MXPA99001544A - Modified acyl-acp desaturase - Google Patents

Abstract

Disclosed is a method for modifying the chain length and double bond positional specificities of a soluble plant fatty acid desaturase. More specifically, the method involves modifying amino acid contact residues in the substrate binding channel of the soluble fatty acid desaturase which contact the fatty acid. Specifically disclosed is the modification of an acyl-ACP desaturase. Amino acid contact residues which lie within the substrate binding channel are identified, and subsequently replaced with different residues to effect the modification of activity.

Description

Desaturase is specific for stearoyl-ACP and introduces a double bond between carbons 9 and 10. Initial desaturation reactions in animals and fungi and subsequent desaturation reactions in plants, are mediated by a different class of fatty acid desaturases that they are integral membrane proteins. Since most plants lack other desaturases that act on the 18: 0 level, the ratio of saturated to unsaturated fatty acids in higher plants is mainly controlled by enzymes that catalyze the conversion of saturated fatty acids into monounsaturated. The? 9 desaturase cDNA codes for precursor proteins that contain an N-terminal transit peptide to target the plastid. For safflower and castor, the 33-residue transit peptide is excised to obtain a mature desaturase olipeptide of 363 amino acid with an apparent molecular weight of 37 kDa per subunit per "SDS-AGE". The enzyme is produced in dimers of about 70 a. The enzymatic reaction requires molecular oxygen, AD (P) H, NAD (P) H ferredoxin oxide-reductase and ferredoxin. Previous studies have shown that both the soluble 9 and the membrane-bound lesaturasas require non-heme b-iron for catalytic activity. More recently, spectroscopic analysis and comparisons of amino acid sequences have established that the? 9 desaturase contains a di-iron group. This class of di-iron proteins is characterized by two occurrences of the sequence motif E-X-X-H, spaced at approximately 100 amino acids, and includes the R2 subunit of the ribonucleotide reductase and a methane monooxygenase hydroxylase component. A greater understanding of the catalytic mechanism of the acyl -ACP desaturases enzymes may allow the exploitation of said enzymes in the manufacture of vegetable seed oil. SUMMARY OF THE INVENTION The invention in question relates to a method for modifying the length of the chain and the specificities of the positions of the double bonds of a plant-soluble fatty acid desaturase. More particularly, the method involves the modification of residues ie contact amino acids in the substrate binding channel ie the soluble desaturase of fatty acids which contact with fatty acid. In preferred embodiments, the soluble fatty acid desaturase is an acyl-ACP desaturase. The contact amino acid residues that are found within the substrate binding channel are identified, for example, by first having the primary amino acid sequence of the acyl-ACP desaturase.

Many such sequences are known and others can be determined through the application of routine experimentation. Said amino acid sequences are then aligned with the primary amino acid sequence of the? 9 iesaturase of Ricinus communis for maximum conservation ie sequence. A three-dimensional model is then constructed for the acyl-ACP desaturase based on the sequence conservation with the 9 desaturase of Ricinus commuriis. The contact amino acid residues in the substrate binding channel of the patterned structure are then identified. A potent acyl-ACP desaturase is then generated having chain length and modified double bond position specificities, substituting one or more of the contact amino acid residues with another amino acid residue. In another aspect, the present invention relates to a mutant acyl-ACP desaturase, which is characterized by the ability to catalyze the desaturation ie a first fatty acid and a second fatty acid, whose first and second fatty acids differ in their length of adena. . This mutant is further characterized by the ability to desaturate both the first and second fatty acids at rates that differ by no more than about 4 times. of amino acids between members of different families, such as? 4,? 6 and? 9 acyl-ACP desaturases. It is known that each of these desaturases catalyzes the formation of double bonds between the carbon atoms of the same fatty acids or the like. The primary difference between the activities of the various acyl-ACP desaturase is the location of the carbon atoms within the substrate fatty acids that are to be desaturated. The conservation of the amino acid sequence is even greater in a particular family of acyl-ACP desaturases, such as? 9. Based on the present disclosure, one skilled in the art will predict that contact residues within the substrate-binding channel of all members of the 9-acyl-ACP desaturases are substantially similar, if not identical, to those identified in the 9-acyl-ACP desaturase described in the following Example 1. The high degree of homology in the amino acid sequence within a family of acyl-ACP desaturases that catalyze the same enzymatic reactions and the sequence homology between acyl-ACP desaturases families that catalyze different enzymatic reactions suggest that certain portions of the enzymes will exhibit similar tertiary structures. This is consistent in the discovery of other molecules, such as antibodies, where the conservation of amino acid residue homology is normally higher in those amino acids involved in maintaining the functional structure of the molecule in question. One such structural region in the acyl-ACP desaturases that is conserved among the different acyl-ACP desaturases is the substrate-binding channel described in the Examples section given below. The substrate attachment channel described below exhibits an architecture that provides an almost perfect accommodation for the fatty acid substrate. Although it has a precedent, this exquisite adaptation is extremely unusual. The fact that this substrate binding channel is highly conserved can be confirmed by alignment for the maximum identity (by conventional techniques) of the amino acid sequences of the members of other families of acyl-ACP desaturases with that of α9. acyl-ACP castor desaturase (i.e., Ricinus co munis) described in the following Example 1. The deduced amino acid sequence of this castorine protein was described by Shanklin and Somerville (Proc. Nati. Acad. Sci. USA 88: 2510 (1991)). After this alignment, a modification of a contact residue can be generated (and, in some cases, other residues, as exemplified by the chimeras of Example 2) can alter the length of the chain and the positional specificities of the doubles links of an acyl-ACP desaturase. For example, as shown in the following Example 2, a chimera was produced in which amino acids 172-202 of the? 6-16: 0 -ACP desaturase were replaced by amino acids 178-207 of the? 9-18: 0-ACP desaturase. This led to the introduction of 9 new amino acids in the α-substrate channel of the 6-16: 0 -ACP desaturase, which differed from the amino acids at the corresponding positions in the α-6-16: 0 -ACP desaturase wild type. The chimera was not only capable of desaturating the fatty acid 16: 0 with which the wild type worked best, but also it could desaturate the 18: 0 as much in the position? 6 omo in the? 9 to equivalent levels. The fact that contact amino acid residues in the substrate-binding channel of an acyl-ACP desaturase play a critical role of this type in the determination of chain length and specificity (ie the positions of the double bonds offers an opportunity for the rational design of acyl-ACP desaturases mutants that have unique and useful properties.These new mutant molecules can be designed, for example, by first identifying the contact residues in the substrate-binding channel (as described above). , through the alignment with the amino acid sequence? 