- Email: [email protected]
Changes in apolar metabolites during in vitro organogenesis of Pancratium maritimum
Plant Physiology and Biochemistry 48 (2010) 827e835
Contents lists available at ScienceDirect
Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy
Research article
Changes in apolar metabolites during in vitro organogenesis of Pancratium maritimum Strahil Berkov a, b, *, Atanas Pavlov c, Vasil Georgiev c, Jost Weber d, Thomas Bley d, Francesc Viladomat a, Jaume Bastida a, Carles Codina a a
Departament de Productes Naturals, Biologia Vegetal i Edafologia, Facultat de Farmàcia, Universitat de Barcelona. Av. Joan XXIII s/n, 08028 Barcelona, Catalonia, Spain AgroBioInstitute, 8 Dragan Tzankov Blvd., 1164- Sofia, Bulgaria Department of Microbial Biosynthesis and Biotechnologies e Laboratory in Plovdiv, Institute of Microbiology, Bulgarian Academy of Sciences, 26 “Maritza” Blvd., 4002 Plovdiv, Bulgaria d Institute of Food Technology and Bioprocess Engineering, Technische Universität Dresden, 01069 Dresden, Germany b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 November 2009 Accepted 12 July 2010 Available online 27 July 2010
Calli, shoot-clumps and regenerated plants were initiated from young fruits of Pancratium maritimum L. Their genetic stability was monitored by flow cytometry before chemical studies. Apolar metabolites (alkaloids extracted at pH > 7, free fatty acids and fatty alcohols, sterols etc.) were qualitatively and quantitatively analyzed by GCeMS. The results clearly demonstrated that alkaloid synthesis in P. maritimum is closely related with tissue differentiation. The highest amounts of alkaloids and presence of homolycorine and tazettine type compounds (end products of the biosynthetic pathway of the Amaryllidaceae alkaloids) were found in highly differentiated tissues. Galanthamine accumulated in the leaves of plantlets. The amount of hordenine, a protoalkaloid, is related with the ability of tissues to synthesize alkaloids. Saturated fatty acids were found in considerably higher levels in undifferentiated callus cultures and partially differentiated shoot-clumps than in regenerated plants. Mono- and dienoic fatty acids were found at higher levels in non-photosynthesizing tissues e calli, and in vitro and intact bulbs, while a-linolenic acid (trienoic acid) was found in higher amounts in the photosynthesizing leaves of shoot-clumps and regenerated plants than in bulbs and calli. Fatty alcohols were found mainly in leaves, while sterols tended to accumulate in photosynthesizing and undifferentiated tissues. Ó 2010 Elsevier Masson SAS. All rights reserved.
Keywords: Amaryllidaceae alkaloids Apolar metabolites Lipids Pancratium maritimum Tissue differentiation
1. Introduction Differentiation of cells into plant tissues and organs leads to metabolic changes, as well as to excretion and/or storage of specific metabolites. The level of metabolites - end products of cellular regulatory processes - can be regarded as the ultimate response of biological systems to genetic or environmental changes. Physiological and chemical changes during plant growth and development are usually investigated by analysing a limited number of target metabolites. A multi-target profiling analysis is a valuable approach to the study of complex biological systems, leading to a better
Abbreviations: FAs, fatty acids; SFAs, saturated fatty acids; MFAs, monoenoic fatty acids; DFAs, dienoic fatty acids; TFAs, trienoic fatty acids; DW, Dry weight; 2,4-D, 2,4-dichlorophenoxyacetic acid; NAA, a-naphthylacetic acid; BAP, 6-benzylaminopurine; TIC, total ion current; IS, Internal standard. * Corresponding author at: Departamentde Productes Naturals, Biologia Vegetal i Edafologia, Facultat de Farmàcia, Universitat de Barcelona. Av. Joan XXIII s/n, 08028 Barcelona, Catalonia, Spain. Tel.: þ34 93 402 02 68; fax: þ34 93 402 90 43. E-mail address: [email protected] (S. Berkov). 0981-9428/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.plaphy.2010.07.002
understanding of their biological role or functions at different developmental stages or under specific environmental factors [13]. Apolar plant metabolites such as lipids and alkaloids are involved in the mechanisms of plant defense and adaptation. Surface plant lipids, deposited in the cuticule, form a functional barrier preventing excessive water loss and entry of pathogens into the plant [25]. One of the main functions of the alkaloids is that of chemical defense against herbivores. Some alkaloids are antibacterial, antifungal, and antiviral; and these properties may extend to toxicity towards animals [28]. Recently, in vitro cultures of amaryllidaceous plants have attracted attention as an alternative source of alkaloids, including galanthamine - an AChE inhibitor marketed for the treatment of Alzheimer’s disease [19]. It was found that galanthamine is synthesized and accumulated in shoot-clumps at considerably higher levels than in unorganised callus cultures of N. confusus and Leucojum aestivum [4,8,10,14,23]. Fewer alkaloids have been found in the alkaloid profiles of calli in comparison with in vitro shoot-clumps of L. aestivum [4]. Pancratium maritimum L. (sea daffodil) is an amaryllidaceaous species growing in extreme environmental conditions - in salty and
828
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
sandy soils along the coastlines of the Mediterranean, the Black and Caspian seas and the Atlantic ocean [26]. In contrast to L. aestivum and N. confusus, the intact plants of this species produce high amounts of structurally diverse alkaloids, which make it a good model to study alkaloid synthesis. Previous phytochemical studies have resulted in the identification of a number of bioactive alkaloids as well as several phenolic compounds [3,16,21]. Pancratium maritimum plantlets propagated in vitro, by twin-scaling in particular, have been reported [5,22]. As a part of our ongoing studies on alkaloid synthesis in in vitro and ex vitro amaryllidaceous plants, we here report on a range of apolar metabolites in P. maritimum in vitro cultures at different stages of differentiation - calli, shoot-clumps and differentiated plantlets. The results provide basic information on the changes in apolar metabolites taking place during in vitro tissue differentiation of P. maritimum.
2. Results and discussions 2.1. In vitro cultures To the best of our knowledge, this is the first time successful propagation of P. maritimum from callus cultures has been reported (Fig. 1). Optimal results for callus formation were observed on MS nutrient medium supplemented with 1 or 4 mg/L 2,4 D and 2 mg/L BAP, in conditions similar to those previously reported for callus formation in L. aestivum [23]. Callus formation started 35 days after the onset of explants. The obtained calli were yellow and after their removal from the plant explants and two further subcultivations on the same media, they became homogenous, with a granular structure. The callus cultures were cultivated for more than a year and they exhibited stable growth and morphological characteristics. Shoot formation was observed when the calli were cultivated on MS medium with 1.15 mg/L NAA and 2 mg/L BAP. In this case, up to 90% of cultured callus pieces developed shoots after 48 days of cultivation. The shoot-clumps showed balanced growth and stable morphological characteristics and they were subcultivated for more than six months before the chemical analysis. Planting slices of P. maritimum fruit on MS medium with 1.15 mg/L NAA and 2 mg/L BAP first led to the formation of an undifferentiated callus cell mass from which the organogenesis (small leaves) appeared 50 days later. Three subcultivations later,
bulb formation was observed and after five more subcultivation cycles, fully regenerated P. maritimum plants were obtained. Calli and in vitro regenerated plants may possess various genome changes (polyploidy, endoreduplication, aneoploidy etc), which can influence the metabolites and lead to somaclonal variability [20]. For that reason, the genetic stability of the studied cultures was checked by flow cytometry (Fig. 2). The histograms of the tissues revealed the existence of one main peak (2C) and in some tissues a second peak of a small fraction of 4C cells. The latter peak is most likely comprised of some 2C cells in the G2 phase of the cell cycle and some 2C- doublets. The estimation of the genome size for the plant and in vitro cultures (Table 1) resulted in values in the range of 56.0e57.6 pg. The slight variations in the genome size were not significant and were probably due to runerun variability. Its value corresponded well with a previous report [27] and confirmed that all of the cultures investigated here were stable diploids, with no differences being observed between intact P. maritimum plants and their in vitro cultures [1]. Therefore, the differences in the metabolite profiles are a result of tissue differentiation but not of any genetic alteration.
2.2. Apolar metabolites Plant apolar (lipid) fractions consist of more than 100 metabolites, which are generally analyzed as TMS derivatives or after methylation of the fatty acids [7]. Recent GCeMS studies have demonstrated that the alkaloid profiles of the Amaryllidaceae plants are also complex and consist of more than 10e15 compounds [4,17]. Many of the alkaloids exist at very low concentrations and their peaks can be easily overlapped by other compounds. For these reasons, this group of metabolites was fractionated and analyzed separately, without derivatization. Both alkaloid and lipid profiles changed during organ differentiation of the P. maritimum in vitro cultures. 2.2.1. Alkaloid profiles Up to the moment, detailed studies on the alkaloid synthesis in amaryllidaceous in vitro cultures have been performed mainly on the galanthamine-producing N. confusus and L. aestivum calli and shootclumps, indicating a relationship between the state of differentiation and galanthamine accumulation [3,4,8,23]. In contrast to these previous studies, we now report results from fully differentiated
Fig. 1. In vitro cultures of P. maritimum: A e callus; B e shoot-clumps; C e regenerated plantlet.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
829
was also in relation to their degree of differentiation. Undifferentiated calli displayed the lowest alkaloid levels, the lowest number of compounds and a predomination (74% of TIC) of the tyramine type protoalkaloid hordenine (1), which is a side-reaction in the biosynthetic pathway of the Amaryllidaceae alkaloids (Tables 2 and 3, Fig. 4). Shoot-clumps showed higher levels of alkaloids, and a wider range of alkaloids. Hordenine was also predominant but in a lower proportion (54%) than in calli. Intact and in vitro plants showed similar alkaloid patterns but with different ratios of the main compound types. Like intact plants, the in vitro bulbs and leaves showed different alkaloid profiles. The predominant compound types were tyramine (35%) and narciclasine (26%) in the bulbs and haemanthamine (42%) and narciclasine (21%) in the leaves (Table 3). The lowest proportion (14%) of the protoalkaloid hordenine (1) was found in the highly differentiated leaves, which had the highest alkaloid content. The proportion of protoalkaloids indicates the ability of tissues to synthezise alkaloids from precursors. It should be noted that the homolycorine and tazettine type alkaloids, which are end products of ortho-pará and paraepará oxidative coupling of O-methylnorbelladine and derived from lycorine and haemanthamine type compounds, respectively (Fig. 4) [2], were detected only in highly (fully) differentiated plant tissues, that is, leaves and bulbs. Galanthamine type compounds accumulated mainly in the in vitro leaves (8%, Table 3). Galanthamine (4), norgalanthamine (6), haemanthamine (11), tazettine (12) and lycorine (14), found in the in vitro cultures, possess potent bio- and pharmacological activities [2]. Their higher accumlation in the differentiated tissues is probably related with their defense properties. It has been reported that shoot-clumps of N. confusus and L. aestivum accumulate higher amounts of their principal alkaloid (galanthamine) than undifferentiated calli, but still much less than intact plants [4,8,14,23]. Shoot-clump systems are used to study the alkaloid synthesis because they show stable growth and their cultivation in bioreactors is easy [23]. Our study, however, indicates that the alkaloid accumulation found in the in vitro leaves (0.20% of DW) and bulbs (0.14% od DW) of P. maritimum is considerably higher than in shoot-clumps (0.03% of DW) and comparable to those found in intact plants. Thus, due to the close relationship between tissue differentiation and alkaloid synthesis, in vitro plantlets should be considered as a perspective model system for optimizing in vitro alkaloid production.
