Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2011 Sep 30;108(41):16944-16949. doi: 10.1073/pnas.1111558108

Improving peppermint essential oil yield and composition by metabolic engineering

Bernd Markus Lange a,b,1, Soheil Seyed Mahmoud a,2, Mark R Wildung a,3, Glenn W Turner a, Edward M Davis a, Iris Lange a, Raymond C Baker c, Rick A Boydston d, Rodney B Croteau a,1
PMCID: PMC3193216  PMID: 21963983

Abstract

Peppermint (Mentha × piperita L.) was transformed with various gene constructs to evaluate the utility of metabolic engineering for improving essential oil yield and composition. Oil yield increases were achieved by overexpressing genes involved in the supply of precursors through the 2C-methyl-D-erythritol 4-phosphate (MEP) pathway. Two-gene combinations to enhance both oil yield and composition in a single transgenic line were assessed as well. The most promising results were obtained by transforming plants expressing an antisense version of (+)-menthofuran synthase, which is critical for adjusting the levels of specific undesirable oil constituents, with a construct for the overexpression of the MEP pathway gene 1-deoxy-D-xylulose 5-phosphate reductoisomerase (up to 61% oil yield increase over wild-type controls with low levels of the undesirable side-product (+)-menthofuran and its intermediate (+)-pulegone). Elite transgenic lines were advanced to multiyear field trials, which demonstrated consistent oil yield increases of up to 78% over wild-type controls and desirable effects on oil composition under commercial growth conditions. The transgenic expression of a gene encoding (+)-limonene synthase was used to accumulate elevated levels of (+)-limonene, which allows oil derived from transgenic plants to be recognized during the processing of commercial formulations containing peppermint oil. Our study illustrates the utility of metabolic engineering for the sustainable agricultural production of high quality essential oils at a competitive cost.

Keywords: 1-deoxy-D-xylulose 5-phosphate reductoisomerase, 2C-methylerythritol 4-phosphate pathway, monoterpene, terpenoid

Peppermint (Mentha × piperita) is the source of commercially valuable essential oil used in numerous mint-flavored consumer products (1). Following a peak production in 1995, the peppermint acreage in the United States has been decreasing (2), and there is an increasing market pressure for lower-cost mint oils. Over the course of several decades improvements in agricultural practices have resulted in substantial yield increases (2). However, progress in varietal improvements has been frustratingly slow, in part because peppermint, which is a sterile hybrid developed from a cross between water mint (Mentha aquatica) and spearmint (Mentha spicata) (3), is not amenable to classical breeding. Thus, metabolic engineering is the only realistic short-term alternative for enhancing peppermint oil yield and composition (4). Fortunately, in comparison to other essential oil plants, the monoterpenoid essential oil biosynthetic pathway in peppermint is exceptionally well understood (reviewed in ref. 5), which has facilitated the development of metabolic engineering strategies.

One approach to potentially increase peppermint oil yield is to manipulate the expression levels of selected genes involved in the 2C-methyl-D-erythritol 4-phosphate (MEP) pathway, which provides the precursors for monoterpene biosynthesis (6, 7) (Fig. S1). Indeed, increased oil yields were reported previously for transgenic plant lines overexpressing the MEP pathway gene 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR; up to 44% above wild-type controls) (8, 9). An alternative strategy to increase peppermint oil yield by metabolic engineering is to overexpress genes that encode enzymes catalyzing seemingly slow steps in monoterpene biosynthesis, but this approach has not been successful thus far. For example, (−)-limonene synthase ((−)-LS) catalyzes the first committed step in the p-menthane monoterpene pathway and, because of its relatively low catalytic efficiency, is considered a potentially rate-limiting enzyme (10). However, elevated expression levels of the gene encoding (−)-limonene synthase ((−)-LS) led to only moderate increases in oil yield in one study (11), whereas two independent studies did not find any positive effects on yield (12, 13). L3H catalyzes an equally slow step in the pathway (14, 15), but the overexpression of the L3H gene also did not result in increased oil yields (9). Metabolic engineering efforts aimed at improving oil composition have focused on reducing the accumulation of (+)-menthofuran and its intermediate (+)-pulegone, particularly under stress conditions. The expression of (+)-menthofuran synthase (MFS) (16) in antisense orientation resulted in reduced (+)-pulegone and (+)-menthofuran levels (9). Interestingly, one of these transgenic antisense lines, designated MFS7A, not only had an improved oil composition but also an increased oil yield by roughly 35%.

In the present study we have investigated the utility of overexpressing additional genes that encode enzymes involved in the precursor supply pathway for monoterpene biosynthesis. We have also generated transgenic plants, tested under greenhouse and commercial-scale field trial conditions, in which a two-gene combination (downregulation of MFS and upregulation of DXR) is used to positively affect both oil yield and composition. We have also produced transgenic lines that accumulate a chemical marker, (+)-limonene, which enables the tracing of oil distilled from our transgenic plants.

Results and Discussion

Increasing Essential Oil Yield.

To evaluate the utility of overexpressing genes involved in monoterpene precursor biosynthesis we generated transgenic peppermint lines using peppermint 1-deoxy-D-xylulose 5-phosphate synthase (DXPS) (17), peppermint isopentenyl diphosphate isomerase (IPPI) (18), and Grand fir (Abies grandis) geranyl diphosphate synthase (GPPS) (19) (Fig. S1), all under control of the constitutive cauliflower mosaic virus 35S promoter (20). The heterologous GPPS from Grand fir was chosen because it is encoded by a single gene (forming a homodimeric protein), whereas the peppermint GPPS is encoded by two genes (forming a heterodimeric protein).