9 of castor, followed by three-dimensional modeling.) Specific point mutations can then be introduced into the molecule of acyl-ACP desaturase ie interest.This is more conveniently done at the genetic level using techniques common, such as site-directed mutagenesis, a variety of mutagenic techniques can be applied A site for introducing a change (ie, substitution) of a specific amino acid codon in said DNA. Care must be taken in selecting a residue to replace an existing contact residue in the substrate binding channel of an acyl-ACP desaturase of wild type. It is important, in general, in the initial studies, for example, to select waste for substitution that does not differ radically with respect to the size or load of the side chain. For example, if a glycine contact residue (characterized by its compact aliphatic side chain) is identified in the substrate binding channel, the substitution of an amino acid residue, such as arginine (characterized by the presence of a side chain) voluminous basic) could serve to block the entry of the fatty acid substrate into the substrate-binding channel through stearic block. In general, substitutions with initial amino acids of contact residues must be made using amino acids that have similar loading characteristics, with relatively small differences in terms of side chain volume. Having said this, it is certainly possible that the substitution of an amino acid having radically different properties with respect to a wild-type contact residue may result in a particularly useful acyl-ACP desaturase mutant. Said molecule would be included in the present invention. The brief discussion of the substitution strategy given above is intended only as a guide for the incremental modification of an acyl-ACP desaturase. Therefore, it is the knowledge of the identity of the contact residues in an acyl-ACP desaturase which allows experts in the art to make modifications of the enzyme that may alter the length of the chain and the position specificities of the doubles enzyme bonds without inhibiting their ability to carry out enzymatic catalysis. This knowledge, in turn, depends on the ability of a person skilled in the art to identify the substrate attachment channel and generate a three-dimensional model. As already indicated, the nucleotide sequences of many acyl-ACP desaturases have been described. Moreover, given their high degree of conservation, the utmost experiments of nucleic acid hybridization carried out with DNA isolated from a plant of interest would probably lead to DNA encoding additional acyl-ACP desaturates. In addition, as indicated above, one skilled in the art would predict that, within the family of the? 9 cil-ACP desaturases, the contact amino acid residues within the substrate binding groove would be substantially similar, if not identical. The contact amino acid residues identified by the X-ray crystallographic work described in Example 1 are residues M114, L115, T117, L118, P179, T181, G188 and F189. The fact that the modification of these residues in a? 9 acyl-ACP desaturase actually alters, in fact, the length of the chain and the positional specificities of the double bonds of the enzyme was confirmed in the experiments described in Example 2. More specifically, preliminary experimental work has revealed that, a simple amino acid substitution at position 118 (Leu to Phe) in the? 9 acyl-ACP desaturase castor results in an approximately 10-fold increase in its activity with 16: 0-ACP. Thus, an amino acid substitution in a contact residue position can generate an acyl-ACP desaturase with new and useful properties. Prior to this invention, the only source of acyl-ACP desaturases variants was the plant tissue that synthesizes non-usual isomers of monounsaturated fatty acids. For example, the 4-16: 0-ACP desaturase was derived from the seed of coriander seeds, a tissue that produces large amounts of petroselinic acid (18: l? S), an unusual monounsaturated fatty acid. The present invention allows the design and production of new types of acyl-ACP desaturases without the need to isolate cDNA for these enzymes and plant sources. In addition, the present invention allows the design of acyl-ACP desaturases that can catalyze the synthesis of economically valuable fatty acids that are not found in nature. In a more specific example, this invention offers an alternative means of producing petroselinic acid in plants. This fatty acid has a number of potential industrial and nutritional uses. The only known route of formation of petroselinic acid in plants involves desaturation? 4 of 16: 0-ACP, followed by elongation of the resulting 16: l? 4-ACP to form 18: l? 6 (or petroselinoyl) -ACP. This route requires, among other things, a new acyl-ACP desaturase and a specific acyl-ACP elongation system. Among the desaturases mutants described below are enzymes that can catalyze the 6? 18: 0-ACP saturation to form petrosinoyl-ACP. Said enzymes are useful for the production of petroselinic acid in transgenic cultivation plants without the need to transfer additional genes for the elongation route of 16:? 4-ACP. This route is a current limitation in the efforts to produce petroselinic acid in crop-growing plants through the introduction of the gene For the? 4-16: 0-ACP desaturase. Therefore, mutants generated by altering the identity of one or more contact residues in the substrate binding channel can be used to generate acyl-ACP desaturases having unique functional characteristics. Said enzymes can be used, for example, to generate vegetable oils rich in monounsaturated fatty acids. These vegetable oils are important in human nutrition and can be used as renewable sources of industrial chemicals. In addition, the ability to manipulate the chain length preferences and the positions of the double bonds of these molecules offers a form of manipulation of the physical properties and commercial uses of conventional vegetable oils. In addition, the development of transgenic crops capable of producing non-usual types of monounsaturated fatty acids can be exploited based on the present description. The mutants described in Example 2 below exhibit certain unique properties. For example, acyl-ACP desaturases of wild type tend to exhibit very strong preferences for a fatty acid of particular chain length and binding position. However, in the experiments described below, amino acid substitutions of the contact residues in the substrate binding channel have been shown to modify this preference. For example, chimeric mutants are disclosed that exhibit the ability to catalyze the desaturation of substrates of different lengths (e.g., 16: 0 and 18: 0) at rates that differ by no more than about 4-fold. The nucleic acid sequences encoding these acyl-ACP desaturases glycosylation can be used. After purification by conventional methods, an acyl-ACP desaturase mutant expressed in a eukaryotic system (e.g., the baculovirus expression system) can be used to modify the length of the chain and the position of double bonds of an acid fatty. This protein can also be used as part of a crude lysate in many circumstances. The mutant acyl-ACP desaturases can also be cloned into an expression vector in plants. These vectors allow the production of a desired protein product, for example the mutant acyl-ACP desaturase, in the medium of the plant cell in which the substrate fatty acid resides. By producing the enzyme itself, the modification of the product can occur before it is collected, allowing the rapid purification of the desired fatty acid with the appropriate double bond position and without the need for expensive manufacturing steps. In some cases, more than one acyl-ACP desaturase mutant may be desired in a particular transgenic plant to produce fatty acids with double bonds in multiple positions. Plants are also easy to grow and grow in large numbers. This sprout can also be used as part of a crude lysate in many circumstances. EXAMPLES Example 1 Results and discussion Map of electron density and model quality: The three-dimensional structure of recombinant homodimeric desaturase? 