Academia.eduAcademia.edu
Essential Oil­ Bearing Grasses The genus Cymbopogon Edited by Anand Akhila Medicinal and Aromatic Plants - Industrial Profiles o ~~~F~~~~~"P Boca Raton London New York CRC Press is an imprint of the Taylor f< Francis Group, an informa business Preface to Preface .:..·,· Author/Edito Contributors CRC Press Taylor & francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, Fl. 33487-2742 2010 by Taylor and francis Group, LLC CRC Press is an imprint of Taylor & francis Group, an Informa business Chapter 2 (i) No claim to original Ll.S. Government works Printed in the United States of America on acid-free paper 10987654321 International Standard Book Number, 978-0-8493-7857-7 (Hardback) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the valid­ ity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. ChapterS Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti­ lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy­ ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. Chapter 6 Cbapter4 for permission to photocopy or use material electronically from this work, please access www.copyrtght.com thttp.r/ www.copyright.coml) or contact the Copyright Clearance Center, Inc. (Ct.C), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. for organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Chapter 7 ; Library of Congress Cataloging-in-Publication Data Chapter 8 Essential oil-bearing grasses' the genus Cymbopogon / editor: Anand Akhila. p. em. -­ (Medicinal and aromatic plants--industrial profiles; v.46) Includes bibliographical references and index. ISBN 978-0-8493-7857-7 (hardcover, alk. paper) 1. Cymbopogon. 2. Cymbopogon--Industrial applications. 3. Cymbopogon--Therapeutic use. I. Akhila. Anand. II. Title. Il l. Series: Medicinal and aromatic plants-vindustrial profiles; v. 46. QK495.G74E7452010 661'.806--dc22 Chapter 9 2009024407 Index Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com . 1 The Genus Cymbopogon Botany, Including Anatomy, Physiology, Biochemistry, and Molecular Biology Cinzia M. Bertea and Massimo E. Maffei CONTENTS 1.1 Introduction 1.2 Anatomy 1.2.1 General Considerations 1.2.1.1 Leaf Anatomy 1.3 Biochemistry 1.3.1 Characterization of the Photosynthetic Variant.. 1.3.2 NADP+Inhibition of NADP-MDH l.3.3 pH and Temperature Dependence of NADPH-MDH and NADP-ME 1.3.4 CO 2 Assimilation and Stomatal Conductance 1.4 Molecular Biology 1.4.1 Randomly Amplified Polymorphic DNA (RAPD) Markers 1.4.2 Simple Sequence Repeat Markers (SSRs) 1.5 Physiology and Ecophysiology 1.5.1 Cymbopogon martinii 1.5.2 Cymbopogon fiexuosus 1.5.3 Cymbopogon winterianus 1.5.4 Other Cymbopogon Species References 1.1 I 4 4 5 7 8 9 9 10 II II 13 14 14 16 17 18 20 INTRODUCTION Among monocots forming the family of Gramineae some grasses produce essential oils that are a valuable source for the flavor industry. Two grasses are known for their industrial potential for essential oil production: Vetiveria zizanioides Stapf, which has been the subject of a monograph in the series Medicinal and Aromatic Plants-Industrial Profiles (Maffei 2002) and Cymbopogon, which is the subject of this book and which updates the monograph edited by Kumar et al. (2000). Cymbopogon is a genus comprising about 180 species, subspecies, varieties, and subvarieties. It is native to warm temperate and tropical regions of the Old World and Oceania. Table 1.1 lists the several species, subspecies, varieties, and subvarieties as reported by the International Plant Names Index (2004), published on the Internet http://www.ipni.org (accessed September 10, 2007). The name Cymbopogon was introduced by Sprengel in 1815 (Sprengel 1815) and at that time the genus consisted of a few species, which were then moved to the genus Andropogon. In fact, both Cymbopogon and Andropogon belong to the tribe Andropogoneae, a monophyletic tribe that 2 Essential Oil-Bearing Grasses: The Genus Cymbopogon TABLE 1.1 List of Cymbopogon Species, Subspecies, Varieties, and Subvarieties as Reported in The International Plant Names Index (2004) Cvmbopogon acutispathaceus De Wild e. afronardus Stapf e. ambiguus A. Camus. e. andongensis Rendle e. angustispica Nakai e. annamensis A. Camus. e. arabicus Nees ex Steud. e. arriani Aitch. e. arundinaceus Schult. e. baginnicus Stapf e. hassacensis A. Camus. e. bequaertii De Wild e. bhutanicus Noltie e. bombycinus A. Camus. e. bombvcinus var. bombvcinus (RBr) Domin. e. bombvcinus var. townsvillensis Domin. e. bombvcinus var. typicus Domin. e. bracteatus Hitchcock e. caesius (Hook and Arn.) Stapf e. caesius subsp. giganteus (Chiov) Sales e. calcicola C.E. Hubb. e. calciphilus Bor e. cambodgiensis E.G. Camus and A. Camus e. chevalieri A. Camus e. chrvsargyreus Stapf e. circinnatus Hochst. ex Hookf. e. citratus Stapf e. citriodorus Link e. claessensii Robyns e. clandestinus Stapf e. coloratus Stapf e. commutatus Stapf e. commutatus var.jammuensis (Gupta) H.B. Naithani e. condensatus Spreng. e. confertiftorus Stapf e. connatus Chiov. e. cyanescens Stapf e. cvmbarius Rendle. e. densiflorus Stapf e. dependens B.K. Simon e. dieterlenii Stapf ex Phillips e. diplandrus De Wild. C. distans (Nees ex Steud.) Will Watson. e. divaricatus Stapf e. eherhardtii A. Camus. e. effusus A. Camus. e. elegans Spreng. e. exaltatus A. Camus. e. exaltatus var. ambiguus Domin. e. exaltatus var. exaltatus (RBr) Domin. e. exaltatus var. genuinus Domin. e. exaltatus var. gracilior Domin. e. exaltatus var. lanatus (RBr) Domin. e. exarmatus Stapf e. excavatus Stapf Cifamiliaris De Wild e.jigarianus Chiov. e. filipendulus Rendle e. finuimus Rendle e.fiexuosus Stapf e. flexuosus var. assamensis S.c. Nath and K.K. Sarma e.floccosus Stapf e. foliosus Roem and Schu It. e. gazensis Rendle e. gidarba [Buch-Harn. ex Steud.] Haines e. giganteus Chiov. e. glundulosus Spreng. e. glaucus Schult. e. globosus Henrard. e. goeringii A. Camus e. goeringii var. hongkongensis S. Soenarko e. graws Domin. e. hamatulus A Camus e. hirtus Stapf ex Burtt Davy e. hirtus subsp. villosum (Pignatti) Pignatti e. hispidus Griff. e. hookeri (Munro ex Hackel) Stapf ex Bor e. humboldtii Spreng. e. iwarancusa Schult. e. jinshaensis R. Zhang and C.H. Li e. jwarancusa subsp. olivieri (Boiss.) S. Soenarko e. kapandensis De Wild e. khasianus (Hackel) Stapf ex Bor e. ladakhensis B.K. Gupta e. lanatus Roberty e. laniger Duthie e. lecomtei RendJe e. lepidus (Nees) Chiov. e. liangshanensis S.M. Phillips and S.L. Chen e. lividus (Thwaires) Willis e. luembensis De Wild e. mandalaiaensis Soenarko e. marginatus Stapf ex Burtt Davy e. martinii Stapf he Genus Cymbopogon The Genus Cymbopogon as Reported TABLE 1.1 (continued) list of Cymbopogon Species, Subspecies, Varieties, and Subvarieties as Reported in The International Plant Names Index (2004) e. martinianus Schult. e. mekongensis A. Camus C. melanocarpus Spreng. e. micratherus Pilg. e. microstachys (Hookf) S. Soenarko e. microthecus A. Camus e. minor B.S. Sun and R. Zhang ex S.M. Phillips and S.L. Chen e. minutiflorus S. Dransf e. modicus De Wild e. motia B.K. Gupta e. munroi (C.B. Clarke) Noltie e. nardus (L.) Rendle e. nardus subvar. bombvcinus (R.Br.) Roberty e. nardus vat. confertiftorus (Steud.) Stapf ex Bar e. nardus subvar. exaltatus (R.Br.) Roberty C. nardus subvar. grandis Roberty. e. nardus subvar. lanatus (R.Br.) Roberty e. nardus var. luridus (Hookf.) Gavade and M.R. Almeida e. nardus subvar. procerus (R.Br.) Roberty e. nardus subvar. refractus (R.Br.) Roberty e. nardus subvar. schultzii Roberty e. nervatus A. Camus e. nyassae Pilg. e. obtectus S.T. Blake e. olivieri (Boiss.) Bor e. osmastonii R. Parker e. pachnodes (Trin.) Will Watson e. papillipes (Hochst. ex A. Rich) Chiov. e. parkeri Stapf e. pendulus Stapf e. phoenix Rendle e. pilosovaginatus De Wild e. pleiarthron Stapf C. plicatus Stapf e. plurinodis Stapf ex Burtt Davy e. polyneuros Stapf e. pospischilii (K. Schum) C.E. Hubb e. princeps Stapf e. procerus A. Camus e. procerus var. genuinus Domin. e. procerus var. procerus (R.Br.) Domin. e. procerus var. schultzii Domin. e. prolixus (Stapt) Phillips e. prostratus Sweet e. proximus Stapf e. proximus var. sennarensis (Hochst.) Tackholm 3 C. pruinosus Chiov. e. pubescens (vis) Fritsch. e. queenslandicus S.T. Blake e. quinhonensis (A. Camus) S.M. Phillips and S.L. Chen [transferred to Andropogon (Phillips and Hua 2(05)J e. ramnagarensis B.K. Gupta e. rectus A. Camus e. reflexus Roem and Schult. e. refractus A. Camus e. rufus Rendle e. ruprechtii Rendle e. scabrimarginatus De Wild e. schimperi Rendle e. schoenanthus Spreng. e. schoenanthus subsp. velutinus Cope e. schultzii Roberty e. sennaarensis Chiov. e. setifer Pilg. e. siamensis Bor e. solutus Stapf e. stipulatus Chiov. e. stolzii Pilg. e. stracheyi (Hookf.) Raizada and Jain e. strictus Bojer. e. stypticus Fritsch. e. suaveolens Pilger e. subcordatifolius De Wild e. tamba Rendle e. tenuis Gilli e. thwaitesii (Hookf) Willis e. tibeticus Bor e. tortilis (Presl.) A. Camus e. tortilis subsp. goeringii (Steud.) TKoyama e. traninhensis (A. Camus) SSoenarko e. transvaalensis Stapf ex Burtt Davy e. travancorensis Bar e. tungmaiensis L. Liu e. umbrosus Pilg. e. validus Stapf ex Burtt Davy e. vandervstii De Wild e. versicolor (Nees ex Steud.) Will Watson e. virgarus Stapf ex Rhind. e. virgatus Stapf ex Bor C. welwitschii Rendle e. winterianus Jowitt e. xichangensis R. Zhang and B.S. Sun Source: Published on the Internet hnp://www.ipni.org [accessed September 10.2007]. 4 Essential Oil-Bearing Grasses: The Genus Cymbopogon includes 85 genera. Mathews and coworkers (2002) found strong support for a core Andropogoneae that includes, among others, Andropogon and Cymbopogon, and support for its relationship with an expanded Saccharinae that includes Microstegium. The limited difference in the plant traits between Andropogon and Cymbopogon, has argued the possibility that species belonging to Cymbopogon might be a subgenus of Andropogon. Most of Andropogoneae have pairs of spikelets in the inflores­ cence, one sessile and one on a pedicel, although in some species one or the other of these spikelets appear to be suppressed. The inflorescences form is also highly variable (Mathews et al. 2002). Morphologically, the main difference in the genus Cymbopogon is the presence of some pair of spikelets, for each spike, with unisexual male flowers, whereas in the Andropogon spikelets are usually sessile and often sterile. Cymbopogon plants are tall (up to and above I m) perennial plants, with narrow and long leaves that are mostly characterized by the presence of silica thorns aligned on the leaf edges. Leaves bear glandular hairs, usually each with a basal cell that is wider than the distal cell (see Section 1.2). Representative of the Andropogoneae exhibit C 4 photosynthesis, with NADP-ME as the primary decarboxylating enzyme (Mathews et al. 2002), they usually have a chro­ mosome number of five, with ploidy levels ranging from tetrapJoids to 24-pJoid. Polyploidy, either as alloploidy or segmental alloploidy, is frequent. Representative specimens of various species of the genus Cymbopogon have been cytogenetically studied by Spies and coworkers (Spies et al. 1994). The monophyly of Cymbopogon has also been clearly demonstrated, and the genus is sister to Heteropogon (Mathews et al. 2002). Among the several aromatic species belonging to the genus Cymbopogon the most important in terms of essential oil production are C. martinii, also known as palmarosa; C. citratus, better known as lernongrass; and the so-called East Indian lemongrass, Cymbopogon flexuosus, native to India, Sri Lanka, Burma, and Thailand; whereas, for the related West Indian C. citratus, a Malesian origin is generally assumed. C. nardus and C. winterianus produce the famous citronella from Sri Lanka and lava, respectively. Also known to produce essential oils are C. schoenanthus, or camel grass; C. caesius, or inchi/kachi grass; C. afronardus, C. clandestinus; C. coloratus; C. exaltatus; C. goeringii; C. giganteus; C. jwarancusa; C. polyneuros; C. procerus; C. proximus; C. rectus; C. sennaarensis; C. stipulatus; and C. virgatus (Guenther 1950b). The main constituents of Cymbopogon essential oils will be described in other chapters of the book. Many laboratories in several countries are deeply involved in studying various aspects of cyrn­ bopogons, using variously derived genetic resources. The work already done covers a wide array of topics, including botanical identification, plant description, cytogenetics, and cell, tissue, and organ in vitro cultures. Physiology and biochemistry of stress tolerance and essential oil biosynthesis, genetics and biotechnology, and agrotechnology involved in crop production and disease and pest control chemistry of terpenes, biological activities of essential oil terpenoids and trade and market­ ing aspects (reviewed by Kumar et al. 2000). The major objective of this monograph is to update the literature and give references on the afore­ mentioned topics of cymbopogons, with particular attention to industrial aspects. In the following sections of this introductory chapter, we will explore the anatomy and biochemistry of the photosyn­ thetic apparatus, also considering the most recent advances in molecular biology of Cymbopogon. We will conclude the chapter with some physiological and ecophysiological considerations. 