A Pathway Where Polyprenyl Diphosphate Elongates in Prenyltransferase

Prenyltransferases catalyze the consecutive condensations of isopentenyl diphosphate to produce linear polyprenyl diphosphates. Each enzyme forms the final product with a specific chain length. The product specificity of an enzyme is thought to be determined by the structure around the unknown path through which the product elongates in the enzyme. To explore the path, we introduced a few mutations at the 5th, the 8th, and/or the 11th positions before the first aspartate-rich motif of geranylgeranyl-diphosphate synthase or farnesyl-diphosphate synthase. The side chains of these amino acids are situated on the same side of an α-helix. In geranylgeranyl-diphosphate synthase, a single mutated enzyme (F77S) mainly produces a C25 product (Ohnuma, S.-I., Hirooka, K., Hemmi, H., Ishida, C., Ohto, C., and Nishino, T. (1996)J. Biol. Chem. 271, 18831–18837). A double mutated enzyme (L74G and F77G) mainly produces a C35 compound with significant amounts of C30 and C40. A triple mutated enzyme (I71G, L74G, and F77G) mainly produces a C40compound with C35 and C45. Mutated farnesyl-diphosphate synthases also show similar patterns. These findings indicate that the elongating product passages on a surface of the side chains of the mutated amino acids, the original bulky amino acids had blocked the elongation, and the path is conserved in prenyltransferases. Moreover, the fact that some double and triple mutated enzymes can also form small amounts of products longer than C50 indicates that the paths in these mutated enzymes can partially access the outer surface of the enzymes.

Prenyltransferases, also referred to as polyprenyl-diphosphate synthases, are indispensable for biosyntheses of more than 20,000 naturally occurring isoprenoids and constitute a broad family of enzymes that catalyze the sequential condensations of isopentenyl diphosphate (IPP, C 5 ) 1 with allylic pre-nyl diphosphates (1). These enzymes are classified into two groups according to the stereochemistry of the E or Z double bond that is formed by the condensation. Although organisms use Z-polyprenyl-diphosphate synthases only for the synthesis of dolichols for N-linked glycoprotein biosynthesis, Z-polyprenols for peptidoglycan biosynthesis in bacteria, and natural rubber, E-polyprenyl-diphosphate synthases are used for the synthesis of a vast variety of important natural isoprenoids ( Fig. 1). A number of enzymes that yield (all-E)-prenyl diphosphate have been isolated from various organisms. Geranyldiphosphate (GPP, C 10 ) synthase, found in plants (2), catalyzes the single condensation of IPP with dimethylallyl diphosphate (DMAPP, C 5 ) to give GPP. The GPP is the precursor of all monoterpenes. Farnesyl-diphosphate (FPP, C 15 ) synthase, which is one of the key enzymes of the biosynthesis of steroids, cholesterol, farnesylated proteins, sesquiterpenes, and so on, catalyzes the consecutive condensation of two molecules of IPP with DMAPP to give FPP as the ultimate product. Geranylgeranyl-diphosphate (GGPP, C 20 ) synthases are thought to be classified in two groups. One type uses DMAPP, GPP, or FPP as the allylic substrate (plant, bacteria, and archaea), and the other utilizes FPP as a substrate (fungi and animals) to give an amphiphilic molecule containing four isoprene units, GGPP, which is a precursor of diterpenes, carotenoids, geranylgeranylated proteins, and ether-linked lipids of archaea. Long E-polyprenyl-diphosphate synthases producing compounds with chain lengths from C 30 to C 50 are involved in respiratory quinone biosynthesis. Chicle and Gutta-percha are high molecular weight compounds like a natural rubber (3,4). These compounds are assumed to be synthesized in plants by specific E-polyprenyl-diphosphate synthases.