9 of the stearoyl-acyl carrier protein, the archetype of soluble vegetable fatty acid iesaturases, has been determined. Saturated fatty acids in unsaturated, by protein crystallographic methods at a resolution of 2.4 A. Lia. Electronic density averaged six times for the main chain and side chains for most of the polypeptide chain is well defined. Exceptions are the first 18 residues at the N-terminus, which are not defined in electronic density and could be flexible in the crystal lattice. Residues 336-347, located in a loop region, are very poorly defined in electron density maps and it is also in this part of the protein structure that the greatest deviations of non-crystallographic symmetry are found. global residual by real spatial correlation of residuals (Br subunit and electronic density map 2Fo-Fc averaged six times) is 0, 76 Criteria such as the crystallographic factor R (R = 22.0%, R = 28.5% with non-crystallographic symmetry restrictions), good stereochemistry of the model (link length rms of 0.008 A), Ramachandran graph (only one Atypical value of the regions allowed per subunit, except for the glycine residues) and observed hydrogen bond pattern, all indicate that the chain check for the fatty acid desaturase is correct. There is a very clear density for the peptide oxygen of Lys 262, the residue with an unaccepted main chain conformation. The high average β factor suggests that the molecule is flexible. The most ordered parts of the molecule are areas involved in the interactions of dimers and hexamers, while the surface loops often have very high β-factors. The major binding sites for the ions u (CN) in the derivatized desaturase crystals are close to the side chains of K56 and C61 on the surface of the molecule. One of the smaller sites is internal, between the side chains of H203 and C222, and the second smaller site is in the area where the N-terminus of the chain is probably located. The overall form of the subunit of? 9 desaturase is a compact cylinder of dimensions 35x35x50, with an accessible surface area of 16773. In addition to a hairpin loop ß at the same C-terminus of the chain, the subunit is mainly composed of helical structures secondaries folded into a large domain. Nine of the eleven total a-helices form an antiparallel helical bundle. The N-terminal part of the chain is disordered, no electron density is observed for the first 18 residues. The next 15 residues lack secondary structure and form an extended chain packing along the helix bundle, with few specific interactions to stabilize their structure. The first helix, al, composed of 23 residues, begins and ends in conformation 310 and is very folded, so that its first half forms a shot at one end of the beam and its second part is the first helix of the beam. The chain continues in the same direction, forming loops linked by hydrogen and a helix 310. The end at the other end of the beam is formed by helices a2 and a? 9 and the C-terminal fork. Four of these helices, a3, a4, a6 and a7, which are very long, of 28, 29, 30 and 31 residues, respectively, contribute with ligands to the center of iron. Although a3 has an interruption in the helical structure in the middle, at residue 107-108, a3 and a4 are symmetric to a6 and a7 and can be superimposed with a r.m. of 1.39 A for 44 residues. Said superposition also aligns the iron atoms to within 1.0 A. The corresponding sequence alignments show that there is little sequence conservation in addition to the residues involved in the binding to the iron group. This superposition also orientates the part of the auction of the envelope a2. The connections between these helices also roughly overlap, although there is no detailed structural similarity. Between a3b and a4 there is an outstanding loop structure stabilized by several turns linked by hydrogen bonds. The a5, a8, alO and all helices, which are very curved, complete the beam. There is a large number of salt bridges, 25, excluding those that interact with iron ions, in the subunit. This corresponds to 0.069 ion pairs per residue, higher than the average number of ion pairs per residue, 0.04, derived from a study of 38 high-resolution protein structures. Seven of the salt bridges in the? 9 desaturase are involved in inter-helical interactions in the beam, ensuring correct mutual packing and, in some cases, the correct orientation of the iron ligands. Eight pairs make intra-helical contacts and three of the others are involved in anchoring the turns between beam-helices. Three pairs are involved in contacts with the 310-helix and the loop after a8. A peculiar characteristic of the subunit? 9 desaturase is a rather flat surface formed by the helices at, a6, a7, alO and all. This surface is not involved in the subunit-subunit contacts in the dimer, but is accessible from the solution. 31 dimer The subunit-subunit interface in the dimer burns a surface area of 5826 A2 per dimer, 17.4% of the area of the dimer. These double interactions include xtensos contacts between the helices in the beams; from a3b nasta the same helix in the second subunit, between a4 and ?? 5, and over a2 and a4 to the corresponding helices in the other subunit. There are also many contacts between the prominent loop between a3b and a4 and the N-terminus,? X3b, and a5. In addition, the residues in the connection between al and a2 lase contacts with a4 and a5 in the second subunit. There are three charged interactions in the contact area of the dimer; two of these involve residues of a5. The iron centers are separated by more than 23 A in the dimer and have no direct contact with each other Non-crystallographic symmetry and crystalline packing The asymmetric unit of the crystal contains three dimers of 9-desaturase In these dimers, the subunits are related by double non-crystallographic axes, which, for one of the dimers, are parallel to one of the crystallographic double axes.At right angles of this, parallel to a, there is a triple non-crystallographic screw axis that relates the three dimers. of the length of a, that is, which is a local axis 31. The contacts between the dimers are not extensive, of the same order as other crystalline contacts, and the influence of the crystalline contacts on the structure seems to be smaller, as judged because of the small deviations observed in the non-crystallographic symmetry, the largest losses are obtained for residues 336-347, where R336, E347 and / or K346 make crystalline contacts, including salt bridges, in some of the subunits. The electronic density in this area is weak, as indicated previously. Another area with deviations from non-ristalographic symmetry includes residues 19 to 50, which are wound around the subunit and also make different weak crystal contacts in the subunits. The packing of subunits corresponding to an asymmetric unit is seen along the triple axis and one of the double axes. The iron core Previous studies have shown that the? 9 desaturase contains four iron atoms per dimer and the optical and Móssbauer spectroscopy indicated that these iron ions consist of di-iron-oxo groups. The dihierro-oxo groups have now been identified in a wide variety of proteins that perform both catalytic and non-catalytic functions. They contain two iron atoms connected by an