1.2 1.2.1 ANATOMY GENERAL CONSIDERATIONS Higher plants can be divided into two groups, C, and C 4 , based on the mechanism utilized for pho­ tosynthetic carbon assimilation related to anatomical and ultrastructural features. A cross section of a typical C 1 leaf reveals essentially one type of photosynthetic, chloroplast-containing cell, the mesophyll, and in these plants atmospheric CO 2 is fixed directly by the primary carbon-fixation enzyme ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco), In contrast, a typical C 41eaf 1.2.1.1 First desc . and C. refi followed by conducted by 1.2.1.1.1 The Genus Cymbopogon core Andropogoneae relationship with an planttraits between ing to Cymbopogon ikelets in the inflores­ her of these spikelets Mathews et al. 2002). nee of some pair of .ropogon spikelets are 1m) perennial plants, f silica thorns aligned · 1that is wider than the 4 photosynthesis, with y usually have a chro­ oid. Polyploidy, either · s of various species of 'COworkers (Spies et al. and the genus is sister on the most important sa; C. citratus, better onflexuosus, native to . C. citratus, a Malesian us citronella from Sri "schoenanthus, or camel • coloratus; C. exauatus; Co proximus; C. rectus; main constituents of various aspects of cym­ covers a wide array of cell, tissue, and organ sential oil biosynthesis, .on and disease and pest ids and trade and markete references on the afore­ aspects. In the following mistryof the photosyn­ ( biology of Cymbopogon. I considerations. hanism utilized for pho­ features. A cross section plast-containing cell, the primary carbon-fixation contrast, a typical C 4 leaf 5 has two distinct chloroplast-containing cell types, the mesophyll and the bundle sheath (or Kranz) cells, and they differ in photosynthetic activities (Hatch 1992; Maurino et al. 1997). The operation of the C 4 photosynthetic mechanism requires the cooperative effort of both cell types, connected by an extensive network of plasmodesmata that provides a pathway for the flow of metabolites between the cells. The C 4 pathway is a complex adaptation of the C 3 pathway that overcomes the limitation of photo­ respiration and is found in a diverse collection of species, many of which grow in hot climates. It was first discovered in tropical grasses (e.g., sugarcane and maize) and is now known to occur in 16plant families. It occurs in both monocotyledonous and dicotyledonous plants, and is particularly prominent in species of the Gramineae, Chenopodiaceae, and Cyperaceae (Edwards and Walker 1983).About half of the species of the Poaceae are included among the C 4 plants (Smith and Brown 1973). The key feature of C 4 photosynthesis is the compartmentalization of activities into two spe­ cialized cell and chloroplast types. Rubisco and C 3 photosynthetic carbon reduction (PCR) cycle are found in the inner ring of bundle sheath cells. These cells are separated from the mesophyll and from the air in the intercellular spaces by a lamella that is highly resistant to the diffusion of CO 2 (Hatch 1988). Thus, by virtue of this two-stage CO 2 fixation pathway, the rnesophyll-located C4 cycle acts as a biochemical pump to increase the concentration of CO 2 in the bundle sheath an estimated lO-fold over atmospheric concentrations. The net result is that the oxygenase activity of Rubisco is effectively suppressed, and the PCR cycle operates more efficiently. C4 plants have two chloroplast types, each found in a specialized cell type. Leaves of C 4 plants show extensive vascularization, with a ring of bundle sheath (BS) cells surrounding each vein and an outer ring of mesophyll (M) cells surrounding the bundle sheath. CO 2 fixation in these plants is a two-step process. There are three variants on the basic C 4 pathway, and the biochemical distinctions are correlated with the ultrastructural differences of Kranz cells (Gutierrez et al. 1974; Hatch et al. 2007; Hatch et al. 1975). The three C 4 variants can be distinguished ultrastructurally by using combinations of two characters of bundle sheath cell chloroplasts, by the degree of granal stacking, and by the chloroplasts position (Gutierrez et al. 1974). For this reason, comparative grass leaf anatomy has become the object of intensive investigation in relation to photosynthesis along with biochemical studies. 1.2.1.1 Leaf Anatomy First descriptions of the genus Cymbopogon were given by Breakwell (1914) on C. bombycinus and C. refractus under the name of Andropogon bombycinus R. Br. and A. refractus, respectively, followed by a leaf structure description done by Vickery (1935) and Prat (1937). Further studies were conducted by Metcalfe (1960). The descriptions given by these authors are very similar and still valid. 1.2.1.1.1 Generic Characters Both adaxial and abaxial epidermises of Cymbopogon species contain short-cells, over the veins, solitary, paired or in short or long rows, the proportion of each type varying with the species. Silica bodies are located over the veins, mostly crossed to dumbell shaped. Microhairs are present usually each with the basal cell wider than the distal cell, the latter frequently tapering to a pointed apex, or hemispherical. Stomata with subsidiary cells range from low or tall dome-shaped to triangular, the proportions of each type varying in different species and sometimes in separate preparations from a single species. The vascular bundles are small, mostly angular, but less conspicuous in some species than in others. The mesophyll presents a distinctly radial chlorenchyma. Bundle sheaths are single (Metcalfe 1960). Figure I.lA shows an electron scanning micrograph of a C. citratus leaf blade. It is possible to observe developed prickle hairs with rather elongated bases over the veins. The abaxial epidermis also reveals several stomata, with narrow guard cells associated with subsidiary cells, typical of grasses (Figure UB). 6 Essential Oil-Bearing Grasses: The Genus Cymbopogon ~\. ~ .' ,:&~, ' , _':«. :..r~ ...?, ,.,.'- ..........4~~.·­ ~'.;.:"~""" r~ A\·',~, iii .J;.(.. . l FIGURE 1.1 (A) Electron scanning micrograph of C. citratus leaf blade. Prickle hairs are evident on the veins (white arrows) (80x). (B) Electron scanning micrograph of a stoma seen from the surface. Guard cells are narrow in the middle and enlarged at the end, while subsidiary cells are triangular (2000x). (C) Semi-thin cross section of C. citratus leaf stained with toluidine blue. The vascular bundle in a minor vein is surrounded by a layer of sheath cells, with chloroplasts arranged in a centrifugal position. (D) Bundle sheath chloroplast without grana and with few starch grains. (E) Mesophyll chloroplast with most of the thylakoid stacked in grana and without starch grains. (F) Bundle sheath chloroplast showing immunolabeling against Rubisco. Colloidal gold particles (white arrows in the high magnification) strongly label the stroma. (G) Enlargement of plasmodesmata connecting mesophyll and bundle sheath cells, providing metabolite flow between the two photosynthetic tissues. 1.3 Genus Cymbopogon The Genus Cymbopogon 7 1.2.1.7.7.7 Leaf Ultrastructure of Cymbopogon citratus The structure of parenchymatic bundle sheath (BS) cells is particularly important in distinguishing C 3 and C 4 species. A commonly men­ tioned anatomical feature of C 4 plants is the orderly arrangement of mesophyll cells with reference to the BS, the two together forming concentric layers around the vascular bundle. The anatomi­ cal differences between plants exhibiting a C 4 photosynthetic carbon assimilation pathway can be disclosed by electron microscopical observations of the BS (Chapman and Hatch 1983; Edwards and Walker 1983; Gutierrez et al. 1974; Hatch 1988; Hatch et al. 1975; Jenkins et al. 1989). In the NADP-ME type, chloroplasts are peripherally arranged and grana are deficient or absent in bundle sheath cells. Two other distinctive features are the presence or absence of a mestome sheath, a layer of cells intervening between metaxylem vessel elements and laterally adjacent BS Kranz cells, and the presence or absence of a cell wall suberized lamella (SL). The mestome sheath occurs in NAD-ME and PCK species, while the SL is present in bundle sheath cell walls of several NADP-ME and PCK species (Eastman et al. 1988; Hattersley and Watson 1976). The number of mitochondria in the NADP-ME subtype is lower than in the NAD-ME one because, in the latter, the enzymes involved in the transformation of aspartate to CO 2 and pyruvate are present in these organelles (Hatch et al. 1975). Anatomical studies conducted on C. citratus leaves indicated the presence of the C 4 Kranz anatomy in this plant along with several ultrastructural features typical of NADP-ME species. In C. citratus cross sections of minor veins, the vascular bundle appears surrounded by one layer of sheath cells, in which the chloroplasts are located in a centrifugal position (Figure 1.lC). No mestome sheath between metaxylem vessel elements and laterally adjacent Kranz cells is observed. In the bundle sheath, ultrastructural analyses show a suberized cell wall lamella, and the presence of agranal chloroplasts, containing numerous starch grains (Figure I.lD). Figure I.L E shows meso­ phyll chloroplasts with most of the thylakoids stacked in grana. The two photosynthetic tissues are connected by a certain number of plasmodesmata that provide a pathway for the flow of metabolites between the mesophyll and bundle sheath cells (Figure I.LG). These observations are in accordance with previous anatomical descriptions related to the genus Cymbopogon reported by Rajendrudu and Das (198 l). These authors reported on the leaf anatomy and photosynthetic carbon assimilation in five species of Cymbopogon (C.fiexuosus, C. martinii var. motia, C. nardus, C. pendulus, and C. winterianus) a Kranz-type leaf anatomy with a centrifugal position of starch-containing chloroplasts in the bundle sheath cells. Starch was exclusively local­ ized in the bundle sheath cells that were typically elongated parallel to the veins and nearly twice as long as wide in the species of Cymbopogon. A narrow leaf interveinal distance was a common feature among the five Cymbopogon species. A xylem-mestome sheath of cells between metaxylem vessels and laterally adjacent bundle sheath cells of primary vascular bundles was totally absent in the five Cymbopogon species (Rajendrudu and Das 1981). 7.2.7.7.2 hairs are evident on the m the surface. Guard cells ular(2000x). (C) Semi-thin oR a minor vein is surrounded . ~ Bundle sheath chloroplast of the thylakoid stacked in ., labeling against Rubisco. :·tIle stroma. (G) Enlargement ite flow between the two Rubisco Immunolocalization High-resolution immunolocalization of Rubisco by electron microscopy showed that labeling occurred only in the bundle sheath chloroplasts of C. citratus. For these experiments, purified rabbit polyclonal antibodies raised against Rubisco were employed. Bound antibodies were then visu­ alized by linking conjugated gold-labeled goat antirabbit polyclonal antibodies. Gold particles appeared to be uniformly distributed throughout the stroma (Figure I.L F). These studies conducted on C. citratus leaves provide evidence for the localization of Rubisco in the stroma of bundle sheath chloroplasts, as expected for a C 4 plant (Bertea et al. 2003). 1.3 BIOCHEMISTRY In a preliminary physiological study conducted by (Maffei et al. 1988) on C. citratus grown in humid temperate climates, some of the PEP-carboxylase kinetic characteristics, and Rubisco and glycolate oxidase activities were found to be comparable to those of C 4 plants. 8 Essential Oil-Bearing Grasses: The Genus Cymbopogon The C, mechanism was also confirmed by the 13C/llC stable isotope ratio analyses (Sl3C = -l3.0). These results are in accordance with sl3e values measured on other species or Cvmbopogon (Rajendrudu and Das 1981) in which Sl3C value of -ll.O for C. flexuosus, -9.7 for C. martinii, -l1.6 for C. nardus, -lO.3 for C. pendulus. and -11.3 for C. winterianus, respectively, were recorded. From a biochemical point of view, the three types of the basic Co pathway differ mainly in the C 4 acid transported into the bundle sheath cells (malate and aspartate) and in the way in which it is decarboxylated; they are named (based on the enzymes that catalyse their decarboxylation) NADP-dependent malic enzyme (NADP-ME) found in the chloroplasts, NAD-dependent malic enzyme (NAD-ME) found in mitochondria, and phosphoenolpyruvate (PEP) carboxykinase (PCK), found the cytosol of the bundle sheath cells (Edwards and Walker 1983; Ghannoum et al. 2001; Hatch et al. 1975; Jenkins et al. 1989; Huang et al. 2001). Furthermore, a characteristic leaf anatomy, biochemistry, and physiology are associated with each of the C4 types (Dengler and Nelson 1999; Hattersley and Watson 1976). A clear indication of the C 4 photosynthetic pathway of C. citratus and the variant to which it helongs was obtained by estimating the activities of NADP-ME (EC 1.1.1.40), NADP-MDH (EC 1.1.1.82), PPDK (EC 2.7.9.1), NAD-ME (EC 1.1.1.39), and PCK (EC 4.1.1.49) as well as some kinetic characteristics of NADP-ME and NADP-MDH. Adaptation to a particular environment is a complex process involving a number of physiological, morphological, and ecologi­ cal factors (Ghannoum et al. 2001; Huang et al. 2001). Therefore, enzyme activities were recorded at the low and high temperatures typical of humid­ temperate climates, in order to evaluate the adaptability of C. citratus. In order to estimate increases or decreases in the reaction rate due to changes in the protonation state, groups involved in the catal­ ysis and/or binding of substrates as a consequence of pH fluctuations, activities were also recorded at different pH values. 