We regenerated a total of 28 transgenic lines harboring the peppermint DXPS construct, 11 of which had a higher DXPS transcript abundance compared to wild-type controls (based on Northern blot analyses) (SI Text). None of the transgenic lines overexpressing DXPS had significant oil yield increases relative to wild-type controls (Table 1). However, three lines (DXPS 15, DXPS23, and DXPS28) had favorable oil compositions (in particular, low (+)-menthofuran levels between 2% and 4% and (+)-pulegone levels below 1%) (Table S1).

Table 1.

Essential oil yield and composition in various greenhouse-grown transgenic peppermint lines, and comparison with wild-type controls (N≥3 ± SD)

Mint Line Oil Yield [Fold changevs. WT] (SD) P-Value (t-test) Monoterpene Composition [% of total monoterpenes]
 Refs
Limonene 1,8-Cineole Menthone Sabinenehydrate Mentho-furan Iso-menthone Menthylacetate Pulegone Menthol Others
Single gene transformations for yield enhancements
WT 1.00 (0.06) n.a. 1.3 5.5 38.1 3.7 9.8 3.9 1.9 2.8 29.3 3.8 p.s.
DXPS2 1.18 (0.12) 0.228 0.7 4.5 45.2 2.2 9.9 4.3 2.1 5.4 22.5 3.2 p.s.
DXPS8 1.16 (0.13) 0.222 1.3 4.8 44.2 2.3 7.8 4.7 1.9 1.5 28.4 3.1 p.s.
DXPS26 1.14 (0.16) 0.300 1.4 5.9 44.8 2.1 8.2 4.1 1.7 3.9 24.8 2.6 p.s.
DXR38 1.44 (0.11) 0.009 2.0 4.6 42.9 2.4 7.5 4.2 1.9 5.3 26.3 2.9 (9)
WT 1.00 (0.03) n.a. 1.4 5.1 44.2 2.0 5.1 4.2 1.9 3.0 29.5 3.6 p.s.
IPPI4 1.26 (0.10) 0.019 0.7 4.5 48.5 2.3 10.6 4.3 1.9 2.7 21.9 2.7 p.s.
IPPI5 1.14 (0.06) 0.043 0.8 4.3 46.2 2.3 11.2 4.2 1.7 3.4 22.9 2.7 p.s.
IPPI9 1.19 (0.06) 0.017 1.0 4.4 45.0 2.6 6.8 4.0 2.1 1.5 29.9 2.6 p.s.
WT 1.00 (0.04) n.a. 1.1 7.2 54.1 1.8 4.8 4.1 2.0 2.7 17.8 4.4 p.s.
GPPS8 1.17 (0.05) 0.001 1.6 6.1 61.9 2.1 4.0 4.6 1.6 0.4 14.9 2.9 p.s.
GPPS10 1.17 (0.04) 0.001 1.5 7.2 61.6 2.4 5.9 4.6 1.7 1.6 10.9 2.8 p.s.
GPPS11 1.18 (0.07) 0.003 1.5 6.3 63.9 1.4 4.1 4.8 1.8 1.3 11.9 3.1 p.s.
LS27 1.09 (0.19) 0.637 1.2 6.1 48.4 2.8 7.5 4.1 1.7 2.8 22.8 2.9 (12)
L3H25 1.01 (0.09) 0.791 1.3 5.2 38.3 2.1 15.1 3.5 1.7 7.1 23.1 3.0 (12)
Singe gene transformations for compositional enhancements
MFS7A 1.35 (0.08) 0.011 2.2 5.5 34.6 2.3 2.4 0.7 1.8 0.8 46.7 2.9 (9)
Two-gene transformations for yield and compositional enhancements
WT 1.00 (0.07) n.a. 0.8 8.9 49.8 1.0 3.9 5.9 0.1 1.8 18.8 9.0 p.s.
BD7A-3 1.61 (0.16) 0.001 1.6 10.1 50.9 1.2 1.4 5.4 0.1 0.2 19.7 9.3 p.s.
BD7A-4 1.53 (0.17) 0.007 1.1 9.9 52.1 0.9 1.6 5.3 0.1 0.2 20.2 8.6 p.s.
BD7A-7 1.49 (0.23) 0.015 1.0 9.8 48.5 0.7 1.5 5.5 0.1 0.2 21.8 10.9 p.s.
Open in a new tab

Data for the three highest yielding lines of each set of transgenic plants are shown. Full data are available in Table S4. Acronyms for transgenes: DXPS, 1-deoxy-D-xylulose 5-phosphate synthase; DXR, 1-deoxy-D-xylulose 5-phosphate reductoisomerase; IPPI, isopentenyl diphosphate isomerase; GPPS, geranyl diphosphate synthase; LH, (−)-limonene synthase; L3H, (−)-limonene 3-hydroxylase; MFS, (+)-menthofuran synthase. The BD7A family of lines was generated by transforming the MFS7A transgenic line (expression of MFS in antisense orientation) with a construct for the overexpression of DXR.

Abbreviations and acronyms: n.a., not applicable; p.s., present study; SD, standard deviation; WT, wild-type peppermint (Black Mitcham cultivar).

Transgenic plants were also regenerated from nine independent transformations with a peppermint IPPI overexpression construct. The abundance of IPPI transcript was extremely low in wild-type controls (below detection limit after one week of photographic film exposure on Northern blots). In contrast, bands were detectable in three transgenics (SI Text). Oil yield increases of up to 26% were measured in the IPPI4 line (Table 1). A quantitative real-time PCR (qPCR) analysis confirmed low endogenous expression levels of IPPI in peppermint leaves, as four times higher tissue amounts than usual had to be processed to obtain any detectable signal. The IPPI transcript levels in the IPPI4 transgenic line were 1.8-fold higher than in wild-type controls (Fig. 1).

Fig. 1.

Fig. 1.