In both Sulfolobus acidocaldarius GGPP synthase and Bacillus stearothermophilus FPP synthase, we have found that the fifth amino acid before the first aspartate-rich motif (FARM) is extremely important for the ultimate chain length determination (21,22). From the analysis of the amino acid, we demonstrated that the terminus of an elongating allylic product directly contacts with this amino acid residue, and the interaction must prevent further condensation of IPP (23).
Although we examined various FPP synthases and GGPP synthases mutated at the 5th position before FARM, there was no mutated enzyme that could yield any products longer than C 30 (21)(22)(23). Therefore, we hypothesized that another amino acid upstream from FARM must block the further elongation and that the alteration of this amino acid would be essential to create from GGPP synthase and FPP synthase an enzyme that produces compounds longer than C 30 . To demonstrate our hypothesis, we introduced mutations in the upstream region from FARM of S. acidocaldarius GGPP synthase and B. stearothermophilus FPP synthase and analyzed their products. This information should elucidate a path through which products formed by long E-polyprenyl-diphosphate synthases elongate and might elucidate a common mechanism of chain termination of long E-polyprenyl-diphosphate synthases. Furthermore, this information might enable investigators to design of a "no stop" prenyltransferase.
Construction of Mutated GGPP Synthases-Site-directed mutagenesis was carried out by the method of Kunkel (25). For the construction of pGGPS-L1, pGGPS-L2, pGGPS-L3, and pGGPS-L4, a single-strand DNA generated from pBS-GGPSmut3 (22) having the gene for the mutated S. acidocaldarius GGPP synthase (F77S) and synthetic oligonucleotides described below were used as follows: oligo-V73A, 5Ј-GGT-GCAGCAATTGAAGCTCTTCATAC-3Ј; oligo-V73G, 5Ј-GGTGCAGCA-ATTGAAGGTCTTCATAC-3Ј; oligo-L74A, 5Ј-GGTGCAGCAATTGAAG-TTGCTCATAC-3Ј; oligo-L74G, 5Ј-GGTGCAGCAATTGAAGTTGGTC-ATAC-3Ј. Mutated nucleotides are indicated by bold letters. Every primer can introduce an additional restriction site of silent mutation, which is indicated by the underline. pGGPS-L5 was constructed using a single-strand DNA template from pBS-GGPS having the gene for the wild type GGPP synthase (13) and a synthetic oligonucleotide of oligo-V74G. For the construction of pGGPS-L6 and pGGPS-L7, a singlestrand DNA template was isolated from Escherichia coli CJ236 containing pGGPS-L4, and then the second mutagenesis was carried out FIG. 1. Consecutive condensation of isopentenyl diphosphates by prenyltransferases. Wild type GGPP synthase of S. acidocaldarius and wild type FPP synthase of B. stearothermophilus catalyzes consecutive condensations of IPP to produce GGPP and FPP as the ultimate product, respectively. Other products longer than GGPP are formed by the specific prenyltransferases such as HexPP synthase and solanesyl-diphosphate synthase.
Preparation of Mutated GGPP Synthases and Measurement of Their Enzyme Activities-E. coli XL1-Blue was transformed with the plasmids containing mutated GGPP synthase genes and cultured according to the method described previously (26). The mutated GGPP synthases were prepared according to the previously reported method (22). In the procedures, we had used heat treatment to eliminate a majority of thermolabile endogenous protein. Thus, we confirmed that all mutated enzymes are still thermostable ( Table I). The assay mixture contained, in a final volume of 1 ml, 25 nmol of [1-14 C]IPP (37 GBq/mol), 25 nmol of the indicated allylic substrate (DMAPP, GPP, (all-E)-FPP, (all-E)-GGPP), 5 mol of MgCl 2 , 10 mol of phosphate buffer (pH 5.8), and the same amount of indicated enzyme. This mixture was incubated at 55°C for 15 min, and the reaction was stopped by chilling quickly in an ice bath. The mixture was shaken with 3.5 ml of 1-butanol that had been saturated with H 2 O. The 1-butanol layer was washed with water saturated with NaCl, and the radioactivity in the 1-butanol layer was determined with a liquid scintillation counter.
