J Chem Ecol (2007) 33:1430–1448 DOI 10.1007/s10886-007-9312-3
Comparative Phloem Chemistry of Manchurian (Fraxinus mandshurica) and Two North American Ash Species (Fraxinus americana and Fraxinus pennsylvanica) Alieta Eyles & William Jones & Ken Riedl & Don Cipollini & Steven Schwartz & Kenneth Chan & Daniel A. Herms & Pierluigi Bonello
Received: 12 March 2007 / Revised: 26 April 2007 / Accepted: 7 May 2007 / Published online: 24 May 2007 # Springer Science + Business Media, LLC 2007
Abstract Recent studies have investigated interspecific variation in resistance of ash (Fraxinus spp.) to the exotic wood-boring beetle, emerald ash borer (EAB, Agrilus planipennis). Manchurian ash (Fraxinus mandshurica) is an Asian species that has coevolved with EAB. It experiences little EAB-induced mortality compared to North American ashes. Host phloem chemistry, both constitutive and induced, might partly explain this interspecific variation in resistance. We analyzed the constitutive phloem chemistry of three ash species: Manchurian ash and North American white (Fraxinus americana) and green (Fraxinus pennsylvanica) ash. Analysis of the crude phloem extracts revealed the presence of an array of phenolic compounds including hydroxycoumarins, a A. Eyles : P. Bonello Department of Plant Pathology, The Ohio State University, 201 Kottman Hall, 2021 Coffey Road, Columbus, OH 43210, USA W. Jones : K. Chan Division Of Pharmaceutics, College Of Pharmacy, The Ohio State University, 308 Comprehensive Cancer Center, 410 W 12th Ave, Columbus, OH 43210, USA K. Riedl : S. Schwartz Department of Food Science and Technology, 110 Parker Food Science, The Ohio State University, 2015 Fyffe Court, Columbus 43210, USA D. Cipollini Department of Biological Sciences, Wright State University, 208 Biological Sciences Building, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA D. A. Herms Department Of Entomology, The Ohio State University/Ohio Agricultural Research and Development Center, 1680 Madison Ave, Wooster, OH 44691, USA Present address: A. Eyles (*) Cooperative Research Centre for Forestry, Private Bag 12, Hobart 7001, Australia e-mail:
[email protected],
[email protected]
J Chem Ecol (2007) 33:1430–1448
1431
monolignol, lignans, phenylethanoids, and secoiridoids. Both qualitative and quantitative differences were observed among the three ash species. Hydroxycoumarins and the phenylethanoids, calceloariosides A and B, were present only in the phloem of Manchurian ash and might represent a mechanism of resistance against EAB. Keywords Agrilus planipennis . LC-MS . Hydroxycoumarins . Phenolics . Coevolution . Host resistance . Invasive species
Introduction Since its accidental importation from Asia, emerald ash borer (EAB Agrilus planipennis Fairmaire (Col: Buprestidae), a phloem-feeding beetle, has infested and killed millions of ash (Fraxinus spp.) trees in Indiana, Ohio, Michigan, and Ontario (Herms et al. 2004; Cappaert et al. 2005; Poland and McCullough 2006). As EAB continues to spread, it has the potential to eliminate ash species from natural and urban forests throughout North America. In Ohio alone, there are 3.8 billion white ash trees (Griffith et al. 1991). Unlike congeneric woodboring beetles endemic to North America, including bronze birch borer (Agrilus anxius Gory) and two-lined chestnut borer (Agrilus bilineatus Weber) that colonize stressed trees (e.g., Anderson 1944; Dunn et al. 1990), EAB is more aggressive, killing even healthy trees on high quality sites (Haack et al. 2002; Herms et al. 2004; Poland and McCullough 2006). In a recent common garden study that investigated interspecific variation in ash resistance to EAB, Manchurian ash (Fraxinus mandshurica Rupr.) was shown to have low levels of EAB colonization and EAB-induced mortality, while North American ashes experienced high rates of colonization and severe mortality (Rebek et al. 2006). These observations are consistent with the hypothesis that Asian ashes have high resistance to EAB by virtue of their coevolutionary history. Reports indicate that EAB is rare in Asia, where Manchurian ash is a primary host (Haack et al. 2002; Liu et al. 2003; Schaefer 2005), and infestations appear restricted to stressed trees (Gould et al. 2005). This implies that Asian ashes may be generally resistant. Relatively little research has addressed mechanisms of resistance of angiosperm trees to wood-borers (e.g., Hanks et al. 1999). We propose that in the case of phloem-feeding buprestids, host resistance results from constitutive and induced physical and phytochemical defenses (Dunn et al. 1990). More specifically, defense may have three main components: (1) constitutive, or pre-attack, physical and chemical defenses, such as thick outer bark and preexisting pools of compounds in the phloem that are either toxic and/or serve as deterrents against the insect; (2) immediate, attack-induced de novo synthesis and accumulation of specific, low-molecular-weight secondary metabolites, phenolic polymers such as lignin and suberin, and defensive proteins (Kemp and Burden 1986; Bostock and Stermer 1989; Fernandes 1990; Pearce 1996; Kehr 2006), which may slow the growth of the invading organism enough to allow for (3) subsequent formation of wound (necrophylactic) periderms (i.e., callus tissue), which serve to isolate the wound, inhibit the spread of the colonizing organism (e.g., by encapsulation of insect larvae), and reestablish phellogen and cambial integrity (Mullick and Jensen 1973; Mullick 1977; Biggs et al. 1984; Bostock and Stermer 1989; Robinson et al. 2004). Allelochemicals are represented by a diverse array of plant secondary metabolites, including terpenes and phenolic derivatives such as tannins, flavonoids, and stilbenes. Whether constitutively expressed or induced in response to herbivory, they may serve a wide range of roles. The genus Fraxinus belongs to the family Oleaceae, which comprises around 600
1432
J Chem Ecol (2007) 33:1430–1448
species in 25 genera (Wallander and Albert 2000; Perez et al. 2005). The olive (Olea europaea L.), cultivated for its fruit and oil, belongs to this family and has been studied extensively with regard to chemical constituents. Compound classes that have been reported to be commonly associated with olives include iridoid glucosides, phenylethanoids, coumarins, and lignan glucosides (Bianco et al. 2001; Jensen et al. 2002; Ryan et al. 2002; Cardoso et al. 2005). As part of a broader study investigating potential mechanisms of Manchurian ash resistance to EAB, we compared the constitutive composition of secondary metabolites in phloem from the main stem of white (Fraxinus americana L.), green (Fraxinus pennsylvanica Marsh.), and Manchurian ash. We hypothesized that: (a) Manchurian ash phloem contains concentrations of particular constitutive secondary metabolites different from white and green ash, and (b) that these compounds may be involved in resistance. To date, relatively few studies have investigated the phenolic chemistry of Fraxinus spp. phloem (Terazawa and Sasaya 1970; Tsukamoto et al. 1984; 1985; Nykolov et al. 1993; Kostova et al. 1995; Iossifova et al. 1998, 1999) with many focused on manna ash (Fraxinus ornus L.) (Steinegger and Brantschen 1959; Kostova 2001). The present paper addresses hypothesis (a), by providing chemical data that will be needed to investigate the role of putative defense compounds in resistance of Manchurian ash to EAB (hypothesis b).
Methods and Materials Plant Material Eight 4-yr-old saplings each of F. americana, F. mandshurica, and F. pennsylvannica were obtained from Bailey Nursery, Inc., St Paul, MN, USA. The ash saplings had a mean stem height of 181±0.03 (SEM), 216±0.06, and 149±0.03 cm, respectively, and a mean stem diameter of 15.8±0.5, 23.2±0.9, and 12.4±0.3 cm, respectively. The dormant bareroot saplings were stored in the dark at 4°C for 2 wk before sampling. Chemicals and Reagents Methanol and acetic acid (both HPLC grade) were purchased from Fisher (Pittsburg, PA, USA). Ultrapure water was prepared using a Milli-Q water system (Millipore Ltd, Bedford, MA, USA). Standards of caffeic acid, esculin, and vanillin were purchased from Sigma Aldrich (St. Louis, MO, USA). Chlorogenic acid, fraxin, oleuropein, and verbascoside, were purchased from Extrasynthese (Genay, France). Pinoresinol was purchased from Arbo Nova (Turku, Finland). Extraction One hundred milligrams of stem phloem were ground in liquid nitrogen and extracted twice with 500 μl of 100% methanol over 48 hr in the dark at 4°C. The pooled extract was transferred to a 1.5-ml microcentrifuge tube and centrifuged (12,000×g for 5 min) to remove solids. Samples were stored at −20°C and analyzed within 1 wk of extraction. Analysis of Phenolics with HPLC-UV, HPLC-ESI-MS, and 1H NMR and 13C NMR HPLC with UV detection and HPLC-ESI-MS were used to analyze phenolics. LC-MS analyses were conducted on a Hewlett Packard HPLC system (Palo Alto, CA, USA) model HP 1100 equipped with an HP DAD G1315A detector coupled to a QTof-1 mass spectrometer (Micromass, Manchester, UK). Chromatographic separation was carried out by using a Waters Xterra™ RP18, 5 μm, 4.6×150 mm column and a Waters Xterra ™ RP18, 3.9 μm, 3.0×20 mm guard column. The autosampler and column temperatures were set at 4 and 30°C, respectively.
