Identification of the diterpene synthase used to synthesize salvinorin A

Identification of the diterpene synthase Salvia divinorum uses to synthesize salvinorin A
March 17, 2008

Abstract
Salvia divinorum is a plant found in Oaxaca, Mexico where it has been traditionally used by the Mazatec Indians for its hallucinogenic and therapeutic effects. It produces a unique hallucinogenic compound called salvinorin A which is a highly selective κ-Opioid receptor agonist. κ-Opioid agonists could conceivably represent treatments for a wide range of diseases including schizophrenia, certain dementias, Alzheimer’s, and even cocaine and amphetamine addiction. Salvinorin A is a diterpenoid produced in the glandular trichomes and at this point there has been very little research on the pathway S. divinorum uses to biosynthesize salvinorin A. This study identifies a cDNA sequence from S. divinorum with high sequence homology to a copalyl diphosphate synthase previously identified and characterized in Scoparia dulcis. This is a very likely candidate for the enzyme that catalyzes the first committed step in pathway to salvinorin A. In addition, salvinorin A in the leaves, stems, and roots of the plant was quantified using liquid chromatography-mass spectrometry.

Project Goal
The goal of this project is to further understand the metabolic pathway that Salvia divinorum uses to biosynthesize salvinorin A. Currently none of the committed enzymes, meaning the enzymes that lead only to salvinorin A, involved in the pathway have been identified. The goal of the project is to identify and characterize the diterpene synthase that is used to catalyze the first committed step in the pathway to salvinorin A.

Background information
   Salvia divinorum is a unique member of the mint (Lamiaceae) family found in Oaxaca, Mexico where it has been traditionally used by the Mazatec Indians for its hallucinogenic and therapeutic effects. Ritual healing ceremonies were held in which S. divinorum was used for its medicinal benefits in regulating eliminatory functions (such as diarrhea), relieving headaches and rheumatic disorders (such as arthritis), and moderating anemia (Valdes et al., 1983). The plant has only been found on isolated plots and is therefore thought to be a cultigen, meaning it lacks a wild or uncultivated counterpart (Reisfield et al., 1993). The active compound in the plant responsible for inducing its unique hallucinogenic state, and probably many of its medicinal effects, is salvinorin A. It is active at extremely low doses (200 to 500 µg) making it the most potent naturally occurring psychoactive known (Valdes, 1994).
   Salvinorin A is highly selective to the κ-Opioid (kappa-Opioid) receptor and because of this it is believed to have tremendous medicinal potential. κ-Opioid receptor selective agonists could conceivably represent treatments for diseases in which hallucinations are prominent, including schizophrenia, depression with psychotic features, and the hallucinosis associated with certain dementias, such as Alzheimer’s, Huntington’s, and Pick diseases, and certain types of drug abuse, such as amphetamine and cocaine psychosis (Roth et al., 2002). In addition, it has been suggested that salvinorin A may be useful for treating HIV or AIDS because a ligand that binds to the κ-Opioid receptor was shown to effectively suppress the expression of a monocytotropic HIV-1 strain in human cells and blood cultures (Peterson et al., 2001).
   Salvinorin A is a secondary metabolite produced by S. divinorum. Secondary metabolites are specialized compounds produced in plants for purposes other than growth, development, and reproduction. These compounds can serve a variety of purposes such as defense agents against microbial pathogens and insect and animal herbivores. Some secondary metabolites are volatile and are released to attract pollinators or even insects that prey on the plant's enemies or repel harmful organisms (Pare et al., 1999; Wittstock et al., 2002). In many cases these compounds are stored in specialized cells or structures, presumably to protect the plant itself from its own toxicity (Duke et al., 2000). Common structures used to house these compounds are called glandular trichomes, which protrude from the surface of the plant. Typically glandular trichomes contain gland cells (or a single cell) that synthesize these compounds and also a cuticular sac that covers the gland cells and releases the contained compounds when it is ruptured. Salvinorin A is localized in the glandular trichomes on the abaxial sides of the leaves (Siebert, 2004). Since salvinorin A is a diterpenoid, a common class of secondary metabolites, and is found in glandular trichomes (Fig. 1), it is therefore presumable that it is also produced within these structures.

Fig. 1 Scanning electron micrographs showing S. divinorum trichomes. (A) An intervenous area of the abaxial (underside) leaf surface with several peltate glandular trichomes clearly visible. (B) A closer view of a peltate glandular trichome on the abaxial leaf surface. Photo by Daniel Siebert (2004), Sage Wisdom.