Construction of Mutated FPP Synthases-The plasmid encoding FPS-F1 was constructed by polymerase chain reaction method using pEX11 (8) as the template and four primers as follows: primer 1, 5Ј-ATCATCATGAATCAAAGAGCCCGTATGGCCCATTTC-3Ј; primer 2, 5Ј-AAACAGACCATGGCGCTTTTC-3Ј; primer 3, 5Ј-CAGCCAAGCT-TTTAATGGTC-3Ј; primer 4, 5Ј-CTTTGATTCATGATGATTTG-3Ј. Mutated nucleotides are indicated by bold letters, and the restriction sites are indicated by an underline. An upstream fragment and a downstream fragment were amplified using a combination of primers 1 and 2 and of primers 3 and 4, respectively. The upstream fragment obtained from the digestion with NcoI and BspHI and the downstream fragment obtained from the digestion with BspHI and HindIII were ligated back into the NcoI and HindIII sites of the original pEX11 plasmid. For the construction of the plasmid encoding FPS-F2, we used two primers as follow: primer 5, 5Ј-ATGGACAACGATGATTTGCG-3Ј and primer 6, 5Ј-CAAATCATCATGGATCAAAG-3Ј. Polymerase chain reaction was carried out using these primers and the plasmid encoding FPS-F1 as the template. The plasmid encoding FPS-F2 was formed by self-ligation of the blunt-ended amplified fragment. We have also confirmed the mutagenesis by the same way as the mutated GGPP synthase.
Preparation of Mutated FPP Synthases and Measurement of Their Enzyme Activities-E. coli XL1-Blue was transformed with the plasmids encoding mutated FPP synthases and cultured according to the method described previously (26). The mutated FPP synthases were prepared according to the previously reported method (8,21,23). The assay mixture contained, in a final volume of 200 l, 25 nmol of [1-14 C]IPP (37 GBq/mol), 25 nmol of the indicated allylic substrate (DMAPP, GPP, (all-E)-FPP, (all-E)-GGPP), 1 mol of MgCl 2 , 10 mol of NH 4 Cl, 10 mol of 2-mercaptoethanol, 10 mol of Tris-HCl buffer (pH 8.5), and a suitable amount of enzyme. After the incubation at 55°C for 20 min, the activity was determined by the 1-butanol extraction method described above.
Product Analysis-After the enzymatic reaction, the polyprenyl diphosphates were extracted with 1-butanol, and then the 1-butanol was evaporated under a N 2 stream. The resulting polyprenyl diphosphates were treated with acid phosphatase according to the method of Fujii et al. (27). The hydrolysates were extracted with n-pentane and analyzed by reversed-phase thin layer chromatography using LKC-18 developed with acetone/H 2 O (9:1). To determine the chain length of long products, polyprenyl diphosphates extracted from 10 ml of the reaction mixture were treated with the similar method and analyzed by reversed-phase thin layer chromatography using HPTLC RP-18 developed with acetone/hexane (19:1). Authentic standard alcohols were visualized with iodine vapor, and the absolute radioactivities of the spots were measured with a Bio-image analyzer BAS2000 (Fuji). Each amount of an allylic product was determined by division of the absolute radioactivity by the number of IPPs incorporated into the product.

FIG. 2. Deduced amino acid sequences of mutated archaeal GGPP synthases.
Amino acid sequences before the first aspartate-rich motif of the wild type and the mutated GGPP synthases from S. acidocaldarius are shown. The mutated amino acid residues are indicated as boxed bold letters. The first aspartate-rich motifs are boxed. Mutant 3 was reported in our previous paper (22).