Open in a new tab

Quantitative real-time PCR analysis of transgenic plants and wild-type controls (fold change vs. control +/− standard deviation). Asterisks indicate the statistical significance level: P-value < 0.05 (*), < 0.01 (**), and < 0.001 (***). Acronyms for transgenes: DXR, 1-deoxy-D-xylulose 5-phosphate reductoisomerase; IPPI, isopentenyl diphosphate isomerase; GPPS, geranyl diphosphate synthase; MFS, (+)-menthofuran synthase. The BD7A family of lines was generated by transforming the MFS7A transgenic line (expression of MFS in antisense orientation) with a construct for the overexpression of DXR (N≥3 ± SD). Expression levels are expressed as fold change of transgene vs. wild-type controls (transgenes: A, IPPI; B, GPPS; C, MFS; D, DXR).

The Grand fir GPPS overexpression construct was integrated into the genome of 22 independently transformed peppermint lines. Corresponding transcript was detected in 12 lines on Northern blots (SI Text), whereas in wild-type controls, which do not contain this particular GPPS gene, the transcript was not detected. Oil yield increases of up to 18% over wild-type controls were measured in some lines expressing GPPS (highest levels in GPPS11). The GPPS11 line also had desirable effects on oil composition (< 5% (+)-menthofuran and < 2% (+)-pulegone; Table 1). A qPCR analysis confirmed relatively high expression levels of the Abies grandis GPPS in the GPPS11 line, while, as expected, only background level amplification of a PCR product was detectable with wild-type controls (Fig. 1).

It is interesting to note that the compositional balance of monoterpenes was affected slightly by overexpressing genes involved in the MEP pathway (e.g., DXPS2 had high (+)-pulegone levels, while IPPI4 and IPPI5 had increased (+)-menthofuran levels). This could potentially be explained by changes in flux distribution through the monoterpene pathway or pleiotropic effects resulting from each transformation event.

Combining High Essential Oil Yield with Favorable Composition.

The overexpression of DXR led to oil yield increases, while the expression of MFS in antisense orientation in transgenic peppermint plants (elite line MFS7A) resulted in desired decreases in the relative amounts of (+)-menthofuran and (+)-pulegone (9). To evaluate if oil yield could be improved while preserving the desirable composition, we transformed the transgenic MFS7A line with an additional construct for the overexpression of DXR. We regenerated 35 transgenic plant lines that contained the DXR overexpression construct, of which six lines had significantly higher oil yields than wild-type controls (Table S1). The essential oil yields in three elite lines, designated BD7A-3, BD7A-4, and BD7A-7, were substantially increased above those in wild-type controls (61%, 53%, and 49%, respectively) (Table 1). Consistently low levels of (+)-menthofuran (≤ 1.9% of the essential oil) and (+)-pulegone (roughly 0.2% of the essential oil) were measured in all of these transgenic lines (Table 1). Based on qPCR assays, the expression levels of MFS were fairly consistent across all BD7A lines at about 0.75-fold the transcript abundance as in wild-type controls. DXR expression levels were significantly increased over wild-type controls in BD7A-24 and BD7A-27, while DXR transcript abundance in all other high yielding transgenics (BD7A-3, BD7A-4, BD7A-7, BD7A-29, and BD7A-33) was similar to that in controls (Fig. 1 and SI Text), indicating a poor correlation of transgene expression level and oil yield. Possible reasons for this phenomenon are discussed below.

Field Testing of Elite Transgenic Lines.

Encouraged by the results obtained with greenhouse-grown transgenics, we performed field trials to evaluate the performance of these lines under commercial growth conditions in the Yakima River Valley growing area of Washington State. The first generation of lines to be advanced to field trials were DXR38 (DXR overexpression line with high oil yield) and MFS7A (MFS antisense-expression line with enhanced oil composition), which were first planted in 2001. Compared with wild-type controls the initial oil yields obtained with the DXR38 line were lower (28% down in 2003), while moderately increased yields were detected in 2004 (11% up) and 2005 (18% up) (Table 2). The oil composition was similar to that of wild-type controls throughout the three growth seasons (2003–2005). Oil yields of MFS7A plants were significantly increased over wild-type controls in all samples (26%, 69%, and 50% increased in 2003, 2004, and 2005, respectively) (Table 2). The composition was similar to that of untransformed controls, with the exception that (+)-menthofuran levels were desirably lower. Fresh hay yields were at 42 to 50 tons per hectare for all years and genotypes (Table 2), indicating that controls and transgenic plants had similar growth characteristics. Three double-transgenics optimized for oil yield and composition were first planted in 2007 (BD7A-3) and 2008 (BD7A-4 and BD7A-7). Highly significant yield enhancements over wild-type controls were determined for BD7A-3 (27% up in 2008 harvest) and BD7A-4 (70 and 78% up in 2008 and 2009 harvests, respectively), while the 26% yield increase of BD7A-7 in 2009 harvest was not significant. The oil composition in all transgenic lines was favorable (similar to wild-type but with significantly lower (+)-menthofuran levels). In 2008, some mint plots were harvested only a few months after establishment. High levels of (+)-menthofuran (> 6% of essential oil) were detected in wild-type controls, whereas in transgenic lines BD7A-4 and BD7A-7 the levels were at < 5% of the oil. Because high (+)-menthofuran levels have also been reported for mature plants grown under low light intensities (cloudy days) or high night temperatures (21–23), the accumulation of undesirable oil components (such as (+)-menthofuran) under adverse weather conditions could potentially be prevented by planting BD7A transgenic lines. It is also important to note that oil yield and composition trends observed in greenhouse-grown lines were substantiated in field trials. Our data also provide evidence that, after a plot establishment period of 1–2 yr, oil yield and composition in transgenic lines are stable for multiple growth seasons.

Table 2.

