Bioinformatics-Aided Identification, Characterization of Fungal Linalool Synthases and Applications in Linalool biosynthesis

: Enzymes empower chemical industries and are the keystone for metabolic engineering. For example, linalool synthases (LSs) are indispensable for the biosynthesis of linalool, an important fragrance used in 60-80% cosmetic and personal care products. However, plant LSs have low activities while expressed in microbes. Aided by bioinformatics analysis, four linal-ool/nerolidol synthases (LNSs) from various Agaricomycetes were accurately predicted and validated experimentally. Further-more, we discovered a novel LS with exceptionally high levels of selectivity and activity from Agrocybe pediades , ideal for linalool bioproduction. It effectively converted glucose into enantiopure (R)-linalool in Escherichia coli , 44-fold and 287-fold more efficient than bacterial and plant counterparts, respectively. Phylogenetic analysis indicated the divergent evolution paths for plant, bacterial and fungal LSs. More critically, structural comparison provided catalytic insights in Ap.LS superior specificity and activity, and mutational experiments validated the key residues responsible for the specificity. 1N1B/1N21) 26 using Modeller software. The binding pockets, consisting of 21-22 residues within 6 angstroms from the substrate, deter- mined by PyMOL software v2.1.1. Structural alignment of proteins was conducted using MUSTANG 27 implemented in YASARA 256 28 . The web implementation of PLIP (Protein-Ligand Interaction Profiler) was used to identify interactions between ligand and 257 the surrounding amino acid residues in the protein 29 . Docking was performed with AutoDock Vina an exhaustiveness PDBQT files homology Ap.LS, Ap.LNS, Sc.LNS Steady-state kinetics of purified Aa.LNS and Ap.LS were determined by measuring PPi release via conversion to phosphate with inorganic pyrophosphatase using the EnzChek® Pyrophosphate Assay Kit (Thermo Fisher Scientific, Singapore). Substrate concentrations of GPP and (E,E)-FPP (Echelon bioscience, USA) were varied between 6.5-40 μM. Reactions were carried with 298 20 μg/ml (460 nM) of Ap.LNS or 10 μg/ml (250 nM) of Aa.LS at 37 °C for 1 hour. PPi concentrations were calculated by linear 299 interpolation of the standard curve using the kit assay (0-60 μM). The scatter plots of initial rate versus substrate concentration 300 were fitted to the equation v=v max [ S ]/( K M +[ S ]) where v max = k cat [E 0 ]. 301