1.3.1 1.3.2 CHARACTERIZATION OF THE PHOTOSYNTHETIC VARIANT Further studies dealing with the characterization of the C 4 variant indicated an NADP-depenqent malic enzyme photosynthetic pathway in C. citratus. The biochemical subtype was established through the estimation of the highest activities of NADP-dependent malic enzyme (NADP-ME, EC 1.1.1.40), NADP-dependent malate dehydroge­ nase (NADP-MDH, E.C 1.1.1.82), pyruvate, orthophosphate dikinase (PPDK, E.C 2.7.9.1), NAD­ dependent malic enzyme (NAD-ME, E.C 1.1.1.39),and phosphoenolpyruvate carboxykinase (PCK, E.C 4.1.1.49) and some kinetic, along with some chemical-physical parameters of NADP-ME and NADP-MDH. Extraction and partial purification sequentially involved precipitation with crystalline ammo­ nium sulfate, dialysis, and anion exchange (DEAE-Sephacell). Both, extraction and assays were conducted according to Ashton (1990). The low activity values of PPDK (90.28 nKat mg:' prot), PCK «1 nKat rng" prot), and NAD-ME (52.51 nKat mg:' prot) in C. citratus leaf extracts did not allow to determine the kinetic characteristics, such as Km and Vmax, and/or other chemical-physical parameters of these enzymes. NADP-MDH and NADP-ME presented relatively high activity values (15.93 mKat mg:' and 12.56 mKat mg" prot, respectively). NADP-ME activity was 239-fold greater than NAD-ME activity. The kinetics of NADP-MDH and NADP-ME were therefore measured to gain a clearer picture of the photosynthetic pathway. The low activity of PPDK found in plant extracts of C. citratus agrees with the literature data for C 4 plants (Ashton 1990) and with data on species of the same photosynthetic subtype (Ashton 1990; Bertea et al. 2001). The low levels of NAD-ME and PCK activities found in our extracts clearly indicated the absence of a C4 variant utilizing these two pathways. The relatively high activities found for NADP-MDH allowed us to determine some of its kinetic characteristics (Km and Vmax), which were comparable to those of plants belonging to the NADP-ME photosynthetic variant (Ashton 1990; Bertea et al. 2003). 1.3.3 The Genus Cymbopogon ratio analyses (b13C = speciesof Cymbopogon 9.7 for C. martinii, -11.6 'vely, were recorded. 'way differ mainly in the :and in the way in which their decarboxylation) , ' NAO-dependent malic ) carboxykinase (PCK), ; Ghannoum et al. 2001; cteristicleaf anatomy, gler and Nelson 1999; thway of C. citratus and NAOP-ME (EC 1.1.1.40), "1lIId PCK (EC 4.1.1.49) as ~aptation to a particular rphological, and ecologi­ tures typical of humid­ er toestimate increases ps involved in the catal­ ivities were also recorded the highest activities of ndent malate dehydroge­ OK, E.C. 2.7.9,1), NAD­ , vatecarboxykinase (PCK, eters of NADP-ME and with crystalline ammo­ 'extraction and assays were . 'K (90.28 nKat mg:' prot), ratus leaf extracts did not r otherchemical-physical , tively high activity values tivity was 239-fold greater were therefore measured " y of PPOK found in plant ton 1990) and with data on 'ial. 2001), The low levels of the absence of a C4 variant NAOP-MDH allowed us to re comparable to those of ; Berteaet al. 2003). 9 Km values obtained for OAA and NADPH (NADP-MDH) were 29.0 (±0.014) mM and 31,67 (±0.09) mM, respectively. With regard to NADP-ME, the apparent Km values for NADP" and malate were 19.40 (±0.08) and 242.0 (±0.008) mM, respectively. In the case of NADP-MDH, Vmax values for OAA and NADPH were 12.52 (±0.021) and 14.97 (±0.012) mKat mg- I prot, respectively. NADP-ME Vmax values for malate and NADP" were 8.63 (±0.507) and 18.60 (±0.007) mKat mg" prot. respectively. In general. relatively high activities of NADP-MDH and NADP-ME allowed a partial characterization of these enzymes and provided evidence for an NADP-ME subtype for C. citratus. The apparent kinetic properties of both enzymes were comparable to those of plants belonging to this subtype (Ashton 1990; Hatch et al. 2.007; Hatch et al. 1975), and were consistent with a high photosynthetic activity. even when the plant was cultivated in a temperate climate. 1.3.2 NADP+ INHIBITION OF NADP-MDH Inhibition studies were carried out by measuring the NADP-MDH-catalyzed reaction at varying concentrations of NADP" and constant concentrations of OAA (1.0 mM) and NADPH (0.2 mM). NADP-MDH activity was increasingly inhibited by increasing NADP" concentrations. The activ­ ity value recorded in the presence of 0.25 mM NADP" was only 38% of the activity measured in absence of the oxidized coenzyme. In Zea mays, activation of NADP-MDH is regulated by oxidation and reduction of cysteine residues (thioredoxin-rnediated system) (Lunn et al. 1995), and interconversion of the reduced and oxidized forms is influenced by the NADPH/NADP" ratio (Trevanion et at. 1997). A high NADPHI NADP" ratio leads to a more active enzyme; thus, high rates of OAA reduction only occur in reduced conditions. The percentage of inhibition caused increasing NADP" concentration in our DEAE preparations was in accordance with the observations reported earlier. The relatively high activities of NADP-ME detected enable us to characterize the enzyme in C. citratus. A low Km value was calculated for free NADf>+. These results confirm the high affinity of NADP" for its binding site in all isoforms of this enzyme (Rothermel and Nelson 1989). A higher Km value was calculated for malate in accordance with literature data. 1.3.3 pH AND TEMPERATURE DEPENDENCE OF NADPH-MDH AND NADP-ME pH studies were carried out by using a buffer system that contained an equimolar mixture of buffers adjusted to different pH values with KOH (Bertea et al. 2001). Assays were performed at different pH values, 6.0 to 10.5 for NADP-MDH, and 6.0 to 10 for NADP-ME, using the DEAE-preparation. Maximal activity of NADP-MDH enzyme was observed at pH 8.3, in agreement with the published data (Ashton 1990). An increase in activity was observed starting from the lowest pH value (6.0) up to pH 8.3. At pH 7.0-7.5, the activity was com­ parable to that at pH values ranging between 9.5 and 10.0. At pH 10.5 the activity was comparable to that recorded at pH 6.0. Temperature changes also affected the reaction rate. The influence of temperature on enzyme activities was determined for both NADP-MDH and NADP-ME by adding the substrates to standard assay mixtures equilibrated at the appropriate temperatures. The assays were performed by using the DEAE-preparation. Apparent activation energy was calculated from Arrhenius plots. When enzyme activity was measured using the standard assay system at temperatures ranging from 20°C to 49°C, maximal activity was detected at 35°C, while at 49°C activity was lower, but still much higher than that at 20°C, in accordance with the typical behavior of C 4 photosynthetic enzymes. From a linear Arrhenius plot of the data, in a temperature range from 20°C to 38°C, the activation energy of the reaction was calculated to be 6970.8 cal mol:'. The highest NADP-ME activity was recorded at pH 8.3, in accordance with the enzyme characteristics (Edwards and f:' I ~'. 10 Essential Oil-Bearing Grasses: The Genus Cymbopogon Andreo 1992), whereas at pH 10.0 the activity was higher than the activity recorded at pH 6.0 and 6.5. With regard to temperature, maximal activity was measured at 45°C, the lowest at 20°e. Also in this case, activity response to temperature changes was typical of C 4 plants (Edwards and Andreo 1992). The activation energy of the reaction, calculated in a temperature range from 20°C to 45°C, was 7605.2 cal mol:". Climatic conditions exert an evident effect on the physiological status of photosynthetic enzymes, and variations in light, temperature, moisture, etc., may influence the cytosolic and stromal pH (Ashton 1990). In C 4 plants, NADP-MDH has subunits of 42 kDa, and the native enzyme apparently occurs as either a tetramer or a dimer. The tetramer is the more active form; it is stable at alkaline pH values and at high temperatures (Ashton 1990). NADP-ME is a tetramer with a molecular weight of about 280 kDa, and it is more stable at pH values above 8.0 (Edwards and Andreo 1992). This enzyme exists as a dimer and a monomer, both of which are active. Differences in pH can dramatically alter the activities of these photosynthetic enzymes. Temperature is another critical parameter. When it is low, photosynthetic rates of C 4 plants may fall below those of C3 ones. The response of enzyme activities to such changes depends on the photosynthetic pathway adopted, resulting in a different optimum range of temperatures over which the highest growth rate can be maintained (Fitter and Hay 1987). Because they originated in tropical and subtropical areas, the optimum temperature for photo­ synthesis in C 4 plants is 30°C-40°C, which is approximately LOoC higher than in C, plants (Leegood 1993). However, C4 photosynthesis is usually sensitive to low temperature; the minimum tempera­ ture for photosynthesis in several C4 tropical grasses is 5°C_10°C (Casati et al. 1997). Activities at different temperatures and pH values of C. citratus NADP-MDH and NADP-ME indicated that this species is a C4 NADP-ME plant, which is able to retain its photosynthetic mechanism even when cultivated in temperate climates. 1.3.4 CO2 ASSIMILATION AND STOMATAL CONDUCTANCE A very low compensation point (between 8 and 15 ppm CO 2 ) was calculated for C. citratus. This result is typical for a C 4 plant. Stomatal opening increased in response to CO 2 concentration up to 157 ppm. However, at higher CO 2 values a decrease was recorded, thus indicating a clear effect of the CO 2-concentrating mechanism present in C4 plants (data not shown). In order to evaluate changes in photosynthesis as a function of leaf age, CO 2 assimilation and stomatal conductance were also measured at different developmental stages of C. citratus leaves. A general increase for both parameters was observed, starting from primordial up to mature leaves, while a decrease in CO 2 assimilation and stomatal conductance was recorded in old leaves. Thus, primordial leaves presented the lowest CO 2 assimilation value (9.01 mmol CO 2 drrr? S-I), while the highest values were recorded in young and mature leaves, without appreciable differences (22.71 and 23.94 mmol CO 2 drrr? S-I, respectively). With regard co stomatal conductance, the lowest value was measured in old leaves (55.00 mM H 20 dm? S-I), while the highest occurred in mature ones (165.27 mM H20 drrr? S-I). From a physiological point of view, the remarkable differences between the photosynthetic responses of C 1 and C4 plants to CO 2 concentration become apparent when calculating the CO 2 compensation point. In plants with CO 2-concentrating mechanisms, including C 4 plants, CO 2 concentrations at the carboxylation sites are often saturating. Plants with C 4 metabolism have a CO 2 compensation point of or close to zero, reflecting their very low levels of photorespiration. The results obtained in C. citratus are in accordance with the values previously recorded on other species of Cymbopogon (Raiendrudu and Das 1981). In addition, the C 4 mechanism allows the plant to maintain high photosynthetic rates at lower partial CO 2 pressures in the intercellular spaces of the leaf, which require lower rates of stomatal conductance for a given ratt: of photosynthesis. For these reasons, measuring the CO 2 compensation point and stomatal conductance can be useful to distinguish between C 3 and C4 pathways. 