J Chem Ecol (2007) 33:1430–1448
1433
The binary mobile phase consisted of water/acetic acid (A) (99.9:0.1, v/v) and methanol/ acetic acid (B) (99.9:0.1, v/v), and the flow rate was set at 1 ml/min. The elution program was as follows (percentages refer to proportion of eluant B): 5–15% (0–15 min); 15–30% (15–35 min); 30–40% (35–40 min); 40–60% (40–50 min); 60–90% (50–55); 100% (55– 60 min). The injection volume for all samples was 5 μl. Compound separation was followed by post-column splitting (1:1), passing of the LC effluent through the Photodiode Array (PDA) Detector (scanning range, 200–400 nm) and an electrospray source (source block temperature, 120°C; desolvation temperature, 400°C). The MS detector was optimized to obtain maximum yields of [M-H]− ions of epicatechin/flavonoid standards. The optimized parameters in the negative ionization mode were as follows: capillary voltage, 3.2 kV; cone voltage, 32 V. Argon was used as the collision gas at 20 psi and nitrogen as the nebulizing gas at 15 l/hr. For full-scan MS analysis, the spectra were recorded in the range of m/z 100–900. We applied MS detection in survey scan mode to identify the major compounds in the HPLC eluate, and we compared these peaks with the PDA trace. In this MS method, full MS scans (survey) were analyzed on-the-fly by the instrument to identify the most abundant parent ions. The mode temporarily switched to MS/MS mode to perform daughter scans of the presumptive parent ions. Data acquisition and processing were performed by using a MassLynx version 3.5 data system (Micromass). The identifications of esculin, fraxin, verbascoside, oleuropein, and apigenin were unequivocally confirmed by spiking samples with the respective standards and comparing retention times and ultraviolet (UV) spectra. Additional mass data for syringin was obtained with a LCQ Classic (Finnigan) by using a slightly modified version of the above LC-MS method. Calceolariosides A and B, oleuropein, and pinoresinol glucoside were isolated by semipreparative HPLC and further purified by normal-phase preparative TLC (1,000 μm thickness silica gel plates) with chloroform:methanol:toluene:acetic acid (80:30:8:2). These compounds were analyzed by 1H NMR and 13C NMR. Spectra were obtained in methanol-d4 by using a Bruker Avance DRX-400 (400 MHz) spectrometer. 1H NMR and 13C NMR spectra were calibrated to the residual solvent peaks at δH 3.31 and δC 49.0, respectively. Quantification of Selected Phenolics After identification, the compounds (Table 4) were quantified against external standards as mg g−1 FW. Lack of suitable external standards made it impossible to quantify all compounds. Contents were expressed as verbascoside equivalents for calceolariosides A and B, fraxin equivalents for fraxetin and mandshurin, esculin equivalents for esculetin and methylesculin, oleuropein equivalents for ligustroside and oleuropein hexoside, pinoresinol equivalents for pinoresinol dihexoside, pinoresinol glucoside, and syringaresinol. Quantitative HPLC analyses were performed on an Alliance 2690 separation module (Waters, Milford, MA, USA) equipped with an autosampler and a 996 Photodiode Array Detector (Waters). The same column and gradient program as described for the LC-MS system were used. The mobile phase consisted of water/acetic acid (A) (98:2, v/v) and methanol/acetic acid (B) (98:2, v/v). Monitoring and processing of data were performed at 280 nm. Statistical Analysis The contents of selected phenolic compounds that were present in two or more ash species were examined by univariate analysis of variance (ANOVA) using SPSS 14.0 for Windows (SPSS Inc., Chicago). Levene’s test was used to examine variance equality. Data that did not have equal variances were log-transformed to satisfy this assumption.
