Steroidal glycoalkaloids in Solanum chacoense

Phytochemistry 75 (2012) 32–40

Contents lists available at SciVerse ScienceDirect

Phytochemistry journal homepage: www.elsevier.com/locate/phytochem

Steroidal glycoalkaloids in Solanum chacoense Alice M. Mweetwa a,1, Danielle Hunter a, Rebecca Poe a, Kim C. Harich b, Idit Ginzberg c, Richard E. Veilleux a, James G. Tokuhisa a,⇑ a

Department of Horticulture, Virginia Tech, Blacksburg, VA 24061, USA Department of Biochemistry, Virginia Tech, Blacksburg, VA 24061, USA c Institute of Plant Sciences, ARO, The Volcani Center, Bet Dagan 50250, Israel b

a r t i c l e

i n f o

Article history: Received 19 September 2011 Received in revised form 28 November 2011 Available online 2 January 2012 Keywords: Solanum chacoense Solanaceae Chaco potato Developmental profile Steroidal glycoalkaloids Leptines hmg pss sgt

a b s t r a c t Potato (Solanum tuberosum L.), a domesticated species that is the fourth most important world agricultural commodity, requires significant management to minimize the effects of herbivore and pathogen damage on crop yield. A wild relative, Solanum chacoense Bitt., has been of interest to plant breeders because it produces an abundance of novel steroidal glycoalkaloid compounds, leptines and leptinines, which are particularly effective deterrents of herbivory by the Colorado potato beetle (Leptinotarsa decemlineata Say). Biochemical approaches were used in this study to investigate the formation and accumulation of SGAs in S. chacoense. SGA contents were determined in various organs at different stages of organ maturity during a time course of plant development. Leptines and leptinines were the main contributors to the increased levels in SGA concentration measured in the aerial versus the subterranean organs of S. chacoense accession 8380-1. Leptines were not detected in aboveground stolons until the stage where shoots had formed mature chlorophyllous leaves. To gain insights into SGA biosynthesis, the abundance of SGAs and steady-state transcripts of genes coding for enzymes of the central terpene and SGA-specific pathways in various plant organs at anthesis were compared. For two genes of primary terpene metabolism, transcript and SGA abundances were correlated, although with some discrepancies. For genes associated with SGA biosynthesis, transcripts were not detected in some tissues containing SGAs; however these transcripts were detected in the progenitor tissues, indicating the possibility that under our standard growth conditions, SGA biosynthesis is largely limited to highly proliferative tissues such as shoot, root and floral meristems. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Steroidal glycoalkaloids (SGAs) are a class of plant natural products found in many species of the genus Solanum including potato and tomato (Milner et al., 2011). They are also found in the genus Veratrum of the monocotyledon family Melanthiaceae (Kaneko et al., 1977). SGAs have been shown to function as defense compounds in non-host resistance by Solanaceous plants against herbivores (Tingey, 1984) and eukaryotic microbial pathogens (Deahl et al., 1973; Fewell and Roddick, 1993). Their biological activity has been attributed to two modes of action: inhibition of (acetyl)cholinesterase activity (Roddick, 1989) and disruption of the eukaryotic cell membrane structure through binding with the sterol component of the membranes (Bouarab et al., 2002). ⇑ Corresponding author. Address: 503 Latham Hall, Department of Horticulture, Virginia Tech, Blacksburg, VA 24060, USA. Tel.: +1 540 231 5653; fax: +1 540 231 3347. E-mail address: [email protected] (J.G. Tokuhisa). 1 Present address: School of Agricultural Science, University of Zambia, Lusaka, Zambia. 0031-9422/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.phytochem.2011.12.003

The biological properties of SGAs can adversely affect humans and are the reason for limits on their abundance in the tubers of cultivated potato. Commercial potato breeders reject cultivars that have SGA concentrations in the tubers exceeding 0.2 lmol g 1 fr. wt. However, SGAs are present in the ‘‘skins’’ of potato tubers (Krits et al., 2007) and unripe tomato fruit (Moco et al., 2007) and give a characteristic bitterness that contributes to the organoleptic properties of the cooked potato tuber and unripe tomato fruit. Typically the SGA levels found in supermarket potatoes are 0.03–0.06 lmol g 1 fr. wt. In other tissues of cultivated and wild potatoes, SGAs accumulate to various and typically greater levels (Nema et al., 2008). For instance, SGAs are present in the leaves of cultivated potato at concentrations up to 1.7 lmol g 1 fr. wt. and in wild potato leaves to 8.5 lmol g 1 fr. wt. (Gregory et al., 1981; Friedman and Dao, 1992). Certain accessions of the diploid wild potato species Solanum chacoense have some of the highest contents of SGAs among wild potato species (Sinden et al., 1986b). Whereas a-chaconine (1) and a-solanine (2) are the SGAs found in cultivated potato (Fig. 1), S. chacoense produces leptinines (3,4) and leptines (5,6), unique classes of SGAs that are the hydroxylated and O-acetylated

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

β -D -Glc-(1-3)

β - D -Glc-(1-3)

α - L -Rha-(1-4) α -L -Rha-(1-2)

β - D -Glc-(1-4) -- β -D -Gal-O-

β -D -Gal-O-

β - D -Glc-O-

α - L -Rha-(1-2)