INTRODUCTION
Nature is the best inventor and breeds versatile enzymes. Among various enzymes, terpene synthases represent a unique 2 class of biocatalysts with fascinating capabilities (e.g. introduction of carbon-carbon bonds, facilitation of cyclization and rear-3 rangement of terpenes) 1 . Terpene synthases are pivotal for the biosynthesis and diversity of terpenoids (>80000 different 4 molecules), which constitute the largest group of natural products 1 . Terpenoids have wide applications, including pharmaceu-5 ticals, nutraceuticals, flavorings, fragrances, and biofuels 2,3 . However, the biosynthesis of most terpenoids has yet to achieve 6 high titers and yields that are vital for commercial production. One major obstacle is that currently identified terpene synthases 7 have low activities and/or low selectivities 2,4 . 8 An example is linalool, a naturally occurring monoterpene alcohol (C10) found in several flowers such as lavender 5 . With a 9 pleasant floral smell, linalool is an important fragrance ingredient widely used in food, beverage and many personal care prod-10 ucts (perfumes, body lotions, etc.). Natural linalool has two stereoisomers with different smells, (S)-linalool and (R)-linalool. (S)-11 linalool is floral, citrus and petitgrain-like (odor threshold 7.4 ppb) and (R)-linalool is woody and lavender-like (odor threshold 0.8 12 ppb) 6 . Natural linalool has higher enantiopurity, thus, superior to synthetic linalool racemates in applications such as high-end 13 perfumes and cosmetics. In 2018, the world consumption of linalool surpassed 11,000 metric tons and its global market is 14 projected to reach 12.3 billion US$ in 2024 7 . Despite great commercial interests, the biosynthesis of linalool has only achieved 15 limited success (mg/L scale) 8,9 . This contrasts with the rapidly growing demand for natural linalool. The plant linalool synthases 16 (LSs, converting geranyl pyrophosphate (GPP) into linalool) are relatively abundant, yet proven to have low activities when 17 expressed in microbial hosts (e.g. yeasts and Escherichia coli) 10 . Recently, a bacterial bifunctional linalool/nerolidol synthase 18 (LNS) has been identified and characterized 11 . However, it produces more nerolidol (a sesquiterpene alcohol, the product of 19 farnesyl pyrophosphate or FPP) than linalool when expressed in microbes. This is because, unlike plants that have special 20 compartments (e.g. plastids) where GPP synthases (GPPSs) are localized 12 , wildtype microbes have neither specialized orga-21 nelles nor dedicated GPPSs. Rather, GPP is merely an intermediate compound of FPP synthases (e.g. ispA of E. coli, ERG20 22 of Saccharomyces cerevisiae) in microbes. Hence, FPP is more abundant than GPP in the cytosol 13 . As such, a specific and 23 more active LS is desired for microbial linalool production. Besides plants and bacteria, fungi such as the agaric mushroom 24 Agrocybe aegerita (recently renamed into Cyclocybe aegerita), are also known to produce linalool and several other monoter-25 penes (a-pinene, p-cymene, limonene, and b-ocimene) 14 . However, fungal LSs have not been identified until very recently. In 26 our previous study 9 , we described the bifunctional role of an LNS from A. aegerita that was used to produce linalool or nerolidol 27 in E. coli but did not elaborate on the method of identification or its kinetic characterization. 28 Here guided by bioinformatics predictions and experimental validation, we identified four bifunctional LNSs and a specific LS 29 (no sesquiterpene activity) from four fungal species. We characterized the kinetic parameters of the LNS from A. aegerita and 30 produced various sesquiterpenes but not linalool 3 . A re-evaluation of the raw genomic data led to an additional putative terpene 48 synthase sequence (AAE3_109435). The gene exists in the Illumina sequencing data, whereas absent in the PacBio results. 49 PCR amplification of the AAE3_109435 with subsequent sequencing confirmed the presence of the gene in the genome (Fig.  50   S1). 51 The subsequent expression of AAE3_109435 in a GPP-accumulating E. coli that co-expressed the native enzymes DXS, IDI 52 and ispA_S80F mutant (GPPS). The resulting strain (GPPS+9435) produced the acyclic monoterpene linalool as the main 53 product and small amount of nerolidol ( Fig. 1; mass spectra in Fig. S2). Geraniol was detected in GPPS+9435 as well as in its 54 control strain (GPPS_ctrl), indicating that geraniol was not the product of AAE3_109435 but of a native E. coli enzyme (such as 55 PhoA, a phosphatase, and NudB, a Nudix hydrolase 22 ). Furthermore, AAE3_109435 was expressed in the E. coli strain that 56 accumulates FPP by overexpressing DXS and IDI (without ispA_S80F), and the strain was named 'FPPS+9435'. Its main prod-57 uct (>96% regarding the peak area of all detected terpenes) was nerolidol and only traces of linalool could be detected (Fig.  58 1A). The control strain (FPPS_ctrl) with an empty vector produced neither linalool nor nerolidol. The results clearly proved that 59 found to be bifunctional LNSs (Fig. 2C), thus renamed Gm.LNS and Hs.LNS, respectively. Moreover, like Agr8, the strain ex-91 pressing Galma_266794 produced germacrene D (1,10 cyclization) as main product and a few minor products (g-muurolene 92 and (+)-d-cadinene, Fig. S4), thus validating our hypothesis. 93 Here, the synergistic use of BLAST search, full-sequence alignment and active-site alignment was explored. As such, we 94 achieved a relatively high predictability of hunting for biocatalysts of the same function (e.g. linalool synthases). BLAST search 95 helps with identifying overall similar enzymes that form our initial screening candidates. Full sequence alignment and phyloge-96 netic tree facilitate the classification of enzymes. Different classes often have distinct catalytic functions (e.g. different cyclization 97 positions). Active-site prediction further supplements the prediction with two main roles. One is to filter out those enzyme can-98 didates of incomplete binding pockets (e.g., Agr11 is missing the NSE triad, Tables S1 and S2). The other is to complement the 99 enzyme classification of full sequence alignment. This is based on the hypothesis that enzymes of the same function may have 100 overall low similarity but more conserved active sites. For example, the Galma_266794 shares comparable full-sequence simi-101 larities with Ap.LS (61%) and Agr8 (62%), however it has much higher active-site identity with Agr8 (95%) than with Ap.LS (71%, 102 Table S1). Indeed, the products of Galma_266794 are similar to that of Agr8. 103 Purification and characterization of fungal LS and LNSs.

104
To date, a number of plant LSs and LNSs have been identified. However, only one bacterial LNS from Streptomyces 105 clavuligerus was recently identified 11 , which only shares 15.2% identity with Aa.LS (Fig. S5). With the fungal enzymes studied 106 in this work, LSs and LNSs in three kingdoms have been identified. Next, we sought to compare their catalytic activities and 107 mechanisms by in vitro, in vivo assays, sequence alignments and 3D structural models. 108 Protein purification is the prerequisite to study the in vitro kinetics of the fungal enzymes. Though all the five bacterial strains 109 expressing fungal enzymes produced linalool, their expression levels in E. coli were largely different. Aa.LNS had the highest 110 expression level, followed by Gm.LNS, Ap.LNS and Ap.LS. The expression of Hs.LNS (Hypsu_148385) was so low that it was 111 not detectable in a protein gel (Fig. S6A). As Aa.LNS had the highest expression level and Ap.LS is the only specific monoter-112 pene synthase, they were chosen as the representatives of fungal LNS and LS for further studies. However, none of them was 113 soluble based on solubility analysis with B-PER II reagent (Thermo Scientific™) (Fig. S6A). Many approaches were tested but 114 failed to improve their solubility (such as abiotic condition optimization: lowering incubation temperature, tuning inducer dosages, 115 media additives and protein fusion). Refolding of insoluble fraction could be another solution which we did not test because it is 116 very time-consuming to optimize the best conditions. The N-terminal fusion of Aa.LNS with a maltose binding protein or thiore-117 doxin did not help (Fig. S6B). Different chaperone systems (DnaK-dnaJ, GroES-GroEL) and trigger factor (TF) in E. coli were 118 further tested. It was found that TF chaperone could slightly improve the solubility of the synthases. With the optimal condition 119 (3.3 mM arabinose to induce TF chaperone and 0.1 mM IPTG to induce Aa.LNS, Fig. S7) and further separation by size exclu-120 sion chromatography, we managed to purify enough soluble Aa.LNS for in vitro characterization. Yet its purity was quite low 121 with ~16.3% (Fig. 3A). In contrast, relatively high purity of soluble Ap.LS (~71.2%) was obtained with the same experimental 122 conditions (Fig. 3B). Consistent with the E. coli cultures producing the respective synthase, purified enzymes reconfirmed that 123 Aa.LNS can use FPP and GPP to produce nerolidol and linalool, respectively. Whereas Ap.LS was only able to use GPP but 124 not FPP (Fig. 3). The Km and kcat values of Aa.LNS for FPP were 9.0±2.3 μM and 3.3±0.3 min -1 , respectively; and slightly lower 125 for GPP with 6.7±4.6 μM and 0.5±0.1 min -1 , respectively ( Table 1). The Km value of Ap.LS for GPP with 3.8±0.7 μM was slightly 126 lower than that for Aa.LNS, whereas kcat was much higher with 6±0.3 min -1 . To compare the catalytic efficiencies among the 127 known LSs and LNSs, kcat/ Km value of Ap.LS was the highest, which is about 21-fold, 29-fold, threefold and fourfold higher than 128 that of Aa.LNS, La.LS (Q2XSC5) from Lavandula angustifolia (Lavender) 5 , Ma.LS (Q8H2B4) from Mentha aquatica 23 , and of 129 the bacterial Sc.LNS 11 , respectively. As for Aa.LNS as a nerolidol synthase, although the kcat value of Aa.LNS for FPP was 130 more than five times lower compared to the bacterial one, the Km value was similar to that of the bacterial Sc.LNS 11 and about 131 half as that of Zm.LNS from Zea mays (Maize) ( In vivo activity comparison of LSs and LNSs from three kingdoms.

133
Due to potential issues such as poor expression and solubility when expressed in cells and the localization difference (cyto-134 solic/membrane-bound in vivo versus a one-pot aqueous reaction), the advantages of in vitro enzyme kinetics (Table 1)  is about 44-fold and 287-fold as high as that of Sc.LNS (bacterial) and Cb.LS (plant), respectively. A previous study also sup-142 ported that the bacterial Sc.LNS is better than plant LSs in terms of in vivo activity in TB media 10 . In the same TB media, the 143 linalool titre reached 601.2 mg/L for Ap.LS stain, about 65% higher than previously reported using Sc.LNS 10 . Our study demon-144 strated that fungal Ap.LS is even superior to the bacterial one, in both activity and selectivity. Although Sc.LNS has a higher 145 activity than plant Cb.LS, it prefers FPP (lower Km and higher kcat) to GPP as the substrate 11 . Therefore, Sc.LNS produced a 146 larger amount of nerolidol than linalool in E. coli whose cytosol contained both FPP and GPP, in contrast, Ap.LS produced 100% 147 linalool. The superior activity and selectivity of Ap.LS makes it more suitable for microbial production of linalool than its plant 148 and bacterial counterparts. High activity contributes to high titers, rates and yields (TRYs) of linalool production and low manu-149 duction cost. Thus, this study paves the way for commercialization of linalool bioproduction which is greener, safer, sustainable 151 and of exceptional enantiopurity ((R)-linalool), as compared to chemical synthesis. 152

153
Next, we generated a phylogenetic tree with 35 plant enzymes (including 4 nerolidol synthases, 9 LNs and 22 LNSs), 1 154 bacterial LNS (D5SL78) and 9 fungal enzymes (Table S3). The enzymes were clearly separated into two major clades (one is 155 plant, clade 1, and the other is microbial, clade 2, Fig. 4). The bacterial LNS was closer to fungal ones, in clade 2. Specifically, 156 the sequence identity among fungal, bacterial and plant LNSs or LSs are only 8%-15%, which includes those metal binding sites 157 ( Fig. S5). Overall, plant LSs/LNSs are larger with 500-900 amino acids than microbial ones with 300-400 amino acids. One 158 particular enzyme D8RNZ9, which is a LNS isolated from the nonseed plant Selaginella moellendorffii (Spikemoss) 25 , is more 159 closely related to the bacterial LNS than plant LNSs. It was hypothesized that it could stem from horizontal gene transfer from 160 microbes to plants or that seed plants lost these LNS enzymes during evolution from nonseed plants 25 (Table S4)  cyclases with a single domain, despite with acyclic products (Fig. S9). Both GPP (its analogue, 2-fluorogeranyl diphosphate) 172 and FPP were docked into the three models. We mainly analyzed the interactions of the three LSs with GPP. With 9 hydrophobic 173 interactions with GPP, Ap.LS had the highest amount, as compared to 7 of Sc.LNS and 6 of Ma.LS (Fig. 5). Except for the 174 negatively charged pyrophosphate (Ppi) head, GPP is largely hydrophobic, thus, these hydrophobic interactions may contribute 175 to the high activity of Ap.LS. The number of hydrogen bonds identified for the three enzymes were similar. In addition, Ap.LS 176 had the highest binding affinity (-7.6 kcal/mol) to GPP, followed by Sc.LNS (-7 kcal/mol) and Ma.LS (-6.4 kcal/mol). The binding 177 affinity inversely correlated with the Km values of the three enzymes (Table 1), where higher binding affinity contributed to a 178 lower Km value. As compared for Ap.LS and Ap.LNS, Ap.LS has higher binding affinity to GPP than Ap.LNS (-7.3 kcal/mol), but 179 lower binding affinity (-8.5 kcal/mol) to FPP than Ap.LNS (-9.0 kcal/mol). The binding affinity data are nicely correlated with their 180 difference in monoterpene and sesquiterpene activities. 181 Furthermore, we superimposed the 3D structures of the active sites of the three enzymes ( Fig. 6A and Table S4) Ap.LS share the highest identity 77.9%, hence, we compared their difference of residues surrounding the substrate binding 192 pocket. In total, five residues were found to be different ( Fig. 6B and Table S5) had effect on the selectivity of Ap.LS (Fig. 6E). We speculated that the specificity might be the synergistic result of multiple 198 residues. The region A59-L60 was particularly interesting, as A59 and L60 are in close proximity to both the Ppi head and 199 hydrocarbon tail of FPP (Fig. 6D). Indeed, the combination of A59S and L60M mutations resulted in the production of a trace 200 amount of nerolidol, ~2% of total amount of linalool and nerolidol produced ( Fig. 6C and E), which indicated the two mutations 201 are sufficient to convert Ap.LS from a monofunctional LS to a bi-functional LNS. Adjacent to A59-L60, another residue is also 202 different between Ap.LS (V61) and Ap.LNS (I60). Although single mutation V61I had no effect on the selectivity of the wildtype 203 Ap.LS, the introduction of V61I enhanced the nerolidol production by 12-fold (~40% of total linalool and nerolidol produced) and 204 decreased linalool production by 45% on the basis of the double mutant A59S-L60M ( Fig. 6C and E). It seems that the mutation 205 L60M-A59S favors sesquiterpene activity (nerolidol formation) by stabilizing the ligand in a favorable position (A59S) and by 206 promoting the easier leave of the Ppi group from the binding pocket (L60M, Fig. 6D). The third mutation V61I further enhances 207 the effect by pushing M60 and S59 closer to FPP (Fig. 6D). 208 sponding residues are more similar to those 59-61 of Ap.LS (S59:A59, L60:L60, V61:V61, Fig. S9), produced the highest amount 211 of linalool (15%) among all the wildtype fungal LNSs (Fig. 2C). As such, we concluded that the region residues (A59-V61) play 212 an essential role for the highly specificity of Ap.LS. 213

215
Sequence validation of AAE3_109435 in the genome. Bioinformatics prediction of fungal linalool synthases. 226 The fungal LS candidate genes were obtained by the combination of BLAST search in JGI fungal genomics and UniProt 227 databases, full sequence alignment and predicted active site alignment. Those homologues with the highest similarity (usually 228 >50% identity, or in the same cluster of the phylogenetic tree or sequence similarity network) to Aa.LNS (accession number 229 MN954676) or Ap.LS in full sequences and in active sites were selected as the targets for experimental validation. 230 Prediction of active sites of fungal terpene synthases.

231
The full sequence alignment was generated by aligning the complete sequences of LS and LNS proteins from across the 232 three kingdoms by Clustal Omega v1.2.2. In contrast, active-site alignment is proceeded by identifying amino acid residues 233 surrounding the predicted active sites of each enzyme in 3D structures and used to extract the relevant residues from the full 234 alignment. We used an in-house developed algorithm, BioTransformer v0.9, to predict and align the active sites. In brief, the 235 algorithm first searches the PDB for appropriate templates. Next, the user gets to select the most appropriate PDB template or 236 templates with the most appropriate ligands (usually the ligands or combination of ligands that maximize the space within the 237 active site). Using this approach, the PDB structures 4LXW (Epi-isozizaene synthase in complex with inorganic pyrophosphate 238 (Ppi) and benzyl triethyl ammonium from Streptomyces coelicolor) and 5NX5 (Linalool/Nerolidol synthase in complex with 2-239 fluorogeranyl diphosphate from Streptomyces clavuligerus) were chosen as templates for the prediction of the active sites. To 240 maximize the number of residues found within the binding pockets of the PDB structures, residues found within 6.0 Angstrom 241 from the substrate are considered as part of the active site, and the union set derived from both structural templates is used as 242 the predicted active site. 243 Phylogenetic analysis and sequence similarity network of terpene synthases.  cam-T7-dxs-idi that overexpresses the enzymes DXS and IDI from E. coli to enhance the supply of terpene precursors 3 . For 266 the monoterpene production study, the mevalonate pathway was overexpressed in p15A vectors under the T7 promoter variants 267 9 . 268 Mutation study of Ap.LS.

269
The targeted mutations were introduced to Ap.LS using in-house methods as described previously  Table  273 S6. 274 Terpenoid production in E. coli.

302
For characterization of fungal terpene synthases, the headspace compounds were sampled at 60 °C for 20 min by SPME 303 with a DVB/CAR/PDMS (50/30 µm divinylbenzene/carboxen/polydimethylsiloxane) fiber (length 1 cm; Supelco, Steinheim, Ger-304 many). Subsequently, the compounds were desorbed for 1 min in the split inlet (250 °C; SPME liner, 0.75mm i.d.; Supelco) and 305 analyzed by an Agilent 7980B gas chromatography equipped with an Agilent 5977B MSD. Samples were injected into Agilent 306 DB5ms column with a split ratio of 40:1 at 240 °C. The oven program started at 80 °C for 1 min, was raised up to 210 °C at 10 307 °C/min, then to 310 °C at 60 °C/min and maintained at 310 °C for another 2 min. Mass spectrometer was operated in EI mode 308 with full scan analysis (m/z 33-300, 9 scans/s). In addition to mass spectra, Kovats retention indices of detected compounds 309 were calculated by calibrating the used GC-MS system with a C8-C30 alkane mix and compared with literature data in the 310 National Institute of Standards and Technology database (Fig. S2). 311 For quantification of linalool and nerolidol of different strains, the organic layer with secreted terpenes was separated and 312 diluted into ethyl acetate by 10-100 times. The samples were then analyzed with the same GC-MS program as the characteri-313 zation method. The concentrations were calculated by interpolation using the standard curve of authentic linalool and nerolidol 314 standards (Sigma-Aldrich, Singapore). 315 Chiral study of linalool produced by fungal LSs.The chirality of linalool produced by fungal LSs was analyzed by the 316 GC chiral CycloSil-B column (30m, 0.25mm, 0.25u, Agilent, Singapore) in the same Agilent 7980B gas chromatography 317 equipped with the 5977B MSD. The oven program started at 80 °C for 2 min, was raised up to 210 °C at 5 °C/min, then to 250 318 °C at 20 °C/min and maintained at 250 °C for another 2 min. Mass spectrometer was operated in EI mode with full scan analysis 319 (m/z 33-300, 5.5 scans/s). The retention times and mass spectra of samples were compared with authentic standards of both 320 (R)-linalool and a mixture of (R/S)-linalool. 321 322   Table S1. Predicted active sites of terpene synthases with 4LXW and 5NX5 as the templates. 323    This study This study 5 23 11     Table S3.

Bioinformatics-Aided Identification, Characterization of Fungal Linalool Synthases and Applications in Linalool biosynthesis
Congqiang Zhang 1* , Xixian Chen 1 , Raphael Tze Chuen Lee 2 , Rehka T 1 , Sebastian Maurer-Stroh 2,3# , Martin Rühl 4# Table of contents  Table S1. Predicted active sites of terpene synthases with 4LXW and 5NX5 as the templates. Table S2. Comparison and combination of the predicted active sites of the two templates 4LXW and 5NX5. Table S3. Summary of LSs and LNSs in this study. Table S4. 3D alignment of active-site residues of Ap.LS, Sc.LNS and Ma.LS. Table S5. Comparison of residues in the substrate-binding pockets of Ap.LS and Ap.LNS. Table S6. Primers used for Ap.LS mutation. Fig. S1. Nucleic acid sequence of AAE3_109435 in the genome. Fig. S2. Mass spectra and retention indices for terpenes detected in this study. Fig. S3. The alignment between Agrped_689675 and Galma_223690 and BLAST search results in UniProt database with Agrped_689675 (or Ap.LS). Fig. S4. GCMS chromatograms and spectra for Galma_266794. Fig. S5. Amino acid sequence alignment and identity table of LSs and LNSs from fungi, bacteria and plants. Fig. S6. Expression and solubility analysis of the fungal LNSs and LS. Fig. S7. The optimization of Aa.LNS solubility with the co-expression of chaperone proteins. Fig. S8. The sequence alignment and their secondary structures of plant terpene synthases. Fig. S9. The sequence alignment and their secondary structures of microbial terpene synthases. Table S1. Predicted active sites of terpene synthases with 4LXW and 5NX5 as the templates.
Murolene/cadinene synthase cluster and LS/LNS cluster predicted or characterized in this study were shown in blue and green, respectively (consistent with Fig. 2). The nonfunctional synthase Agr10 and Agr11 were highlighted in red. 1 The two proteins were used as templates: D5SL78, Sc.LNS from Streptomyces clavuligerus, (PDB ID: 5NX5) and Q9K499, Epi-isozizaene synthase from Streptomyces coelicolor (PDB ID: 4LXW). For 4LXW, the ligands BTM, POP and MG were used for prediction of active sites. For 5NX5, the ligands 0FV and MG were used for prediction of active sites. 2 Analyzed by PyMOL software version 2.1.1. The homologue model of Q8H2B4 was built on the structure of (+)-bornyl diphosphate synthase from Salvia officinalis (PDB entry ID, 1n1b/1n21) with Modeller software.

consensus X L X M X F V D E T D V E Y R T X G X X X X N D S Y X E X X H N W X W S R Y
Agrped1_694262

F L A M Y F V D E T D V E Y R T S A G T I I N D S F R E C G H N W N W S R Y 14
Agrped1_749682