1434
J Chem Ecol (2007) 33:1430–1448
Results Characterization of Phenolic Compounds Among the three ash species, a total of 38 individual compounds were analyzed in detail. The characterization of unknown phenolic compounds was based on comparison of UV, mass spectrometry (MS), MS fragmentation behavior, and published NMR data. Liquid chromatography/tandem mass spectrometry (LC-TMS) with electrospray ionization has proven to be a powerful tool for the characterization of nonvolatile, polar, and thermally labile phenolic compounds obtained from plant extracts (Eyles et al. 2003a, b; Cardoso et al. 2005; March et al. 2006). Comparison of the ESI-MS/MS data with published data and standards revealed a diverse range of phenolic compounds in the crude phloem extracts including hydroxycoumarins, lignans, phenylethanoids, and secoiridoids. Tables 1, 2, 3 summarize the tentative identification of the chromatographic peaks on the basis of retention times, molecular weights, and the main ions obtained by TMS for compounds in each species. Given that the majority of the compounds detected are known and have been characterized in previous studies (Tables 1, 2 and 3), only brief explanations of the structural analyses of the major compounds are presented (Fig. 1). Hydroxycoumarins Seven compounds were identified as hydroxycoumarins (5, 7, 8, 12, 14, 15, and 16). Most were identified as a glycoside containing one sugar moiety, usually a hexoside as characterized by a neutral loss of 162 Da (Fig. 2). Esculin (5) and fraxin (14) were identified by comparing with standards. Compound 8 exhibited a [M-H]− ion at m/z 353 and UV spectrum maxima at 290 and 335 nm. Its MS/MS spectrum included ions at m/z 191, which corresponds to a loss of hexose (162 Da) and at m/z 176, which corresponds to a loss of a methoxy group (15 Da) from m/z 191. These findings are consistent with those reported for methylesculin (Steinegger and Brantschen 1959; Vereecke et al. 1997). Compounds 12 and 15 were tentatively identified as fraxidin hexoside and mandshurin, respectively (Table 2, Fig. 3). Despite both compounds exhibiting the same [M-H]− ion at m/z 383 and MS/MS spectrum (m/z 221, 206), it was possible to identify each compound by comparing their UV spectra to those previously reported (Terazawa and Sasaya 1970; Jensen and Nielsen 1976). Lignans Compounds 19, 20, 26, and 28 showed maxima of absorption bands at around 281 nm, characteristic of furofuran lignans (Tsukamoto et al. 1984) and similar to the pinoresinol standard. Compounds 19 and 26 were identified as pinoresinol dihexoside and pinoresinol glucoside, respectively. Pinoresinol dihexoside (19) displayed a [M-H]− ion at m/z 681, and its MS/MS spectrum gave ions at m/z 519 and 357, originating from successive losses of 162 Da, suggesting the presence of two hexosyl residues. By contrast, pinoresinol glucoside (26) yielded only one neutral loss of 162 Da. Furthermore, the 1H NMR and 13C NMR data of this compound were consistent with the literature values (Bodesheim and Holzl 1997). Compound 22 exhibited a [M-H]− ion at m/z 523, corresponding to a neutral loss of a hexose moiety (162 Da) and the MS/MS spectrum (m/z 329, 359) was consistent with those reported for lariciresinol (Smeds et al. 2006). Thus, this compound was tentatively characterized as lariciresinol hexoside. Monolignol Compound 10 exhibited an [M-H]− ion at m/z 371 and its MS/MS gave a [MH-162]− ion at m/z 209, corresponding to the loss of a hexose. Furthermore, the product ion spectrum of m/z 371 yielded fragments at m/z 194 and 179, indicating the successive losses of 15 Da and the presence of two methoxyl groups (loss of methyl radicals) (Ye et al. 2005).