Fig. 2. Histograms of the relative DNA content of P. martinunm samples: (A) shoot meristem, (B) shoot leaves, (C) in vitro plant bottom plate, (D) in vitro plant bulb, (E) in vivo bottom plate, (F) in vivo leaf, (G) in vivo bulb scales, (H) callus.
in vitro plantlets of a species (P. maritimum) in which alkaloid synthesis is dominated by haemanthamine, tazettine and lycorine type compounds. Sixteen alkaloids were identified in the in vitro cultures and intact plants of P. maritimum (Table 2, Fig. 3). The ability of P. maritimum in vitro cultures to accumulate/synthesize alkaloids Table 1 Genome size of P. maritimum in vitro cultures and intact plants. Sample
Sample DNA-content [pg] SD
Callus Shoots In vitro plants Intact plants
57.0 56.7 56.0 57.6
1.74 1.66 1.61 1.66
2.2.2. Lipid profiles More than one hundred compounds were detected in the lipid fractions (Fig. 3). Identified and unknown metabolites comprising more than 0.2% of TIC are listed in Table 4. Lipid profiles consisted of free fatty acids including saturated fatty acids (SFAs, C9:0 - C32:0; mainly C16:0 and C18:0), monoenoic acids (MFAs, C14:1 e C24:1; mainly C16:1 and C18:1), dienoic acids (DFAs, C18:2 and C20:2), one trienoic acid (TFA, C18:3), 2(3)-hydroxy fatty acids, free fatty alcohols, sterols, etc. The predominant compounds in the lipid fraction were fatty acids and sterols (Tables 4 and 5). The in vitro and intact plants showed differences in their lipid profiles. Considerable amounts of tocopherol (90) were found only in intact leaves while other compounds such as 20, 21, 44, 86, and 94 were detected only in in vitro cultures. On the other hand, metabolites like 90 and phytyl derivatives 31-34 were only detected in photosynthesizing tissues. The highest content of lipid fractions was observed in intact leaves, which may be a result of the adaptation of the intact plant to its naturally arid conditions of growth. The in vitro leaves showed considerably fewer lipids (Table 4). Cunha and Fernandes-Ferreira [9] also found the highest content of lipids in primary explants (obtained from intact plants) but they observed a progressively decreasing lipid content with dedifferentiation of tissues during somatic embriogenesis of flax. The lipid content of
830
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
Table 2 .Alkaloidsidentified in in vitro cultures and intact plants P. maritimum of (mg/g DW, mean SD). Alkaloids
RI
Base/Mþ Ions (m/z)
Intact plants Bulbs
Hordenine (1)a Ismine (2)a Trisphaeridine (3)b Galanthamine (4)a Buphanisine (5)b,c Norgalanthamine (6)a Vittatine (7)a Anhydrolycorine (8)c 6-Deoxytazettine (9)c 11,12-Dehydroanhydrolycorine (10)b Haemanthamine (11)a Tazettine (12)a 11-Hydroxyvittatine (13)a Lycorine (14)a N-Formylnorgalanthamine (15)a 8-O-Demethylhomolycorine (16)a Percentage of DW a b c d
1460 2280 2283 2406 2436 2443 2473 2501 2539 2608 2641 2653 2707 2747 2816 2839
58/165 238/257 223/223 286/287 285/285 272/273 271/271 250/251 231/315 248/249 272/301 247/331 258/287 226/287 301/301 109/301() 0.10%
45 13 47 13 33 112 54 21 91 15 128 96 284 35 0.16%
In vitro cultures Leaves
Callus
Shoot-clumps
Bulbs
Leaves
tr 48 157 65 14 28 66
148 178
151 30
408 217
24 22
89 70 tr 41 17 15 81
349 210 trd 12 5
305 32 375 17 39 136 47 28
tr 293 43 441 48 137 88 185
trd
trd 50 56
26 29 trd
24 15 15 6 tr
311 207 128 99
trd
51 31
45 41 32 19
0.03 0.01%
0.14 0.09%
219 13 57 5 2 41 13 13
238 41 738 104 13 2 34 13 18 2 91
0.20 0.05%
Isolated standard. Literature data. Standard from NIST 05 database. Detected in only one sample.
P. maritimum callus cultures was higher than in differentiated shoot-clumps and in vitro bulbs and leaves (Table 4). As SFAs tend to decrease with the increase of tissue differentiation of in vitro cultures, the highest amounts were found in calli. Very long FAs (C32:0), associated with the cuticule components, were detected in the leaves of intact plants. Although rarely described in plant tissues [9], free C15:0 and C17:0 were found in all samples, which is atributed to an a-oxidation system [24]. 2-Hydroxy FAs, possessing higher polarity than SFAs and associated with cuticule [18], were found in higher amounts in leaves than bulbs. MFAs and DFAs showed a higher content in non-photosynthesizing tissues, that is, calli and bulbs, than in shoot-clumps and leaves. On the contrary, TFAs, associated with photosinthesis and thylacoidal membranes [6], accumulated mainly in green tissues. The unsaturation level, measured by the ratio of total FAs/unsaturated FAs, increased significantly with dedifferentiation and callus formation, reaching values of 1.93 and 1.92 in calli and shoot-clumps, respectively. These values for in vitro bulbs and leaves were 1.65 and 1.74, respectively. Like hydroxy FAs, fatty alkohols (mainly long chain: C26 and C28) accumulated in
shoot-clumps and leaves. They are also considered as constituents of the cuticule [18]. Accumulation of long-chain (C25-C31) alkanes was found in the calli, decreasing during the tissue differentiation. As sterol content (dominated by b-sitosterol) tends to be higher in photosynthesizing and undifferentiated tissues, calli and leaves showed higher sterol levels than bulbs and shoot-clumps, respectively (Table 4). The high capacity of P. maritimium cultures to biosynthesize alkaloids is clearly related to cell differentiation but the different metabolite profiles of the in vitro leaves and bulbs cannot be explained simply by the different cell organization in these organs. During in vitro organogenesis, photosynthesis starts with the differentiation of leaves in the shoot-clumps and plantlets. Further in vitro experiments need to be performed to elucidate the influence of light on both lipid and alkaloid synthesis. The close relationship between alkaloid accumulation and organ differentiation suggests that strategies involving differentiated plants rather than partially differentiated shoot-clumps should be applied for in vitro alkaloid production.
Fig. 3. GCeMS chromatograms of alkaloid (A) and lipid (B) fractions from in vitro leavs of P. maritimum. The numbers of compounds are congruent with those of Tables 2 and 4.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
831
Table 3 Contribution of the different alkaloid types to the alkaloid mixture of P. maritimum presented as a percentage of TIC (mean SD). Alkaloid type
Intact plants
Tyramine Narciclasine Galanthamine Haemanthamine Lycorine Tazettine Homolycorine
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
Leaves
4.5 5.3 0.5 32.3 51.1 2.7 3.3
<0.1 12.1 10.7 10.1 25.3 26.0 10.9
73.5 6a 10.3 e 6.2 10.0 3a e e
53.1 11.6a,b 26.8 15.5a 6.1 7.8a 6.5 4a 7.5 2.8a e e
35.3 17.2b,c 25.6 5.2a 0.2 10.1 5.6a 26.9 13.6a e 2.0 1.0a
14.3 20.7 7.8 42.3 14.4 0.3 0.1
Values followed by different superscripts within a line are significantly different according to Turkey’s multiple comparison test (P < 0.05).
L-Phe
OH
L-Tyr
HO HO HO
Norbelladine
CHO
H2N Me
OH O N
HO
3 O
1 NH O-Methylnorbelladine
R1
para-para' NHMe
O
2
para-ortho'
ortho-para'
O
OH
Narcicasine type
R2
O H
N
HO
O
Haemanthamine type 5 R1=OMe, R2=H 7 R1=OH, R2=H 11 R1=OMe, R2=OH 13 R1=OH, R2=OH
H 14
N
MeN
N 8
HO
O 16
O Homolycorine type
OMe
O
H
MeO
O O
H NMe
O
Galanthamine type 4 R1=OH, R2=Me 6 R1=OH, R2=H 15 R1=OH, R2=CHO
H
O
N R2
MeO
OH R1
O
Me
Tyramine type
MeO
O
N
Tyramine
Protocatechuic aldehyde
O R1 13 R2 R3
O O
N 10
Tazettine type 9 R1= R2=R3=H 12 R1=OH, R2=R3=H Fig. 4. Biosynthetic relationship of the alkaloids found in P. maritimum.
