Ginkgo Biloba - PDF Free Download (2024)

GINKGO BILOBA

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

Medicinal and Aromatic Plants—Industrial Profiles Individual volumes in this series provide both industry and academia with in-depth coverage of one major medicinal or aromatic plant of industrial importance. Edited by Dr Roland Hardman Volume 1 Valerian edited by Peter J.Houghton Volume 2 Perilla edited by He-Ci Yu, Kenichi Kosuna and Megumi Haga Volume 3 Poppy edited by Jenö Bernáth Volume 4 Cannabis edited by David T.Brown Volume 5 Neem H.S.Puri Volume 6 Ergot edited by Vladimír Kren and Ladislav Cvak Volume 7 Caraway edited by Éva Németh Volume 8 Saffron edited by Moshe Negbi Volume 9 Tea Tree edited by Ian Southwell and Robert Lowe Volume 10 Basil edited by Raimo Hiltunen and Yvonne Holm Volume 11 Fenugreek edited by Georgious Petropoulos Volume 12 Ginkgo biloba edited by Teris A.van Beek Other volumes in preparation Allium, edited by K.Chan Alloes, edited by A.Dweck and R.George Artemisia, edited by C.Wright Cardamom, edited by P.N.Ravindran and K.J.Madusoodanan Chamomile, edited by R.Franke and H.Schilcher Cinnamon and Cassia, edited by P.N.Ravindran and S.Ravindran Colchicum, edited by V.Šimánek Curcuma, edited by B.A.Nagasampagi and A.P.Purohit Please see the back of this book for other volumes in preparation in Medicinal and Aromatic Plants—Industial Profiles. Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA

Edited by Teris A.van Beek Laboratory of Organic Chemistry Wageningen Agricultural University The Netherlands

harwood academic publishers

Australia • Canada • France • Germany • India • Japan Luxembourg • Malaysia • The Netherlands • Russia • Singapore Switzerland

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

This edition published in the Taylor & Francis e-Library, 2006. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group. All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, without permission in writing from the publisher. Amsteldijk 166 1st Floor 1079 LH Amsterdam The Netherlands

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library. ISBN 0-203-30494-2 Master e-book ISBN

ISBN 0-203-34306-9 (Adobe eReader Format) ISBN: 90-5702-488-8 (Print Edition) ISSN: 1027-4502

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CONTENTS

Foreword

vii

Preface to the Series

ix

Contributors

xi

Introduction Teris A.van Beek

1

1 The Evolution, Ecology, and Cultivation of Ginkgo biloba Peter Del Tredici

7

2 Ginkgo biloba L.: Aspects of the Systematical and Applied Botany Volker Melzheimer and Johannes J.Lichius

25

3 Lignification of Xylem Cell Walls of Ginkgo biloba Noritsugu Terashima and Kazuhiko f*ckushima

49

4 Cultivation of Ginkgo biloba on a Large Scale Dominique Laurain

63

5 Plant Cell Biotechnology of Ginkgo Danielle Julie Carrier and Dominique Laurain

81

6 Ginkgo biloba—Large Scale Extraction and Processing Joe O’Reilly

99

7 Chemical Constituents of Ginkgo biloba Andreas Hasler

109

8 A Personal Account of the Early Ginkgolide Structural Studies Koji Nakanishi

143

9 Chemical Analysis of Ginkgo Terpene Trilactones Teris A.van Beek

151

10 The Analysis of Ginkgo Flavonoids Otto Sticher, Beat Meier and Andreas Hasler

179

11 Occurrence and Analysis of Alkyl Phenols in Ginkgo biloba Luisella Verotta, Paolo Morazzoni and Federico Peterlongo

203

12 Occurrence and Analysis of Ginkgo Polyprenols Hoon Huh

215

13 Considerations in the Development of the U.S. Pharmacopoeia’s Monograph on Ginkgo biloba L. V.Srini Srinivasan

229

v Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

vi

CONTENTS

14 Industrial Quality Control of Ginkgo Products Fabrizio F.Camponovo and Fabio Soldati

245

15 History, Development and Constituents of EGb 761 Katy Drieu and Hermann Jaggy

267

16 In Vitro Studies of the Pharmacological and Biochemical Activities of Ginkgo biloba Extract (EGb 761) and its Constituents Francis V.DeFeudis and Katy Drieu

279

17 In Vivo Studies of the Pharmacological and Biochemical Activities of Ginkgo biloba Extract (EGb 761) and its Constituents Katy Drieu and Francis V.DeFeudis

303

18 The Neuroprotective Properties of Ginkgo Extracts Cynthia L.Darlington, Paul F.Smith and Karyn Maclennan

331

19 Ginkgo biloba Extracts for the Treatment of Cerebral Insufficiency and Dementia Joerg Schulz, Peter Halama and Robert ho*rr

345

20 Clinical Uses of Ginkgo biloba Extract in the Field of Peripheral Arterial Occlusive Disease (PAOD) Bernard Bulling, Norbert Clemens and Volker Dankers

371

21 Efficacy of Ginkgo biloba Special Extracts—Evidence from Randomized Clinical Trials Martien C.J.M.van Dongen, Erik van Rossum and Paul Knipschild

385

22 Adverse Effects and Toxicity of Ginkgo Extracts Herman J.Woerdenbag and Peter A.G.M.De Smet

443

23 Food Poisoning by Ginkgo Seeds: The Role of 4-O-Methylpyridoxine Keiji Wada

453

24 Homeopathic Uses of Ginkgo biloba Frans M.van den Dungen

467

25 Cosmetical Uses of Ginkgo Extracts and Constituents Ezio Bombardelli, Aldo Cristoni and Paolo Morazzoni

475

26 Patents of Ginkgo biloba and its Constituents Marian A.J.T.van Gessel

491

27 The Marketing and Economic Aspects of Ginkgo biloba Products Gopi Warrier and Amy Corzine

517

28 The Ginkgo, Past, Present and Future Yves Christen

523

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

FOREWORD

It is a very encouraging and healthy sign that folk medicinal knowledge and practice has started to gain recognition among the populace and the medical professionals, although there still exist hard core opponents who only think in terms of so-called Western drugs. The revival in interest is reflected in the recent establishment of chairs in alternative medicine or natural medicine at universities even in the USA where recognition lagged far behind other continents. There is no doubt that folk medicinal herbs, which have been used for thousands of years, are effective and complementary to modern medicine. The knowledge and practice in natural medicine is comfortably accepted in the households of most countries independent of the stage of development of that nation. The interest has been traditionally high in countries such as China, Germany and Japan, and certainly in undeveloped areas healing through herb treatment is the norm. The official record of herbs, which started in China around 2800 BC, continues to provide the Chinese medicine and pharmacognosy community with a wealth of precious knowledge. Folk medicinal treatment is also widespread in the South American and African rain forests and similar areas, but here the knowledge is transmitted to the next generation through the medicine person or sharman and not through writing. The mode of action of plant extracts and other folk medicines are far more complicated than that of a single compound because they are mixtures of numerous constituents (a crude plant extract may contain hundreds of constituents). It is probably true that only a few of the constituents are responsible for the healing effects, while others are merely present without exerting any effects. Understanding the mode of action of a Western drug is already complicated enough: what is the receptor(s) and what chain of events ensue after the interaction leading to healing? Rationalization and understanding of the effects of folk medicine is no doubt a most challenging but worthwhile target for future interdisciplinary studies. A broad interdisciplinary approach to clarify how folk medicine works was almost impossible until quite recently because of the lack of sophisticated technology and general knowledge. However, we can now start to be engaged in such approaches and disentangle the role of respective constituents and their synergism. Ginkgo biloba, one of the oldest of the documented Chinese herbs, has a proven remedial record through its hundreds of years of use. At least one group of constituents, ginkgolide B and related terpenoid lactones have been shown by modern assays to selectively inhibit platelet activation although this has not led to a single-component drug. Nevertheless, the sale of the crude extract is increasing at a phenomenal rate throughout the world. Not many people would resist taking G.biloba extract when statistics have shown that it helps enhance memory and has a positive effect in slowing vii Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

viii

FOREWORD

Alzheimer’s disease, with no known adverse effects. It is most opportune that this plant has been selected to be included in the present series and that the editor has taken the time and effort to select the topics and contributors for an updated picture of this fantastic plant from nature. Koji Nakanishi

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PREFACE TO THE SERIES

There is increasing interest in industry, academia and the health sciences in medicinal and aromatic plants. In passing from plant production to the eventual product used by the public, many sciences are involved. This series brings together information which is currently scattered through an ever increasing number of journals. Each volume gives an in-depth look at one plant genus, about which an area specialist has assembled information ranging from the production of the plant to market trends and quality control. Many industries are involved such as forestry, agriculture, chemical, food, flavour, beverage, pharmaceutical, cosmetic and fragrance. The plant raw materials are roots, rhizomes, bulbs, leaves, stems, barks, wood, flowers, fruits and seeds. These yield gums, resins, essential (volatile) oils, fixed oils, waxes, juices, extracts and spices for medicinal and aromatic purposes. All these commodities are traded worldwide. A dealer’s market report for an item may say “Drought in the country of origin has forced up prices”. Natural products do not mean safe products and account of this has to be taken by the above industries, which are subject to regulation. For example, a number of plants which are approved for use in medicine must not be used in cosmetic products. The assessment of safe to use starts with the harvested plant material which has to comply with an official monograph. This may require absence of, or prescribed limits of, radioactive material, heavy metals, aflatoxins, pesticide residue, as well as the required level of active principle. This analytical control is costly and tends to exclude small batches of plant material. Large scale contracted mechanised cultivation with designated seed or plantlets is now preferable. Today, plant selection is not only for the yield of active principle, but for the plant’s ability to overcome disease, climatic stress and the hazards caused by mankind. Such methods as in vitro fertilisation, meristem cultures and somatic embryogenesis are used. The transfer of sections of DNA is giving rise to controversy in the case of some end-uses of the plant material. Some suppliers of plant raw material are now able to certify that they are supplying organically-farmed medicinal plants, herbs and spices. The Economic Union directive (CVO/EU No 2092/91) details the specifications for the obligatory quality controls to be carried out at all stages of production and processing of organic products. Fascinating plant folklore and ethnopharmacology leads to medicinal potential. Examples are the muscle relaxants based on the arrow poison, curare, from species of Chondrodendron, and the antimalarials derived from species of Cinchona and Artemisia. The methods of detection of pharmacological activity have become increasingly reliable and specific, frequently involving enzymes in bioassays and avoiding the use of laboratory animals. By using bioassay linked fractionation of ix Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

x

PREFACE TO THE SERIES

crude plant juices or extracts, compounds can be specifically targeted which, for example, inhibit blood platelet aggregation, or have antitumour, or antiviral, or any other required activity. With the assistance of robotic devices, all the members of a genus may be readily screened. However, the plant material must be fully authenticated by a specialist. The medicinal traditions of ancient civilisations such as those of China and India have a large armamentarium of plants in their pharmacopoeias which are used throughout South East Asia. A similar situation exists in Africa and South America. Thus, a very high percentage of the world’s population relies on medicinal and aromatic plants for their medicine. Western medicine is also responding. Already in Germany all medical practitioners have to pass an examination in phytotherapy before being allowed to practise. It is noticeable that throughout Europe and the USA, medical, pharmacy and health related schools are increasingly offering training in phytotherapy. Multinational pharmaceutical companies have become less enamoured of the single compound magic bullet cure. The high costs of such ventures and the endless competition from me too compounds from rival companies often discourage the attempt. Independent phytomedicine companies have been very strong in Germany. However, by the end of 1995, eleven (almost all) had been acquired by the multinational pharmaceutical firms, acknowledging the lay public’s growing demand for phytomedicines in the Western World. The business of dietary supplements in the Western World has expanded from the Health Store to the pharmacy. Alternative medicine includes plant based products. Appropriate measures to ensure the quality, safety and efficacy of these either already exists or are being answered by greater legislative control by such bodies as the Food and Drug Administration of the USA and the recently created European Agency for the Evaluation of Medicinal Products, based in London. In the USA, the Dietary Supplement and Health Education Act of 1994 recognised the class of phytotherapeutic agents derived from medicinal and aromatic plants. Furthermore, under public pressure, the US Congress set up an Office of Alternative Medicine and this office in 1994 assisted the filing of several Investigational New Drug (IND) applications, required for clinical trials of some Chinese herbal preparations. The significance of these applications was that each Chinese preparation involved several plants and yet was handled as a single IND. A demonstration of the contribution to efficacy, of each ingredient of each plant, was not required. This was a major step forward towards more sensible regulations in regard to phytomedicines. My thanks are due to the staff of Harwood Academic Publishers who have made this series possible and especially to the volume editors and their chapter contributors for the authoritative information. Roland Hardman

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CONTRIBUTORS

Teris A.van Beek Laboratory of Organic Chemistry Phytochemical Section Wageningen Agricultural University Dreijenplein 8 6703 HB Wageningen The Netherlands

Norbert Clemens Intersan GmbH Einsteinstrasse 30 76275 Ettlingen Germany Amy Corzine McAlpine Thorpe and Warrier Limited 50 Penywern Road London SW5 9SX UK

Ezio Bombardelli Indena SpA Scientific Department Viale Ortles 12 20139 Milano Italy Bernard Bulling Aachener Strasse 312 50933 Köln Germany

Aldo Cristoni Indena SpA Scientific Department Viale Ortles 12 20139 Milano Italy

Fabrizio F.Camponovo Research and Development Pharmaton SA 6903 Lugano Switzerland

Volker Dankers Intersan GmbH Einsteinstrasse 30 76275 Ettlingen Germany

Danielle Julie Carrier Department of Agricultural and Bioresource Engineering University of Saskatchewan 57 Campus Drive Saskatoon, Saskatchewan Canada S7N 5A9

Cynthia L.Darlington Department of Psychology and the Neuroscience Research Centre University of Otago Dunedin New Zealand

Yves Christen Institut IPSEN 24 rue Erlanger 75781 Paris Cedex 16 France

Francis V.DeFeudis Institute for Bioscience 153 West Main Street Westboro, MA 01581 USA xi

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

xii

CONTRIBUTORS

Peter Del Tredici Arnold Arboretum of Harvard University 125 Arborway Jamaica Plain, MA 02130–3519 USA Peter A.G.M.De Smet Scientific Institute Dutch Pharmacists Alexanderstraat 11 2514 JL Den Haag The Netherlands Martien C.J.M.van Dongen Department of Epidemiology Maastricht University P.O. Box 616 6200 MD Maastricht The Netherlands Katy Drieu Institut Henri Beaufour-IPSEN 24 rue Erlanger 75116 Paris France Frans M.van den Dungen Biohorma B.V. Medical Research P.O. Box 33 8081 HH Elburg The Netherlands Kazuhiko f*ckushima School of Agriculture Nagoya University Nagoya 464–01 Japan Marian A.J.T.van Gessel Hamsestraat 79 4043 LG Opheusden The Netherlands Peter Halama Berner Heerweg 175 22159 Hamburg Germany

Andreas Hasler Zeller AG Herbal Remedies 8590 Romanshorn Switzerland Robert ho*rr Dr. Willmar Schwabe Pharmaceuticals Clinical Research Department Willmar Schwabe Strasse 4 76227 Karlsruhe Germany Hoon Huh Laboratory of Pharmacognosy College of Pharmacy Seoul National University 56–1 Shinlim-Dong Kwanak-Gu Seoul 151–742 Korea Hermann Jaggy Dr. Willmar Schwabe Arzneimittel Postfach 410925 7500 Karlsruhe 41 Germany Paul Knipschild Department of General Practice Maastricht University P.O. Box 616 6200 MD Maastricht The Netherlands Dominique Laurain Laboratoire de Pharmacognosie Faculté de Pharmacie Université de Picardie Jules Verne 1 rue des Louvels 80037 Amiens Cedex 1 France Johannes J.Lichius Institut für Pharmazeutische Biologie Philipps-Universität Marburg Deutschhausstrasse 17A 35032 Marburg Germany

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CONTRIBUTORS

Karyn Maclennan Department of Pharmacology School of Medical Sciences University of Otago Medical School Dunedin New Zealand

Erik van Rossum Department of Epidemiology Maastricht University P.O. Box 616 6200 MD Maastricht The Netherlands

Beat Meier Zeller AG, Herbal Remedies 8590 Romanshorn Switzerland

Joerg Schulz Klinikum Buch Geriatrische Klinik Zepernicker Strasse 1 13125 Berlin Germany

Volker Melzheimer Botanischer Garten Philipps-Universität Marburg Karl von Frisch Strasse 35032 Marburg Germany Paolo Morazzoni Indena SpA Scientific Department Viale Ortles 12 20139 Milano Italy Koji Nakanishi Department of Chemistry Columbia University 3000 Broadway Mail Code 3114 New York, NY 10027 USA Joe O’Reilly Cara Partners Little Island, Co. Cork Ireland Federico Peterlongo Indena SpA Laboratori Ricerca e Sviluppo Via Don Minzoni 6 20090 Settala (Milano) Italy

xiii

Paul F.Smith Department of Pharmacology School of Medical Sciences University of Otago Medical School Dunedin New Zealand Fabio Soldati Research and Development Pharmaton SA 6903 Lugano Switzerland V.Srini Srinivasan United States Pharmacopoeia 12601 Twinbrook Parkway Rockville, MD 20852 USA Otto Sticher Department of Pharmacy Federal Institute of Technology (ETH) Winterthurerstrasse 190 8057 Zürich Switzerland Noritsugu Terashima 2–610 Uedayama, Tenpaku Nagoya 468 Japan

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

xiv

CONTRIBUTORS

Luisella Verotta Dipartimento di Chimica Organica e Industriale Università degli Studi di Milano Via Venezian 21 20133 Milano Italy Keiji Wada Faculty of Pharmaceutical Sciences Health Sciences University of Hokkaido Ishikari–Tobetsu Hokkaido 061–02 Japan

Gopi Warrier McAlpine Thorpe and Warrier Limited 50 Penywern Road London SW5 9SX UK

Herman J.Woerdenbag University Centre for Pharmacy Antonius Deusinglaan 2 9713 AW Groningen The Netherlands

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

1. THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA PETER DEL TREDICI Arnold Arboretum of Harvard University 125 Arborway Jamaica Plain, Massachusetts 02130–3519, USA

EVOLUTIONARY HISTORY Ginkgo biloba, which is not closely related to any other living plant, is generally classified in its own division, the Ginkgophyta. This taxon is distinguished from the Coniferophyta (conifers) on the basis of its reproductive structures, most notably its multiflagellated sperm cells, and from the Cycadophyta (cycads) on the basis of its vegetative anatomy (Wang and Chen, 1983; Gifford and Foster, 1987). Recent molecular analysis of the Ginkgo genome, while far from complete, suggests a much closer relationship to the cycads than to the conifers (Hasebe, 1997). The fossil record of the genus Ginkgo is extensive, with numerous reports of “Ginkgophyte” foliage and wood from many stratigraphic regions in both the northern and southern hemispheres. Because most of this material is sterile, however, its precise relationship to the extant species has always been conjectural. This uncertainty has recently been resolved by the discovery in Henan Province, China of fossils from the Middle Jurassic (180 million years ago) that possessed Ginkgo-like ovule-bearing organs (Zhou and Zhang, 1989). These are the earliest, unequivocal representatives of the genus Ginkgo and have been described as a new species, G.yimaensis, that differs from G.biloba in having more highly dissected leaves and much smaller ovules clustered on branched peduncles (Figure 1). Another extinct “species,” G.adiantoides, had an extensive distribution in the northern hemisphere from the lower Cretaceous through the Pliocene (141–1.8 mya), and many authors consider this taxon to be the likely ancestor of G.biloba because it had a similar leaf morphology and ovule structure (Tralau, 1968; Zhou, 1994). The genus Ginkgo appears to have reached the peak of its diversity during the lower Cretaceous (141–98 mya), with several distinct species occupying a more or less circumpolar distribution in the northern hemisphere which also extended into several parts of the southern hemisphere. During the upper Cretaceous (98–65 mya), the fossil record for Ginkgo shows a decline in diversity and distribution, particularly toward the end of the period when worldwide temperatures decreased dramatically. The diminution of Ginkgo’s range continued into the Tertiary, and was particularly striking from the Oligocene (38–26 mya), when the genus disappeared from polar 7 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

8

PETER DEL TREDICI

Figure 1 A reconstruction of Ginkgo yimaensis: left, leaves of various shapes and sizes believed to have been produced by a single long shoot; right, a portion of a long shoot with two dwarf shoots bearing both leaves and ovule-bearing organs. Bar at lower right equals three centimeters. Reprinted with permission from Zhou and Zhang, 1989.

areas, through the end of the Miocene (24–7 mya), when it disappeared from estern North America. These dramatic changes were most likely the result of the extensive cooling that occurred throughout the Northern hemisphere during these time periods. The genus Ginkgo was gone from Europe by the end of the Pliocene (1.8 mya) as temperatures dropped and the rainfall regimen gradually shifted from one of summerwet to one of summer-dry. The only known Pleistocene (1.8 mya to present) occurrences of the genus Ginkgo are from southwestern Japan. Based on their leaf characteristics, these fossils have been classified under the name of the extant species, G.biloba (Uemura, 1997). In general, the distribution of Ginkgo fossils indicates that the genus has been relatively consistent in its ecological tolerances since the Cretaceous, preferring to grow in warm temperate climates characterized by moist summers and

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

9

cool winters (Tralau, 1968; Uemura, 1997). As regards the morphology of its reproductive organs, the trend within the genus Ginkgo seems to be one of reduction, with the ovules decreasing in number while increasing in size, and the pedicels disappearing to leave sessile ovules connected directly to the peduncle (Zhou and Zhang, 1989; Zhou, 1991; 1994). ECOLOGY As a wild species, Ginkgo biloba is native to China and was probably a member of the mixed-mesophytic forest community that once covered the hill country bordering the Yangtze River valley. Most of this warm-temperate forest has now been cut down with the exception of a few remnants located in isolated valleys and on steep mountain slopes (Wang, 1961; Zheng, 1992a). One of Ginkgo’s last wild refugia is thought to be in Zhejiang Province, China, on the west peak of Tianmu Mountain (Xitianmu Shan: 1506 meters elevation; 119° 25' E; 30° 20' N; mean annual rainfall 1767 millimeters; mean January temperature -3.2°C; mean July temperature 20.5°C). There are also reports of “wild” Ginkgo populations in other parts of China, including Guangxi, Guizhou and Sichuan Provinces, but these claims have yet to be substantiated by careful field research (Liang, 1993). Botanists have long debated the “wildness” of the Ginkgo population growing on Tianmu Mountain. While the long history of human habitation in the area makes it difficult to determine whether or not the trees are truly wild, the exceptional speciesrichness of the surrounding forest and the large size of many of its trees suggests that they may well be (Del Tredici et al., 1992; Zheng, 1992b). Recent isozyme studies on the population, indicate a relatively low degree of genetic diversity among the plants, an observation that led the authors to speculate that the population may be descended from cultivated trees (Wu et al., 1992). In 1984, the Tianmu Ginkgo population consisted of approximately 244 individuals with a mean diameter at breast height (DBH) of 45 centimeters and a mean height of 18.4 meters. Most of the trees were growing on disturbed sites such as along stream beds, on rocky slopes, and on the edges of exposed cliffs (Figure 2). Ginkgo seedlings were quite rare on Tianmu Mountain and are typically found in areas of the forest that have been opened up by disturbance, an observation that provides support for the idea that Ginkgo acts as a “pioneer” species in its native habitat (Del Tredici et al., 1992). Approximately 40% of the Tianmu Ginkgos possessed more than one trunk greater than 10 centimeters diameter at breast height (DBH). Most of these secondary trunks originated from lignotubers located at, or just below, ground level (Figure 3). Secondary trunk formation was most apparent in specimens that were damaged or under severe stress (Del Tredici et al., 1992). The sprouting ability of Ginkgo is an important factor that has enabled the species to persist on the mountain’s badly eroded slopes, and may well have played a role in the survival and morphological stability of the genus since the Tertiary (Figure 4).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

10

PETER DEL TREDICI

Figure 2 The famous “living fossil” Ginkgo tree growing on Tianmu Mountain, Zhejiang Province, China, photographed in 1989. This ancient, ovulate specimen occupies an area of approximately 20 square meters and consists of 15 stems greater than 10 centimeters diameter. The largest trunk has a diameter of 110 centimeters. The Chinese describe this tree, perched on the edge of a steep cliff at 950 meters elevation, an “an old dragon trying to fly.” The fence protecting the tree was built in 1980. Reprinted from Del Tredici et al., 1992.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

11

Figure 3 The lignotuber-developed shoot system of an old Ginkgo growing on top of a stone wall on Tianmu Mountain in Zhejiang Province, China. At least three generations of stems can be seen: the oldest represented by the cut trunks A, B, and C (with diameters of 55, 40 and 37 centimeters respectively); the second by the living trunks A, and B, (with diameters of 26 and 20cm); and the third by suckers arising from the distal portions of the lignotuber (stippled). Reprinted from Del Tredici et al., 1992.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

12

PETER DEL TREDICI

Figure 4 A schematic representation of the life cycle of Ginkgo biloba on Tianmu Mountain.