Chacotriosyl

33

Solatriosyl

β -D -Glc-(1-2) Commertetraosyl

Fig. 1. Structures and molecular weight of the major SGAs of Solanum chacoense accession 8380-1. The basic solanidine aglycone is modified by hydroxylation and Oacetylation at C23 (R2) and named as indicated. The glycosides are formed by the addition of a chacotriosyl, solatriosyl or commertetraosyl group at C3 (R1).

derivatives, respectively, of a-chaconine (1) and a-solanine (2; Kuhn and Löw, 1961). Leptines (5,6) and leptinines (3,4) are absent from cultivated potato varieties and even among wild species this class of SGAs has a limited distribution (Distl and Wink, 2009). These derivative compounds have been shown to be more effective than the parental SGAs as deterrents against the Colorado potato beetle (Sinden et al., 1988). Furthermore, leptines (5,6) and leptinines (3,4) accumulate only in the aerial parts of the plant. S. chacoense was targeted as an experimental system to explore the role of SGAs in the natural chemical defense system of wild potato. In this study it was assumed that the reduction of SGA content to ensure the safety of tubers for human consumption (Johns and Alonso, 1990) has also reduced the herbivore and pathogen defenses of cultivated potato. An understanding of the formation and function of SGAs in an entire plant whose genotype has been defined by natural selection rather than domestication will allow us to more fully exploit the defense potential of SGAs in cultivated potato. As an initial investigation, described herein are: (i) a time course survey of SGA concentration and composition in various organs of the developing plant, and (ii) transcript levels of genes coding for selected isoprenoid biosynthetic enzymes including some genes directly involved with SGA biosynthesis.

2. Results and discussion 2.1. Qualitative differences in the SGA profile of S. chacoense Different accessions of S. chacoense vary in SGA profile and content (Ronning et al., 2000). In this study, the high leptine accumulating line S. chacoense accession 8380-1 was chosen to investigate the chemical defense biochemistry because it has been used previously for analytical and ecological studies (Sinden et al.,

1986a). In this study, the accession under controlled growing conditions was cultivated to analyze SGA accumulation at the organ level and to monitor the dynamics of these plant defense compounds at different developmental stages. SGAs were extracted from plant organs at various developmental stages (Table 1) and fractionated by reversed phase HPLC with monitoring at A202nm (Fig. 2). The chromatographic profiles differed most significantly between the aerial (flowers and leaves) and subterranean (roots and tubers) organ extracts. However, all samples had two peaks with retention times (RTs) of 7.9 and 8.4 min that co-migrated with authentic standards of a-solanine (2) and a-chaconine (1), respectively. To confirm the identity of the additional HPLC peaks of the tissue extracts, leaf and root samples as well as subfractions of the HPLC eluate of leaf samples were analyzed by LC–MS. For the major peaks of the leaf extract, the values of the molecular [M + H]+ and fragment ions were consistent with those observed for a-chaconine (1), a-solanine (2), leptines (5,6) and their closely related compounds, leptinines (3,4) (Table 2; Distl and Wink, 2009). The specific 8380-1 accession used has been maintained through clonal propagation and has been the source material for analyses of leptine I (5) and II (6) and leptinine I (3) and II (4) by thin layer chromatography and HPLC (Sinden et al., 1986b; Kowalski et al., 2000), and for the characterization of their respective aglycones by NMR and GC–MS (Lawson et al., 1992, 1997). Taken together, the identities in leaf sample extracts of leptinine II (4; RT 4.4 min), leptinine I (3; RT 4.7 min), a-solanine (2; RT 7.9 min), a-chaconine (1; RT 8.4 min), leptine II (6; RT10.6 min) and leptine I (5; RT 11.0 min) were confirmed. Root extracts analyzed by HPLC exhibited a distinct peak at RT 7.7 min. LC–MS analyses indicated m/z values of 1046.4 for the molecular ion [M + H]+, and 398, 560, and 722 for the fragment ions. This pattern was identical to the LC–MS characterization of dehydrocommersonine (7; Distl and Wink, 2009). Accession 8380-1 has

34

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

Table 1 Organs harvested as indicated (⁄) at six developmental stages of S. chacoense. Stage

DATa

Roots

Early Vegb Mid Veg Late Veg Anthesis 26-DPAc 65-DPA

30 44 58 71 97 136

⁄

⁄

⁄

⁄

⁄

⁄

⁄

Tubers

Leaves Mature

a b c

Stolons Expand

Emerge

Below

Flowers Above

Mature

Stems Emerge ⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄

⁄ ⁄ ⁄

⁄

⁄

⁄

⁄

⁄

⁄ ⁄

DAT: Days after transplant. Veg: Vegetative. DPA: Days post-anthesis.

been shown to accumulate dehydrocommersonine (7) in root cultures in addition to the a-chaconine (1) and a-solanine (2) that are found in all parts of the plant (Zacharius and Osman, 1977). Taken together, this identification of seven SGAs (Fig. 2) in 8380-1 was consistent with previous characterizations of the accession.