TABLE I Prenyltransferase activity of the mutated GGPP synthase
Each name of the mutated GGPP synthases is shown with the carbon number of the main long chain product and the amino acid replacement(s) in parentheses. Enzyme activities were measured as described under "Experimental Procedures." These activities were determined from the amount of radioactivity incorporated into 1-butanol-extractable materials when 25 M [1-14 C]IPP and 25 M of the indicated allylic substrate were used. 2nd column indicates the specific activity obtained from the reaction using GPP as an allylic substrate. Columns 3-6 indicate the relative activity of DMAPP, GPP, FPP, or GGPP, which is based on GPP activity. Thermostability indicates the remaining activity after 7 h of heat treatment of enzymes at 70°C. All figures show mean Ϯ S.D. (n ϭ 3).

RESULTS
Design of Double Mutated GGPP Synthases-In our previous study, we made two mutated GGPP synthases that have the replacement of phenylalanine by serine or leucine at position 77, which is the 5th position before FARM (22). These two artificial enzymes can mainly produce geranylfarnesyl diphosphate (GFPP, C 25 ) with a small amount of HexPP (C 30 ). However, we failed to create an enzyme yielding a product longer than C 30 . We have also reported that, in B. stearothermophilus FPP synthase, the replacement of tyrosine at the corresponding position causes a similar change in the ultimate product and that the amino acid at this position seems to contact directly with the terminus of FPP (21,23). Based on the fact that the mutated GGPP synthases did not yield long products, we have hypothesized that an amino acid that is situated 1 isoprene unit distant from the above mutated site directly blocks the further condensation.
Tarshis et al. (28) have determined the three-dimensional structure of avian FPP synthase, which shows a considerable amino acid sequence similarity to S. acidocaldarius GGPP synthase and B. stearothermophilus FPP synthase. In the structure, the region around FARM forms an ␣-helix. The average number of residues is 3.6 per turn in an ␣-helix, which corresponds to 5.4 Å. This helix pitch is almost the same as the length of 1 isoprene unit of ϳ5.1 Å, which was calculated based on the length of carbon bonds. Thus, assuming that a growing polyprenyl diphosphate elongates along the ␣-helix and that the elimination of the blocking side chain of the amino acid situated at one pitch of ␣-helix distant from the fifth amino acid before FARM enables the mutated enzyme to catalyze the synthesis of products longer than C 30 , we constructed several double mutated GGPP synthases; the first mutation is at the fifth amino acid before FARM, the second mutation is at the third or the fourth amino acid before the first mutation (Fig. 2).
If one of the double mutated GGPP synthases can produce polyprenyl diphosphate longer than C 30 , it will be strong evidence for our hypothesis and will indicate that the elongating product directs to the side chain of the second mutated amino acid.
Characterization of Double Mutated GGPP Synthases-Cells of E. coli XL1-Blue were transformed with the plasmids encoding the double mutated GGPP synthases, and the mutated enzymes were prepared according to the method reported previously. The specific activities of the mutated enzymes were determined by measuring the radioactivity in 1-butanol extractable materials of the reaction using DMAPP, GPP, FPP, or GGPP as the primer substrate (Table I). When assayed using GPP, all enzymes except for GGPS-L7 and GGPS-L4 showed comparable prenyltransferase activities. All mutated GGPP synthases and the wild type enzyme had similar relative activities for DMAPP, GPP, and FPP. On the other hand, the relative activities for GGPP based on GPP activities vary from 1.7 to 44%. The wild type enzyme hardly accepts GGPP as the primer substrate. The activity of GGPS-L5 for GGPP is also low, suggesting that the single mutation at position 74 is not enough to change the substrate specificity much. The relative activities of GGPS-L1, GGPS-L2, and GGPS-L3 are very similar to that of mutant 3 concerning total radioactive products, whereas the relative activities for GGPP of GGPS-L4, GGPS-L6, and GGPS-L7 are higher than those of GGPS-L1, GGPS-L2, and GGPS-L3. These results suggest that the mutation at position 73 has no effect on the chain length of product, whereas the leucine to glycine mutation at position 74 for mutants GGPS-L4, -6, and -7 is effective in changing the product specificity in such a way that L74G yields longer products than does mutant 3. To confirm these observations, the reaction products were analyzed.