oxo or hydroxo bridge-forming ligand and have been classified based on different motifs of the primary sequence provided by the group ligands and the structural differences elucidated by X-ray crystallography. Four classes have been described , one containing hemeritrin and myohemeritrin; a second containing the R2 subunit of the ribonucleotide reductase, bacterial hydrocarbon hydroxylases and the? 9 desaturase; a third containing rubreritrin, and a fourth containing Fe (III) -Zn (II) purple acid phosphatase (Fe (II) - (Fe (II) phosphatase Strmamífero acids) In addition to these soluble proteins, there is a distinct class of functionally related integral membrane proteins, including the fatty acids desaturases and the hydrocarbon hydroxylases, which contain non-heme iron centers activated by oxygen, which have yet to be structurally characterized.The crystalline structure of the β9 reveals that the enzyme belongs to the proteins dihierro of class II and that contains a metallic group.The distance between the iron ions is of 4.2 A and the geometry of coordination of the iron ions is a distorted octahedron, where one of the positions of the ligands is without ocular The structure of the group is highly symmetric: E143 of a4 and E229 of a7 both act as bridge-forming ligands.E105 of a3a is a bidentate ligand for an iron ion and corresponds E166 from a6 is a bidentate ligand for the second iron ion. Each iron ion is also linked by a nitrogen atom, Ndl in H146 of a4 and H232 of a7, respectively. The orientation of the iron ligands is maintained, in some cases, by hydrogen bonds of the side chains; E105 interacts with H203, E143 with the Nel atom in W139, Ne2 in H146 with the side chain of D228, which, in turn, interacts with the side chains of R145 and 62, Ne2 in H232 with the side chain of E143, which , in turn interacts with the side chain of R231. Further away from one of the iron ions is the Nel atom at 139, which could be considered as a second ligand of the cortex. In the vicinity of the iron group, there is an electron density corresponding to a solvent molecule. Its distances to iron ions are 3.2 and 3.4 A, respectively, and is therefore not part of the first coordination shell of the metal center. Form of the desaturase in the crystal structure The presence of a dihierro group with bridges μ- DXO in the diferric state of the? 9 desaturase has been shown unambiguously using Móssbauer Y spectroscopy of Raman resonance. It is, therefore, surprising that no μ-oxo bridge was observed in the electron density map of the? 9 desaturase, since the enzyme used for the experiments was in the oxidized state and no reducing agent was added to the mother liquor . In addition, the difference between iron ions (4.2 A) is greater than expected for a di-iron group with an intact μ-oxo bridge. -Cn the oxidized form of the ribonucleotide reductase with the μ-oxo bridge present, the iron-iron distance is 3.3 A. The geometry observed in the? 9 desaturase is surprisingly similar to that observed in the reduced form of the ribonucleotide reductase, where, with the chemical reduction of R2, the distance between the iron ions increases to 3.8 A, the μ-oxo bridge and the arrangement of ligands becomes very symmetric, as seen by protein crystallography and Móssbauer spectroscopy. It may be suggested that the exposure of the desaturase crystals to X-ray radiation results in the photochemical reduction of the metal center, which is accompanied by loss of the μ-oxo bridge and redistribution of ligands. By therefore, the structure of the? 9 desaturase presented herein most likely represents the reduced form of the enzyme. The crystal structure of the? 9 desaturase reveals a highly symmetric ligand arrangement of the iron group in the diferrosa form of the enzyme, according to the Raman resonance studies. Deviations in the symmetric distribution of ligands in the reduced state of the enzyme, as suggested by the previous data of Móssbauer temperature dependents could be due to variations in link lengths and bond angles in the two metal sites, too small for be observable in electronic density maps at the current resolution. Active site and interactions with other proteins The structure of the? 9 desaturase described here strong and elongated in the electron density maps 2Fo-Fc averaged in this channel, which had not been assigned to solvent or protein atoms. Based on the shape of this density and the hydrophobic character of the pocket, it can be assumed that this electron density may represent the hydrophobic acyl tail of a β-octylglucoside molecule. The hydrocarbon tail of octylglucoside would adapt well to this density, but the density corresponding to the rest of sugar is poorly defined. This putative octylglucoside molecule is oriented with its tail deep in the hydrophobic pocket, close to the dihierro group, and the carbohydrate moiety extending to the surface. The weak electron density for this part of the molecule could indicate a local disorder resulting from less specific interactions with the enzyme. The modeling of a stearic acid in the so-called substrate binding pocket places the C9 carbon atom at about 5.5 A of one of the iron ions. This carbon atom, where the double bond will be formed, is also close to the small pocket with the solvent molecule attached; in fact, the water molecule forms a bridge in the distance between the carbon C9 of the substrate and the nearest iron ion. In the active enzyme, this pocket is probably occupied by an oxygen molecule bound to one or both iron atoms. During the catalysis, a peroxide radical capable of removing one of the hydrogen atoms at the C9 position of the fatty acid could be generated. Comparison with other dihydric proteins A superposition of the structure of the? 9 desaturase on the three-dimensional structures of other di-iron proteins, the R2 subunit of the ribonucleotide reductase of Escherichia coli and the MMO subunit of Methylococcus capsullate, shows that the global structures they are quite similar, with an rms setting from 1.90 A for 144 Ca atoms (? 9 desaturase versus R2) and an adjustment r.m.s. from 1.98 A for 117 Ca atoms equivalent (? 9 iesaturase versus MMO). The folds are very similar, most of the a-helices, a a8 and aO, have their counterparts in R2 and MMO. There are few amino acids conserved in addition to the iron ligands, but there can be little doubt that these enzymes are evolutionarily related. There are significant differences in the (structure of the iron centers in the three proteins.) In general, the metal center in the? 9 desaturase is (considerably more symmetric than in the other two proteins.) However, when compared to the structure of The reduced form of R2, the coordination geometries of the dinuclear iron center in the 9-desaturase and R2 are more similar.The most significant difference is that, in the 9-desaturase, the terminal carboxylates E105 and E196, respectively, act as bidentate ligands, whereas, in R2, the equivalent side chains are monodentate ligands for iron ions, R2 is unique among these enzymes, in that it forms a stable radical at position Y122.The corresponding residue in the? 9 desaturase is L150, located in the hydrophobic group that makes packaging interactions in the four-helix bundle that binds to the iron group and there is no evidence that which could indicate that this residue is necessary for the catalytic activity. There are very few amino acid residues that are conserved in all three enzymes. Among them are ligands for metal ions, with the exception of E105, which is substituted by anaspartic acid in R2. The only other invariant residues are 1225 and D228. Residue 1225 is in the neighborhood of the diolier group (closest distance 4.6 A) on the opposite side of the substrate channel.

The side chain is packaged between H203, H146 and 62 in the three-dimensional structure and a more detailed examination of its function has to await the results of site-directed mutagenesis studies. The other invariant protein residue in the three enzymes, D228, is part of an electron transfer path from the dinuclear iron center to the surface of the protein, which has been suggested for R2. In R2, this route goes from one of the iron ions, through the side chain of H118, D237, to W48, which is located on the surface of the protein. These residues are conserved in the? 9 desaturase and a similar route for electron transfer can be postulated to include the structurally equivalent residues H146, D228 and W62, as indicated above. Moreover, a slightly modified route for electron transfer could also be suggested for MMOs. In this case, the iron ligand (H147) and the aspartic acid residues (D242) are preserved; however, the structure on the surface is different. However, an aromatic lateral adena (Y67) on the surface is in the vicinity of the side chain of D242. Most of the other residues conserved between the? 9 desaturase and R2, on the one hand, and the? 9 iesaturase and MMO on the other, are located on the surface of the protein, or are involved in packaging interactions. The conserved residues common between R2 and the? 9 desaturase in the vicinity of the iron site are residues W135 and W139. While 135 and W139 are strictly conserved in the desaturase, the corresponding residues W107 and III in R2 are not strictly conserved. With the exception of T4 and the E. coli protein, W135 is replaced by a side chain of phenylalanine or tyrosine. Similarly, 139 is replaced by a glutamine residue. Materials and methods Purification and crystallization of the enzyme. Recombinant? 9 expressed in E. coli cells was purified as previously described (Fox et al., Biochemistry 33: 127766 (1993)). The crystallization of the enzyme was achieved according to Schneider et al., J. Mol. Biol. 225: 561 (1992), with slight modifications. The enzyme samples were concentrated at 12-18 mg / ml. An aliquot of 7.5 ml of protein solution was mixed with the same amount of the reservoir solution, put on coverslips And it was left to equilibrate on 1 ml of the solution of the well at 20 ° C. The reservoir solution contained 0.08 cacodylate buffer, pH 5.4, 200 mM Mg-acetate, 75 mM ammonium sulfate, 2 mM LiCl, 1 mM KCl, 0.2% ß-octylglucoside and 12-15% PEG 4000 as Data frames were processed with DENZO and converted to scale with SCALEPACK. Determination of phases, construction of models and crystallographic refinement. Most of the crystallographic calculations were performed using the series of CCP4 programs (Collaborative Computational Project, Number 4, Acta Crystallogr, D50: 760, (1994)). The initial crystallographic analysis was carried out with the data sets collected in the multifilar detector at a resolution of 3.1 A. The Patterson differences map for the gold derivative was analyzed using RSPS (Knight, PhD thesis, Swedish University of Agricultural Sciences, Uppsala 1989). Two sites were used, related by a strong cross peak in Patterson's difference map for the calculation of Fourier difference maps and new sites were identified. Finally, 6 main sites and 12 minor sites were found and the heavy metal parameters were refined using MLPHARE. By the results of the calculations of the rotation function and the positions of the metal ions, it was possible to determine the direction and position of the local symmetry operators, which relate the six subunits of the? 9 desaturase in the asymmetric unit. The sextuple averaging of non-crystallographic symmetry was then used using the RAVE program (Jones, in CCP4 Study Weekend 1992: Molecular Replacement (Dodson, EJ, Gover, S. and olf, W., eds.), Pp. 91-105, Daresbury Laboratory, Daresbury, UK (1992)) and a spherical shell, centered on the presumed position of a subunit of? 9 desaturase, to refine the initial SIR phases. Thanks to the electronic density map at low resolution, based on these phases, it was possible to identify part of the central beam of four helices, which coordinates the iron center and the iron atoms. The coordinates of the iron atoms were refined from the anomalous native data and new phases were calculated based on the Au derivative and the anomalous contribution of the iron atoms. A new envelope was made for the subunit using MAMA (Kleywegt and Jones, Acta Cryst, D50: 178 (1994)) by orienting approximately one subunit of R2 in the correct position for the propeller bundle. After the non-crystallographic averaging, it was possible to construct an initial model of the desaturase from the electronic density map. Model building cycles, refinement in XPLOR (Brunger, A., Crystallogr Act A45: 50 (1989)) (Brunger, A., The X-PLOR manual, Yale University, New Haven, CT (1990)) , the redefinition of the envelope, the refining of the symmetry operators using IMP (Kleywegt and Jones, Acta Cryst, D50: 171 (1994)) and the averaging were carried out until no electron density appeared in the averaged maps. In this stage, a further loop was created that seemed to have a different structure in the subunits from the 2Fo-Fc maps. Crystallographic refining was carried out with XPLOR, using the Engh and Huber force field (Engh and Huber, -Acta Crystallogr, A47: 392 (1991)) and restrictions of non-crystallographic symmetry. Due to the low resolution (3.1 A) of the data set, a global B value was used. The model in this stage had a crystallographic R factor of 26.7%, with restrictions of sextuple non-crystallographic symmetry imposed in the refining. In this stage of refining, a new set of native data could be available for a resolution of 2.4 A, collected in NSLS, and refining was continued with this data set. The refining procedure was monitored by 2.5% of the reflections that were not included in the refining, but were used to calculate Rlibre (Brunger, A., Nature 355: 412 (1992)). Even at the resolution of 2, 4 A, the observation of the parameter ratio is just about one and the refining problem is poorly determined. Therefore, throughout the refining process, restrictions of non-crystallographic symmetry were employed to avoid over-adjustment of the diffraction data. They were only without restricting those parts of the structure that were judged, by the averaged electronic density maps, which did not obey the non-crystallographic symmetry. This includes residues 19-50, 121-122, 127-129, 208-212, 241-253, 259-260, 308-319, 336-348 and some side chains. The electron density for some residues in the 336-347 region is so weak that their positions must be considered arbitrary and the occupancies for these atoms were, therefore, adjusted to zero. The global anisotropic refining reduced the free R factor by approximately 2%. In this stage, water molecules were added to the model. Individual B factors were also refined, but restricted by non-crystallographic symmetry. The final model has a crystallographic R factor of 22.0% (free R 28.5%). The deviations r.m. for the restricted Ca positions (263 atoms) of subunit A to the corresponding parts of the other subunits are 0.06 and, for all atoms Ca (345 atoms), 0.26, 0.23, 0.24, 0 , 32, 0.25, respectively. The protein model was analyzed using the PEPFLIP and RSFIT options in O (Jones et al., Acta Crystalllog.A47: 100 (1991)) and with the PROCHECK program (Laskowski et al., J. Appl. Crystallogr. 282 (1993)). The atomic coordinates will be deposited in the Brookhaven Protein Data Bank. Structural comparisons All structural overlays were performed by least squares methods using O (Jones et al., Acta Crystallogr, A47: 100 (1991)) and were made in pairs. The superposition was made by selecting an initial group of equivalent Ca atoms, consisting of four sections of the polypeptide chain (approximately 10 residues each) of the four helices containing the ligands for the center of iron. This initial alignment was later maximized including all the Ca atoms of the atomic models. The residues were considered structurally equivalent if they were within 3, 8 of each other and within a consecutive section of more than three equivalent residues. Example 2 Results and discussion The approach of combining the elements of the amino acid sequence of structurally related enzymes with different properties has proven to be effective in the characterization of the substrate specificities and positions of the fatty acid modifying enzymes, such as the lipoxygenases of mammal and acyl-ACP plant thioesterases. This approach was used here to identify the residues responsible for the differences in properties of a? 9-18: 0-ACP desaturase and a? 6-16: 0-ACP desaturase encoded by the T. alata cDNAs pTAD2 and pTA? 4, respectively. The mature polypeptides encoded by these cDNAs share 65% amino acid sequence identity. Initially, two chimeric mutants were constructed: (a) Chimera 1 contained the first 171 amino acids of the mature? S-16: 0-ACP desaturase bound to the remaining 185 amino acids of the? 9-18: 0-ACP desaturase and ( b) Chimera 2 contained the first 227 amino acids of the mature? 9-18: 0-ACP desaturase bound to the remaining 134 amino acids of the? S-16: 0-ACP desaturase Both enzymes showed only detectable 9-18: 0-ACP desaturase activity. In addition to catalyzing a similar activity, these mutants share a region of 50 overlapping residues (residues 178-227) of the? 9-18: 0-ACP desaturase. This suggested that the determinants of chain length and positional specificities of double bonds are present in this portion of the 9-18: 0-ACP desaturase. Therefore, a Chimera 3 was constructed in which residues 172-221 of the? S-16: 0-ACP desaturase were substituted with the corresponding 50 amino acid region of the? 9-18: 0-ACP desaturase. The enzyme The resulting catalyst catalyzed the? 6 or? 9 desaturation of both 16: 0-ACP and 18: 0-ACP. Almost identical activity was obtained for Chimera 4, in which a subgroup of 30 amino acids of this domain was transposed (residues 178-207 of the? 9-18: 0- ACP desaturase) to the? 6-16: 0-ACP desaturase. As shown in Figure 1, in stark contrast to the activity of wild-type? 6-16: 0-ACP desaturase, this enzyme it catalyzed the desaturation? 6 and? 9 in a ratio of almost 3: 1 and 1: 1 with 16: 0 -ACP and 18: 0 -ACP, respectively. Moreover, the specific activity with 18: 0 -ACP as a substrate was practically double that detected with 16: 0 -ACP. These results are in stark contrast to the activity of the wild type 6-16: 0 -ACP desaturase. However, this chimeric enzyme is capable of catalyzing the insertion of a double bond in more than one position of 18: 0-ACP, whereas the wild-type? 6-16: 0-ACP desaturase only has detectable activity? S desaturase with 16: 0 -ACP. In addition, the wild type enzyme was approximately 6 times more active with 16: 0 -ACP than with 18: 0 -ACP. To better characterize the 50 amino acid region of the? 9-18: 0-ACP desaturase, a smaller portion of this sequence (residues 178-202) was transposed to the? 6-16: 0-ACP desaturase (Chimera 5) . In contrast to the? 6-16: 0-wild-type desaturase ACP, the specific activity of the resulting enzyme was practically the same with 16: 0- and 18: 0 -ACP. In addition to an extended specificity of fatty acid chain length, the mutant desaturase catalyzed the insertion of a double bond almost exclusively at the? 6 position of 16: 0- and 18: 0 -ACP. Moreover, the specific activity of this enzyme was more than twice as high as that of wild-type? S-16: 0 -ACP desaturase. This may partly reflect the greater stability of the mutant enzyme in E. coli (i.e., that the mutant desaturase was expressed at higher levels and exhibited higher solubility than the wild-type? 6-16: 0-ACP desaturase). The 178-207 region of the? 9-18: 0-ACP desaturase contains nine amino acids that are different from those found in the analogous portion of the? 6-16: 0-ACP desaturase. Through site-directed mutagenesis of the? 6-16: 0-ACP desaturase, each of these residues, either singly or in combination, was converted to the present 9-18: 0-ACP desaturase. An activity qualitatively similar to that of Chimera 4 was obtained by the following mutation of the? 6-16: 0 -ACP desaturase: A181T / A188G / Y189F / S205N / L206T / G207A. (Note: The amino acid numbering is given with respect to the? 9- 18: 0 -ACP desaturase). In addition, the phenotype of Chimera 5 (ie, extended chain length specificity) was achieved qualitatively by the mutation A188G / Y189F of the? S-16: 0-ACP desaturase. Mutant desaturases were also obtained with unexpected activities in these experiments. For example, the mutation A181T / A200F of the? 6-16: 0-ACP desaturase gave rise to an enzyme that catalysed primarily the desaturation? 9 of 18: 0-ACP, but that worked as a? s desaturase with 16: 0 -ACP. The specific activity of this enzyme with 18: 0-ACP, however, was approximately 3 times lower than that detected with 16: 0-ACP. Moreover, the A181T / A200F / S205N / L206T / G207A mutation of the? 6-16: 0 -ACP desaturase gave rise to an enzyme that possessed only? 9 desaturase activity detectable with 18: 0 -ACP and was almost four times more active with this substrate that with 16: 0 -ACP.

The substrate therefore establishes severe restrictions on the length of the aliphatic chain beyond the double bond introduced, which may, in part, explain the differences in specificity for the enzymes in this family. As can be seen, variants of the enzyme that accept substrates with fewer carbon atoms beyond the double bond have their junctional grooves closed by amino acid substitutions with bulky side chains. The amino acids involved in determining the specificity in this part of the binding site are 114-115, 117-118, 179, 181 and 188-189. In the absence of a structural model for the enzyme-substrate-ACP complex, the determinants of chain length specificities on the other side of the double bond, towards the acyl carrier protein, are not so simple to deduce. Assuming that ACP binds in the same way in different enzymes of this type, the differences in the amino acid side chains in the upper part of the substrate channel and in the surface entry of the subunit would allow the enzymes to accommodate different lengths of the alkyl chain between the double bond and the phosphopantein prosthetic group of the ACP. However, the amino acids that line the top of the binding site, from the double bond to the surface of the protein, are conserved in the available enzymatic sequences and it is very likely that determinants of specificity are found at the entrance of the substrate channel. and on the enzymatic surface that interacts with acyl-ACP. Here, the joint pocket becomes wider and it has not been possible to model the phosphopantein part of stearoyl-ACP. Residues 280, 283, 286 and 294 in this area are not conserved between the different enzymes and could be involved in the determination of substrate specificity. By the structure of the binding site in this area, it is possible to rationalize some of the results of chimeras and mutants. All chimeras and mutants include determinant 179-189 (indeed, residues 179, 181, 188-189) and it is therefore not surprising to find effects on specificity. Both Chimera 1 and 2 have very little residual activity, probably due to some steric hindrance to their formation. Chimera 1 has this determinant of the? 9-18: 0-ACP desaturase in the deep pocket and also of the specific surface determinant of the? 9-18: 0-ACP desaturase, only a determi¬ nant, residues 114-115 and 117-118 specific to the? 6-16.-0-ACP desaturase and, therefore, the little remaining activity of this chimera is that of the? 9-18: 0-ACP desaturase Chimera 2 has the complete determinant of the? 9-18: 0-ACP desaturase in the buried pocket area and the known determinant of the? S-16: 0-ACP desaturase at the surface end; this chimera also has activity? 