell in usage of nucleotide codons to specify a given amino acid. To determine the frequency of usage of a particular codon in a gene, the number of occurrences of that codon in the gene is divided by the total number of occurrences of all codons specifying the same amino acid in the gene. Table 1, for example, gives the frequency of codon usage for Bt genes, which was obtained by analysis of four Bt genes whose sequences are publicly available. Similarly, the frequency of preferred codon usage exhibited by a host cell can be calculated by averaging frequency of preferred codon usage in a large number of genes expressed by the host cell. It is preferable that this analysis be limited to genes that are highly expressed by the host cell. Table 1, for example, gives the frequency of codon usage by highly expressed genes exhibited by dicotyledonous plants, and monocotyledonous plants. The dicot codon usage was calculated using 154 highly expressed coding sequences obtained from Genbank which are listed in Table 1. Monocot codon usage was calculated using 53 monocot nuclear gene coding sequences obtained from Genbank and listed in Table 1, located in Example 1.
When synthesizing a gene for improved expression in a host cell it is desirable to design the gene such that its frequency of codon usage approaches the frequency of preferred codon usage of the host cell.
The percent deviation of the frequency of preferred codon usage for a synthetic gene from that employed by a host cell is calculated first by determining the percent deviation of the frequency of usage of a single codon from that of the host cell followed by obtaining the average deviation over all codons. As defined herein this calculation includes unique codons (i.e., ATG and TGG). The frequency of preferred codon usage of the synthetic Btt gene, whose sequence is given in FIG. 1, is given in Table 1. The frequency of preferred usage of the codon `GTA` for valine in the synthetic gene (0.10) deviates from that preferred by dicots (0.12) by 0.02/0.12=0.167 or 16.7%. The average deviation over all amino acid codons of the Btt synthetic gene codon usage from that of dicot plants is 7.8%. In general terms the overall average deviation of the codon usage of a synthetic gene from that of a host cell is calculated using the equation ##EQU1## where X.sub.n =frequency of usage for codon n in the host cell; Y.sub.n =frequency of usage for codon n in the synthetic gene. Where n represents an individual codon that specifies an amino acid, the total number of codons is Z, which in the preferred embodiment is 61. The overall deviation of the frequency of codon usage, A, for all amino acids should preferably be less than about 25%, and more preferably less than about 10%.
Derived from is used to mean taken, obtained, received, traced, replicated or descended from a source (chemical and/or biological). A derivative may be produced by chemical or biological manipulation (including but not limited to substitution, addition, insertion, deletion, extraction, isolation, mutation and replication) of the original source.
Chemically synthesized, as related to a sequence of DNA, means that the component nucleotides were assembled an vitro. Manual chemical synthesis of DNA may be accomplished using well established procedures (Caruthers, M. (1983) in Methodology of DNA and RNA Sequencing, Weissman (ed.), Praeger Publishers, New York, Chapter 1), or automated chemical synthesis can be performed using one of a number of commercially available machines.
The term, designed to be highly expressed as used herein refers to a level of expression of a designed gene wherein the amount of its specific mRNA transcripts produced is sufficient to be quantified in Northern blots and, thus, represents a level of specific mRNA expressed corresponding to greater than or equal to approximately 0.001% of the poly(A)+ mRNA. To date, natural Bt genes are transcribed at a level wherein the amount of specific mRNA produced is insufficient to be estimated using the Northern blot technique. However, in the present invention, transcription of a synthetic Bt gene designed to be highly expressed not only allows quantification of the specific mRNA transcripts produced but also results in enhanced expression of the translation product which is measured in insecticidal bioassays.
Crystal protein or insecticidal crystal protein or crystal toxin refers to the major protein component of the parasporal crystals formed in strains of Bt. This protein component exhibits selective pathogenicity to different species of insects. The molecular size of the major protein isolated from parasporal crystals varies depending on the strain of Bt from which it is derived. Crystal proteins having molecular weights of approximately 132, 65, and 28 kDa have been reported. It has been shown that the approximately 132 kDa protein is a protoxin that is cleaved to form an approximately 65 kDa toxin.
The crystal protein gene refers to the DNA sequence encoding the insecticidal crystal protein in either full length protoxin or toxin form, depending on the strain of Bt from which the gene is derived.
The authors of this invention observed that expression in plants of Bt crystal protein mRNA occurs at levels that are not routinely detectable in Northern blots and that low levels of Bt crystal protein expression correspond to this low level of mRNA expression. It is preferred for exploitation of these genes as potential biocontrol methods that the level of expression of Bt genes in plant cells be improved and that the stability of Bt mRNA in plants be optimized. This will allow greater levels of Bt mRNA to accumulate and will result in an increase in the amount of insecticidal protein in plant tissues. This is essential for the control of insects that are relatively resistant to Bt protein.
Thus, this invention is based on the recognition that expression levels of desired, recombinant insecticidal protein in transgenic plants can be improved via increased expression of stabilized mRNA transcripts; and that, conversely, detection of these stabilized RNA transcripts may be utilized to measure expression of translational product (protein). This invention provides a means of resolving the problem of low expression of insecticidal protein RNA in plants and, therefore, of low protein expression through the use of an improved, synthetic gene specifying an insecticidal crystal protein from Bt.
Attempts to improve the levels of expression of Bt genes in plants have centered on comparative studies evaluating parameters such as gene type, gene length, choice of promoters, addition of plant viral untranslated RNA leader, addition of intron sequence and modification of nucleotides surrounding the initiation ATG codon. To date, changes in these parameters have not led to significant enhancement of Bt protein expression in plants. Applicants find that, surprisingly, to express Bt proteins at the desired level in plants, modifications in the coding region of the gene were effective. Structural-function relationships can be studied using site-specific mutagenesis by replacement of restriction fragments with synthetic DNA duplexes containing the desired nucleotide changes (Lo et al. (1984) Proc. Natl. Acad. Sci. 81:2285-2289). However, recent advances in recombinant DNA technology now make it feasible to chemically synthesize an entire gene designed specifically for a desired function. Thus, the Btt coding region was chemically synthesized, modified in such a way as to improve its expression in plants. Also, gene synthesis provides the opportunity to design the gene so as to facilitate its subsequent mutagenesis by incorporating a number of appropriately positioned restriction endonuclease sites into the gene.
The present invention provides a synthetic Bt gene for a crystal protein toxic to an insect. As exemplified herein, this protein is toxic to coleopteran insects. To the end of improving expression of this insecticidal protein in plants, this invention provides a DNA segment homologous to a Btt structural gene and, as exemplified herein, having approximately 85% homology to the Btt structural gene in p544Pst-Met5. In this embodiment the structural gene encoding a Btt insecticidal protein is obtained through chemical synthesis of the coding region. A chemically synthesized gene is used in this embodiment because it best allows for easy and efficacious accommodation of modifications in nucleotide sequences required to achieve improved levels of cross-expression.
Today, in general, chemical synthesis is a preferred method to obtain a desired modified gene. However, to date, no plant protein gene has been chemically synthesized nor has any synthetic gene for a bacterial protein been expressed in plants. In this invention, the approach adopted for synthesizing the gene consists of designing an improved nucleotide sequence for the coding region and assembling the gene from chemically synthesized oligonucleotide segments. In designing the gene, the coding region of the naturally-occurring gene, preferably from the Btt subclone, p544Pst-Met5, encoding a 65 kDa polypeptide having coleoperan toxicity, is scanned for possible modifications which would result in improved expression of the synthetic gene in plants. For example, to optimize the efficiency of translation, codons preferred in highly expressed proteins of the host cell are utilized.