Determination of the Chain Lengths of Products by the Dou- ble Mutated GGPP Synthases-After treatment of 1-butanol extractable products with acid phosphatase, the chain length distribution of the resultant polyisoprenols and their amounts were analyzed by reversed-phase TLC. Prenyltransferases are supposed to utilize the intermediates that are initially formed and released in the reaction solution as the primer substrates again. To determine the chain length of the initially formed products as much as possible, the prenyltransferase reaction was terminated before 10% of the substrates were consumed. When the enzyme reaction was carried out using 25 M [1-14 C]IPP and 25 M DMAPP, the wild type GGPP synthase mainly produced GGPP and did not give any products longer than GGPP (Fig. 3A). Mutant 3, which has only a mutation of F77S, mainly produced GFPP with small amounts of FPP, GGPP, and HexPP. The product distribution was similar to that previously reported (22). The product distribution did not change when the second substitutions at the position 73 were introduced as shown in the results of GGPS-L1 and GGPS-L2. On the other hand, GGPS-L3 gave a small amount of HepPP although the main product is still C 25 . GGPS-L4 gave C 30 product as the main product in the long chain region (ϾC 20 ), with a large amount of HepPP. Moreover, GGPS-L4 can form small amounts of several polyprenyl diphosphates longer than C 35 . These data show that the second mutation at position 74 can influence the chain length determination. Furthermore, they show that the mutated enzyme with a smaller amino acid at this position tends to produce longer products (Leu-74 Ͻ Ala-74 Ͻ Gly-74) (Fig. 3A).
We have reported in a previous paper (23) that, in the case of B. stearothermophilus FPP synthase, the chain length of product was dramatically changed by the replacement of the amino acid at the fifth amino acid before FARM and that the average chain length of products is almost inversely proportional to the accessible surface area of the substituted amino acid. Moreover, we also observed similar phenomena in S. acidocaldarius GGPP synthase. Main products of single mutated enzymes of F77S and F77G are C 25 and C 30 , respectively (data not shown). By analogy of these results, we have constructed two more additional artificial GGPP synthases, GGPS-L6 and GGPS-L7, each of which has the replacement with a small amino acid, alanine or glycine, at position 77 in addition to L74G. When the assay was carried out using DMAPP as the primer substrate, the relative amount of HexPP formed in the reactions by GGPS-L6 is about twice greater than that by GGPS-L4. GGPS-L7 mainly forms HepPP (C 35 ) regarding the products over C 20 , indicating that the bulk of the side chain of the first mutated amino acid still affects the chain length of product (Ser-77 Ͻ Ala-77 Ͻ Gly-77). GGPS-L5, which has only one replacement at position 74 (L74G), could give a significant amount of GFPP although the relative activity for GGPP was very low (Fig. 3A, Table I).
Similar product distributions were observed even when GPP, FPP, or GGPP was used as the allylic substrate (Fig. 3, B-D).