9-18: 0-ACP. Chimera 3 and 4 retain their activity, one of the determinants in the deep pocket is that of? 9-18: 0-ACP, residue A181 replaces the major threonine side chain, but, at the same time, A188 it replaces glycine and Y189 with phenylalanine, really making more space available in the deep cavity and thus allowing even activity? S-18: 0-ACP. Chimera 5 differs from Chimera 4 only in that it has retained the sequence of? 6-16: 0-ACP desaturase for residues 203-207. These residues are in the upper part of the substrate channel, but do not make direct contact with the substrates and it is difficult to understand the effect on substrate specificity. These residues are quite conserved among the known desaturasas of this family; only the? 6-16: 0-ACP desaturase has a different sequence for residue 205 to 207 and this region is probably not part of the natural determinant for substrate specificity. In the case of mutant A181T / A200F, the decrease in activity? S-16: 0-ACP compared to the wild-type enzyme is consistent with structural changes in the substrate channel due to a reduction in the size of this cavity by the change from A181 to threonine. The effect of A200F is not possible to rationalize; this residue is on the surface of the subunit, pointing out of the substrate channel. In all the sequenced desaturases of this family, except in? 6-16: 0-ACP, this residue is a phenylalanine. From the above discussion, it is clear that the activity of A181T / A200F / S205N / L206T / G207A is impossible to explain in structural terms; we can not rationalize the effects of changes in waste 200 and 205-207. Therefore, it has been seen that the region and chain length specificities of the fatty acid desaturase can be changed by specific amino acid substitutions. The determinants for the chain length specificity have the map partially in the region of the three-dimensional structure that defines the shape and size of the substrate-binding channel. However, some of these residues are outside the bonding channel to the substrate and changes in such waste can lead to new and useful activities. With the availability of the three-dimensional structure of the fatty acid ACP desaturase, the fruitful attempts to change the substrate specificities presented here can now be extended to rationally engineered variants of the enzyme having different length and region specificities. However, this will only be successful if, from the crystalline structure of a substrate complex and the availability of multiple amino acid sequences of the enzymes of this family, we can resolve which are the determinants of the specificity at the entrance of the channel of substrate. Material and methods The names of the fatty acids are abbreviated in the format x: ydz, where x is the length of chain or the numbers of carbon atoms in the fatty acid, and is the number of double bonds and z is the position of the double linkage in the fatty acid in relation to the carboxyl end of the molecule (for example, oleic acid or 18: l? 9 is an 18-carbon fatty acid with a double bond, which is located at the ninth carbon atom in relation to the extreme carboxyl of the molecule). Preparation of chimeric mutants Chimeric mutants were prepared by joining portions of the coding sequence of the mature T. ss. 16: 0- and? 9: 18: 0-ACP desaturases through native sites of restriction enzymes or sites of restriction enzymes generated by PCR. Site-specific mutations were introduced into the coding sequence of amino acids 178-202 of the? 9-18: 0-ACP desaturase (equivalent to residues 172-196 of the? S-16: 0-ACP desaturase) by extension and amplification of overlapping oligonucleotide primers using PCR with Pfu polymerase (Stratagene). Mutations A181T / A188G / Y189F were made with the following oligonucleotides: 5'ATGGATCCTGGCACGGATAACAACCCGTAC3 '(Primer AY), 5' ACGAGGTGTAGATAAATCCGAGGTACGGGTTGTTATCCG3 '(Primer 2A), 5' TATCTACACCTCGTATCAGGAGAGGGCGACA3 '(Primer 3A), 5' TTGAATTCCATGGGAAATCGCTGTCGCCCTCTCCTG3 '(Primer 4A). A188G / Y189F mutations were introduced using the following oligonucleotides: • ATGGATCCTGGCGCGGATAACAACCCGTAC3 ' (Primer IB), Primer 2A, Primer 3A, Primer 4A. A181T / A200F mutations were generated with the following: Primer A, 5 'ACGAGGTGTAGATATATGCGAGGTACGGGTTGTTATCCG3' (Primer 2B), Primer 3A, 5 'TTGAATTCCATGGGAAATGAATGTCGCCCTCTCCTG3' (Primer 4B). PCR reactions were carried out without adding template, using 12.5 pmoles of the A or B and 4A or B primers and 6.25 pmoles of the 2A or B and 3A primers. For the first 10 CPR cycles, an annealing temperature of 37 ° C and an extension temperature of 72 ° C were used. This was followed by 20 additional cycles, with the annealing temperature rising to 55 ° C. The products of the PCR reactions were digested with Ba Hl and EcoRI and inserted into the corresponding sites of pBluescript II KS (-) (Stratagene), from which the nucleotide sequence was determined using a Sequenase 2.0 kit (Amersham). This plasmid was then digested with BamHl and EcoRI and the recovered insert was ligated to the coding sequence of amino acids 1-171 of the mature? 6-16: 0-ACP desaturase in the pET3a expression vector (Novagen). The resulting construct (which now contains the coding sequence for amino acids 1-196 of the mutant or wild-type? S-16: 0-ACP desaturase) was restricted with Ncol and EcoRI and ligated to a Ncol / EcoRI fragment containing the sequence encoding the remaining amino acids (residues 197-355) of the? 6-16: 0-ACP desaturase and a portion of the pET3d plasmid (nucleotides). The S205N / L206T / G207A mutation was generated by PCR amplification of the coding sequence of amino acids 197-355 of the? 6-16: 0-ACP desaturase using as template the original cDNA for this enzyme in pBluescript SK (-). The 5 'oligonucleotide (5' TTTCCATGGGAACACGGCTCGGCTAGCGAGGCAGAAGG3 •) contained the appropriate mutant codons and the T7 primer was used as the 3 'oligonucleotide for the PCR reactions. The amplification product was digested with Ncol and Bell and inserted into the Ncol / BamHI site of pET 3d. An Ncol / EcoRI fragment of this construct was then ligated to the coding sequence of amino acids 1-196 of the appropriate mutant? 6-16: 0-ACP desaturase (e.g., A181T / A200F) to generate a full-length coding sequence . The products of the PCR reactions were sequenced to confirm the presence of the desired mutations. Production of acyl-ACP desaturases Acyl-ACP desaturases of wild type and mutants were obtained by expression of the coding sequences in E. coli BL21 (DE3) behind the promoter of the T7 RNA polymerase using the vectors pET3a or pET3d. The recombinant enzymes, whose activities are described in Figure 1, were purified from bacterial cultures of 6 to 9 liters induced at 20 to 25 ° C. Protein purification was performed using DEAE-anion exchange chromatography, followed by 20HS cation exchange chromatography (Perseptive Biosystems), using a Biocad Sprint HPLC (Perseptive Biosystems). Mutant desaturases were obtained at a r-purity of 90% and wild-type? S-16: 0-ACP desaturase was recovered at approximately 80% purity. After purification, the enzymes were changed to a buffer consisting of 40 mM Tris-HCl (pH 7.5), 40 mM NaCl and 10% glycerol and stored in aliquots at -75 ° C after lyophilization in liquid nitrogen . Testing and analysis of acyl-ACP desaturases The acyl-ACP desaturation tests and the analysis of the reaction products were carried out as previously described (Cahoon, E.B. et al., Proc. Nat. Acad. Sci. , USA, 89: 1184 (1994)), with the following modifications: Anaebena recombinant vegetative ferredoxin (22 fg / assay) and corn root FNR (0.4 U / assay) were used in place of ferredoxin and FNR of spinach and the amounts of NADPH and [1-14C} 16: 0- or 18: 0-ACP by assay at 2.5 mM and 178 pmoles (or 1.2 fM), respectively. The ACP used in the synthesis of substrates was recombinant spinach ACP-I. The specific activity of [1-14C} 16: 0 and 18: 0 (American Radiolabeled Chemicals) was 55 mCi / mmol. The enzymatic activity was measured by determining the percentage of monounsaturated product generated in the desaturation tests. The distribution of the radioactivity between the products and the unreacted substrate was measured from phosphor images of CCF separations using an ImageQuant program and by liquid scintillation counting in CCF scrapings. Determination of the positions of the double bonds The positions of the double bonds of monounsaturated fatty acid products were determined by the mobility of the methyl ester derivatives in 15% ARF plaques and by CG-MS analysis of the adducts of dimethyl disulfide of these derivatives. The desaturation assays for the GC-MS analyzes were carried out using 16: 0-, 17: 0- and 18: 0 -ACP not labeled as substrates and purified enzymes. In addition to the results presented in the text, approximately 15% of the desaturation products formed by the reaction of 17: 0-ACP with the wild-type? 6-16: 0-ACP desaturase was detected as isomer 17: l? 7. The rest of the product it was 17: l? 6, also detecting trace quantities of 17: l? 9.

Claims (1)

1. a) having the primary amino acid sequence of the acyl-ACP desaturase; b) aligning the primary amino acid sequence of the acyl-ACP desaturase with the primary amino acid sequence of the? 9 desaturase of Ricinus communis for maximum sequence conservation; c) construct a three-dimensional model for acyl-ACP desaturase based on sequence conservation with the? 9 desaturase of Ricinus communis; d) identifying the contact amino acid residues in the substrate binding channel of the modeled structure in step c), and e) generating a mutant acyl-ACP desaturase having modified chain length and double link position specificities by substituting one or more of the contact amino acid residues identified in step d) with another amino acid residue. 7. An acyl-ACP desaturase mutant that is characterized by the ability to catalyze the desaturation of a first fatty acid and a second fatty acid, the first and second fatty acids differing in their chain length, differing the rates of desaturation of the first and the second fatty acids in no more than about 4 times. 8. The acyl-ACP desaturase mutant of claim 7, which contains a point mutation in a contact amino acid residue in the substrate binding channel. 9. The acyl-ACP desaturase mutant of claim 7, wherein the first fatty acid has a chain length of 16: 0 and the second fatty acid has a chain length of 18: 0. 10. A mutant acyl-ACP desaturase having an amino acid substitution in a contact residue in the attachment-to-substrate channel. 11. The acyl-ACP desaturase mutant of Claim 10, which is characterized by changes in the chain length specificity and position of double bonds compared to the counterpart of wild-type acyl-ACP desaturase. 12. The acyl-ACP desaturase mutant of the Claim 11, wherein the acyl-ACP desaturase is the? 9 acyl-ACP desaturase and the contact residues in the substrate-binding channel are selected from the group consisting of 114, 115, 117, 118, 179, 181, 188 and 189. 13. The mutant of Claim 12, wherein the 9 acyl-ACP desaturase is produced by mutagenesis of cloned nucleic acid from Thunbergia alata or Ricinus communi s. 14. A nucleic acid sequence encoding a mutant acyl-ACP desaturase that is characterized by the ability to catalyze the desaturation of a first fatty acid and a second fatty acid, the first and second fatty acids differing in their chain length, deferring the rates of desaturation of the first and second fatty acids by no more than about 4 times. 15. The nucleic acid sequence of Claim 14, wherein the mutant acyl-ACP desaturase contains a point mutation in a contact amino acid residue in the substrate binding channel. 16. The nucleic acid sequence of the Claim 15, wherein the acyl-ACP desaturase is a? 9 desaturase. 17. The nucleic acid sequence of Claim 16, wherein the contact amino acid residues are selected from the group consisting of residues 114, 115, 117, 118, 179, 181, 188 and 189. 18. A DNA expression construct, consisting, in an expressible form, in a nucleic acid sequence encoding a mutant acyl-ACP desaturase, which is characterized by the ability to catalyze the desaturation of a first fatty acid and a second fatty acid, differing the first and second fatty acids in their chain length, the desaturation rates of the first and second fatty acids differing by no more than about 4 times. 19. The DNA expression construct of Claim 18, wherein the mutant acyl-ACP desaturase contains a point mutation in a contact amino acid residue in the substrate binding channel. 20. The DNA expression construct of the Claim 19, wherein the acyl-ACP desaturase is a? 9 desaturase. 21. The DNA expression construct of Claim 20, wherein the contact amino acid residues are selected from the group consisting of residues 114, 115, 117, 118, 179, 181, 188, and 189. 22. A transformed cell with a DNA expression construct, consisting, in an expressible manner, in a nucleic acid sequence encoding a mutant acyl-ACP desaturase, which is characterized by the ability to laugh the first and second fatty acids in their chain length, deferring the rates of desaturation of the first and second fatty acids by no more than about 4 times. 30. The chimeric acyl-ACP desaturase of claim 29, consisting of? 6-16: 0, wherein amino acids 172-201 are substituted by amino acids 178-207 of? 9-18: 0 -ACP desaturase. 31. The chimeric acyl-ACP desaturase of Claim 29, consisting of? S-16: 0, wherein amino acids 172-196 are substituted by amino acids 178-202 of? 9-18: 0-ACP desaturase. 32. The chimeric acyl-ACP desaturase of claim 29, consisting of? 6-16: 0, wherein amino acids 176, 183, 184, 200, 201 and 202 are substituted by amino acids 181, 188, 189, 205, 206 and 207, respectively, of the? 9-18: 0 -ACP desaturase. 33. The chimeric acyl-ACP desaturase of Claim 29, consisting of? 6-16: 0, wherein amino acids 183 and 184 are substituted by amino acids 188 and 189, respectively, of the? 9-18: 0-ACP desaturase 34. The chimeric acyl-ACP desaturase of Claim 32, consisting of? S-16: 0, wherein amino acids 176 and 195 are substituted by amino acids 181 and 200, respectively, of the? 9-18: 0-ACP desaturase 35. The chimeric acyl-ACP desaturase of Claim 32, consisting of? 6-16: 0, wherein amino acids 176, 195, 200, 201 and 202 are substituted by amino acids 181, 200, 205, 206 and 207, respectively, of the? 9-18: 0-ACP desaturase.