   In plants, terpenoids, such as salvinorin A, are produced via the assembly of isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP) building blocks (Eisenreich et al., 2004). These precursor chemicals were previously thought to be biosynthesized only through the mevalonic acid (MVA) pathway. However, it was discovered that some plants utilize a second pathway involving the monophosphate of 1-deoxy-d-xylulose (DOX) in addition to the MVA pathway (Broers, 1994; Schwarz, 1994). This alternative biosynthetic route is now known as the DOXP/MEP (or non-mevalonate) pathway (Rohmer, 1999). S. divinorum is one of the plants found to utilize the DOXP/MEP pathway, which it uses to biosynthesize salvinorin A (Kutrzeba et al., 2007). This was found by incorporating stable isotopes, in this case 13C (2H is also commonly used), into the plant; a common practice used in the study of biosynthetic pathways of many different metabolites (Simpson, 1998).
   Apart from the determination that salvinorin A is produced through the MEP/DOXP pathway, very little has been studied about the biosynthesis of salvinorin A. None of the enzymes involved in the pathway to salvinorin A after the synthesis of IPP and DMAPP have been identified. When tracing a pathway it is usually easiest to begin by investigating the initial and final steps. After IPP and DMAPP are synthesized, various terpene synthases catalyze the initial steps that eventually lead to the production of monoterpenoids, diterpenoids, and other terpenoids (Croteau, 1995). Since salvinorin A is a diterpenoid, the first committed step along its pathway will require a diterpene synthase.

Methods
Plant Material
   Live Salvia divinorum plants were obtained from a plant supplier in Shirley, Arkansas. The plants were grown hydroponically at 20°C under 16 hours of florescent light per day.

Construction of cDNA Library and Sequencing
   Glandular trichomes isolated from S. divinorum were obtained from the University of Mississippi. The RNA used to make the cDNA library was isolated from the trichomes with RNeasy kit using the spin column method. A modified version of the Clontech SMART cDNA library kit protocol was used to create the cDNA library. A QIAquick PCR purification kit was used to do a cDNA cleanup. The cDNA library was sent to the Michigan State University Research Technology Support Facility to be sequenced via 454 sequencing. The sequencing yielded 213,458 reads after the clean-up.

BLAST searching and assembly
   The sequences, from the GenBank database, of several diterpene synthases that have been previously identified and characterized were BLAST searched against the sequenced S. divinorum cDNA library. This pulled up 32 non-redundant fragments from the S. divinorum cDNA library which were assembled using the Seqlab fragment assembler. The two most complete assembled contigs were Netblast searched against the NCBI protein database. The sequences for several proteins found with the highest homology to the contigs were BLAST searched against the S. divinorum cDNA library. This search yielded 127 non-redundant fragments that assembled into 9 contigs. The longest contig produced by the assembly was 1184 base pairs long. This contig was Netblast searched against the NCBI protein database and found to be homologus with several diterpene synthases.

Quantification of Salvinorin A in Plant Tissues
   A stock solution was prepared with 1 mg of ~92% pure salvinorin A (obtained from Daniel Siebert) in 1 mL of methanol. Samples were prepared containing the following amounts of salvinorin A: 0.255, 0.510, 0.765, 1.02, 1.28, and 1.53 µg. Three 5 µL injections of each sample were carried out on the liquid chromatography-mass spectrometer (LC-MS). The three peak areas for each amount of salvinorin A were averaged and used to create the standard curve.
Plant tissue was harvested from the leaves, stems, and roots, lyophilized, and then ground into powder. Three equal samples of each type of tissue were extracted in chloroform (50 mL per 1 g of tissue) for 30 min. The extract was filtered and then evaporated with a nitrogen evaporator. The dry solids in each sample were redissolved in 1 mL methanol using sonication to assist in dissolving all solid material.
   For each sample, 5 µL of reconstituted methanol extract was injected into the LC-MS and the peak area due to salvinorin A was calculated. This value was plugged into the linear equation obtained from the standard curve in order to calculate the amount of salvinorin A in the sample.

Results
BLAST searching and assembly
   The original BLAST search against the S. divinorum cDNA library sequence data was performed with the following diterpene synthase protein sequences: P59287 (casbene synthase), Q41594 (taxadiene synthase), AAK83566 (taxadiene synthase), AAG02257 (taxadiene synthase), Q38710 (abietadiene synthase), AAS47691 ( levopimaradiene/abietadiene synthase), AAK83563 (abietadiene synthase), CAA75244 (copalyl diphosphate synthase), and AAS47690 (isopimaradiene synthase). The 32 non-redundant fragments found from the S. divinorum cDNA library were assembled into 13 contigs ranging from 200-550 base pairs in length. The proteins showing the highest homology found after performing a Netblast (against the GenBank protein database) with the two most complete contigs (481 and 553 base pairs long) are shown in Table 1.