Sexual Reproduction Ginkgo is a dioecious species, with separate male and female individuals occurring at a roughly 1:1 ratio, although occasional monoecious individuals are reported to occur (Santamour et al., 1983b). Ginkgo shows a long juvenile period, typically not reaching sexual maturity until 20 to 30 years of age. Male and female sex organs are produced on short shoots, in the axils of bud scales and leaves. The male catkins emerge before the leaves and fall off immediately after shedding their pollen. Wind pollination occurs anywhere from early April in areas with mild winters to late May in areas with severe winters. As to pollination distance, it is difficult to say what the maximum is, but in the Boston area 400 meters between male and female trees does not inhibit seed set (Del Tredici, 1989). Ginkgo ovules are 2 to 3mm long and produced in pairs at the ends of stalks 1 to 1.5 centimeters long. When the ovule is receptive, it secretes a small droplet of mucilaginous fluid from its micropyle which functions to capture airborne pollen. Retraction of this droplet at the end of the day brings the pollen into the pollen chamber. Once inside the ovule, the male gametophyte commences a four-month long development period that culminates with the production of a pair of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

13

multiflagellated spermatozoids, one of which fertilizes a waiting egg cell (Friedman, 1987) while the ovules are still on the tree (Holt and Rothwell, 1997). Depending on the date of pollination, this union can occur anytime between late August to late September. The mature seed of Ginkgo is relatively large (20–30mm×16–24mm) and consists of an embryo embedded in the tissue of the female gametophyte surrounded by a thick seed coat. This seed coat consists of a soft, fleshy outer layer (the sarcotesta), a hard, stony middle layer, and a thin, membranous inner layer. The seed, devoid of the fleshy sarcotesta, is generally referred to as the Ginkgo “nut,” with dimensions of 19–30mm×11–14mm. The developing ovules are green until they mature in the autumn when, in response to cold temperatures, they turn the same yellow color that the leaves do. Typically they fall from the tree about a month after fertilization. The foul odor associated with Ginkgo seeds develops only when they are fully mature, and is the result of the presence of two volatile compounds, butanoic and hexanoic acids, localized in the sarcotesta (Parliment, 1995), also contains phenolic compounds known to cause contact dermatitis in humans (Kochibe, 1997). Seed Dispersal and Establishment The paucity of fossilized Ginkgo seeds has not deterred speculation as to what animals might have dispersed Ginkgo seeds over the course of its long evolution. Several authors have proposed that dinosaurs might have dispersed Ginkgo seeds, but none of them have provided any anatomical evidence that would support the claim or specified what type of dinosaur it might have been (Janzen and Martin, 1982; van der Pijl, 1982; Tiffney, 1984; Rothwell and Holt, 1997). A second proposal is that early mammals in the extinct family Mutituberculata could have been effective dispersal agents (Del Tredici, 1989). These marsupial-like creatures, often referred to as the “rodents of the Mesozoic,” were widespread in the temperate parts of the northern hemisphere from the Late Jurassic through the Oligocene. Many multituberculates possessed teeth that were adapted to cracking hard food objects, such as seeds, and a skeletal structure that was adapted to living in trees (Krause and Jenkins, 1983). In modern times, a number of different mammals have been observed feeding on, and presumably dispersing, the odoriferous, nutrient-rich seeds of G.biloba. In the order Rodentia, these include the red-bellied squirrel (Callosciurus flavimanus var. ningpoensis, family Sciuridae) on Tianmu Mountain (Del Tredici et al., 1992), and the gray squirrel (Sciurus carolinensis, family Sciuridae) in eastern North America (Del Tredici, 1989). In the order Carnivora, potential dispersal agents include the masked palm civet (Paguma larvata, family Viveridae) on Tianmu Mountain (Del Tredici et al., 1992), the leopard cat (Felis bengalensis, family Felidae) in Hubei Province, China (Jiang et al., 1990), and the raccoon dog (Nyctereutes procyonoides, family Canidae) in Japan (Hori, 1996). The existence of three independent reports of carnivores consuming whole Ginkgo seeds and defecating intact nuts, raises the possibility that the foul smelling sarcotesta may be attracting these animals by

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

14

PETER DEL TREDICI

mimicking the smell of rotting flesh (Del Tredici et al., 1992). Projecting this line of speculation back into evolutionary time, it seems likely that if dinosaurs were involved in the dispersal of Ginkgo seeds, then it was probably done by carrion feeding scavengers, with teeth adapted to tearing flesh, rather than by herbivores with dentition adapted to grinding vegetation (Del Tredici, 1989). Ginkgo seeds are dormant when they fall from the tree because the embryo is not fully developed, being only about 4 to 5 millimeters in length. If seeds are collected shortly after dispersal, cleaned, and placed in a warm greenhouse, the embryo will grow to its full size—10 to 12 millimeters in length—and germinate within eight to ten weeks (Li and Chen, 1934; Holt and Rothwell, 1997). This type of germination behaviour has important implications in terms of seedling establishment in the field. In warm-temperate climates Ginkgo seeds are shed in late summer or early fall, and the embryo is able to make considerable growth during the mild weather that follows. In cold-temperate climates, on the other hand, seeds are shed later in the season and the colder autumn temperatures delay full embryo development until the following spring. This differential timing of embryo maturation means that seeds in warm climates will be ready to germinate during the favourable conditions of mid-to late spring (April through early June), while those in cold climates will not germinate until later in the summer (late June through early August), when conditions for establishment are much less favorable (Del Tredici, 1991a). These phenological differences suggest that the innate germination requirements of Ginkgo seed may account for the species’ warm-temperate distribution in recent times, as well as the warm-temperate distribution of its fossil ancestors. Tree Architecture Ginkgo is an extremely long-lived deciduous tree that, in cultivation in China, is capable or reaching ages in excess of a thousand years, with stem diameters between one and four meters (Lin, 1995). Depending on the growing conditions, mature trees typically reach heights of between 20 to 40 meters, although one exceptional specimen in Korea has been reported with a height of 64 meters and a girth of 14 meters (Del Tredici, 1991b). The form of vigorous young Ginkgos is distinctly pyramidal, with a dominant central leader and widely spaced whorls of lateral branches that grow out at a diagonal orientation to the trunk. With the onset of sexual maturity, at around 25 years of age, height growth generally slows down and the tree fills in its sparsely branched juvenile structure, forming a broad, spreading crown (Gunkle et al., 1949; Del Tredici et al., 1991c). Ginkgo trees produce two types of shoots: long shoots with widely spaced leaves that subtend axillary buds; and short, or spur, shoots with clustered leaves that lack both internodes and axillary buds. Long shoots are responsible for building up the basic framework of the tree and generating new growing points, while the short shoots produce the majority of leaves and all of the reproductive structures. In response to a variety of environmental and physiological factors, the growth pattern of these shoots can be reversed, such that a short shoot can proliferate into a long shoot, and

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

15

the terminal growth of a long shoot can fail to elongate and become a short shoot (Gunkle et al., 1949). This flexibility provides Ginkgo with a simple mechanism for modulating carbohydrate allocation between sexual reproduction and vegetative growth (Del Tredici, 1991c). Under stressful growing conditions, Ginkgo is capable of producing secondary trunks at or just below ground level. Typically, these secondary stems originate from root-like, positively geotropic shoots known as lignotubers or “basal chichi.” Anatomically lignotubers develop in all Ginkgo seedlings as part of their normal ontogeny from buds located in the axils of the two cotyledons (Figure 5). When stimulated by some traumatic event, one of these cotyledonary buds often grows out from the trunk to form a woody, positively-geotropic lignotuber which has the capability of producing both aerial shoots and adventitious roots (Del Tredici, 1992; 1997). Similar structures, known as “aerial chichi,” often develop along the trunk and branches of very old Ginkgo trees in response to traumatic injury or environmental stress. These stalactite-like growths originate from embedded axillary buds and can produce both roots and shoots when they come in contact the soil (Del Tredici, 1992). CULTIVATION AND UTILIZATION While Ginkgo has been cultivated in China for several thousand years, the earliest written references to the plant that have been specifically cited in the literature date back only to Song dynasty documents of the early eleventh century (Li, 1956). According to these ancient texts, the tree came from an area south of the Yangtze River, in what is now the Ningguo District of southern Anhui Province. Most of these early writers praise the Ginkgo for the unique beauty of its leaves and for its edible, and medicinally active, nuts. Contrary to numerous reports in the literature, it is these attributes, rather than any special religious significance, that probably account for the tree’s rapid spread in cultivation throughout China and into Korea (Li, 1956). It should be noted, however, that the oldest Ginkgos in China are generally found growing near Daoist and Buddhist temples, and that these ancient specimens have played an important role in the preservation and dissemination of the species (Figure 6). Ginkgo was introduced into western Japan from eastern China about eight hundred years ago. As is the case in China, the largest trees are growing in the vicinity of Buddhist or Daoist temples and shrines (Tsumura et al., 1992). From Japan, Ginkgo was introduced into Europe at the Botanic Garden in Utrecht, Netherlands, about 1730, and into Kew Gardens, near London, England, around 1754. From England the tree was imported into North America in 1784 at Philadelphia, Pennsylvania (Del Tredici, 1991b). Ornamental Uses Ginkgo is cultivated throughout the temperate zones of the world for ornamental purposes. This includes areas with a Mediterranean-type climate as well as those

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

16

PETER DEL TREDICI

Figure 5 The early stages of lignotuber formation in two- to three-year-old Ginkgo seedlings. A. A partially debarked two year old seedling show the unequal development of the cotyledonary buds. Scale in millimeters. B. Longitudinal section of a partially debarked two-year-old seedling showing the strongly kinked stem that is often associated with lignotuber formation. Scale in millimeters. C. A three-year-old seedling in which one of the cotyledonary buds has formed a prominent lignotuber while the other has not (arrow). Bar equals one centimeter. D. A debarked three-year-old seedling showing the xylem traces of the numerous dormant shoot buds on a well-developed lignotuber. Scale in millimeters. Reprinted from Del Tredici, 1992.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

17

Figure 6 A two thousand year-old Ginkgo growing on the grounds of an ancient temple in Xin Cun Village, Tan Chen County, Shandong Province, China. In 1985, the tree was 37 meters tall, with a trunk DBH of 2.3 meters. It is a staminate tree, with a large ovulate branch grafted onto it.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

18

PETER DEL TREDICI

with a cold-temperate climate where minimum winter temperatures can reach -30°C. While Ginkgo grows best when planted in full sun, it shows the ability to persist indefinitely under low light and low nutrient conditions, such as when planted along the streets of densely populated cities (Handa et al., 1997). In the eastern United States, Ginkgo grows rapidly within USD A hardiness zones 5 to 8, a region characterized by abundant, year-round moisture, high temperatures during the growing season, cold winters, and acid to neutral soils. Ginkgo does not perform particularly well in subtropical climates or on soils that are overly wet or dry during the growing season (Santamour et al., 1983b; Del Tredici, 1991b; Liang, 1993). Relative to other commonly cultivated trees, Ginkgo possesses a high degree of resistance to insect damage and to fungal, viral, and bacterial diseases, as well as to ozone and sulfur dioxide pollution, making it an excellent choice for planting in urban areas (Sinclair et al., 1987; Honda, 1997). Numerous selections of Ginkgo have been made for ornamental purposes during its long history. Older horticultural forms that are still cultivated include: “Epiphylla” an ovulate tree which produces seeds attached to the leaves; “Fastigiata” with a narrow, upright growth habit; “Pendula” with a spreading growth habit composed entirely of horizontal branches; and “Variegata”, a generally unstable form with leaves striped with yellow or white (Santamour et al., 1983a). These, as well as other more recent cultivars can be propagated from rooted soft-wood cuttings in summer, hardwood cuttings in winter, twig grafting in late winter, or bud grafting in late summer (Del Tredici, 1991b; Liang, 1993). Because of complex epigenetic effects, however, both rooted cuttings and grafted plants generally fail to reproduce the upright form of the seed-grown parent they were propagated from. Instead, they tend to produce low-branched, vase-shaped trees that lack a dominant central leader (Figure 7) (Del Tredici, 1991b, 1991c). At the present time, fastigiate male clones are the only vegetatively propagated selections that are widely available from commercial nurseries in Europe and North America. Seed Production Each Ginkgo seed contains a single, thin-shelled “nut,” that is traditionally consumed as both a food and a medicine throughout Asia. Dry nuts, which constitute 59% of the fresh nut weight, have a composition of roughly 6% sucrose, 68% starch, 13% protein, and 3% fat (Duke, 1989). The commercial production of Ginkgo nuts has been established in China for over 600 years, and at least 44 cultivars have been selected in that country based on the size and shape of their nuts, and on their productivity (Santamour et al., 1983a; Liang, 1993; He et al., 1997). These cultivars are usually propagated through grafting on seedling rootstocks, and they begin producing nuts at about 5 years of age (Figure 7). For the widely grown cultivar “Dafushon,” annual yields in Jiangsu Province, China vary between 5 and 10 kilograms of nuts for 15-year-old trees and between 50 and 100 kilograms for 50year-old trees. In general, Ginkgo produces a heavy crop of seeds every other year, with relatively light crops in alternate years (Del Tredici, 1991b; Liang, 1993). While precise figures are difficult to come by, one recent estimate of the Chinese crop suggests

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

19

Figure 7 A grove of grafted Ginkgo trees cultivated for their edible nuts on Dongting Shan, Jiangsu Province, China. Note the spreading, vase-shaped form of the tree. Reprinted from Del Tredici, 1991b.

that 700,000 to 800,000 trees produce an average of between six to seven thousand tons of dried nuts per year (He et al., 1997). Raw Ginkgo nuts, cleaned of their fleshy pulp, have long been used in traditional Chinese medicine to treat a variety of lung-related ailments such as asthma and bronchitis, as well as for the treatment of kidney and bladder disorders. They also display antibiotic effects against a variety of bacterial pathogens (Perry, 1980; Bensky and Gamble, 1986). Raw Ginkgo nuts contain a toxin that has been identified as 4O-methylpyridoxine (MPN), whose primary mode of action is to antagonize the activity of vitamin B6 (Wada and Haga, 1997). Typically, children are much more susceptible to poisoning from Ginkgo nuts than adults. For culinary purposes, Ginkgo seeds must be cooked to eliminate the potential for poisoning. They are usually steamed until the hard shell cracks open, allowing the kernel to be removed. In China, these kernels are either boiled in sugar water to make a sweet soup or roasted and eaten plain. Because of their potential toxicity, however, people are generally advised not to eat too many Ginkgo nuts at one sitting. Leaf Production Ginkgo leaves are known to contain a wide variety of medicinally active chemicals, most notably terpenoids (ginkgolides and bilobalide) and flavonoids (glycosides of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

20

PETER DEL TREDICI

kaempferol, quercetin, isorhamnetin, etc.) (Boralle et al., 1988; DeFeudis, 1998). The use of Ginkgo leaf extract for pharmaceutical purposes was originally developed in Germany in 1965, and the first commercially available Ginkgo leaf extract was registered for human use in 1974 in France, under the code-name “EGb 761” (DeFeudis, 1998). This extract is made from dried Ginkgo leaves and has a standardized content of 22–27% flavonol glycosides and 5–7% terpene trilactones. This extract is taken internally for the treatment of cerebral and peripheral vascular diseases, as well as to alleviate some of the ailments associated with ageing, including dizziness, ringing in the ears, and short-term memory deterioration. So far as is known, the extract has only minimal side effects, even after prolonged use (DeFeudis, 1998; Juretzek, 1997). Contrary to numerous reports in the literature, Ginkgo leaf preparations have played a very minor role in the practice of traditional Chinese medicine, and have only recently been listed in official Materia Medica. Since 1982, Ginkgo leaves have been cultivated on a large scale in both France (480 hectares) and the United States (460 hectares) specifically for the production of the EGb 761 extract. When grown for this purpose, seedlings are spaced 40 centimeters apart in rows 1 meters apart, producing a density of approximately 25,000 plants per hectare (Figure 8). Special pruning techniques are used to keep the trees below 3 meters tall, thereby allowing the use of mechanical harvesting equipment. Green leaves are generally harvested in mid- to late summer, after which the plants are cutback to near ground level once every four or five years. The new growth following

Figure 8 The Ginkgo plantation in Sumter, South Carolina, in early spring. For scale, the individual segments of the irrigation system are about 45 meters long. Reprinted from Del Tredici, 1991b.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

21

such low cutbacks originates from buds located at the base of the stem or on basal lignotubers. When provided with an adequate supply of moisture and fertilizer during the growing season, the annual height growth from the point of cutting is typically a meter or more (Del Tredici, 1991b). In China, yellow leaves are typically harvested at the same time that the nuts are collected, in early autumn, and sold for the manufacture of leaf extracts. These older leaves are not as rich in ginkgolides and other medicinally active compounds as green leaves, and consequently command a lower price in the international market. Since 1992, over 2000 hectares of Ginkgo seedlings have been planted in eastern China specifically for the production of green leaves. The first of these plantations began producing marketable quantities of leaves in 1996, and projections are for yields to increase dramatically over the next five to ten years (Balz, 1997; He et al., 1997). The use of Ginkgo leaf extract for therapeutic as well as other purposes has grown dramatically since its introduction in 1974. While accurate figures are difficult to come by, recent estimates of the world wide sales of Ginkgo leaf products suggest that it is around half a billion US dollars (see Warrier and Corzine, this volume), a truly remarkable figure given that the product has only been on the market for 25 years. REFERENCES Balz, J.-P. (1997) Agronomic aspects of Ginkgo biloba leaves production. In Proceedings of ’97 International Seminar on Ginkgo, Nov. 10–12, 1997, Beijing, China, pp. 101–104. Bensky, D. and Gamble, A. (1986) Chinese Herbal Medicine: Materia Medica. Eastland Press, Seattle, Washington (translated from the Chinese). Boralle, N., Braquet, P., Gottlieb, O.R. (1988) Ginkgo biloba: a review of its chemical composition. In P.Braquet (ed.) Ginkgolides-Chemistry, Biology, Pharmacology and Clinical Perspectives, Vol. 1, J.R.Prous, Barcelona, pp. 9–25. DeFeudis, F.V. (1998) Ginkgo biloba Extract (EGb 761): From Chemistry to Clinic, Ullstein Medical, Weisbaden. Del Tredici, P. (1989) Ginkgos and multituberculates: evolutionary interactions in the Tertiary. Biosystems, 22, 327–339. Del Tredici, P. (1991a) Evolution and Natural History of Ginkgo biloba. PhD thesis, Boston Univ. Del Tredici, P. (1991b) Ginkgos and people: a thousand years of interaction. Arnoldia, 51, 2– 15. Del Tredici, P. (1991c) The architecture of Ginkgo biloba L. In C.Edelin (ed.), L’Arbre, Biologie et Developpement. Naturalia Monspeliansia n°h.s., pp. 155–168. Del Tredici, P. (1992) Natural regeneration of Ginkgo biloba from downward growing cotyledonary buds (basal chichi). Am. J. Bot., 79, 522–530. Del Tredici, P. (1997) Lignotuber formation in Ginkgo biloba. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 119–126. Del Tredici, P., Ling, H., Yang, G. (1992) The Ginkgos of Tian Mu Shan. Conserv. Biol., 6, 202–209. Duke, J.A. (1989) CRC Handbook of Nuts, CRC Press, Boca Raton, Florida. Friedman, W.E. (1987) Growth and development of the male gametophyte of Ginkgo biloba within the ovule (in vitro). Am. J. Bot., 74, 1797–1815. Gifford, E.M. and Foster, A.S. (1987) Morphology and Evolution of Vascular Plants, W.H.Freeman and Company, New York.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

22

PETER DEL TREDICI

Gunkle, J.E., Thimann, K.V., and Wetmore, R.H. (1949) Studies of development in long shoots and short shoots of Ginkgo biloba L., part IV. Growth habit, shoot expression and the mechanism of its control. Am. J. Bot., 36, 309–316. Handa, M., Iizuka, Y., and Fujiwara, N. (1997) Ginkgo landscapes. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 259–283. Hasebe, M. 1997. Moleculary phylogeny of Ginkgo biloba: close relationship between Ginkgo biloba and cycads. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 173–182. He, S.-A., Yin, G., and Pang, Z.-J. (1997) Resources and prospects of Ginkgo biloba in China. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 373–383. Holt, B.F. and Rothwell, G.W. (1997) Is Ginkgo biloba (Ginkgoaceae) really an oviparous plant? Am. J. Bot., 84, 870–872. Honda, H. (1997) Ginkgos and insects. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. SpringerVerlag, Tokyo, pp. 243–250. Hori, T. (1996) Ginkgo to the Japanese people. Microscopia, 13, 184–185 (in Japanese). Janzen, D.H. and Martin, P.S. (1982) Neotropical anachronisms: the fruits the gomphotheres ate. Science, 215, 19–27. Jiang, M., Jin, Y. and Zhang, Q. (1990) Preliminary study on Ginkgo biloba in Dahongshan region, Hubei. J. Wuhan Bot. Res., 8, 191–193 (in Chinese). Juretzek, W. (1997) Recent advances in Ginkgo biloba extract (Egb 761). In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 341–358. Kochibe, N. (1997) Allergic substances of Ginkgo biloba. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 301–307. Krause, D.W. and Jenkins, F.A. (1983) The postcranial skeleton of North American multituberculates. Bull. Mus. Comp. Zool., 150, 199–246. Li, H.L. (1956) A horticultural and botanical history of Ginkgo. Bull. Morris Arb., 7, 3–12. Li, T.T. and Chen, S.M. (1934) Temperature and the development of the Ginkgo embryo. Sci. Rep. Nat. Tsing Hua Univ., ser. B, 2, 37–39. Liang, L. (1993) The Contemporary Ginkgo Encyclopedia of China, Beijing Agric. Univ. Press (in Chinese). Lin, J.-X. (1995) Old Ginkgo trees in China. International Dendrological Society Yearbook, 1995, pp. 32–37. Parliment, T. (1995) Characterization of the putrid aroma compounds of Ginkgo biloba fruits. In R.Rouseff and M.Leahy (eds.) Fruit Flavors: Biogenesis, Characterization, and Authentication, Am. Chem. Soc. Symp. Ser., 596, pp. 276–279. Perry, L.M. (1980) Medicinal Plant of East and Southeast Asia: Attributed Properties and Uses, MIT Press, Cambridge, Massachusetts. van der Pijl, L. (1982) Principles of Dispersal in Higher Plants, 3rd ed., Springer-Verlag, Berlin. Rothwell, G.W. and Holt, B. (1997) Fossils and phenology in the evolution of Ginkgo biloba. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 223–230. Santamour, F.S., He, S.A., McArdle, A.J. (1983a) Checklist of cultivated Ginkgo. J. Arboriculture 9:88–92. Santamour, F.S., He, S.A., Ewert, T.E. (1983b) Growth, survival and sex expression in Ginkgo. J. Arboriculture 9:170–171. Sinclair, W.A., Lyon, H.H., Johnson, W.T. (1987) Diseases of Trees and Shrubs, Comstock Publishing Associates, Ithaca. Tiffney, B.H. (1984) Seed size, dispersal syndrome and the rise of the angiosperms: evidence and hypotheses. Ann. Missouri Bot. Gard. 71:551–576.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

THE EVOLUTION, ECOLOGY, AND CULTIVATION OF GINKGO BILOBA

23

Tralau, H. (1968) Evolutionary trends in the genus Ginkgo. Lethaia 1:63–101. Tsumura, Y., Motoike, H., Ohba, K. (1992) Allozyme variation of old Ginkgo biloba memorial trees in western Japan. Can. J. For. Res. 22:939–944. Uemura, K. (1997) Cenozoic history of East Asia. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 207–221. Wada, K. and Haga, M. (1997) Food poisoning by Ginkgo biloba seeds. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Tremouillaux-Guiller, and H.Tobe (eds.), Ginkgo biloba—A Global Treasure. Springer-Verlag, Tokyo, pp. 309–321. Wang, C.W. (1961) The Forests of China, Maria Moors Cabot Found., publ. 5. Harvard Univ., Cambridge, Mass. Wang, F.H. and Chen, Z.K. (1983) A contribution to the embryology of Ginkgo with a discussion of the affinity of the Ginkgoales. Acta Botanica Sinica, 25, 199–211 (in Chinese). Wu, J., Cheng, P., and Tang, S. (1992) Isozyme analysis of the genetic variation of Ginkgo biloba L. population in Tian Mu Mountain. J. Plant Resources Environment, 1, 20–23 (in Chinese). Zheng, C.Z. (1992a) A preliminary analysis of flora in Tianmu Mountain Reserve. In F.Yang (ed.) Comprehensive Investigation Report on Natural Resource of Tianmu Mountain Nature Reserve, Science and Technology Press, Hangzhou, pp. 89–93 (in Chinese). Zheng, C.Z. (1992b) A catalogue of seed-plants in Tianmu Nature Reserve. In F.Yang (ed.) Comprehensive Investigation Report on Natural Resource of Tianmu Mountain Nature Reserve, Science and Technology Press, Hangzhou, pp. 94–128 (in Chinese). Zhou, Z. (1991) Phylogeny and evolutionary trends of Mesozoic ginkgoaleans—a preliminary assessment. Rev. Palaeobot. Palynol., 68, 203–216. Zhou, Z. (1994) Heterochronic origin of Ginkgo biloba-type ovule organs. Acta Palaeontologica Sinica, 33, 131–139 (in Chinese). Zhou, Z. and Zhang, B. (1989) A middle-Jurassic Ginkgo with ovule-bearing organs from Henan, China. Palaeontographica, B 211, 113–133.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

2. GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY VOLKER MELZHEIMER1 and JOHANNES J.LICHIUS2 1

Botanischer Garten, Philipps-Universität Marburg, Karl-von-Frisch-Str., D-35032 Marburg, Germany 2 Institut für Pharmazeutische Biologie, Philipps-Universität Marburg, Deutschhausstr. 17 A, D-35032 Marburg, Germany

NOMENCLATURE AND SYSTEMATICS The first definitive description of the genus and species of Ginkgo biloba L. was given by Linnaeus (1771). For this reason, other names like Salisburia adiantifolia Smith and Pterophyllus salisburiensis Nelson are either invalid or are regarded as synonyms according to the rules of nomenclature with particular reference to the rule of priority (Smith, 1797; Nelson, 1866). The spelling of proper names is also determined by these rules. Consequently, Mayr’s correction of the spelling into Ginkgyo in 1906, which was justified in some respects, could not be accepted (Mayr, 1906). A German name is actually not in use. The English term “Maidenhair tree” is based on the similarity to the foliage of the “Maidenhair fern” (Adiantum). In Japan it is called “ginkyo”, whereas the French names are “l’arbre aux quarante écus” and “noyer du Japon”. In older systems the genus Ginkgo was included in the family of Taxaceae. Originally this family included Podocarpaceae and Cephalotaxaceae and has always been considered very artificial. The inclusion of the genus Ginkgo in this family mainly resulted from ignorance. Hirase’s discovery in 1895 that Ginkgo possesses multiciliated spermatozoids was the basis of Engler’s classification of a particular family and class of Ginkgoopsida (Hirase, 1895; Pilger, 1926). This class, Ginkgo and Ginkgo-like precursors (Vozenin-Serra et al., 1991), can be traced back to the Lower Permian (see Fig. 1) and is likely to extend into the Upper Devonian. According to Tralau (1967), the oldest reliable proof of the genus Ginkgo reaches back as far as the Lower Jurassic (see Fig. 2). With regard to the number of Ginkgo species, it is much smaller than has so far been believed because many fossil finds, each one of which has been regarded as a different species, can now be included in 2–3 different Ginkgo species (Tralau, 1967). Thus it is likely that the more highly developed representatives [progressive in that the ovules become sessile and reduced in number but increased in size (Zhou, 1991)] from the lower Cretaceous can be considered as the direct precursors of the recent Ginkgo biloba (also see Sp*rne, 1965). As far back as the Upper Palaeozoic there are a few primitive plants whose characteristics seem to align 25 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

26

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 1 Possible phylogenetic connections between related groups of seed plants and their development during different geological eras (numbers at the beginning of geological formations stand for millions of years). Unreliable relations not documented by fossil finds are shown by broken lines or left white. (Strasburger, 1998) P=Pentoxylidae, E=Ephedridae, G=Gnetidae, W=Welwitschiidae

them with this family. During the Mesozoic, especially during the Jurassic and the early part of the Cretaceous, Ginkgo attained its greatest prominence (see Fig. 2). This was not only true of the number of different species and genera occurring, but also of their nearly world-wide distribution and high number of habitats. Reaching its climax in the Upper Jurassic and Lower Cretaceous the group declined rather rapidly before the end of the Cretaceous. By the Oligocene all but two of the 19 genera with nearly 60 species of the family became extinct. From then on only a few species have persisted albeit in gradually lessening numbers and in gradually narrowing geographic range: Today they have disappeared from all but one continent, in which only one single species has survived (Dorf, 1958) (see Fig. 3). Whether or not this decline is due to natural or culture-related causes remains difficult to decide today (Del Tredici et al., 1992). Thus Ginkgo biloba is the prime botanical example of a “living fossil”, an expression first used by Darwin for the King crab (Limulus sp.), an animal alive today which is very similar to the fossil species Mesolimulus sulcatus. The currently used systematic classification into division and subdivision is as follows (Strasburger, 1998):

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

27

Figure 2 Chart of the geologic ranges of valid genera of the Ginkgo family, also showing the relative abundance of species within each genus, and the composite range and relative importance of the entire family in each successive geologic epoch (Dorf, 1958).