1.4 1.4.1 The Genus Cymbopogon ded at pH 6.0 and :lowest at 20°e. Also wards and Andreo from 20°C to 45°C, syntheticenzymes, lie and stromal pH enzyme apparently ·itis stable at alkaline . a molecular weight Andreo 1992). .Differences in pH can ,. re is another critical those of C3 ones. The ic pathway adopted, st growth rate can be temperature for photo­ in C3 plants (Leegood the minimum tempera­ al. 1997). Activities at -ME indicated that this mechanism even when d for C. citratus. This CO2 concentration up to indicating a clear effect n). In order to evaluate stomatal conductance . A general increase for Yes, while a decrease in Thus, primordial leaves while the highest values (22.71 and 23.94 mmol st value was measured in ure ones (165.27 mM u.o . tween the photosynthetic when calculating the CO 2 including C 4 plants, CO 2 with C4 metabolism have levelsof photorespiration. viously recorded on other mechanism allows the plant m the intercellular spaces of , rate of photosynthesis. For Conductance can be useful to 1.4 11 MOLECULAR BIOLOGY The morphological variation and oil characteristics of various species and varieties of Cymbopogon have been reported, but such information is not sufficient to precisely define the relatedness among the morphotypes and chemotypes. For instance, C. martinii var. sofia and C. martinii val'. motia are morphologically almost indistinguishable, but show distinct chernotypic characteristics in terms of oil constituents (Guenther 1950a). Conversely, phenotypically and taxonomically well distinguish­ able species produce oils of almost identical chemical compositions, such as lemongrass oils from C. citraius and C. fiexuosus (Khanuja et al. 2005). Such phenotypic traits, whether morphological or chemotypic, are basically the phenotypic expression of the genotype, while DNA markers are independent of environment, age, and tissue, and expected to reveal the genetic variation more conclusively in assessing such variations. Introgression of various traits, intermittent mutations, and selection through human intervention may lead to variation in chemotypic characters across geographical distributions (Kuriakose (995). While natural hybridization may lead to the formation of morphological or chemotypic intermediates, defining taxa purely on this basis may not be appro­ priate. Molecular markers provide extensive polymorphism at DNA level used for differentiating closely related genotypes (Pecchioni et al. 1996) and also to find out the extent of genetic diversity (Jain et al. 2003). . Different types of molecular markers have been developed and used in various plant species including grasses in the recent past years. 1.4.1 RANDOMLY AMPLIFIED POLYMORPHIC DNA (RAPD) MARKERS DNA-based markers such as randomly amplified polymorphic DNA (RAPD) (Welsh and McClelland 1990) have been employed not only for cultivar identification but also for phylogenetic and pedigree studies in a number of food, forage, and fiber crops (Chalmers et al. 1992; Kresovich et al. 1994). RAPDs have provided rare biotype specific markers in medicinal and aromatic plants such as vetiver (Adams and Daffern 1997) and Artemisia annua (Sangwan et al. 1999). Randomly primed polymerase chain reaction provides a simple and fast approach to detecting DNA polymorphism, with allelic RAPD marker variations being detected as a plus or minus allele (Welsh and McClelland 1990). In particular, the approach provides multi locus profiling of DNA sequence differences of genotypes when genetic knowledge is lacking. Several studies have been carried out on Cymbopogon species by employing the RAPD approach. A study using RAPD markers was carried out by Shasany and coworkers (2000) to trace the ancestors of cultivar Java II within C. winterianus. The species C. winterianus Jowitt is believed to have originated from the well-known species C. nardus, type Maha Pengiri, referred to as Ceylonese (Sri Lankan) commercial citronella. It was introduced into Indonesia and became commercially known as the Javanese citronella. The Javanese type C. winterianus material was introduced into India for the commercial cultivation of this crop during 1959. Varieties of this species have been developed later by the use of breeding procedures from the same introduced material. The authors carried out a comparative analysis of the morphological characters, chemical traits (oil percent­ age and constituents), and RAPD profiles to assess the diversity and relationships among the Java citronella cultivated forms, which were systematically developed for their suitability to different climatic regions, and also their differences and similarities to the believed parent species C. nardus (Purseglove 1975). All these accessions were analyzed at the molecular level for the similarity and genetic distances through RAPD profiling, using 20 random primers. More than 50% divergence was observed for all the C. winterianus accessions in relation to C. nardus accession CN2. The clus­ tering based on the similarity matrices showed a major cluster of six accessions, consisting of two subclusters. The accession C. nardus CN2 got carved out along with two C. winterianus accessions, CW2 and CW6. On the other hand, the accessions CW2 and CW6 demonstrated distinct identities compared to CN2 at the DNA level (Shasany et al. 2000). 12 Essential Oil-Bearing Grasses: The Genus Cymbopogon The same approach was used by Sangwan et al. (2001) on eleven elite and popular Indian cul­ tivars of Cymbopogon aromatic grasses of essential oil trade types-citronella, palmarosa, and lemongrass. They were characterized by means of RAPOs to discern the extent of diversity at the DNA level between and within the oil biotypes. Primary allelic variability and the genetic bases of the cultivated germplasm were computed through parameters of gene diversity, expected heterozy­ gosity, allele number per locus, SENA, and Shannon's information indices. The allelic diversity was found to be in this order: lemongrass > palmarosa > citronella. Lemongrasses displayed higher (1.89) allelic variability per locus than palmarosa (1.63) and citronella (1040). Also, RAPOs of diag­ nostic and curatorial importance were discerned as "stand-along" molecular descriptors. Principal component analysis (PCA) resolved the cultivars into four clusters: one each of citronella and pal­ rnarosa, and two of lemongrasses (one of C. flexuosus and another of C. pendulus and its hybrid with C. khasianus). Proximity of the two species-groups of lemongrasses was also revealed as they shared the same dimension in the three-dimensional PCA (Sang wan et al. 2001). The same authors analyzed the elite and popular cultivars of C. martinii for genomic and expressed molecular diversity using RAPO, enzyme, and SOS-PAGE protein polymorphisms. The allelic score at each locus of the enzymes, as well as presence and absence profiling in RAPOs, and overall occurrence of band types were subjected to computation of gene diversity, expected heterozygosity, allele number per locus, and similarity matrix. These, in turn, provide inputs to derive primary account of allelic variability, genetic bases of the cultivated germplasm, putative need for gene/trait introgression from the wild or geographically diverse habitat in elite selections. 'PRel' possessed the highest number of unique bands based on RAPD polymorphism. In variety 'IW31245E,' diaphorase and glutamate oxaloacetate transaminase isozymes generated two unique bands as dia-III2 and got-II4. 'RRL(B)77' exhibited three unique bands; one produced by esterase as allele est-III and two by malic enzyme (me-IIIt3). Only one unique band was generated by malic enzyme in variety 'Trishna.' But sofia had three unique bands, two contributed by diaphorase (dia-II3 and dia-II4) and one by glutamate oxaloacetate transaminase (got-II2). SDS-PAGE analysis revealed the presence of unique polypeptide fragments (97.7 to 31.6 kOa) in varieties 'IW31245E,' 'RRL(B)77,' 'Tripta,' 'Trishna,' 'PRCl,' and sofia, generated as a diagnostic marker. In general, molecular distinctions associated with varieties. motia and sofia were clearly noticed in C. martinii (Sangwan et al. 2003). Khanuja et al. (2005) analyzed 19 Cymbopogon taxa belonging to 11 species, 2 varieties, I hybrid taxon, and 4 unidentified species for their essential oil constituents and RAPO profiles to determine the extent of genetic similarity and thereby the phylogenetic relationships among them. Remarkable variation was observed in the essential oil yield ranging from 0.3% in Cymbopogon travancorensis Bor to 1.2% in Cymbopogon martinii (Roxb.) Wats. var. motia. Citral, a major essen­ tial oil constituent, was employed as the base marker for chemotypic clustering. Based on genetic analysis, elevation of Cymbopogon fiexuosus var. microstachys (Hook. E) Soenarko to species sta­ tus and separate species status for C. travancorensis Bor, which has been merged under C.fiexuosus (Steud.) Wats., were suggested toward resolving some of the taxonomic complexes in Cymbopogon. The separate species status for the earlier proposed varieties of C. martinii tmotia and sofia) is fur­ ther substantiated by these analyses. The unidentified species of Cymbopogon have been observed as intermediate forms in the development of new taxa (Khanuja et al. 2005). Somaclonal variants that arise through the tissue culture have been reported in a large number of species. The significance of somaclonal variation in crop improvement depends upon establishing a genetic basis for variation (Larkin and Scowcroft 1981). The use of the molecular marker is becom­ ing widespread for the identification of somaclonal variant. In particular, RAPD markers have proved useful for this purpose owing to its ability to analyze DNA variation at many loci using small amounts of tissue (Munthali et al. 1996; Wallner et al. 1996). Screening of somaclonal variants with improved oil yield and quality have been reported in two species of Cymbopogon, C. winterianus and C. martinii (Mathur et al. 1988; Patnaik et al. 1999), but RAPDs were not used to establish The Genus Cymbopogon popular Indian cul­ ella, palmarosa, and tent of diversity at the d the genetic bases of ity, expectedheterozy­ . The allelic diversity ssesdisplayed higher . Also, RAPOs of diag­ . r descriptors. Principal . h of citronella and pal­ pendulus and its hybrid also revealed as they 2001). rtinii for genomic and in polymorphisms. The e profiling in RAPOs, 'gene diversity, expected turn, provide inputs to ed germplasm, putative bitat in elite selections. lymorphism. In variety sgenerated two unique one produced by esterase band was generated by eontributed by diaphorase , -(12). S-OS-PAGE analysis . ~in varieties 'IW31245E,' stic marker. In general, y noticedin C martinii II species, 2 varieties, I ts and RAPD profiles to 'selationships among them. 0.3% in Cymbopogon tia. Citral,a major essen­ , ustering. Based on genetic 'F.) Soenarko to species sta­ "merged under Cflexuosus eomplexes in Cymbopogon. , ii (motia and sofia) is fur­ pagan have been observed 5). rted in a large number of depends upon establishing a molecularmarker is becom­ lar, RAPD markers have 'on at manyloci using small of somaclonalvariants with . ymbopogon, C winterianus were not used to establish 13 the genetic basis of this somaclonal variation. The paper of Nayak and coworkers (2003) reports the quantitative and qualitative analysis of selected somaclones of jamrosa (a hybrid Cymbopogon) in the field, screening and selection of agronomically useful somaclonal variants with high oil yield and desirable quality, and detection of gross genetic changes through RAPD analysis. In this study. the high oil yield somaclonal variants SCI and SC2 were subjected to RAPD analy­ sis and the result was compared with RAPD profile of the control. A total of 22 arbitrary primers were utilized for initial screening for their amplifying ability. Of these, 12 primers successfully amplified jamrosa DNA with reproducible banding pattern. In general, 2-11 amplified fragments were scored, depending upon primers, ranging in molecular sizes from 266 bp to 1.9 Kb. The test samples SCI, SC2. and the control could be suitably distinguished by the presence of specific mark­ ers or by their absence. Out of the two somaclones analyzed, relatively less distinctness in the ampli­ fied DNA of SC2 was detected using the primers tested. Banding pattern of this somaclone SC2 and the control plant was similar whereas in other variants, somaclone (SCI) DNA polymorphism was observed by having distinct banding pattern. As indicated by RAPDs gross genetic changes have occurred in somaclone (SCl). The results obtained by Nayak and coworkers are in agreement with detection of somaclonal variants by RAPD analysis in Populus deltoides (Rani et al. 1995), garlic (AI-Zahim et al. 1999) and in rice (Yang et al. 1999). Taylor et al. (1995) also reported that RAPD analysis proved suitable for detecting gross genetic changes occurring in sugarcane tissues subjected to prolonged in vitro culture. This work has demonstrated the scope of selecting improved clones of jamrosa with high oil yield and quality through somaclonal variation and suitability of RAPDs for detecting gross genetic changes in somaclonal variants at DNA level. 