Table 1

When the proteins showing the highest homology (Table 1) were BLAST searched against the S. divinorum sequence data 127 non-redundant fragments were found and assembled into 9 contigs. One contig was significantly more complete than the others at 1184 base pairs long and assembled from 112 fragments. The second longest contig was 553 base pairs and assembled from only 5 fragments. When the largest contig was Netblast searched it was found to be 56% similar and 68% positive to putative copalyl diphosphate synthase (BAD91286) from Scoparia dulcis. The contig sequence (Contig1) compared with the the sequence for BAD91286 is shown in Fig 2. The contig also showed high similarity to other previously identified diterpene synthases (Table 2).

Fig. 2 BLAST result comparing the sequence of the most complete contig assembled from the S. divinorum cDNA library (Contig1) to putative copalyl diphosphate synthase (BAD91286) from Scoparia dulcis. Identical matches are labeled in green and positive matches are labeled in turquoise.

Table 2

Quantification of Salvinorin A in Plant Tissues
   Two standard curves were created, one using the peak area of original 433 m/z ion of salvinorin A and another using the first ion it breaks down into (373 m/z), which was much more abundant. The 373 m/z standard curve was preferable because it had less error. Using the 373 m/z standard curve the concentration of salvinorin A was calculated in the leaves, stems, and roots at 72.63, 3.66, and 0.46 µg/g of dried tissue, respectively. The calculated concentration in the leaves had a percent error of 24% and for the roots it was 10%. The stems had an unacceptably high percent error of 90%.

Discussion
   A sequence encoding for a protein in S. divinorum (Contig1) was found that showed high sequence homology to the putative copalyl diphosphate synthase (BAD91286) identified and characterized by Lee et al. (2005) in S. dulcis. Contig1, assembled from the S. divinorum cDNA library, is 1184 base pairs long and BAD91286 is a little over double that, so it is unlikely that Contig1 contains the sequence for the complete diterpene synthase gene in S. divinorum. Contig1 was the only contig of significant size found when BLAST searching was performed. There are probably other diterpene synthases in S. divinorum, however the one encoding for the diterpene synthase that catalyzes the first committed step in the pathway to salvinorin A is likely to be the most expressed because salvinorin A seems to be the most abundant secondary metabolite. Therefore, since the sequenced cDNA library reflects the level gene expression in the plant, this gene is likely to be the easiest to find.
   Lee et al. (2005) found the copalyl diphosphate synthase expressed in E. coli to produce ent-copalyl diphosphate from the substrate GGPP, as is the typically the case for copalyl diphosphate synthases found in plants (Bohlmann et al., 1998). Therefore, if Contig1 does encode for a copalyl diphosphate synthase in S. divinorum then it is likely that ent-copalyl diphosphate is a precursor to salvinorin A.
   Since salvinorin A is localized in the glandular trichomes (Siebert, 2004) it was expected that there would be a significantly lower concentration in the roots. The results are in agreement with this, however there was a 90% error for the concentration in the stems and a previous study that quantified salvinorin A in S. divinorum (Gruber et al., 1999) found significantly higher concentrations of salvinorin A in both the stems and leaves. Although their values varied widely from plant to plant, their lowest value (0.89 mg/g) is over ten times higher than what the results show (72.63 µg/g). There are a number of possible reasons for this discrepancy. Due to low availability of tissue only three samples (for each tissue type) of only 0.25 grams each were used to calculate salvinorin A concentration. The tissue was also relatively unhealthy and not selected for uniformity. The quantification will be repeated, but these results do allow for speculation about the nature of the diterpene synthase of interest. Since it is clear that the leaves contain the most salvinorin A it is likely that this is where protein expression will be highest. The concentration was dramatically reduced in the roots, so there will likely be little to no enzyme expression in this part of the plant.

Conclusion
The diterpene synthase that catalyzes the first committed step in the pathway to salvinorin A is likely to be a copalyl diphosphate synthase. The next step is to attempt to find the complete sequence for this diterpene synthase. One way this could be acomplished is by assembling the entire S. divinorum cDNA library and then BLAST searching Contig1 against the assembled library. Once the complete sequence is found it will be expressed in E. coli and assayed. When expressed, the protein should create a diterpenoid precursor from GGPP.