Figure 3 “Natural” occurrence of Ginkgo biloba. There are relatively dense growths of Ginkgo of different ages. The more important occurrence is in the provinces of Anhwei and Chekiang, here mainly in the Tian-Mu-Shan Mountains (according to Li, (1982), slightly modified).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

28

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

17. division: Spermatophyta 1. subdivision: Coniferophytina 1. class: Ginkgoopsida 2. class: Pinopsida (with three subclass) 2. subdivision: Cycadophytina 1. class Lyginopteridopsida (seed-ferns) 2. class Cycadopsida (Cycadaceae a.o.) 3. class Bennettitopsida (only known in fossil form) 4. class Gnetopsida (Ephedra, Welwitschia, Gnetum) 3. subdivision: Magnoliophytina (Angiospermae)

SHORT DESCRIPTION Tree, height up to 40m, multiform habit, slim and conical or spread out, bark grey, deeply furrowed on old trunks, alternate leaves on long shoots, fan-shaped, tough and leathery, 5–8cm wide with long petiole, on the wider side often incised or lobed, parallel and forked nervation, colour: a fresh green, golden yellow in autumn before being shed, dioecious, seeds resembling small yellow plums, yellowish-green, about 2.5cm long, fleshy outside, disagreeable smell, 2–3-edged stones, indigenous to China, frequently grown in Japan as temple tree, today also as roadside tree in Japan, Europe and America. Compared with other groups of plants, however, the Ginkgo has some peculiarities requiring a more detailed explanation. VEGETATIVE ORGANS The Ginkgo grows into a stately tree with a height of 30–40m and a girth of 3–4m; the bark is scabby, dark grey (for details of bark morphology see Kim et al., 1992); at first of pyramidal growth, it later forms an expanding crown. The trees can reach an age of several hundred years. In Europe, the oldest are about 245 years old, whereas in East Asia there are said to be some specimens of 1000 years with a girth of up to 20 m measured 1 m above the ground (see Fig. 4). The famous Ginkgo of Sendai in Japan is allegedly 1200 years old. Its branches cover the enormous area of 250m2; it owes its fame to the voluminous “chichi” growing on the underside of the big branches (see Fig. 5). This word means “mother’s breast”, which is why Japanese women consider trees with such phenomena to be symbols of numerous offspring and good breast-feeding ability (Schmid, 1994). Anatomically these are strictly positive geotropic sprouting growths (origin and possible early development see Del Tredici, 1992), developing roots upon reaching the ground. The growth rings of such stalactite-like branches are usually narrower than normal ones and there are sometimes zones of irregularly oriented tracheids (Li and Lin, 1991). Subsequently, new lateral branches can develop on this “trunk”. Similar phenomena are also known in different types of the genus Ficus (gum tree) where they help the trunk to expand, thus increasing the stability of the plant.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

29

Figure 4 Giant Ginkgo in the Zensho-ji Temple District, Fuchizawa (Japan) (Kato, 1994).

Branching is achieved by short and long shoots (see Fig. 6). On the long shoots leaves are placed relatively far from each other in 2/5 or 3/8 spirals (also see Gunckel and Wetmore, 1946). The end bud, more often the origin of another long shoot rather than a short shoot, consists of about 15 scales, the exterior ones are small, broad, and almost entirely suberized and more or less densely pilose on the edge. There is a decreasing suberization of the interior scales towards their tips, so that the interior scales are green and not suberized; the innermost ones can have rudimentary blades as appendages. On the axial buds, normally the starting points of short shoots and rarely of long shoots, several pairs of decussated scales can be found. When the long shoots of old trees stop growing, there can be a spontaneous development of short shoots into long shoots. These short shoots can have leaves for many years up to several decades; their annual increase is low, however, and at an advanced age they appear warty on account of the densely placed bases of the shed leaves. According to

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

30

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 5 Branch with “Chichi” in the initial stage on the Ginkgo in ji*zo-an near Itano (Kato, 1994).

earlier investigations the Ginkgo has no separation zone (which already exists in the family of Pinaceae), and the separation of the leaf stalk takes place in a layer of round cells between the stalk and the bark and is facilitated by segmentation at the base of the leaf. The leaves of the Ginkgo are summer green, i.e. they are active only during one vegetation period. The form of the leaves varies considerably (see Fig. 7). The unlobed or bilobed early leaves of short shoots are preformed in the autumn bud, and their nearly synchronous-expansion in the spring is not accompanied by stem elongation. The others are multilobed late leaves which develop at intervals of several days and their production can go on possibly throughout the summer. According to Critchfield (1970), the developmental events strongly correlated with the pattern of auxin production or distinct concentration in short and long shoots respectively. The stalk is thin, about 4–9cm long, more or less pilose, furrowed on the upper side and only a little broader at the base; near the base of the blade there is a short inconspicuous widening of the stalk, or else it becomes distinctly wedge-shaped. The blade itself is fan-shaped and approximately semi-circular, the edge has several irregular lobes and a narrow more or less deep incision opposite the stalk; or the blade is completely undivided and only the edge is retuse. The young leaves, on the other hand, have several deep incisions. After sprouting, the leaves are sparsely hairy. The width of the

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

31

Figure 6 Ginkgo biloba L.: A branch with male blossoms on a short shoot; B-D staminodium; C with open sporangia; D rear view; E branch with female blossom on short shoot; F female blossom; G peduncle with developed seed-bud—fractions indicate magnification or reduction (Pilger, 1926).

blade is 7–10 (max. 12) cm for leaves of long shoots and 4–7cm for those of short shoots. Anatomy of the Vegetative Organs As with all cormophytes, the primordial leaves of gymnosperms develop at the vegetative cone of equivalent initial cells (=initial field). Up until Strasburger’s discoveries around 1870, the opinion was held that in gymnosperms the apical cell of a shoot was comparable to ferns (pteridophytes), i.e. the original meristem develops from one single apical cell. Ginkgo, however, has developed no tunica layer (=later epidermis). Some gymnosperms, as well as the monocotyledons, have a single-layered tunica whereas a very large number of dicotyledons have a double layered one. This so-called pinus type of primordial leaf is limited almost entirely to gymnosperms and is characterised by periclinal (in addition to anticlinal) divisions in the outer layer of cells of the vegetative cone (in an apical position as well as in the development of primordial leaves). As with all seed plants, the longitudinal growth of primordial leaves has two components: apical growth, which is typical of most ferns, and intercalary growth, which prevails in angiosperms. It can be generally said that

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

32

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 7 Outlines of Ginkgo leaves, the upper ones of long shoots, the lower ones of short shoots (1/3 of natural size) (Chamberlain, 1935).

intercalary growth begins before apical growth comes to an end. This is especially true of Ginkgo which—together with various Cycadaceae—is an exception among the gymnosperms because the development of the leaf stalk precedes the development of the blade. The stalk finishes its apical growth only when the growth of the leaves comes to an end altogether. In the upper epidermal layer, the cells are more or less regularly arranged, especially above the longitudinal vascular bundles. In contrast, the cells of the lower layer are very irregularly arranged (Li et al., 1989b). In a cross section, the epidermal cells on the upper and underside are approximately isodiametric; a top view shows slightly wavy walls. The epidermal cells of Ginkgo and a few other gymnosperms are bigger on the upper side than those on the underside. The exterior walls of the epidermal cells are covered by a more or less thin cuticle covered with wax structures (chemical composition see Guelz et al., 1992). A palisade parenchyma is generally found—only the smaller leaves of flowering short shoots have no palisade parenchyma. Gymnosperms are characterised by a wide variety of excretory organs, both with regard to the substances secreted in the metabolic process and the manner of excretion. Excreted substances are mainly resins, essential oils, gums, slime, tannins and calcium oxalate. Three kinds of excretory organs can be distinguished in the leaves of Ginkgo:

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

33

1. excretory organs of lysogenic origin alternating in rows of 1–7mm in length with the vascular bundles. 2. secretory cells in curved and annular groups close to the vascular cylinder. 3. tannic idioblasts which can become thick-walled and very long by stretching or fusing. Excretory organs also occur in the leaf stalk, the primary bark and occasionally in the pith. In most cases the stomata are arranged irregularly only on the underside of the leaf. Ginkgo thus differs from the longitudinal orientation of the stomata in the group of gymnosperms. The first stomata are still oriented parallel to the nerves which is no longer true for the later ones. Intact stoma mother cells on the upper side of the leaf are rare, more frequent are those which are not fully developed.

Figure 8 Anomalies in the vascular bundles of leaves. I. After separating the two branches of a vascular bundle meet again and remain one until they reach the edge of the leaf. II. After separating the two branches of a vascular bundle meet again and then separate again. III. Two vascular bundles meet and remain one until they reach the edge of the leaf. IV. Two vascular bundles only touch each other at one point. (Arnott 1959; quoted from: Napp-Zinn, 1966)

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

34

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

These non-functioning stomata are interesting because fossil leaves of Ginkgoaceae from earlier geological periods are amphistomatic (=stomata on both sides) instead of hypostomatic (=stomata only on the underside of the leaves) as in Ginkgo biloba (Kanis and Karstens, 1963). On the other hand the stomata of the cotyledons are mainly found on the upper side of the leaf. The leaf stalk has stomata on all sides. The stomata are only a little retracted, and the exterior walls of the contiguous cells are on the same level as those of the other epidermal cells. The leaf is fed by two vascular cylinders which, interestingly enough, stem from two different (!) cords of vascular bundles. These two vascular bundles converge with the xylem parts in the stalk; both ducts have bifurcations which develop dichotomously in the transitional zone between stalk and blade and feed one of the two halves of the blade. Each bifurcation normally develops only a few anastomoses (see Fig. 8). In general, these anastomoses are more frequent in leaves of long shoots than in short ones. Occasionally there are cords of vascular bundles which begin spontaneously, i.e. without any connection to the rest of the vascular system. This can be explained by a discontinuous development of protoxylem within the blade. The vascular bundles of the Ginkgo leaf are split by numerous parachymal pith rays; in gymnosperms these are absent only in the Cycadaceae. The cells of the phloem sometimes contain crystallised calcium oxalate. As is the case with all gymnosperms, the wood consists only of tracheids which have both a conductive and stabilising function. The secondary wood of the trunk forms growth rings. There are intercellular spaces between the tracheid. Bordered pits are also found on the tangential cell walls between the autumn and spring wood. The pith rays have a width of one row of cells and a height of up to five rows of cells; they contain starch in every section and remain viable for up to 30 years. Toward the lumen of the tracheids there are big pits which converge conically from a broad base so that a top view shows the picture of bordered pits. The pith rays continue in the bark. BLOSSOMS The blossoms are always dioecious, male and female blossoms always occurring separately on two different trees. The Male Blossom The male blossoms grow from the axil of scale leaves on short shoots (see Fig. 6); they are catkin-like, short-stalked, scattered, about 2cm long, with irregularly spiral stamina placed at some distance from each other. The stalk has no leaves, only rarely with 1–2 prophylls. The stamina consist of a thin filament which has a short rotund apical extension or “knob”, from which hang two (in some cases up to four) microsporangia (=pollen sac) (see Fig. 13). In connection with surplus microsporangia, this extension was formerly interpreted as a rudimentary microsporangium. Among other things, this interpretation is based on the fact that the extinct genus Baiera, which was closely related to Ginkgo, normally had stamina with several

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

35

microsporangia. The microsporangia develop on the dorsal side of the staminodium; they hang down without contact next to each other in parallel rows and split open longitudinally on the sides turned towards each other; finally they open wide, almost horizontally, so that the pollen grains can be easily carried away by the wind. Contrary to the Cycadaceae and conifers which form an exothecium, the microsporangium also has an endothecium which is typical of angiosperms. Thus there is a thickwalled endothecium of three layers of cells under the epidermis of microsporangia (see Fig. 9). Towards the interior these are followed by parietal cells containing chlorophyll. The boat-shaped pollen grains measure 30 by 10µm with a typical median groove along their monopodium (see Fig. 10). The outermost layer of exine, also called ektexine or sexine, is three-partite: a thick tectum, a narrow infratectum with small, irregular spaces, and a distinct footlayer. The inmost layer of exine, also known

Figure 9 Microsporangium and part of the microsporophyll of Ginkgo biloba, longitudinal section with multilayered endothecium. (Jeffrey and Torrey, 1916; quoted from: Singh, 1978)

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

36

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 10 Pollen grain of Ginkgo biloba (magnification 4000 times) (Michel, 1986).

as endexine or nexine, is relatively thick and the lamellae are closely appressed to each other (Kurman, 1992). The development of the male gametophyte The development of the male gametophyte (see Fig. 11) begins with the four-cellstage of the mature pollen grain, i.e. it is composed of 2 prothallial cells, one antheridial cell, and one vegetative pollen tube cell. When a pollen grain reaches the pollination droplet (in Central Europe approximately in May), a further development of the gametophyte begins (see below). The pollen grain absorbs water and other substances from the ovular liquid; in this way it becomes heavier and penetrates through the micropyle. On reaching the pollen chamber, the vegetative pollen tube cell germinates and expands. Rhizoid-like protuberances develop on the side turned away from the archegonium chamber, thus fixing the gametophyte in the nuclear tissue. During this morphological development of the male gametophyte of Ginkgo biloba Friedman (1987) distinguishes three distinct steps: 1) After germination a brief period of diffuse growth of the tube cell resulting in a bulbous protusion of the tube cell through the sulcus, so far only known in Ginkgo biloba. 2) In Ginkgo biloba the initiation of the tip growth and the formation of a tubulagametophytic body (=normal in seed plants) is accompanied by an extensive degree of branching, giving rise to an extensive intercellular haustorial system.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

37

Figure 11 Ginkgo biloba, development of the male gametophyte. A-E First divisions up to the four-cell-stage (pc prothallial cells; cc central cell; ai antheridial initial; ac antheridial cell; tn tube nucleus; E-H development of the prothallium in the pollen chamber, in G the two spermatozoids have already developed (sc stalk cell, bc body cell, nt nuclear tissue, tp “tent pole”, pc pollen chamber, ach archegoneal chamber). (A-G Chamberlain, 1935, Favre-Ducharte, 1956; quoted from: Singh, 1978; H according to Hirase; quoted from: Pilger, 1926).

3) The late radical swelling of the proximal end of the gametophyte (resulting in the metamorphosis of an originally tubular structure into a spherical or subspherical structure) appears to be unique in Ginkgo and cycads. The antheridial cell divides into a stalk cell and a spermatogenic cell. Two spermatozoids of 70–90µm finally develop from the spermagenic cell. Among the seed plants only Cycadaceae and Ginkgo have flagellated gametes. The flagellate apparatus shows the typical 92+2 substructure (Li et al., 1989a). The discovery of the spermatozoids of Ginkgo by Hirase in 1895 and 1898 was a milestone in the history of gymnosperm embryology (Hirase, 1895; Hirase, 1898). The Female Blossom Although female blossoms grow individually in the axils of the leaves or in those of the topmost scale leaves of the spurs, there are always several blossoms placed together on one spur (see Fig. 6). A relatively thin stalk of several cm and somewhat thickened at its end has here two transversally placed sessile seed-buds. These are surrounded

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

38

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 12 Seed-bud before and after fertilisation, longitudinal, o resin-like fleshy exterior layer (sacrotesta, white), s woody central layer (sclerotesta, black), i interior layer (endotesta, broken lines) still juicy on the left, on the right preserved only as a paper like skin; c collar (double size) (Chamberlain, 1935).

at their base by a swelling called “collar” (see Fig. 12). As a rule the stalk has two seed-buds; there are occasional exceptions as, for example, with only one or several additional seed-buds, either stemless or individually petiolate. It is interesting to see that the number of vascular bundles in the stalk is always twice the number of seedbuds on the stalk. In the past, this fact led to a great variety of interpretations—also in connection with the “collar”—regarding the question if the stalk and the seedbuds should be assigned to the leaf or the shoot. The development of the female gametophyte The seed-bud of gymnosperms consists of an integument and the nucellus included in it (bitegmental seed-buds occur in Ephedra, Welwitschia, and angiosperms; Ephedra has even three integuments). In young seed-buds the nucellus is not yet close-fitting (see Fig. 13); it is only by further cell divisions in the chalaza region that the nucellus is increased so that it closely fits the interior wall of the integument. A narrow interspace is only retained in the micropylar region. The meiosis of the megaspore mother cell results in a tetrad of four haploid daughter cells. The further development of the megaprothallial starts at the lowest cell. The remaining three cells degenerate. Approximately in July, i.e. two months after pollination, this megaspore cell (=embryo

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

39

Figure 13 Left: longitudinal section through a male inflorescence (magnified 20 times). Right: longitudinal section through a young seed-bud (magnified 17 times) (Bracegirdle and Miles, 1973).

sac cell) divides in a process of free nuclear division. The result is a syncytium, a multinuclear object. It is only at a level of 200 free nuclei that the development of the cell wall membranes begins. A generically typical pollen chamber develops in the nucellus at the apical end. The pollen chamber results from the degeneration of nucellus cells in the micropylar region. The interior cells dissolve first, the epidermis cells do so last (i.e. the pollen chamber does not result from the fragmentation of the nucellus tissue during the growth of the apical region!). During the ripening of the seed-buds the nucellar tissue is reduced to a thin membrane, now wrapping the big embryo sac. At the time of ripening (1-) 2 (-3), archegonia develop from a number of marginal or central cells. The archegonia consist of a central ovum, 2–4 neck cells and a ventral canalicular cell. It is worth noting that the cells of the embryo sac contain chlorophyll. So the female gametophyte of Ginkgo biloba is the only seed plant gametophyte known to contain chlorophyll (Friedman and Goliber, 1986). As a rule there are two potential seed-buds per pedice, only one of which fully develops in most cases. EMBRYOLOGY, SEED, GERMINATION During the ripening of the ovum, the pollen chamber is extended by the development of the archegonial chamber. What happens is that a column-like structure of cells

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

40

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

called “tent pole” (see Fig. 11) develops in the embryo sac among the archegonia. It supports the nucellar cap above it, because the channel-like opening existing during pollination between the two fusing halves of the micropyle has disappeared. The two tips of the integument have formed the nucellar cap by growing together. After being set free, the two spermatozoids move through the pollen and archegonial chambers penetrating the ova where nuclear fusion takes place. There is another free nuclear division. Then parietal development begins, and an undifferentiated tissue, the proembryo, fills the ovum. Now there are special cell divisions at the end away from the micropyle (chalza region) and a massive accumulation of tissue penetrates into the endosperm without the development of a suspensory zone otherwise quite typical of gymnosperms. According to Schneckenburger (1989), however, no definite proof has been provided so far that the suspensor is not developed. From this tissue a root, a small trunk-like structure of cells and two cotyledons, gradually develops. Now and then polyembryony can be observed. In this case two possible forms can be distinguished: polyzygotic polyembryony, if there are several fertilised archegonia. The developing germs occasionally reach maturity simultaneously. The second possibility is split polyembryony which has been repeatedly observed; in this case a second embryo forms on a fully developed embryo by budding. The accessory germs develop above all at the base of the original germ; together with the latter they can reach maturity and develop into intact plants. A second embryo also develops occasionally. With regard to the chronological sequence, it is sometimes claimed that fertilisation does not take place before the seedbuds are shed (see Fig. 14). This is admittedly possible, but it is not the rule; on the other hand, there is reliable proof of the fertilisation of seed-buds still on the tree. The seed-buds are shed in autumn, they have an integument whose exterior layer, the sarcotesta, has become a fleshy resin-like substance which, when ripe, exudes an unpleasant smell of butyric acid. Further inside, there is a ligneous layer called sclerotesta. It surrounds the stone which is about 2 cm long, ellipsoid, slightly compressed, and clearly 2–3-edged [in rare cases there can be one-edged or four-edged seeds (Kartsens, 1945)]. Adherent to the stony layer is an interior thin layer of thinwalled cells, the endotesta. In the seed-buds it is soft and juicy, dries up later on, and can be identified only as a paper-like skin in the ripe seed. The abundant nutritional tissue contains starch. The relatively big embryo has two cotyledons whose upper sides are placed close to each other. In this context Ginkgo biloba shows another feature. Ginkgo—and the cycads—do not show fixed dormancy, a characteristic of the seeds of higher gymnosperms (Dogra, 1992). During germination the cotyledons remain largely enclosed by the seed. The first two leaves following on the shoots of young plants are small and scaly; contrary to the later leaves, the first leaves have several deep incisions. KARYOLOGY In the case of Ginkgo the number of chromosomes is 2n=24. This number is the same as that of other gymnosperms. Surprisingly, minimal differences have been found in

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

41

Figure 14 Diagram to illustrate the reproductive cycle: haploid phases in broken lines, diploid phases in solid lines. The total length for a 2-year-reproductive cycle is not exceptional in gymnosperms. The only unusual fact—this is not only true of Ginkgo—is that the embryonal development is completed only several months after the shedding of the seed-buds. (Chamberlain 1935; Favre-Ducharte 1962, quoted from: Singh, 1978, with additions and minor modifications).

the karyotype of one and the same plant. This also applies to the precise localisation of the centromeres and the satellites. Of course, Newcomer (1954) and Lee (1954), like other cytologists before them, tried to answer the question if perhaps a genotypical sex identification is available for the dioecious Ginkgo, i.e. if the karyogram reveals sex chromosomes of the type XY as e.g. in Melandrium (=Silene). According to Newcomer (1954), the two longest chromosomes, and according to Lee (1954), the two short satellite chromosomes (see Fig. 15), correspond to the assumed sex chromosomes of the XX and XY type. This did not yet provide a definite answer to the question of sex chromosomes. On the other hand, these different findings do not come as a surprise because the chromosomes of Ginkgo are not easily made visible

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

42

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Figure 15 Different female karyograms of Ginkgo biloba with 2n=24 chromosomes and their possible sex chromosomes. The possible sex chromosomes from the male karyogram are placed in a frame next to the possible female sex chromosomes. A Karyogram according to Newcomer (1954). For better visual quality Newcomer’s results were transferred into Lee’s (1954) karyogram. B Karyogram according to Lee (1954). C Karyogram according to Chen et al. (1993).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

43

(Chen et al., 1993; Kim, 1995). Quite recently, Chen and his collaborators have examined the chromosomes of Ginkgo with special methods (Giemsa C-banding and Ag-staining). They were able to confirm the genotypical sex identification, but not the supposed XX/XY type. On the basis of their results, Chen et al. conclude that the sex determination mechanism of Ginkgo is explained by the WZ/ZZ type, i.e. the male set of chromosomes is 2n=24=22A+ ZZ-NOR, and the female set is 2n=24=22A+WZ-NOR (NOR=Nucleolus Organisator Region). The short arm of one of the large pairs of chromosomes possesses only one obvious large satellite and NOR (named W-chromosome), while the other possesses a very small satellite and NOR (named Z-chromosome). This Z-chromosome is completely identical with the two large chromosomes (ZZ chromosomes) in the cell of the male plant. The update (Hizume, 1997) about the problem of sex-chromosomes shows that the satellites varied in occurrence and size in different trees (!). Using methods of molecular genetics Hizume pointed out that the hypothetical occurrence of sex chromosomes cannot be confirmed. It is clear that more investigations are necessary to confirm Hizume’s opinion. Otherwise another explanation of sex determination remains to be found. Several references (Ikeno, 1901; Miyoshi, 1931; Dietrich pers. communication, 1997), describing phenomena which cannot be definitely explained so far, can now be discussed in this context. The spontaneous growth of female branches on male trees without any outside influence has been observed (!). One such case, which has not yet been reported, was found in the Botanical Garden of Jena (Germany). There the development of a female branch on a 170-year-old male tree is under observation. The number of seed buds on this branch increases year by year (1993=34, 1994=132, 1995=144, 1996=538). Kato (pers. communication 1998) named three trees with female branches on male trees found in the province Iwateken, Japan. This phenomenon might be explained by point-mutation or atavism (the Ginkgo and Ginkgo-like precursors were monoecious) (personal communication received in 1997 from Dietrich, curator of the Jena Botanical Garden, Germany). This unusual feature—seeds under a male tree have so far only been reported as occurring in male trees—is probably due to the fact that it is extremely conspicuous in contrast to what can be found in female trees. PRACTICAL ASPECTS OF SYSTEMATICAL AND APPLIED BOTANY Michel (1986) pointed out that the crushed seed-buds were used as washing powder. After removing or washing off the fleshy outer layer (which smells unpleasantly rancid because of its content of butyric acid) the seed-buds can easily be removed from the stone. Eaten raw, their taste resembles that of raw potatoes with a resin-like aftertaste. Roasted Ginkgo seed-buds are regarded as a delicacy all over the Far East and are used like pistachio nuts. However, as they have stronger traditional associations, it is customary to offer them at wedding ceremonies—in that case dyed red. The wood is bright-coloured, rather hard and brittle. It is only of negligible importance, although in China miscellaneous objects are made of it. Experiments have shown that with