2.2. Patterns of SGA distribution and content in aerial versus subterranean organs Total SGA levels varied with both tissue type and developmental stage (Fig. 3A, Suppl. Table 1). Two accumulation patterns were evident. Tubers, roots and stems had small concentrations of SGAs that did not change during the developmental time course. In contrast, leaves had an initial lag followed by a linear increase in SGA accumulation. Because the modifications associated with leptine (5,6) and leptinine (3,4) formation at C23 probably occur with an aglycone precursor that is used for all SGAs found in S. chacoense (Fig. 1; Lawson et al., 1993; Kuhn and Löw, 1961), a-chaconine (1) plus a-solanine (2) content were plotted separately from leptine plus leptinine content for the different organs (Fig. 3B and C; Suppl. Tables 2 and 3). The separation of these two classes of SGAs highlighted the observation that the increased SGA content during aerial tissue development was due to increases in leptine and leptinine content while a-chaconine (1) and a-solanine (2) contents remained relatively constant. Roots had the lowest SGA concentrations observed, with a range of 1.0–8.3 lmol g 1 dry wt across the developmental stages (Fig. 3A). In general there was no significant change in SGA accumulation with time and there was no major change in the proportions of the three SGAs detected, a-chaconine (1), a-solanine (2) and dehydrocommersonine (7) in roots (data not shown). Tubers were first harvested at the late vegetative stage of plant development and showed total SGA concentrations ranging from 7.0 to 15.0 lmol g 1 dry wt. (Fig. 3A). SGA levels increased slightly from the late vegetative stage to 26-d post-anthesis and then declined to 11.9 lmol g 1 dry wt at 65-d post-anthesis. All changes in a-chaconine (1) and a-solanine (2) content in the tubers maintained a 1.7:1 (±0.4, n = 8) molar ratio of a-chaconine to a-solanine.

2.3. Pattern of SGA distribution during the transition from underground stolon to aboveground shoot Stolons, lateral outgrowths of axillary meristems in the basal region of the stem that originate both below- and above-ground were also examined. Depending on the photoperiod and growth conditions, these grow agravitropically below-ground and produce tubers under short day conditions or grow upright, emerging from the soil to form aerial shoots. The stolon developmental program is mediated by both hormones and light (Kumar and Wareing, 1972).

Because stolons originate in the stem, detection of leptines (3– 6) were expected; however, the stolon profile was simple like in the tuber, consisting of only a-chaconine (1) and a-solanine (2; Fig. 4A) yet the total SGA concentrations in below- and aboveground stolons were between 20 and 70 lmol g 1 dry wt (Fig. 5), more typical of SGA concentrations found in aerial tissue extracts. The a-chaconine (1) to a-solanine (2) ratios for below- and aboveground stolons was 1.7:1 (±0.1, n = 15) and 1.8:1 (±0.1, n = 4), respectively, and were more similar to the tuber ratio of 1.7:1. Stolon SGA abundance varied inconsistently across plant developmental stages and between replicate plant samples (Fig. 5). Light has been shown to influence SGA accumulation in S. chacoense and other potato species, and light gradients and plant shading under our growth conditions may have contributed to the observed variation (Deahl et al., 1991; Lafta and Lorenzen, 2000). To determine when leptines and leptinines (3–6) appear in stolon development, the transition of belowground stolons into aerial shoots was monitored (Fig. 4A–D). The underground stolons consisted of achlorophyllous branched stems with emerging roots and leaf primordia. As the stolons emerged from the soil, they began to green at the apex. The first leaves were small but subsequent leaves expanded to sizes that were more comparable to those that developed on the main stem. As shown in Fig. 4D, leptines (3–6) were detected only late in the transition at a stage where the first leaves were fully expanded and well beyond the initial greening and development of chlorophyllous shoots. 2.4. Patterns of SGA distribution and content in aerial organs during plant development Among the aboveground tissues, stems had the smallest total SGA concentrations ranging from 6 to 13 lmol g 1 dry wt (Fig. 3A). Although the total SGA abundance did not change significantly across the developmental stages, the proportion of leptines and leptinines (3–6) relative to a-chaconine and a-solanine (1,2) increased (Fig. 3A–C). The a-chaconine to a-solanine ratio in stem tissue extracts was 1.1:1 (±0.2, n = 27). Leaf samples were pooled at each plant developmental stage by leaf developmental stage, either emerging (with shoot meristem), expanding, or mature leaves. At the early- and mid-vegetative stages, the total concentration of SGAs in the expanding and mature leaves was similar and did not change significantly with time (Fig. 3A). However, coincident with the change from a 10- to a 14-h light photoperiod at 65 DAT to induce flowering, the emerging leaves accumulated more SGAs at each subsequent plant developmental stage, and this increase was reflected in the same leaves as they were sampled at expanding and mature stages later in plant development. There was no time point where the SGA content of expanding or mature leaves exceeding the content of the emerging leaves. This trend continued through the last sampling at 65-d postanthesis where between anthesis and 65-d post-anthesis stages,

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

35

Fig. 2. HPLC chromatograms (A202 nm) of various tissues of S. chacoense at anthesis. SGAs were extracted from (A) flowers, (B) leaves, (C) roots, and (D) tubers, and fractionated by reversed phase chromatography. The compound identities were confirmed by LC–MS analysis of the same extracts and fractions thereof following HPLC separation. The peaks correspond to (u) unknown, (4) leptinine II, (3) leptinine I, (2) a-solanine, (1) a-chaconine, (6) leptine II, (5) leptine I, and (7) dehydrocommersonine.