Bias in codon choice within genes in a single species appears related to the level of expression of the protein encoded by that gene. Codon bias is most extreme in highly expressed proteins of E. coli and yeast. In these organisms, a strong positive correlation has been reported between the abundance of an isoaccepting tRNA species and the favored synonymous codon. In one group of highly expressed proteins in yeast, over 96% of the amino acids are encoded by only 25 of the 61 available codons (Bennetzen and Hall (1982) J. Biol. Chem. 257:3026-3031). These 25 codons are preferred in all sequenced yeast genes, but the degree of preference varies with the level of expression of the genes. Recently, Hoekema and colleagues (1987) Mol. Cell. Biol. 7:2914-2924 reported that replacement of these 25 preferred codons by minor codons in the 5' end of the highly expressed yeast gene PGK1 results in a decreased level of both protein and mRNA. They concluded that biased codon choice in highly expressed genes enhances translation and is required for maintaining mRNA stability in yeast. Without doubt, the degree of codon bias is an important factor to consider when engineering high expression of heterologous genes in yeast and other systems.
Experimental evidence obtained from point mutations and deletion analysis has indicated that in eukaryotic genes specific sequences are associated with post-transcriptional processing, RNA destabilization, translational termination, intron splicing and the like. These are preferably employed in the synthetic genes of this invention. In designing a bacterial gene for expression in plants, sequences which interfere with the efficacy of gene expression are eliminated.
In designing a synthetic gene, modifications in nucleotide sequence of the coding region are made to modify the A+T content in DNA base composition of the synthetic gene to reflect that normally found in genes for highly expressed proteins native to the host cell. Preferably the A+T content of the synthetic gene is substantially equal to that of said genes for highly expressed proteins. In genes encoding highly expressed plant proteins, the A+T content is approximately 55%. It is preferred that the synthetic gene have an A+T content near this value, and not sufficiently high as to cause destabilization of RNA and, therefore, lower the protein expression levels. More preferably, the A+T content is no more than about 60% and most preferably is about 55%. Also, for ultimate express ion in plants, the synthetic gene nucleotide sequence is preferably modified to form a plant initiation sequence at the 5' end of the coding region. In addition, particular attention is preferably given to assure that unique restriction sites are placed in strategic positions to allow efficient assembly of oligonucleotide segments during construction of the synthetic gene and to facilitate subsequent nucleotide modification. As a result of these modifications in coding region of the native Bt gene, the preferred synthetic gene is expressed in plants at an enhanced level when compared to that observed with natural Bt structural genes.
In specific embodiments, the synthetic Bt gene of this invention encodes a Btt protein toxic to coleopteran insects. Preferably, the toxic polypeptide is about 598 amino acids in length, is at least 75% homologous to a Btt polypeptide, and, as exemplified herein, is essentially identical to the protein encoded by p544Pst-Met5, except for replacement of threonine by alanine at residue 2. This amino acid substitution results as a consequence of the necessity to introduce a guanine base at position +4 in the coding sequence.
In designing the synthetic gene of this invention, the coding region from the Btt subclone, p544Pst-Met5, encoding a 65 kDa polypeptide having coleopteran toxicity, is scanned for possible modifications which would result in improved expression of the synthetic gene in plants. For example, in preferred embodiments, the synthetic insecticidal protein is strongly expressed in dicot plants, e.g., tobacco, tomato, cotton, etc., and hence, a synthetic gene under these conditions is designed to incorporate to advantage codons used preferentially by highly expressed dicot proteins. In embodiments where enhanced expression of insecticidal protein is desired in a monocot, codons preferred by highly expressed monocot proteins (given in Table 1) are employed in designing the synthetic gene.
In general, genes within a taxonomic group exhibit similarities in codon choice, regardless of the function of these genes. Thus an estimate of the overall use of the genetic code by a taxonomic group can be obtained by summing codon frequencies of all its sequenced genes. This species-specific codon choice is reported in this invention from analysis of 208 plant genes. Both monocot and dicot plants are analyzed individually to determine whether these broader taxonomic groups are characterized by different patterns of synonymous codon preference. The 208 plant genes included in the codon analysis code for proteins having a wide range of functions and they represent 6 monocot and 36 dicot species. These proteins are present in different plant tissues at varying levels of expression.
In this invention it is shown that the relative use of synonymous codons differs between the monocots and the dicots. In general, the most important factor in discriminating between monocot and dicot patterns of codon usage is the percentage G+C content of the degenerate third base. In monocots, 16 of 18 amino acids favor G+C in this position, while dicots only favor G+C in 7 of 18 amino acids.
The G ending codons for Thr, Pro, Ala and Ser are avoided in both monocots and dicots because they contain C in codon position II. The CG dinucleotide is strongly avoided in plants (Boudraa (1987) Genet. Sel. Evol. 19:143-154) and other eukaryotes (Grantham et al. (1985) Bull. Inst. Pasteur 83:95-148), possibly due to regulation involving methylation. In dicots, XCG is always the least favored codon, while in monocots this is not the case. The doublet TA is also avoided in codon positions II and III in most eukaryotes, and this is true of both monocots and dicots.
Grantham and colleagues (1986) Oxford Surveys in Evol. Biol. 3:48-81 have developed two codon choice indices to quantify CG and TA doublet avoidance in codon positions II and III. XCG/XCC is the ratio of codons having C as base II of G-ending to C-ending triplets, while XTA/XTT is the ratio of A-ending to T-ending triplets with T as the second base. These indices have been calculated for the plant data in this paper (Table 2) and support the conclusion that monocot and dicot species differ in their use of these dinucleotides.
TABLE 2______________________________________Avoidance of CG and TA doublets in codons position II-III.XCG/XCC and XTA/XAA values are multiplied by 100. Mono- Soy- RuBPCGroup Plants Dicots cots Maize bean SSU CAB______________________________________XCG/ 40 30 61 67 37 18 22XCCXTA/ 37 35 47 43 41 9 13XTT______________________________________ RuBPC SSU = ribulose 1,5 bisphosphate small subunit CAB = chlorophyll a/b binding protein
Additionally, for two species, soybean and maize, species-specific codon usage profiles were calculated (not shown). The maize codon usage pattern resembles that of monocots in general, since these sequences represent over half of the monocot sequences available. The codon profile of the maize subsample is even more strikingly biased in its preference for G+C in codon position III. On the other hand, the soybean codon usage pattern is almost identical to the general dicot pattern, even though it represents a much smaller portion of the entire dicot sample.
In order to determine whether the coding strategy of highly expressed genes such as the ribulose 1,5 bisphosphate small subunit (RuBPC SSU) and chlorophyll a/b binding protein (CAB) is more biased than that of plant genes in general, codon usage profiles for subsets of these genes (19 and 17 sequences, respectively) were calculated (not shown). The RuBPC SSU and CAB pooled samples are characterized by stronger avoidance of the codons XCG and XTA than in the larger monocot and dicot samples (Table 2). Although most of the genes in these subsamples are dicot in origin (17/19 and 15/17), their codon profile resembles that of the monocots in that G+C is utilized in the degenerate base III.
The use of pooled data for highly expressed genes may obscure identification of species-specific patterns in codon choice. Therefore, the codon choices of individual genes for RuBPC SSU and CAB were tabulated. The preferred codons of the maize and wheat genes for RuBPC SSU and CAB are more restricted in general than are those of the dicot species. This is in agreement with Matsuoka et al. (1987) J. Biochem. 102:673-676) who noted the extreme codon bias of the maize RuBPC SSU gene as well as two other highly expressed genes in maize leaves, CAB and phosphoenolpyruvate carboxylase. These genes almost completely avoid the use of A+T in codon position III, although this codon bias was not as pronounced in non-leaf proteins such as alcohol dehydrogenase, zein 22 kDa subunit, sucrose synthetase and ATP/ADP translocator. Since the wheat SSU and CAB genes have a similar pattern of codon preference, this may reflect a common monocot pattern for these highly expressed genes in leaves. The CAB gene for Lemna and the RuBPC SSU genes for Chlamdomonas share a similar extreme preference for G+C in codon position III. In dicot CAB genes, however, A+T degenerate bases are preferred by some synonymous codons (e.g., GCT for Ala, CTT for Leu, GGA and GGT for Gly). In general, the G+C preference is less pronounced for both RuBPC SSU and CAB genes in dicots than in monocots.