Construction of Triple Mutated GGPP Synthases and Their Product Analysis-In the previous section, we demonstrated that the double mutated enzymes can form the products longer than those formed by the single mutated enzyme. However, these enzymes could not produce polyprenyl diphosphate longer than C 35 as the main product, although small amounts of very long products were formed as the minor products. These results indicate that some amino acid around the exit of polyprenyl diphosphate still disturbs the complete effluence of the product, suggesting that improvement around this region might confer a complete no stop enzyme that produces long polyprenyl diphosphates such as Chicle and Gutta-percha. Thus, we made four triple mutated GGPP synthases that have the third mutation at position 71 or 72 in addition to the mutations at position 74 and position 77 (Fig. 2), and we analyzed the products (Fig. 4). The third mutation is situated one pitch of ␣-helix distant from the second mutated position 74. When the assay was carried out using 25 M DMAPP and 25 M IPP, triple mutated GGPS-L10 (I71G, L74G, and F77G) increased the amount of OPP relative to HepPP, and then OPP became the major product in the products over C 20 (OPP/ HepPP ϭ 2.3), which is 1 isoprene unit longer than that of the corresponding double mutated GGPS-L7 (OPP/HepPP ϭ 0.34). Moreover, the other triple mutated enzyme, GGPS-L8 (I71G, L74G, F77S), increased the ratio of HepPP to HexPP (HepPP/ HexPP ϭ 2.6), compared with that of the corresponding double mutated GGPS-L4 (HepPP/HexPP ϭ 0.29). HepPP was the major long product of GGPS-L8, and it is 1 isoprene unit longer  6. Deduced amino acid sequences of mutated bacterial FPP synthases and their product analysis. A, amino acid sequences before the first aspartate-rich motif of wild type and the mutated FPP synthases from B. stearothermophilus are shown. Y81G was reported in our previous paper (23). B, the sample from incubation of [1-14 C]IPP with the indicated enzyme and the indicated allylic primer substrate was analyzed by reversed-phase LKC-18 TLC as described under "Experimental Procedures." The carbon numbers of the products were indicated at left. ori., origin; s.f., solvent front. than that of GGPS-L4. On the other hand, in the triple mutated GGPS-L9 and GGPS-L11, there was no dramatic change of product distribution, although the ratio of products over C 25 was slightly increased compared with the corresponding double mutated GGPS-L4 and GGPS-L7 (Fig. 4A). These results indicate that the amino acid at 71 can also interact with the elongating products. Similar product distributions were observed when GPP, FPP, or GGPP was used as the primer substrate (Fig. 4, B--D).
Analysis of Products Longer Than C 35 -As shown in Figs. 3 and 4, the double and triple mutated GGPP synthases, GGPS-L3, GGPS-L4, GGPS-L6, GGPS-L7, GGPS-L8, GGPS-L9, GGPS-L10, and GGPS-L11, can produce extremely long polyprenyl diphosphates, and the ladder of the products yielded in the reactions by some mutated enzymes appears to continue to the origin. We tried to confirm the longer limit of the product that can be synthesized by the mutated enzymes as follows. The reaction was carried out using a large volume incubation, and the products were analyzed by using a different TLC system (Fig. 5). As a result, these eight mutated GGPP synthases were found to be able to yield the products longer than C 35 , but the longest products were different. To our surprise, the longest detectable products of GGPS-L3, GGPS-L4, GGPS-L6, GGPS-L7, GGPS-L8, GGPS-L9, GGPS-L10, and GGPS-L11 were C 65 , C 120 , C 105 , C 110 , C 90 , C 80 , C 80 , and C 70 , respectively, suggesting that the structure around the modified region affects the production of the long products, and the paths in these mutated enzymes have partially accessed the outer surfaces of the enzymes.