1.4.2 SIMPLE SEQUENCE REPEAT MARKERS (SSRs) Kumar and coworkers (2007) developed a set of simple sequence repeat markers from a genomic library of Cymbopogon jwarancusa to help in the precise identification of the species (includ­ ing accessions) of Cymbopogon. For this purpose, they isolated 16 simple sequence repeats containing genomic deoxyribonucleic acid clones of C. jwarancusa, which contained a total of 32 simple sequence repeats with a range of I to 3 simple sequence repeats per clone, The major­ ity (68.8%) of the 32 simple sequence repeats comprised dinucleotide repeat motifs followed by simple sequence repeats with trinucleotide (21.8%) and other higher-order repeat motifs. Eighteen (81.8%) of the 22 designed primers for the above simple sequence repeats amplified products of expected sizes, when tried with genomic DNA of C. jwarancusa. Thirteen (72.2%) of the 18 functional primers detected polymorphism among the three species of Cymbopogon (C jlexuosus, C. pendulus, and C. jwarancusai and amplified a total of 95 alleles (range 1-18 alleles) with a PIC value of 0.44 to 0.96 per simple sequence repeat. Thus, the higher allelic range and high level of polymorphism demonstrated by the developed simple sequence repeat markers are likely to have many applications such as in improvement of essential oil quality by authentication of Cymbopogon species and varieties, and mapping or tagging the genes control­ ling agronomically important traits of essential oils, which can further be utilized in marker assisted breeding (Kumar et al. 2007). Considering the high reproducibility and polymorphic nature of the SSRs, the SSRs developed by Kumar and coworkers (2007) may be utilized for identification/authentication of superior accessions/species of the genus Cymbopogon with cor­ rectness and certainty to ensure production of high-quality oil. The SSR markers developed during this kind of study might also be used to resolve the taxonomic disputes, study the genetic diversity, and for genetic mapping and QTL (quantitative trait loci) analysis. The SSRs due to their codominant nature may be specifically useful for the identification of interspecific hybrids, which have been shown to be superior in terms of both their yielding high quantity and better quality of essential oils. 14 1.5 Essential Oil-Bearing Grasses: The Genus Cymbopogon PHYSIOLOGY AND ECOPHYSIOLOGY As discussed, Cymbopogon has a photosynthetic machinery that allows the plant to perform high rates of carbon assimilation and, at the same time, save water. In species that produce essential oil, the biogenesis of terpenoids relies on photosynthetic carbon dioxide reduction on the one hand and availability of water and nutrients, on the other. For this reason several studies have been conducted in order to assess which nutrients and at what conditions were required for an optimal production of both biomass and essential oils. In this section, we will discuss the most important Cymhopogon species in terms of yield of biomass and essential oil production as related to nutrition. Furthermore, when available, references to biotechnological applications will be also reported. 1.5.1 CYMBOPOGON MART/Nil Water requirement, productivity, and water use efficiency of palmarosa (c. martiniis were studied under different levels of irrigation (0.1, 0.3, 0.5, 0.7, 0.9, 1.1, 1.3, and 1.5 IW:CPE ratio). Growth, herb, and essential oil yield increased significantly up to 0.5 IW:CPE ratio. At 0.5 IW:CPE ratio palmarosa produced 47.3 t ha" yr' of fresh herb and 227.3 kg ha' yr' of essential oil. Further increase in irrigation levels caused an adverse effect on growth and yield of palmarosa. Irrigation levels did not affect the quality of oil in terms of its geraniol and geranyl acetate contents. Water requirement of palmarosa was worked out to be 89.1 cm. The highest water use efficiency of 2.97 kg ha- l crrr ' oil was recorded at 0.1 IW:CPE ratio, at 0.5 IW:CPE ratio (optimum) it was 2.55 kg ha" crrr" oil. Irrigation scheduled at 0.5 IW:CPE ratio gave the highest net return of Rs 51 963 ha' yr" (Singh et al. 1997). In C. martinii the application of 160 kg N/ha per year produced the high­ est amount of biomass and essential oil, and increased the net profit and NPK uptake by the crop (Rao et al. 1988); furthermore, dressing of 40 kg K/ha enhanced the yield of biomass by 13.6% and 6.5% and that of oil by 12.9% and 6.1%, compared with 20 and 80 kg Kzha, respectively (Singh et al. 1992). In the same species, harvesting the crop at early seeding (l12-l 15 days after planting) gave 25% more herbage and 5l% more oil yield over harvesting vegetative stage, while the oil so produced had higher content (90.1%) geraniol (Maheshwari et al. 1992). Highest dry-matter yield, essential oil yield, and maximum net return of palmarosa were recorded by applying Azotobacter at 2 kg/ha together with 20 kg N + 20 kg P/ha under rainfed condition in a shallow black soil (Maheshwari et al. 1998). Intercropping of blackgram-blackgram or sorghum fodder-ratoon with palmarosa gave additional yields of 660 kg/ha seed and 16.6 t/ha fodder, respectively, compared with the sole crop of palmarosa (Rao et al. 1994). Moreover, sowing of pigeon pea in alternate rows parallel to palmarosa proved most efficient and economic, as it provided higher economic returns, bonus income, and monetary advantage, and the oil content and quality in terms of total geraniols of palmarosa were not adversely affected by adoption of intercropping (Maheshwari et al. 1995). However, in palmarosa-pigeon pea intercropping systems, competition exists mainly for light rather than for nutrients and moisture, possibly because the two crop components acquire their nutrients and moisture from different soil layers (Singh et al. 1998). Concerning essential oil production of palmarosa, changes in fresh weight, dry weight, chlorophyll, and essential oil content and its major constituents, such as geraniol and geranyl acetate, were examined for both racemes and spathe at various stages of spikelet development (Dubey et al. 2000). The essential oil content was maximal at the unopened spikelets stage and decreased significantly thereafter. At unopened spikelets stage, the proportion of geranyl acetate (58.6%) in the raceme oil was relatively greater compared with geraniol (37.2%), whereas the spathe oil contained more geraniol (61.9%) compared with geranyl acetate (33.4%). The relative percentage of geranyl acetate in both the oils, however, decreased sig­ nificantly with development, and this is accompanied by a corresponding increase in the percentage of geraniol. Analysis of the volatile constituents from racemes and spathes (from mature spikelets) and seeds by capillary GC indicated 28 minor constituents besides the major constituent geraniol. (E)-Nerolidol was detected for the first time in an essential oil from this species. The geraniol ~ The Genus Cymbopogon 'the plant to perform high 'that produce essential oil, tion on the one hand and ies have been conducted for an optimal production important Cymbopogon to nutrition. Furthermore, rted. (c. martiniit were studied 1.5 IW:CPE ratio). Growth, ratio. At 0.5 IW:CPE ratio I of essential oil. Further d of palmarosa. Irrigation 'yl acetate contents. Water r use efficiency of 2.97 kg . imum) it was 2.55 kg na' return of Rs 51 963 ha' r year produced the high­ NPK uptake by the crop ld of biomass by 13.6% and K/ha, respectively (Singh 12-115 days after planting) ive stage, while the oil so ), Highest dry-matter yield, by applying Azotobacter 'on in a shallow black soil rghum fodder-ratoon with r, respectively, compared pigeon pea in alternate rows higher economic returns, y in terms of total geraniols (Maheshwari et al. 1995). exists mainly for light rather ents acquire their nutrients essential oil production of tial oil content and its major both racemes and spathe at ial oil content was maximal . ,At unopened spikelets stage, ively greater compared with ,9%) compared with geranyl oils, however, decreased sig­ ing increase in the percentage thes (from mature spikelets) major constituent geraniol. this species. The geraniol 15 content predominated in the seed oil, whereas the geranyl acetate content was higher in the raceme oil (Dubey et al. 2000). Biotechnology is a powerful and consolidated technique for understanding plant growth and development as well as for improving biomass and yield of crops. Callus could be induced from nodal explant of mature tillering plant of C. martinii in different basal media supplemented with 2,4-dichlorophenoxy acetic acid (2,4-D) and kinetin (Kin). Shoot bud was regenerated from such calli in MS and B5 basal media modified with various combinations of phytohormones, vitamins, and amino acids. Root formation was induced either in white basal medium or half-strength MS or B5media containing naphthalene acetic acid (NAA) or indole-3-butyric acid (IBA). High survival percentage of regenerated plants in soil was obtained after acclimatization in normal environment (Baruah and Bordoloi 1991). A detailed characterization of chromosomal status was carried out in callus, somatic embryos, and regenerants derived from in vitro cultured nodal and inflorescence explants of C. martinii (2n = 20). Both the callus lines revealed considerable ploidy variations (tet­ raploids to octoploids and hyperoctoploids), and the degree of polyploidization increased with the culture age.Prequencies of various polyploid cells were significantly higher in nodal callus lines (3.6% to 46.3%) than the inflorescence callus lines (1.9% to 23.6%) when analyzed over 520 days of culture. Somatic embryos derived from both the callus lines retained a predominantly diploid chromosome status throughout (99.0% to 93.1%). Root tip analysis of about 70 regenerants ran­ domly taken from cultures of various ages (days 20 to 520) revealed only diploid chromosome numbers (2n = 20) implying a strong relative stability of diploidy among the regenerants (Patnaik et al. 1996). Chromosome counts of cells in suspensions, calli, and somatic embryos derived from cultures of different ages revealed the presence of diploids, tetraploids, and octaploids (Parnaik et al. 1997). Sodium chloride tolerant callus lines of C. martinii were obtained by exposing the cal­ lus to increasing concentrations of NaCI (0-350 mM) in the MS medium. The tolerant lines grew better than the sensitive wild-type lines in all concentrations of NaCl tested up to 300 mM. Callus survival and growth were completely inhibited, resulting in tissue browning and subsequent death at 350 mM NaCl. The selected lines retained their salt tolerance after 3-4 subcultures on salt-free medium, indicating the stability of the induced salt tolerance. The growth behavior, the Na", K+, and proline contents of the selected callus lines were characterized and compared with those of the NaCl-sensitive lines. The Na" levels increased sharply, while the K+ level declined continu­ ously with the corresponding increase in external NaCI concentrations in both lines, but the NaCl­ tolerant callus lines always maintained higher Na" and K+ levels than that of the sensitive lines. The NaCI-selected callus line accumulated high levels of proline under salt stress. The degree of NaCl tolerance of the selected lines was in negative correlation with the K+/Na+ ratio and in positive correlation with proline accumulation (Patnaik and Debata 1997a). The embryogenic potential of NaCl-tolerant callus selected even at 300 mM could be improved significantly by the incorporation of gibberellic acid (GA(3)) and abscisic acid (ABA), in the medium where, with 2 mg/L of GAOl and I mg/L of ABA, the highest rates of embryogenesis (44.5%, 28.8%, and 18.6%) were achieved against 17.5%,8.2%, and 1.8% on medium devoid of GAiJ), and ABA at 50%, 150%, and 250 mM of NaCl, respectively (Patnaik and Debata 1997b). Finally, plants regenerated from cell suspension cultures of palmarosa were analyzed for somaclonal variation in five clonal generations. A wide range of variation in important quantitative traits, for example, plant yield, height, tiller number, oil content and qualitative changes in essential oil constituents geraniol, geranyl acetate, geranyl formate, and linalool, were observed among the 120 somaclones screened. Eight somaclones were selected on the basis of high herb and oil yield over the donor line and high geraniol content in the oil. Based on performance in the field trials, three superior lines were selected, and maintained for five clonal generations. The superior lines exhibited a reasonable degree of stability in the traits selected (Patnaik et al. 1999). Palmarosa was also found to be associated with a vesicular-arbuscular mycorrhizal (VAM) fungus, Glomus aggregatum. Glasshouse experiments showed that inoculation of palmarosa with G. aggregatum caused a twofold and threefold biomass production as compared to nonmycorrhizal Essential Oil-Bearing Grasses: The Genus Cymbopogon 16 plants. These findings indicate the potential use of YAM-fungi for improving the production of this essential oil-bearing plant (Gupta and Janardhanan 1991). Furthermore, when the interactive effects of phosphate solubilizing bacteria, N-2 or fixing bacteria, and arbuscular mycorrhizal fungi (AMF) were studied in a low phosphate alkaline soil amended with a tricalcium insoluble source of inor­ ganic phosphate on the growth of C. martinii, The rhizobacteria behaved as a "mycorrhiza helper" and enhanced root colonization by G. aggregatum in presence of tricalcium phosphate at the rate of 200 mg kg:' soil (PI level) (Ratti et a1. 20m). A dramatic increase in PEP carboxylase activity and oil biosynthesis was observed under drought conditions in C. martinii (Sangwan et al. 1993). The physiological and biochemical basis of drought tolerance in C. martinii has been elucidated on the basis of growth and metabolic responses (Fatima et a1. 