Field trial evaluation of essential oil yield and composition in elite transgenic peppermint lines and wild-type controls (N≥3 ± SD)

Geno-type Essential Oil Yield
 Hay Yield
 Mono- and Sesquiterpene Composition [% of total volatile terpenes]
Fold Change vs. WT (+/−SD) P-Value (t-test) [kg per hectare](+/−SD) [tons per hectare](+/−SD) Menthone Mentho-furan Menthylacetate Pulegone Menthol beta-Caryo-phyllene Germa-crene D
2003, July 17 Harvest, First Cut (4 replicate plots), 2001 Planting
WT 1.00 (0.17) n.a. 110.2 (18.7) 50.2 (4.0) 21.7 4.8 5.8 1.7 43.2 3.2 4.0
DXR38 0.82 (0.11) 0.110 90.5 (10.6) 44.8 (3.4) 19.9 4.8 6.4 1.6 46.4 3.2 4.1
MFS7A* 1.26 (0.14) 0.005 139.2 (20.2) 45.9 (3.1) 24.6 1.7 5.0 0.8 41.6 3.5 5.2
2004, July 7 Harvest, First Cut (4 replicate plots), 2001 Planting
WT 1.00 (0.14) n.a. 85.9 (11.8) 47.3 (3.4) 23.9 1.4 3.5 0.4 41.9 3.8 4.0
DXR38 1.11 (0.11) 0.280 95.2 (10.1) 41.7 (4.5) 21.4 1.1 3.5 0.3 45.2 2.1 3.8
MFS7A 1.69 (0.17) 0.004 145.2 (24.3) 45.2 (8.1) 23.4 0.6 3.0 0.4 40.3 4.1 4.9
2005, July 12 Harvest, First Cut (4 replicate plots), 2001 Planting
WT 1.00 (0.03) n.a. 127.0 (4.0) 49.3 (4.5) 26.1 2.3 4.0 0.6 40.4 2.7 3.7
DXR38 1.18 (0.13) 0.065 150.2 (20.1) 48.9 (3.1) 25.2 2.3 4.1 0.5 42.8 2.6 3.9
MFS7A 1.50 (0.07) 0.00001 191.1 (13.2) 48.9 (3.8) 27.0 1.0 3.5 0.5 38.8 2.9 4.4
2008, July 18 Harvest, First Cut (3 replicate plots), 2007 Planting
WT 1.00 (0.05) n.a. 90.6 (4.8) 52.5 (1.6) 28.0 2.2 4.0 0.5 39.6 3.0 4.2
BD7A-3 1.27 (0.10) 0.026 115.4 (11.5) 39.0 (3.4) 26.8 1.6 3.4 0.6 36.9 2.5 3.6
2008, August 27 Harvest, First Cut (3 replicate plots), 2008 Planting
WT 1.00 (0.16) n.a. 65.0 (10.8) 18.7 (2.4) 25.0 6.1 5.4 0.4 40.0 2.1 2.7
BD7A-4 1.70 (0.10) 0.007 110.7 (11.1) 20.8 (4.3) 24.9 4.6 5.6 0.3 39.0 1.9 2.5
BD7A-7 1.21 (0.13) 0.180 78.9 (10.4) 16.1 (1.0) 25.1 4.9 4.9 0.2 37.0 2.4 2.4
2009, July 31 Harvest, First Cut (3 replicate plots), 2008 Planting
WT 1.00 (0.18) n.a. 93.3 (16.6) 43.5 (4.0) 24.4 2.4 4.1 0.5 40.3 3.4 3.7
BD7A-4 1.78 (0.09) 0.003 165.5 (10.5) 48.0 (2.7) 23.7 1.2 4.0 0.5 41.1 3.5 3.8
BD7A-7 1.26 (0.09) 0.099 117.8 (10.9) 37.0 (2.9) 24.1 0.9 3.9 0.4 41.7 3.2 4.2
Open in a new tab

For acronyms used as identifiers for transgenic lines see Table 1.

Abbreviations and acronyms: n.a., not applicable; SD, standard deviation; WT, wild-type peppermint (Black Mitcham cultivar).

Introduction of a Chemical Marker for Essential Oil Derived from Transgenic Plants.

Commercial essential oil from field-grown mint plants is collected by steam distillation. This oil does not contain genetic material, because DNA is not volatile and remains in the mint hay residue. However, regulatory agencies in some countries require labeling of products of transgenic provenance. To enable the tracing of oil from transgenic peppermint plants it would thus be advantageous if it contained a chemical marker. Such a marker would have to be present at low levels in the oil, should be easily detectable by standard analytical methods, and should not affect the organoleptic properties of the oil. Commercial peppermint oils grown in various US locations contain readily detectable amounts of (−)-limonene (1.1–2.7% of the essential oil), the precursor to all p-menthane monoterpenes (Fig. S1) but only very small amounts (ranging between 0.07% and 0.25% of the essential oil) of its enantiomer, (+)-limonene (24). The Federal Drug Administration of the United States lists (+)-limonene under the chemicals with “generally recognized as safe” (GRAS) status (25), and enhanced levels of (+)-limonene would thus be a highly suitable chemical marker for transgenic plants. Identification of a gene that encodes a (+)-limonene synthase (PLS) would provide for such a marker gene.