Product Analysis of Mutated FPP Synthases-In the previously sections, we described that the double and triple mutated GGPP synthases can catalyze the synthesis of polyprenyl diphosphates much longer than those yielded by single mutated enzymes. Is this effect common to all prenyltransferases? To answer this question, we constructed two kinds of double mutated bacterial FPP synthases, FPS-F1 and FPS-F2 (Fig.  6A). FPS-F1 has two replacements at positions 78 and 81, I78G and Y81G, and this corresponds to GGPS-L7. FPS-F2 has a deletion of two amino acids in FARM in addition to the two mutations of I78G and Y81G. This mutation was designed to compare eukaryotic FPP synthases and bacterial FPP synthases, because the bacterial enzymes have an extra insertion of two amino acids in this region, which affects the product specificity (29). These mutated enzymes produced the polyprenyl diphosphates much longer than those of the wild type and a single-mutated FPP synthase, Y81G (Fig. 6B). FPS-F1 mainly gave significant amounts of products ranging from C 40 to C 65 in addition to three major products, GFPP, HexPP, and HepPP. When DMAPP was used as the primer, the relative amounts of the products were as follows: GPP (2.4%), FPP (10%), GGPP (3.7%), GFPP (15%), HexPP (44%), HepPP (16%), OPP (4.3%), C 45 PP (1.9%), C 50 PP (0.68%), and longer products (1.7%). The product distributions of FPS-F1 and FPS-F2 are almost similar to that of GGPS-L6, indicating that bacterial FPP synthase and archaeal GGPP synthase have similar structures before FARM and that the chain lengths of the products of mutated enzymes are regulated by a common mechanism. The similarity of the product distributions of FPS-F1 and FPS-F2 indicates that the extra insertion of two amino acids in FARM does not cause a significant change in the structure of the upstream region. Moreover, the ratios of longer products of the mutated FPP synthases were slightly increased as the enzyme concentration was increased. Similar observations were made in the case of mutated GGPP synthases. Ultimately, these mutated FPP synthases can catalyze the formation of up to C 65 compound although the main product is still C 30 .
Further Improvement of GGPP Synthase-We have succeeded in converting short chain prenyl-diphosphate synthases such as GGPP synthase and FPP synthase to long chain prenyl diphosphate synthases. However, almost all of these enzymes also formed significant amounts of short products. Many prenyltransferases that catalyze formation of long polyprenyl diphosphates, including undecaprenyl diphosphate, solanesyl diphosphate, and natural rubber, require additional factors such as detergents, carrier proteins, and latex molecules (30 -35). These cofactors seem to work to remove the long chain product from the enzyme effectively. Therefore, we investigated the effect of detergent on the ratio of long products. The activities of wild type and all mutated GGPP synthases were enhanced several times by the addition of 2.5% of Triton X-100, indicating that the products of these enzymes were removed efficiently. However, no change of the product distribution was observed (data not shown). Other detergents, such as Tween 80 and CHAPS, did not have any effect on the chain length of products either.
Moreover, our previous study showed that, in some prenyltransferases, the product distribution depends on the concentration of the substrates. At a low concentration of IPP or Mg 2ϩ ions, solanesyl-diphosphate synthase mainly produces short chain products of FPP and GGPP with significant amounts of C 35 and C 40 , but the amounts of intermediate products are low (24). This product distribution is similar to those of our mutated enzymes. When an allylic substrate is decreased or IPP is increased, the prenyltransferases increase the ratio of long products (24,29). Thus, we examined whether these mutated GGPP synthases produced larger amounts of long products under the lower concentration of an allylic substrate with the addition of detergent. When the reaction was carried out with 2.5 M GPP and 25 M IPP in the presence of 2.5% Triton X-100, the main product of every mutated GGPP synthase was shifted to longer chain region (C 25 to C 40 ) although the chain length of final product was not extended. The main products of these mutated GGPP synthases were FPP except for GGPS-L9, which produced GFPP as the main product, in the reaction with L9, GGPS-L10, and GGPS-L11 were GFPP (73%), HexPP (41%), HexPP (65%), HepPP (29%), HepPP (28%), HexPP (59%), OPP (31%), and HexPP (34%), respectively, in reaction with 25 M IPP, 2.5 M GPP, and 2.5% Triton X-100 (Fig. 7). These results indicate that product distribution is affected by the substrate concentration, but cofactors or substrate concentration is not important for determination of the ultimate chain length of the products. These characters of the mutated GGPP synthases are very similar to those of natural long chain Epolyprenyl-diphosphate synthases such as solanesyl-diphosphate synthase (24). DISCUSSION Chain Elongation Mechanism of Prenyltransferases-On the basis of our present results and the three-dimensional structure of avian FPP synthase (28), we postulate a common mechanism of chain length determination in E-prenyltransferase reactions (Fig. 8). In all wild type FPP synthases or GGPP synthases, the bulky amino acid residue at the 5th position before FARM directly blocks the elongation beyond FPP or GGPP (Fig. 8A). The difference between the product specifici-ties of FPP synthase and GGPP synthase is mainly given by the difference of the fourth amino acid before FARM and the insertion of two amino acids in the motif (29).