In designing a synthetic gene for expression in plants, attempts are also made to eliminate sequences which interfere with the efficacy of gene expression. Sequences such as the plant polyadenylation signals, e.g., AATAAA, polymerase II termination sequence, e.g., CAN.sub.(7-9) AGTNNAA, UCUUCGG hairpins and plant consensus splice sites are highlighted and, if present in the native Btt coding sequence, are modified so as to eliminate potentially deleterious sequences.
Modifications in nucleotide sequence of the Btt coding region are also preferably made to reduce the A+T content in DNA base composition. The Btt coding region has an A+T content of 64%, which is about 10% higher than that found in a typical plant coding region. Since A+T-rich regions typify plant intergenic regions and plant regulatory regions, it is deemed prudent to reduce the A+T content. The synthetic Btt gene is designed to have an A+T content of 55%, in keeping with values usually found in plants.
Also, a single modification (to introduce guanine in lieu of adenine) at the fourth nucleotide position in the Btt coding sequence is made in the preferred embodiment to form a sequence consonant with that believed to function as a plant initiation sequence (Taylor et al. (1987) Mol. Gen. Genet. 210:572-577) in optimization of expression. In addition, in exemplifying this invention thirty-nine nucleotides (thirteen codons) are added to the coding region of the synthetic gene in an attempt to stabilize primary transcripts. However, it appears that equally stable transcripts are obtained in the absence of this extension polypeptide containing thirty-nine nucleotides.
Not all of the above-mentioned modifications of the natural Bt gene must be made in constructing a synthetic Bt gene in order to obtain enhanced expression. For example, a synthetic gene may be synthesized for other purposes in addition to that of achieving enhanced levels of expression. Under these conditions, the original sequence of the natural Bt gene may be preserved within a region of DNA corresponding to one or more, but not all, segments used to construct the synthetic gene. Depending on the desired purpose of the gene, modification may encompass substitution of one or more, but not all, of the oligonucleotide segments used to construct the synthetic gene by a corresponding region of natural Bt sequence.
As is known to those skilled in the art of synthesizing genes (Mandecki et al. (1985) Proc. Natl. Acad. Sci. 82:3543-3547; Feretti et al. (1986) Proc. Natl. Acad. Sci. 83:599-603), the DNA sequence to be synthesized is divided into segment lengths which can be synthesized conveniently and without undue complication. As exemplified herein, in preparing to synthesize the Btt gene, the coding region is divided into thirteen segments (A-M). Each segment has unique restriction sequences at the cohesive ends. Segment A, for example, is 228 base pairs in length and is constructed from six oligonucleotide sections, each containing approximately 75 bases. Single-stranded oligonucleotides are annealed and ligated to form DNA segments. The length of the protruding cohesive ends in complementary oligonucleotide segments is four to five residues. In the strategy evolved for gene synthesis, the sites designed for the joining of oligonucleotide pieces and DNA segments are different from the restriction sites created in the gene.
In the specific embodiment, each DNA segment is cloned into a pIC-20 vector for amplification of the DNA. The nucleotide sequence of each fragment is determined at this stage by the dideoxy method using the recombinant phage DNA as templates and selected synthetic oligonucleotides as primers.
As exemplified herein and illustrated schematically in FIGS. 3 and 4, each segment individually (e.g., segment M) is excised at the flanking restriction sites from its cloning vector and spliced into the vector containing segment A. Most often, segments are added as a paired segment instead of as a single segment to increase efficiency. Thus, the entire gene is constructed in the original plasmid harboring segment A. The nucleotide sequence of the entire gene is determined and found to correspond exactly to that shown in FIG. 1.
In preferred embodiments the synthetic Btt gene is expressed in plants at an enhanced level when compared to that observed with natural Btt structural genes. To that end, the synthetic structural gene is combined with a promoter functional in plants, the structural gene and the promoter region being in such position and orientation with respect to each other that the structural gene can be expressed in a cell in which the promoter region is active, thereby forming a functional gene. The promoter regions include, but are not limited to, bacterial and plant promoter regions. To express the promoter region/structural gene combination, the DNA segment carrying the combination is contained by a cell. Combinations which include plant promoter regions are contained by plant cells, which, in turn, may be contained by plants or seeds. Combinations which include bacterial promoter regions are contained by bacteria, e.g., Bt or E. coli. Those in the art will recognize that expression in types of micro-organisms other than bacteria may in some circumstances be desirable and, given the present disclosure, feasible without undue experimentation.
The recombinant DNA molecule carrying a synthetic structural gene under promoter control can be introduced into plant tissue by any means known to those skilled in the art. The technique used for a given plant species or specific type of plant tissue depends on the known successful techniques. As novel means are developed for the stable insertion of foreign genes into plant cells and for manipulating the modified cells, skilled artisans will be able to select from known means to achieve a desired result. Means for introducing recombinant DNA into plant tissue include, but are not limited to, direct DNA uptake (Paszkowski, J. et al. (1984) EMBO J. 3:2717), electroporation (Fromm, M. et al. (1985) Proc. Natl. Acad. Sci. USA 82:5824), microinjection (Crossway, A. et al. (1986) Mol. Gen. Genet. 202:179), or T-DNA mediated transfer from Agrobacterium tumefaciens to the plant tissue. There appears to be no fundamental limitation of T-DNA transformation to the natural host range of Agrobacterium. Successful T-DNA-mediated transformation of monocots (Hooykaas-Van Slogteren, G. et al. (1984) Nature 311:763), gymnosperm (Dandekar, A. et al. (1987) Biotechnology 5:587) and algae (Ausich, R., EPO application 108,580) has been reported. Representative T-DNA vector systems are described in the following references: An, G. et al. (1985) EMBO J. 4:277; Herrera-Estrella, L. et al. (1983) Nature 303:209; Herrera-Estrella, L. et al. (1983) EMBO J. 2:987; Herrera-Estrella, L. et al. (1985) in Plant Genetic Engineering, New York: Cambridge University Press, p. 63. Once introduced into the plant tissue, the expression of the structural gene may be assayed by any means known to the art, and expression may be measured as mRNA transcribed or as protein synthesized. Techniques are known for the in vitro culture of plant tissue, and in a number of cases, for regeneration into whole plants. Procedures for transferring the introduced expression complex to commercially useful cultivars are known to those skilled in the art.
In one of its preferred embodiments the invention disclosed herein comprises expression in plant cells of a synthetic insecticidal structural gene under control of a plant expressible promoter, that is to say, by inserting the insecticide structural gene into T-DNA under control of a plant expressible promoter and introducing the T-DNA containing the insert into a plant cell using known means. Once plant cells expressing a synthetic insecticidal structural gene under control of a plant expressible promoter are obtained, plant tissues and whole plants can be regenerated therefrom using methods and techniques well-known in the art. The regenerated plants are then reproduced by conventional means and the introduced genes can be transferred to other strains and cultivars by conventional plant breeding techniques.
The introduction and expression of the synthetic structural gene for an insecticidal protein can be used to protect a crop from infestation with common insect pests. Other uses of the invention, exploiting the properties of other insecticide structural genes introduced into other plant species will be readily apparent to those skilled in the art. The invention in principle applies to introduction of any synthetic insecticide structural gene into any plant species into which foreign DNA (in the preferred embodiment T-DNA) can be introduced and in which said DNA can remain stably replicated. In general, these taxa presently include, but are not limited to, gymnosperms and dicotyledonous plants, such as sunflower (family Compositeae), tobacco (family Solanaceae), alfalfa, soybeans and other legumes (family Leguminoseae), cotton (family Malvaceae), and most vegetables, as well as monocotyledonous plants. A plant containing in its tissues increased levels of insecticidal protein will control less susceptible types of insect, thus providing advantage over present insecticidal uses of Bt. By incorporation of the insecticidal protein into the tissues of a plant, the present invention additionally provides advantage over present uses of insecticides by eliminating instances of nonuniform application and the costs of buying and applying insecticidal preparations to a field. Also, the present invention eliminates the need for careful timing of application of such preparations since small larvae are most sensitive to insecticidal protein and the protein is always present, minimizing crop damage that would otherwise result from preapplication larval foraging.