A terpene cyclase structure/function study using (−)-LS from spearmint (10) as the scaffold revealed several potential candidate mutant enzymes that generate largely monoterpenes of the (+)-enantiomeric series in cell-free assays using geranyl diphosphate as substrate (SI Text). One promising candidate (M458C; methionine 458 changed to a cysteine) showed a relative reaction velocity approaching wild-type cyclase values and was thus further evaluated for potential use as a marker gene. GC and chiral GC analyses determined the product composition of the M458C mutant to be 57.3% limonene (5.4∶1 (+/−)), 38.3% α-terpineol (7.3∶1 (+/−)), 4% linalool (4∶1 (+/−)), 1.1% (−)-β-pinene and 1.5% (+)-β-pinene (Fig. 2A). Given that all the assay products of M458C mutant are listed under GRAS status (25), peppermint plants were transformed with a M458C mutant overexpression construct, from which 29 transgenic lines (designated PLS1-29) were regenerated. Within this family of transgenics, six lines had significantly increased (> 50% up) levels of (+)-limonene (Fig. 2 B and C). The highest (+)-limonene levels were detected in the PLS3 line but, for as yet unknown reasons, the concentrations of all (−)-limonene-derived monoterpenes decreased dramatically and the oil yield dropped to 18% of wild-type controls (Fig. 2 B and C). However, four other (+)-limonene-accumulating transgenic lines (PLS10, PLS11, PLS12, and PLS28) had a generally desirable oil composition and oil yields comparable to those of wild-type controls. These data indicate that (+)-limonene can be introduced as a chemical marker, without adverse effects on oil yield.

Fig. 2.

Fig. 2.

Open in a new tab

Introduction of a chemical marker into peppermint essential oil. (A) Product profile obtained with a (+)-limonene synthase (PLS M458C) that was generated by site-directed mutagenesis of spearmint (−)-limonene synthase. (B) Comparison of chiral GC-FID traces of essential oil distilled from the transgenic line PLS3 and wild-type plants. (C) Monoterpene profiles of transgenic plants expressing PLS, and comparison with wild-type controls (N≥3 ± SD).

Strategies for Further Yield Enhancements.

Overall, our metabolic engineering approaches to modulate the expression of genes of the essential oil precursor pathway (up) and those with relevance for oil composition (down) were highly successful. However, transgene expression levels did not always correlate directly with oil yield (Table 1; Fig. 1). Because future improvements will depend on understanding the factors that determine oil yield in peppermint, it should be noted that we have recently made substantial progress in that regard. Using a combination of mathematical modeling and experimental testing (26, 27) we found that, across several transgenic lines and under various growth conditions, oil yield correlated strongly with the total number and developmental distribution patterns of glandular trichomes, the specialized anatomical structures that synthesize and accumulate the oil (28–31). We had previously hypothesized that a “push-pull” mechanism (increased expression of a biosynthetic gene induces the initiation of glandular trichomes) could potentially allow more essential oil to be produced and stored (27). Such a mechanism is only partially supported by the current dataset (transgenics with higher oil yields had increased expression levels of the precursor pathway transgene but there was no tight correlation between the degree of transgene overexpression and oil yield). It is conceivable that the tissue culture manipulations used in the peppermint transformation protocol also have effects on the oil yield capacity of the regenerated plants, independent of the expression levels of the introduced transgene. Our transgenic plants did not show any obvious visible phenotypic differences (leaf and stem morphology) compared to wild-type controls and an important next step in furthering our understanding of the mechanisms controlling essential oil yield in peppermint will thus be to evaluate the density and size of glandular trichomes on leaves of the highest yielding transgenic lines. The mechanisms underlying glandular trichome distribution in peppermint are presently unknown, but positive regulators of trichome development/initiation have been characterized in other plants. These include transcription factors with various DNA binding domains (32–37) and protein–protein interaction partners of these transcriptional regulators (38). An analysis of expressed sequenced tags from isolated peppermint glandular trichome secretory cells indicated the presence of several genes with very high homology to the above-mentioned transcription factors (18), and we are currently testing their utility for affecting glandular trichome numbers and developmental distribution patterns. Our analyses of transgenic peppermint plants with varying essential oil yield and composition have yielded highly valuable insights that now enable the knowledge-based metabolic engineering of other commercially important essential oil plants.

Materials and Methods

Plant Growth Conditions.

Peppermint (Mentha × piperita cv. Black Mitcham) plants were grown on soil (Sunshine Mix LC1, SunGro Horticulture) in a greenhouse with supplemental lighting from sodium vapor lights (850 μmol m-2 s-1 of photosynthetically active radiation at plant canopy level) with a 16 h photoperiod and a temperature cycle of 27 °C/21 °C (day/night). Plants were watered daily with a fertilizer mix (N∶P∶K 20∶20∶20, v/v/v; plus iron chelate and micronutrients). The growth and analysis of field trial plants is detailed in SI Text and supporting information regarding weather conditions, irrigation schedule, and pesticide applications are available in Tables S2–S4.

Construct Design, Plant Transformations, Transformant Selection, and Transcript Analysis.

The pGAdekG/NIb.L parent vector was a gift from J.C. Carrington. It is a derivative of the pGA482 vector developed by G. An’s group (39). The target DNA was amplified from suitable plasmid DNA using primers that introduced a 5′-EcoRI site upstream of the start codon and a 3′-KpnI site downstream of the stop codon. The pGAdekG/NIb.L vector was digested with EcoRI/KpnI, gel-purified, and the EcoRI/KpnI-predigested target cDNA ligated into the vector. The vector used for the overexpression of (+)-limonene synthase was based on the backbone of pCAMBIA1300 (Cambia) with a Gateway cloning cassette from pMDC32 (40). The Agrobacterium-mediated transformation of peppermint leaf discs and regeneration of transgenic plants was performed according to previously published protocols (41, 42). The construction of the BD7A transgenics used three plasmid vectors (pEGAD (43), pRT-GUS (44) and pCAMBIA0380 (AF234290)) and involved several steps outlined in SI Text. To test for the integration of the transgene-containing construct into the peppermint genome we assayed for the activity of the neomycin phosphotransferase (45) in crude protein extracts (9) or amplified the appropriate selectable marker gene from isolated genomic DNA (46) using PCR assays. Preliminary transcript abundance analyses were performed by Northern blotting (according to ref. 9). Elite lines were also assayed by qPCR (according to ref. 27). A more detailed outline of molecular biological methods and a complete list of all primers are available in SI Text.