In the case of the single mutated GGPP synthases and FPP synthases described in Fig. 8B, the substituted amino acid at the fifth amino acid before FARM allows the mutated enzymes to catalyze the consecutive condensations of IPP beyond the native chain length of the product. An amino acid at this position must be biologically significant in the chain length determination of all natural E-prenyltransferases because all reported E-prenyltransferases that catalyze the condensation beyond GGPP have alanine at this position without exception (15)(16)(17)(18)(19).
The double mutated GGPP synthases, GGPS-L3, GGPS-L4, and GGPS-L6, can produce C 30 as the main long product, which is 1 unit longer than that of the corresponding single mutated enzyme, demonstrating that, in single mutated enzymes, the side chain of Leu-74 interferes with further elongation of their products. GGPS-L7 produced C 35 compound as the main long product, indicating that the bulk of the first mutated amino acid at position 77 still affects the product length of the double mutated enzyme. Moreover, these enzymes also can catalyze further condensations of IPP with GFPP to produce longer products (C 50 to C 110 ). In the case of avian FPP synthase, the amino acid residues from position 94 to position 119 form an ␣-helix, and the region from the bottom of the ␣-helix to the first aspartate of FARM contains about 7 pitches of ␣-helix. This distance corresponds to the chain length of the hydrocarbon moiety of solanesyl diphosphate (C 45 ). Thus, these data indicate that the very long products have already protruded from the enzymes (Fig. 8C).
The triple mutated GGPP synthases of GGPS-L8 and GGPS-L10 mainly produced C 35 and C 40 compounds regarding the long chain products, respectively, indicating that the isoleucine 71 also interacts with the elongating hydrocarbon region (Fig.  8D). Also, these data indicate that the first replaced amino acid still affects the chain length.
In this paper, we assigned the three amino acids that are situated in the pathway and directly interact with the elongating product, and we demonstrated that the replacement of these amino acids causes dramatic change of product chain length. Although we obtained the enzymes that can produce longer products than C 100 , the efficiency is still low. The further modification around the path through which products pass might improve the efficiency and produce real no stop prenyltransferase.
Our results also suggest that the product specificities of medium chain polyprenyl-diphosphate synthases, HexPP (C 30 ) synthase (36), HepPP (C 35 ) synthase (37), OPP (C 40 ) synthase (38), solanesyl-diphosphate (C 45 ) synthase (24,39), and decaprenyl-diphosphate (C 50 ) synthase (40), must be determined by the overall structure around the region we modulated in this study (Fig. 8). On the other hand, there is a difference between these natural enzymes and our mutated enzymes although the product specificities are similar. The natural enzymes cannot accept a short chain allylic diphosphate, DMAPP, as the allylic substrate, whereas our mutated enzymes can utilize DMAPP with almost the same reactivity as GPP. Some difference of the structure of a putative allylic substrate binding region (41) might be responsible for the substrate specificity because our group recently found that some mutations in the putative allylic substrate binding region of GGPP synthases from S. acidocaldarius cause a change of allylic substrate specificity without change of the product specificity (42).
Finally, Tarshis et al. (43) has recently found that a double mutated avian FPP synthase that has two replacements with small amino acids at the 5th and the 4th positions before FARM yields mainly polyprenyl diphosphates ranging from C 30 to C 45 with small amounts of products longer than C 45 . Based on the crystal structure, they have demonstrated that the growing allylic product directly interacts with the mutated amino acids. Although the second mutated position is different from those of our mutated enzymes, these results strongly support our mechanism and indicate that the structure of eukaryotic FPP synthase around the upstream region is slightly different from those of the bacterial FPP synthase and archaeal GGPP synthase.