This invention combines the specific teachings of the present disclosure with a variety of techniques and expedients known in the art. The choice of expedients depends on variables such as the choice of insecticidal protein from a Bt strain, the extent of modification in preferred codon usage, manipulation of sequences considered to be destabilizing to RNA or sequences prematurely terminating transcription, insertions of restriction sites within the design of the synthetic gene to allow future nucleotide modifications, addition of introns or enhancer sequences to the 5' and/or 3' ends of the synthetic structural gene, the promoter region, the host in which a promoter region/structural gene combination is expressed, and the like. As novel insecticidal proteins and toxic polypeptides are discovered, and as sequences responsible for enhanced cross-expression (expression of a foreign structural gene in a given host) are elucidated, those of ordinary skill will be able to select among those elements to produce "improved" synthetic genes for desired proteins having agronomic value. The fundamental aspect of the present invention is the ability to synthesize a novel gene coding for an insecticidal protein, designed so that the protein will be expressed at an enhanced level in plants, yet so that it will retain its inherent property of insect toxicity and retain or increase its specific insecticidal activity.
EXAMPLES
The following Examples are presented as illustrations of embodiments of the present invention. They do not limit the scope of this invention, which is determined by the claims.
The following strains were deposited with the Patent Culture Collection, Northern Regional Research Center, 1815 N. University Street, Peoria, Ill. 61604.
______________________________________Strain Deposited on Accession #______________________________________E. coli MC1061 (p544-HindIII) 6 October 1987 NRRL B-18257E. coli MC1061 (p544Pst-Met5) 6 October 1987 NRRL B-18258______________________________________
The deposited strains are provided for the convenience of those in the art, and are not necessary to practice the present invention, which may be practiced with the present disclosure in combination with publicly available protocols, information, and materials. E. coli MC1061, a good host for plasmid transformations, was disclosed by Casadaban, M. J. and Cohen, S. N. (1980) J. Mol. Biol. 138:179-207.
Example 1
Design of the synthetic insecticidal crystal protein gene
(i) Preparation of toxic subclones of the Btt gene
Construction, isolation, and characterization of pNSB544 is disclosed by Sekar, V. et al. (1987) Proc. Natl. Acad. Sci. USA 84:7036-7040, and Sekar, V. and Adang, M. J., U.S. patent application Ser. No. 108,285, filed Oct. 13, 1987, which is hereby incorporated by reference. A 3.0 kbp HindIII fragment carrying the crystal protein gene of pNSBP544 is inserted into the HindIII site of pIC-2OH (Marsh, J. L. et al. (1984) Gene 32:481-485), thereby yielding a plasmid designated p544-HindIII, which is on deposit. Expression in E. coli yields a 73 kDa crystal protein in addition to the 65 kDa species characteristic of the crystal protein obtained from Btt isolates.
A 5.9 kbp BamHI fragment carrying the crystal protein gene is removed from pNSBP544 and inserted into BamHI-linearized pIC-2OH DNA. The resulting plasmid, p405/44-7, is digested with BglII and religated, thereby removing Bacillus sequences flanking the 3'-end of the crystal protein gene. The resulting plasmid, p405/54-12, is digested with pstI and religated, thereby removing Bacillus sequences flanking the 5'-end of the crystal protein and about 150 bp from the 5'-end of the crystal protein structural gene. The resulting plasmid, p405/81-4, is digested with SpBI and PstI and is mixed with and ligated to a synthetic linker having the following structure: ##STR1## (SD indicates the location of a Shine-Dalgarno prokaryotic ribosome binding site.) The resulting plasmid, p544Pst-Met5, contains a structural gene encoding a protein identical to one encoded by pNSBP544 except for a deletion of the amino-terminal 47 amino acid residues. The nucleotide sequence of the Btt coding region in p544Pst-Met5 is presented in FIG. 1. In bioassays (Sekar and Adang, U.S. patent application Ser. No. 108,285, supra), the proteins encoded by the full-length Btt gene in pNSBP544 and the N-terminal deletion derivative, p544Pst-Met5, were shown to be equally toxic. All of the plasmids mentioned above have their crystal protein genes in the same orientation as the lacZ gene of the vector.
(ii) Modification of preferred codon usage
Table 1 presents the frequency of codon usage for (A) dicot proteins, (B) Bt proteins, (C) the synthetic Btt gene, and (D) monocot proteins. Although some codons for a particular amino acid are utilized to approximately the same extent by both dicot and St proteins (e.g., the codons for serine), for the most part, the distribution of codon frequency varies significantly between dicot and Bt proteins, as illustrated in columns A and B in Table 1.
TABLE 1______________________________________Frequency of Codon Usage Distribution FractionAmino (A) Dicot (B) Bt (C) Synthetic (D) Mono-Acid Codon Genes Genes Btt Gene cot Genes______________________________________Gly GGG 0.12 0.08 0.13 0.21Gly GGA 0.38 0.53 0.37 0.17Gly GGT 0.33 0.24 0.34 0.18Gly GGC 0.16 0.16 0.16 0.43Glu GAG 0.51 0.13 0.52 0.75Glu GAA 0.49 0.87 0.48 0.25Asp GAT 0.58 0.68 0.56 0.27Asp GAC 0.42 0.32 0.44 0.73Val GTG 0.29 0.15 0.30 0.36Val GTA 0.12 0.32 0.10 0.08Val GTT 0.39 0.29 0.35 0.19Val GTC 0.20 0.24 0.25 0.37Ala GCG 0.06 0.12 0.06 0.22Ala GCA 0.25 0.50 0.24 0.16Ala GCT 0.42 0.32 0.41 0.24Ala GCC 0.27 0.06 0.29 0.38Lys AAG 0.61 0.13 0.58 0.86Lys AAA 0.39 0.87 0.42 0.14Asn AAT 0.45 0.79 0.44 0.25Asn AAC 0.55 0.21 0.56 0.75Met ATG 1.00 1.00 1.00 1.00Ile ATA 0.18 0.30 0.20 0.11Ile ATT 0.45 0.57 0.43 0.24Ile ATC 0.37 0.13 0.37 0.64Thr ACG 0.08 0.14 0.07 0.20Thr ACA 0.27 0.68 0.27 0.14Thr ACT 0.35 0.14 0.34 0.19Thr ACC 0.30 0.05 0.32 0.46Trp TGG 1.00 1.00 1.00 1.00End TGA 0.33 0.00 0.00 0.34Cys TGT 0.44 0.33 0.33 0.30Cys TGC 0.56 0.67 0.67 0.70End TAG 0.19 0.00 0.00 0.36End TAA 0.48 1.00 1.00 0.30Tyr TAT 0.43 0.81 0.43 0.21Tyr TAC 0.57 0.19 0.57 0.79Phe TTT 0.45 0.75 0.44 0.25Phe TTC 0.55 0.25 0.56 0.75Ser AGT 0.14 0.25 0.13 0.08Ser AGC 0.18 0.13 0.19 0.26Ser TCG 0.06 0.08 0.06 0.14Ser TCA 0.19 0.19 0.17 0.11Ser TCT 0.25 0.25 0.27 0.15Ser TCC 0.18 0.10 0.17 0.25Arg AGG 0.25 0.09 0.23 0.26Arg AGA 0.30 0.50 0.32 0.09Arg CGG 0.04 0.14 0.05 0.13Arg CGA 0.08 0.14 0.09 0.04Arg CGT 0.21 0.09 0.23 0.12Arg CGC 0.11 0.05 0.09 0.36Gln CAG 0.41 0.18 0.39 0.46Gln CAA 0.59 0.82 0.61 0.54His CAT 0.54 0.90 0.50 0.33His CAC 0.46 0.10 0.50 0.67Leu TTG 0.26 0.08 0.27 0.14Leu TTA 0.10 0.46 0.12 0.03Leu CTG 0.09 0.04 0.10 0.28Leu CTA 0.08 0.21 0.10 0.10Leu CTT 0.28 0.15 0.18 0.15Leu CTC 0.19 0.06 0.22 0.31Pro CCG 0.09 0.20 0.08 0.23Pro CCA 0.42 0.56 0.44 0.34Pro CCT 0.32 0.24 0.32 0.17Pro CCC 0.17 0.00 0.16 0.26______________________________________154 coding sequences of dicot nuclear genes were used tocompile the codon usage table. The pooled dicot codingsequences, obtained from Genbank (release 55) or, when noGenbank file name is specified, directly from the publishedsource, were:GENUS/SPECIES GENBANK PROTEIN REF______________________________________Antirrhinum majus AMACHS Chalcone synthetaseArabidopsis thaliana ATHADH Alcohol dehydrogenase ATHH3GA Histone 3 gene 1 ATHH3GB Histone 3 gene 2 ATHH4GA Histone 4 gene 1 ATHLHCP1 CAB ATHTUBA .alpha. tubulin 5-enolpyruvyl4hifate 1 3-phosphate synthetaseBertholletia excelsa High methionine 2 storage proteinBrassica campestris Acyl carrier protein 3Brassica napus BNANAP NapinBrassica oleacea BOLSLSGR S-locus specific glycoproteinCanavalia ensiformis CENCONA Concanavalin ACarica papaya CPAPAP PapainChlamdomonas CREC552 Preapocytochromereinhardtii CRERBCS1 RuBPC small sub- unit gene 1 CRERBCS2 RuBPC small sub- unit gene 2Cucurbita pepo CUCPHT PhytochromeCucumis sativus CUSGMS Glyoxosomal malate synthetase CUSLHCPA CAB CUSSSU RuBPC small subunitDaucus carota DAREXT Extensin DAREXTR 33 kD extensin related proteinDolichos biflorus DBILECS seed lectinFlaveria trinervia FTRBCR RuBPC small subunitGlycine max SOY7SAA 7S storage protein SOYACT1G Actin 1 SOYCIIPI CII protease inhibitor SOYGLYA1A Glycinin Ala Bx subunits SOYGLYAAB Glycinin A5A4B3 subunits SOYGLYAB Glycinin A3/b4 subunits SOYGLYR Glycinin A2B1a subunits SOYHSP175 Low M W heat shock proteins SOYLGBI Leghemoglobin SOYLEA Lectin SOYLOX Lipoxygenase 1 SOYNOD20G 20 kDa nodulin SOYNOD23G 23 kDa nodulin SOYNOD24H 24 kDa nodulin SOYNOD26B 26 kDa nodulin SOYNOD26R 26 kDa nodulin SOYNOD27R 27 kDa nodulin SOYNOD35M 35 kDa nodulin SOYNOD75 75 kDa nodulin SOYNODR1 Nodulin C51 SOYNODR2 Nodulin E27 SOYPRP1 Proline rich protein SOYRUBP RuBPC small subunit SOYURA Urease SOYHSP26A Heat shock protein 26A Nuclear-encoded 4 chloroplast heat shock protein 5 22 kDa nodulin 6 .beta.1 tubulin 6 .beta.2 tubulinGossypium hirsutum Seed .alpha. 7 globulin (vicilin) Seed .beta. 7 globulin (vicilin)Helianthus annus HNNRUBCS RuBPC small subunit 2S albumin seed 8 storage proteinIpomoea batatas Wound-induced 9 catalaseLemna gibba LGIAB19 CAB LGIR5BPC RuBPC small subunitLupinus luteus LUPLBR leghemogloblin ILycopersicon TOMBIOBR Biotin binding proteinesculentum TOMETHYBR Ethylene biosyn- thesis protein TOMPG2AR Polygalacturo- nase-2a TOMPSI Tomato photosystem I protein TOMRBCSA RuBPC small subunit TOMRBCSB RuBPC small subunit TOMRBCSC RuBPC small subunit TOMRBCSD RuBPC small subunit TOMRRD Ripening related protein TOMWIPIG Wound induced proteinase inhibitor I TOMWIPII Wound induced proteinase inhibitor II CAB 1A 10 CAB 1B 10 CAB 3C 10 CAB 4 11 CAB 5 11Medicago sativa ALFLB3R Leghemoglobin IIIMesembryanthemum RuBPC small 12 subunitcrystallinumNicotiana TOBATP21 Mitochondrialplumbaginifolia ATP synthase .beta. subunit Nitrate reductase 13 Glutamine 14 synthetaseNicotiana tabacum TOBECH Endochitinase TOBGAPA A subunit of chloroplast G3PD TOBGAPB B subunit of chloroplast G3PD TOBGAPC C subunit of chloroplast G3PD TOBPR1AR Pathogenesis related protein 1a TOBPR1CR Pathogenesis-related protein 1c TOBPRPR Pathogenesis related protein 1b TOBPXDLF Peroxidase TOBRBPCO RuBPC small subunit TOBTHAUR TMV-induced protein homologous to thaumatinPerseus americana AVOCEL CellulasePetroselinum PHOCHL Chalcone synthasehortensePetunia Sp. PETCAB13 CAB 13 PETCAB22L CAB 22L PETCAB22R CAB 22R PETCAB25 CAB 25 PETCAB37 CAB 37 PETCAB91R CAB 91R PETCHSR Chalcone synthase PETGCR1 Glycine-rich protein PETRBCS08 RuBPC small subunit PETRBCS11 RuBPC small subunit 70 kDa heat shock 15 proteinPhascolus vulgaris PHVCHM Chitinase PHVDLECA Phytohemagglutinin E PHVDLECB Phytohemagglutinin L PHVGSR1 Glutamine synthetase 1 PHVGSR2 Glutamine synthetase 2 PHVLBA Leghemoglobin PHVLECT Lectin PHVPAL Phenylalanine ammonia lyase PHVPHASAR .alpha. phaseolin PHVPHASBR .beta. phaseolin Arcelin seed protein 16 Chalcone synthase 17Pisum sativum PEAALB2 Seed albumin PEACAB80 CAB PEAGSR1 Glutamine syn- thetase (nodule) PEALECA Lectin PEALEGA Legumin PEARUBPS RuBPC small subunit PEAVIC2 Vicilin PEAVIC4 Vicilin PEAVIC7 Vicilin Alcohol 18 dehydrogenase 1 Glutamine synthe- 19 tase (leaf) Glutamine synthe- 19 tase (root) Histone 1 20 Nuclear encoded 4 chloroplast heat shock proteinRaphanus sativus RuBPC small 21 subunitRicinus communis RCCAGG Agglutinin RCCRICIN Ricin RCCICL4 Isocitrate lyaseSilene pratensis SIPFDX Ferrodoxin precursor SIPPCY Plastocyanin precursorSinapis alba SALGAPDH Nuclear gene for G3PDSolanum tuberosun: POTPAT Patatin POTINHWI Wound-induced proteinase inhibitor POTLS1G Light-inducible tissue specific ST-LS1 gene POTP12G Wound-induced proteinase inhibitor II POTRBCS RuBPC small subunit Sucrose synthetase 22Spinacia oleracea SPIACPI Acyl carrier protein I SPIOEC16 16 kDa photosynthetic oxygen-evolving protein SPIOEC23 23 kDa photo- synthetic oxygen- evolving protein SPIPCG Plastocyanin SPIPS33 33 kDa photo- synthetic water oxidation complex precursor Glycolate oxidase 23Vicia faba VFALBA Legthemoglobin VFALEB4 Legumin B 24 VicillinPooled 53 monocot coding sequences obtained from Genbank(release 55) or, when no Genbank file name is specified,directly from the published source, were:Avena sativa ASTAP3R Phytochrome 3Hordeum vulgare BLYALR Aleurain BLYAMY1 .alpha. amylase 1 BLYAMY2 .alpha. amylase 2 BLYCHORD1 Hordein C BLYGLUCB .beta. glucanase BLYHORB .beta.1 hordein BLYPAPI Amylase/protease inhibitor BLYTH1AR Toxin .alpha. hordothionin BLYUBIQR Ubiquitin Histone 3 25 Leaf specific 26 thionin 1 Leaf specific 26 thionin 2 Plastocyanin 27Oryza sativa RICGLUTG Glutelin 28 GlutelinTriticum aestivum WHTAMYA .alpha. amylase WHTCAB CAB WHTEMR Em protein WHTGIR gibberellin re- sponsive protein WHTGLGB .gamma. gliadin WHTGLIABA .alpha./.beta. gliadin Class All WHTGLUT1 High MW glutenin WHTH3 Histone 3 WHTH4091 Histone 4 WHTRBCB RuBPC small subunitSecale cereale RYESECGSR .gamma. secalinZea mays MZEA1G 40.1 kD A1 protein (NADPH-dependent reductase) MZEACT1G Actin MZEADH11F Alcohol dehydrogenase 1 MZEADH2NR Alcohol dehydrogenase 2 MZEALD Aldolase MZEANT ATP/ADP translocator MZEEG2R Glutelin 2 MZEGGST3B Glutathione S transferase MZEH3C2 Histone 3 MZEH4C14 Histone 4 MZEHSP701 70 kD Heat shock protein, exon 1 MZEHSP702 70 kD Heat shock protein, exon 2 MZELHCP CAB MZEMPL3 Lipid body surface protein L3 MZEPEPCR Phosphoenolyruvate carboxylase MZERBCS RuBPC small subunit MZESUSYSG Sucrose synthetase MZETP12 Triosephosphate isomerase 1 MZEZEA20M 19 kD zein MZEZEA30M 19 kD zein MZEZE15A3 15 kD zein MZEZE16 16 kD zein MZEZE19A 19 kD zein MZEZE22A 22 kD zein MZEZE22B 22 kD zein Catalase 2 29 Regulatory C1 locus 30Bt codons were obtained from analysis of coding sequencesof the following genes: Bt var. kurstaki HD-73, 6.6 kbHindIII fragment (Kronstad et al. (1983) J. Bacteriol.154:419-428); Bt var. kurstaki HD-1, 5.3 kb fragment (Adanget al. (1987) in Biotechnology in Invertebrate Pathologyand Cell Culture, K. Maramorosh (ed.), Academic Press, Inc.New York, pp. 85-99); Bt var. kurstaki HD-1, 4.5 kbfragment (Schnepf and Whiteley (1985) J. Biol. Chem.260:6273-6280); and Bt var. tenebrionis, 3.0 kb HindIIIfragment (Sekar et al. (1987) Proc. Natl. Acad. Sci.84:7036-7040).REFERENCES1. Klee, H. J. et al. (1987) Mol. Gen. Genet. 210:437- 442.2. Altenbach, S. B. et al. (1987) Plant Mol. Biol. 8:239- 250.3. Rose, R. E. et al. (1987) Nucl. Acids Res. 15:7197.4. Vierling, E. et al. (1988) EMBO J. 7:575-581.5. Sandal, N. N. et al. (1987) Nucl. Acids Res. 15:1507- 1519.6. Tingey, S. V. et al. (1987) EMBO J. 6:1-9.7. Chlan, C. A. et al. (1987) Plant Mol. Biol. 9:533-546.8. Allen, R. D. et al. (1987) Mol. Gen. Genet. 210:211- 218.9. Sakajo, S. et al. (1987) Eur. J. Biochem. 165:437-442.10. Pirersky, E. et al. (1987) Plant Mol. Biol. 9:109-120.11. Ray, J. et al. (1987) Nucl. Acids Res. 15:10587.12. DeRocjer, E. J. et al. (1987) Nucl. Acids Res. 15:6301.13. Calza, R. et al. (1987) Mol. Gen. Genet. 209:552-562.14. Tingey, S. V. and Coruzzi, G. M. (1987) Plant Phys. 84:366-373.15. Winter, J. et al. (1988) Mol. Gen. Genet. 211:315-319.16. Osborn, T. C. et al. (1988) Science 240:207-210.17. Ryder, T. B. et al. (1987) Mol. Gen. Genet. 210:219- 233.18. Llewellyn, D. J. et al. (1987) J. Mol. Biol. 195:115- 123.19. Tingey, S. V. et al. (1987) EMBO J. 6:1-9.20. Gantt, J. S. and Key, J. L. (1987) Eur. J. Biochem. 166:119-125.21. Guidet, F. and Fourcroy, P. (1988) Nucl. Acids Res. 16:2336.22. Salanoubat, M. and Belliard, G. (1987) Gene 60:47-56.23. Volokita, M. and Somerville, C. R. (1987) J. Biol. Chem. 262:15825-15828.24. Bassner, R. et al. (1987) Nucl. Acids Res. 15:9609.25. Chojecki, J. (1986) Carlsberg Res. Commun. 51:211-217.26. Bohlmann, H. and Apel, K. (1987) Mol. Gen. Genet. 207:446-454.27. Nielsen, P. S. and Gausing, K. (1987) FEBS Lett. 225:159-162.28. Higuchi, W. and Fukazawa, C. (1987) Gene 55:245-253.29. Bethards, L. A. et al. (1987) Proc. Natl. Acad. Sci. USA 84:6830-6834.30. Paz-Ares, J. et al. (1987) EMBO J. 6:3553-3558.______________________________________
For example, dicots utilize the AAG codon for lysine with a frequency of 61% and the AAA codon with a frequency of 39%. In contrast, in Bt proteins the lysine codons AAG and AAA are used with a frequency of 13% and 87% respectively. It is known in the art that seldom used codons are generally detrimental to that system and must be avoided or used judiciously. Thus, in designing a synthetic gene encoding the Btt crystal protein, individual amino acid codons found in the original Btt gene are altered to reflect the codons preferred by dicot genes for a particular amino acid. However, attention is given to maintaining the overall distribution of codons for each amino acid within the coding region of the gene. For example, in the case of alanine, it can be seen from Table 1 that the codon GCA is used in Bt proteins with a frequency of 50%, whereas the codon GCT is the preferred codon in dicot proteins. In designing the synthetic Btt gene, not all codons for alanine in the original Bt gene are replaced by GCT; instead, only some alanine codons are changed to GCT while others are replaced with different alanine codons in an attempt to preserve the overall distribution of codons for alanine used in dicot proteins. Column C in Table 1 documents that this goal is achieved; the frequency of codon usage in dicot proteins (column A) corresponds very closely to that used in the synthetic Btt gene (column C).
In similar manner, a synthetic gene coding for insecticidal crystal protein can be optimized for enhanced expression in monocot plants. In Table 1, column D, is presented the frequency of codon usage of highly expressed monocot proteins.
Because of the degenerate nature of the genetic code, only part of the variation contained in a gene is expressed in this protein. It is clear that variation between degenerate base frequencies is not a neutral phenomenon since systematic codon preferences have been reported for bacterial, yeast and mammalian genes. Analysis of a large group of plant gene sequences indicates that synonymous codons are used differently by monocots and dicots. These patterns are also distinct from those reported for E. coli, yeast and man.
In general, the plant codon usage pattern more closely resembles that of man and other higher eukaryotes than unicellular organisms, due to the overall preference for G+C content in codon position III. Monocots in this sample share the most commonly used codon for 13 of 18 amino acids as that reported for a sample of human genes (Grantham et al. (1986 supra), although dicots favor the most commonly used human codon in only 7 of 18 amino acids.
Discussions of plant codon usage have focused on the differences between codon choice in plant nuclear genes and in chloroplasts. Chloroplasts differ from higher plants in that they encode only 30 tRNA species. Since chloroplasts have restricted their tRNA genes, the use of preferred codons by chloroplast-encoded proteins appears more extreme. However, a positive correlation has been reported between the level of isoaccepting tRNA for a given amino acid and the frequency with which this codon is used in the chloroplast genome (Pfitzinger et al. (1987) Nucl. Acids Res. 15:1377-1386).
Our analysis of the plant genes sample confirms earlier reports that the nuclear and chloroplast genomes in plants have distinct coding strategies. The codon usage of monocots in this sample is distinct from chloroplast usage, sharing the most commonly used codon for only 1 of 18 amino acids. Dicots in this sample share the most commonly used codon of chloroplasts in only 4 of 18 amino acids. In general, the chloroplast codon profile more closely resembles that of unicellular organisms, with a strong bias towards the use of A+T in the degenerate third base.
In unicellular organisms, highly expressed genes use a smaller subset of codons than do weakly expressed genes although the codons preferred are distinct in some cases. Sharp and Li (1986) Nucl. Acids Res. 14:7734-7749 report that codon usage in 165 E. coli genes reveals a positive correlation between high expression and increased codon bias. Bennetzen and Hall (1982) supra have described a similar trend in codon selection in yeast. Codon usage in these highly expressed genes correlates with the abundance of isoaccepting tRNAs in both yeast and E coli. It has been proposed that the good fit of abundant yeast and E. coli mRNA codon usage to isoacceptor tRNA abundance promotes high translation levels and high steady state levels of these proteins. This strongly suggests that the potential for high levels of expression of plant genes in yeast or E. coli is limited by their codon usage. Hoekema et al. (1987) supra report that replacement of the 25 most favored yeast codons with rare codons in the 5' end of the highly expressed gene PGK1 leads to a decrease in both mRNA and protein. These results indicate that codon bias should be emphasized when engineering high expression of foreign genes in yeast and other systems.
(iii) Sequences within the Btt coding region having potentially destabilizing influences
Analysis of the Btt gene reveals that the A+T content represents 64% of the DNA base composition of the coding region. This level of A +T is about 10% higher than that found in a typical plant coding region. Most often, high A+T regions are found in intergenic regions. Also, many plant regulatory sequences are observed to be AT-rich. These observations lead to the consideration that an elevated A+T content within the Btt coding region may be contributing to a low expression level in plants. Consequently, in designing a synthetic Btt gene, the A+T content is decreased to more closely approximate the A+T levels found in plant proteins. As illustrated in Table 3, the A+T content is lowered to a level in keeping with that found in coding regions of plant nuclear genes. The synthetic Btt gene of this invention has an A+T content of 55%.
TABLE 3______________________________________Adenine + Thymine Content in Btt Coding RegionCoding BaseRegion G A T C % G + C % A + T______________________________________Natural Btt 341 633 514 306 36 64geneSynthetic Btt 392 530 483 428 45 55gene______________________________________
In addition, the natural Btt gene is scanned for sequences that are potentially destabilizing to Btt RNA. These sequences, when identified in the original Btt gene, are eliminated through modification of nucleotide sequences. Included in this group of potentially destabilizing sequences are:
(a) plant polyadenylation signals (as described by Joshi (1987) Nucl. Acids Res. 15:9627-9640). In eukaryotes, the primary transcripts of nuclear genes are extensively processed (steps including 5'-capping, intron splicing, polyadenylation) to form mature and translatable mRNAs. In higher plants, polyadenylation involves endocucleotylic cleavage at the polyA site followed by the addition of several A residues to the cleaved end. The selection of the polyA site is presumed to be cis-regulated. During expression of Bt protein and RNA in different plants, the present inventors have observed that the polyadenylated mRNA isolated from these expression systems is not full-length but instead is truncated or degraded. Hence, in the present invention it was decided to minimize possible destabilization of RNA through elimination of potential polyadenylation signals within the coding region of the synthetic Btt gene. Plant polyadenylation signals including AATAAA, AATGAA, AATAAT, AATATT, GATAAA, GATAAA, and AATAAG motifs do not appear in the synthetic Btt gene when scanned for 0 mismatches of the sequences.
(b) polymerase II termination sequence, CAN.sub.7-9 AGTNNAA. This sequence was shown (Vankan and Filipowicz (1988) EMBO J. 7:791-799) to be next to the 3' end of the coding region of the U2 snRNA genes of Arabidopsis thaliana and is believed to be important for transcription termination upon 3' end processing. The synthetic Btt gene is devoid of this termination sequence.
(c) CUUCGG hairpins, responsible for extraordinarily stable RNA secondary structures associated with various biochemical processes (Tuerk et al. (1988) Proc. Natl. Acad. Sci. 85:1364-1368). The exceptional stability of CUUCGG hairpins suggests that they have an unusual structure and may function in organizing the proper folding of complex RNA structures. CUUCGG hairpin sequences are not found with either 0 or 1 mismatches in the Btt coding region.
(d) plant consensus splice sites, 5'=AAG:GTAAGT and 3'=TTTT(Pu)TTT(Pu)T(Pu)T(Pu)T(Pu)TGCAG:C, as described by Brown et al. (1986) EMBO J. 5:2749-2758. Consensus sequences for the 5'and 3' splice junctions have been derived from 20 and 30 plant intron sequences, respectively. Although it is not likely that such potential splice sequences are present in Bt genes, a search was initiated for sequences resembling plant consensus splice sites in the synthetic Btt gene. For the 5' splice site, the closest match was with three mismatches. This gave 12 sequences of which two had G:GT. Only position 948 was changed because 1323 has the KpnI site needed for reconstruction. The 3'splice site is not found in the synthetic Btt gene.
Thus, by highlighting potential RNA-destabilizing sequences, the synthetic Btt gene is designed to eliminate known eukaryotic regulatory sequences that affect RNA synthesis and processing.
Example 2
Chemical synthesis of a modified Btt structural gene
(i) Synthesis Strategy
The general plant for synthesizing linear double-stranded DNA sequences coding for the crystal protein from Btt is schematically simplified in FIG. 2. The optimized DNA coding sequence (FIG. 1) is divided into thirteen segments (segments A-M) to be synthesized individually, isolated and purified. As shown in FIG. 2, the general strategy begins by enzymatically joining segments A and M to form segments AM to which is added segment BL to form segment ABLM. Segment CK is then added enzymatically to make segment ABCKLM which is enlarged through addition of segments DJ, EI and RFH sequentially to give finally the total segment ABCDEFGHIJKLM, representing the entire coding region of the Btt gene.
FIG. 3 outlines in more detail the strategy used in combining individual DNA segments in order to effect the synthesis of a gene having unique restriction sites integrated into a defined nucleotide sequence. Each of the thirteen segments (A to M) has unique restriction sites at both ends, allowing the segment to be strategically spliced into a growing DNA polymer. Also, unique sites are placed at each end of the gene to enable easy transfer from one vector to another.
The thirteen segments (A to M) used to construct the synthetic gene vary in size. Oligonucleotide pairs of approximately 75 nucleotides each are used to construct larger segments having approximately 225 nucleotide pairs. FIG. 3 documents the number of base pairs contained within each segment and specifies the unique restriction sites bordering each segment. Also, the overall strategy to incorporate specific segments at appropriate splice sites is detailed in FIG. 3.
(ii) Preparation of oligodeoxynucleotides
Preparation of oligodeoxynucleotides for use in the synthesis of a DNA sequence comprising a gene for Btt is carried out according to the general procedures described by Matteucci et al. (1981) J. Am. Chem. Soc. 103:3185-3192 and Beaucage et al. (1981) Tetrahedron Lett. 22:1859-1862. All oligonucleotides are prepared by the solid-phase phosphoramidite triester coupling approach, using an Applied Biosystems Model 380A DNA synthesizer. Deprotection and cleavage of the oligomers from the solid support are carried out according to standard procedures. Crude oligonucleotide mixtures are purified using an oligonucleotide purification cartridge (OTC, Applied Biosystems) as described by McBride et al. (1988) Biotechniques 6:362-367.