Micro-Scale Distillation and Terpenoid Analysis.

Leaves were directly (without prior freezing) steam-distilled and solvent-extracted using 10 mL of n-hexane in a condenser-cooled Likens-Nickerson apparatus (47). For the analysis of chiral essential oil constituents (in particular for determining the ratio of (−)-limonene and (+)-limonene in PLS transgenics), the steam-distilled and solvent-extracted oil was injected onto an Agilent Technologies 7890A GC-FID instrument equipped with a Cyclodex B column (J&W Scientific 112-2532; 30 m × 250 μm × 0.25 μm). GC settings were as follows: inlet 250 °C, split 2∶1, split flow 4 mL/ min, column flow (He) 2.0 mL/ min, FID detector at 250 °C (H2 flow at 30 mL/ min, air flow at 400 mL/ min, makeup flow (He) at 25 mL/ min). The GC oven program had an initial temperature of 62 °C, which was held for 1 min, a slow ramp to 65 °C at 0.1 °C per min, a fast ramp to 200 °C at 50 °C, a hold at the final temperature for 10 min. Student’s t tests to assess statistical significance of differences between control and experimental samples were performed in Microsoft Excel.

Characterization of a (+)-Limonene Synthase Generated from Spearmint (−)-Limonene Synthase by Site-Directed Mutagenesis.

Mutagenesis of (−)-(4S)-limonene synthase utilized a truncated form of the cDNA in which the transit peptide coding sequence had been removed and an ATG inserted directly upstream to generate the LS-M57 inserted into the pSBET vector (48), a construct that facilitated both recombinant expression and enzyme activity (49, 50). One mutant clone (M458C) yielded substantial amounts of (+)-limonene (SI Text). The clone was overexpressed in Escherichia coli and the recombinant protein purified (SI Text). Typical enzyme assays contained 100 μM geranyl diphosphate, 7.5 mM MgCl2, 50 mM MOPSO at pH 7.0, 10 % (v/v) glycerol, and 10 mM DTT in a 500 μL volume. Assays were carried out for 5–30 min at 31 °C and terminated by extraction with 500 μL pentane. The pentane layer was used directly for chiral GC analyses as described above.

Supplementary Material

Supporting Information
supp_108_41_16944__index.html (1.1KB, html)

Acknowledgments.

We would like to thank Julia Gothard-Szamosfalvi and Sue Vogtman for growing plants and Darlene Croteau, Dr. Kumkum Rani, Lyuba Yurgel, and Blake Herron for technical assistance. We would also like to thank Dr. R. E. B. Ketchum for providing the pREBKs vector. Financial support from the Mint Industry Research Council (grants to B.M.L. and R.B.C.), and the US Department of Energy (DE-FG02-09ER16054 to B.M.L.) is greatly appreciated.

Footnotes

Conflict of interest statement: Cocorresponding author B. M. Lange serves as chief scientific officer for Ajuga Biosciences, a biotechnology company using mint as a host for the production of terpenoid-based fine chemicals. However, it is important to note that Ajuga Biosciences did not sponsor the current study, nor did the company have any involvement in the generation or interpretation of the data.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1111558108/-/DCSupplemental.