2002). 1.5.2 CYMBOPOGON FLEXUOSUS Cymbopogon fiexuosus (also known as 1emongrass) is a perennial, multicut aromatic grass that yields an essential oil used in perfumery and pharmaceutical industries and vitamin A. It has a long initial lag phase. The growth and herbage and oil production of C. jiexuosus in response to different levels of irrigation water (IW) 0.1, 0.3, 0.5, 0.7, 0.9, 1.1, 1.3, and 1.5 times cumulative pan evaporation CPE evaluated on deep sandy soils showed that an increment in the level of irriga­ tion increased the plant height up to 0.7 IW:CPE ratio. However, the response of irrigation levels on tiller production of lemongrass differed with the season of harvest. Oil content had an inverse relationship with the levels of irrigation, whereas significantly higher herb and essential oil yields were recorded at 0.7 IW:CPE ratio, irrespective of season of harvest (Singh et al. 2000). Application of nitrogen (0, 50, 100, and 150 kg N ha- l yr ') and phosphorus to C. fiexuosus crops maintained the fertility of the soil, while potassium depletion was noticed (Singh 2001). When the effects of phosphorus (at 0, 17.75, and 35.50 kg ha- l yell, potassium (at 0,33.2,66.4, and 99.6 kg ha- l yr') and nitrogen (at 100 and 200 kg ha- l yell and potassium (at 0, 33.2 and 66.4 kg ha- l yr") were studied on herbage and oil yield of Ci fiexuosus. it was found that plants produced significantly higher herb­ age and oil yields compared with controls (Singh et al. 2005; Singh and Shivaraj 1999). Spraying of iron-complexed additives on C. flexuosus increased iron translocation and the dry-matter produc­ tion. Application of iron chelates and salts increased the vegetative herb yield, and oil and citral content. While maximum geraniol and less citral were obtained in the chlorotic plants, Fe recov­ ered plants possessed more citral and less geraniol. The maximum recovery of total chlorophyll and nitrate reductase activity were recorded in the crop when Fe-EDTA chelates were sprayed at 22.4 ppm (Misra and Khan 1992). In Ci jiexuosus. a closer plant spacing of 45 x 45 ern resulted in higher herb and oil yields compared to wider spacing of60 x 60 em. Application of 150 kg N ha- l yr- l resulted in higher herb and oil yields. Higher nitrogen applications also increased the plant height and number of tillers per clump. The oil content and quality were not influenced by spacing and nitrogen levels (Singh et al. 1996b). As for C. martinii, intercropping of C. flexuosus with the food legumes such as blackgram (Vigna mungo (L) Hepper), cowpea (Vigna unguiculata (L) Walp), or soybean (Glycine max (L) Merr.) prompted extra yields over and above that of pure cultures, without affecting the oil yield (Singh and Shivaraj 1998). The influence of different foliar applications of the triacontanol (Tria.)-based plant growth regu­ lator Miraculan on growth, CO 2 exchange, and essential oil accumulation in C. fiexuosus showed increased rates in plant height, tillers per plant, biomass yield, accumulation of essential oil, net CO 2, and exchange and transpiration compared to the untreated control, but the number of leaves per tiller remained unaffected. Application of Miraculan also increased micronutrient uptake and total chlorophyll and citral content but decreased chlorophyll alb ratio and stomatal resistance. Increase in shoot biomass, photosynthesis, and chlorophyll were significantly correlated with essential oil content (Misra and Srivastava 1991). Only young and rapidly expanding C. fiexuosus leaves were 1.5.3 . The Genus Cymbopogon ing the production of this whenthe interactive effects 'mycorrhizal fungi (AMF) insoluble source of inor­ as a "mycorrhiza helper" m phosphate at the rate of f r sis was observed under ogical and biochemical 's of growth and metabolic lticut aromatic grass that and vitamin A. [t has a C. flexuosus in response to 1.5 times cumulative pan nt in the level of irriga­ ponse of irrigation levels Oil content had an inverse herb and essential oil yields , h et aL 2000). Application "jtexuosus crops maintained 2001). When the effects of .A. and99.6kg ha' yr ') and 4 kg ha' yr') were studied significantly higher herbShivaraj 1999). Spraying of and the dry-matter produc­ b yield, and oil and citral chlorotic plants, Fe recov­ overy of total chlorophyll A chelates were sprayed at :, g of 45 x 45 em resulted in ;Application of 150 kg N ha' ns also increased the plant not influenced by spacing legumes such as blackgram .,. soybean (Glycine max (L) ithout affecting the oil yield f ·a.)-based plant growth regu­ 'on in C. flexuosus showed ulation of essential oil, net butthe number of leaves per . 'icronutrient uptake and total stomatal resistance. Increase correlated with essential oil ing C. flexuosus leaves were The Genus Cymbopogon 17 found to have the capacity to synthesize and accumulate essential oil and citral. The pattern of the ratio of the label incorporated in citral to that in geraniol, during leaf ontogeny, evinced parallelism with the geraniol dehydrogenase activity. The elevated levels of glucose-6-phosphate dehydrogenase, 6-phosphogluconate dehydrogenase, NAOP"-malic enzyme, and NAOP"-isocitrate dehydrogenase coincided with the period of active essential oil biogenesis accompanying early leaf growth (Singh et al. 1990). Thus, there is an active involvement of oxidative pathways in essential oil biosynthesis. The time-course (12 h light followed by 12 h dark) monitoring of the C-14 radioactivity in starch and essential oil, after exposure of the immature (\5 days after emergence) leaf to (CO)-C-14, revealed a progressive loss of label from starch and a parallel increase in radioactivity in essential oil. Thus, there was indication of a possible degradation of transitory starch serving as the source of carbon precursor for essential oil (monoterpene) biogenesis in the tissue (Singh et al. 1991). Biotechnological applications revealed that C. fiexuosus plants derived from somatic embryoids were more uniform in all the characteristics examined when compared with the field performance of plants raised through slips by standard propagation procedures (Nayak et al. 1996). A fungal endophyte, Balansia sclerotica (Pat.) Hohn., has been found to establish a perennial association with the commercially grown East Indian C. fiexuosus cv. Kerala local (syn. = 00-19). Endophyte-infected plants produced 195% more shoot biomass and 185% more essential oil than the endophyte-free control plants when grown experimentally under glasshouse conditions. The essential oil extracted from the endophyte-infected plants is qualitatively identical with that of endophyte-free plants and is free of toxic ergot alkaloids. Thus, B. sclerotica-infected East Indian C. flexuosus has potential for agricultural exploitation (Ahmad et al. 2001). 1.5.3 CYMBOPOGON WINTERIANUS Java citronella (c. winterianus) is a perennial, multiharvest aromatic grass, the shoot biomass of which, on steam distillation, yields an essential oil extensively used in fragrance and flavor industries. Fresh C. winterianus (Java citronella) herbage and essential oil yields were significantly influenced by application of N up to 200 kg ha:' yr', while tissue N concentration and N uptake increased only to ISO kg N ha", The oil yields with neem cake-coated urea (urea granules coated with neem cake) and urea super granules were 22 and 9% higher over that with prilled urea, and urea supergranules were significantly increased up to 200 kg N ha' while with neem cake-coated urea, response was observed only to 150 kg N ha'! Estimated recovery of N during two years from neem cake-coated urea, urea supergranules, and prilled urea were 38%. 31%, and 21%, respectively (Singh and Singh 1992).The interaction between N doses and nitrification inhibitors was also significant. Nitrification inhibitors performed better at the highest N dose (450 kg N ha- ' yr"), and the increase in the essen­ tial oil yields was to an extent of 27.3% to 34.6% when compared with "N alone" treatment. The nitrification inhibitors also increased the apparent N recoveries by citronella considerably. The oil content in the herb and its quality were not affected by the treatments. The nitrification inhibitors increased citronella yields and improved N economy (Puttanna et al. 2001). In Java citronella sig­ nificant positive correlations were observed between fresh matter, citronellol content, dry and fresh matter yields, and total essential oil content (Omisra and Srivastava 1994). When the effect of depth (25,37.5, and 50 mm) and methods (ridge and furrow, and broad bed and furrow method) of irriga­ tion acid nitrogen levels (0, 200, and 400 kg N ha' yr') were studied on herb and oil yields of Java citronella, highest herb and oil yields were achieved with the application of 400 kg N, maintaining 25 mm depth of irrigation, while the content and quality of oil were not affected either by irrigation or nitrogen (Singh et al. 1996a). Among food legumes, greengram (Vigna radiata (L.) Wilez.), and among vegetables, c1usterbean (Cyamopsis psoraloides O. c., syn. Cyamopsis tetragonoloba (L.) Taub.), tomato (Lycopersicon esculentum Mill.) and lady's finger (Abelmoschus esculentus Moench.) as intercrops of C. win­ terianus did not decrease its biomass and essential oil yield and produced bonus yields of these crops over and above that of Java citronella. Maximum monetary returns were recorded by Java Essential Oil-Bearing Grasses: The Genus Cymbopogon 18 citronella intercropped with tomato or greengram. However, Java citronella intercropped with red­ gram iCajanus cajan (L.) Millsp.), horsegram (Macrotyloma uniflorum (Lam.) Verd, syn. Dolichos biflorus Roxb.), and brinjal (Solanum melongena L.) suffered significant biomass and essential oil yield reductions. Horsegram proved to be the most competitive iruercrop, producing least yields and minimum monetary returns (Rao 2000). Changes in the utilization pattern of primary substrate, viz. [U-C-14] acetate. (C0 2)-C-14 and [U-C-14] saccharose, and the contents of C-14 fixation products in photosynthetic metabo­ lites (sugars. amino acids, and organic acids) were determined in Fe-deficient Java citronella in relation to the essential oil accumulation. An overall decrease in photosynthetic efficiency of the Fe-deficient plants as evidenced by lower levels of incorporation into the sugar fraction and essen­ tial oil after (C0 2)-C-14 had been supplied was observed. When acetate and saccharose were fed to the Fe-deficient plants, despite a higher incorporation of label into sugars, amino acids, and organic acids, there was a lower incorporation of these metabolites into essential oils than in control plants. Thus, the availability of precursors and the translocation to a site of synthesis/accumulation, severely affected by Fe deficiency, is equally important for the essential oil biosynthesis in citro­ nella (Srivastava et al. 1998). Lal and coworkers (2001) observed that improvement of oil quality with high citronellal content and low elemol content in Java citronella is believed to be achievable. although some compromise will have to be made in oil yield. Nutrient acquisition and growth of Java citronella was also studied in a P-deficient sandy soil to determine the effects of mycorrhizal symbiosis and soil compaction. When a pasteurized sandy loam soil was inoculated either with rhizosphere microorganisms excluding VAM fungi (nonmycor­ rhizal) or with the VAM fungus, Glomus intraradices Schenck and Smith (mycorrhizal) and sup­ plied with 0, 50, or 100 mg P kg' soil, G. intraradices was found to substantially increase root and shoot biomass, root length, nutrient (P, Zn, and Cu) uptake per unit root length, and nutrient concen­ trations in the plant, compared to inoculation with rhizosphere microorganisms when the soil was at the low bulk density and not amended with P. Little or no plant response to the VAM fungus was observed when the soil was supplied with 50 or 100 mg P kg:" soil and/or compacted to the highest bulk density. At higher soil compaction and P supply, the VAM fungus significantly reduced root length. Nonmycorrhizal plants at higher soil compaction produced relatively thinner roots and had higher concentrations and uptake of P, Zn, and Cu than at lower soil compaction, particularly under conditions of P deficiency (Kothari and Singh 1996). Pythium aphanidermatum was the predominant fungus recovered from the roots of Java citro­ nella showing lethal yellowing in the northern part of India. Roots of infected plants showed marked discoloration, and the cortical region was completely disintegrated and sloughed from the vascular tissue. Diseased plants were chlorotic and stunted. Rotting was often found to spread from roots to stem, leading to severe chlorosis and death of the infected plants. The pathogenicity of the fungus was established. The disease is a potential constraint to citronella cultivation in nonarid climates where the crop is irrigated extensively (Alam et al. 1992). Another disease affecting commercial plantations of Java citronella is a collar rot and wilt disease. The causal organism was identified as Fusarium moniliforme, anamorph of Gibberellafujikuroi. Isolates of the pathogen differed in their pathogenicity on the host plant under glasshouse conditions. Differences were also observed in growth rates, pigment production. and sporulation between isolates (Alam et al. 1994). 