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

44

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

regard to the pollution of settled areas (cities), the Ginkgo can be reckoned among the relatively fume resistant plants (Kiermeier, 1984). That is why it is planted as a roadside tree more and more frequently in the USA, Europe and Japan. In the meantime, research carried out with a scanning electron microscope (Kim and Lee, 1990) has shown that the waxy cuticle of Ginkgo leaves is far less sensitive to pollution, e.g. to sulphur dioxide, than the needles of other gymnosperms (Kim and Lee, 1992; Kim and Lee, 1990). According to them, however, it is wrong to maintain that the Ginkgo is one of the most-planted roadside trees in the USA. The truth is that after several types of maple the Ginkgo holds a share of only 0.2–2% among the total number of roadside trees. Compared with European conditions this share is still quite considerable. Ginkgo is also often cultivated for its bright golden autumnal foliage. This is caused by changing optical properties associated with senescence which are due to the breakdown of chlorophyll in conjunction with a remarkably high retention of carotenoids (Matile et al., 1992). Depending on different possible uses, the cultivation and selection of certain varieties (especially with regard to growth) was begun as early as the middle of the last century. According to Kiermeier (1984), there are at present 18–20 European-American varieties and 25 Chinese fruit varieties which are of no importance to Europeans because there is no demand for roasted Ginkgo seeds. As the available seed-buds offer no clue whether or not the pollination, fertilisation, and embryonic development have taken place, a considerable loss has to be reckoned with. Consequently seed-buds of female branches grafted on male trees have a greater rate of success than when the two sexes stand at a great distance from one another. At the beginning the growth rate is quite high. Under favourable conditions young plants quickly develop a vertical axis with less robust horizontal branches. After 5 or 6 years the plants can reach a height of 2–3m. Then the growth slows down, and it takes several decades before the tree achieves a certain size and beauty. The tree likes light and sunshine, is considered to be undemanding and takes a lot of heat and drought. Flesch et al. (1991) reported that higher light intensity and higher temperatures activate the growth of long shoots and the sprouting of lateral buds. The Ginkgo thrives in a great variety of soils such as sandy humus soils or heavy loamy loess. According to Kiermeier (1984), Ginkgo tolerates a soil of pH=5.57.7. Young trees are more sensitive than older ones to late frost. In the literature on the subject it is underlined again and again that the species is largely resistant against diverse fungal diseases and that pest-infestation is hardly of any importance. Only in connection with propagation by cuttings can the roots be damaged by larvas (e.g. Bradysia paupera), but the damage can be reduced to a minimum by suitable countermeasures. SUMMARY Ginkgo biloba, a “living fossil” is the only recent representative of its genus, order, and class. The family developed its greatest morphological variety between the Triassic and Cretaceous. Contrary to a widespread opinion, Ginkgo itself was probably only represented with 2–3 species which consequently can be regarded as the direct

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

45

precursors of the recent Ginkgo biloba. Definitely the dioecious Ginkgo can morphologically and anatomously be included in the Coniferophytina. The spermatozoids of Ginkgo (within the Spermatophyta only to be found in Cycadatae) give it an exceptional position. The reproductive cycle is characterised by chronological delays: pollination in May, beginning of the ripening process of seed-buds and the ova in July, fertilisation and beginning of the embryonic development in September/ October, October/November shedding of the ripe seed-buds with or without embryo, continuing embryonic development on the ground, in late spring germination of the seedlings. The problem of sex identification of non-blossoming young plants has not been solved at present because reliable phenotypical sex characteristics are not known. Nor has genotypical sex identification according to the XY mechanism been proved beyond all doubt and is thus not yet applicable to sex identification. Consequently the safest method of obtaining young plants of the desired sex is the unproblematical propagation by cuttings. ACKNOWLEDGEMENTS We wish to thank H.Reinstadler for his support in the formulation of this manuscript. The assistance in preparing the figures from M.Thumberger is gratefully acknowledged. We appreciate the permission to use photos from the University of Chicago Press, Intersan GmbH, Gebrüder Borntraeger Verlag, A.Kato (Fürth, Germany), and Wagner Free Institute of Science (Philadelphia). REFERENCES Bracegirdle, B. and Miles, P.H. (1973) An Atlas of Plant Structure, Vol. 2., Heinemann Educational Books, London. Chamberlain, J.C. (1935) Gymnosperms, Structure and Evolution, University of Chicago Press, Chicago—Illinois. Chen, R.-Y., Song, W.-Q., Li, X.-L., An, Z.-P. (1993) A study on the sex chromosomes of Ginkgo biloba. Cathaya, 5, 41–48. Critchfield, W.B. (1970) Shoot growth and Heterophylly in Ginkgo biloba. Bot. Gaz., 131, 150–162. Del Tredici, P. (1992) Natural regeneration of Ginkgo biloba from downward growing cotyledonary buds (basal chichi). Am. J. Bot., 79, 522–530. Del Tredici, P., Ling, H., Yang, G. (1992) The Ginkgos of Tian Mu Shan. Conserv. Biol., 6, 202–209. Dogra, P.D. (1992) Embryogeny of primitive gymnosperms Ginkgo and Cycads—pro-embryo— basal plan and evolutionary trends. Phytomorphology, 42, 157–184. Dorf, E. (1958) The geological distribution of the Ginkgo family. Bull. Wagner Free Inst. Sci., 33, 1–10. Ehrendorfer, F. (1998) Gymnospermae. In Strasburger, E., Ehrendorfer, F., Lehrbuch der Botanik, 34. edition, Fischer Verlag, Stuttgart, pp. 712–731. Flesch, V., Jaques, M., Cosson, L., Petiard, V., Balz, J.P. (1991) Effects of light and temperature on growth of Ginkgo biloba cultivated under controlled long day conditions. Ann. Set. Forr. (Paris), 48, 133–148. Cited from: Biol. Abstr., 92, AB–407, Ref. No. 51988.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

46

VOLKER MELZHEIMER and JOHANNES J.LICHIUS

Friedman, W.E. (1987) Growth and development of the male gametophyte of Ginkgo biloba within the ovule (in vivo). Amer. J. Bot., 74, 1797–1815. Friedman, W.E., Goliber T.E. (1986) Photosynthesis in the female gametophyte of Ginkgo biloba. Amer. J. Bot., 73, 1261–1266. Guelz, P.G., Mueller, E., Schmitz, K., Gueth, S. (1992) Chemical composition and surface structures of epicuticular leaf waxes of Ginkgo biloba, Magnolia grandiflora and Liriodendron tulipifera. Z. Naturforsch., 47 C, 516–526. Gunckel, J.E., Wetmore, R.H. (1946) Studies of development in long shoots and short shoots of Ginkgo biloba L. II. Phyllotaxis and the organization of the primary vascular system, primary phloem and primary xylem. Am. J. Bot., 33, 532–543. Hirase, S. (1895) Etude sur la fécondation et l’embryogenie du Ginkgo biloba. Jour. Coll. Sci. Univ. Tokyo, 8, 307–822. Hirase, S. (1898) Etude sur la fécondation et l’embryogenie du Ginkgo biloba, second memoire. Jour. Coll. Sci. Univ. Tokyo, 12, 103–149. Hizume, M. (1997) Chromosomes of Ginkgo biloba. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Trémouillaux-Guiller, H.Tobe, (eds.), Ginkgo biloba—A global Treasure. From Biology to Medicine, Springer, Tokyo, pp. 109–118. Ikeno, S. (1901) Contribution à l’étude de la fécondation chez le Ginkgo biloba. Ann. Sci. Nat. Bot., 8, 305–318. Kanis, A., Karstens, W.K.H. (1963) On the occurrence of amphistomatic leaves in Ginkgo biloba L. Acta Bot. Neerl., 12, 281–286. Karstens, W.K.H. (1945) Variability of the female reproductive organs in Ginkgo biloba L. Blumea, 5, 532–553. Kiermeier, P. (1984) Zur Problematik stadtfester Gehölze. Gartenamt, 33, 239–244. Kim, S.I. (1995) Studies on the chromosome types of Ginkgo species. J. Korean For. Soc., 84, 131–144. Kim, K., Whang, S.S., Sun B.Y. (1992) Bark morphology of some Korean gymnosperms. Korean J. Bot., 35, 339–358. Cited from: Biol. Abstr., 96, AB–694, Ref. No. 6327. Kim, M.H., Lee S.W. (1992) Resistance function of woody landscape plants to air pollutants. J. Korean For. Soc., 81, 234–246. Cited from: Biol. Abstr., 95, AB–1061, Ref. No. 113187. Kim, Y.S., Lee, J.K. (1990) Chemical and structural characteristics of conifer needles exposed to ambient air pollution. Eur. J. Pathol., 20, 193–200. Kurmann H. (1992) Exine stratification in extant gymnosperms: a review of published transmission electron micrographs. Kew Bulletin, 47, 25–39. Lee, C.L. (1954) Sex Chromosomes in Ginkgo biloba. Am. J. Bot., 41, 545–549. Li, H.L. (1982) A Horticultural and Botanical History of Ginkgo. In H.L.Li, (ed.), Contributions to Botany, Epoch Publishing, Taiwan, pp. 448–457. Li, Y., Wang, F.H., Knox, R.B. (1989a) Ultrastructural analysis of the flagellar apparatus in sperm cells of Ginkgo biloba. Protoplasma, 149, 57–63. Li, Z.L., Lin, J. (1991) Wood anatomy of the stalactite-like branches of Ginkgo. Int. Assoc. Wood Anat. Bull., 12, 251–255. Cited from: Biol. Abstr., 93, AB–417, Ref. No. 27813. Li, Z.L., He, X, Xu, B.W. (1989b) The epidermal structure of Ginkgo leaf. Acta Bot. Sin., 31 427–431. Cited from: Biol. Abstr., 89, AB–573, Ref. No. 49934. Linnaeus, C. (1771) Mantissa Plantarum altera, vol. 2, p. 313. Matile, P., Flach, B.M.-P., Eller, B.M. (1992) Autumn leaves of Ginkgo biloba L.: Optical properties, pigments and optical brighteners. Bot. Acta, 105, 13–17. Mayr, H. (1906) Fremdländische Wald- und Parkbäume, Parey, Berlin, p. 288. Michel, P.F. (1986) Ginkgo biloba, ein Baum besiegt die Zeit, Intersan, Ettlingen. Miyoshi, M. (1931) Merkwürdige Ginkgo biloba in Japan. Mitt. Deut. Dendr. Ges., Jahrbuch 1931, 21–22. Napp-Zinn, K. (1966) Anatomie des Blattes: I. Blattanatomie der Gymnospermen, Gebr. Borntraeger, Berlin. Nelson, J. (1866) Pinaceae, Hatchard and Co., London, p. 163.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA L.: ASPECTS OF THE SYSTEMATICAL AND APPLIED BOTANY

47

Newcomer, H. (1954) The karyotype and possible sex chromosomes of Ginkgo biloba. Am. J. Bot., 41, 542–545. Pilger, R. (1926) Ginkgoaceae. In A.Engler and K.Prantl, (eds.), Die natürlichen Pflanzenfamilien, Vol. 13., W.Engelmann Verlag, Leipzig, pp. 98–109. Schneckenburger, S. (1989) Studien zur Embryogenese und Keimung verschiedener Gymnospermen unter besonderer Berücksichtigung der Suspensorbildung und Keimwurzelgenese. Palmarum Hortus Francofortensis, 1, 56–68. Singh, H. (1978) Embryology of Gymnosperms, Gebr. Borntraeger, Berlin. Smith, J.E. (1797) Trans. Linn. Soc., 3, 330. Sp*rne, K.R. (1965) The Morphology of Gymnosperms, Hutchinson and Co. Publ., London, pp. 164–171. Strasburger, E., Noll, F., Schenck, H., Schimper, A.F.W. (founded) Sitte, P., Ziegler, H., Ehrendorfer, F., Bresinsky, A. (1998) Lehrbuch der Botanik, 34. edition, Fischer Verlag, Stuttgart, p. 697, 658–716. Tralau, H. (1967) The phytogeographic evolution of the genus Ginkgo L. Bot. Not., 120, 409– 422. Vozenin-Serra, C., Broutin, J., Toutin-Morin, N. (1991) Permian woods of southwest Spain and southeast France: Implications for Paleozoic gymnospermous taxonomy and Ginkgophyta phylogeny, Palaeontogr. Abt. B Palaeophytol., 221, 1–26. Cited from: Biol. Abstr., 91, AB–809, Ref. No. 133453. Zhou, Z. (1991) Phylogeny and evolutionary trends of Mesozoic Ginkgoaleans: A preliminary assessment. Rev. Palaeobot. Palynol., 68, 203–216. Cited from: Biol. Abstr., 92, AB-846, Ref. No. 90978.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

3. LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA NORITSUGU TERASHIMA1 and KAZUHIKO f*ckUSHIMA2 1

2

2–610 Uedayama, Tenpaku, Nagoya 468–0001 Japan School of Agriculture, Nagoya University, Nagoya, 464–8601 Japan

INTRODUCTION More knowledge about the chemical structure of lignin in plant cell walls and its formation mechanism can lead to a better understanding and developments in a wide area of science and technology related to woody plants as follows: (1) From the viewpoint of plant science, lignin is one of the major cell wall polymers, which makes difference between trees (woody plants) and non-woody plants. The word “lignin” is derived from the Latin “Lignum” meaning wood. In the course of the evolution of plants, some species acquired a metabolic pathway to produce lignin, and their cell walls were endowed by lignification with characteristic physical, chemical and biological properties as follows: (a) lignification makes cell walls hydrophobic so that aqueous nutrients can be conducted through tissues, (b) lignification makes cell walls mechanically strong so that plants can grow higher and can extend branches to receive more sunlight, (c) lignification makes cell walls resistant to attacks by microorganisms and animals, and (d) lignification protects living cells from physicochemical effects by sunlight. These unique properties of lignified cell walls enabled trees to grow and survive for hundreds of years, and the largest portion of organic substances (fixed carbon dioxide in stable form) on the earth including humic substances in the soil originate from lignified cell walls of woody plants. (2) For the development in technology related to forestry and forest industry such as the wood, pulp and paper industry, the formation mechanism and structures of lignin and polysaccharides in the cell wall are the key fundamental knowledge, because physical, chemical and biological properties of wood and wood derived products depend primarily on the chemical structure of these polymers and their three dimensional (3D) assembly in the cell wall. Ginkgo biloba is one of the oldest still living trees on earth. Because it retains primitive characteristic features of trees appearing in the early stage of evolution of trees, it is a suitable tree species for investigation of the lignification mechanism of plant cell walls and the chemical structure of lignin. Except for the irregularity in size and shape of tracheids, the anatomical features of vascular tissue and xylem tissue of this tree are similar to those of conifers which is one of the major classes of trees on earth (Timell, 1986). Except for some of the extraneous substances, the major chemical 49 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

50

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

components of Ginkgo wood are similar to those of coniferous wood (Timell, 1986) and lignification of Ginkgo cell walls and chemical structure of Ginkgo lignin are also similar to those of conifers. In this chapter, the general feature of lignification of tree cell walls and its investigation methods will be introduced briefly, then lignification of Ginkgo xylem cell walls will be described in detail. General Feature of Lignification in Tree Xylem Cell Walls Pectin, hemicellulose, cellulose and lignin are the major components of cell walls of tree xylem. Lignified xylem cell wall is formed by successive deposition of these cell wall polymers to generate a composite in which component polymers are physically and chemically bound to each other in a biochemically regulated manner. Evolution of plants involved an evolution of biosynthetic pathways of cell wall components. This evolution of biosynthetic pathways is especially extensive in secondary metabolites including lignin. Two aromatic amino acids in the primary metabolic pathway, phenylalanine (1) and tyrosine (2) are the precursors of various aromatic secondary metabolites (Figure 1). Lignin monomers (monolignols, 5, 8 and 11) are derived from phenylalanine (1) and tyrosine (2), and polymerize to form a macromolecular lignin in the cell wall (Neish, 1968; Sarkanen, 1971; Higuchi, 1985; Haslam, 1993). Most gymnosperms, which appeared in an earlier stage of evolution than angiosperms, form p-hydroxyphenyl-guaiacyl lignin, a polymer of p-coumaryl alcohol (5) (minor monomer) and coniferyl alcohol (8) (major monomer). On the other hand, more evolved trees belonging to angiosperms form guaiacyl-syringyl lignin from coniferyl alcohol (8) and sinapyl alcohols (11). Even in the course of differentiation of a cell wall, the type of monomers supplied to the lignifying cell walls changes with the stage of differentiation. At an early stage, p-hydroxyphenyl-guaiacyl lignin is formed in the compound middle lamella (combined layers of middle lamella between cells and primary wall), while at a later stage guaiacylsyringyl lignin is formed in the secondary walls (Terashima and f*ckushima, 1989). The polymerization of these monomers is effected partly by laccase-type monolignol oxidase at an early stage and mainly by peroxidase/hydrogen peroxide in later stages of cell wall differentiation (Freudenberg, 1968; Sarkanen, 1971; Higuchi, 1985; Savidge and Udagama-Randeniya, 1992; Sterjiades et al., 1992; Bao et al., 1993; O’Malley et al., 1993; McDougall and Morrison, 1996). These monolignols are dehydrogenated to their phenoxy radicals, and coupling of their mesomeric radicals give polylignols in which monomeric units are bound by various types of linkages as shown in Fig. 2. (Freudenberg, 1968; Sarkanen, 1971). The distribution of these linkages is assumed not to be uniform in the lignin macromolecule. METHODS FOR INVESTIGATING LIGNIFICATION Non-destructive Approaches Lignin exists as a 3D macromolecule in the cell wall, and its chemical structure is not hom*ogeneous. The major part of lignin is intimately associated with polysaccharides

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

51

Figure 1 Biosynthetic pathway of lignin, and radio-labeled precursors used for specific labeling of monomeric units in Ginkgo lignin. 5: p-coumaryl alcohol, 8: coniferyl alcohol, 11: sinapyl alcohol.

by chemical and physical bonds. Therefore, it is impossible to isolate lignin from the cell wall retaining its 3D heterogeneous structure, and lignin as it is in the cell wall is often called protolignin to discriminate it from isolated lignin preparations. Because destructive analysis of protolignin causes losses in information on its 3D heterogeneous structure, investigation of the structure must be carried out by non-destructive

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

52

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

Figure 2 Lignin substructures showing various type of linkages between monomeric units.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

53

methods. One of the effective approaches is to observe the lignification process of differentiating cell walls from the early stage to the final stage. Available nondestructive methods are in two categories: (1) microscopy-spectroscopic methods and (2) tracer methods. The former category includes various techniques such as electron microscopy, UV and visible light microspectroscopy, infrared and Raman microspectroscopy, and solid state NMR. The latter includes radio- and stable isotope tracer methods. Combinations of techniques in two categories provide useful information on the formation and structure of protolignin in the cell wall. Tracer Methods Employing Radio- and Stable Isotopes Tracer methods employing radioisotopes, 3H or 14C, and the stable isotope 13C can provide useful information which cannot be obtained by any other methods. A technique of selective radio-labeling of protolignin or polysaccharides combined with a detection technique of the radioactivity by photomicrography is called microautoradiography. This enables one to visualize the deposition process of the labeled component during the formation of cell walls (Saleh et al., 1967; Fujita and Harada, 1979; Takabe et al., 1981; Terashima et al., 1988; Terashima and f*ckushima, 1989). Solid state 13C-NMR analysis combined with a technique of selective 13Cenrichment of a specific carbon in protolignin is also a useful non-destructive approach for elucidation of the chemical structure of protolignin in the cell wall (Lewis et al., 1989; Eberhardt et al., 1993; Terashima et al., 1997a; 1997b). Selective labeling of lignin Among various precursors of lignin biosynthesis, the monolignol glucosides, pglucocoumaryl alcohol (6), coniferin (9) and syringin (12) are found to be suitable precursors for selective labeling of protolignin in the cell wall. A specific hydrogen or carbon in the precursor is replaced by an isotope, and the labeled precursor is administered to a growing stem of a tree. The precursor is incorporated into the newly formed lignin in the differentiating cell wall, and the polymeric lignin is specifically labeled at a certain hydrogen or carbon corresponding to the administered monolignol. Selective labeling of polysaccharides Pectin and hemicelluloses can also be labeled selectively by administration of suitable precursors, and their incorporation into differentiating cell wall can be visualized by microautoradiography (Terashima et al., 1988; Imai and Terashima, 1990; Imai et al., 1996a; 1996b). The deposition of cellulose microfibrils can be visualized by observation of a cross section of differentiating xylem with a polarized light microscope (Takabe et al., 1981). Thus the order of deposition of major cell wall polymers, pectin, hemicellulose, cellulose microfibrils and lignin during the cell wall formation can be visualized by microautoradiography and polarized light microscopy.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

54

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

LIGNIFICATION OF XYLEM CELL WALL OF GINKGO Monolignol glucosides specifically labeled with 3H or 14C (Fig. 1, 6–1, 6–2, 9–1, 9–2, 9–3, 12–1) were administered to a growing stem of a Ginkgo tree to selectively label the corresponding position in the lignin. The process of incorporation of different monomer units and formation of different types of linkages were estimated by determination of the radioactivity in the differentiating cell walls using the technique of microautoradiography (f*ckushima and Terashima, 1991). Fig. 3 shows the microautoradiogram of differentiating xylem of Ginkgo after administration of coniferin-[OCH3-14C] (9–1). The silver grains indicate deposition of polylignol formed from coniferyl alcohol-[OCH3-14C] derived from the administered precursor. Because the major part of Ginkgo lignin is composed of a polymer of coniferyl alcohol (guaiacyl lignin) (Timell, 1986; f*ckushima and Terashima, 1991), this microautoradiogram shows the process of formation of Ginkgo lignin during the cell wall differentiation. The deposition of lignin occurs in three distinct stages (Fig. 3). The first deposition of lignin occurs at the cell corner and compound middle lamella regions just after the formation of the primary wall has finished and formation of the outer layer of secondary wall (S1) has started (Fig. 3). It has been shown by microautoradiography that pectic substances are deposited as the major polysaccharide in the newly formed cell walls at the cambium region of pine (Terashima et al., 1988) and Ginkgo (Imai et al., 1996b) prior to the deposition of lignin. Therefore, the polymerization of monolignol to polylignol takes place in a gel of pectic substances. The lignification period in this region is rather short. In the next stage, deposition of lignin is very slow. During this time, the major part of cellulose microfibrils and hemicellulose deposits in the middle layer of the secondary wall (S2), and the cell wall becomes thick. It has been shown by microautoradiography that galactoglucomannan deposits at the same time as the cellulose microfibrils (Imai et al., 1996a), and hemicelluloses may associate with cellulose microfibrils to keep them separately in a hemicellulose gel (Terashima et al., 1993). The third stage of lignin deposition occurs after the start of the formation of the inner layer of secondary wall, S3 (Fig. 3). The major part of lignin deposits at this stage in the secondary wall. Polymerization of monolignols at this stage takes place in the mannan gel. Fig. 4 shows the microautoradiograms of differentiating xylem of Ginkgo after administration of p-glucocoumaryl alcohol-[arom. ring-2–3H] (6–1), coniferin-[arom. ring-2–3H] (9–2), or coniferin-[arom. ring-5–3H] (9–3). Because the tissue section labeled with the low-energy beta emitter 3H can provide a high resolution microautoradiogram, it is possible to estimate semi-quantitatively the incorporation of labeled units in different morphological regions by counting the number of silver grains (Takabe et al., 1981). The distribution is shown in Fig. 5 (f*ckushima and Terashima, 1991). Incorporation of p-coumaryl alcohol into polylignol (p-hydroxyphenyl lignin moiety) occurs at an early stage of cell wall differentiation mainly in the middle amella region (Fig. 4a, and Fig. 5a). These observations are consistent with the results

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

55

Figure 3 Microautoradiogram of differentiating xylem of Ginkgo biloba after administration of coniferin-[OCH3–14C]. Different stages of cell wall formation can be observed in a cross section. Deposition of lignin is visualized by silver grains on the autoradiogram. The first lignification starts at the middle lamella and the cell corners after the start of S1 formation (䉰). Intensive lignification occurs after the start of S3 formation (䉳).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

56

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

Figure 4 Microautoradiogram of differentiating xylem of Ginkgo biloba after administration of (a) p-glucocoumaryl alcohol-[arom. ring-2–3H], (b) coniferin-[arom. ring-2–3H], (c) coniferin[arom. ring-5–3H]. The number in the autoradiograms shows the number of cells counted from the cell in which formation of the S1 layer started.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

57

of determination of the methoxyl content (Whiting and Goring, 1982) and UV spectra (Fergus and Goring, 1970) of lignin in the middle lamella tissue fraction. It should be noticed that 3H at position 2 of the aromatic ring of monolignol glucosides (6–1, 9– 2, 12–1) is retained after polylignol is formed, because the new bond is scarcely formed at this position during the polymerization. On the other hand, 3H at position 3 or 5 of the aromatic ring (6–2, 9–3) is removed when a new bond is introduced at this position to form condensed structural units (15, 16, 17, 18 in Fig. 2) or when a methoxyl is introduced to form guaiacyl units or syringyl units via the routes, 6–5–8 or 6–5–4–7–8, or 9–8–11 or 9–8–7–10–11 (Matsui et al., 1994). However approximately 75 and 97% of 6 and 9, respectively, were not methoxylated (f*ckushima and Terashima, 1991). Therefore, the activity remaining after incorporation into lignin may be ascribed to the presence of a non-condensed structural moiety (13, 14, 19, 20 in Fig. 2). Differences in the distribution of silver grains in Fig. 5a and 5b indicate that a considerable part of p-hydroxyphenyl lignin in the middle lamella region may be of the condensed type. A high content of condensed phydroxyphenyl lignin in the middle lamella region is also consistent with the results that oxidative analysis of this tissue fraction gave only trace amounts of phydroxybenzaldehyde and p-hydroxybenzoic acid which are not produced from condensed units (Westermark, 1985). Fig. 4b and 4c show microautoradiograms of ginkgo xylem administered with coniferin-[aromatic ring-2–3H] (9–2) and coniferin[aromatic ring-5–3H] (9–3). The distribution of silver grains is shown in Fig. 5c and 5d. It is obvious that deposition of total guaiacyl lignin (Fig. 4b and 5c) occurs in three distinct stages with two peaks of deposition as observed on the microautoradiogram of Ginkgo xylem administered with coniferin-[OCH3–14C] (9– 1) (Fig. 3). Comparison of Fig. 5c and 5d indicates that removal of 3H at position 5 of the guaiacyl ring due to the formation of condensed units (15, 16, 17, 18) is more frequent in the middle lamella region. When syringin-[aromatic ring-2–3H] (12–1) was administered, the incorporation was low and mainly in the secondary wall (Fig. 5e). Thus different types of monomer units are incorporated into lignin at the different stages of cell wall formation. As a result, the structure of Ginkgo lignin is heterogeneous with respect to the morphological regions in the cell wall. When Ginkgo wood was subjected to alkaline nitrobenzene oxidation, combined yields of aldehydes and acids were as follows; p-hydroxybenzaldehyde and phydroxybenzoic acid derived from p-hydroxyphenyl lignin: 3.5%, vanillin and vanillic acid derived from guaiacyl lignin: 95.3%, syringaldehyde and syringic acid derived from syringyl lignin: 1.3% based on the total aromatic aldehydes and acids (f*ckushima and Terashima, 1991). Pyrolysis-gas chromatography of Ginkgo wood also showed that the content of syringyl units in Ginkgo lignin is very low (Obst and Landucci, 1986). These results indicate that the major part of protolignin in Ginkgo wood consists of guaiacyl lignin containing a small amount of p-hydroxyphenyl lignin in the middle lamella region and a small amount of syringyl lignin in the secondary wall. Solid state NMR analysis of Ginkgo xylem tissue in which the side-chain ß-carbon of protolignin is specifically 13C-enriched provides information on the type and content

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

58

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

Figure 5 Distribution of silver grains in microautoradiograms of Ginkgo xylem after administration of labeled precursors. The cell number shows the number of cells counted from the cell in which the formation of the S1 layer started. CML: compound middle lamella, SW: secondary wall.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

59

of linkages between monomeric units (Terashima et al., 1997b). The content of ß-O4 substructure (14) including substructures (15), (16) and (17) in guaiacyl lignin was estimated to be 59±1.5% and combined ß-5 (18), ß-ß (19) and ß-1 (20) substructures, 36±1.5% and coniferyl alcohol/aldehyde end group (13), 4 ±1.5%. A part of ß-1 (20) substructure may be present in wood as its precursor, cyclohexadienone structure (21) (Brunow and Lundquist, 1991). Because the ß’ carbon of 21 gives a signal in the same region as the ß carbon of ß-O-4 structures in solid state NMR, the above estimated percentage for ß-O-4 includes one half of the content of structure 21 (Terashima et al., 1997b). These features on the heterogeneous formation and structure of Ginkgo lignin are very close to those observed for pine lignin (Terashima and f*ckushima, 1988). The reason why different kinds of labeled precursor were incorporated into different specific morphological regions may be explained partly by the substrate specificity of ß-glucosidase which catalyzes the hydrolysis of monolignol glucosides to corresponding alcohols. It has been shown that the cell-wall bound ß-glucosidase in spruce seedlings has different substrate specificity towards three kinds of monolignol glucosides (Marcinowski and Griesebach, 1978), and ß-glucosidase specific for coniferin has been found in lignifying xylem of lodge-pole pine (Dharmawardhana et al., 1995). Heterogeneity in lignification may be related to the development of biosynthetic pathways not only of lignin but also of polysaccharides during the differentiation of cell wall (Terashima et al., 1993). As shown above, studies on lignification of Ginkgo xylem provided important information on the lignification mechanism and the lignin structure in tree cell walls. Based on the results including those obtained with Ginkgo, a hypothetical general struc-ture of lignin has been proposed (Terashima et al., 1998). In the future, determination of the frequency and distribution of bonds between lignin and polysaccharides will be important for better understanding the 3D assembly of major components in the cell wall. Evolution of plants also involved formation of compression wood at the lower side of the inclined stem or branch of trees belonging to conifers. Morphology and chemical composition of compression wood differs from normal wood, and the quality of compression wood is generally inferior for industrial use. Ginkgo also forms compression wood, however the difference between compression wood and normal wood is not as significant as the differences observed in pine (Timell, 1983; Tomimura et al., 1980; f*ckushima and Terashima, 1991). Detailed studies on the differences in lignification mechanism of compression wood between Ginkgo and other conifers may provide useful information for the improvement of forest trees in the future.

CONCLUSION The process of lignification of Ginkgo xylem cell wall is similar to that of conifer xylem cell wall. The characteristic features of formation and structure of Ginkgo lignin are summarized as follows:

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

60

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

1. Lignified cell walls are formed by successive deposition of pectin, hemicellulose, cellulose and lignin in differentiating xylem cell walls. The kind of polysaccharide and type of monolignol supplied to the cell wall changes with the age of the cell. Then the lignin formation takes place in different ways in the compound middle lamella and the secondary wall. 2. The major part of Ginkgo lignin is guaiacyl lignin. At an early stage of cell wall formation, guaiacyl lignin containing a small amount of p-hydroxyphenyl-propane units deposit rather quickly in a gel of pectic substances in the compound middle lamella region. At a later stage, the major part of guaiacyl lignin containing a small amount of syringylpropane units deposits rather slowly in mannan gel in the secondary wall. 3. The lignin formed in the compound middle lamella contains more condensed type substructures than the lignin formed in the secondary wall. About 60% of monomer units are bonded by a ß-O-4 type linkage. 4. Thus the lignification of Ginkgo xylem proceeds heterogeneously with respect to the morphological region of the cell wall and with respect to the 3D macromolecular lignin structure. Further extensive studies on the lignification of Ginkgo cell walls in the future will promote better understanding in a wide area of science and technology related to forestry and forest industry. REFERENCES Bao, W., O’Malley, D.M., Whetten, R. and Sederoff, R.R. (1993) A laccase associated with lignification in loblolly pine xylem. Science, 260, 672–674. Brunow, G. and Lundquist, K. (1991) On the acid catalysed alkylation of lignins. Holzforschung, 45, 37–40. Dharmawardhana, D.P., Ellis, B.E. and Carlson, J.E. (1995) A ß-glucosidase from lodge-pole pine xylem specific for the lignin precursor coniferin. Plant Physiol., 107, 331–339. Eberhardt, T.L., Bernards, M.A., He, L-F., Davin, L.B, Wooten, J.B. and Lewis N.G. (1993) Lignification in cell suspension culture of Pinus taeda. In situ characterization of a gymnosperm lignin. J. Biol. Chem., 268, 21088–21096. Fergus, B.J. and Goring, D.A.I. (1970). The distribution of lignin in birch wood as determined by ultraviolet microscopy. Holzforschung, 24, 118–124. Freudenberg, K. (1968) The constitution and biosynthesis of lignin. In K.Freudenberg and A.C.Neish, Constitution and Biosynthesis of Lignin, Springer-Verlag, Berlin, pp. 85–97. Fujita, M. and Harada, H. (1979) Autoradiographic investigation of cell wall development II. Tritiated phenylalanine and ferulic acid assimilation in relation to lignification. Mokuzai Gakkaishi, 25, 89–94. f*ckushima, K. and Terashima, N. (1991) Heterogeneity in formation of lignin. XIV. Formation and structure of lignin in differentiating xylem of Ginkgo biloba. Holzforschung, 45, 87– 94. Haslam, E. (1993) Shikimic acid-metabolism and metabolites, John Wiley & Sons, Chichester, pp. 227–238. Higuchi, T. (1985) Biosynthesis of lignin. In T.Higuchi (ed.), Biosynthesis and Biodegradation of Wood Components, Academic Press, Orlando, pp. 141–160. Imai, T. and Terashima, N. (1990) Determination of the distribution and reaction of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

LIGNIFICATION OF XYLEM CELL WALLS OF GINKGO BILOBA

61

polysaccharides in wood cell walls by the isotope tracer technique. I. Selective radiolabeling of cell wall polysaccharides in Magnolia kobus. Mokuzai Gakkaishi, 36, 917– 922. Imai, T., Goto, N., Yasuda, S., and Terashima, N. (1996a) Visualization of deposition process of mannan in differentiating cell wall of Ginkgo and pine by microautoradiography. Proc. 46th Conf. Wood Res. Soc. Japan, Kumamoto, April 1996, p. 328. Imai, T., Goto, N., Yasuda, S., and Terashima, N. (1996b) Selective labeling of non-cellulose polysaccharides of Ginkgo and pine. Proc. 46th Conf. Wood Res. Soc. Japan, Kumamoto, April 1996, p. 329. Lewis, N.G., Razal, R.A., Yamamoto, E., Bokelman, G.H. and Wooten, J.B. (1989) Carbon13 specific labeling of lignin in intact plants. In N.G.Lewis, and M.G.Paice, (eds.), Plant cell wall polymers, biogenesis and biodegradation; ACS Symposium Series 399, American Chemical Society, Washington DC, pp. 169–181. Matsui, N., f*ckushima, K., Yasuda, S. and Terashima, N. (1994) On the behavior of monolignol glucosides in lignin biosynthesis. II. Synthesis of monolignol glucosides labeled with 3H at the hydroxymethyl group of side chain, and incorporation of label into Magnolia and Ginkgo lignin. Holzforschung, 48, 375–380. McDougall, G.J. and Morrison, I.M. (1996) Extraction and partial purification of cell-wallassociated coniferyl alcohol oxidase from developing xylem of sitka spruce. Holzforschung, 50, 549–553. Marcinowski, S. and Griesebach, H. (1978) Enzymology of lignification. Cell-wall-bound ßglucosidase for coniferin from spruce (Picea abies) seedling. Eur. J. Biochem., 87, 37–44. Neish, A.C. (1968) Monomeric intermediates in the biosynthesis of lignin. In K.Freudenberg and A.C. Neish (eds.), Constitution and Biosynthesis of Lignin. Springer-Verlag, Berlin, pp. 3–43. O’Malley, D.M., Whetten, R., Bao, W., Chen, C-L. and Sederoff, R.R. (1993) The role of laccase in lignification. Plant J., 4, 751–757. Obst, J.R. and Landucci, L.L. (1986) The syringyl content of soft wood lignin. J. Wood Chem. Technol., 6, 311–327. Saleh, T.M., Leny, L. and Sarkanen, K.V. (1967) Radioautographic study of cotton wood, Douglas fir and wheat plants. Holzforschung, 21, 116–120. Sarkanen, K.V. (1971). Precursors and their polymerization. In K.V.Sarkanen and C.H. Ludwig (eds.), Lignins, Occurrence, Formation, Structure and Reactions. Wiley-Interscience, New York, pp. 95–163. Savidge, R. and Udagama-Randeniya, P. (1992). Cell wall bound coniferyl alcohol oxidase associated with lignification in conifers. Phytochemistry, 31, 2959–2966. Sterjiades, R., Dean, J.F.D., and Ericksson, K.-E.L. (1992). Laccase from sycamore maple (Acer pseudoplatanus) oxidizes monolignols. Plant Physiol., 99, 1162–1168. Takabe, K., Fujita, M., Harada, H. and Saiki, H. (1981) Lignification process of Japanese black pine (Pinus thunbergii Parl) tracheids. Mokuzai Gakkaishi, 27, 813–820. Terashima, N., f*ckushima, K., Sano, Y. and Takabe, K. (1988) Heterogeneity in formation of lignin. X. Visualization of lignification process in differentiating xylem of pine by microautoradiography. Holzforschung, 42, 347–350. Terashima, N. and f*ckushima, K. (1988) Heterogeneity in formation of lignin. XI. An autoradiographic study of the heterogeneous formation and structure of pine lignin. Wood Sci. Technol. 22, 259–270. Terashima, N. and f*ckushima, K. (1989) Biogenesis and structure of macromolecular lignin in the cell wall of tree xylem as studied by microautoradiography. In N.G.Lewis and M.G.Paice, (eds.), Plant Cell Wall Polymers, Biogenesis and Biodegradation, ACS Symp. Ser. 399, Am. Chem. Soc., Washington, DC. pp. 160–168. Terashima, N., f*ckushima, K., He, L-F., and Takabe, K. (1993) Comprehensive model of the lignified plant cell wall. In H.G.Jung, D.R.Buxton, R.D.Hatfield and J.Ralph (eds.), Forage Cell Wall Structure and Digestibility, Am. Soc. Agronomy, Madison, Wisconsin, pp. 247– 270.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

62

NORITSUGU TERASHIMA and KAZUHIKO f*ckUSHIMA

Terashima, N., Atalla, R.H. and Van der Hart, D.L. (1997a) Solid state NMR spectroscopy of specifically 13C-enriched lignin in wheat straw from coniferin. Phytochemistry, 46, 863– 870. Terashima, N., Hafrén, J., Westermark, U. and Van der Hart, D.L. (1997b) Structure of lignin in Ginkgo wood determined by a combination of specific 13C-enrichment technique and solid state NMR spectroscopy. In Proc. 9th Intern. Symp. Wood and Pulping Chem., CPPA, Montreal, pp. H1-1–H1-5. Terashima, N., Nakashima, J. and Takabe, K. (1998) Hypothetical structure of protolignin in plant cell walls. In N.G.Lewis and S.Sarkanen, (eds.), Lignin and Lignan Biosynthesis, ACS Symp. Ser. 697, Am. Chem. Soc., Washington, DC, pp. 180–193. Timell, T.E. (1983) Origin and evolution of compression wood. Holzforschung, 37, 1–10. Timell, T.E. (1986) Origin and evolution of compression wood. In Compression Wood in Gymnosperms, Springer-Verlag, Berlin, Vol. 1, pp. 597–621. Tomimura, Y., Yokoi, T. and Terashima, N. (1980) Heterogeneity in formation of lignin. V. Degree of condensation in guaiacyl nucleus. Mokuzai Gakkaishi, 22, 259–270. Westermark, U. (1985) The occurrence of p-hydroxyphenylpropane units in the middle lamella lignin of spruce (Picea abies). Wood Sci. Technol., 19, 223–232. Whiting, P. and Goring, D.A.I. (1982) Chemical characterization of tissue fractions from middle lamella and secondary wall of black spruce tracheids. Wood Sci. Technol., 16, 261–267.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

4. CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE DOMINIQUE LAURAIN Laboratory of Pharmacognosy, Faculty of Pharmacy, 1 rue des Louvels, 80 037 Amiens Cédex 1, France

INTRODUCTION Ginkgo biloba, the unique living representative of the order Ginkgoales which appeared in the Permian era, is an extraordinary tree because of its numerous botanical and physiological characteristics. Some of its properties, known since time immemorial in China, are nowadays exploited in two very different fields, i.e. in horticulture and in therapeutics. Its interest for horticulture is linked to its wonderful foliage which takes on a bright yellow colour in autumn and to its resistance to all serious pests, bacteria, fungi, viruses and equally to its tolerance to air pollution (Rensselaer, 1969). Another example of its strength is shown by the regeneration of a shoot of Ginkgo the year following its destruction by the Hiroshima bomb (Michel, 1985). That is why this tree has become a favourite ornamental plant in parks and streets in Europe and the U.S.A. Ginkgo is also of great interest for therapeutics because of its leaves which possess pharmaceutical properties such as radical scavenging, improved blood flow, vasoprotection and anti-PAF activity (DeFeudis, 1991; O’Reilly, 1993). Therefore the pharmaceutical industry needs huge quantities of leaves, two thousand tonnes of leaves are used each year in the world (Masood, 1997). China is the major producer in the world of Ginkgo leaves (Fusheng et al., 1997). In this country leaves are harvested by hand from young trees growing for an ornamental use or from trees 5–20 m high (O’Reilly, 1993). Also fallen leaves can be collected during the harvesting of Ginkgo ovules (Fusheng et al., 1997). These methods of harvesting pose however problems in terms of regular production and leaf quality. The chemical synthesis of some active molecules, like ginkgolide B (Corey et al., 1988), could permit a production without trees but it is a long process and is therefore unprofitable. So to meet the growing demand for Ginkgo biloba trees and leaves it is necessary to set up plantations (Fig. 1). Only few plantations of Ginkgo biloba with an industrial vocation are established in the world. In Europe, one sole plantation is established in the Bordeaux region (France) since 1982. Depending on the aim of the plantation, growing methods may be different, especially propagation, pruning and mineral nutrition. This chapter first deals with some aspects of the biology and the architecture 63 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

64

DOMINIQUE LAURAIN

Figure 1 Ginkgo biloba plantation in Champaubert (France)

of the tree that are instrumental in managing those plantations efficiently. Then methods of propagating Ginkgo, those of culture in plantations and harvesting of leaves for pharmaceutical use are described. BIOLOGY OF GINKGO Ginkgo biloba is dioecious, possesses deciduous leaves and can exceed 30 m in height. The male reproductive organs consist of catkins situated at the level of short shoots which carry leaf clusters. The stamina carry two pollen bags and the anthesis takes place around mid-April in France. The female reproductive organs consist of spherical ovules in two’s at the level of one petiole. Usually one of the two aborts before reaching maturity. Meiosis takes place in mid-April and gives rise to four haploid cells. Only the lowest of the four cells will evolve progressively toward a female gametophyte. Early in September, three layers of the ovule tegument differentiate inwards to form the sarcotesta, sclerotesta and endotesta. The reproductive organs appear in spring on trees that are at least 20 years old. However sexes can be distinguished before the trees are mature because leaves of male trees develop before the female ones and in autumn the males lose their leaves two or three weeks earlier than the female ones (Michel, 1985). Ginkgo trees have characteristic leaves with one or several lobes, dichotomic nervation and a long petiole. Trees present archaic

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

65

characteristics with a zoidogamous fertilization and with a chlorophyll female gametophyte which accumulates reserves and parts with the sporophyte before fertilization (Favre-Duchartre, 1956). In France fertilization happens in September, 4.5 months after anthesis. The embryo develops immediately without a latent phase. During the Quaternary era Ginkgo biloba was found only in China but nowadays it is planted in any temperate geographic region. This tree was introduced in Europe during the first half of the 18th century and it reached America in 1784 (Rohr, 1989). ARCHITECTURE OF THE TREE Ginkgo biloba trees raised from ovules are composed of an orthotropic monopodial trunk with plagiotropic branches per tier. Its architecture conforms to Massart’s Model according to the classification system of Halle and Oldeman (1970). The trunk is formed by a single meristem that is active throughout the life of the tree. The form of vigorous young trees is quite conifer-like, with a dominant central leader and whorled lateral branches. As the tree ages the pyramidal shape is lost and a broad, spreading Table 1 Different cultivars of Ginkgo biloba used in horticulture and their properties (Nurseryman Adeline, La Chapelle Montlinard, France).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

66

DOMINIQUE LAURAIN

form develops (Del Tredici, 1991). Depending on cultivars the shape of the tree will be modified (Table 1). The shoot system of Ginkgo biloba is composed of long and short shoots. Long shoots make up the woody framework of the tree, they increase in length from 2 cm to more than 1 m per year (Del Tredici, 1991). Leaves are arranged spirally in a 5/13 phyllotaxis (Gunckel and Wetmore, 1946). On the long shoots, leaves are alternate, they carry an axillary bud and a terminal bud is present at the top of the shoot. A short shoot is formed when an axillary bud of an auxiblast develops bringing about the formation of a rosette of leaves with a terminal bud (Camefort and Boue, 1969). The growth of these short shoots is limited to about 1 to 2 mm per year and the internodes are short. These produce the majority of leaves on the tree and when the Ginkgo is 25–30 years old, they bear the reproductive structures developed within the rosettes of leaves. These short shoot leaves lack axillary buds (Critchfield, 1970). Under certain conditions, some short shoots are transformed into long shoots. This ability is frequent in the conifers when a reduction or a deletion of the carrier axis dominance appears after a traumatism (Nozeran, 1986). In Ginkgo biloba this event, which is in relation with a late partial reiteration (defined as the repetition of the seedling model), happens spontaneously on trees physiologically in senescence phase (Edelin, 1986). This phenomenon has been observed also on young trees of G.biloba in growing phase (Laurain, unpublished). The alteration of the auxin content of the shoot tip under extended light, high nutrition or other conditions could explain this phenomenon (Collins, 1903; Gunkel and Thimann, 1949). Another mode of reiteration in Ginkgo biloba is the development of secondary stems directly from the base of old Ginkgos. According to Del Tredici (1991) this behaviour is common in China but not often seen in plants cultivated in Europe and in North America. However some cases have been observed in France, in particular on trees in the Plant Garden in Nantes and in private people’s gardens in Mouleydier and in Creysse. Another distinctive feature of Ginkgo trees growing in their natural habitat is the production of distinct organs called aerial “chichi” and basal “chichi” according to where they are located on the tree. Aerial “chichi” look like “air roots” or “burls” produced along the underside of large lateral branches of old trees (Del Tredici, 1991). Basal “chichi” can be defined as positively geotropic aggregates of suppressed shoot buds, they are located in the cotyledonary axils of all Ginkgo seedlings. “Chichi” constitute a particular system of vegetative rejuvenation of Ginkgo biloba. In fact they are capable of generating both aerial shoots and adventitious roots under appropriate conditions (Del Tredici, 1991) or when cut off from the parent trunk and planted upside down in soil (Hu, 1987; Li and Lin, 1990). Thus basal “chichi” could be a unique mechanism of clonal regeneration of this tree and interesting for commercial purposes (Del Tredici, 1992). PROPAGATION The establishment of large scale Ginkgo plantations for ornamental or pharmaceutical use requires a propagation system making it possible to get many new trees quickly.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

67

The Ginkgo biloba can be multiplied by vegetative propagation or sowing. Micropropagation by plant tissue culture systems has not yet been published for Ginkgo trees. Ginkgo biloba somatic embryos have been obtained from immature zygotic embryos (Laurain et al., 1996) but they did not lead to plantlets. Vegetative Propagation In the horticultural field, male individuals are chosen for asexual reproduction because of the foul odor of ovules that contain butyric acid along with other compounds (Santamour et al., 1983) making the female plant considerably less desirable. No relationship between plant sex and metabolite content in leaves has been shown. Concerning leaf production for the pharmaceutical industry, male or female plants are indifferently used. On the other hand vegetative propagation is effective for multiplying a hyperproductive tree which bears leaves rich in active components. Three vegetative propagation techniques have been described for Ginkgo: cutting, running and grafting. Cutting is one technique of propagation but is not recommended for Ginkgo because it generally brings about weak vigour in trees which causes limited vegetative development. Several parameters have been shown to influence the rooting of cuttings. Thus the use of plant growth regulators has increased the rooting percentage and decreased the time required to achieve rooting (Huh and Staba, 1992). Treatments of basal stem cuttings with indolebutyric acid (50mg/l for 23 hours) led them to root within 30 days whereas non-treated cuttings required 60 days to root (Doran, 1954). Skirvin and Chu (1979) have obtained similar results. The nature of shoots used as cutting material and the period of the year when cutting takes place influence the success of the vegetative multiplication. Short lateral shoots root well (Teuscher, 1951) but their growth is not satisfactory (Vermeulen, 1960). Apical shoots are better but less numerous. Woody cuttings gathered in March–April and treated with growth regulators give interesting results (Michel, 1985). The young age of the shoots is also a factor of success in cuttings. In this way cuttings, taken from a Ginkgo which had many young shoots after being drastically trimmed, have given the best results in rooting cuttings (Laurain, 1990). Rejuvenating mother plants will therefore be an effective means for obtaining cuttings. Layering is unacceptable because layers do not grow in a straight line and trunks are twisted (Krussmann, 1968). On the other hand, grafting is the technique used to ensure the propagation of Ginkgo cultivars known in horticulture (Table 1). This technique is also used in nurseries for propagating only male trees and the cuttings are taken exclusively from male trees. Grafting also makes it possible to obtain trees bearing male and female characteristics on a single tree. Putting together male and female sexes on a single tree improves the fertility when male trees are missing. In France, female Ginkgos bearing a grafted male branch have been observed in the “Jardin des Plantes” in Paris and in the “Jardin Botanique” in Tours. Grafting is variable: the age of the receiving tree and the type of transplant as well as the period of the year are all of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

68

DOMINIQUE LAURAIN

influence. Grafting can be carried out in winter, after leaf fall and on sown potted plants during their first year of vegetation (Krussmann, 1968) or on older trees (2 to 3 years old). In mid-August shield-grafting or most often inlaid-grafting can be carried out at the level of the neck of a receiving tree in its second or third year of vegetation. The used graft should have two eyes. A scion one meter high is obtained within the year and the tree remains three years in a nursery before being transplanted. Studies on vegetative multiplication of Ginkgo are few and date back several decades. However one may conclude that the most adapted technique for Ginkgo is grafting. Sexual Reproduction Through sexual reproduction of Ginkgo a large number of plants can be obtained quickly. Moreover, they present a faster vegetative development than trees from cutting or grafting. The main drawback with this propagation technique is the genetic variability concerning both the vegetative traits and the secondary metabolite content of the leaves. Ginkgo biloba is dioecious and in order to propagate it through sowing, one must harvest ovules which come from female trees situated at a favourable distance from male trees for ovule fertilisation. In France, ovules are usually collected in November, after they have spontaneously fallen down from trees. It is equally possible to buy ovules imported from China, Korea and Japan. The price of 1kg of ovules is variable according to the tradesman, it can reach 40 dollars. The size of the ovules is variable (0.76 to 4.17g per ovule) depending on the trees which were bearing them and is not related to their germination potential. After the harvest, ovules are cleared of sarcotesta (yellow fleshy layer which emits a foul odour due to butyric acid) and washed with water. Usually the embryo cannot be seen with the naked eye, the fertilisation having only taken place in September. Then there is no latent phase, the embryo carries on developing according to external temperature: the higher the temperature, the faster the development. When the ovules have to wait for adequate temperature they must be cleaned and treated with a fungicide and kept in a mixture of moist peat and sand (80/20). Stocking takes place in the dark at a temperature over 4°C during winter until the time comes when they can be sown in the open-air after the last frosts in spring. In the north-east of France, sowing takes place at the end of April-beginning of May and the shoots appear in June. South of the Loire, in spring young shoots spontaneously burst out under the tree. Before sowing, soaking the ovules in hot water should provide injuries in the sclerotesta which should allow speedy germination (Krussmann, 1968). Two sowing methods can be used for large scale cultivation of Ginkgo biloba. The first possibility consists of letting the plants remain for a year in the nursery before transferring them outside (Krussmann, 1968), but it is better to replant them after the second year of vegetation in countries where the winter is particularly rigorous. Ovules are sown in nurseries in soil which has been previously desinfected with methyl bromide (O’Reilly, 1993) every 10cm in furrows 50cm from one another. The principal

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

69

drawbacks related to this method are the loss of plants after transplantation and slowing of vegetative development for several years compared to a non-transferred plant. It has been noted that rooting of transferred plants is different to that of plants which have not been transferred, following a cut in the radical pivot when it is uprooted (Laurain, 1990). One may note an emission of numerous roots producing a network of rootlets which is bigger and more ramified than those of plants with an uncut pivot. The health of transplanted plants is influenced by several external factors which are difficult to control: humidity, temperature and nature of soil. Replanting the trees is interesting because they can be spaced regularly at 50 cm on rows one meter from one another. However the space between the rows and the number of plants per linear meter can be modified, thus plantations contain 20,000 to 30,000 trees per hectare. One year old plantlets can be bought for 3 to 5 francs the piece according to the number bought and be planted directly in the plantation. The second possibility is directly sowing on the plantation using a machine such as a garlic sower. The main drawback of this method is an irregular germination due to the variable origin of the ovules. A fertility rate of 70% has been observed for ovules harvested in France (Laurain, unpublished). Concerning ovules native from China, Korea and Japan, a 50 per cent germination was considered good (O’Reilly, 1993). This method has the advantage of limiting plant loss when replanting. Besides plant development is not disturbed. In Champaubert (France), plants reaching about 20 cm high in the first year of vegetation, 50 cm in the second year and a meter in the third year have been observed. On the other hand, after the second year of vegetation plant density must be regular, that is to say two plants per linear meter by uprooting superfluous plants. PRACTICES OF CULTIVATION Pruning According to the aim of the plantation, the practice of prunning is different. Ornamental Ginkgos require no particular pruning. However, large urban Ginkgos may require drasting pruning according to size. This happened to 12 Ginkgos planted in 1864 in Saint Sulpice Laurière (France). In 1920, when the trees were 56 years old, they were topped at 7 meters from the ground and cut all around the trunk. After that operation, they developed several new shoots at the top of the trunk as well as numerous lateral shoots. One should note that this traumatism did not provoke new shoots at the base of the trunk. This is a good example of Ginkgo’s resistance to pruning and its big ability to regenerate new stems. These properties are exploited in Ginkgo plantations whose aim is leaf production. Ginkgos are pruned every year in winter in order to limit vegetative development as it is realized for the upkeep of hornbeam hedges for example. Trees should not exceed a height of 1.20 meter to allow mechanized leaf harvesting. Trees can bear these annual prunings at least during 15 years, the older plantations, with an industrial vocation, being set up in 1982 (O’Reilly, 1993). Every five years a major pruning is carried out at the base of the

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

70

DOMINIQUE LAURAIN

tree in order to rejuvenate the plants and make the shoots vigorous. Between 2 to 5 new shoots can reach a height of 1.20 meter the first year after the major pruning. Likewise it has been noted that after the cutting of the hypocotyle or the stem of a young plant, one or two new shoots may develop again. In the years following the pruning, many new shoots grow but they do not reach a height of 1.20meter. This large regenerative capacity might be explained by the presence of basal chichi reported by Del Tredici (1992) which are located in the cotyledonary axils of all Ginkgo seedlings and which are able to produce aerial stems in response to a traumatic injury. Culture Conditions Ginkgo biloba grows in a large variety of soils. However for best growth it should have a deep, fairly rich soil, limestone, silicious-argilo calcareous with neutral pH. It may also develop in acidic soils and tolerates alkaline soils although its growth is then reduced. It does not grow in soaked soils, however occasional deep watering gives best results and it needs sufficient rainfall, approximately 90cm/year (Franklin, 1959). Ginkgo also shows an extraordinary ability to adjust to a wide variety of climates. It grows in southern Canada and in other regions where winter temperature may drop to -25°C (Rensselaer, 1969). The tree is indigenous to China where natural populations supposedly exist in the Chekiang province of eastern China (Chen, 1933; Li 1956). Currently Ginkgo trees can be found in Asia, Europe, North America and in temperate regions of New Zealand and Argentina (Huh and Staba, 1992). It grows up to an altitude of 1000meters (Michel, 1985). Ginkgo is known to be a slow grower. For example the trees at St Sulpice Laurière (France) are 135 years old and have a circumference of between 1.65 and 3.10 meters which corresponds to an annual growth of about 1–2cm. This result is comparable to oak-growth. However Ginkgo growth may be quicker. For example the Ginkgo in Mouleydier (France) is about 150 years old and has a circumference of 4.45meters which corresponds to an average of 3cm a year. Depending on the situation, climate and soil, Ginkgo’s growth is more or less speedy. Fertilisers can improve the vegetative development of the tree. On the other hand, like fruit trees, a female Ginkgo which has an intense vegetative development has been noted to produce no ovules. About the growth of Ginkgo, some criteria as the elongation speed of the main axis, the final size of the main axis, the plastochrone and the number of leaves formed, can be influenced by light and temperature under controlled long day conditions. On the other hand the alternation between long and short internodes and the insertion level of ramifications are independent (Flesch et al., 1991). Mineral Nutrition Few studies have been published on the mineral nutrition of Ginkgo biloba. Observations on a field Ginkgo plantation and assays (Laurain, 1990) realized in the I.N.R.A. Agronomy Station in Angers (France) on plants in 2.5 l containers (containing a mixture of grinded pine-bark and yellow peat in equal quantities with the pH

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

71

Table 2 Composition (macronutrients) of nutritious solutions (Coïc Lesaint type) used for the mineral nutrition of Ginkgo biloba trees in containers.

adjusted to 6–6.3) have shown that Ginkgo biloba reacts well to additions of fertilizers. “Two year old Ginkgo plants in containers have received nutrient solutions of the type Coïc Lesaint (Table 2) containing increasing doses of nitrogen (29.4, 58.8, 116.2, 175, 233.8mg/l) but with a constant ratio NO3/NH4 of 12/2 (approximately).”. This study has shown an increase of plant growth as a function of increasing doses of nitrogen and particularly an increase of the number of lateral ramifications and a decrease of the apical dominance at the concentration of 116.2mg/l of nitrogen. Following this study, the N, P, K, Ca and Mg contents in leaves and in one year old shoots have been analysed as a function of nitrogen doses. The results are presented in Table 3. In leaves the Mg content is stable at the different nitrogen doses while the N, P, K and Ca contents increase with the increasing doses of nitrogen. In shoots, the P and Ca contents do not increase, the Mg content decreases and the N and K contents increase with the increasing doses of nitrogen. Nitrogen fertilization effects on foliar N and P dynamics and autumnal resorption in Ginkgo biloba have been studied (Brinkman and Boerner, 1994). Foliar N and P concentrations were more similar to those of angiosperms than of gymnosperms. On the other hand the levels of N and P resorption were similar to those reported for a wide range of gymnosperms and greater than those reported for most angiosperms. Weed Control The plantations are kept weed-free with a minimum use of herbicides in view to harvest leaves without pesticide residues. The soil of the nurseries and of the plantations are treated with a pre-emergent herbicide (diflufenicanil+isoproturon) before sowing. The same herbicides are sprayed in March before the leaves appear. During the growing

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

72

DOMINIQUE LAURAIN

Table 3 Contents (%) of N, P, K, Ca, Mg in leaves and in annual shoots of Ginkgo biloba trees two years old, cultivated in container, as a function of nitrogen doses.

season, the nurseries and the plantations are kept clean by burning and also weeding by hand. Local and pointed applications of glyphosate can be carried out during the growing period. Some herbicides have been tested on Ginkgo biloba (C.N.I.H., 1985), and the following active substances can be used in the plantations: alloxydim sodium, chloroxuron, lenacil+neburon, oxadiazon granules, oxadiazon+carbetamid, propyzamid, trifluralin. It is also possible to cultivate herbaceous plants between the rows occupied by Ginkgo trees in winter to keep down the weeds (O’Reilly, 1993). Pests Ginkgo is reputed to be resistant to all serious pests (Franklin, 1959). This tree has not acquired predator fauna in North America and it is unusual that the plant has no predator in its native habitat of East Asia (Wheeler, 1975). However some predators have been identified in French Ginkgo plantations. They are different according to the age of the plant and the situation of the plantation. As soon as a shoot appears on the soil surface it runs the risk of being destroyed by birds, such as pigeons, doves and partridges. During the first years of vegetation, snails, slugs, mouses, rabbits and hares are principal predators. These pests eat the bark at the bottom of the tree which can cause death if the bark is removed all around the trunk. When the plantation is situated next to a forest, stags break branches and destroy the bark. To prevent attack by rodents it is advisable to put a plastic netting around each tree trunk up to a height of 50 cm or to fence off the plantation. It is recommended to cover nurseries with netting to protect young plants from birds

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

73

and to spread metaldehyde, thiodicarb, mercaptodimethur or bensultap granules on the ground against slugs and snails. Ginkgo is one of the least vulnerable trees as far as insects are concerned. They cause only minimal damage and none is specifically seen as a threat. The leaves are used in China and Japan as bookmarks to protect books from silverfish and larvae of other insects (Major, 1967). However a weasy insect, Cacoecimorpha pronubana, proved obnoxious to leaves (Martinez and Chambon, 1987). In addition some insects, Brachytrupes portentosus, Agrotis ypsilon and Gulcula panterinaria, were found to be the main causes of death of seedlings of Ginkgo biloba (Zhiquan et al., 1991). The larvae of the corn pyral (Pyrausta mubilalis) can eat Ginkgo leaves but if an aqueous extract of the same leaves is given to the larvae, their growth is considerably inhibited (Major, 1967). In addition, Major and Tietz (1962) observed that Japanese beetles died of starvation rather than eat fresh leaves of Ginkgo biloba. Investigations have been carried out on the insecticidal properties of the tree. It has been shown that the leaves are very acidic and that 2-hexenal is produced when Ginkgo leaves are damaged in the presence of oxygen (Major et al., 1963). Bevan et al. (1961) have reported that this aldehyde is an insect repellent. Ginkgolide A has been found the most active of several compounds which included bilobalide and ginkgolic acids, possessing antifeedant activity against larvae of the cabbage butterfly (Pieris rapae crucivora) (Matsumoto and Sei, 1987). Roots and stems of G.biloba are also toxic for insects (Major, 1967). Diseases Ginkgo is highly resistant to diseases. Only one phytopathological problem has been encountered in the plantation located near Bordeaux (France), a mould that grows at the bottom of the young plants (O’Reilly, 1993). This disease might be explained by soil fungi which infect ovules left on the ground for too long a time. Embryos grown from ovules taken immediately when they fall from the trees have been cultivated in vitro with no contamination, whereas embryos grown from ovules collected a long time after they have fallen on the ground were contamined with mould despite ovule decontamination (Laurain, 1994). These results show how important it is to harvest healthy ovules. Ovule soil contamination would affect the germination rate and the development of healthy young plants. Death of Ginkgo biloba seedlings can be caused by some fungi (Fusarium sp., Macrophomina phaseoli) which generate root and stem rot (Zhiquan et al., 1991). On the other hand Ginkgo, like the tree Eleagnus, appears to be resistant even after inoculation with the fungus Verticillium dahliae (Smith and Neely, 1979). The resistance of G.biloba to fungi could be explained by the 2-hexenal isolated from the steam distillate of its leaves that had an antifungal activity in low concentrations (Major et al., 1960). But the unusual resistance of Ginkgo leaves to fungi cannot be explained completely on the basis of its 2-hexenal content (Major, 1967). Other investigations have shown that a waxy material in the cuticle of leaves reduced spore germination, inhibited germ-tube growth of certain fungi so that fungi did not penetrate the cuticle (Christensen, 1972a). However while the leaves did not

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

74

DOMINIQUE LAURAIN

appear to be damaged, the chemical nature of the epidermal wall of the leaves was changed (Adams et al., 1962). Christensen (1972b) reported a phytoalexin production in Ginkgo leaves induced by Botrytis allii which selectively inhibited the process of cuticle penetration while allowing germination and growth of the fungi. In addition the Ginkgo tree is remarkably free of bacteria and virus attack. The acidity of the leaves could explain the resistance to bacteria like Escherichia coli, Pseudomonas phaseolicola, Xanthom*onas phaseoli and Bacillus pumilus (Major, 1967). A study had shown that extracts of the root of Ginkgo trees inhibited the growth of bean mosaic virus and tobacco mosaic virus (Major, 1967). HARVESTING OF LEAVES AND QUALITY Leaves are manually picked from trees. Concerning the few plantations specially established for the industrial culture of Ginkgo biloba, leaves are harvested with a modified cotton picker (O’Reilly, 1993). In France leaves are harvested in September/ October before the first frost, which is responsible for the leaf fall, and before they turn yellow. Then they contain a maximum of active secondary metabolites. After the harvesting leaves have to be air dried within 24 hours in order to assure a good preservation. Leaves can be dried by mixing during a few minutes in propane heated rotary drum dryers. The moisture content of the leaves prior to drying is 75% and less than 10% after drying. 3.6kg of green leaves correspond to 1kg of dried leaves. Dried leaves are baled and can be stocked during months. The production of leaves from 4 to 5 years old plantations after transplanting is either 12–16 tons of fresh leaves per hectare or about 4 tons of dried leaves per hectare (Balz, 1997). The price of one kg of Ginkgo biloba dried leaves can reach 14 dollars in the medicinal plant trade. Leaf quality varies according to harvest time. Van Beek et al. (1991) showed that the time of harvest plays an important role in the yield of terpenoids. The percentages of bilobalide and ginkgolides in Dutch Ginkgo leaves were lowest in spring and then gradually increased until a maximum was reached in late summer or early autumn (Fig. 2A, 2B). The concentration declined until leaf fall (van Beek and Lelyveld, 1992). The flavonol content is also seasonal with a highest flavonol glycoside content in the leaves in spring, decreasing throughout the season to a minimum before the leaves turn yellow (Fig. 2C) and then increasing after yellowing (Lobstein et al., 1991; Sticher, 1993). The soil type and the environmental conditions in which the tree is grown influence the flavonol content too (O’Reilly, 1993). The age of the tree (van Beek and Lelyveld, 1992) and genetic diversity, also influence leaf quality. The content of flavonoids in leaves decreases with an increase of the age of the tree (Fusheng, 1997). Leaves harvested in September 1989 from a Dutch and a German Ginkgo showed a percentage of 0.196% and 0.032% terpene trilactones respectively (van Beek et al., 1991). A ginkgolide content of 0.154, 0.244 and 0.136% has been obtained following different cultivation methods experimented in a plantation of Ginkgo in Champaubert (France) (Laurain, unpublished results). Moreover terpenoids have also been found in non-lignified shoots and in woody shoots two to four years old in variable amounts:

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

75

Figure 2 Relation between harvest time of leaves and bilobalide (A), ginkgolides (B) and flavonol glycoside (C) contents of female Ginkgo trees (van Beek and Lelyveld, 1992; Sticher, 1993).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

76

DOMINIQUE LAURAIN

0.058% or 0.148% terpenoids in fresh shoots and 0.044% or 0.109% terpenoids in woody shoots according to the method of cultivation used. The nutritional quality of Ginkgo leaves has been analysed also. This study has been realised by the Laboratory of Nutrition and Feeding—National Veterinary School of Alfort (France) on dried leaves of Ginkgo biloba, harvested in June from trees growing in Champaubert (France). The dry matter, cellulose, ash, calcium and phosphorus content were analysed and were found to be 93.2, 11, 6.5, 1.8 and 0.5% respectively. These results showed a good nutritional value near that of wheat bran with a better Ca/P ratio.

CONCLUSION Nowadays, Ginkgo biloba is an ornamental tree which can be found in numerous cities throughout the world and in recent decades the therapeutic use of its leaves in the West has given rise to the establishment of a few large-scale Ginkgo plantations (France, South Carolina and China). A report by a London-based management consultancy reports that some medicinal plants like Ginkgo biloba are species in urgent need of conservation because of the growing popularity of herbal medicines (Masood, 1997). In fact, demand of Ginkgo leaves in the world is increasing at 26 per cent a year. So urgent measures are needed to cultivate Ginkgo biloba at a large scale. Trees of plantations established with the aim to harvest leaves are grown from ovules. On the other hand in horticulture, vegetative propagation by grafting is principally employed for obtaining on the one hand cultivars possessing desired ornamental characteristics and on the other hand for propagating only male trees, thus eliminating annoyances caused by ovules of female Ginkgos. The major problem in relation with Ginkgo multiplication by sowing is the diversity observed at the vegetative development level of young plants as well as at the level of their secondary metabolite content. Grafting could lead to a plantation made of hom*ogeneous plants but it is a time-consuming technique and more expensive to achieve for millions of trees than multiplication by sowing. The first studies made on cultivation methods on a Ginkgo plantation (Laurain, unpublished) could solve the problem of secondary metabolite variability. They have shown an influence on ginkgolide content. Such studies have to be continued because they could lead to plantations with Ginkgo biloba trees, obtained by sowing, rich in active compounds.

ACKNOWLEDGEMENT I would like to thank Mr. R.Laurain, Dr. J.Mattar, Prof. A.Jacquin and Dr. F.Beaujard for information and advice given during the preparation of this paper and Mrs. M.Pelabon for the translation into English.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

77

REFERENCES Adams, P.B., T.Sproston, H.Tietz and R.T.Major (1962). Phytopathol., 52, 233. Cited from: Major, R.T. (1967). The Ginkgo, the most ancient living tree. Science, 157, 1270–1273. Balz, J.P. (1997). Agronomic aspects of Ginkgo biloba leaves production. Proceedings of ’97 International Seminar on Ginkgo, November 10–12, Beijing, China, 101–104. Beek, T.A. van and Lelyveld, G.P. (1992). Concentration of ginkgolides and bilobalide in Ginkgo biloba leaves in relation to the time of year. Planta Med., 58, 413–416. Beek, T.A. van, Scheeren, H.A., Rantio, T., Melger, W.Ch. and Lelyveld, G.P. (1991). Determination of ginkgolides and bilobalide in Ginkgo biloba leaves and phytopharmaceuticals. J. Chromatogr., 543, 375–387. Bevan, C.W.L., Birch, A.J. and Caswell, H. (1961). An insect repellant from black co*cktail ants. J. Chem. Soc., 96, 488. Brinkman, J.A. and Boerner, R.E.J. (1994). Nitrogen fertilization effects on foliar nutrient dynamics and autumnal resorption in maidenhair tree (Ginkgo biloba L.). J. Plant Nutr., 17, 433–443. Camefort, H. and Boue, H. (1969). Reproduction et biologie des principaux groupes végétaux. Les Cormophytes ou Archégoniates. Edition Doin, 45–50. Chen 1933 cited from: Gaussen, H. (1946). Les Ginkgoales. In: Gaussen H. (ed.). Les gymnospermes actuelles et fossiles. Fac. Sci., Toulouse, 1–54. Christensen, T.G. (1972a). Phytoalexin production in Ginkgo biloba in relation to inhibition of fungal penetration. Phytopathol., 62, 493–494. Christensen, T.G. (1972b). Resistance of Ginkgo to fungi. Phytoalexin production induced by Botrytis allii. Ciss. Abstr. Int. B., 32, 4340. Collins, G.N. (1903). Dimorphism in the shoots of the Ginkgo. The Plant World, 6, 9–11. C.N.I.H. (1985). Conseils pratiques pour le désherbage des pépinières. Comité National Interprofessionnel de l’Horticulture (ed.), Paris, 159 p. Corey, E.J., M.C.Kang, N.C.Desai, A.K.Ghosh and I.N.Houpis. (1988). Total synthesis of (+)ginkgolide B. J. Am. Chem. Soc., 110, 649–651. Critchfield, W.B. (1970). Shoot growth and heterophylly in Ginkgo biloba. Bot. Gaz., 131, 2, 150–162. De Feudis, F.V. (1991). Ginkgo biloba extract EGb 761: Pharmacological activities and clinical applications. Elsevier, Paris. Del Tredici, P. (1991). The architecture of Ginkgo biloba L. L’Arbre. Biologie et développement. C.Edelin ed., Naturalia Monspeliensia, 155–167. Del Tredici, P. (1992). Natural regeneration of Ginkgo biloba from downward growing cotyledonary buds (basal chichi). Am. J. Botany, 79, 5, 522–530. Doran, W.L. (1954). The vegetative propagation of Ginkgo. J. For., 52, 176–177. Edelin, C. (1986). Stratégic de réitération et édification de la cime chez les conifères. Naturalia Monspeliensia. Colloque International sur l’Arbre. 139–158. Favre-Duchartre, M. (1956). Contribution à l’étude de la reproduction chez le Ginkgo biloba. Rev. Cytol. Biol. Veg., XVII, 1–2, 214 p. Flesch, V., Jacques, M., Cosson, L., Pétiard, V. and Balz, J.P. (1991). Influences de l’éclairement et de la température sur divers paramètres de croissance de jeunes Ginkgo biloba L. cultivés en conditions contrôlées de jours longs. Ann. Sci. For., 48, 133–147. Franklin, A.H. (1959). Ginkgo biloba L.: Historical summary and bibliography. Virginia J. Sci. 10, 131–176. Fusheng, P., Wangou, L., Fugui, H., Shanan, H. and Zijie, P. (1997). Study on the optimizing of the resources of the Ginkgo biloba leaves. Proceedings of ’97 International Seminar on Ginkgo, November 10–12, Beijing, China, pp. 105–112. Gunckel, J.E. and Thimann, K.V. (1949). Studies of development in long shoots and short

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

78

DOMINIQUE LAURAIN

shoots of Ginkgo biloba. III. Auxin production in shoot growth. Am. J. Bot., 36, 145– 151. Gunckel, J.E. and Wetmore, R.H. (1946). Studies of development in long shoots and short shoots of Ginkgo biloba L. II. Phyllotaxis and the organization of the primary vascular system, primary phloem and primary xylem. Am. J. Bot., 33, 532–543. Halle, F. and Oldeman, R. (1970). Essai sur l’architecture et la dynamique de croissance des arbres tropicaux. Masson, Paris. Hu, Y.H. (1987). Chinese Penjing. Timber Press, Portland, Oregon. Huh, H. and Staba, E.J. (1992). The botany and chemistry of Ginkgo biloba L. J. Herbs Spices Med. Plants, 1, 91–124. Krussmann, G. (1968). La pépinière. Tome 2: Organisation des exploitations. La Maison Rustique, Paris. Laurain, D. (1990). Etude de la croissance, de la multiplication végétative et de la nutrition minérale chez le Ginkgo biloba. Mémoire de D.E.A., Station d’Agronomie, INRA Angers. Laurain, D. (1994). Etablissem*nt de cultures cellulaires haploïdes, diploïdes et transformées de Ginkgo biloba par diverses méthodes variabilisantes. Thèse de Doctorat, Université Fraçois Rabelais, Tours, France. Laurain, D., Chénieux, J.C. and Trémouillaux-Guiller, J. (1996). Somatic embryogenesis from immature zygotic embryos of Ginkgo biloba. Plant Cell Tissue Organ Culture, 44, 19–24. Li, H.L. (1956). A horticultural and botanical history of Ginkgo. Bull. Morris Arb., 7, 3–12. Li, Z. and Lin, J. (1990). Anatomical studies of the “burls” of Ginkgo (abstract). I.A.W.A. Bull., 2, 129. Lobstein, A., Rietsch-Jako, L., Haag-Berrurier, M. and Anton, R. (1991). Seasonal variations of the flavonoid content from Ginkgo biloba leaves. Planta Med., 57, 430–433. Major, R.T. (1967). The Ginkgo, the most ancient living tree. Science, 157, 1270–1273. Major, R.T., Marchini, P. and Boulton, A.J. (1963). Observations on the production of ahexenal by leaves of certain plants. J. Biol. Chem., 238, 1813–1816. Major, R.T., Marchini, P. and Sproston, T. (1960). Isolation from Ginkgo biloba of an inhibitor of fungus growth. J. Biol. Chem., 235, 3298–3299. Major, R.T. and Tietz, H.J. (1962). Modification of the resistance of Ginkgo biloba leaves to attack by Japanese beetles. J. Econ. Entomol., 55, 272. Martinez, M. and Chambon, J.P. (1987). Le point sur quelques ravageurs nouveaux, autochtones ou récemment introduits en France. In Conférence internationale sur les ravageurs en agriculture, T. 1, Paris, ANPP, 9 pp. Masood, E. (1997). Medicinal plants threatened by over-use. Nature, 66, 570. Matsumoto, T. and Sei, T. (1987). Antifeedant activities of Ginkgo biloba L. components against the larva of Pieris rapae crucivora. Agric. Biol. Chem., 51, 249–250. Michel, P.F. (1985). Ginkgo biloba-L’arbre qui a vaincu le temps. Editions du Félin, Paris. Nozeran, R. (1986). Le mouvement morphogénétique spécialement chez les végétaux supérieurs pérennes. Naturalia Monspeliensia. Colloque International sur l’Arbre. 415–428. O’Reilly, J. (1993). Ginkgo biloba—Cultivation, extraction, and therapeutic use of the extract. In T.A.van Beek and H.Breteler, (eds.) Phytochemistry and Agriculture. Clarendon Press Oxford, pp. 253–270. Rensselaer, M. (1969). The remarkable Ginkgo. Plants Garden, 25, 50–53. Rohr, R. (1989). Maidenhair Tree (Ginkgo biloba L.). In Y.P.S.Bajaj (ed.), Biotechnology in Agriculture and Forestry, Trees II, Springer Verlag, Berlin, 5, 574–590. Santamour, F.S., He, S. and McArdle, A.J. (1983). Checklist of cultivated Ginkgo. J. Arboric., 9, 88–92. Skirvin, R.M. and Chu, M.C. (1979). Ginkgo: a beautiful tree with edible seeds. Res., Univ.: III Agric. Exp. Stn. 21, 10–11. Smith, L.D. and Neely, D. (1979). Relative susceptibility of trees species to Verticillium dahliae. Plant Disease Reporter, 63, 328–332.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

CULTIVATION OF GINKGO BILOBA ON A LARGE SCALE

79

Sticher, O. (1993). Quality of Ginkgo preparations. Planta Med., 59, 2–11. Teuscher, H. (1951). Ginkgo biloba from cuttings. Am. Nurs., 15, 7. Vermeulen, J. (1960). Propagation of Ginkgo biloba by cuttings. Proc. Plant Propag. Soc., 10, 127–130. Wheeler, A.G. (1975). Insect associates of Ginkgo biloba. Entomol. News, 86, 37–44. Zhiquan, Z., Dongrong, J., Guangquan, Z. and Yongmei L. (1991). A study on the causes of death of seedling of Ginkgo biloba. Guihaia, 11, 334–338.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

5. PLANT CELL BIOTECHNOLOGY OF GINKGO DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN1 Department of Agricultural and Bioresource Engineering, University of Saskatchewan, 57 Campus Drive, Saskatoon, Saskatchewan, Canada, S7N 5A9. 1 Faculté de Pharmacie, Laboratoire de Pharmacognosie, 1 rue des Louvels, 80037 Amiens Cédex 1, France.

INTRODUCTION The Ginkgo biloba tree is the sole representative of the Ginkgoales order (Engler, 1954) and its origins trace back to the Jurassic period. Parts of this unique tree have been used as a phytotherapeutic agent since at least 1300 AD (Braquet, 1988; Michel and Hosford, 1988). This tree produces a wide range of secondary metabolites (Boralle et al., 1988), among which, the terpene trilactones ginkgolide A, B, C, J and M further abbreviated as G-A, G-B, G-C, G-J and G-M, respectively. Bilobalide is the major terpene of leaves (Nakanishi et al., 1971); it has also been detected in roots (van Beek, unpublished). Ginkgolides have been detected in roots and leaves (see Nakanishi, 1988); their ecological significance as well as their site of synthesis within the G.biloba tree has yet to be resolved. Seasonal variation of their concentration has been reported (van Beek and Lelyveld, 1992; Flesch et al., 1992). Ginkgolides have significant pharmacological properties (Braquet et al., 1987). Although they have been chemically synthesized (Corey et al., 1988), their commercial supply can only arise from biomass sources. Ginkgolides have been detected in undifferentiated cell cultures (Carrier et al., 1991; Huh and Staba, 1993; Jeon et al., 1995). However, the low amounts of secondary metabolite obtained from tissue culture, from ng gdw-1 (Carrier et al., 1991) to µg gdw-1 (Jeon et al., 1995), preclude the commercial use of such methods when compared to the amounts of ginkgolides measured in the leaves of the tree: mg gdw-1 (Flesch et al., 1992). Cellular differentiation is a factor which influences secondary metabolite synthesis. Galewsky and Nessler (1986) have reported that thebaine was not detected in Papaver somniferum undifferentiated cell cultures, but was detected in their somatic embryos. It is possible that G.biloba embryogenic cell cultures could lead to an increase in secondary metabolite production from such in vitro grown biomass. In additon to knowledge on G.biloba differentiated cell cultures, an understanding of the ginkgolide biosynthetic pathway would, most likely, facilitate their production in an in vitro environment. Through their astute pioneering work, Nakanishi and 81 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

82

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Habaguchi (1971) have established that ginkgolides were of terpenoid nature. Gaining further knowledge about the major biosynthetical steps, from carbohydrates to ginkgolides, would enable the understanding of the parameters affecting and controlling their biosynthesis. Once these controlling steps are understood, productive in vitro cultivation strategies could be contemplated. In this work, a compilation of the most recent results pertaining to differentiated and undifferentiated G.biloba cell cultures is presented. BIOTECHNOLOGICAL APPROACH Cultivation of Undifferentiated Cells Modifications of ammonium/nitrate medium composition ratio Ginkgolides were detected in embryo-derived G.biloba cells cultivated in shake flasks and 6 L bioreactors using Murashige and Skoog (1962) medium (MS) (Carrier et al., 1994). The detected quantities were extremely low, ranging in the ng gdw-1. In an attempt to increase ginkgolide production, modifications were made to MS with respect to the ammonium/nitrate ratio. As a control, cells were first grown for 34 days in MS, where the ammonium/nitrate ratio was 1:2. Nutrient uptake was as previously described (Carrier et al., 1991). The lyophilised cells were analysed, and yielded 14.7 and 12.3 ng gdw-1 of G-A and G-B, respectively. Cells were also cultured in a MS based medium modified with an ammonium/nitrate ratio of 1:1 (omission of KNO3). Cells were cultured for 34 days, and were pale green and sometimes yellowish. Under these conditions, the cells did not entirely consume their extracellular glucose, where 5 g L-1 remained at the end of the cultivation cycle. Of the 20 mM of nitrate initially present, 3 mM were detected after 34 days. No traces of either G-A or G-B were detected in the lyophilised extracted biomass. Cells cultivated in an MS medium with an ammonium/nitrate ratio of 0:1 (omission of NH4NO3) were emerald green. Carbohydrates were entirely consumed by day 32, following the pattern of the control cultures. At the end of the culture period, 3 mM of extracellular nitrate was detected. The ginkgolide content of the lyophilised biomass was 2.6 and 7.2 ng gdw-1 of G-A and G-B, respectively. The highest concentration of ginkgolide was found in cells cultivated in MS. These results agree with those presented by Jeon et al. (1995) where the maximum G-B production was obtained in media with an ammonium/nitrogen ratio of 1:2 and 1:3. Kinetics of cells cultured in Murashige and Skoog medium The kinetics of ginkgolide production in MS in cells cultured in shake flasks and in a 2 L immobilisation bioreactor (Archambault et al., 1990) cultures was evaluated. Cultivation vessels, either shake flasks or bioreactors, were simultaneously inoculated, but dismantled at different times throughout the cultivation period. The lyophilised biomass, either from the shake flask or from the bioreactor cultures, was taken for ginkgolide analysis. Fig. 1 presents biomass and ginkgolide concentrations as well as phosphate consumption of shake flask cultivated cells. Ginkgolides were detected

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

83

concomitantly with dry weight increases. Extracellular phosphate depletion and the onset of ginkgolide biosynthesis occurred simultaneously. The onset of their production began with the depletion of extracellular phosphate, and peaked on day 21. The increase in extracellular phosphate concentrations can most likely be accounted for by release of phosphate during cell lysis. Fig. 2 presents biomass and ginkgolide concentrations as well as phosphate consumption of cells cultivated in 2 L immobilization bioreactors. Bioreactors dismantled on days 20, 22, 27 and 29 yielded biomass concentrations of about 13 gdw L-1 and the maximum biomass production, of about 16 gdw L-1, was obtained on day 31 and on day 34. Ginkgolides were first detected when phosphate was depleted, corresponding to the bioreactor dismantled on day 22. This possibly indicates that production of ginkgolides may commence when cells cease their division and initiate their increases in mass. Results presented by Jeon et al. (1995) also show that the maximum amount of ginkgolides produced corresponded to the middle of the exponential growth phase. Pépin et al. (1995) demonstrated that cultured plant cells undergo two distinct types of growth: a period of high cell division followed by a period of cell mass increase. The combined results of Carrier (1992), Pépin et al. (1995) and Jeon et al. (1995) show that there is a possible link between the offset of the high cell division period and ginkgolide production. When phosphate is depleted from the cultivation medium, cells may halt division and initiate secondary metabolite production. Large scale undifferentiated Ginkgo cell cultures are not the method of choice to industrially produce ginkgolides. However, differentiated cultures may possibly be more amenable to such practices. Nakanishi and Habaguchi (1971) have detected ginkgolides in zygotic G.biloba embryos; hence, embryogenic Ginkgo cell cultures may yield higher ginkgolide concentrations than undifferentiated cell cultures. Ginkgo biloba Embryogenesis Ginkgo biloba haploid tissue culture was first reported by Tulecke in the early 1950’s (Tulecke, 1997, and references therein). Yates (1986) investigated Ginkgo embryogenesis from mature zygotic embryos, and was successful in obtaining callus formation. Laurain et al. (1993a, 1993b, and 1996) obtained embryogenesis in Ginkgo from male (1993a) and female (1993b) tissues on one hand and from immature zygotic embryos on the other hand (1996). Variabilising methodologies, such as cellular clonings and methodologies capable of leading to embryogenesis, have been used and are presented below. Type of explants leading to an embryogenesis As research on somatic embryogenesis progressed, it became apparent that immature zygotic embryos seemed to be most suitable for initiating embryogenic cultures in both angiosperms (Vasil, 1987) and gymnosperms (Attree and Fowke, 1993). Recently, immature Ginkgo zygotic embryos were shown to be suitable for initiating somatic embryogenesis (Laurain et al., 1996). It was determined that the stage of development of the immature zygotic embryo influenced the rate of induction of embryogenic lines.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

84

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Figure 1 Embryo-Derived G.biloba Cells Cultivated in Shake Flasks. A. Biomass (䉬) and phosphate (䊏). B. G-A (䊐) and G-B (䊏).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

85

Figure 2 Embryo-Derived G.biloba Cells Cultivated in a 2 L Immobilization Bioreactor. A. Biomass (䉬) and phosphate (䊏). B. G-A (䊐) and G-B (䊏).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

86

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

The four stages of zygotic embryo development are: globular, torpedo, precotyledonary and early cotyledonary. Of these four developmental periods, the precotyledonary and early cotyledonary stages, characterized by the initiation and the development of cotyledons, respectively, were suitable for the initiation of embryogenic tissue. As expected, cotyledonary stage embryos were longer than precotyledonary stage embryos with lengths of 3.6mm and 2.2mm, respectively. Globular, torpedo, precotyledonary and cotyledonary stage embryos led to cellular proliferations of 5%, 69%, 98% and 99%, respectively. These frequencies included both embryogenic tissue and callus, of which the proportions varied according to the hormonal composition of the induction media. These results indicated that young cotyledonary stage embryos were found to be most appropriate for induction of G.biloba embryogenic tissues. Similar results were obtained for gymnosperms: Picea glauca (Lu and Thorpe, 1987), Pinus elliottii (Jain et al., 1989), Larix occidentalis (Thompson and von Aderkas, 1992), and Pinus patula (Jones et al., 1993). Interest in embryogenic callus or direct embryogenesis, obtained from haploid tissue, is justified by the production of hom*ozygous plants and by the isolation of mutants producing large quantities of medicinal substances (Sangwan-Norreel et al., 1986). In Ginkgo, direct embryogenesis has been obtained from both male and female gametophytes. The rate of induction of haploid embryogenic tissue was limited by the degree of maturation of the explant material. For male Ginkgo gametophytic material, only microspores isolated at the uninucleate stage lead to embryogenesis (Laurain et al., 1993a), forming embryogenic clusters that enabled the establishment of a pure male cell line (Trémouillaux-Guiller et al., 1996). In Graminae, the developmental stage of the explant was also related to the success rate of the establishment of embryogenic clusters (Coumans et al., 1989). For the embryogenesis derived from female Ginkgo (Laurain et al., 1993b), the prothallus used was characterized by a large cavity, which occupied its centre. At this stage of development, the prothallus was either at the coenocytic-stage or at the early septation-stage (Favre-Duchartre, 1956). Only such Ginkgo prothallus could produce haploid protoplasts, characterized by an ovoid hollow bag reaching a length of 0.5 to 3mm. In France, ovules presenting such prothallus can be harvested from mid May to the end of June. Somatic embryogenesis Immature zygotic embryos were cultured on solid Murashige and Tücker (1969) medium prepared either at half strength (MT/2) or at full strength (MT) (Laurain et al., 1996). The induction of embryogenic tissue was obtained by supplementing MT/ 2 or MT with either solely 10µM Benzylaminopurine (BA) or with 5µM (BA) combined with 5, 10 or 20µM Naphthaleneacetic acid (NAA) (Table 1). After two weeks of culture, direct embryogenesis was observed on hypertrophic cotyledons cultivated on MT or MT/2 induction medium supplemented with solely 10µM of BA. The latter combination yielded, on average, 9.6 somatic embryos per explant. This was the highest quantity of embryos generated in this study (Table 1). Unfortunately,

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

87

Table 1 Induction media used for obtaining somatic embryogenesis from Ginkgo biloba immature zygotic embryos (Laurain et al., 1996).

MT: Murashige and Tücker (1969); MT/2: half strength mineral MT medium; BA: Benzylaminopurine; NAA: Naphthaleneacetic acid. Numbers in parenthesis indicate the µM concentration of the growth regulator.

direct embryogenesis was not observed for the majority of treatments; and, transfers on fresh MT/2 and MT media, devoid of or supplemented with 10µM of BA, were necessary to obtain somatic embryos via indirect embryogenesis. Such embryos developed to the cotyledonary stage; however, they were unable to complete their development when isolated from their tissue of origin, and plated on MT medium devoid of growth regulators. As for the developing embryos, cultures could only be maintained for six months; after this time, they showed browning and necrosed cells were observed. These results showed that immature zygotic G.biloba embryos could be induced to produce embryogenic material, using MT medium supplemented with solely BA. Cytokinins were also instrumental in inducing embryogenesis in the Abies genera (Schuller et al., 1989; Norgaard and Krogstrup, 1991). In contrast to Abies, some varieties of woody species required exogenous auxins for the induction of somatic embryogenesis (Lu and Thorpe, 1987; von Arnold, 1987; Woods et al., 1992). For Ginkgo, direct or indirect embryogenesis was observed depending on the nature of the exogenous growth regulators used in the induction medium. On hypertrophic cotyledons, direct embryogenesis was induced only in the presence of BA, while indirect

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

88

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

embryogenesis seemed to be correlated with the presence of exogenous auxins. Embryogenic induction frequencies in the order of 90–95% were obtained, showing that Ginkgo is amendable to this technique. These results were high in comparison to those reported for Pinus patula (2.6%) (Jones et al., 1993), Larix occidentalis (3%) (Thompson and von Aderkas, 1992), and Pinus purgeus (16%) (Afele et al., 1992). Gametophytic embryogenesis Cell cultures stemming from both isolated uninucleate microspores and from protoplasts, isolated either from female prothallus or from microspore-derived cell lines, were amendable to the formation of embryos via direct embryogenesis. G.biloba uninucleate microspores and protoplasts, isolated from female gametophytes, were cultured in Bourgin and Nitsch (1967) (BN) liquid medium devoid of growth regulators, but supplemented with coconut milk. The microspores were also cultured in BN medium supplemented with coconut milk and with Indoleacetic acid (IAA) and Kinetin (KIN). Protoplasts, isolated from female gametophytes, were also cultured in MT liquid medium devoid of ammonium ions; and supplemented with glutamine, BA and NAA. Protoplasts isolated from a microspore-derived cell line were cultured in either BN, MT/2, or Gamborg et al. (1976) (B5) supplemented with combinations of NAA, IAA, KIN, BA or Dichlorophenoxyacetic acid (2, 4-D). The reader is referred to Tremouillaux-Guiller et al. (1996) and Laurain et al. (1993b) for exact media compositions. Microspores as well as protoplasts isolated from female prothallus and from microspore-derived cell suspensions exhibited various modes of development before showing embryogenesis. In these three types of cultures, the gametophytic embryos were produced either from a unicellular microclone by endomitosis or from a microclone by typical divisions. A particularity, observed during embryogenesis from isolated microspores, was the formation of fourfold asymmetrically celled microspores. These asymmetrical divisions were comparable to those observed during the in vivo development of the pollen grain with the II prothallial cell, reproductive cell and germinal cell formation, while the I prothallial cell rapidly degenerated. The germinal cell increased in volume and in length, developing a pollen tube of variable length (up to 480 micron long), and occasionally provided rhizoid structures. Similarly, microspore-derived protoplasts produced long cytoplasmic extensions comparable to pollen tubes at the beginning the of culture cycle (Trémouillaux-Guiller et al., 1996). The celled microspores could evolve to embryogenesis by ejecting enveloped nuclei into the extracellular medium through a phenomenon that is called inverted endocytosis (Laurain et al., 1993b). Nuclei expulsions, producing small cells leading to microclones, have also been observed during cultivation of female haploid protoplast and microspore-derived cell suspension protoplasts. In the presence or in the absence of exogenous growth regulators, microspores, female protoplasts and microspore-derived protoplasts isolated from G.biloba divided and formed microclones, which directly evolved into embryos. The formation of pro-embryos, from microspore cells cultivated for four weeks, was observed in the presence or the absence of growth regulators. After ten weeks of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

89

culture, embryogenesis was observed with an efficiency factor (defined as the number of embryos divided by the number of microspores originally plated) of 1.8%. After cultivation for a period of four to five months in liquid media, slow growing embryoclusters and embryos, isolated at different stages of development, were transferred onto BN solid media devoid of growth regulators or containing various combinations of KIN and IAA. Embryo-clusters showed signs of growth particularly when cultured on BN solid medium supplemented with 11.4µM KIN; after one month, some of these embryo-clusters developed, exhibiting a club shape. Cell suspensions, established from these embryogenic clusters, produced embryogenic cells in MT liquid medium devoid of growth regulators or supplemented with 18.8µM NAA (TrémouillauxGuiller et al., 1996). Protoplasts were isolated from a six-day-old subculture of these cell suspensions. After two weeks of culture, microcolonies derived pro-embryos were obtained in the media preparations shown in Table 2 (MT, B5, and BN liquid media devoid of growth regulators or supplemented with various combinations of growth regulators). After four weeks of cultivation, the efficiency of pro-embryo and embryo formation ranged from 0.6% to 4%, depending on the media tested (Table 2). Table 2 Efficiency of embryo formation obtained from female protoplast culture, protoplasts isolated from a microspore-derived cell suspension and microspore culture of Ginkgo biloba (Laurain et al. (1993b); Trémouillaux-Guiller et al. (1996)).

MT: Murashige and Tücker (1969); MT’: Murashige and Tücker (1969) basal medium modified by omitting NH4NO3; BN: Bourgin and Nitsch (1967); B5: Gamborg et al. (1976). The concentration of growth regulator is given in µM. MT (1): NAA (10.74); MT (2): NAA (10.74), KIN (0.93); MT (3): NAA (2.69), BA (8.87); MT (4): BA (8.87); MT (5): IAA (11.42), KIN (0.93); BN (2): IAA (11.42), KIN (0.46); BN (3): IAA (17.13), KIN (0.46); B5 (1) 2,4-D (0.45); MT’ (1): NAA (26.85), BA (4.44); MT’ (2): NAA (16.11), BA (4.44); MT’ (3): NAA (10.74), BA (4.44); MT’ (4): NAA (5.37), BA (4.44).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

90

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Between day 45 and day 60 of female protoplast culture, pro-embryos and embryos appeared in every Petri dish. Embryogenesis was observed with frequencies between 0.7% to 1.9% when using modified MT (without ammonium) liquid medium and of 0.2% when using BN liquid medium (Table 2). Most embryos had reached the oblong stage while some heart-shaped embryos were also present in the cultures. After three months, the number of embryos ranged from 165 to 1900 embryos mL-1 depending on the culture medium used and decreased to 50 to 470 embryos mL-1 one month later (Laurain et al., 1993b). The decline in embryo number could be related to browning which appeared after the addition of fresh medium. However, after five months of culture, a supply of fresh medium was no longer detrimental to the embryos. As for the embryos derived from the microspore cultures, embryos obtained from protoplasts of male and female gametophytes exhibited slow growth, rendering the transfer to solid media impossible. These results show that isolated microspores and protoplasts isolated from both microspore-derived cell suspensions and from female prothallus did not require exogenously supplemented growth regulators for their culture. G.biloba female protoplasts produced new cell walls and divided in BN medium supplemented solely with coconut milk. G.biloba cell suspension-derived protoplasts divided in BN, MT and B5 with or without growth regulators. This result contrasts with the work of Zrÿd (1988) and references therein, but is similar to that observed by Bekkaoui et al. (1987) and Trémouillaux-Guiller et al. (1987). Growth in media devoid of growth regulators may be due to the high endogenous levels of growth regulators. Similar results were observed during Zea mays androgenesis (Mitchell and Petolino, 1991; Coumans et al., 1989), and during Brassica campestris microspore-derived embryogenesis (Sato et al., 1989). It is possible, however, that the coconut milk, supplemented to the BN medium, contained cytokinins which acted as a supply, and contributed to the initiation of embryogenesis in G.biloba microspore and female prothallus protoplasts cell lines. Sangwan-Norreel et al. (1986) have reported that low levels of growth regulators, such as in coconut milk, stimulated increases in the number of gametophytic plants. This contrasts with Keller and Stringam (1978) who have shown that cytokinins were essential for maximum microspore response in Datura and Solanum. It is well known that the application of electrical currents stimulates cellular division and differentiation, in particular for protoplasts of woody species regarded as recalcitrant to culture (Chand et al., 1988). Combinations of electric field (50 to 1000V cm-1), pulse-duration (20, 30 and 100µs) and pulse-number (1 or 3 pulses) were applied to G.biloba microspores (Laurain et al., 1993a). The cells survived these high voltage treatments. After 24 hours of culture, the viability of the control and of the electrostimulated microspores was similar with values of 41%. Electrostimulated microspore cells showed embryogenesis earlier than that of the control. After four months of culture, embryos appeared in the electrostimulated microspore cultures, while only pro-embryos were present in the control cultures. These results showed that electrical pulses applied to G.biloba microspores cultures stimulated differentiation similar to the application of electrical currents on protoplasts (Davey and Power, 1987).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

91

Ginkgolide content of various cell suspensions Recently, Laurain et al. (1997) have established cell cultures from various G.biloba explants, of which, two were found to contain ginkgolides. The first one was established from roots, which developed from zygotic embryos inoculated with Agrobacterium rhizogenes A4. The second cell suspension was derived from female prothallus. The combined amounts of bilobalide, G-A, G-B, G-C and G-J were determined to be 650µg gdw-1 in the putatively transformed suspension culture and 870µg gdw-1 in the female prothallus derived suspension. These results indicated that synthesis of bilobalide and ginkgolide in cell cultures is dependent on the nature of the initial explant, and that transformation with A. rhizogenes A4 is favourable for the production of G.biloba terpene trilactones. Shunan et al. (1997) reported the appearance of hairy roots on Ginkgo leaves infected with A. rhizogenes A4. Terpene trilactones were not quantified; however, these cultures were successfully grown in 2.5 L to 5 L flasks, indicating scale-up possibilities. Ginkgolides and Bilobalide Ginkgolide and bilobalide biosynthesis Through their astute pioneering work, Nakanishi and Habaguchi (1971) established that ginkgolides were of terpenoid nature, assuming that they were formed following the classical acetate-pathway, namely through the biosynthesis of mevalonic acid and the primary terpenes (geranyl (GPP), farnesyl (FPP) and geranylgeranyl pyrophosphate (GGPP)). With the work of Schwarz (1994), important advances were made in the elucidation of the biosynthetical pathway of ginkgolide and bilobalide. This work has shown that, in G.biloba, parallel and mechanistically distinct biosynthetic pathways govern the formation of IPP. The IPP units required for the formation of sitosterol would be formed from the acetate-mevalonic acid steps, occurring in the cytosol. Surprisingly, ginkgolides and bilobalide would be synthesized from IPP units arising from a new pathway, namely the “triose/pyruvate”-pathway, which would possibly be located in the plastid. Schwarz (1994) suggested that bilobalide would not be a genuine sesquiterpene, but would be derived from the ginkgolide biosynthetical pathway. Carrier et al. (1996) reported that undifferentiated G.biloba cell cultures showed mainly FPP synthase activity. These results combined with those of Schwarz (1994) possibly suggest that the plastid “triose/pyruvate” pathway is not very active in undifferentiated cultured cells, justifying the minute ginkgolide concentrations detected in such cultures (Carrier et al., 1991). Recently, Neau et al. (1997), pursuing the work of Schwarz (1994), have elegantly shown, with the use of inhibitors of cytochrome P-450 dependent oxygenases, that ginkgolide and bilobalide biosynthesis proceeds via dehydroabietane. A precursor product relationship would exist between dehydroabietane and ginkgolides, including bilobalide. For the most recent schematics of the biosynthetic pathway of ginkgolides and bilobalide, the reader is referred to Neau et al. (1997). Note added in proof: very recently an updated article on ginkgolide biosynthesis appeared (Schwarz and Arigoni, 1999).