J Chem Ecol (2007) 33:1430–1448
1435
Table 1 Characterization of phenolic compounds from the phloem of green ash, Fraxinus pennsylvanica No. Rt [M-H]− Fragments m/z (Ordered (min) by Decreasing Intensity)
UV Maxima Tentative Identificationa
References
1
5.78
315
153,135, 119, 123
278
Cardoso et al. 2005
3
7.88
299
4
11.10 431
6
12.67 601
119, 113, 137, 131, 149, 276 179, 161 149, 299, 179, 191, 161, 276 131, 251, 119 403, 223, 179, 161 sh 240
10 13
13.96 371 16.60 377
209, 194, 179, 151 197, 153
265 260
18
22.17 353
191
325
19
27.42 681
519, 357, 161
274
20
29.16 535
21 24
31.50 523 33.10 391
26
35.39 519
373, 211, 343, 181, 193, 278 179, 151 361 278 179, 161, 135, 311 sh 290, 331 357, 151 278
28
37.16 417
181, 402, 387, 166
277
29
37.94 701
539, 377, 275, 307, 437
sh 280
31
39.64 685
32
40.06 623
33
41.37 539
36 37
44.62 523 46.37 609
38
54.44 269
523, 453, 421, 299, 223, 330 179 461, 161 sh 285, 331 377, 275, 307, 223, 345, sh 281 149, 179 291, 361, 259, 223, 179 sh 281 301, 271 255, 353 149, 151, 225 267, 339
Hydroxytyrosol hexoside Tyrosol hexoside
Kammerer et al. 2005
Unknown Elenolic acid derivative Syringin Oleuropein aglycone Caffeoyl-quinic acid Pinoresinol dihexoside Hydroxypinoresinol hexoside Unknown Unknown Pinoresinol glucosideb Syringaresinol Oleuropein hexoside Nuzhenide Verbascosidec Oleuropeinb,c Ligustroside Quercetin diglycoside Apigeninc
Wang and Sun 2005 Cardoso et al. 2005 Cardoso et al. 2005 Boyer et al. 2005 Iossifova et al. 1998
Ye et al. 2005 Evtuguin and Amado 2003 Bianco et al. 2001 Ryan et al. 2002; Mulinacci et al. 2005 Cardoso et al. 2005 Tanahashi et al. 1998; Bianco et al. 2001 Ryan et al. 2002 Cardoso et al. 2005 Fabre et al. 2001
sh Shoulder a
Tentative identification based on MS data and UV spectrum consistent with literature
b
Identification confirmed by NMR
c
Identification based on retention time, mass spectral and UV spectrum consistent with those of standard
Based on these observations and a characteristic UV spectrum with a peak of 265 nm, 10 was tentatively identified as syringin (Wang and Sun 2005). Secoiridoids Oleuropein (33) was identified by comparing with the standard. The 1H NMR and 13C NMR data for this compound were consistent with those found in the literature
1436
J Chem Ecol (2007) 33:1430–1448
Table 2 Characterization of phenolic compounds from the phloem of Manchurian ash, Fraxinus Mandshurica No.
Rt (min)
[M-H]−
Fragments m/z (Ordered by Decreasing Intensity)
UV Maxima
Tentative Identificationa
References
1
5.69
315
135, 119, 153, 113, 123
280
Cardoso et al. 2005
2
7.03
167
152, 122, 121
276
3
7.77
299
276
5
11.82
339
119, 113, 137, 131, 149, 179, 161 177
Hydroxytyrosol hexoside Homovanillic alcohol Tyrosol hexoside Esculinb
Parejo et al. 2004
7
12.74
177
133, 149, 105
Esculetin
Concannon et al. 2000
8
13.13
353
191, 176
Methylesculin
Vereecke et al. 1997
9
13.36
389
Oleoside
Cardoso et al. 2005
11
13.89
353
165, 183, 121, 209, 345, 119 191, 179, 135, 173
Cardoso et al. 2005
12
15.57
383
221, 191, 163, 135, 206
14
17.01
369
207, 191, 354
15 16 19
17.37 20.49 26.89
383 207 681
221, 206, 191, 161 193, 165, 109 519, 357, 161
292, sh 340 sh 300, 341 329 335 276
Caffeoyl-quinic acid Fraxidin hexoside Fraxinb
21 22
30.78 32.26
523 521
361 329, 359
278 278
25 26
33.18 35.01
335 519
179, 135, 161 357, 151, 342, 136
276 278
27
36.77
477
161, 179, 135, 315
290, 328
29
37.10
701
539, 377, 275, 307, 437
sh 278
31
39.11
685
32
39.38
623
523, 453, 421, 299, 223, 179 461, 161
33
41.14
539
not detected sh 290, 331 sh 281
35
42.07
477
36 37
44.50 46.01
523 609
377, 275, 307, 223, 345, 149, 179 161, 315, 135, 179, 251, 281, 221 361, 291, 259, 223, 127 301, 271
290, 336 sh 290, 335 sh 290, 335 Not detected 287, 335
285, 328 sh 280 256, 353
Mandshurin Fraxetin Pinoresinol dihexoside Unknown Lariciresinol hexoside Unknown Pinoresinol glucoside b Calceolarioside Ac Oleuropein hexoside Nuzhenide Verbascosideb Oleuropeinb,c Calceolarioside Bc Ligustroside Quercetin diglycoside
Kammerer et al. 2005
Jensen and Nielsen 1976 Godecke et al. 2005 Terazawa and Sasaya 1970 Liu et al. 2005 Boyer et al. 2005
Smeds et al. 2006
Ye et al. 2005 Nicoletti et al. 1986 Bianco et al. 2001 Ryan et al. 2002; Mulinacci et al. 2005 Cardoso et al. 2005 Tanahashi et al. 1998; Bianco et al. 2001 Nicoletti et al. 1986; Ersöz et al. 2002 Ryan et al. 2002 Cardoso et al. 2005
sh Shoulder a
Tentative identification based on MS data and UV spectrum consistent with literature
b
Identification based on retention time, mass spectral and UV spectrum consistent with those of standard
c
Identification confirmed by NMR
J Chem Ecol (2007) 33:1430–1448
1437
Table 3 Characterization of phenolic compounds from the phloem of white ash, Fraxinus americana No. Rt [M-H]− Fragments m/z (Ordered (min) by Decreasing intensity)
UV Maxima Tentative Identificationa
References
1
5.82
315
153, 123, 135
278
Cardoso et al. 2005
3
7.77
299
6 8
10.98 431 12.27 565
119, 113, 137, 131, 149, 179, 276 161 149, 299, 179, 191, 221, 251 275 403, 179, 223, 161, 119, 143 239
9 10 13
13.36 389 13.36 371 16.47 377
165, 183, 121, 209, 345, 119 209, 194, 179, 151 197, 153
266 265 262
17
21.60 537
375, 327, 195, 179
19
27.02 681
519, 357, 161
270, sh 305 276
20
28.70 535
373, 343, 211, 181, 193
278
21 22
30.96 523 32.10 521
361 329, 341, 359
278 278
23
32.58 555
26
35.00 519
273, 393, 307, 361, 179, 181, sh 290, 375, 419 332 357, 151, 136, 342 278
28
37.16 417
181, 402, 387, 166
277
29
37.18 701
539, 377, 275, 307, 437
sh 278
30
38.40 621
179, 161, 459, 487
32
39.11 623
33
40.98 539
34
42.00 685
286, 331 461, 161 sh 285, 331 377, 275, 307, 223, 345, 149, sh 281 179 361, 291, 259, 179 277
36 37
44.37 523 45.98 609
291, 361, 223, 179, 127 301, 271
38
54.45 269
149, 151, 225, 201
280 255, 353 267, 337
Hydroxytyrosol hexoside Tyrosol hexoside Unknown Elenolic acid derivative Oleosideb Syringinb Oleuropein aglycone Unknown
Kammerer et al. 2005
Cardoso et al. 2005 Wang and Sun 2005 Cardoso et al. 2005
Pinoresinol Boyer et al. 2005 dihexoside Hydroxypinoresinol Iossifova et al. 1998 hexoside Unknown Lariciresinol Smeds et al. 2006 hexoside Unknown Pinoresinol glucosidec Syringaresinol Oleuropein hexoside Unknown
Ye et al. 2005 Evtuguin and Amado 2003 Bianco et al. 2001
Verbascosided
Cardoso et al. 2005
Oleuropeinc,d
Tanahashi et al. 1998; Bianco et al. 2001
Ligustroside derivative Ligustroside Quercetin diglycoside Apigenind
Ryan et al. 2002 Cardoso et al. 2005 Fabre et al. 2001
sh Shoulder a
Tentative identification based on MS data and UV spectrum consistent with literature
b
Peaks co-eluted
c
Identification confirmed by NMR
d
Identification based on retention time, mass spectral and UV spectrum consistent with those of standard
1438
J Chem Ecol (2007) 33:1430–1448
Hydroxycoumarins: R3 R1
R2
O
O
R4
Esculin: Fraxin: Fraxidin: Mandshurin:
R1
R2
R3
R4
Glc OMe OMe Glc
OH OH OMe OMe
H H H OMe
H Glc OH H
Lignans: OH
R
O H O
O H
O HO
Pinoresinol: Syringaresinol:
R
R=H R = OMe
Fig. 1 Chemical structure of major compounds in crude ash extracts
(Tanahashi et al. 1998). Compound 29 exhibited the same characteristic UV spectrum as oleuropein but gave an [M-H]− ion at m/z 701. Its MS/MS spectrum yielded an intense ion at 539, consistent with a neutral loss of a hexose (162 Da) plus other ions (m/z 377, 275, 307) identical to those for oleuropein. Therefore, this compound was tentatively identified as oleuropein hexoside. The MS/MS spectrum of compound 36 yielded ions at m/z 291, 361, 223, 179, and 127. These findings are consistent with ligustroside (Ryan et al. 2002). Compound 34 exhibited a similar MS/MS spectrum as ligustroside (m/z 361, 291, 259, 159) but with a [M-H]− ion at m/z 685, which corresponds to an addition of 162 Da. Therefore, it was tentatively identified as a ligustroside derivative. Phenylethanoids Compound 32 was characterized by the [M-H]− ion at m/z 623 and CID gave two major ions at m/z 461 and 161 that are in accordance with the fragmentation of
J Chem Ecol (2007) 33:1430–1448
1439
Monolignol: O OH
O O O HO
OH HO OH
Syringin
Secoiridoid glycosides: O
O
R
O O O
HO O
O
HO
HO
OH OH
Ligustroside: Oleuropein:
R=H R = OH
Fig. 1 (continued)
verbascoside (Cardoso et al. 2005). Compounds 27 and 35 exhibited [M-H]− ion at m/z 477 with almost identical UV and mass spectra suggesting them to be isomers of the same compound (Fig. 4). In both cases, their MS/MS spectra gave an [M-H-162]− ion at m/z 315,
1440
J Chem Ecol (2007) 33:1430–1448
Phenylethanoids:
OR 2 O
R1 O
O
OH
OH
OH
OH
Calceolarioside A: R1 = Caffeoyl Calceolarioside B: R1 = H
R2 = H R2 = Caffeoyl
Fig. 1 (continued)
indicating the loss of a caffeic acid moiety. In addition, caffeic acid (m/z 179) and its decarboxylated product (m/z 135) generated signals of low intensity (Fig. 4). The intense fragment ion at m/z 161 is most likely due to the fragmentation of the hexose moiety by the loss of water (Tolonen et al. 2004). Thus, compounds 27 and 35 were assigned as calceolariosides. The 1H NMR and 13C NMR data showed them to be identical with those reported for calceolariosides A (Nicoletti et al. 1986) and B (Ersöz et al. 2002), respectively. Comparison of the Three Ash Species Both qualitative and quantitative differences were observed among the three species (Fig. 2, Tables 1, 2, 3 and 4). Comparisons of 17 individual phenolic compounds revealed a number of differences in composition as listed in Table 4. In particular, Manchurian ash was distinguished from the North American ash species by the presence of hydroxycoumarins, particularly mandshurin (15), fraxin (14), and fraxidin hexoside (12), in decreasing order of abundance. Other minor hydroxycoumarins that were not detected in either green or white ash included esculin (5), esculetin (7), methylesculin (8), and fraxetin (16). The phenylethanoids, calceolariosides A (27) and B (35), also were only found in Manchurian ash. The main lignan in green and white ash was syringaresinol (28) followed by smaller amounts of pinoresinol glucoside (26) and pinoresinol dihexoside (19). Manchurian ash contained amounts of pinoresinol dihexoside (19) that were 11-fold higher than in green and 10-fold higher than in white ash (F2,23 =238, P