7.1c 3.1a 0.8a 6.5b 0.6a 0.1 0.1b
832
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
Table 4 Metabolites in the apolar fraction. Results represent the means SD of responce ratios using 100 mg of nonadecanoic acid as the quantitative internal standard. Compound
Nonanoic acid (C9:0, 17) Decanoic acid (C10:0, 18) Undecanoic acid (C11:0, 19) Octanedioic acid (20) Unknown (21) Lauric acid (C12:0, 22) Azelaic acid (23) Unknown (24) Tridecanoic acid (C13:0, 25) Tetradecenoic acid (C14:1, 26) Myristic acid (C14:0, 27) 9-Methyltetradecanoate (28) Unknown (29) Pentadecanoic acid (C15:0, 30) Phytyl derivative (31) Phytyl derivative (32) Phytyl derivative (33) Phytyl derivative (34) Hexadecanol (35) Hexadecenoic acid (C16:1, 36) Palmitic acid (C16:0, 37) Unknown (38) Unknown (39) Cyclopropaneoctanoic acid, 2-hexyl (40) Margaric acid (C17:0, 41) Unknown (42) Hexadecanoic acid, 2-hydroxy (43) Unknown (44) Linoleic acid (C18:2, 45) a-Linolenic acid (C18:3, 46) Oleic acid (C18:1n-9, 47) Stearic acid (C18:0, 48) Fatty acid, 3-hydroxy (49) Unknown (50) Octadecanoic acid, 2-hyroxy (51) Eicosadienoic acid (C20:2 n-6, 52) Eicosenoic acid (C20:1 n-9, 53) Eicosatrienoic acid (C20:3, 54) Arachidic acid (C20:0, 55) 4,8,12,16-Tetramethylheptadecan-4-olide (56) Unknown (57) Heneicosanoic acid (C21:0, 58) Unknown (59) Unknown (60) Unknown (61) Eicosanoic acid, 2-hydroxy (62) Unknown (63) Pentacosane (C25, 64) Docosenoic acid (C22:1, 65) Unknown (66) Behenic acid (C22:0, 67) Hexacosane (C26, 68) Tricosanoic acid (C23:0, 69) Docosanoic acid, 2-hydroxy (70) Heptacosanol (71) Heptacosane (C27, 72) Tetracosenoic acid (C24:1, 73) Lignoceric acid (C24:0, 74) Tricosanoic acid, 2-hydroxy (75) Octacosane (C28, 76) Pentacosanoic acid (C25:0, 77) Tetracosanoic acid, 2-hydroxy (78) Unknown (79) Nonacosane (C29, 80) Hexacosanol (81) Cerotinic acid (C26:0, 82) Unknown (83) Heptacosanoic acid (C27:0, 84) Hexacosanoic acid, 2-hydroxy (85) Unknown (86) Hentriacontane (87) b-Sitosterol (acetate?, 88) Montanic acid (C28:0, 89)
Rt
5.19 6.19 7.12 7.33 7.79 8.05 8.31 8.84 9.13 10.12 10.39 11.34 11.39 11.72 11.96 12.09 12.32 12.40 12.56 12.89 13.28 13.65 14.02 14.45 14.82 15.01 15.15 15.47 16.01 16.20 16.25 16.45 17.08 17.29 18.41 19.14 19.20 19.25 19.61 20.04 20.64 21.17 21.38 21.50 21.54 21.58 22.08 22.23 22.32 22.34 22.68 23.70 24.17 24.61 24.80 25.15 25.26 25.59 26.06 26.60 27.00 27.46 27.59 27.89 27.92 28.36 28.92 29.66 30.13 30.22 30.48 30.49 30.93
Base/Mþ Ions (m/z) 74/172 74/186 74/200 129/202() 75/74/214 152/216() 75/74/228 74/240 74/242 74/256 75/74/256 68/123/82/123/55/74/268 74/270 99/99/55/282 74/284 85/227/286 85/428 81/294 79/292 55/296 74/298 103/117/255/314 67/322 55/324 79/320 74/326 99/324 115/e 74/340 115/e 141/e 154/e 283/342 154/e 57/352 55/352 240/e 74/354 57/366 74/368 311/370 57/396() 57/380 55/380 74/382 325/384 57/394 74/396 339/398 97/e 57/408 97/e 74/410 75/e 74/424 367/426 97/e 57/436() 396/e 74/438
Intact plants
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
9 5 tr
21
tr tr
22 4 88 9 199 28
53 41 7 11 24 4 46 1 24 1 139 8 269 24
10 1 118 26 86 33 28 123 13
92 109 9 tr 43 9 91 72
11 30 134 tr 14 69 8
204 7 98 51 13 973 tr
75
112 75 102 285 180
tr 12 1 17 1 17 3 65 15 tr 12 1 84 11 51 15 5 94 8
tr
17 9 204 27 4649 38
tr 15 5 179 53 4072 250
77 173 26 2746 204
29 20 237 46
20 6 230 24
23 3 181 9
52 16
57 7
33 6
3742 409 2259 238 1062 302 777 90
4525 520 687 7 1925 287 657 163
27 4 95 30 43 9 tr 297 225
25 7 50 27 26 15 54 149 37
17 5 44 4 33 33
9
33 6 29 12
135
36 6
7
50
19 1
15
64
101 22 17 7
32 2 20 1 19 1 29 2 24 4 22 3 30 7 82 2
12 2 22 4 62 82 73 15 5 12 2 44 3
tr 172 2254
34 91 39 3687 394 1480 502
15 58 23 tr 54 21 6
69 6201 159 171 148 610 298 66 4105 10792 831 1586 46 645 19 17 12 89 138
76 tr 28 14
116 tr 57 65
34
17
93 tr 5 18 33
151 6 11 65 110
tr
604 380 628 7 8
18 22 5
78 70
33
5142 1627 2104 1301
794 133 914 727
382 224 20 6 130 50 33 5 44 26 250 13 38 20 578 389 12 1 15 1 95 57 51 39 84 32 181 17
192 16 17 1 127 66 47 34 25 3 142 24 92 274 35 13 58 61 45
3 6 3 4
133 36
21 3 149 31 71 70 9 15 5 tr 99 4 208 39 71 58 8 28 1 11 2 47 14
Leaves tr tr 11 10 38 30 13 1 52 40 130 108 tr 10 2 80 8 tr 25 11 57 5 22 4 18 14 75 24 38 26 137 16 3786 636 90 23 99 22 17 2 82 10 88 15 74 21 132 20 3820 864 2389 682 1028 261 493 86 22 3 199 49 35 6 77 28 75 20 94 71 11
13 2
81 tr 264 59 59 8 108 41 15 13 62 194 29 85 23 1 68 18
201 143
261 24 143 123
111 34
750 243 48 5 19 8
17 13 99 221 96
61 87 125 3
72 18 11 29 6
20 5
44 16 151 94
38 10 59 18
37 17 98 20
81 23 15 17 8
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
833
Table 4 (continued ) Compound
Rt
Tocopherol (90) Unknown (91) Campesterol (92) Sterol, unknown (93) Unknown (94) b-Sitosterol (95) Stigmastan-3-ol (96) Stigmasta-5,24-dien-3-ol (97) Dotriacontanoic acid (C32:0, 98)
31.04 31.47 32.21 32.62 32.70 33.43 33.54 34.01 36.59
Base/Mþ Ions (m/z) 165/430 337/e 400/400 396/e 97/e 414/414 416/416 314/412 74/494()
Intact plants
In vitro cultures
Bulbs
Callus
Shoot-clumps
Bulbs
Leaves
48 8 263 23 101 10
48 11 249 57 72 5 69 20 2025 271 39 8
44 7 128 10 65 23 29 19 1705 107 22 16
34 5 217 66 32 12 2007 503 92
17882 ± 1093
14597 ± 1443
17156 ± 3633
193 41
Total
Leaves 1939 64 347
753
4180 53 47 116
2599 158 29 3 40 17
10646
37880
22334 ± 2317
3. Materials and methods 3.1. Plant material Young fruits of P. maritimum L. plants (at the early stage of seed formation) were collected from a single cluster at the “Silistar” beach, Bulgaria in July 2005. They were carefully washed with tap water, surface sterilized by treatment with 70% of EtOH for 10 s and then with 7% of Ca(ClO2) (Sigma, USA) for 15 min. They were then thoroughly rinsed with sterile distilled water, dried on sterile tissue paper and used as explants in the experiments. Leaves and bulb from one plant were lyophilized and used for GC/MS analysis as described below. A voucher specimen (SOM Co-974) was authenticated by Dr. Luba Evstatieva and deposited in the Herbarium of the Institute of Botany (Bulgarian Academy of Sciences), Sofia.
3.2. In vitro cultures 3.2.1. Callus induction Surface sterilized young fruits were sliced and placed on Murashige and Skoog (MS) (Duchefa, The Netherlands) medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar” (Duchefa) and different concentrations and combinations of growth regulators as follows: auxins - 2,4-dichlorphenoxyacetic acid (2,4-D) (Sigma, USA) e 0.5; 1.0; 2.0; 3.0; 4.0 mg/L; a-naphtylacetic acid (NAA) (Duchefa, The Netherlands) e 1.0; 2.0 mg/L and cytokines - kinetin (Sigma, USA) e 0.5; 1.0; 2.0 mg/L; 6-benzylamynopurine (BAP) (Duchefa, The Netherlands) - 0.5; 1.0; 2.0 mg/L. Cultivation was carried out at 26 C in darkness.
3.2.2. Shoot formation The shoot cultures were established by planting the obtained calli on MS nutrient medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar”, 1.15 mg/L NAA and 2.0 mg/L BAP. The cultivation was carried out at 26 C under illumination 16:8 (light:dark). The further subcultivations of obtained shoot cultures were performed every 28 days under the same conditions. 3.2.3. In vitro plant regeneration In vitro plants were obtained from shoot cultures established by planting slices of young fruits on MS nutrient medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar” and three combinations of growth regulators as follows: 0.2 mg/L 2,4D and 1.0 mg/L BAP; 1.15 mg/L NAA and 2.0 mg/L BAP; 5.0 mg/L NAA and 2.0 mg/L isopentenyladenosine (iPA) (Duchefa, The Netherlands). The cultivation was carried out at 26 C in darkness. The further subcultivations of obtained shoot cultures were performed every 28 days under the same conditions. At the end of cultivation period, the in vitro cultures were lyophilized before metabolite extraction.
3.3. Metabolite extraction The lipophilic fractions were obtained from 300 mg plant samples extracted exhaustively with 10 ml methanol (3 48 h) at room temperature until discoloration of the tissues. Before extraction, 50 mg of codeine and 100 mg of nonadecanoic acid were added as internal standards (IS). The extracts were evaporated and the dry residue was resolved in 3 mL of 2% H2SO4. The lipid fraction was first extracted with diethyl ether (3 5 mL) and then the
Table 5 Contribution of the main groups of compounds to the apolar fraction of P. maritimum in vitro cultures and intact plants. Results represent the means SD of response ratios using 0.100 mg of nonadecanoic acid as the quantitative internal standard. Intact plants
SFA 2(3)-Hydroxy FA MFA DFA TFA Total FA Fatty alcohols Alkanes Sterols Tocopherols Isoprenes Unknowns
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
Leaves
3.35 0.15 1.69 3.75 0.40 9.32 tr 0.10 0.99
10.85 0.52 0.93 4.12 10.88 27.30 0.60 0.10 4.63 1.94 0.66 2.56
8.42 2.08a 0.23 0.09a 2.42 0.98a 5.24 0.82a 1.63 0.13a 17.944.10 0.06 0.04a 0.57 0.06a 3.10 0.21a
6.52 0.78a,b 0.24 0.07a 1.29 0.37a 3.79 0.44a 2.26 0.24a 14.101.90 0.30 0.03b 0.25 0.03b 2.39 0.34a
4.66 0.59b 0.15 0.04a 2.15 0.35a 4.57 0.53a 0.69 0.01b 12.211.50 0.01 0.01a 0.21 0.02b 1.92 0.16b
5.27 0.92b 0.35 0.10b 1.26 0.30a 3.90 0.90a 2.40 0.69a 13.182.90 0.75 0.24c 0.02 0.02c 2.27 0.58a
0.25 0.5
0.15 0.07 0.85 0.29
0.20
0.91 0.20
tr 0.78 0.09
Values followed by different superscripts within a line are significantly different according to Turkey’s multiple comparison test (P < 0.05).
834
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
alkaloids by ethylacetate (3 5 mL) after basification of the solution with 25% of ammonia to pH 9-10. The organic solvent of the alkaloid fraction was evaporated and the residue was resolved in 100 mL methanol. The organic solvent of the lipid fraction was evaporated and the fatty acids were methylated with a solution of 5% of H2SO4 in methanol (0.5 mL) for 16 h at 70 C. The compounds were extracted with 3 mL n-hexane after cooling of the mixture and addition of 2 mL of aqueous solution (6%) of NaCl. The n-hexane was rinsed twice with 2 mL of aqueous solution (6%) Na2CO3 and evaporated. The dry residue was resolved in 100 mL of chloroform for further GC/MS analysis. 3.4. GC/MS analysis 3.4.1. Chromatographic conditions The GC/MS analyses were carried out on Hewlett Packard 6890 þ MSD 5975 equipment (Hewlett Packard, Palo Alto, CA, USA) operating in EI mode at 70 eV. An HP-5 MS column (30 m 0.25 mm 0.25 mm) was used. The temperature program was: 100e180 C at 15 C min1, 180e300 C at 5 C min1 and 10 min hold at 300 C. Injector temperature was 250 C. The flow rate of carrier gas (Helium) was 0.8 mL min1. Split ratio of 1:20 was used. One mL of the solutions were injected. 3.4.2. Alkaloid quantification A calibration graph for galanthamine quantification was obtained using solutions of 5, 10, 20, 50, 100, 200 and 400 mg/mL of galanthamine containing 50 mg/mL of IS (codeine). The ratios of the peak areas of selected ions (TIC mode) of galanthamine (m/z at 286) versus those of the IS (m/z at 299) were plotted against the corresponding concentration of galanthamine to obtain the calibration graph. The other alkaloids were quantified as galanthamine equivalents using the galanthamine calibration graph. This is not a true quantification, but it could be used for comparative studies. 3.5. Identification of the metabolites 3.5.1. Alkaloids Hordenine (1), ismine (2), trisphaeridine (3) galanthamine (4), norgalanthamine (6), vittatine (7), haemanthamine (11), tazettine (12), 11-hydroxyvittatine (13), lycorine (14), N-formylnorgalanthamine (15) and 8-O-demethylgalanthamine (16) were identified by comparing their GCeMS spectra and Kovats retention indices (RI) with those of authentic compounds isolated and identified by other spectrometric methods (NMR, UV, CD) by the Group of Natural Products, University of Barcelona. Buphanisine (5), anhydrolycorine (8), 11-deoxytazettine (9) and 11,12-didehydroanhydrolycorine (10) were identified by comparing their mass spectral fragmentation with standard reference spectra from NIST 05 database (NIST Mass Spectral Database, PC-Version 5.0e2005, National Institute of Standardization and Technology, Gaithersburg, MD). The configuration of the 5,10b-ethano bridge of the haemanthamine type alkaloids, which can be established by CD analysis of isolated compounds, was tentatively assigned at the -position like the compounds of this type previously reported for this species [16]. 3.5.2. Lipids The fatty acids in the apolar fraction were identified as fatty acid methyl esters (FAME) according to the characteristic mass spectral fragmentation of each subgroup (saturated, hydroxy, monoenoic, dienoic and trienoic FA), molecular ion, retention time and mass spectra available in the lipid library [18] and NIST 05 database. Fatty alcohols were identified by comparison of their mass spectra with those of the lipid library, NIST 05 database, chromatographic retention and characteristic [M]þ-18 fragments. Hydrocarbons,
sterols and tocopherol were identified by their RI and comparison of their spectra with those of authentic compounds in NIST 05 database and online available plant-specific database (The Golm Metabolome Database; http://csbdb.mpimp-golm.mpg.de/csbdb/ gmd/home/gmd_sm.html). The measured mass spectra were deconvoluted by the Automated Mass Spectral Deconvolution and Identification System (AMDIS) before comparison with the databases. The spectra of individual components were transferred to the NIST Mass Spectral Search Program MS Search 2.0 where they were matched against reference compounds. Response ratio was applied for quantification of the metabolites in the lipid fractions. Response ratios represented the peak area ratio using nonadecanoic acid (100 mg) as the quantitative IS and the respective metabolite in a sample whose weight was normalized to 1 g DW. Base peaks were used for MS deconvolution before peak area calculation (TIC mode). The RIs of the compounds were recorded with a standard n-hydrocarbon calibration mixture (C9-C36) (Restek, Cat no. 31614, supplied by Teknokroma - Spain) using AMDIS 3.6 software. 3.5.3. Statistical analysis Samples for GCeMS analysis were prepared from pooled explants from three independent cultivations of the in vitro cultures. The data were analyzed for statistical significance using analysis of variance (ANOVA single factor) test. P-values less than 0.05 were considered to be significant. All data were processed using GraphPad Prism version 3.00 for Windows (GraphPad Software, San DiegoCalifornia USA, www.graphpad.com). 3.6. Flow cytometry The flow cytometric analyses were carried out using a CyFlow SL Blue (20 mW solid state laser at 488 nm) Flow Cytometer (Partec GmbH, Münster, Germany), operated with FloMax for Windows XP (Partec GmbH, Münster, Germany) software. DNA histograms were recorded on a semi-logarithmic scale, since the histogram peaks on such a scale are evenly distributed along the abscissa [15] and thus the peak heights can be considered to be proportional to the proportions of nuclei at the corresponding ploidy levels. The cell nuclei were extracted from 20 to 100 mg of fresh leaves and/or in vitro culture tissues, and stained using the Otto buffers as described by Dolezel [11]. Every tissue was analyzed at least three times on different days. The resulting peaks typically had CV of 5 %e9% depending on the age and type of tissue used [1]. Prior to analysis, the instrument was checked for linearity with fluorescent beads (AlignFlow 2.5 mm, Invtrogen, Eugene, OR, USA), and the amplification settings were kept constant throughout the experiment. The nuclei were gated in an FSC vs. FL3 density plot to eliminate much of the debris. The genome sizes were determined by co-processing the sample and an internal standard with (Pisum sativum cv. Ctirad) with the known genome size of 9.09 pg per 2C nucleous [12]. To exclude a possible influence of secondary metabolites on the staining, the sample and the standard were first processed alone and the position of the peaks were compared to their positions when processed together. Acknowledgements This work was partially financed by the Generalitat de Catalunya (2005SGR-00020) and National Science Fund of Bulgaria (TK-Ƃ1605/2006). S. Berkov thanks the Spanish Ministerio de Educacion y Ciencia for a research fellowship (SB2004-0062). The authors also thank Dr. Asunción Marín, Serveis Cientificotècnics, Universitat de Barcelona, for performing the GCeMS analyses.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
References [1] M. Barow, A. Meister, Endopolyploidy in seed plants is differently correlated to systematics, organ, life strategy and genome size, Plant Cell Environ. 26 (2003) 571e584. [2] J. Bastida, R. Lavilla, F. Viladomat, Amaryllidaceae alkaloids. in: G. Cordell (Ed.), The Alkaloids, vol. 63. Elsevier Inc, 2006, pp. 87e179. [3] S. Berkov, L. Evstatieva, S. Popov, Alkaloids in Bulgarian Pancratium maritimum L. Z. Naturforsch.C. 59 (2004) 65e69. [4] S. Berkov, A. Pavlov, V. Georgiev, J. Bastida, M. Burrus, M. Ilieva, C. Codina, Alkaloid synthesis and accumulation in Leucojum aestivum in vitro cultures, Nat. Prod. Commun. 4 (2009) 359e364. [5] Y. Bogdanova, M. Stanilova, Ch. Gussev, Y. Bosseva, T. Stoeva, In vitro propagation of Pancratium maritimum L. (Amaryllidaceae) by liquid cultures, Propag. Ornam. Plants 8 (2008) 45e46. [6] A. Blanckaert, L. Belingheri, J. Vasseur, J.-L. Hilbert, Changes in lipid composition during embriogenesis in leaves of Cichorium, Plant Sci. 157 (2000) 165e172. [7] W. Christie, Lipid Analysis. The Oily Press, Bridgewater, England, 2003. [8] C. Codina, Production of galanthamine by Narcissus tissues in vitro. in: G. Hanks (Ed.), Medicinal and Aromatic Plants e Industrial Profiles: Narcissus and Daffodil (The Genus Narcissus). Taylor and Francis, London and New York, 2002, pp. 215e241. [9] A. Cunha, M. Fernandes-Ferreira, Ontogenic variation in free and esterified fatty acids during somatic embriogenesis of flaz (Linum usitatissimum L.), Plant Sci. 164 (2003) 863e872. [10] M. Diop, A. Ptak, F. Chrétien, M. Henry, Y. Chapleur, D. Laurain-Mattar, Galanthamine content of bulbs and in vitro cultures of Leucojum aestivum, Nat. Prod. Commun. 1 (2006) 475e479. [11] J. Dolezel, W. Göhde, Sex determination in dioecious plants Melandrium album and M. rubrum using high-resolution flow-cytometry, Cytometry 19 (1995) 103e106. [12] J. Dolezel, J. Greilhuber, S. Lucretti, A. Meister, M.A. Lysak, L. Nardi, R. Obermayer, Plant genome size estimation by flow cytometry: inter-laboratory comparison, Ann. Bot. 82 (1998) 17e26. [13] O. Fiehn, Metabolomics e the link between genotypes and phenotypes, Plant Mol. Biol. 48 (2002) 155e171. [14] L. Georgieva, S. Berkov, V. Kondakova, J. Bastida, F. Viladomat, A. Atanassov, C. Codina, Alkaloid variability in Leucojum aestivum from wild populations, Z. Naturforsch. C 62 (2007) 627e635.
835
[15] L.J.W. Gilissen, M.J. van Staveren, J.C. Hakkert, M.J.M. Smulders, H.A. Verhoeven, J. Creemers-Molenaar, The competence of cells for cell division and regeneration in Tobacco explants depends on cellular location, cell cycle phase and ploidy level, Plant Sci. 103 (1994) 81e91. [16] O. Hoshino, The Amaryllidaceae alkaloids. in: G. Cordell (Ed.), The Alkaloids, vol. 51. Academic Press, 1998, pp. 324e417. [17] M. Kreh, R. Matusch, L. Witte, Capillary gas chromatography-mass spectrometry of Amaryllidaceae alkaloids, Phytochemistry 38 (1995) 773e776. [18] The Lipid library: http://www.lipidlibrary.co.uk/index.html [19] A. Maelicke, M. Samochocki, R. Jostok, A. Feherbacher, J. Ludwig, E.X. Albuquerque, M. Zerlin, Allosteric sensitization of nicotinic receptors by galanthamine, a new treatment strategy for the Alzheimer’s disease, Biol. Psychiatry 26 (2001) 279e288. [20] M. Nakano, T. Nomizu, K. Mizunashi, M. Suzuki, S. Mori, S. Kuwayama, M. Hayashi, H. Umehara, E. Oka, K. Kobayashi, Somaclonal variation in Tricyrtis hirta plants regenerated from 1-year-old embryogenic callus cultures, Sci.Hortic. 110 (2006) 366e371. [21] M. Nikolova, R. Gevrenova, Determination of phenolic acids in amaryllidaceae species by high performance liquid chromatography, Pharm. Biol. 43 (2005) 289e291. [22] D. Nikopoulos, A. Alexopoulos, In vitro propagation of an endangered medicinal plant: Pancratium maritimum L, J. Food Agr. Environ. 6 (2008) 390e398. [23] A. Pavlov, S. Berkov, E. Courot, T. Gocheva, D. Tuneva, B. Pandova, V. Georgiev, S. Yanev, M. Burrus, M. Ilieva, Galanthamine production by Leucojum aestivum in vitro systems, Process Biochem. 42 (2007) 734e739. [24] Y. Poirier, G. Ventre, D. Caldelary, Increased flow of fatty acids towards b-oxidation in developing seeds of Arabidopsis deficient in diacylglycerol acyltransferase activity or synthesizing medium-chain fatty acids, Plant Physiol. 121 (1999) 1359e1366. [25] J. Reina-Pinto, A. Yephremov, Surface lipids and plant defenses, Plant Physiol. Biochem. 47 (2009) 540e549. [26] H. Zahreddine, C. Clubbe, R. Baalbaki, A. Ghalayini, S.N. Talhouk, Status of native species in threatened Mediterranean habitats: the case of Pancratium maritimum L. (sea daffodil) in Lebanon, Biol. Conserv. 120 (2004) 11e18. [27] B.J.M. Zonneveld, I.J. Leitch, M.D. Bennett, First nuclear DNA amounts in more than 300 angiosperms, Ann. Bot. 96 (2005) 229e244. [28] M. Wink, Ecological role of alkaloids. in: E. Fattorusso, O. Taglialatela-Scafati (Eds.), Modern Alkaloids. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2008, pp. 3e24.
Contents lists available at ScienceDirect
Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy
Research article
Changes in apolar metabolites during in vitro organogenesis of Pancratium maritimum Strahil Berkov a, b, *, Atanas Pavlov c, Vasil Georgiev c, Jost Weber d, Thomas Bley d, Francesc Viladomat a, Jaume Bastida a, Carles Codina a a
Departament de Productes Naturals, Biologia Vegetal i Edafologia, Facultat de Farmàcia, Universitat de Barcelona. Av. Joan XXIII s/n, 08028 Barcelona, Catalonia, Spain AgroBioInstitute, 8 Dragan Tzankov Blvd., 1164- Sofia, Bulgaria Department of Microbial Biosynthesis and Biotechnologies e Laboratory in Plovdiv, Institute of Microbiology, Bulgarian Academy of Sciences, 26 “Maritza” Blvd., 4002 Plovdiv, Bulgaria d Institute of Food Technology and Bioprocess Engineering, Technische Universität Dresden, 01069 Dresden, Germany b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 November 2009 Accepted 12 July 2010 Available online 27 July 2010
Calli, shoot-clumps and regenerated plants were initiated from young fruits of Pancratium maritimum L. Their genetic stability was monitored by flow cytometry before chemical studies. Apolar metabolites (alkaloids extracted at pH > 7, free fatty acids and fatty alcohols, sterols etc.) were qualitatively and quantitatively analyzed by GCeMS. The results clearly demonstrated that alkaloid synthesis in P. maritimum is closely related with tissue differentiation. The highest amounts of alkaloids and presence of homolycorine and tazettine type compounds (end products of the biosynthetic pathway of the Amaryllidaceae alkaloids) were found in highly differentiated tissues. Galanthamine accumulated in the leaves of plantlets. The amount of hordenine, a protoalkaloid, is related with the ability of tissues to synthesize alkaloids. Saturated fatty acids were found in considerably higher levels in undifferentiated callus cultures and partially differentiated shoot-clumps than in regenerated plants. Mono- and dienoic fatty acids were found at higher levels in non-photosynthesizing tissues e calli, and in vitro and intact bulbs, while a-linolenic acid (trienoic acid) was found in higher amounts in the photosynthesizing leaves of shoot-clumps and regenerated plants than in bulbs and calli. Fatty alcohols were found mainly in leaves, while sterols tended to accumulate in photosynthesizing and undifferentiated tissues. Ó 2010 Elsevier Masson SAS. All rights reserved.
Keywords: Amaryllidaceae alkaloids Apolar metabolites Lipids Pancratium maritimum Tissue differentiation
1. Introduction Differentiation of cells into plant tissues and organs leads to metabolic changes, as well as to excretion and/or storage of specific metabolites. The level of metabolites - end products of cellular regulatory processes - can be regarded as the ultimate response of biological systems to genetic or environmental changes. Physiological and chemical changes during plant growth and development are usually investigated by analysing a limited number of target metabolites. A multi-target profiling analysis is a valuable approach to the study of complex biological systems, leading to a better
Abbreviations: FAs, fatty acids; SFAs, saturated fatty acids; MFAs, monoenoic fatty acids; DFAs, dienoic fatty acids; TFAs, trienoic fatty acids; DW, Dry weight; 2,4-D, 2,4-dichlorophenoxyacetic acid; NAA, a-naphthylacetic acid; BAP, 6-benzylaminopurine; TIC, total ion current; IS, Internal standard. * Corresponding author at: Departamentde Productes Naturals, Biologia Vegetal i Edafologia, Facultat de Farmàcia, Universitat de Barcelona. Av. Joan XXIII s/n, 08028 Barcelona, Catalonia, Spain. Tel.: þ34 93 402 02 68; fax: þ34 93 402 90 43. E-mail address: [email protected] (S. Berkov). 0981-9428/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.plaphy.2010.07.002
understanding of their biological role or functions at different developmental stages or under specific environmental factors [13]. Apolar plant metabolites such as lipids and alkaloids are involved in the mechanisms of plant defense and adaptation. Surface plant lipids, deposited in the cuticule, form a functional barrier preventing excessive water loss and entry of pathogens into the plant [25]. One of the main functions of the alkaloids is that of chemical defense against herbivores. Some alkaloids are antibacterial, antifungal, and antiviral; and these properties may extend to toxicity towards animals [28]. Recently, in vitro cultures of amaryllidaceous plants have attracted attention as an alternative source of alkaloids, including galanthamine - an AChE inhibitor marketed for the treatment of Alzheimer’s disease [19]. It was found that galanthamine is synthesized and accumulated in shoot-clumps at considerably higher levels than in unorganised callus cultures of N. confusus and Leucojum aestivum [4,8,10,14,23]. Fewer alkaloids have been found in the alkaloid profiles of calli in comparison with in vitro shoot-clumps of L. aestivum [4]. Pancratium maritimum L. (sea daffodil) is an amaryllidaceaous species growing in extreme environmental conditions - in salty and
828
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
sandy soils along the coastlines of the Mediterranean, the Black and Caspian seas and the Atlantic ocean [26]. In contrast to L. aestivum and N. confusus, the intact plants of this species produce high amounts of structurally diverse alkaloids, which make it a good model to study alkaloid synthesis. Previous phytochemical studies have resulted in the identification of a number of bioactive alkaloids as well as several phenolic compounds [3,16,21]. Pancratium maritimum plantlets propagated in vitro, by twin-scaling in particular, have been reported [5,22]. As a part of our ongoing studies on alkaloid synthesis in in vitro and ex vitro amaryllidaceous plants, we here report on a range of apolar metabolites in P. maritimum in vitro cultures at different stages of differentiation - calli, shoot-clumps and differentiated plantlets. The results provide basic information on the changes in apolar metabolites taking place during in vitro tissue differentiation of P. maritimum.
2. Results and discussions 2.1. In vitro cultures To the best of our knowledge, this is the first time successful propagation of P. maritimum from callus cultures has been reported (Fig. 1). Optimal results for callus formation were observed on MS nutrient medium supplemented with 1 or 4 mg/L 2,4 D and 2 mg/L BAP, in conditions similar to those previously reported for callus formation in L. aestivum [23]. Callus formation started 35 days after the onset of explants. The obtained calli were yellow and after their removal from the plant explants and two further subcultivations on the same media, they became homogenous, with a granular structure. The callus cultures were cultivated for more than a year and they exhibited stable growth and morphological characteristics. Shoot formation was observed when the calli were cultivated on MS medium with 1.15 mg/L NAA and 2 mg/L BAP. In this case, up to 90% of cultured callus pieces developed shoots after 48 days of cultivation. The shoot-clumps showed balanced growth and stable morphological characteristics and they were subcultivated for more than six months before the chemical analysis. Planting slices of P. maritimum fruit on MS medium with 1.15 mg/L NAA and 2 mg/L BAP first led to the formation of an undifferentiated callus cell mass from which the organogenesis (small leaves) appeared 50 days later. Three subcultivations later,
bulb formation was observed and after five more subcultivation cycles, fully regenerated P. maritimum plants were obtained. Calli and in vitro regenerated plants may possess various genome changes (polyploidy, endoreduplication, aneoploidy etc), which can influence the metabolites and lead to somaclonal variability [20]. For that reason, the genetic stability of the studied cultures was checked by flow cytometry (Fig. 2). The histograms of the tissues revealed the existence of one main peak (2C) and in some tissues a second peak of a small fraction of 4C cells. The latter peak is most likely comprised of some 2C cells in the G2 phase of the cell cycle and some 2C- doublets. The estimation of the genome size for the plant and in vitro cultures (Table 1) resulted in values in the range of 56.0e57.6 pg. The slight variations in the genome size were not significant and were probably due to runerun variability. Its value corresponded well with a previous report [27] and confirmed that all of the cultures investigated here were stable diploids, with no differences being observed between intact P. maritimum plants and their in vitro cultures [1]. Therefore, the differences in the metabolite profiles are a result of tissue differentiation but not of any genetic alteration.
2.2. Apolar metabolites Plant apolar (lipid) fractions consist of more than 100 metabolites, which are generally analyzed as TMS derivatives or after methylation of the fatty acids [7]. Recent GCeMS studies have demonstrated that the alkaloid profiles of the Amaryllidaceae plants are also complex and consist of more than 10e15 compounds [4,17]. Many of the alkaloids exist at very low concentrations and their peaks can be easily overlapped by other compounds. For these reasons, this group of metabolites was fractionated and analyzed separately, without derivatization. Both alkaloid and lipid profiles changed during organ differentiation of the P. maritimum in vitro cultures. 2.2.1. Alkaloid profiles Up to the moment, detailed studies on the alkaloid synthesis in amaryllidaceous in vitro cultures have been performed mainly on the galanthamine-producing N. confusus and L. aestivum calli and shootclumps, indicating a relationship between the state of differentiation and galanthamine accumulation [3,4,8,23]. In contrast to these previous studies, we now report results from fully differentiated
Fig. 1. In vitro cultures of P. maritimum: A e callus; B e shoot-clumps; C e regenerated plantlet.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
829
was also in relation to their degree of differentiation. Undifferentiated calli displayed the lowest alkaloid levels, the lowest number of compounds and a predomination (74% of TIC) of the tyramine type protoalkaloid hordenine (1), which is a side-reaction in the biosynthetic pathway of the Amaryllidaceae alkaloids (Tables 2 and 3, Fig. 4). Shoot-clumps showed higher levels of alkaloids, and a wider range of alkaloids. Hordenine was also predominant but in a lower proportion (54%) than in calli. Intact and in vitro plants showed similar alkaloid patterns but with different ratios of the main compound types. Like intact plants, the in vitro bulbs and leaves showed different alkaloid profiles. The predominant compound types were tyramine (35%) and narciclasine (26%) in the bulbs and haemanthamine (42%) and narciclasine (21%) in the leaves (Table 3). The lowest proportion (14%) of the protoalkaloid hordenine (1) was found in the highly differentiated leaves, which had the highest alkaloid content. The proportion of protoalkaloids indicates the ability of tissues to synthezise alkaloids from precursors. It should be noted that the homolycorine and tazettine type alkaloids, which are end products of ortho-pará and paraepará oxidative coupling of O-methylnorbelladine and derived from lycorine and haemanthamine type compounds, respectively (Fig. 4) [2], were detected only in highly (fully) differentiated plant tissues, that is, leaves and bulbs. Galanthamine type compounds accumulated mainly in the in vitro leaves (8%, Table 3). Galanthamine (4), norgalanthamine (6), haemanthamine (11), tazettine (12) and lycorine (14), found in the in vitro cultures, possess potent bio- and pharmacological activities [2]. Their higher accumlation in the differentiated tissues is probably related with their defense properties. It has been reported that shoot-clumps of N. confusus and L. aestivum accumulate higher amounts of their principal alkaloid (galanthamine) than undifferentiated calli, but still much less than intact plants [4,8,14,23]. Shoot-clump systems are used to study the alkaloid synthesis because they show stable growth and their cultivation in bioreactors is easy [23]. Our study, however, indicates that the alkaloid accumulation found in the in vitro leaves (0.20% of DW) and bulbs (0.14% od DW) of P. maritimum is considerably higher than in shoot-clumps (0.03% of DW) and comparable to those found in intact plants. Thus, due to the close relationship between tissue differentiation and alkaloid synthesis, in vitro plantlets should be considered as a perspective model system for optimizing in vitro alkaloid production.
Fig. 2. Histograms of the relative DNA content of P. martinunm samples: (A) shoot meristem, (B) shoot leaves, (C) in vitro plant bottom plate, (D) in vitro plant bulb, (E) in vivo bottom plate, (F) in vivo leaf, (G) in vivo bulb scales, (H) callus.
in vitro plantlets of a species (P. maritimum) in which alkaloid synthesis is dominated by haemanthamine, tazettine and lycorine type compounds. Sixteen alkaloids were identified in the in vitro cultures and intact plants of P. maritimum (Table 2, Fig. 3). The ability of P. maritimum in vitro cultures to accumulate/synthesize alkaloids Table 1 Genome size of P. maritimum in vitro cultures and intact plants. Sample
Sample DNA-content [pg] SD
Callus Shoots In vitro plants Intact plants
57.0 56.7 56.0 57.6
1.74 1.66 1.61 1.66
2.2.2. Lipid profiles More than one hundred compounds were detected in the lipid fractions (Fig. 3). Identified and unknown metabolites comprising more than 0.2% of TIC are listed in Table 4. Lipid profiles consisted of free fatty acids including saturated fatty acids (SFAs, C9:0 - C32:0; mainly C16:0 and C18:0), monoenoic acids (MFAs, C14:1 e C24:1; mainly C16:1 and C18:1), dienoic acids (DFAs, C18:2 and C20:2), one trienoic acid (TFA, C18:3), 2(3)-hydroxy fatty acids, free fatty alcohols, sterols, etc. The predominant compounds in the lipid fraction were fatty acids and sterols (Tables 4 and 5). The in vitro and intact plants showed differences in their lipid profiles. Considerable amounts of tocopherol (90) were found only in intact leaves while other compounds such as 20, 21, 44, 86, and 94 were detected only in in vitro cultures. On the other hand, metabolites like 90 and phytyl derivatives 31-34 were only detected in photosynthesizing tissues. The highest content of lipid fractions was observed in intact leaves, which may be a result of the adaptation of the intact plant to its naturally arid conditions of growth. The in vitro leaves showed considerably fewer lipids (Table 4). Cunha and Fernandes-Ferreira [9] also found the highest content of lipids in primary explants (obtained from intact plants) but they observed a progressively decreasing lipid content with dedifferentiation of tissues during somatic embriogenesis of flax. The lipid content of
830
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
Table 2 .Alkaloidsidentified in in vitro cultures and intact plants P. maritimum of (mg/g DW, mean SD). Alkaloids
RI
Base/Mþ Ions (m/z)
Intact plants Bulbs
Hordenine (1)a Ismine (2)a Trisphaeridine (3)b Galanthamine (4)a Buphanisine (5)b,c Norgalanthamine (6)a Vittatine (7)a Anhydrolycorine (8)c 6-Deoxytazettine (9)c 11,12-Dehydroanhydrolycorine (10)b Haemanthamine (11)a Tazettine (12)a 11-Hydroxyvittatine (13)a Lycorine (14)a N-Formylnorgalanthamine (15)a 8-O-Demethylhomolycorine (16)a Percentage of DW a b c d
1460 2280 2283 2406 2436 2443 2473 2501 2539 2608 2641 2653 2707 2747 2816 2839
58/165 238/257 223/223 286/287 285/285 272/273 271/271 250/251 231/315 248/249 272/301 247/331 258/287 226/287 301/301 109/301() 0.10%
45 13 47 13 33 112 54 21 91 15 128 96 284 35 0.16%
In vitro cultures Leaves
Callus
Shoot-clumps
Bulbs
Leaves
tr 48 157 65 14 28 66
148 178
151 30
408 217
24 22
89 70 tr 41 17 15 81
349 210 trd 12 5
305 32 375 17 39 136 47 28
tr 293 43 441 48 137 88 185
trd
trd 50 56
26 29 trd
24 15 15 6 tr
311 207 128 99
trd
51 31
45 41 32 19
0.03 0.01%
0.14 0.09%
219 13 57 5 2 41 13 13
238 41 738 104 13 2 34 13 18 2 91
0.20 0.05%
Isolated standard. Literature data. Standard from NIST 05 database. Detected in only one sample.
P. maritimum callus cultures was higher than in differentiated shoot-clumps and in vitro bulbs and leaves (Table 4). As SFAs tend to decrease with the increase of tissue differentiation of in vitro cultures, the highest amounts were found in calli. Very long FAs (C32:0), associated with the cuticule components, were detected in the leaves of intact plants. Although rarely described in plant tissues [9], free C15:0 and C17:0 were found in all samples, which is atributed to an a-oxidation system [24]. 2-Hydroxy FAs, possessing higher polarity than SFAs and associated with cuticule [18], were found in higher amounts in leaves than bulbs. MFAs and DFAs showed a higher content in non-photosynthesizing tissues, that is, calli and bulbs, than in shoot-clumps and leaves. On the contrary, TFAs, associated with photosinthesis and thylacoidal membranes [6], accumulated mainly in green tissues. The unsaturation level, measured by the ratio of total FAs/unsaturated FAs, increased significantly with dedifferentiation and callus formation, reaching values of 1.93 and 1.92 in calli and shoot-clumps, respectively. These values for in vitro bulbs and leaves were 1.65 and 1.74, respectively. Like hydroxy FAs, fatty alkohols (mainly long chain: C26 and C28) accumulated in
shoot-clumps and leaves. They are also considered as constituents of the cuticule [18]. Accumulation of long-chain (C25-C31) alkanes was found in the calli, decreasing during the tissue differentiation. As sterol content (dominated by b-sitosterol) tends to be higher in photosynthesizing and undifferentiated tissues, calli and leaves showed higher sterol levels than bulbs and shoot-clumps, respectively (Table 4). The high capacity of P. maritimium cultures to biosynthesize alkaloids is clearly related to cell differentiation but the different metabolite profiles of the in vitro leaves and bulbs cannot be explained simply by the different cell organization in these organs. During in vitro organogenesis, photosynthesis starts with the differentiation of leaves in the shoot-clumps and plantlets. Further in vitro experiments need to be performed to elucidate the influence of light on both lipid and alkaloid synthesis. The close relationship between alkaloid accumulation and organ differentiation suggests that strategies involving differentiated plants rather than partially differentiated shoot-clumps should be applied for in vitro alkaloid production.
Fig. 3. GCeMS chromatograms of alkaloid (A) and lipid (B) fractions from in vitro leavs of P. maritimum. The numbers of compounds are congruent with those of Tables 2 and 4.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
831
Table 3 Contribution of the different alkaloid types to the alkaloid mixture of P. maritimum presented as a percentage of TIC (mean SD). Alkaloid type
Intact plants
Tyramine Narciclasine Galanthamine Haemanthamine Lycorine Tazettine Homolycorine
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
Leaves
4.5 5.3 0.5 32.3 51.1 2.7 3.3
<0.1 12.1 10.7 10.1 25.3 26.0 10.9
73.5 6a 10.3 e 6.2 10.0 3a e e
53.1 11.6a,b 26.8 15.5a 6.1 7.8a 6.5 4a 7.5 2.8a e e
35.3 17.2b,c 25.6 5.2a 0.2 10.1 5.6a 26.9 13.6a e 2.0 1.0a
14.3 20.7 7.8 42.3 14.4 0.3 0.1
Values followed by different superscripts within a line are significantly different according to Turkey’s multiple comparison test (P < 0.05).
L-Phe
OH
L-Tyr
HO HO HO
Norbelladine
CHO
H2N Me
OH O N
HO
3 O
1 NH O-Methylnorbelladine
R1
para-para' NHMe
O
2
para-ortho'
ortho-para'
O
OH
Narcicasine type
R2
O H
N
HO
O
Haemanthamine type 5 R1=OMe, R2=H 7 R1=OH, R2=H 11 R1=OMe, R2=OH 13 R1=OH, R2=OH
H 14
N
MeN
N 8
HO
O 16
O Homolycorine type
OMe
O
H
MeO
O O
H NMe
O
Galanthamine type 4 R1=OH, R2=Me 6 R1=OH, R2=H 15 R1=OH, R2=CHO
H
O
N R2
MeO
OH R1
O
Me
Tyramine type
MeO
O
N
Tyramine
Protocatechuic aldehyde
O R1 13 R2 R3
O O
N 10
Tazettine type 9 R1= R2=R3=H 12 R1=OH, R2=R3=H Fig. 4. Biosynthetic relationship of the alkaloids found in P. maritimum.
7.1c 3.1a 0.8a 6.5b 0.6a 0.1 0.1b
832
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
Table 4 Metabolites in the apolar fraction. Results represent the means SD of responce ratios using 100 mg of nonadecanoic acid as the quantitative internal standard. Compound
Nonanoic acid (C9:0, 17) Decanoic acid (C10:0, 18) Undecanoic acid (C11:0, 19) Octanedioic acid (20) Unknown (21) Lauric acid (C12:0, 22) Azelaic acid (23) Unknown (24) Tridecanoic acid (C13:0, 25) Tetradecenoic acid (C14:1, 26) Myristic acid (C14:0, 27) 9-Methyltetradecanoate (28) Unknown (29) Pentadecanoic acid (C15:0, 30) Phytyl derivative (31) Phytyl derivative (32) Phytyl derivative (33) Phytyl derivative (34) Hexadecanol (35) Hexadecenoic acid (C16:1, 36) Palmitic acid (C16:0, 37) Unknown (38) Unknown (39) Cyclopropaneoctanoic acid, 2-hexyl (40) Margaric acid (C17:0, 41) Unknown (42) Hexadecanoic acid, 2-hydroxy (43) Unknown (44) Linoleic acid (C18:2, 45) a-Linolenic acid (C18:3, 46) Oleic acid (C18:1n-9, 47) Stearic acid (C18:0, 48) Fatty acid, 3-hydroxy (49) Unknown (50) Octadecanoic acid, 2-hyroxy (51) Eicosadienoic acid (C20:2 n-6, 52) Eicosenoic acid (C20:1 n-9, 53) Eicosatrienoic acid (C20:3, 54) Arachidic acid (C20:0, 55) 4,8,12,16-Tetramethylheptadecan-4-olide (56) Unknown (57) Heneicosanoic acid (C21:0, 58) Unknown (59) Unknown (60) Unknown (61) Eicosanoic acid, 2-hydroxy (62) Unknown (63) Pentacosane (C25, 64) Docosenoic acid (C22:1, 65) Unknown (66) Behenic acid (C22:0, 67) Hexacosane (C26, 68) Tricosanoic acid (C23:0, 69) Docosanoic acid, 2-hydroxy (70) Heptacosanol (71) Heptacosane (C27, 72) Tetracosenoic acid (C24:1, 73) Lignoceric acid (C24:0, 74) Tricosanoic acid, 2-hydroxy (75) Octacosane (C28, 76) Pentacosanoic acid (C25:0, 77) Tetracosanoic acid, 2-hydroxy (78) Unknown (79) Nonacosane (C29, 80) Hexacosanol (81) Cerotinic acid (C26:0, 82) Unknown (83) Heptacosanoic acid (C27:0, 84) Hexacosanoic acid, 2-hydroxy (85) Unknown (86) Hentriacontane (87) b-Sitosterol (acetate?, 88) Montanic acid (C28:0, 89)
Rt
5.19 6.19 7.12 7.33 7.79 8.05 8.31 8.84 9.13 10.12 10.39 11.34 11.39 11.72 11.96 12.09 12.32 12.40 12.56 12.89 13.28 13.65 14.02 14.45 14.82 15.01 15.15 15.47 16.01 16.20 16.25 16.45 17.08 17.29 18.41 19.14 19.20 19.25 19.61 20.04 20.64 21.17 21.38 21.50 21.54 21.58 22.08 22.23 22.32 22.34 22.68 23.70 24.17 24.61 24.80 25.15 25.26 25.59 26.06 26.60 27.00 27.46 27.59 27.89 27.92 28.36 28.92 29.66 30.13 30.22 30.48 30.49 30.93
Base/Mþ Ions (m/z) 74/172 74/186 74/200 129/202() 75/74/214 152/216() 75/74/228 74/240 74/242 74/256 75/74/256 68/123/82/123/55/74/268 74/270 99/99/55/282 74/284 85/227/286 85/428 81/294 79/292 55/296 74/298 103/117/255/314 67/322 55/324 79/320 74/326 99/324 115/e 74/340 115/e 141/e 154/e 283/342 154/e 57/352 55/352 240/e 74/354 57/366 74/368 311/370 57/396() 57/380 55/380 74/382 325/384 57/394 74/396 339/398 97/e 57/408 97/e 74/410 75/e 74/424 367/426 97/e 57/436() 396/e 74/438
Intact plants
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
9 5 tr
21
tr tr
22 4 88 9 199 28
53 41 7 11 24 4 46 1 24 1 139 8 269 24
10 1 118 26 86 33 28 123 13
92 109 9 tr 43 9 91 72
11 30 134 tr 14 69 8
204 7 98 51 13 973 tr
75
112 75 102 285 180
tr 12 1 17 1 17 3 65 15 tr 12 1 84 11 51 15 5 94 8
tr
17 9 204 27 4649 38
tr 15 5 179 53 4072 250
77 173 26 2746 204
29 20 237 46
20 6 230 24
23 3 181 9
52 16
57 7
33 6
3742 409 2259 238 1062 302 777 90
4525 520 687 7 1925 287 657 163
27 4 95 30 43 9 tr 297 225
25 7 50 27 26 15 54 149 37
17 5 44 4 33 33
9
33 6 29 12
135
36 6
7
50
19 1
15
64
101 22 17 7
32 2 20 1 19 1 29 2 24 4 22 3 30 7 82 2
12 2 22 4 62 82 73 15 5 12 2 44 3
tr 172 2254
34 91 39 3687 394 1480 502
15 58 23 tr 54 21 6
69 6201 159 171 148 610 298 66 4105 10792 831 1586 46 645 19 17 12 89 138
76 tr 28 14
116 tr 57 65
34
17
93 tr 5 18 33
151 6 11 65 110
tr
604 380 628 7 8
18 22 5
78 70
33
5142 1627 2104 1301
794 133 914 727
382 224 20 6 130 50 33 5 44 26 250 13 38 20 578 389 12 1 15 1 95 57 51 39 84 32 181 17
192 16 17 1 127 66 47 34 25 3 142 24 92 274 35 13 58 61 45
3 6 3 4
133 36
21 3 149 31 71 70 9 15 5 tr 99 4 208 39 71 58 8 28 1 11 2 47 14
Leaves tr tr 11 10 38 30 13 1 52 40 130 108 tr 10 2 80 8 tr 25 11 57 5 22 4 18 14 75 24 38 26 137 16 3786 636 90 23 99 22 17 2 82 10 88 15 74 21 132 20 3820 864 2389 682 1028 261 493 86 22 3 199 49 35 6 77 28 75 20 94 71 11
13 2
81 tr 264 59 59 8 108 41 15 13 62 194 29 85 23 1 68 18
201 143
261 24 143 123
111 34
750 243 48 5 19 8
17 13 99 221 96
61 87 125 3
72 18 11 29 6
20 5
44 16 151 94
38 10 59 18
37 17 98 20
81 23 15 17 8
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
833
Table 4 (continued ) Compound
Rt
Tocopherol (90) Unknown (91) Campesterol (92) Sterol, unknown (93) Unknown (94) b-Sitosterol (95) Stigmastan-3-ol (96) Stigmasta-5,24-dien-3-ol (97) Dotriacontanoic acid (C32:0, 98)
31.04 31.47 32.21 32.62 32.70 33.43 33.54 34.01 36.59
Base/Mþ Ions (m/z) 165/430 337/e 400/400 396/e 97/e 414/414 416/416 314/412 74/494()
Intact plants
In vitro cultures
Bulbs
Callus
Shoot-clumps
Bulbs
Leaves
48 8 263 23 101 10
48 11 249 57 72 5 69 20 2025 271 39 8
44 7 128 10 65 23 29 19 1705 107 22 16
34 5 217 66 32 12 2007 503 92
17882 ± 1093
14597 ± 1443
17156 ± 3633
193 41
Total
Leaves 1939 64 347
753
4180 53 47 116
2599 158 29 3 40 17
10646
37880
22334 ± 2317
3. Materials and methods 3.1. Plant material Young fruits of P. maritimum L. plants (at the early stage of seed formation) were collected from a single cluster at the “Silistar” beach, Bulgaria in July 2005. They were carefully washed with tap water, surface sterilized by treatment with 70% of EtOH for 10 s and then with 7% of Ca(ClO2) (Sigma, USA) for 15 min. They were then thoroughly rinsed with sterile distilled water, dried on sterile tissue paper and used as explants in the experiments. Leaves and bulb from one plant were lyophilized and used for GC/MS analysis as described below. A voucher specimen (SOM Co-974) was authenticated by Dr. Luba Evstatieva and deposited in the Herbarium of the Institute of Botany (Bulgarian Academy of Sciences), Sofia.
3.2. In vitro cultures 3.2.1. Callus induction Surface sterilized young fruits were sliced and placed on Murashige and Skoog (MS) (Duchefa, The Netherlands) medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar” (Duchefa) and different concentrations and combinations of growth regulators as follows: auxins - 2,4-dichlorphenoxyacetic acid (2,4-D) (Sigma, USA) e 0.5; 1.0; 2.0; 3.0; 4.0 mg/L; a-naphtylacetic acid (NAA) (Duchefa, The Netherlands) e 1.0; 2.0 mg/L and cytokines - kinetin (Sigma, USA) e 0.5; 1.0; 2.0 mg/L; 6-benzylamynopurine (BAP) (Duchefa, The Netherlands) - 0.5; 1.0; 2.0 mg/L. Cultivation was carried out at 26 C in darkness.
3.2.2. Shoot formation The shoot cultures were established by planting the obtained calli on MS nutrient medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar”, 1.15 mg/L NAA and 2.0 mg/L BAP. The cultivation was carried out at 26 C under illumination 16:8 (light:dark). The further subcultivations of obtained shoot cultures were performed every 28 days under the same conditions. 3.2.3. In vitro plant regeneration In vitro plants were obtained from shoot cultures established by planting slices of young fruits on MS nutrient medium supplemented with 30 g/L sucrose, 5.5 g/L “Plant agar” and three combinations of growth regulators as follows: 0.2 mg/L 2,4D and 1.0 mg/L BAP; 1.15 mg/L NAA and 2.0 mg/L BAP; 5.0 mg/L NAA and 2.0 mg/L isopentenyladenosine (iPA) (Duchefa, The Netherlands). The cultivation was carried out at 26 C in darkness. The further subcultivations of obtained shoot cultures were performed every 28 days under the same conditions. At the end of cultivation period, the in vitro cultures were lyophilized before metabolite extraction.
3.3. Metabolite extraction The lipophilic fractions were obtained from 300 mg plant samples extracted exhaustively with 10 ml methanol (3 48 h) at room temperature until discoloration of the tissues. Before extraction, 50 mg of codeine and 100 mg of nonadecanoic acid were added as internal standards (IS). The extracts were evaporated and the dry residue was resolved in 3 mL of 2% H2SO4. The lipid fraction was first extracted with diethyl ether (3 5 mL) and then the
Table 5 Contribution of the main groups of compounds to the apolar fraction of P. maritimum in vitro cultures and intact plants. Results represent the means SD of response ratios using 0.100 mg of nonadecanoic acid as the quantitative internal standard. Intact plants
SFA 2(3)-Hydroxy FA MFA DFA TFA Total FA Fatty alcohols Alkanes Sterols Tocopherols Isoprenes Unknowns
In vitro cultures
Bulbs
Leaves
Callus
Shoot-clumps
Bulbs
Leaves
3.35 0.15 1.69 3.75 0.40 9.32 tr 0.10 0.99
10.85 0.52 0.93 4.12 10.88 27.30 0.60 0.10 4.63 1.94 0.66 2.56
8.42 2.08a 0.23 0.09a 2.42 0.98a 5.24 0.82a 1.63 0.13a 17.944.10 0.06 0.04a 0.57 0.06a 3.10 0.21a
6.52 0.78a,b 0.24 0.07a 1.29 0.37a 3.79 0.44a 2.26 0.24a 14.101.90 0.30 0.03b 0.25 0.03b 2.39 0.34a
4.66 0.59b 0.15 0.04a 2.15 0.35a 4.57 0.53a 0.69 0.01b 12.211.50 0.01 0.01a 0.21 0.02b 1.92 0.16b
5.27 0.92b 0.35 0.10b 1.26 0.30a 3.90 0.90a 2.40 0.69a 13.182.90 0.75 0.24c 0.02 0.02c 2.27 0.58a
0.25 0.5
0.15 0.07 0.85 0.29
0.20
0.91 0.20
tr 0.78 0.09
Values followed by different superscripts within a line are significantly different according to Turkey’s multiple comparison test (P < 0.05).
834
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
alkaloids by ethylacetate (3 5 mL) after basification of the solution with 25% of ammonia to pH 9-10. The organic solvent of the alkaloid fraction was evaporated and the residue was resolved in 100 mL methanol. The organic solvent of the lipid fraction was evaporated and the fatty acids were methylated with a solution of 5% of H2SO4 in methanol (0.5 mL) for 16 h at 70 C. The compounds were extracted with 3 mL n-hexane after cooling of the mixture and addition of 2 mL of aqueous solution (6%) of NaCl. The n-hexane was rinsed twice with 2 mL of aqueous solution (6%) Na2CO3 and evaporated. The dry residue was resolved in 100 mL of chloroform for further GC/MS analysis. 3.4. GC/MS analysis 3.4.1. Chromatographic conditions The GC/MS analyses were carried out on Hewlett Packard 6890 þ MSD 5975 equipment (Hewlett Packard, Palo Alto, CA, USA) operating in EI mode at 70 eV. An HP-5 MS column (30 m 0.25 mm 0.25 mm) was used. The temperature program was: 100e180 C at 15 C min1, 180e300 C at 5 C min1 and 10 min hold at 300 C. Injector temperature was 250 C. The flow rate of carrier gas (Helium) was 0.8 mL min1. Split ratio of 1:20 was used. One mL of the solutions were injected. 3.4.2. Alkaloid quantification A calibration graph for galanthamine quantification was obtained using solutions of 5, 10, 20, 50, 100, 200 and 400 mg/mL of galanthamine containing 50 mg/mL of IS (codeine). The ratios of the peak areas of selected ions (TIC mode) of galanthamine (m/z at 286) versus those of the IS (m/z at 299) were plotted against the corresponding concentration of galanthamine to obtain the calibration graph. The other alkaloids were quantified as galanthamine equivalents using the galanthamine calibration graph. This is not a true quantification, but it could be used for comparative studies. 3.5. Identification of the metabolites 3.5.1. Alkaloids Hordenine (1), ismine (2), trisphaeridine (3) galanthamine (4), norgalanthamine (6), vittatine (7), haemanthamine (11), tazettine (12), 11-hydroxyvittatine (13), lycorine (14), N-formylnorgalanthamine (15) and 8-O-demethylgalanthamine (16) were identified by comparing their GCeMS spectra and Kovats retention indices (RI) with those of authentic compounds isolated and identified by other spectrometric methods (NMR, UV, CD) by the Group of Natural Products, University of Barcelona. Buphanisine (5), anhydrolycorine (8), 11-deoxytazettine (9) and 11,12-didehydroanhydrolycorine (10) were identified by comparing their mass spectral fragmentation with standard reference spectra from NIST 05 database (NIST Mass Spectral Database, PC-Version 5.0e2005, National Institute of Standardization and Technology, Gaithersburg, MD). The configuration of the 5,10b-ethano bridge of the haemanthamine type alkaloids, which can be established by CD analysis of isolated compounds, was tentatively assigned at the -position like the compounds of this type previously reported for this species [16]. 3.5.2. Lipids The fatty acids in the apolar fraction were identified as fatty acid methyl esters (FAME) according to the characteristic mass spectral fragmentation of each subgroup (saturated, hydroxy, monoenoic, dienoic and trienoic FA), molecular ion, retention time and mass spectra available in the lipid library [18] and NIST 05 database. Fatty alcohols were identified by comparison of their mass spectra with those of the lipid library, NIST 05 database, chromatographic retention and characteristic [M]þ-18 fragments. Hydrocarbons,
sterols and tocopherol were identified by their RI and comparison of their spectra with those of authentic compounds in NIST 05 database and online available plant-specific database (The Golm Metabolome Database; http://csbdb.mpimp-golm.mpg.de/csbdb/ gmd/home/gmd_sm.html). The measured mass spectra were deconvoluted by the Automated Mass Spectral Deconvolution and Identification System (AMDIS) before comparison with the databases. The spectra of individual components were transferred to the NIST Mass Spectral Search Program MS Search 2.0 where they were matched against reference compounds. Response ratio was applied for quantification of the metabolites in the lipid fractions. Response ratios represented the peak area ratio using nonadecanoic acid (100 mg) as the quantitative IS and the respective metabolite in a sample whose weight was normalized to 1 g DW. Base peaks were used for MS deconvolution before peak area calculation (TIC mode). The RIs of the compounds were recorded with a standard n-hydrocarbon calibration mixture (C9-C36) (Restek, Cat no. 31614, supplied by Teknokroma - Spain) using AMDIS 3.6 software. 3.5.3. Statistical analysis Samples for GCeMS analysis were prepared from pooled explants from three independent cultivations of the in vitro cultures. The data were analyzed for statistical significance using analysis of variance (ANOVA single factor) test. P-values less than 0.05 were considered to be significant. All data were processed using GraphPad Prism version 3.00 for Windows (GraphPad Software, San DiegoCalifornia USA, www.graphpad.com). 3.6. Flow cytometry The flow cytometric analyses were carried out using a CyFlow SL Blue (20 mW solid state laser at 488 nm) Flow Cytometer (Partec GmbH, Münster, Germany), operated with FloMax for Windows XP (Partec GmbH, Münster, Germany) software. DNA histograms were recorded on a semi-logarithmic scale, since the histogram peaks on such a scale are evenly distributed along the abscissa [15] and thus the peak heights can be considered to be proportional to the proportions of nuclei at the corresponding ploidy levels. The cell nuclei were extracted from 20 to 100 mg of fresh leaves and/or in vitro culture tissues, and stained using the Otto buffers as described by Dolezel [11]. Every tissue was analyzed at least three times on different days. The resulting peaks typically had CV of 5 %e9% depending on the age and type of tissue used [1]. Prior to analysis, the instrument was checked for linearity with fluorescent beads (AlignFlow 2.5 mm, Invtrogen, Eugene, OR, USA), and the amplification settings were kept constant throughout the experiment. The nuclei were gated in an FSC vs. FL3 density plot to eliminate much of the debris. The genome sizes were determined by co-processing the sample and an internal standard with (Pisum sativum cv. Ctirad) with the known genome size of 9.09 pg per 2C nucleous [12]. To exclude a possible influence of secondary metabolites on the staining, the sample and the standard were first processed alone and the position of the peaks were compared to their positions when processed together. Acknowledgements This work was partially financed by the Generalitat de Catalunya (2005SGR-00020) and National Science Fund of Bulgaria (TK-Ƃ1605/2006). S. Berkov thanks the Spanish Ministerio de Educacion y Ciencia for a research fellowship (SB2004-0062). The authors also thank Dr. Asunción Marín, Serveis Cientificotècnics, Universitat de Barcelona, for performing the GCeMS analyses.
S. Berkov et al. / Plant Physiology and Biochemistry 48 (2010) 827e835
References [1] M. Barow, A. Meister, Endopolyploidy in seed plants is differently correlated to systematics, organ, life strategy and genome size, Plant Cell Environ. 26 (2003) 571e584. [2] J. Bastida, R. Lavilla, F. Viladomat, Amaryllidaceae alkaloids. in: G. Cordell (Ed.), The Alkaloids, vol. 63. Elsevier Inc, 2006, pp. 87e179. [3] S. Berkov, L. Evstatieva, S. Popov, Alkaloids in Bulgarian Pancratium maritimum L. Z. Naturforsch.C. 59 (2004) 65e69. [4] S. Berkov, A. Pavlov, V. Georgiev, J. Bastida, M. Burrus, M. Ilieva, C. Codina, Alkaloid synthesis and accumulation in Leucojum aestivum in vitro cultures, Nat. Prod. Commun. 4 (2009) 359e364. [5] Y. Bogdanova, M. Stanilova, Ch. Gussev, Y. Bosseva, T. Stoeva, In vitro propagation of Pancratium maritimum L. (Amaryllidaceae) by liquid cultures, Propag. Ornam. Plants 8 (2008) 45e46. [6] A. Blanckaert, L. Belingheri, J. Vasseur, J.-L. Hilbert, Changes in lipid composition during embriogenesis in leaves of Cichorium, Plant Sci. 157 (2000) 165e172. [7] W. Christie, Lipid Analysis. The Oily Press, Bridgewater, England, 2003. [8] C. Codina, Production of galanthamine by Narcissus tissues in vitro. in: G. Hanks (Ed.), Medicinal and Aromatic Plants e Industrial Profiles: Narcissus and Daffodil (The Genus Narcissus). Taylor and Francis, London and New York, 2002, pp. 215e241. [9] A. Cunha, M. Fernandes-Ferreira, Ontogenic variation in free and esterified fatty acids during somatic embriogenesis of flaz (Linum usitatissimum L.), Plant Sci. 164 (2003) 863e872. [10] M. Diop, A. Ptak, F. Chrétien, M. Henry, Y. Chapleur, D. Laurain-Mattar, Galanthamine content of bulbs and in vitro cultures of Leucojum aestivum, Nat. Prod. Commun. 1 (2006) 475e479. [11] J. Dolezel, W. Göhde, Sex determination in dioecious plants Melandrium album and M. rubrum using high-resolution flow-cytometry, Cytometry 19 (1995) 103e106. [12] J. Dolezel, J. Greilhuber, S. Lucretti, A. Meister, M.A. Lysak, L. Nardi, R. Obermayer, Plant genome size estimation by flow cytometry: inter-laboratory comparison, Ann. Bot. 82 (1998) 17e26. [13] O. Fiehn, Metabolomics e the link between genotypes and phenotypes, Plant Mol. Biol. 48 (2002) 155e171. [14] L. Georgieva, S. Berkov, V. Kondakova, J. Bastida, F. Viladomat, A. Atanassov, C. Codina, Alkaloid variability in Leucojum aestivum from wild populations, Z. Naturforsch. C 62 (2007) 627e635.
835
[15] L.J.W. Gilissen, M.J. van Staveren, J.C. Hakkert, M.J.M. Smulders, H.A. Verhoeven, J. Creemers-Molenaar, The competence of cells for cell division and regeneration in Tobacco explants depends on cellular location, cell cycle phase and ploidy level, Plant Sci. 103 (1994) 81e91. [16] O. Hoshino, The Amaryllidaceae alkaloids. in: G. Cordell (Ed.), The Alkaloids, vol. 51. Academic Press, 1998, pp. 324e417. [17] M. Kreh, R. Matusch, L. Witte, Capillary gas chromatography-mass spectrometry of Amaryllidaceae alkaloids, Phytochemistry 38 (1995) 773e776. [18] The Lipid library: http://www.lipidlibrary.co.uk/index.html [19] A. Maelicke, M. Samochocki, R. Jostok, A. Feherbacher, J. Ludwig, E.X. Albuquerque, M. Zerlin, Allosteric sensitization of nicotinic receptors by galanthamine, a new treatment strategy for the Alzheimer’s disease, Biol. Psychiatry 26 (2001) 279e288. [20] M. Nakano, T. Nomizu, K. Mizunashi, M. Suzuki, S. Mori, S. Kuwayama, M. Hayashi, H. Umehara, E. Oka, K. Kobayashi, Somaclonal variation in Tricyrtis hirta plants regenerated from 1-year-old embryogenic callus cultures, Sci.Hortic. 110 (2006) 366e371. [21] M. Nikolova, R. Gevrenova, Determination of phenolic acids in amaryllidaceae species by high performance liquid chromatography, Pharm. Biol. 43 (2005) 289e291. [22] D. Nikopoulos, A. Alexopoulos, In vitro propagation of an endangered medicinal plant: Pancratium maritimum L, J. Food Agr. Environ. 6 (2008) 390e398. [23] A. Pavlov, S. Berkov, E. Courot, T. Gocheva, D. Tuneva, B. Pandova, V. Georgiev, S. Yanev, M. Burrus, M. Ilieva, Galanthamine production by Leucojum aestivum in vitro systems, Process Biochem. 42 (2007) 734e739. [24] Y. Poirier, G. Ventre, D. Caldelary, Increased flow of fatty acids towards b-oxidation in developing seeds of Arabidopsis deficient in diacylglycerol acyltransferase activity or synthesizing medium-chain fatty acids, Plant Physiol. 121 (1999) 1359e1366. [25] J. Reina-Pinto, A. Yephremov, Surface lipids and plant defenses, Plant Physiol. Biochem. 47 (2009) 540e549. [26] H. Zahreddine, C. Clubbe, R. Baalbaki, A. Ghalayini, S.N. Talhouk, Status of native species in threatened Mediterranean habitats: the case of Pancratium maritimum L. (sea daffodil) in Lebanon, Biol. Conserv. 120 (2004) 11e18. [27] B.J.M. Zonneveld, I.J. Leitch, M.D. Bennett, First nuclear DNA amounts in more than 300 angiosperms, Ann. Bot. 96 (2005) 229e244. [28] M. Wink, Ecological role of alkaloids. in: E. Fattorusso, O. Taglialatela-Scafati (Eds.), Modern Alkaloids. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2008, pp. 3e24.