SGA levels had increased 2.4-, 3.0- and 1.4-fold in emerging, expanding, and mature leaves, respectively. The increase was due to leptine and leptinine accumulation (3–6; Fig. 3C). The a-chaconine (1) to a-solanine (2) ratio in leaf tissue extracts was 1.0:1 (±0.3, n = 48).

Although the sexual reproductive tissues were not abundant in S. chacoense, we compared SGA content at anthesis in a developmental series of inflorescence maturation. The greatest SGA concentration of 85.6 lmol g 1 dry wt (n = 1) was measured in emerging inflorescences with lower levels in the expanding

36

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

Table 2 LC–MS identification of SGAs in extracts and HPLC fractions of S. chacoense leaves.

a b

LC–MS Retention time (min)

Major fragment ions and molecular ion (m/z)

10.83 11.42 17.97 18.92 19.98 20.30

414.5, 414.5, 398.5, 398.5, 456.5, 456.5,

576.6, 576.6, 560.6, 560.6, 618.6, 618.6,

722.6, 704.6, 706.7, 706.7, 764.6, 764.6,

738.7, 722.8, 722.7, 852.8 780.7, 910.8

Compound 884.8 868.7 868.8 926.8

Leptinine II (4)a Leptinine I (3)a a-Solanine (2) a-Chaconine (1) Leptine II (6)b Leptine I (5)b

Detected in HPLC fractions collected prior to a-solanine elution. Detected in HPLC fractions collected after a-chaconine elution.

Table 3 Oligonucleotide primer pairs for semi-quantitative PCR of cDNA derived from RNA extractions of various tissues of S. chacoense. Primer

Target gene

Sequence

GenBank identifier

Amplified fragment length (bp)

1HMG1 2HMG1 1HMG2 2HMG2 1HMG3 2HMG3 1SGT1 2SGT1 1SGT2 2SGT2 1ActStuS 2ActStuT 1SQS1 2SQS1

hmg1 hmg1 hmg2 hmg2 hmg3 hmg3 sgt1 sgt1 sgt2 sgt2 PoAc101 PoAc101 pss1 pss1

TCTTGTGGAGCTGAACATGC AAGCAAGCTGACTGTGATGC TTGTTCGTGAAGGTCCGTG TTATGTCTTTGCTATGTTAG TGTGTCAAGGTGGATGACC TAAGCACCACTGCATACCAG TCTACAACGAGAAGGTAGTC TGAGAGCAGTGAGATTGTTC ATGCAGTGGAAGAAGGTGGG ACATACCGAAACTTGAATCG TGAGTTACCAGATGGTCAGG TCCTTGCTCATACGATCAGC CGAGGAGAACTCGGTTAAGG ATAGCACATAGCTAATGTCCC

GI: 169484

267

GI: 12082119 245 GI: 7415989

249

GI: 82802846 250 GI: 78191093 240 GI: 21533

259

Unpublished

295

inflorescences (50.9 ± 5.7 lmol g 1 dry wt) followed by open flowers (24.1 lmol g 1 dry wt, n = 1; Suppl. Table 1). The a-chaconine (1) to a-solanine (2) ratio in floral extracts was 1.3:1 (±0.2, n = 6). 2.5. Patterns of SGA distribution and content in tissues at different stages of maturity These data, taken together, demonstrate patterns of SGA accumulation in leaf and floral development where the youngest tissues of an organ have the greatest abundance of SGAs while the maturing tissues having progressively less SGA content. As nascent cells expand and mature, increasing in dry wt, the decreased SGA concentrations (per g dry wt, Fig. 3, Suppl. Table 1) in maturing tissues suggest there is little or no additional SGA accumulation following the initial biosynthesis in proliferative tissues. A pattern of SGA biosynthesis restricted to the proliferative tissues could explain the detection of leptines and leptinines (3–6) at late stages of the transition of belowground stolons to aboveground shoots. If leptine and leptinine (3–6) formation was only induced at a particular developmental stage in the apical tissue and extracted with an additional 4 cm of stolon, the leptines and leptinines (3–6) would not appear in the SGA extract profiles until the tissue constituted a majority of the harvested organ. The a-chaconine (1) to a-solanine (2) ratios varied between organs. Stems and leaves had a ratio of 1.0:1, floral tissue had a ratio of 1.3:1 whereas tubers and belowand above-ground stolons had a ratio closer to 1.7:1. The ratios were inversely correlated with the proportion of leptines and leptinines (3–6) in the extracts; at anthesis leptines and leptinines (3– 6) represented 66%, 28%, and 0% of the total SGAs in expanding leaves, expanding flowers and tubers, respectively. These results suggest there is an effect of leptine and leptinine (3–6) formation on the a-chaconine (1) to a-solanine (2) ratio.