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

92

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Site of biosynthesis of ginkgolides and bilobalide Analysis of aerial portions of G.biloba plants revealed that ginkgolides and bilobalide were present in equal concentration ratios. Of the ginkgolides, G-A occurred at the highest concentration, followed by relatively equal amounts of G-B and G-C (Flesch et al., 1992; van Beek and Lelyveld, 1992; Huh and Staba, 1993; Cartayrade et al., 1997; Carrier et al., 1998). Interestingly, the underground portions of the plant, were practically free of bilobalide, and yielded in decreasing order of concentration, G-A, G-C, G-B, and G-J (Flesch et al., 1992; Carrier et al., 1998) (Table 3). This contrasted with G-C being detected as the most abundant ginkgolide in root bark of Gingko trees with trunks exceeding 30 cm in diameter (Nakanishi, 1988). The age of the plants extracted could possibly account for the differences in the ginkgolide concentration distribution patterns. In an attempt to shed light on the possible site of ginkgolide biosynthesis, cell free extracts prepared from stem and root bark, root and root meristem, as well as terminal buds, rosettes and side branches of three year old G.biloba plants were prepared (Carrier et al., 1998) (Table 4). The formation of farnesyl and geranylgeranyl pyrophosphate, were monitored by the incorporation of [1– 14C]-isopentenyl pyrophosphate ([1–14C]-IPP) by these cell free extracts. FPP and GGPP were detected as the corresponding alcohols: farnesol and geranylgeraniol, respectively. Stem bark, terminal leaf as well as root and root meristem cell free extract preparations did incorporate the labeled substrate into terpenoid product(s). However, rosette leaf cell free extracts incubated with [1–14C]-IPP did not yield terpenoid products (Carrier et al., 1998). Extracts from the terminal bud region contained high terpene concentrations and showed GGPP and FPP synthase activity. The production of GGPP in these growing tissues may be necessary for biosynthesis, of among others, gibberellic acid, carotenoids, the phytol chain of chlorophyll and diterpenes (Gray, 1987; Kleinig, 1989). Similarly, enzymatic activity of FPP synthase may be necessary for the synthesis of sterol derived constituents. Interestingly, the concentration of Ginkgo terpene trilactones in leaves has been documented to increase throughout the summer season (Flesch et al., 1992; van Beek and Lelyveld, 1992; Huh and Staba, 1993). In the work reported by Carrier et al. (1998), rosette leaves contained high concentrations of terpene trilactones, but yielded crude enzyme preparations incapable of incorporation of the labeled substrate. Results obtained with Capsicum (Kutz et al., 1992) have shown that the activity of GGPP synthase and its mRNA expression varied as a function of plant development. Carrier et al. (1998) suggested that the high terpene concentration in rosette leaves could possibly be explained by the following or a combination of the following: (a) synthesized elsewhere in the plant and transported, (b) synthesized when the tissues were actively growing and then sequestered or (c) obtained from the catabolism of other moieties also derived from GGPP. Recently, Cartayrade et al. (1997), through intricate labeling experiments, shed light on the site of ginkgolide synthesis within the G.biloba plant. In distinct 14CO2 and (U-14C) glucose labeling experiments, labeled ginkgolides were detected in roots before being traced in the stems and the leaves.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

93

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

Table 3 Bilobalide, G-A, G-B, G-C and G-J and total terpene content with standard deviation of various parts obtained from 3 year old G.biloba plants. Contents are expressed as mg terpene 100mg-1 dry wt (Carrier et al., 1998). Numbers in italic represent standard deviations.

94

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Table 4 Formation of terpenoid products, monitored as the corresponding alcohols (Farnesol and Geranylgeraniol) in cell free extracts prepared from various parts obtained from 3 year old Ginkgo biloba plants (Carrier et al., 1998).

Numbers represent the peak area (counts per minute (CPM)/mg of protein). Numbers in italic represent standard deviations. IPOH=isopentenol, GGOH=geranylgeraniol, FOH=farnesol.

Examination of the distribution pattern of the label, within a 27 day time frame, indicated that the terpenes were translocated from the roots to the stems and the leaves. Cartayrade et al. (1997) concluded that the aerial portion of the plant acted as a ginkgolide sink for the diterpene producing underground portions. CONCLUSIONS AND PERSPECTIVES In conclusion, undifferentiated G.biloba cell cultures can easily be scaled-up from shake flask to bioreactors; however, they produce trace amounts of ginkgolides, and are uninteresting from an industrial perspective. Higher ginkgolide yields can be obtained from differentiated cultures. The knowledge of the ginkgolide site of synthesis, being that of the root, combined to the expertise in differentiated cell culture techniques leads to interesting and exciting industrial prospects. However, many barriers remain to be crossed before this becomes reality. In the short term it is now imperative to further investigate cultivation protocols that pertain with root development. REFERENCES Afele, J.C., Senaratna, T., McKersie, B.D., Saxena, P.K. (1992) Somatic embryogenesis and plant regeneration from zygotic embryo culture in blue spruce (Picea pungens Engelman). Plant Cell Rep., 11, 299–303. Archambault, J., Volesky, B., Kurz, W. (1990) Development of bioreactors for the culture of surface immobilized plant cells. Biotechnol. Bioeng., 35, 702–711. von Arnold, S. (1987) Improved efficiency of somatic embryogenesis in mature embryos of Picea abies L. Karst. J. Plant Physiol., 128, 233–244. Attree, S.M., Fowke, L.C. (1993) Embryogeny of gymnosperms: advances in synthetic seed technology of conifers. Plant Cell Tissue Organ Cult., 35, 1–35. van Beek, T.A., Lelyveld, G.P. (1992) Concentration of ginkgolides and bilobalide in Ginkgo biloba leaves in relation to the time of year. Planta Med., 58, 413–416.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

95

Bekkaoui, F., Saxena, P.K., Attree, S.M., Fowke, L.C., Dunstan, D.I. (1987) The isolation and culture of the protoplasts from an embryogenic cell suspension culture of Picea glauca (Moench) Voss. Plant Cell Rep., 6, 476–479. Boralle, N., Braquet and P., Gottlieb, O. (1988) Ginkgo biloba; a review of its chemical composition. In P.Braquet (ed.), Ginkgolides—Chemistry, Biology, Pharmacology and Clinical Perspectives, Volume 1, J.R.Prous Science Publishers, Barcelona, pp. 9–26. Bourgin, J.P., Nitsch, J.P. (1967) Obtention de Nicotiana haploïdes à partir d’étamines cultivées in vitro. Ann. Physiol. Vég., 9, 377–382. Braquet, P. (1988) The ginkgolides. From Chinese pharmacopeia to a new class of pharmacological agents: The antagonists of platelet-activating factor. In P.Braquet (ed.), Ginkgolides—Chemistry, Biology, Pharmacology and Clinical Perspectives, Volume 1, J.R.Prous Science Publishers, Barcelona, pp. XV–XXXIV. Braquet, P., Touqui, L., Shen, T., Vargaftig, B. (1987) Perspectives in platelet-activating factor research. Pharmacol. Rev., 39, 97–145. Carrier J., Chauret N., Coulombe P., Mancini M., Neufeld R., Weber M., Archambault J. (1991) Detection of ginkgolide A in Ginkgo biloba cell cultures. Plant Cell Rep., 10, 256–259. Carrier, D.J. (1992) Immobilization and ginkgolide production of Ginkgo biloba cells. PhD thesis, McGill University, Canada. Carrier J., Chauret, N., Neufeld, R. and Archambault, J. (1994) Ginkgo biloba L. (maiden hair Tree): In Vitro culture and the formation of ginkgolides. In Y.Bajaj (ed.), Biotechnology in Agriculture and forestry, Volume 26, Springer-Verlag, Berlin, pp. 136–145. Carrier, D.J., Archambault, J., van der Heijden, R., Verpoorte, R. (1996) Formation of terpenoid products in Ginkgo biloba L. cultivated cells. Plant Cell Rep., 15, 888–891. Carrier, D.J., van Beek, T.A., van der Heijden, R., Verpoorte, R. (1998) Formation of terpenoid products in Ginkgo biloba L. cultivated cells. Phytochemistry, 48, 89–92. Cartayrade, A., Neau, E., Sohier, C., Balz, J.P., Carde, J.P., Walter, J. (1997) Ginkgolide and bilobalide biosynthesis in Ginkgo biloba. I: Sites of synthesis, translocation and accumulation of ginkgolides and bilobalide. Plant Physiol. Biochem., 35, 859–868. Chand, P.K., Occhatt, S.J., Rech, E.L., Power, J.B., Davey, M.R. (1988) Electroporation stimulates plant regeneration from protoplasts of the woody medicinal species Solanum dulcamara L. J. Exp. Bot., 39, 1267–1274. Corey, E., Kang, M., Desai, C., Ghosh, A., Houpis, I. (1988) Total synthesis of (±) ginkgolide B. J. Am. Chem. Soc., 110, 649–651. Coumans, M.P., Sohato, S., Swanson, E.B. (1989) Plant development from isolated microspores of Zea mays L. Plant Cell Rep., 7, 618–621. Davey, M.R. and Power, J.B. (1987) Aspects of protoplasts and plant regeneration. In K.J.Puite, J.J.Dons, H.J.Huizing, A.J.Kool, M.Koornneef and F.A.Krens (eds.), Progress in Plant Protoplast Research, Kluwer Academic Publishers, Dordrecht. pp. 15–25. Engler, A. (1954) Syllabus der Pflanzenfamilien, Ed. 12, Band 1. Gebrüder Borntraeger, Berlin. Favre-Duchartre, M. (1956) Contribution à l’étude de la reproduction chez le Ginkgo biloba. Revue Cytol. Biol Vég., 17, 1–212. Flesch, V., Jacques, M., Cosson, L., Teng, B., Pétiard, V., Balz, J. (1992) Relative importance of growth and light level on terpene content of Ginkgo biloba. Phytochemistry, 31, 1941– 1945. Galewsky, S., Nessler, C.L. (1986) Synthesis of morphine alkaloids during opium poppy somatic embyrogenesis. Plant Science, 45, 215–222. Gamborg, O.L., Murashige, T., Thorpe, T.A., Vasil, I.K. (1976) Plant tissue culture media. In vitro, 12, 473–478. Gray J. (1987) Control of isoprenoid biosynthesis in higher plants. Adv. Bot. Res., 14, 25–91. Huh, H., Staba, J. (1993) Ontogenic aspects of ginkgolide production in Ginkgo biloba. Planta Med., 59, 232–239.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

96

DANIELLE JULIE CARRIER and DOMINIQUE LAURAIN

Jain, S.M., Dong, N., Newton, R.J. (1989) Somatic embryogenesis in slash pine (Pinus elliottii) from immature embryos cultured in vitro. Plant Science, 65, 233–241. Jeon, M., Sung, S., Huh, H., Kim, Y. (1995) Ginkgolide B production in cultured cells derived from Gingko biloba L. leaves. Plant Cell Rep., 14, 501–504. Jones, N.B., Vanstaden, J., Bayley, A.D. (1993) Somatic embryogenesis in Pinus patula. Plant Physiol., 142, 366–372. Keller, W.A. and Stringam, G.R. (1978) Production and utilization of microspore-derived haploïd plants. In T.A.Thorpe (ed.), Frontiers of Plant Cell culture,. IAPTC, Canada. pp. 113– 122. Kleinig H. (1989) The role of plastids in isoprenoid biosynthesis. Annual Rev. Plant Physiol. Plant Molec. Biol, 40, 39–59. Kutz, M., Romer, S., Suire, C, Hugueney, P., Well, J., Schantz, R., Camara, B. (1992) Identification of a cDNA for the plastid-located geranylgeranyl pyrophosphate synthase from Capsicum annuum: correlative increase in enzyme activity and transcript level during fruit ripening. The Plant J., 2, 25–34. Laurain, D., Trémouillaux-Guiller, J. Chénieux, J.C. (1993a) Embryogenesis from microspores of Ginkgo biloba L., a medicinal woody species. Plant Cell Rep., 12, 501–504. Laurain, D., Chénieux, J.C., Trémouillaux-Guiller, J. (1993b) Direct embryogenesis from female haploid protoplasts of Ginkgo biloba L., a medicinal woody species. Plant Cell Rep., 12, 656–660. Laurain, D., Chénieux, J.C., Trémouillaux-Guiller, J. (1996) Somatic embryogenesis from immature zygotic embryos of Ginkgo biloba. Plant Cell Tissue Organ Cult., 44, 19–24. Laurain, D., Trémouillaux-Guiller, J. Chénieux, J.C, van Beek T.A. (1997) Production of ginkgolide and bilobalide in transformed and gametophyte derived cell cultures of Ginkgo biloba, Phytochemistry, 46, 127–130. Lu, C.Y., Thorpe, T.A. (1987) Somatic embryogenesis and plantlet regeneration in cultured immature embryos of Picea glauca. J. Plant Physiol., 128, 297–302. Michel, P. and Hosford, D. (1988) Ginkgo biloba: from “living fossil” to modern therapeutic agent. In P.Braquet (ed.), Ginkgolides—Chemistry, Biology, Pharmacology and Clinical Perspectives, Volume 1, J.R.Prous Science Publishers, Barcelona, pp. 1–8. Mitchell, J.C., Petolino, J.F. (1991) Plant regeneration from haploid suspension and protoplast cultures from isolated microspores of maize. J. Plant Physiol., 137, 530–536. Murashige, T., Skoog, F. (1962) A revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiologia Plantarum, 28, 473–497. Murashige, T. and Tücker, D.P. (1969) Growth factor requirements of Citrus tissues cultures. In J.V.Chapman (ed.), Proceedings of the First International Citrus Symposium, University of California, Riverside, pp. 1155–1161. Nakanishi, K. (1988) Ginkgolides-Isolation and structural studies carried out in the mid 1960’s. In P.Braquet (ed.), Ginkgolides—Chemistry, Biology, Pharmacology and Clinical Perspectives, Volume 1, J.R.Prous Science Publishers, Barcelona, pp. 27–36. Nakanishi, K, Habaguchi, K. (1971) Biosynthesis of ginkgolide B, its diterpenoid nature and origin of the tert-butyl group. J. Am. Chem. Soc., 93, 3544–3546. Nakanishi, K., Habaguchi, K., Nakadaira, Y., Woods, M., Maruyama, M., Major, R., Alauddin, M., Patel, A., Weinges, K., Bähr, W. (1971) Structure of bilobalide, a rare tert-butyl containing sesquiterpenoid related to the C20-ginkgolides. J. Am. Chem. Soc., 93, 3546–3547. Neau, E., Cartayrade, A., Balz, J.P., Carde, J.P., Walter, J. (1997) Ginkgolide and bilobalide biosyntheisis in Ginkgo biloba. II: Identification of a possible intermediate compound by using inhibitors of cytochrome P-450-dependent oxygenases. Plant Physiol. Biochem. 35, 869–879. Norgaard, J.V., Krogstrup, P. (1991) Cytokinin induced somatic embryogenesis from immature embryos of Abies nordmanniana LK. Plant Cell Rep., 9, 509–513.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

PLANT CELL BIOTECHNOLOGY OF GINKGO

97

Pépin M.F., Archambault J., Chavarie C, Cormier C. (1995) Growth kinetics of Vitis vinifera cell suspension cultures: 1. Shake flask cultures. Biotechnol. Bioeng., 47, 131–138. Sangwan-Norreel, B.S., Sangwan, R.S., Pare, J. (1986) Haploïdie et embryogénèse provoquée in vitro. Bull. Soc. Bot. France, 4, 7–39. Sato, T., Nishio, T., Harai, M. (1989) Plant regeneration from isolated microspore cultures of Chinese cabbage (Brassica campestris spp. pekinensis). Plant Cell Rep., 8, 486–488. Schuller, A., Reuther, G., Geiert, T. (1989) Somatic embryogenesis from seed explants of Abies alba. Plant Cell Tissue Organ Cult., 17, 53–58. Schwarz, M. (1994) Terpen-Biosynthese in Ginkgo biloba. Eine überraschende Geschichte. PhD thesis, Eidgenössischen Technischen Hochschule Zürich, Switzerland (ETH Nr. 10951). Schwarz, M. and Arigoni, D. (1999) Ginkgolide biosynthesis. In D.E.Cane (ed.) Isoprenoids including carotenoids and steroids, Vol. 2 in D.Barton, K.Nakanishi and O.Meth-Cohn (eds.) Comprehensive Natural Products Chemistry. Elsevier, Amsterdam, pp. 367–400. Shunan, L., Tianen, S., Genbao, L. (1997) Transformation of Ginkgo hairy root and establishment of its suspension culture clone. Wuhan Univ. J. Nat. Sci., 2, 493–495. Thompson, R.G., von Aderkas, P. (1992) Somatic embryogenesis and plant regeneration from immature embryos of western larch. Plant Cell Rep., 11, 379–385. Trémouillaux-Guiller, J., Andreu, F., Crèche, J., Chénieux, J.C., Rideau, M. (1987) Variability in tissues cultures of Choisya ternata. Alkaloid accumulation in protoclones and aggregate clones obtained from established strains. Plant Cell Rep., 6, 375–378. Trémouillaux-Guiller, J., Laurain, D. and Chénieux J.C. (1996) Microspore and protoplast culture in Ginkgo biloba. In S.M.Jain, S.K.Sopory and R.E.Veilleux (eds.), In vitro haploid Production in Higher Plants, Volume 3, Kluwer Academic Publishers, The Netherlands, pp. 277–295. Tulecke, W. (1997) Tissue culture studies on Ginkgo biloba. In T.Hori, R.W.Ridge, W.Tulecke, P.Del Tredici, J.Trémouillaux-Guiller and H.Tobe (eds), Ginkgo biloba a global treasure. From biology to medicine. Springer, Heidelberg, pp. 141–156. Vasil, I.K. (1987) Developing cell and tissue culture systems for the improvement of cereal and grass crops. J. Plant Physiol., 128, 193–218. Woods, S.H., Phillips, G.C., Woods, J.E., Collins, G.B. (1992) Somatic embryogenesis and plant regeneration from zygotic embryo explants in mexican weeping bamboo, Otatea acuminata aztecorum. Plant Cell Rep., 11, 257–261. Yates, W.F. (1986) Induction of embryogenesis in embryo-derived callus of Ginkgo biloba L. Proceedings of the VIth International Congress of Plant Tissue and Cell Culture, University of Minnesota, pp. 43. Zrÿd, J.P. (1988) Cultures de cellules, tissus et organes végétaux. Presses Polytech Romandes Lausanne, pp. 1–308.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

6. GINKGO BILOBA—LARGE SCALE EXTRACTION AND PROCESSING JOE O’REILLY Cara Partners, Little Island, Co. Cork, Ireland

DEVELOPMENT OF EXTRACT OF GINKGO BILOBA—EGB 761 In 1965 Dr. Willmar Schwabe Arzneimittel discovered that extracts from the leaves of Ginkgo biloba were effective for the treatment of peripheral and cerebral arterial disturbances of the blood supply especially in elderly patients. As a result they proceeded to develop a special multistage process producing a standardized extract. This product was designated the code name EGb 761. A first patent was granted for the extract in 1971 in Germany and 1972 in France (Dr Willmar Schwabe, 1971 and 1972). The process was specifically designed to enrich the main active ingredients of the extract known at this time i.e. the flavonol glycosides to 24% (see Chapter 9 “Chemical Constituents of Ginkgo biloba” by A.Hasler). A further patent was granted in 1989 (Dr Willmar Schwabe, 1989) that describes the enrichment of terpene trilactones to 6% and also the removal of undesirable components from the extract such as the ginkgolic acids which are known allergens. LABORATORY SCALE TO INDUSTRIAL SCALE The process for the preparation of EGb 761 was first developed on laboratory scale. However when a process moves from laboratory scale to a commercial manufacturing operation, unexpected problems of a physical and chemical nature are often encountered. These problems multiply when dealing with a complex mixture such as EGb 761, where it is necessary to ensure that the uniqueness of the product is not compromised in the scale up. In order to be successful, large scale production must process the leaves to produce an extract at planned production rates, at the projected manufacturing cost and to clearly defined quality standards. Manufacturing cost is not only based on obvious factors such as costs of raw materials, product yield and return on capital but also the overall safety of the operation with respect to personnel at the facility, the public and the environment. Technology, Plant and Equipment Some extraction and purification techniques for the preparation of medicines from natural sources go back more than 1000 years. However, in the pharmaceutical 99 Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

100

JOE O’REILLY

industry, innovation and new technology are essential for the continuous development of the process and result in optimized yield, lower operating costs, reduction in solvent usage and reduction of wastes and energy consumption. Similarly, in laboratory new and improved analytical techniques are developed and adopted to provide a better service to production and as a consequence higher specification of the product. In order to ensure the integrity of EGb 761 during large scale extraction and processing, extensive studies were carried out on pilot plant scale using test equipment supplied by vendors. Data from these studies was used to specify full scale process equipment and materials of construction i.e. high quality stainless steel and borosilicate glass. Likewise mechanical seals on pumps and agitators and all gasket materials were specified to be compatible with the extract and all solvents used in the process. Safety Safety has a high priority. Personnel are trained in all aspects of chemical handling and supplied with the necessary protective clothing. They also undergo regular medical check-ups to ensure the adequacy of personnel protective equipment and procedures. Large volumes of flammable solvents which are used in the process pose high risks and extensive controls are in place to guard against fire and explosion. These control measures include: 1. Elimination of ignition sources by use of explosion proof electrical equipment, earthing and bonding of equipment, “no smoking” regulations and wearing of anti-static shoes and gloves. 2. Elimination of flammable atmospheres by ventilation of process areas, sealed solvent handling systems, pressurisation of control rooms and explosion-proof electrical switch rooms. 3. Fire detection/protection systems which include heat and smoke detectors, Sprinkler systems, CO2 suppression systems and automatic shut-off valves on solvent lines. 4. Solvent tanks are fitted with flame arresters. Over-pressure relief devices are provided on vessels, vent headers and dust collection systems to safely vent any overpressure. These devices include bursting discs, relief valves, explosion panels and isolation valves. 5. Trained emergency response teams and first aiders are available on site in case of emergencies. Environmental Protection As with all manufacturing facilities, wastes are generated during the production of EGb 761 and it is essential that waste streams from the production facility be minimized. Waste leaves generated after extraction are used as mulch or compost. Waste water which may contain minute quantities of Ginkgo extract and solvents from the production facility is treated in an aerobic treatment plant for removal of

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA—LARGE SCALE EXTRACTION AND PROCESSING

101

organic substances. Also abatement systems are in place to reduce emissions to the atmosphere. Optimization of yield results in energy conservation and reduced solvent usage. There is a fire water retention pond on site which is designed to retain all water from the Sprinkler Protection system in the event of a fire. The water collected in the pond is analysed and disposed of in an environmentally safe manner. An environmental management programme is in place. The objective of this programme is to assess operations on site and to review all practical options for the use of cleaner technology and cleaner production to minimise material consumption and waste generation. The programme identifies areas for investigation and a proposed timetable for completion of specific projects and studies.

MANUFACTURE OF EGB 761 With the increasing demand for EGb 761 in the 1970’s, Dr. Willmar Schwabe set up a joint venture with Laboratoires Beaufour of France for the production of EGb 761. The location chosen was a green-field site in Cork, Ireland. The process involves nine main steps (Diagram 1) each of which incorporates various industrial operations so that in all there are over 50 operations from start to finish. The manufacturing process for EGb 761 has been developed to produce a standardized extract and the quality of the extract is defined by: 1. Quality of leaves 2. Process for extraction and purification of the extract using specific techniques and solvents. This gives a standardized extract containing a dosage of known active substances while also eliminating undesirable substances contained in the leaves and materials used during manufacturing. The main specifications for EGb 761 are set out in Table 1. In order to meet these specifications the facility is operated to current Good Manufacturing Practices (cGMP) with emphasis on: 1. 2. 3. 4.

Leaf Quality Quality Control of Raw Materials and Consumables Quality Control of Process Operations Quality Control of Finished Product

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

102

JOE O’REILLY

Diagram 1 Process Flow Diagram for Extract of Ginkgo biloba—EGb 761

Leaf Quality The percentage active ingredients in the leaf varies significantly and can be influenced by natural and human factors. The natural influences which affect the quality of the leaves are: (a) The age of the plant The percentage active ingredients in young plants (1–3 years) is generally very high. However as the plant matures, there is a decrease in the active ingredients, levelling off after 6–8 years and thereafter remaining relatively constant. (b) Seasonal variation The percentage flavonol glycosides, ginkgolide A and bilobalide in the leaves decreases during the growing season whereas ginkgolide B remains relatively constant and ginkgolide C increases. (c) Type of Soil The type of soil affects each of the active ingredients differently. As a result the relative ratios of active ingredients vary from area to area. (d) Climate As with the type of soil, the climate affects the ratio of the different active ingredients in the leaves. Also during the growing season climatic conditions, e.g. drought or excessive rainfall affects the quality of the leaves.

Copyright © 2000 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group.

GINKGO BILOBA—LARGE SCALE EXTRACTION AND PROCESSING

103

The human influences which affect the quality of the leaves are: (a) Pruning Large scale cultivation of Ginkgo biloba requires that the plant be pruned on a regular basis. The percentage active ingredients in the leaves is affected by the severity of the pruning. (b) Use of fertilizers As well as increasing the yield of leaves, the use of fertilizers increases the percentage active ingredients in the leaves. (c) Drying of leaves The leaves on harvesting contain 70–80% moisture. It is essential that leaves are dried within hours of harvesting thus preventing mould growth and fermentation which results in degradation of the active ingredients, Due to the influences outlined above, the percentage active ingredients in the leaves varies considerably. In order to produce a high quality product and meet the specifications set out in Table 1 it is necessary to have the correct blend of leaves for production. From experience gained over the last 20 years the correct blend is achieved by irrigation, varying the time of harvest, use of fertilizer and pruning. As part of our quality control, the quantity and quality of our leaf stock is continuously monitored and the specifications for active ingredients in the leaves are set for the following harvest.

Table 1 Main specifications for extract of Ginkgo biloba–EGb 761

Flavonol Glycosides Terpene trilactones Ginkgolides A+B+C Bilobalide HPLC Profile TLC Proanthocyanidins Ginkgolic Acids Aspect Solubility Odour Water Solvents Sulphate Ash

21.6–26.4% 5.4–6.6% 2.8–3.4 % 2.6–3.2% Same as Ref. Std. Same as Ref. Std.

Ginkgo Biloba - PDF Free Download (2024)
Top Articles
Latest Posts
Article information

Author: Arline Emard IV

Last Updated:

Views: 5813

Rating: 4.1 / 5 (52 voted)

Reviews: 83% of readers found this page helpful

Author information

Name: Arline Emard IV

Birthday: 1996-07-10

Address: 8912 Hintz Shore, West Louie, AZ 69363-0747

Phone: +13454700762376

Job: Administration Technician

Hobby: Paintball, Horseback riding, Cycling, Running, Macrame, Playing musical instruments, Soapmaking

Introduction: My name is Arline Emard IV, I am a cheerful, gorgeous, colorful, joyous, excited, super, inquisitive person who loves writing and wants to share my knowledge and understanding with you.