References

  • 1.Morris MA. In: Mint: The Genus Mentha. Laurence BM, editor. Boca Raton, FL: CRC Press; 2006. pp. 87–136. [Google Scholar]
  • 2.Morris M, Robbins E, Todd AM. Mint landscape: From field to flavor. The mint industry’s latest trends, challenges and outlook. Perfum Flavor. 2005;30:46–51. [Google Scholar]
  • 3.Murray MJ, Lincoln D, Marble PM. Oil composition of Mentha aquatica—M. spicata F1 hybrids in relation to the origin of M. × piperita. Can J Genet Cytol. 1972;14:13–29. [Google Scholar]
  • 4.Wildung MR, Croteau RB. Genetic engineering of peppermint for improved essential oil composition and yield. Transgenic Res. 2005;14:365–72. doi: 10.1007/s11248-005-5475-2. [DOI] [PubMed] [Google Scholar]
  • 5.Croteau R, Davis EM, Ringer KL, Wildung MR. (−)-Menthol biosynthesis and molecular genetics. Naturwiss. 2005;92:562–577. doi: 10.1007/s00114-005-0055-0. [DOI] [PubMed] [Google Scholar]
  • 6.McCaskill D, Croteau R. Monoterpene and sesquiterpene biosynthesis in glandular trichomes of peppermint (Mentha × piperita) rely exclusively on plastid-derived isopentenyl diphosphate. Planta. 1995;197:49–56. [Google Scholar]
  • 7.Eisenreich W, Sagner S, Zenk MH, Bacher A. Monoterpenoid essential oils are not of mevalonoid origin. Tetrahedron Lett. 1997;38:3889–3892. [Google Scholar]
  • 8.Lange BM, Croteau R. Isoprenoid biosynthesis via a mevalonate-independent pathway in plants: Cloning and heterologous expression of 1-deoxyxylulose-5-phosphate reductoisomerase from peppermint. Arch Biochem Biophys. 1999;365:170–174. doi: 10.1006/abbi.1999.1168. [DOI] [PubMed] [Google Scholar]
  • 9.Mahmoud SS, Croteau RB. Metabolic engineering of essential oil yield and composition in mint by altering expression of deoxyxylulose phosphate reductoisomerase and menthofuran synthase. Proc Natl Acad Sci USA. 2001;98:8915–8920. doi: 10.1073/pnas.141237298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Colby SM, Alonso WR, Katahira E, McGarvey DJ, Croteau R. 4S-Limonene synthase from the oil glands of spearmint (Mentha spicata): cDNA isolation, characterization and bacterial expression of the catalytically active monoterpene cyclase. J Biol Chem. 1993;268:23016–23024. [PubMed] [Google Scholar]
  • 11.Diemer F, Caissard JC, Moja S, Calchat JC, Jullien F. Altered monoterpene composition in transgenic mint following the introduction of 4S-limonene synthase. Plant Physiol Biochem. 2001;39:603–614. [Google Scholar]
  • 12.Mahmoud SS, Williams M, Croteau R. Cosuppression of limonene-3-hydroxylase in peppermint promotes accumulation of limonene in the essential oil. Phytochemistry. 2004;65:547–554. doi: 10.1016/j.phytochem.2004.01.005. [DOI] [PubMed] [Google Scholar]
  • 13.Krasnyansky S, May RA, Loskutov A, Ball TM, Sink KC. Transformation of the limonene synthase gene into peppermint (Mentha piperita L.) and preliminary studies on the essential oil profiles of single transgenic plants. Theor Appl Genet. 1999;99:676–682. doi: 10.1007/s001220051284. [DOI] [PubMed] [Google Scholar]
  • 14.Lupien S, Karp F, Wildung M, Croteau R. Regiospecific cytochrome P450 limonene hydroxylases from mint (Mentha) species: cDNA isolation, characterization, and functional expression of (−)-4S-limonene-3-hydroxylase and (−)-4S-limonene-6-hydroxylase. Arch Biochem Biophys. 1999;368:181–192. doi: 10.1006/abbi.1999.1298. [DOI] [PubMed] [Google Scholar]
  • 15.Bertea CM, Schalk M, Karp F, Maffei M, Croteau R. Demonstration that menthofuran synthase of mint (Mentha) is a cytochrome P450 monooxygenase: cloning, functional expression, and characterization of the responsible gene. Arch Biochem Biophys. 2001;390:279–86. doi: 10.1006/abbi.2001.2378. [DOI] [PubMed] [Google Scholar]
  • 16.McConkey ME, Gershenzon J, Croteau RB. Developmental regulation of monoterpene biosynthesis in the glandular trichomes of peppermint. Plant Physiol. 2000;122:215–24. doi: 10.1104/pp.122.1.215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lange BM, Wildung MR, McCaskill D, Croteau R. A family of transketolases that directs isoprenoid biosynthesis via a mevalonate-independent pathway. Proc Natl Acad Sci USA. 1998;95:2100–2104. doi: 10.1073/pnas.95.5.2100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Lange BM, et al. Probing essential oil biosynthesis and secretion by functional evaluation of expressed sequence tags from mint glandular trichomes. Proc Natl Acad Sci USA. 2000;97:2934–2939. doi: 10.1073/pnas.97.6.2934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Burke C, Croteau R. Geranyl diphosphate synthase from Abies grandis: cDNA isolation, functional expression, and characterization. Arch Biochem Biophys. 2002;405:130–136. doi: 10.1016/s0003-9861(02)00335-1. [DOI] [PubMed] [Google Scholar]
  • 20.Odell JT, Nagy F, Chua NH. Identification of DNA sequences required for activity of the cauliflower mosaic virus 35S promoter. Nature. 1985;313:810–812. doi: 10.1038/313810a0. [DOI] [PubMed] [Google Scholar]
  • 21.Burbott AJ, Loomis WD. Effects of light and temperature on the monoterpenes of peppermint. Plant Physiol. 1967;42:20–28. doi: 10.1104/pp.42.1.20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Clark RJ, Menary RC. Environmental effects on peppermint, I. Effect of day length, photon flux density, night temperature and day temperature on the yield and composition of peppermint oil. Aust J Plant Physiol. 1980;7:685–692. [Google Scholar]
  • 23.Court WA, Roy RC, Pocs R. Effect of harvest date on the yield and quality of the essential oil of peppermint. Can J Plant Sci. 1993;73:815–824. [Google Scholar]
  • 24.Mosandl A, et al. Stereoisomeric flavor compounds. 48. Chirospecific analysis of natural flavors and essential oil using multidimensional gas chromatography. J Agric Food Chem. 1991;39:1131–1134. [Google Scholar]
  • 25.Hall RL, Oser BL. Recent progress in the consideration of flavoring ingredients under the food additives amendment. III. GRAS Substances. Food Technol. 1965;19:151–197. [Google Scholar]