1.5.4 OTHER CYMBOPOGON SPECIES Application of graded levels of lime up to 10 t/ha on acid soil (pH 4.2) raised the pH up to 6.7. It increased the dry herbage of C. khasianus linearly. Increase of soil pH decreased N, P, K, Fe, and Zn contents in dry herbage significantly but increased the Ca and Mg contents. Liming showed a positive effect on the uptake of N, P, K, and Ca. However, Fe and Mg declined beyond lime levels of 23 and 5.0 t, respectively. Uptake of Zn was found fluctuating. Oil content (2.00%-2.07%; DWB) and geraniol (80.2%-81.0%) in the oil were unaffected by the lime treatments (Choudhury and •The Genus Cymbopogon lla intercropped with red­ .(Lam.) Verd, syn. Dolichos biomass and essential oil producing least yields and " . -14] acetate, (C0 2)-C-14 in photosynthetic metabo­ ficient Java citronella in nthetic efficiency of the sugar fraction and essen­ and saccharose were fed sugars, amino acids, and ntialoils than in control of synthesis/accumulation, oil biosynthesis in citro­ improvement of oil quality believed to be achievable, 'in a P-deficient sandy soi I ,When a pasteurized sandy ing YAM fungi (nonmycor­ ith (mycorrhizal) and sup­ . tantially increase root and length, and nutrient concen­ isms when the soil was at to the YAM fungus was compacted to the highest significantly reduced root ively thinner roots and had action, particularly under m the roots of Java citro­ ted plants showed marked sloughedfrom the vascular nd to spread from roots to pathogenicity of the fungus ivation in nonarid climates isease affecting commercial organism was identified as pathogendiffered in their es were also observed in et al. 1994). 2) raised the pH up to 6.7. It decreased N, P, K, Fe, and contents. Liming showed a declined beyond lime levels ntent(2.00%-2.07%; OWB) .treatments (Choudhury and The Genus Cymbopogon 19 Bordoloi 1992). Experiments were also conducted to measure the rate of C. caesius litter decompo­ sition and to identify fungal flora associated with the litter during different stages of decomposition in a tropical grassland. Rate of litter decomposition was several times higher than in temperate grasslands. Buried litter decayed more rapidly, and this rate was not influenced by climatic condi­ tions. In contrast, surface litter recorded a lower decomposition rate, which was dependent on tem­ poral (seasonal) fluctuations. Total nitrogen, available phosphorus, and potassium contents of the stem litter decreased during the initial stages of incubation. Thirty-five species of fungi were isolated from the litter during the different stages of litter degradation. Most belonged to Hyphornycetes, which are active decomposers (Senthilkumar et al. 1992). C. nardus var. confertiflorus and C. pendulus were grown under mild and moderate water stress for 45 and 90 d to investigate the impact of in situ drought stress on plants in terms of relative water content, psi, concentration of proline, activities of PEP carboxylase and geraniol dehydroge­ nase, and geraniol and citral biogenesis. The results revealed that the species exhibited differential responses under mild and moderate stress treatments. In general, plant growth was reduced consid­ erably, while the level of essential oils was maintained or enhanced. Significant induction in catalytic activity of PEP carboxylase under water stress was one of the consistent metabolic responses of the aromatic grasses. The major oil constituents, geraniol and citral, increased substantially in both the species. Activity of geraniol dehydrogenase was also modulated under moisture stress. The responses varied depending upon the level and duration of moisture stress. The observations have been analyzed in terms of possible relevance of some of these responses to their drought stress adaptability/tolerance (Singhsangwan et al. 1994). In vitro plants of C. citratus were established, starting from shoot apices derived from plants cultivated under field conditions. The effect of the immersion frequency (two, four, and six immersions per day) on the production of biomass in temporary immersion systems (TIS) of I L capacity was stud­ ied. The highest multiplication coefficient (12.3) was obtained when six immersions per day were used. The maximum values of fresh weight (FW; 62.2 and 66.2 g) were obtained with a frequency of four and six immersions per day, respectively. However, the values for dry weight (OW; 6.4 g) and height (8.97 ern) were greater in the treatment with four immersions per day. The TIS used in this work for the production of lemongrass biomass may offer the possibility of manipulating the culture parameters, which can influence the production of biomass and the accumulation of sec­ ondary metabolites. We describe for the first time the in vitro production of Cymbopogon citratus biomass in TIS. In vitro regeneration of C. polyneuros was obtained through callus culture using leaf base, node, and root as explants. Callus was induced from different explants with 2-5 mg/L alpha-naphthalene acetic acid (NAA) and 1-2 mg/L kinetin in Murashige and Skoog's (MS) basal medium. High frequency shoots were noticed from leaf-base callus supplemented with 3.5 mg/L 6-benzylaminopurine (BA), I-arginine, adenine, and a low level of NAA (0.2 mg/L). About 80-85 shoot buds were obtained from ca. 200 mg of callus per culture. The individual shoots produced root in the presence of 0.5-3 mg/L indole 3-butyric acid or its potassium salt. Regenerated plants were cytologically and phenotypically stable. Regenerants were transplanted into soil and subsequently transferred to the field (Das 1999). C. nardus could be propagated via tissue culture using axillary buds as explants. The aseptic bud explants obtained using double steril­ ization methods produced stunted abnormal multiple shoots when they were cultured on Murashige and Skoog (MS) medium supplemented with 1.0 mg L-I or 2.0 mg L-I benzyladenine (BA). Stunted shoots that cultured on MS + 1.0 mg L -I RA + 1.0 mg L-I N-6-isopentenyl-adenine (2iP) could induce elongation of shoots from about 60% of the stunted shoots. Normal multiple shoots could be induced at the highest (19.7 shoots per bud) from the bud explants within 6 weeks when cul­ tured on proliferation medium consisted of MS supplemented with 0.3 mg L -I BA and 0.1 mg L-I indole-3-butyric acid (IBA). The separated individual shoot produced roots when transferred to basic MS solid medium. The essential oils that were contained in the mature plants namely citro­ nellal, geraniol, and citronellol, were also found in the in vitro C. nardus plantlets. Citronellal was the main essential oil component in the matured plants, while geraniol was the main component in 20 Essential Oil-Bearing Grasses: The Genus Cymbopogon the in vitro plant lets (Chan et al. 2005). The occurrence, mode of infection, and the extent of dam­ age caused by Psilocybe kashmeriensis sp. nov. Abraham on oil grass C. jawarancusa in Kashmir valley is discussed herein. A brief description of the new agaric species is also offered (Abraham 1995). Dormant vegetative slips of jamrosa (c. nardus var. confertifiorus x C. jwarancusa) were subjected to various doses of gamma rays. Plants raised from them were screened with a view to iso­ late improved clones of the crop. Five mutant clones isolated exhibited variation in quality/quantity of essential oil. These changes in oil characters were attributed to microlevel mutations induced by gamma rays (Kak et al. 2000). REFERENCES Abraham SP. 1995. Notes on the occurrence of an unusual agaric on Cymbopogon in Kashmir Valley. Nova Hedwigia 60: 227-232. Adams RP, Dafforn MR. 1997. DNA sampling of the pan-tropical vetiver grass uncovers genetic uniformity in erosion control germplasm. Diversity 13: 27,28. Ahmad A, Alam M, Janardhanan KK. 200 I. Fungal endophyte enhances biomass production and essential oil yield of East Indian lemongrass. Symbiosis 30: 275-285. Al-Zahim MA, Ford-Lloyd BV, Newbury HJ. 1999. Detection of somaclonal variation in garlic tAlllum sativum L.) using RAPD and cytological analysis. Plant Cell Reports 18: 473-477. Alam M, Chourasia HK, Sattar A, Janardhanan KK. 1994. Collar rot and wilt-a new disease of Java-citronella (Cymbopogon winterianuss caused by Fusarium-moniliforme sheldon. Plant Pathology 43: 1057-1061. Alarn M, Sattar A, Janardhanan KK, Husain A. 1992. Lethal yellowing of Java citronella (Cymbopogon winte­ rianusi caused by Pvthium aphanidermatum. Plant Disease 76: 1074-1076. Ashton AR. 1990. Enzymes of C, photosynthesis. In Dey PMHJB, Ed. Methods in Plant Biochemistry. London: Academic Press, pp. 39-72. Baruah A, Bordoloi ON. 1991. Growth and regeneration of palmarosa (Cymbopogon martinii [Roxb.] Wats.) callus tissues under varied nutritional status. Indian Journal of Experimental Biology 29: 582. 583. Bertea CM, Scannerini S, 0' Agostino G, Mucciarelli M, Camusso W, Bossi S, Buffa G, Maffei M. 200 I. Evidence for a C4 NADP-ME photosynthetic pathway in vetiveria rizanioides Stapf. Plant Biosystems 135: 249-262. Bertea CM, Tesio M, 0' Agostino G, Buffa G, Camusso W, Bossi S, MucciareJli M, Scannerini S, Maffei M. 2003. The C-4 biochemical pathway, and the anatomy of lemongrass tCymbopogon citratus (DC) Stapf.) cultivated in temperate climates. Plant Biosystems 137: 175-184. Breakwell E. /914. A study of the leaf anatomy of some native species of the genus Andropogon, N.O. Gramineae. Proceedings of the Linnean Society, N.S. W 39: 385-394. Casati P, Spampinato CP, Andreo CS. 1997. Characteristics and physiological function of NADP-malic enzyme from wheat. Plant Cell Physiology 38: 928-934. Chalmers KJ, Waugh R, Simons AJ, Powell W. 1992. Detection of genetic variations between and within popu­ lations of Gliricidia sepium and C. maculata using RAPD markers. Heredity 69: 465-472. Chan LK, Dewi PR, Boey PL. 2005. Effect of plant growth regulators on regeneration of plantlets from bud cultures of Cymbopogon nardus L. and the detection of essential oils from the in vitro plantlets. Journal (If Plant Biology 48: 142-146. Chapman KSR, Hatch MD. 1983. Intracellular location of phosphoenolpyruvate carboxykinase and other C 4 photosynthetic enzymes in mesophyll and bundle sheath protoplasts of Panicum maxymum. Plant Science Letters 29: 145-154. Choudhury SN, Bordoloi ON. 1992. Effect of liming on the uptake of nutrients and yield performance of Cymbopogon khasianus in acid soils of Northeast India. Indian Journal ofAgronomy 37: 518-522. Das AB. 1999. Effects of different physiochemical factors on regeneration of Cymbopogon polyneuros Stapf. via callus culture and subsequent chromosomal stability. Israel Journal of Plant Sciences 47: 195-198. Dengler NG, Nelson T. 1999. Leaf structure and development in C4 plants. In Sage RF, Monson RK, Eds. The Biology of C4 Plants. San Diego: Academic Press, pp. 133-172. Dubey VS, Mallavarapu GR, Luthra R. 2000. Changes in the essential oil content and its composition during palmarosa (Cymbopogon martinii (Roxb.) Wats. var. motia) inflorescence development. Flavour and Fragrance Journal 15: 309-314. Eastman PAK, Dengler NG, Peterson CA. 1988. Suberized bundle sheaths in grasses (Poaceae) of different photosynthetic types I. Anatomy, ultrastructure, and histochemistry. Protoplasma 142: 92-111. Edwards GE, Andreo CS. 1992. NADP-Malic enzyme from plants. Phytochemistry 31: 1845-1857. The Genus Cymbopogon extent of dam­ sa in Kashmir ered (Abraham arancusa) were ith a view to iso­ quality/quantity ions induced by of Java-citronella gy43: 1057-1061. , Cymbopogonwinte­ inii [Roxb.] Wats.) ., 29: 582, 583. . M.2001. Evidence terns 135: 249-262. nerini S, Maffei M. ~itratus (DC) Stapf.) n and within popu­ H72. Ill'plantlets from bud o plantlets. Journal .yield performance of y37: 518-522. polyneuros Stapf. iences47: 195-198. 'MonsonRK, Eds. The (Poaceae) of different 142: 92-111. 1845-1857. 