Fig. 3. Time course of SGA accumulation during development of S. chacoense. Stolons were excised from mature plants, transplanted to fresh potting mix and allowed to root and establish. Organs were harvested on various days after transplant (DAT) at stages defined as early vegetative (30 DAT), mid-vegetative (44 DAT), late vegetative (58 DAT), anthesis (71 DAT), 26-days post-anthesis (97 DAT), and 65-days post-anthesis (136 DAT). The photoperiod was adjusted from short day (10 h) to long day (14 h) at 65 DAT to induce flowering. The harvested organs included emerging flowers (yellow-filled triangle), mature flowers (yellow-filled circle), emerging leaves (light green, triangle), expanding leaves (olive-green, square), mature leaves (dark green, diamond), stem (orange, horizontal bar), belowground stolons (black, horizontal bar), roots (brown, ‘‘x’’), and tubers (tan, unfilled circle). SGAs were fractionated by reversed phase HPLC, quantified as described in ‘‘Experimental’’ and plotted as (A) total SGAs, (B) a-solanine (2) and achaconine (1), and (C) leptines (5,6) and leptinines (3,4). Each data point represents the mean of three biological replicates ± one S.D. except for emerging flowers (n = 1). The values are listed in Suppl. Tables 1–3. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.6. Comparison of transcript profiles of terpene and SGA pathway genes, and SGA abundance Early steps in the SGA biosynthetic pathway are shared with pathways leading to the formation of sesquiterpenes, sterols, triterpenes and phytosterols. The specialized metabolic pathway required to convert the primary metabolite cholesterol to SGAs is largely unknown except for the last steps of glycoside formation

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

37

Fig. 4. SGA profiles during the transition from belowground stolon to aerial shoot. Organs were harvested as belowground branching stolons (A), stolons as emerging shoots (B), shoots with expanding leaves (C), and shoots with mature leaves (D). The HPLC eluate was monitored at A202 nm and the numbered peaks correspond to (4) leptinine II, (3) leptinine I, (2) a-solanine, (1) a-chaconine, (6) leptine II, and (5) leptine I.

(McCue et al., 2007). We compared SGA abundance with the transcript levels of genes either shared by terpene/sterol pathways or those associated only with SGA biosynthesis. We used semiquantitative RT-PCR to estimate steady-state transcript levels of the genes coding for 3-hydroxymethylglutaryl-CoA reductase (hmg1, hmg2, hmg3), squalene synthase (pss), solanidine glycosyltransferase (sgt1, sgt2) and actin (PoAc101) in the various organs (Fig. 6). Transcripts of hmg2 and hmg3 were not detected in any organ, as expected, as the plants were grown under minimal stress conditions (Choi et al., 1992; Yang et al., 1991). The transcript abundance profiles for hmg1 and pss correlated approximately with the SGA levels across the various organs (Fig. 6). Similar observations and conclusions were reported in a survey of wild and cultivated potato genotypes limited to leaf and tuber organs (Krits et al., 2007). For the genes directly associated with SGA biosynthesis, sgt transcript was not detected in several organs where SGAs were detected including tubers (sgt1 and sgt2), roots (sgt2), mature and

expanding leaves (sgt1), and mature flowers (sgt1 and sgt2). A similar result was observed by Belknap and coworkers (Moehs et al., 1997) in the initial characterization of sgt1. Thus, the transcript levels for sgt1 and sgt2, genes most closely associated with SGA biosynthesis, did not correlate with SGA abundance whereas genes associated with central terpene metabolism showed a correlation. Transcripts of sgt genes were detected in the emerging leaf primordia and floral buds, tissues that develop into the leaf and floral organs which lack sgt transcripts but contain SGAs. These results suggest that uncharacterized sgt gene family members encode solanidine glycosyltransferase activity in those organs lacking detectable sgt1 and sgt2 transcript or that SGT1 and SGT2 enzyme activity persists in mature tissues although the transcripts of the encoding genes were not detected. Alternatively, SGA biosynthesis may occur primarily in proliferating tissues such as the root apical meristems, emerging floral buds and leaf primordia and the SGAs

38

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

3. Concluding remarks

Fig. 5. Total SGA concentration of stolons at different developmental stages. SGA extracts were prepared from belowground (black bar) and aboveground (white bar) stolons. Each data point represents the mean of three biological replicates ± one S.D. except aboveground stolons at anthesis (n = 1). The absence of a bar indicates no sample analyzed. Abbreviations: dPA, days post-anthesis.

detected in other tissues represent the initial biosynthesis or shortrange transport. Long-range transport of SGAs is unlikely as reciprocal rootstock and scion grafts of potato and tomato have demonstrated that species-specific SGAs are not transported across the graft junction into the tissues of the other species (Roddick, 1982).

Leptines and leptinines (3–6) are the dominant SGAs in the leaf, stem and floral organs of S. chacoense. It was found that their accumulation accounts for the large increase in SGA content after floral induction. Although leptines and leptinines (3–6) were present in most aerial tissues, their absence in developing stolons indicated that factors other than organ specificity and light control their accumulation. Correlations were also found for the various plant organs between SGA concentrations and the transcript levels for two major genes in central terpene metabolism, but not between SGA concentrations and the transcript levels for genes closely associated with SGA biosynthesis. These results could indicate that SGA biosynthesis occurs in most tissues but they also do not preclude the possibility of SGA biosynthesis localized to proliferative tissues with little or no additional biosynthesis in the expanding and mature tissues of the organ under normal growth conditions. 4. Experimental 4.1. Plant material and reagents Solanum chacoense accession 8380-1 (derived from PI 458310, Potato Germplasm Introduction Station, Sturgeon Bay, WI; (Sinden et al., 1986b) was the original source of plant material which has been maintained in tissue culture for several years. a-Solanine (1; Cat. # 045K7350) and a-chaconine standards (2; Cat. #

Fig. 6. Comparison of SGA accumulation and steady-state transcript levels for genes of terpene/sterol and SGA biosyntheses in various organs at anthesis. (A) Total SGA abundance (lmol g 1 dry wt) in extracts of tubers (Tbr), belowground stolons (BG Stl), roots (Rt), mature leaves (Mat Lf), expanding leaves (Exp Lf), emerging leaves (Emr Lf), stems (Stm), open flowers (Mat Flr), and emerging flowers (Emr Flr). Values for total SGA concentrations are listed in Suppl. Table 1. (B) Semiquantitative reverse transcriptase-PCR of cDNA prepared from RNA extracts of the same tissues. Oligonucleotide primer pairs were specific for gene family members coding for 3-hydroxy-3methylglutaryl-CoA-reductase (hmg1, hmg2, hmg3), squalene synthase (pss), solanidine glycosyltransferase (sgt1, sgt2) and actin (PoAc101). The primer name, gene abbreviation, oligonucleotide primer sequences, GenBank identifiers and expected DNA product sizes for the genes are listed in Table 3.

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

089H7032; 99% purity, from potato sprouts) were obtained from Sigma–Aldrich (St. Louis, MO). Oligonucleotide primers were from Integrated DNA Technologies (Coralville, IA, Table 2).

39

was analyzed directly using the 3200 Q TRAP Linear Ion Trap Quadrupole LC/MS/MS System (AB Sciex, Carlsbad, CA), equipped with a TurboSpray ion source.

4.2. Plant growth conditions 4.6. SGA identification II: HPLC and LC/MS Plants were grown in an environmental growth chamber (Conviron, Winnipeg, Manitoba) under a 10 h photoperiod at 24 °C and night temperature of 19 °C. A combination of fluorescent and incandescent light sources provided photosynthetically active radiation of approximately 275 lmol photons m 2 s 1. The relative humidity was maintained at 60%. Plants were fertilized every 14 d with All Purpose Miracle-Gro (NPK of 24-8-16, Scotts, Marysville, OH) at a concentration of 1 g l 1. Stolon cuttings were planted in cell packs containing Sunshine Mix # 1 (Sun Gro Horticulture, Bellevue, WA) and then transferred to 10-cm pots after 28 d. Flowering was induced by lengthening the photoperiod to 14 h at 65 days after transplanting. 4.3. Plant developmental stages and organs for SGA analysis Organs were harvested at different developmental stages representing the entire life cycle of the plant as summarized in Table 1. Plants were harvested at the same time of day and separated into various organs. Organ fresh weights were determined after which the samples were lyophilized (Labconco, Kansas City, MO) for 72 h (Sagredo et al., 2006). The number of plants representing each biological replicate varied with developmental stage but there were at least three biological replicates for each sample. 4.4. SGA extraction Freeze-dried samples were ground by mortar and pestle with liq. N2. Extraction and profile analyses using high performance liquid chromatography (HPLC) followed the procedure of Edwards and Cobb (1996). In brief, SGAs were extracted with 0.02 M heptanesulfonic acid in 1% aq. CH3COOH (v/v), then concentrated and enriched by solid phase extraction using C18 Classic cartridge columns (Waters, Milford, MA). 4.5. SGA identification and quantification I SGAs were fractionated by HPLC (Agilent HP 1200 Series, Santa Clara, CA) on a C18 reversed-phase column (Agilent Eclipse XDBC18, 5 lm pore size and 4.6  150 mm). SGAs were eluted using a gradient of Solvent A, CH3CN/0.01 M Tris–HCl (30:70, v/v pH 8.0), and Solvent B, CH3CN/0.01 M Tris–HCl (80:20, v/v, pH 8.0) at a flow rate of 1 ml min 1 at 25 °C. Gradient elution was 0–10 min, 0–40% B; 10–13 min, 40–100% B; 13–15 min, 100% B; 15–15.5 min, 100–0% B; 15.5–20 min, 0% B. The eluent was monitored by diode array detection from 190 to 240 nm. Quantification of SGAs was based on peak absorbance area at A202 nm, which was converted to molar concentrations using relative response factors generated from calibration curves for a-solanine and a-chaconine made with purified standards. The plots were linear within the range of 125–2000 ng and generated response factors of 3.0 ng mAU 1 and 3.4 ng mAU 1, respectively. For quantification of other SGAs, which all contained the aglycone solanidine with side modifications at C3 and C23 (Fig. 2), the response factor was used for a-chaconine (1). For LC–MS, extracted SGAs (2 ll) were injected directly onto the analytical column described above with a flow rate of 0.5 ml min 1 at 50 °C and a gradient consisting of 0.1% aq. HCO2H (Solvent A) and 0.1% aq. HCO2H in CH3CN (Solvent B). The run consisted of 0.0–2.00 min, 30% solvent B; 2.00– 10.00 min, 30–60% solvent B; 10–15 min, 60% solvent B; 15– 15.01 min, 60–30% solvent B; 15.01–17 min, 30% solvent B. Eluant

Additional HPLC analyses were conducted with a smaller format C18 reversed-phase column (Agilent Eclipse XDB-C18, 2.1  100 mm, 3.5 lm pore size) and MS-compatible Solvent A, 10 mM (NH4)CO3 pH 8.0, and Solvent B, CH3CN/80 mM NH4CO3 (7:1) with a gradient elution of 0–0.2 min, 0% B; 0.2–1.0 min, 0–36% B; 1.0– 16.0 min, 36–48% B; 16.0–22.0 min, 48–100% B; 22.0–25.0 min, 100% B; 25.0–25.5 min, 100–0% B; and 25.5–32.0 min, 0% B. These same column and conditions were used for LC–MS with the eluent analyzed by mass spectrometry as described above (Section 4.5).

4.7. RNA extraction and cDNA synthesis Tubers, stolons, roots, leaves, stems, flowers and floral buds of S. chacoense were harvested at anthesis, chilled with liquid nitrogen and stored at 80 °C. Total RNA was extracted from the tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) according to instructions. RNA was quantified by spectrophotometry and assessed by separation on formaldehyde-denaturing agarose gels. First-strand cDNA was prepared from 2 lg of total RNA using 200 U of M-MLV reverse transcriptase SuperScript II (Invitrogen), and oligo (dT)20 as the oligonucleotide primer in a reaction volume of 50 ll.

4.8. Semi-quantitative reverse transcription polymerase chain reaction (RT-PCR) Relative steady-state transcript levels of selected genes were determined from first-strand cDNA product from 40 ng of RNA. The oligonucleotide primers and corresponding genes are listed in Table 2. The cycle conditions were an initial denaturation at 95 °C for 3 min, and 27 cycles at 94 °C for 30 s, 58 °C for 30 s, 72 °C for 1 min, followed by 5 min incubation at 72 °C. Shorter and longer cycles (25, 30, 35 and 40) were conducted to ensure that the PCR was in a linear response range at 27 cycles. The PCR products were stained with ethidium bromide following 1% agarose gel electrophoresis. Restriction enzyme digests confirmed DNA product identity. The gene PoAc101 (Drouin and Dover, 1990) coding for actin was used for normalization.

4.9. Statistical analysis Analysis of variance for SGA content was conducted with SAS software (Version 9.1, SAS Institute Inc., Cary, NC) with mean separation using the Ryan-Einot-Gabriel-Welsch multiple range test (REGWQ) at P 6 0.05.

Acknowledgements Generous support was provided by the Thomas F. and Kate Miller Jeffress Memorial Trust (J-871) and the Binational Agricultural Research and Development Fund (IS-4134-08).

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.phytochem.2011.12.003.

40

A.M. Mweetwa et al. / Phytochemistry 75 (2012) 32–40

References Bouarab, K., Melton, R., Peart, J., Baulcombe, D., Osbourn, A., 2002. A saponindetoxifying enzyme mediates suppression of plant defences. Nature 418, 889– 892. Choi, D., Ward, B.L., Bostock, R.M., 1992. Differential induction and suppression of potato 3-hydroxy-3-methylglutaryl Coenzyme A reductase genes in response to Phytophthora infestans and to its elicitor arachidonic acid. Plant Cell 4, 1333– 1344. Deahl, K.L., Cantelo, W.W., Sinden, S.L., Sanford, L.L., 1991. The effect of lightintensity on Colorado potato beetle resistance and foliar glycoalkaloid concentration of four Solanum chacoense clones. Am. Potato J. 68, 659–666. Deahl, K.L., Young, R.J., Sinden, S.L., 1973. A study of the relationship of late blight resistance to glycoalkaloid content in fifteen potato clones. Am. Potato J. 50, 248–253. Distl, M., Wink, M., 2009. Identification and quantification of steroidal alkaloids from wild tuber-bearing Solanum species by HPLC and LC-ESI-MS. Potato Res. 52, 79–104. Drouin, G., Dover, G.A., 1990. Independent gene evolution in the potato actin gene family demonstrated by phylogenetic procedures for resolving gene conversions and the phylogeny of angiosperm actin genes. J. Mol. Evol. 31, 132–150. Edwards, E.J., Cobb, A.H., 1996. Improved high-performance liquid chromatographic method for the analysis of potato (Solanum tuberosum) glycoalkaloids. J. Agric. Food Chem. 44, 2705–2709. Fewell, A.M., Roddick, J.G., 1993. Interactive antifungal activity of the glycoalkaloids a-solanine and a-chaconine. Phytochemistry 33, 323–328. Friedman, M., Dao, L., 1992. Distribution of glycoalkaloids in potato plants and commercial potato products. J. Agric. Food Chem. 40, 419–423. Gregory, P., Sinden, S.L., Osman, S.F., Tingey, W.M., Chessin, D.A., 1981. Glycoalkaloids of wild, tuber-bearing Solanum species. J. Agric. Food Chem. 29, 1212–1215. Johns, T., Alonso, J.G., 1990. Glycoalkaloid change during the domestication of the potato, Solanum section Petota. Euphytica 50, 203–210. Kaneko, K., Tanaka, M.W., Takahashi, E., Mitsuhashi, H., 1977. Teinemine and isoteinemine. Two new alkaloids from Veratrum grandiflorum. Phytochemistry 16, 1620–1622. Kowalski, S.P., Perez, F.G., Sanford, L.L., Deahl, K.L., 2000. Partial preparative purification of leptine I from foliage of the wild potato, Solanum chacoense (Bitt.). Prep. Biochem. Biotechnol. 30, 133–144. Krits, P., Fogelman, E., Ginzberg, I., 2007. Potato steroidal glycoalkaloid levels and the expression of key isoprenoid metabolic genes. Planta 227, 143–150. Kuhn, R., Löw, I., 1961. Zur Konstitution der Leptine. Chem. Ber. 94, 1088–1095. Kumar, D., Wareing, P.F., 1972. Factors controlling stolon development in the potato plant. New Phytol. 71, 639–648. Lafta, A.M., Lorenzen, J.H., 2000. Influence of high temperature and reduced irradiance on glycoalkaloid levels in potato leaves. J. Am. Soc. Hort. Sci. 125, 563–566. Lawson, D.R., Erb, W.A., Miller, A.R., 1992. Analysis of Solanum alkaloids using internal standardization and capillary gas-chromatography. J. Agric. Food Chem. 40, 2186–2191.

Lawson, D.R., Green, T.P., Haynes, L.W., Miller, A.R., 1997. Nuclear magnetic resonance spectroscopy and mass spectrometry of solanidine, leptinidine, and acetylleptinidine. Steroidal alkaloids from Solanum chacoense Bitter. J. Agric. Food Chem. 45, 4122–4126. Lawson, D.R., Veilleux, R.E., Miller, A., 1993. Biochemistry and genetics of Solanum chacoense steroidal alkaloids: Natural resistance factors to the Colorado potato beetle. Curr. Topics Bot. Res. 1, 335–352. McCue, K.F., Allen, P.V., Shepherd, L.V.T., Blake, A., MacCree, M.M., Rockhold, D.R., Novy, R.G., Stewart, D., Davies, H.V., Belknap, W.R., 2007. Potato glycosterol rhamnosyltransferase, the terminal step in triose side-chain biosynthesis. Phytochemistry 68, 327–334. Milner, S.E., Brunton, N.P., Jones, P.W., O’ Brien, N.M., Collins, S.G., Maguire, A.R., 2011. Bioactivities of glycoalkaloids and their aglycones from Solanum species. J. Agric. Food Chem. 59, 3454–3484. Moco, S., Capanoglu, E., Tikunov, Y., Bino, R.J., Boyacioglu, D., Hall, R.D., Vervoort, J., De Vos, R.C.H., 2007. Tissue specialization at the metabolite level is perceived during the development of tomato fruit. J. Exp. Bot. 58, 4131–4146. Moehs, C.P., Allen, P.V., Friedman, M., Belknap, W.R., 1997. Cloning and expression of solanidine UDP-glucose glucosyltransferase from potato. Plant J. 11, 227– 236. Nema, P.K., Ramayya, N., Duncan, E., Niranjan, K., 2008. Potato glycoalkaloids: formation and strategies for mitigation. J. Sci. Food Agric. 88, 1869–1881. Roddick, J.G., 1982. Distribution of steroidal glycoalkaloids in reciprocal grafts of Solanum tuberosum L. and Lycopersicon esculentum Mill. Experientia 38, 460– 462. Roddick, J.G., 1989. The acetylcholinesterase-inhibitory activity of steroidal glycoalkaloids and their aglycones. Phytochemistry 28, 2631–2634. Ronning, C.M., Kowalski, S.P., Sanford, L.L., Stommel, J.R., 2000. Geographical variation of solanidane aglycone glycoalkaloids in the wild potato species Solanum chacoense Bitter. Genet. Resourc. Crop Evol. 47, 359–369. Sagredo, B., Lafta, A., Casper, H., Lorenzen, J., 2006. Mapping of genes associated with leptine content of tetraploid potato. Theor. Appl. Genet. 114, 131–142. Sinden, S.L., Sanford, L.L., Cantelo, W.W., Deahl, K., 1988. Bioassays of segregating plants: a strategy for studying chemical defenses. J. Chem. Ecol. 14, 1941–1950. Sinden, S.L., Sanford, L.L., Cantelo, W.W., Deahl, K.L., 1986a. Leptine glycoalkaloids and resistance to the Colorado potato beetle (Coleoptera, Chrysomelidae) in Solanum chacoense. Environ. Entomol. 15, 1057–1062. Sinden, S.L., Sanford, L.L., Deahl, K.L., 1986b. Segregation of leptine glycoalkaloids in Solanum chacoense Bitter. J. Agric. Food Chem. 34, 372–377. Tingey, W.M., 1984. Glycoalkaloids as pest resistance factors. Am. Potato J. 61, 157– 167. Yang, Z.B., Park, H.S., Lacy, G.H., Cramer, C.L., 1991. Differential activation of potato 3-hydroxy-3-methylglutaryl coenzyme A reductase genes by wounding and pathogen challenge. Plant Cell 3, 397–405. Zacharius, R.M., Osman, S.F., 1977. Glycoalkaloids in tissue culture of Solanum species. Dehydrocommersonine from cultured roots of Solanum chacoense. Plant Sci. Lett. 10, 283–287.