  • 26.Rios-Estepa R, Turner GW, Lee JM, Croteau RB, Lange BM. A systems biology approach identifies biochemical mechanisms regulating monoterpenoid essential oil composition in peppermint. Proc Natl Acad Sci USA. 2008;105:2818–2823. doi: 10.1073/pnas.0712314105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Rios-Estepa R, Lange I, Lee JM, Lange BM. Mathematical modeling-guided evaluation of biochemical, developmental, environmental and genotypic determinants of essential oil composition and yield in peppermint leaves. Plant Physiol. 2010;152:2105–2119. doi: 10.1104/pp.109.152256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Amelunxen F. Electron microscopy analysis of glandular trichomes of Mentha piperita L. Planta Med. 1965;13:457–473. (Translated from German) [Google Scholar]
  • 29.Amelunxen F, Wahlig T, Arbeiter H. Detection of essential oil in isolated glandular trichomes of Mentha piperita L. Pflanzenphysiol. 1969;61:68–72. (Translated from German) [Google Scholar]
  • 30.Gershenzon J, Maffei M, Croteau R. Biochemical and histochemical localization of monoterpene biosynthesis in the glandular trichomes of spearmint (Mentha spicata) Plant Physiol. 1989;89:1351–1357. doi: 10.1104/pp.89.4.1351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.McCaskill D, Gershenzon J, Croteau R. Morphology and monoterpene biosynthetic capabilities of secretory cell clusters isolated from glandular trichomes of peppermint (Mentha × piperita L.) Planta. 1992;187:445–454. doi: 10.1007/BF00199962. [DOI] [PubMed] [Google Scholar]
  • 32.Oppenheimer DG, Herman PL, Sivakumaran S, Esch J, Marks MD. A myb gene required for leaf trichome differentiation in Arabidopsis is expressed in stipules. Cell. 1991;67:483–493. doi: 10.1016/0092-8674(91)90523-2. [DOI] [PubMed] [Google Scholar]
  • 33.Wang S, et al. Control of plant trichome development by a cotton fiber MYB gene. Plant Cell. 2004;16:2323–2334. doi: 10.1105/tpc.104.024844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Machado A, Wu Y, Yang Y, Llewellyn DJ, Dennis ES. The MYB transcription factor GhMYB25 regulates early fiber and trichome development. Plant J. 2009;59:52–62. doi: 10.1111/j.1365-313X.2009.03847.x. [DOI] [PubMed] [Google Scholar]
  • 35.Payne CT, Zhang F, Lloyd AM. GL3 encodes a bHLH protein that regulates trichome development in Arabidopsis through interaction with GL1 and TTG1. Genetics. 2000;156:1349–1362. doi: 10.1093/genetics/156.3.1349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ma D, et al. Isolation and characterization of AaWRKY1, an Artemisia annua transcription factor that regulates the amorpha-1,4-diene synthase gene, a key gene of artemisinin biosynthesis. Plant Cell Physiol. 2009;50:2146–2161. doi: 10.1093/pcp/pcp149. [DOI] [PubMed] [Google Scholar]
  • 37.Rerie WG, Feldmann KA, Marks MD. The GLABRA2 gene encodes a homeo domain protein required for normal trichome development in Arabidopsis. Genes Dev. 1994;8:1388–1399. doi: 10.1101/gad.8.12.1388. [DOI] [PubMed] [Google Scholar]
  • 38.Walker AR, et al. The TTG1 (transparent testa glabra 1) locus which regulated trichome differentiation and anthocyanin biosynthesis in Arabidopsis encodes a WD40-repeat protein. Plant Cell. 1999;11:1337–1350. doi: 10.1105/tpc.11.7.1337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.An G. Binary Ti vectors for plant transformation and promoter analysis. Methods Enzymol. 1987;153:293–305. [Google Scholar]
  • 40.Curtis MD, Grossniklaus U. A Gateway cloning vector set for high-throughput functional analysis of genes in planta. Plant Physiol. 2003;133:462–469. doi: 10.1104/pp.103.027979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Niu X, Lin K, Hasegawa PM, Bressan RA, Weller SC. Transgenic peppermint (Mentha × piperita L.) plants obtained by cocultivation with Agrobacterium tumefaciens. Plant Cell Rep. 1998;17:165–171. doi: 10.1007/s002990050372. [DOI] [PubMed] [Google Scholar]
  • 42.Niu X, et al. Factors affecting Agrobacterium tumefaciens -mediated transformation of peppermint. Plant Cell Rep. 2000;19:304–310. doi: 10.1007/s002990050017. [DOI] [PubMed] [Google Scholar]
  • 43.Cutler SR, Ehrhardt DW, Griffitts JS, Somerville CR. Structures in cells of Arabidopsis at a high frequency. Proc Natl Acad Sci USA. 2000;97:3718–3723. doi: 10.1073/pnas.97.7.3718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Carrington JC, Freed DD. Cap-independent enhancement of translation by a plant potyvirus 5′-nontranslated region. J Virol. 1990;64:1590–1597. doi: 10.1128/jvi.64.4.1590-1597.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Platt SG, Yang NS. Dot assay for neomycin phosphotransferase activity in crude cell extracts. Anal Biochem. 1987;162:529–535. doi: 10.1016/0003-2697(87)90429-5. [DOI] [PubMed] [Google Scholar]
  • 46.Porebski S, Bailey LG, Baum BR. Modification of a CTAB DNA extraction protocol for plant containing high polysaccharide and polyphenol components. Plant Mol Biol Rep. 1997;15:8–15. [Google Scholar]
  • 47.Gershenzon J, McConkey ME, Croteau RB. Regulation of monoterpene accumulation in leaves of peppermint. Plant Physiol. 2000;122:205–14. doi: 10.1104/pp.122.1.205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Schenk PM, Baumann S, Mattes R, Steinbiss H. Improved high-level expression system for eukaryotic genes in Escherichia coli using T7 RNA polymerase and rare ArgtRNAs. BioTechniques. 1995;19:196–200. [PubMed] [Google Scholar]
  • 49.Williams DC, McGarvey DJ, Katahira EJ, Croteau R. Truncation of limonene synthase preprotein provides a fully active ‘pseudomature’ form of this monoterpene cyclase and reveals the function of the amino-terminal arginine pair. Biochemistry. 1998;37:12213–12220. doi: 10.1021/bi980854k. [DOI] [PubMed] [Google Scholar]
  • 50.Schalk M, Croteau R. A single amino acid substitution (F363I) converts the regiochemistry of the spearmint (−)-limonene hydroxylase from a C6- to a C3-hydroxylase. Proc Natl Acad Sci USA. 2000;97:11948–11953. doi: 10.1073/pnas.97.22.11948. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_108_41_16944__index.html (1.1KB, html)
1111558108_pnas.1111558108_SI.pdf (1.5MB, pdf)
1111558108_ST01.docx (28.4KB, docx)
1111558108_ST02.xls (1,021.5KB, xls)
1111558108_ST03.xls (30KB, xls)
1111558108_ST04.xls (90KB, xls)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES