Biologia, Bratislava, 61/4: 473478, 2006 Section Botany DOI: 10.

2478/s11756-006-0079-8

Secondary metabolites during ontogenetic phase of reproductive structures in Hypericum maculatum

Pavol Mrtonfi1, Miroslav Repk1 & Lenka Mrtonfiov2
1 P. J. afrik University, Faculty of Science, Institute of Biology and Ecology, Department of Botany, Mnesova 23, SK-04154 Koice, Slovakia; tel.: +421-55-2342309, fax: +421-55-6337353, e-mail: pavol.marton@upjs.sk 2 P. J. afrik University, Botanical Garden, Mnesova 23, SK-04352 Koice, Slovakia

Abstract: The distribution patterns of avonoids hyperoside, isoquercitrin, quercitrin, quercetin, I3,II8-biapigenin and naphtodianthrones hypericin and pseudohypericin were studied in reproductive structures during ontogenetic phase of owering in Hypericum maculatum CRANTZ. Considerable dierences in the content of these secondary metabolites, in the particular ower parts were found. The content of all the metabolites studied is stable during the whole period of owering in green ower parts (sepals). In petals, stamens and pistils their content undergoes considerable change associated with the biological functions of particular metabolites. The most conspicuous changes during ontogenetic phase of owering were the decrease of hyperoside and isoquercitrin content in petals (average content in buds 1.589 mg g1 dry weight, average content in overblown owers 0.891 mg g1 dry weight), the decrease of the I3,II8-biapigenin content in stamens (in buds 1.189 mg g1 dry weight, in overblown owers 0.319 mg g1 dry weight), and the increase of hypericin and pseudohypericin content in both petals (total average content of hypericins in the buds 0.547 mg g1 dry weight; in overblown owers 0.792 mg g1 dry weight) and stamens (in buds 0.189 mg g1 dry weight; in overblown owers 0.431 mg g1 dry weight). Hypericins are absent in the pistil. The avonoids hyperoside and isoquercitrin, the content of which decreased during ontogenetic phase of owering, reach the highest contents in the pistil. Key words: ontogenetic phase of owering, Hypericum maculatum, secondary metabolites

Introduction The genus Hypericum includes over 450 species (Robson, 2002). The tetraploid species Hypericum perforatum is the best studied one from the aspect of sec lzl & Petersen, ondary metabolites (Kaul, 2000; Ho 2003). This is connected with the fact that H. perforatum is very popular in both folk medicine and pharmaceutical industry. This species is facultatively apomictic and, according to Robson (2002) it is a result of an ancient hybridization between H. maculatum subsp. hl. and H. attenuatum immaculatum (Murb.) A. Fro Choisy. Although the probable parent species, diploid H. maculatum Crantz is given in several pharmacopoeias together with H. perforatum as a component of the drug Herba hyperici, rather little attention was paid to its secondary metabolites and their dynamics. Brockmann & Sanne (1957) have already found a higher content of hypericin in the owers of H. maculatum (2.1 g kg1 dry weight) than in those of H. perforatum (1.38 g kg1 dry weight). The avonoids hyperoside, quercitrin and quercetin were identied by Michaluk (1961) and Leifertov (1966) in H. maculatum. Kment et al. (1990) also found isoquercitrin and Umek et al. (1999) determined that I3,II8-biapigenin

and amentoavone were also present. The avonoid rutin is often reported as absent from the avonoid pattern of H. maculatum or present only in traces (Michaluk, 1961; Leifertov, 1966; Brantner et al., 1994; Mrtonfi et al., 1996; Umek et al., 1999; Raduiene & Bagdonaite, 2002). However, a chemotype without rutin was also determined in the species H. perforatum (Mrtonfi et al., 2001). Literature data on the avonoid content in H. maculatum are sporadic only. At the same time, the amounts recorded dier substantially (Kment et al., 1990; Umek et al., 1999; Raduiene & Bagdonaite, 2002). This may be caused by dierent methodological approaches, above all in the sampling for analyses and in the subsequent processing. Such dierences can be found in the papers concerning hypericin content in the species H. maculatum (but also in many other papers concerning the species H. perforatum). The authors dier in the methods and in the plant material used for the subsequent analyses: non-owering vs. owering herb (the whole inorescences with certain amount of leaves and stems), or individual owers only. In a collected herb the ratio of buds, blooming and overblown owers is usually not quantied. This may contribute substantially to the variability of the data obtained. In addition, it is nec-

c 2006 Institute of Botany, Slovak Academy of Sciences

474 essary to stress that secondary metabolite content can vary largely already as a result of dierent ecological conditions, in which the plants grow. Also the subsequent laboratory processing after the sampling inuences the values obtained (Avato & Guglielmi, 2004). Not only older data but also the data in recent papers dier considerably. The above mentioned situation and our knowledge of variation in the secondary metabolites content during ontogenetic phase of owering in H. perforatum (Mrtonfi & Repk, 1994) as well as in particular ower parts (Repk & Mrtonfi, 1997), led us to the attempt to characterize more accurately the content of the selected secondary metabolites of H. maculatum (hyperoside, isoquercitrin, quercitrin, quercetin, I3,II8biapigenin, hypericin and pseudohypericin) in particular ower parts (sepals, petals, stamens and pistils) during ower ontogeny.
Material and methods Plant material Samples of Hypericum maculatum plants were collected in three localities in eastern Slovakia: 1. the meadow below the railway in the village of Mlynky, ca. 750 m a.s.l., 31. VII. 2001, leg. MRTONFI 23012303; 2. the meadow near the road in the village of Hnilk, ca. 700 m a.s.l., 31. VII. 2001, leg. MRTONFI 23042306; 3. the meadow near the trbsk Pleso lake in the High Tatra Mts, ca 1400 m a.s.l., 30. VIII. 2001, leg. MRTONFI 25052506. Together 8 plant populations were studied and six to eight plants were collected from each population. From these plants, samples of generative parts in 3 (in the case of pistil in 4) ontogenetic phases were taken: 1. ower buds of the size 61 mm; 2. owers in the phase of full opening of petals, anther had not released pollen; 3. owers overblown, petals are almost dried, but not fallen so far, anthers almost without pollen, pistil beginning to enlarge; 4. (only for pistil) pistil enlarged, green, seeds still white, sappy. The sepals, petals, stamens and pistils were excised from the owers in each ontogenetic phase. Five owers, or, as the case may be, buds were employed for preparation of one sample. Together 96 samples of 1.3. ontogenetic phases and 8 samples of pistils in ontogenetic phase 4 were prepared. The samples were air-dried and subsequently analyzed by HPLC. Voucher specimens of plants studied are deposited in KO (Herbarium of Botanical Garden of P. J. afrik University in Koice, Slovakia). Extraction and HPLC estimation of secondary metabolites Flower parts were extracted in methanol and immediately injected by analytic sample injector - 20 L (Ecom, Prague). The modied gradient method was used for resolving the avonoids, naphthodianthrones and acylphloroglu LZL & OSTROWSKI, 1987). We used the gradicinols (HO ent pump LCP 3001 (Ecom, Prague), variable UV/VIS detector LCD 2040 (Ecom, Prague), column 3.3150 mm, Separon SGX C18, 7 m (Tessek, Prague), the mobile phase: (A) acetonitrile:water:H3 PO4 (19:80:1), (B) acetonitrile:methanol:H3 PO4 (59:40:1) (Merck, Darmstadt), the ow rate 0.5 mL/min, wavelength 254 nm, gradient program 0 min-25% B, 5 min-30% B, 10 min-55% B, 15 min100% B, 25 min-100% B, integrator CSW (Ecom, Prague).

P. Mrtonfi et al.
The following standard compounds were used for comparative identication and quantitative determination: hyperoside, isoquercitrin, quercitrin, quercetin and hypericin (Roth). The 3,8-biapigenin was isolated and puried by partition and by column chromatography. UV spectral analysis was carried out to conrm the compound identication.

Results and discussion Evaluating our results (Tabs 13) we recorded signicant dierences in the content of particular metabolites between dierent ower parts as well as dierences in dynamics of their content during ontogeny of owering. Prior to giving other results we consider important to mention two ndings: i) From the aspect of metabolite content dynamics during ontogenetical development, the content of any metabolite in the calyx shows up invariable; we did not record any case of signicant dierence in metabolite content during ower ontogeny. The calyx thus appears to be a structure, in which secondary metabolites do not undergo substantial changes during owering. ii) Quercitrin content does not change signicantly during ontogenetical development in particular ower parts, although the differences among the parts are present. In owers and stamens its content varies from 0.040 to 0.267 mg g1 dry weight with average values 0.0810.149 mg g1 dry weight. In the calices it is approximately half-value, and in the pistils it is very low (0.028 mg g1 dry weight was the maximum value found, for average data see Tab. 1). The contents of the other avonoids depend on both ower part studied and ontogenetic phase of the ower. These diversities are probably associated with specic functions of avonoids and biavonoids e.g. in pollen development and germination, in plant fertilization, or assumptive natural endogenous regulators of polar auxin transport (Ylstra et al., 1992; Vogt & Taylor, 1995; Jacobs & Rubery, 1988; Geiger, 1994). The content of quercetin glycosides hyperoside and isoquercitrin is relatively high and in sepals, petals and pistils it varies approximately within the range of 1-2 mg g1 dry weight. In the stamens their content is only about one third of this value. In all ower parts (except for calyx) the contents vary during ontogeny (Tabs 2, 3); generally decreasing with the progressing ower development. This is the most conspicuous in the petals: from the average content 1.589 in the phase of bud it decreased to 0.891 mg g1 dry weight in the phase of overblowing. In the pistils it is the avonoid with the highest content while the content of other avonoids is very low. The biavonoid I3,II8-biapigenin has a specic position between avonoids. It occurs in high amounts in stamens while in other ower parts its content is low (cf. Tab. 1). Its dynamism in the course of ontogeny is considerable. The highest relative content is in the stamens of buds (1.554 mg g1 dry weight is the maximum value

Secondary metabolites in Hypericum maculatum

475

Table 1. Average content and coecient of variation for secondary metabolites in particular ower parts during the ower developmental stage of Hypericum maculatum. Part of ower Phase xa cvb x cv x cv x cv x cv x cv x cv x cv x cv x cv x cv x cv x cv Hyperoside + isoquercitrin 1.494 0.234 1.583 0.135 1.585 0.150 1.589 0.112 1.237 0.194 0.891 0.285 0.534 0.169 0.586 0.327 0.279 0.239 1.548 0.187 1.627 0.213 1.240 0.251 0.686 0.364 Quercitrin Quercetin I3,II8-biapigenin Pseudohypericin Hypericin

Sepals

1. buds 2. owers 3. overblown

0.046 0.277 0.054 0.537 0.049 0.379 0.105 0.368 0.081 0.315 0.149 0.515 0.109 0.343 0.101 0.355 0.119 0.287 0.007 0.821 0.008 0.768 0.010 0.755 0.007 0.936

0.003 0.361 0.004 0.302 0.004 0.423 0.020 0.531 0.013 0.391 0.264 0.457 0.018 0.774 0.036 0.384 0.137 0.299 Traces 0.002 0.573 0.012 0.298 0.001 2.048

0.008 0.641 0.017 0.895 0.017 0.451 0.016 0.236 0.067 0.650 0.061 0.331 1.189 0.163 0.646 0.430 0.319 0.238 0.003 1.500 0.008 0.874 0.010 1.029 0.002 2.033

0.031 0.800 0.038 0.762 0.034 0.728 0.451 0.388 0.363 0.307 0.640 0.440 0.161 0.285 0.180 0.285 0.373 0.333

0.007 1.127 0.010 0.884 0.009 0.855 0.096 0.271 0.089 0.338 0.152 0.268 0.028 0.308 0.032 0.260 0.058 0.181

Petals

1. buds 2. owers 3. overblown

Stamens

1. buds 2. owers 3. overblown

Pistils

1. buds 2. owers 3. overblown 4. rippening

a b

Average content (mg g1 dry mass) of the metabolite Coecient of variation

Table 2. Shortened results of ANOVA for 3 (pistils 4) dierent ower ontogenetic stages of sepals, petals, stamens and pistils and secondary metabolites. Part of ower Fa Sb F S F S F S Hyperoside + isoquercitrin 0.25 0.781 16.71 <0.001 11.50 <0.001 14.05 <0.001 Quercitrin Quercetin I3,II8-biapigenin Pseudohypericin Hypericin

Sepals Petals Stamens Pistils

a b

0.24 0.792 3.13 0.065 0.48 0.625 0.34 0.797

0.60 0.560 29.30 <0.001 41.48 <0.001 48.37 <0.001

1.99 0.162 7.05 0.005 33.60 <0.001 1.36 0.275

0.15 0.864 3.42 0.052 14.28 <0.001

0.29 0.753 7.80 0.003 22.84 <0.001

F-statistic Signicance, statistically signicant values (<0.05) are marked as boldface

obtained), it decreases later, also in association with the loss of pollen during blooming. The pollen is probably the place where biapigenin is considerably accumulated (similarly to H. perforatum, Repk, unpubl. results). In contrast to these processes, its content is lowest in the petals of bud and remains steady during the period of blooming (Tab. 3). The content of quercetin aglycone is predominantly relatively low (hundredths to thousandths of mg g1 dry weight), except for the third phenological phase

(overblown owers). In this phase its content in petals and stamens is many times higher, which is obviously connected with the degradation of quercetine glycosides. As stated by several authors (Brantner et al., 1994; Kartnig et al., 1997; Seidler-Lozykowska et al., 1999, Tekeov et al., 2000; SeidlerLozykowska, 2003), from the whole-plant aspect, the largest amount of avonoids in the herb was detected at the beginning of blooming or just before the blooming

476

P. Mrtonfi et al.

Table 3. Results of Sche test (numbers indicate the homogeneous groups) for statistically signicant results of ANOVA for 3 (pistils 4) dierent ower ontogenetic stages of petals, stamens and pistils and secondary metabolites. Petals Hyperoside + isoquercitrin buds owers overblown rippening 3 2 1 Petals Quercetin Petals I3,II8biapigenin 1 2 2 2 Petals Hypericin Pistils Hyperoside + isoquercitrin 2 2 2 1 Pistils Quercetin

1 1

1 1 2

1 1 2 1

Stamens Hyperoside + isoquercitrin buds owers overblown 2 2 1

Stamens Quercetin

Stamens I3,II8biapigenin 3 2

Stamens Pseudohypericin 1 1 2

Stamens Hypericin

1 1 2 1

1 1 2

the phase of well-developed bud, and then decreased as the plant bloomed and overbloomed. As it is clear from our results, these processes are not universal depending on the function of particular avonoids in biological processes and they are localized in dierent plant parts. Similar ndings were published in our earlier papers concerning H. perforatum (Mrtonfi & Repk, 1994; Repk & Mrtonfi, 1997, Tekeov et al., 2000). The naphtodianthrones hypericin and pseudohypericin are compounds frequently studied in the genus Hypericum. Data in articles, however, can dier considerably. For example, in ve recent years (1999-2004) we recorded 4 papers giving the content of hypericins for the species H. maculatum: the lowest content of hypericins (0.058% hypericin + pseudohypericin) was given by Kitanov (2001): however, it is not clear which plant part was used for the analyses. A similar average value 0.55% hypericin (however, it is not clear if the total content of hypericin and pseudohypericin was concerned) was given by Kireeva et al. (1999) for a vegetative developmental stage of the plants, 1.06% for the massive ower buds, 0.71% for the massive owering, and 0.33% for the stage of seed capsule formation. Bagdonaite and Raduiene (2002) determined the content of hypericin between 0.2660.495 mg g1 and the content of pseudohypericin between 0.6092.098 mg g1 for ower extract. Very high values of the pseudohypericin content in the owers were published by Umek et al. (1999): 5.0311.03 mg g1 dry weight with average 8.03 mg g1 and of the hypericin 1.252.48 mg g1 dry weight. Umek et al. (1999) unambiguously separated the data on owers from the data on vegetative parts where they found an average content of hypericin 0.46 mg g1 and pseudohypericin 1.99 mg g1 dry weight. Naphtodianthrones are accumulated mainly in dark tubular and spheroid nodules in plant tissues, similarly as in H. perforatum (Ciccarelli et al., 2001; Onelli et al., 2002). These nodules do not occur in the pistils of the species H. maculatum, therefore the hyper-

icins are found only in calyx, corolla and stamens. In the green plant parts the content of hypericins is lower than that in owers, which was conrmed also in this study. In sepals the content of hypericins ranged (regardless ontogenetic phase) from trace amounts to 0.073 mg g1 dry weight (pseudohypericin) and 0.025 mg g1 dry weight (hypericin), the sample with the highest value contained 0.098 mg g1 dry weight (pseudohypericin + hypericin). In stamens and petals the content of hypericins was substantially higher and expressed also a higher dynamics. The change in pseudohypericin content in petals during the ontogenetic phases was on the edge of signicance (p = 0.052), while the change in pseudohypericin content in stamens and hypericin in petals was highly signicant (Tabs 2, 3). When compared with the rst and the second phase (pseudohypericin in stamens: minimum 0.094 mg g1 dry weight to maximum 0.267 mg g1 dry weight; hypericin: minimum 0.017 mg g1 dry weight in stamens to 0.135 mg g1 dry weight in petals), in the third phase (the phase of overblowing), the content of hypericins was higher (pseudohypericin: petals 0.2761.070 mg g1 dry weight, stamens 0.2030.548 mg g1 dry weight; hypericin: petals 0.0760.197 mg g1 dry weight, stamens 0.0400.077 mg g1 dry weight; for average values see Tab. 1). This result seems to be inconsistent with many other data, although they concern mainly the species H. perforatum. These data conrmed that the highest content of hypericin was determined either in the herb harvested in full blossom and it decreased when the plant overbloomed (Brantner et al., 1994; Kartnig et al., 1997; Seidler-Lozykowska et al., 1999, Tekeov et al., 2000; Seidler-Lozykowska, 2003) or in the ower buds and it decreased when the ower bloomed and overbloomed (Mrtonfi & Repk, 1994). The explanation resides in methodological approach. The evaluation of whole owers (or herb) is dierent from that of ower parts, as it was in our case. In the evaluation of whole owers in the period of overblowing the volume of green parts of ripening pistil is growing,

Secondary metabolites in Hypericum maculatum therefore total hypericin content in dry mass is lower than that in buds and blooming owers. The evaluation of particular ower parts showed that in the course of blooming the accumulation of hypericins in petals and stamens increased gradually in relation to slight changes in dry mass value. This process is documented by our data on the relative weight of dry mass of particular ower parts during 3 (or 4) ontogenetic phases studied. The values for particular phases are given gradually as follows: average dry weight of one corolla (5 petals): 2.653.662.64 mg; average dry weight of androecium in ower (the set of many stamens): 2.62 2.601.65 mg; average dry weight of pistil in ower: 1.121.482.257.20 mg. The above mentioned data indicate that with overblowing the ratio of dry weight of particular ower parts has been changed. While the dry weight of petals decreases because of degradation processes, similarly to the dry weight of stamens (also owing to the release of pollen), the dry weight of pistil increased. In the phase 2 (blooming) the ratio of pistil dry weight to total ower dry weight was 19.1% while in the phase 3 (overblown ower) this ratio was already 34.4%. These results conrmed the necessity of being careful when comparing the values obtained from evaluation of the secondary metabolite contents in the plants. Another methodical approach can provide dierent results for the same material, which is particularly important to be considered in wide ecological, seasonal and topological variation in metabolite content, which was documented e.g. for the species H. perforatum (Pietta et al., 2001; Southwell & Bourke, 2001; Walker et al., 2001; Sirvent et al., 2002; Male et al., 2004). On the other hand, particular factors of environment do not necessarily inuence secondary metabolite content. When the correlation between altitude of the localities studied and secondary metabolite content was examined, no statistically signicant dierence was found in H. maculatum owers.
Acknowledgements Support for this research was provided by the Grant Agency for Sciences (VEGA), Bratislava, Slovakia (grant No. 1/9206/02). Thanks are also due to Mrs. Anna MICHALOV for technical help in HPLC analyses. References
AVATO, P. & GUGLIELMI, G. 2004. Determination of major constituents in St. Johns wort under dierent extraction conditions. Pharm. Biol. 42: 8389. BAGDONAITE, E. & RADUIENE, J. 2002. Analysis of variation of phenotypical characters in Hypericum maculatum CRANTZ wild populations. Biologija, pp. 4649. BRANTNER, A., KARTNIG, Th. & OUEHENBERGER, F. 1994. Vergleichende phytochemische Untersuchungen an Hypericum perforatum L. und Hypericum maculatum CRANTZ. Sci. Pharm. 62: 261276. BROCKMANN, H. & SANNE, W. 1957. Zur Kenntnis des Hypericins und Pseudohypericins. Chem. Ber. 90: 24802491.

477
CICCARELLI, D., ANDREUCCI, A.C. & PAGNI, A.M. 2001. The black nodules of Hypericum perforatum L. subsp. perforatum: morphological, anatomical and histochemical studies during the course of ontogenesis. Israel J. Plant Sci. 49: 33 40. GEIGER, H. 1994. Biavonoids and triavonoids, pp. 95115. In: HARBORNE, J.B. (ed.), The avonoids. Advances in research since 1985, Chapman & Hall, London. LZL, J. & OSTROWSKI, H. 1987. Johanniskraut (Hypericum HO perforatum L.) HPLC-Analyse der wichtigen Inhaltsstoe und deren Variabilit at in einer Population. Deutsche Apoth.Ztg., Berlin, 127: 12271230. LZL, J. & PETERSEN, M. 2003. Chemical constituents of HyHO pericum ssp., pp. 7793. In: ERNST, E. (ed.), Hypericum. The genus Hypericum. Taylor and Francis, London & New York. JACOBS, M. & RUBERY, P.H. 1988. Naturally occurring auxin transport regulation. Science 241: 346349. SSER, L. 1997. Johanniskraut KARTNIG, Th., HEYDEL, B. & LA aus Schweizer Arzneipanzenkultur. Agrarforschung 4: 299 302. atsKAUL, R. 2000. Johanniskraut: Botanik, Inhaltsstoe, Qualit kontrolle, Pharmakologie, Toxikologie und Klinik. Wiss. Verl.-Ges., Stuttgart, 187pp. KIREEVA, T.B., SHARANOV, U.L. & LETCHAMO, W. 1999. Biochemical and eco-physiological studies on Hypericum spp., pp. 467468. In: JANICK, J. (ed.), Perspectives on new crops and new uses, ASHS Press, Alexandria. KITANOV, G.M. 2001. Hypericin and pseudohypericin in some Hypericum species. Biochem. Syst. Ecol. 29: 171178. KMENT, V., PNKOV, I. & STAR, F. 1990. Maten rostliny drogy Herba hyperici. eskosl. Farm. 39: 323326. LEIFERTOV, I. 1966. Studium avonoid v eskoslovenskch druzch rodu Hypericum. Preslia 38: 386390. , V. B., ZUNTAR, I. & MALE, Z., PLAZIBAT, M., VUNDAC , K.H. 2004. Thin-layer chromatographic analysis of PILEPIC avonoids, phenolic acids, and amino acids in some Croatian Hypericum taxa. J. Plan. Chromatogr. 17: 280285. MRTONFI, P. & REPK, M. 1994. Secondary metabolites during ower ontogenesis of Hypericum perforatum L. Zahradnictv 21: 3744. MRTONFI, P., REPK M. & MIHOKOV, L. 1996. Hypericum maculatum CRANTZ subsp. maculatum H. perforatum L. (Hypericaceae): Corroboration of natural hybridization by secondary metabolite analysis. Folia Geobot. Phytotax. 31: 245250. MRTONFI, P., REPK, M., CICCARELLI, D. & GARBARI, F.2001. Hypericum perforatum L. chemotype without rutin from Italy. Biochem. Syst. Ecol. 29: 659661. MICHALUK, A. 1961. Badania nad avonoidami w gatunkach rodzaju Hypericum II. Diss. Pharm. 13: 7379. ONELLI, E., RIVETTA, A., GIORGI, A., BIGNAMI, M., COCUCCI, M. & PATRIGNANI, G. 2002. Ultrastructural studies on the developing secretory nodules of Hypericum perforatum. Flora 197: 92102. PIETTA, P., GARDANA, C. & PIETTA, A. 2001. Comparative evaluation of St. Johns wort from dierent Italian regions. Farmaco 56: 491496. RADUIENE, J. & BAGDONAITE, E. 2002. Phenotypic variation in Hypericum perforatum L. and H. maculatum CRANTZ wild populations in Lithuania. J. Herbs Spices Med. Plants 9: 345 351. REPK, M. & MRTONFI, P. 1997. The localization of secondary substances in Hypericum perforatum ower. Biologia, Bratislava 52: 9194. ROBSON, N.K.B. 2002. Studies in the genus Hypericum L. (Guttiferae) 4(2). Section 9. Hypericum sensu lato (part 2): subsection 1. Hypericum series 1. Hypericum. Bull. Nat. Hist. Mus. London, Bot. 32: 61123. SEIDLER-LOZYKOWSKA, K. 2003. Secondary metabolites content of Hypericum sp. in dierent stages and plant parts, pp. 100 105. In: ERNST. E. (ed.): Hypericum. The genus Hypericum. Taylor and Francis, London & New York.

478
SEIDLER-LOZYKOWSKA, K., DABROWSKA, J. & ZYGMUNT, B. 1999. Content of active substances in herb of St. Johns wort (Hypericum perforatum L.) cvar. Topaz in dierent vegetation phases. Herba Pol. 45: 169172. SIRVENT, T.M., WALKER, L., VANCE, N. & GIBSON, D.M. 2002. Variation in hypericins from wild populations of Hypericum perforatum L. in the Pacic Northwest of the U.S.A. Econ. Bot. 56: 4148. SOUTHWELL, I.A., & BOURKE, C.A. 2001. Seasonal variation in hypericin content of Hypericum perforatum L. (St. Johns wort). Phytochem. 56: 437441. TEKEOV, D., REPK, M., ZEMKOV, E. & TTH, J. 2000. Quantitative changes of dianthrones, hyperforin and avonoids content in the ower ontogenesis of Hypericum perforatum. Planta Med. 66: 778780. UMEK, A., KREFT, S., KARTNIG, Th. & HEYDEL, B. 1999. Quantitative phytochemical analyses of six Hypericum species growing in Slovenia. Planta Med. 65: 388390.

P. Mrtonfi et al.
VOGT, T. & TAYLOR, L.P. 1995. Flavonol 3-O-glycosyltransferases associated with Petunia pollen produce gametophytespecic avonol diglycosides. Plant Physiol. 108: 903911. WALKER, L., SIRVENT, T., GIBSON, D. & VANCE, N. 2001. Regional dierences in hypericin and pseudohypericin concentrations and ve morphological traits among Hypericum perforatum plants in the northwestern United States. Can. J. Bot. 79: 12481255. GER, YLSTRA, B., TOURAEV, A., BENITO MORENO, R.M., STO E., VAN TUNEN, A.J., VICENTE, O., MOL, J.N.M. & HEBERLE-BORS, E. 1992. Flavonols stimulate development, germination, and tube growth of tobacco pollen. Plant Physiol. 100: 902907. Received Nov. 15, 2005 Accepted May 4, 2006