References
1.Bohlmann, J.; Meyer-Gauen, G.; Croteau, R. 1998. Plant terpenoid synthases: molecular biology and phylogenetic analysis, Proc. Natl. Acad. Sci. U.S.A. 95: 4126–4133.
2. Broers, S. T. J. 1994. Regarding the early steps of the biosynthesis of isoprenoids in Escherichia coli. ETH-Zurich Dissertation. #10978.
3.Chen, F.; Tholl, D.; D’auria, J. C.; Farooq, A.; Pichersky, E.; Gershenzon, J. 2003. Biosynthesis and emission of terpenoid volatiles from Arabidopsis flowers. The Plant Cell. 15: 1–15.
4.Croteau, R; McGarvey, D. J. 1995. Terpenoid Metabolism. The Plant Cell. 7: 1015-1026.
5.Duke, S. O.; Canel, C.; Rimondo, A. M.; Tellez, M. R.; Duke, M. V.; Paul, R. N. 2000. Current and Potential Exploitation of Plant Glandular Trichome Productivity. Advances in Botanical Research. 31: 121–151.
6.Eisenreich, W.; Bacher, A.; Arigoni, D.; Rohdich, F. 2004. Biosynthesis of isoprenoids via the non-mevalonate pathway. Cell. Mol. Life Sci. 61: 1401–1426.
7.Gang, D. R.; Wang, J.; Dudareva, N.; Nam, K. H.; Simon, J. E.; Lewinsohn, E.; Pichersky, E. 2001. An investigation of the storage and biosynthesis of phenylpropenes in sweet basil. Plant Physiol. 125: 539–555.
8.Gruber, J. W.; Siebert, D. J.; Der Marderosian, A. H.; Hock, R. S. 1999. High Performance Liquid Chromatographic Quantification of Salvinorin A from Tissues of Salvia divinorum Epling & Játiva-M. Phytochem. Anal. 10: 22–25.
9.Kutrzeba, L.; Dayan, F. E.; Howell, J.; Feng, J.; Giner, J.; Zjawiony, J. K. 2007. Biosynthesis of salvinorin A proceeds via the deoxyxylulose phosphate pathway. Phytochemistry. 68(14): 1872-1881.
10.Lee, J.; Nakagiri, T.; Hayashi, T. 2005. cDNA cloning, functional expression and characterization of ent-copalyl diphosphate synthase from Scoparia dulcis L. Plant Sci. 169: 760-767.
11.Pare, P. W.; Tumlinson, J. H. 1999. Plant volatiles as a defense against insect herbivores. Plant Physiol. 121: 325–332.
12.Peterson, P. K.; Gekker, G.; Lokensgard, J. R.; Bidlack, J. M.; Chang, A.; Fang, X.; Portoghese, P. S. 2001. κ-Opioid receptor agonist suppression of HIV-1 expression in CD4+ lymphocytes. Biochemical Pharmacology. 61(9): 1145-1151.
13.Reisfield, A. S. 1993. The Botany of Salvia divinorum (Labiatae). SIDA. 15: 349-366.
14.Rohmer, M. 1999. The discovery of a mevalonate-independent pathway for isoprenoid biosynthesis in bacteria, algae and higher plants. Nat. Prod. Rep. 16: 565–574.
15.Roth, B. L.; Baner, K.; Westkaemper, R.; Siebert, D.; Rice, K. C.; Steinberg, S.; Ernsberger, P.; Rothman, R. B. 2002. Salvinorin A: A Potent Naturally Occurring Nonnitrogenous Kappa Opioid Selective Agonist. PNAS. 99(18): 11934–11939.
16.Schwarz, M. K. 1994. Terpene biosynthesis in Gingko biloba: a surprising story. ETH-Zurich Dissertation. #10951.
17.Siebert, D. J. 2004. Localization of Salvinorin A and Related Compounds in Glandular Trichomes of the Psychoactive Sage, Salvia divinorum. Annals of Botany. 93(6): 763–771.
18.Simpson, T. J. 1998. Application of isotopic methods to secondary metabolic pathways. Top. Cur. Chem. 195: 1–48.
19.Valdes, L. J., III. 1994. Salvia divinorum and the Unique Diterpene Hallucinogen,
Salvinorin (Divinorin) A. Journal of Psychoactive Drugs. 26, 277-283.
20.Valdes, L. J., III; Diaz, J. L.; Paul, A. G. 1983. Ethnopharamacology of Ska Maria Pastora (Salvia divinorum, Epling and Jativa-M). Journal of Ethnopharmacology.
21.Wittstock, U.; Gershenzon, J. 2002. Constitutive plant toxins and their role in defense against herbivores and pathogens. Curr Opin Plant Biol. 5: 300–307.