5'-phosphorylation of oligonucleotides is performed with T4 polynucleotide kinase. The reaction contains 2 .mu.g oligonucleotide and 18.2 units polynucleotide kinase (Pharmacia) in linker kinase buffer (Maniatis (1982) Cloning Manual, Fritsch and Sambrook (eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). The reaction is incubated at 37.degree. C. for 1 hour.
Oligonucleotides are annealed by first heating to 95.degree. C. for 5 min. and then allowing complementary pairs to cool slowly to room temperature. Annealed pairs are reheated to 65.degree. C., solutions are combined, cooled slowly to room temperature and kept on ice until used. The ligated mixture may be purified by electrophoresis through a 4% NuSieve agarose (FMC) gel. The band corresponding to the ligated duplex is excised, the DNA is extracted from the agarose and ethanol precipitated.
Ligations are carried out as exemplified by that used in M segment ligations. M segment DNA is brought to 65.degree. C. for 25 min, the desired vector is added and the reaction mixture is incubated at 65.degree. C. for 15 min. The reaction is slow cooled over 11/2 hours to room temperature. ATP to 0.5 mM and 3.5 units of T4 DNA ligase salts are added and the reaction mixture is incubated for 2 hr at room temperature and then maintained overnight at 15.degree. C. The next morning, vectors which had not been ligated to M block DNA were removed upon linearization by EcoRI digestion. Vectors ligated to the M segment DNA are used to transform E. coli MC1061. Colonies containing inserted blocks are identified by colony hybridization with .sup.32 P-labelled oligonucleotide probes. The sequence of the DNA segment is confirmed by isolating plasmid DNA and sequencing using the dideoxy method of Sanger et al. (1977) Proc. Natl. Acad. Sci. 74:5463-5467.
(iii) Synthesis of Segment AM
Three oligonucleotide pairs (A1 and its complementary strand A1c, A2 and A2c and A3 and A3c) are assembled and ligated as described above to make up segment A. The nucleotide sequence of segment A is as follows:
TABLE 4__________________________________________________________________________Nucleotide Sequence of Segment A__________________________________________________________________________ ##STR2## ##STR3## ##STR4## ##STR5## ##STR6## ##STR7## ##STR8##__________________________________________________________________________ *c = complementary strand.
In Table 4, bold lines demarcate the individual oligonucleotides. Fragment A1 contains 71 bases, A1c has 76 bases, A2 has 75 bases, A2c has 76 bases, A3 has 82 bases and A3c has 76 bases. In all, segment A is composed of 228 base pairs and is contained between EcoRI restriction enzyme site and one destroyed EcoRI site (5')J. (Additional restriction sites within Segment A are indicated.) The EcoRI single-stranded cohesive ends allow segment A to be annealed and then ligated to the EcoRI-cut cloning vector, pIC20K.
Segment M comprises three oligonucleotide pairs: M1, 80 bases, M1c, 86 bases, M2, 87 bases, M2c, 87 bases, M3, 85 bases and M3c 79 bases. The individual oligonucleotides are annealed and ligated according to standard procedures as described above. The overall nucleotide sequence of segment M is:
TABLE 5__________________________________________________________________________Nucleotide Sequence of Segment M__________________________________________________________________________ ##STR9## ##STR10## ##STR11## ##STR12## ##STR13## ##STR14## ##STR15## ##STR16## 252 BASES (TOTAL)__________________________________________________________________________ *c = complementary strand
In Table 5 bold lines demarcate the individual oligonucleotides. Segment M contains 252 base pairs and has destroyed EcoRI, restriction sites at both ends. (Additional restriction sites within segment M are indicated). Segment M is inserted into vector pIC20R at an EcoRI restriction site and cloned.
As proposed in FIG. 3, segment M is joined to segment A in the plasmid in which it is contained. Segment M is excised at the flanking restrictions sites from its cloning vector and spliced into pIC20K, harboring segment A, through successive digestions with HindIII followed by BglII. The pIC20K vector now comprises segment A joined to segment M with a HindIII site at the splice site (see FIG. 3). Plasmid pIC20K is derived from pIC2OR by removing the ScaI-NdeI DNA fragment and inserting a HincII fragment containing an NPTI coding region. The resulting plasmid of 4.44 kb confers resistance to kanamycin on E. coli.
Example 3
Expression of synthetic crystal protein gene in bacterial systems
The synthetic Btt gene is designed so that it is expressed in the pIC20R-kan vector in which it is constructed. This expression is produced utilizing the initiation methionine of the lacZ protein of pIC20K. The wild-type Btt crystal protein sequence expressed in this manner has full insecticidal activity. In addition, the synthetic gene is designed to contain a BamHI site 5' proximal to the initiating methionine codon and a BglII site 3' to the terminal TAG translation stop codon. This facilitates the cloning of the insecticidal crystal protein coding region into bacterial expression vectors such as pDR540 (Russell and Bennett, 1982). Plasmid pDR540 contains the TAC promoter which allows the production of proteins including Btt crystal protein under controlled conditions in amounts up to 10% of the total bacterial protein. This promoter functions in many gram-negative bacteria including E. coli and Pseudomonas.
Production of Bt insecticidal crystal protein from the synthetic gene in bacteria demonstrates that the protein produced has the expected toxicity to coleopteran insects. These recombinant bacterial strains in themselves have potential value as microbial insecticides, product of the synthetic gene.
Example 4
Expression of a synthetic crystal protein gene in plants
The synthetic Btt crystal protein gene is designed to facilitate cloning into the expression cassettes. These utilize sites compatible with the BamHI and BglII restriction sites flanking the synthetic gene. Cassettes are available that utilize plant promoters including CaMV 35S, CaMV 19S and the ORF 24 promoter from T-DNA. These cassettes provide the recognition signals essential for expression of proteins in plants. These cassettes are utilized in the micro Ti plasmids such as pH575. Plasmids such as pH575 containing the synthetic Btt gene directed by plant expression signals are utilized in disarmed Agrobacterium tumefaciens to introduce the synthetic gene into plant genomic DNA. This system has been described previously by Adang et al. (1987) to express Bt var. kurstaki crystal protein gene in tobacco plants. These tobacco plants were toxic to feeding tobacco hornworms.
Example 5
Assay for insecticidal activity
Bioassays are conducted essentially as described by Sekar, V. et al. supra. Toxicity is assessed by an estimate of the LD.sub.50. Plasmids are grown in E. coli JM105 (Yanisch-Perron, C. et al. (1985) Gene 33:103-119). On a molar basis, no significant differences in toxicity are observed between crystal proteins encoded by p544Pst-Met5, p544-HindIII, and pNSBP544. When expressed in plants under identical conditions, cells containing protein encoded by the synthetic gene are observed to be more toxic than those containing protein encoded by the native Btt gene. Immunoblots ("western" blots) of cell cultures indicated that those that are more toxic have more crystal protein antigen. Improved expression of the synthetic Btt gene relative to that of a natural Btt gene is seen as the ability to quantitate specific mRNA transcripts from expression of synthetic Btt genes on Northern blot assays.

Number	Name	Date
4356270	Itakura	Oct 1982
4888282	Beremand et al.	Dec 1989
5254799	DeGreve	Oct 1993
5276268	Strauch et al.	Jan 1994
5380831	Adang et al.	Jan 1995

	Number	Date	Country
Parent	827844	Jan 1992
Parent	242482	Sep 1988

	Number	Date
Parent	57191	May 1993
Parent	848733	Apr 1986
Parent	535354	Sep 1983

Synthetic insecticidal crystal protein gene

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

CROSS REFERENCES TO RELATED APPLICATIONS

US Referenced Citations (5)

Foreign Referenced Citations (1)

Non-Patent Literature Citations (20)

Continuations (2)

Continuation in Parts (3)