21 Edwards GE, Wal ker OA. 1983. CJ. C4: Mechanism. and Cellular and Environmental Regulation of Photosynthesis. London: Blackwell Scientific Publications. Fatima S, Farooqi AHA, Sharma S. 2002. Physiological and metabolic responses of different genotypes of Cvmbopogon martinii and C. winterianus to water stress. Plant Growth Regulation 37: 143-149. Fitter AH, Hay RKM. 1987. Environmental Physiology of Plants. London: Academic Press. Ghannuum O. van Caemmerer S, Conroy JP. 200 I. Carbon and water economy of Australian NAD-ME and NADP-ME C4 grasses. Australian Journal of Plant Physiology 28: 213-223. Guenther E. 1950a. The Essential Oils (TV). New York, Van Nostrand. Ref Type: Serial (Book, Monograph). Guenther E. 1950b. The Essential Oils. Princeton: Van Nostrand. Gupta ML, Janardhanan KK. 1991. Mycorrhizal association of Glomus aggregatum with palrnarosa enhances growth and biomass. Plant and Soil 131: 261-263. Gutierrez M, Gracen VE, Edwards GE. 1974. Biochemical and cytological relationships in C 4 plants. Planta 119: 279-300. Hatch MD. 1988. C 4 photosynthesis: A unique blend of modified biochemistry, anatomy and ultrastructure. Biochimica et Biophysica Acta 895: 81-106. Hatch MD. 1992. C 4 photosynthesis: An unlikely process full of surprises. Plant and Cell Physiology 33: 333-342. Hatch MD. Kagawa T, Craig S. 1975. Subdivision ofCj-pathway species based on differing C 4 acid decarboxy­ lating systems and ultrastructural features. Australian Journal of Plant Physiology 2: 111-128. Hatch MD, Kagawa T, Craig S. 2007. Subdivision of C 4-pathway species based on differing C 4 acid decarboxy­ lating systems and ultrastructural features. Australian Journal of Plant Physiology 2: 111-128. Hattersley PW, Watson L. 1976. C 4 grasses: An anatomical criterion for distinguishing between NADP-malic . enzyme species and PCK or NAD-malic enzyme species. Australian Journal of Botany 24: 297-308. HuangY,Street-Perrot FA, Metcalfe SE, Brenner M, Moreland M, Freeman KH. 200 I. Climate change as the dom­ inant control on glacial-interglacial variations in C3 and C4 plant abundance. Science 293: 1647-1651. lain N, Shasany AK, Sundaresan V, Rajkumar S, Darokar MP. Bagchi GO, Gupta AK. Kumar S, Khanuja SPS. 2003. Molecular diversity in Phyllanthus amarus assessed through RAPD analysis. Current Science 85: 1454--1458. Jenkins CLD, Furbank RT, Hatch MD. 1989. Mechanism of C 4 photosynthesis. Plant Physiology 91: 1372-1381. Kak SN, Bhan MK, Rekha K. 2000. Development of improved clones of jamrosa (Cymbopogon nardus [L.] Rendle var. confertijlorus [Steud.] Bor. x C. jwarancusa Jones Schult.) through induced mutations. Journal of Essential Oil Research 12: 108-110. Khanuja SPS, Shasany AK, Pawar A. Lal RK, Darokar MP, Naqvi AA, Rajkumar S, Sundaresan V, Lal N, Kumar S. 2005. Essential oil constituents and RAPD markers to establish species relationship in Cymbopogon Spreng. (Poaceae). Biochemical Systematics and Ecology 33: 171-186. Kothari SK, Singh UB. 1996. Response of lava citronella (Cymbopogon winterianus Jowitt) to VA mycorrhizal fungi and soil compaction in relation to P supply. Plant and Soil 178: 231-237. Kresovich S, Lamboy RL, Li R. Szewc-McFadden AK, Blick SM. 1994. Application of molecular methods and statistical analysis for discrimination of accessions and clones of vetiver grass. Crop Science 34: 80S-809. Kumar I, Verma V, Shahi AK, Qazi GN, Balyan HS. 2007. Development of simple sequence repeat markers in Cymhopogon species. Planta Medica 73: 262-266. KumarS. Dwivedi S, Kukreja AK, Sharma IR, Bagchi GD. 2000. Cymhopogon: The Aromatic Grass. Lucknow, India: Central Institute of Medicinal and Aromatic Plants. Kuriakose KP. 1995. Genetic variability in East Indian lemongrass (Cymbopogon flexuosus Stapf). Indian Perfumer 39: 76--83. Lal RK, Sharma IR, Misra HO, Sharma S, Naqvi AA. 200\. Genetic variability and relationship in quantita­ tive and qualitative traits of Java citronella (Cymbopogon winterianus Jowitt). Journal of Essential Oil Research 13: 158-162. Larkin Pl. Scowcroft WR. 1981. Somaclonal variation a novel source of variability from cell cultures for plant improvement. Theoretical and Applied Genetics 60: 197-214. LeegoodRC. 1993. Carbon dioxide-concentrating mechanisms. In Lea PI, Leegood RC, Eds. Plant Biochemistrv and Molecular Biology. Chichester, U.K: Wiley & Sons, pp. 47-72. LunnJE, Agostino A, Hatch MD. 1995. Regulation of NADP-malate dehydrogenase in C 4 plants-activity and properties of maize thioredoxin M and the significance of non-active site thiol groups. Australian Journal of Plant Physiology 55: 577-584. Maffei M. 2002. vetiveria-s-The Genus vetiveria. London: Taylor & Francis. 22 Essential Oil-Bearing Grasses: The Genus Cymbopogon Maffei M, Codignola A, Fieschi M. 1988. Photosynthetic enzyme activities in lemongrass cultivated in temper­ ate climates. Biochemical Systematics and Ecology 16: 263, 264. Maheshwari SK, Chouhan GS, Trivedi KC, Gangrade SK. 1992. Effect of irrigation and stage of crop harvest on oil yield and quality of palrnarosa (Cymhopogon martiniii oil grass. Indian Journal of Agronomy 37: 514-517. Maheshwari SK, Sharma RK, Gangrade SK. 1995. Effect of spatial arrangement on performance of palmarosa (Cymhopogon martinii var. motia)-pigeonpea (Cajanus cajan) intercropping on a black cotton soil (vertisol), Indian Journal ofAgronomy 40: 181-185. Maheshwari SK, Sharma RK, Gangrade SK. 1998. Response of palmarosa tCymbopogon martinii var. motia) to biofertilizers, nitrogen, and phosphorus in a shallow black soil under rainfed condition. Indian Journal ()f Agronomy 43: 175-178. Mathews S, Spangler RE, Mason-Gamer RJ, Kellogg EA. 2002. Phylogeny of Andropogoneae inferred from phytochrome B, GBSSI, and NDHF. International Journal of Plant Sciences 163: 441-450. Mathur AK, Ahuja PS, Pandey B, Kukreja AK, Mandai S. 1988. Screening and evaluation of somaclonal varia­ tions for quantitative and qualitative traits in aromatic grass, Cymhopogon winterianus Jowitt. Plant Breeding 101: 321-334. Maurino VG, Drincovich MF, Casati P, Andreo CS, Edwards GE, Ku MSB, Gupta SK, Franceschi YR. 1997. NADP-malic enzyme: Immunolocalization in different tissues of the C 4 plant maize and the C) plant wheal. Journal of Experimental Botany 48: 799-81 I. Metcalfe CR. 1960. Anatomy of the Monocotyledons. I. Gramineae. London: Oxford University Press. Misra A, Khan A. 1992. Correction of iron-deficiency chlorosis in lemongrass (Cymhopogon jiexuosus Steud.) Watts. Agrochimica 36: 349-360. Misra A, Srivastava NK. 1991. Effect of the triacontanol formulation miraculan on photosynthesis, growth, nutrient uptake, and essential oil yield of lemongrass (Cymhopogon flexuosusi Steud. Watts. Plant Growth Regulation 10: 57-63. Munthali MT, Newbury HJ, Ford-Loyd BY. 1996. The detection of sornaclonal variants of beet using RAPD. Plant Cell Reports IS: 474-478. Nayak S, Debata BK, Sahoo S. 1996. Rapid propagation of lemongrass (Cymbopogon ftexuosus [Nees] Wats.) through somatic embryogenesis in vitro. Plant Cell Reports IS: 367-370. Nayak S, Debata BK, Srivastava VK, Sangwan NS. 2003. Evaluation of agronomically useful somaclonal vari­ ants in jamrosa (a hybrid Cymhopogon) and detection of genetic changes through RAPD. Plant Science 164: 1029-1035. Omisra A, Srivastava NK. 1994. Influence of iron nutrition on chlorophyll contents, photosynthesis, and essential monoterpene oil(s) in Java citronella (Cymhopogon winterianus Jowitt). Photosynthetica 30: 425-434. Patnaik J, Debata BK. 1997a. In vitro selection of NaCI tolerant callus Iines of Cymhopogon martinii (Roxb.) Wats. Plant Science 124: 203-210. Patnaik J, Debata BK. 1997b. Regeneration of plantlets from NaCI tolerant callus lines of Cymhopogon marti­ nii (Roxb.) Wats. Plant Science 128: 67-74. Patnaik J, Sahoo S, Debata BK. 1996. Cytology of callus, somatic embryos and regenerated plants of palmarosa grass, Cymhopogon martinii (Roxb.) Wats. Cytobios 87: 79-88. Patnaik J, Sahoo S, Debata BK. 1997. Somatic embryogenesis and plantlet regeneration from cell suspension cultures of palmarosa grass (Cymhopogon martiniii. Plant Cell Reports 16: 430-434. Patnaik J, Sahoo S, Debata BK. 1999. Somaclonal variation in cell suspension culture-derived regenerants of Cymhopogon martinii (Roxb.) Wats. var. motia. Plant Breeding U8: 351-354. Pecchioni N, Faccioli P, Monetti A, Stanca AM, Terzi Y. 1996. Molecular markers for genotype identification in small grain cereals. Journal of Genetic Breeding SO: 203-219. Phillips SM, Hua P. 2005. Notes on grasses (Poaceae) for the flora of China, Y. New species in Cymbopogon. Novon IS: 471-473. Prat H. 1937. Caracteres anatomique et histologiques de quelques Andropogonees de I' Afrique occidentale. Ann. Mus. Colon. Marseille 5: 25-28. Purseglove JW. 1975. Tropical Crops: Monocotyledons. London: Longman Group Ltd. Puttanna K, Gowda NMN, Rao EVSP. 200 I. Effects of applications of N fertilizers and nitrification inhibitors on dry matter and essential oil yields of Java citronella (Cymhopogon winterianus Jowitt.). Journal of Agricultural Science 136: 427-431. Quiala E, Barbon R, Jimenez E, De Feria M, Chavez M, Capote A, Perez N. 2006. Biomass production of Cymhopogon citratus (DC) Stapf., a medicinal plant, in temporary immersion systems. In Vitro Cellular and Developmental Biology-Plant 42: 298-300. The Genus Cymbopogon s cultivated in temper­ goneae inferred from : 441~50. 'on of somaclonal varia­ interianus Jowitt. Plant " tion from cell suspension 43~34. ture-derived regenerants of 54. for genotype identi fication pLtd. and nitrification inhibitors . erianus Iowitt.). Journal oj 23 Rajendrudu G. Das VSR. 1981. C4 photosynthetic carbon metabolism in the leaves of aromatic tropical grasses-I. Leaf anatomy. CO, compensation point and CO, assimilation. Photosvnthesis Research 2: 225-233. Rani V, Parida A, Raina SN. 1995. Random amplified polymorphic DNA (RAPD) micropropagated plants of Populus deltoides Marsh. Plant Cell Reports 14: 459-462. Rao BRR. 2000. Biomass yield and essential oil yield variations in Java citronella (Cymbopogon winteri­ anus Jowitt.), intercropped with food legumes and vegetables. Journal ofAgronomv and Crop Science­ Zeitschriftfur Acker und Pfianzenbau 185: 99-103. Rao EVSp, Singh M. Chandrasekhara G. 1988. Effect of nitrogen application on herb yield, nitrogen uptake and nitrogen recovery in Java citronella (Cvmbopogon winterianus Jowitt). Indian Journal ojAgronomy 33: 412-415. Rao EVSP, Singh M, Rao RSG. 1994. Performance of intercropping systems based on palmarosa tCymbopogon martinii var. motia). Indian Journal ofAgricultural Sciences 64: 442-445. Ratti N, Kumar S, Verma HN, Gautam SP. 2001. Improvement in bioavailability of tricalcium phosphate to Cymbopogon martinii var. motia by rhizobacteria, AMF and Azospirillum inoculation. Microbiological Research 156: 145-149. Rothermel BA, Nelson T. 1989. Primary structure of the maize NADP-dependent malic enzyme. Journal (~f Biological Chemistry 264: 19587-19592. Sangwan NS, Yadav U. Sangwan RS. 200 I. Molecular analysis of genetic diversity in elite Indian cultivars of essential oil trade types of aromatic grasses (Cvmbopogon species). Plant Cell Reports 20: 437-444. Sangwan NS, Yadav U, Sangwan RS. 2003. Genetic diversity among elite varieties of the aromatic grasses. Cymbopogon martinii. Euphytica 130: 117-130. Sangwan RS, Farooqi AHA, Bansal RP, Singhsangwan N. 1993. Interspecific variation in physiological and metabolic responses of 5 species of Cymbopogon to water stress. Journal of Plant Physiology 142: 618-622. Sangwan RS, Sangwan NS, Jain DC, Kumar S, Ranade SA. 1999. RAPD profile-based genetic characteriza­ tion of chernotypic variants of Artemisia annua L. Biochemistry and Molecular Biology International 47: 935-944. Senthilkumar K, Udaiyan K, Manian S. 1992. Rate of litter decomposition in a tropical grassland dominated by Cymbopogon caesius in Southern India. Tropical Grasslands 26: 235-242. Shasany AK, Lal RK, Patra NK, Darokar MP, Garg A, Kumar S, Khanuja SPS. 2000. Phenotypic and RAP£} diversity among Cymbopogon winterianus Jowitt accessions in relation to Cvmbopogon nardus Rendle. Genetic Resources and Crop Evolution 47: 553-559. Singh A, Singh M, Singh K. 1998. Productivity and economic viabi lity of a palmarosa-pigeonpea intercropping system in the subtropical climate of North India. Journal ofAgricultural Science 130: 149-154. Singh K, Singh DY. 1992. Effect of rates and sources of nitrogen application on yield and nutrient-uptake of Java citronella tCymbopogon winterianus Jowitt). Fertilizer Research 33: 187-191. Singh M. 2001. Long-term studies on yield, quality, and soil fertility of lemongrass tCvmbopogon jlexuosusv in relation to nitrogen application. Journal of Horticultural Science and Biotechnology 76: 180-182. Singh M, Rao RSG, Ramesh S. 2005. Effects of nitrogen, phosphorus, and potassium on herbage, oil yield, oil quality, and soil fertility status of lernongrass in a semi-arid tropical region of India. Journal (if Horticultural Science and Biotechnology 80: 493-497. Singh M, Rao RSG, Rao EVSP. 1996a. Effect of depth and method of irrigation and nitrogen application on herb and oil yields of Java citronella (Cymbopogon winterianus Jowitt) under semi-arid tropical condi­ tions. Journal ojAg ronomy and Crop Science-i-Zeitschrift fur Acker und Pjlanrenbau 177: 61-64. Singh M, Shivaraj B. 1998. Intercropping studies in lemongrass tCymbopogon flexuosusi (Steud, Wats.). Journal ofAgronomy and Crop Science-i-Zeitschrift fur Acker und Pfianzenbau 180: 23-26. Singh M, Shivaraj B. 1999. Effect of irrigation regimes on growth, herbage and oil yields of lemongrass tCymbopogon fiexuosusi under semi-arid tropical conditions. Indian Journal oj Agricultural Sciences 69: 700-702. Singh M, Shivaraj B, Sridhara S. 1996b. Effect of plant spacing and nitrogen levels on growth, herb and oil yields of lemongrass (Cymbopogon flexuosus (Steud) Wats. var. cauvery). Journal ofAgronomy and Crop Science-s-Zeitschriftfur Acker und Pfianzenbau 177: 10 1-105 . Singh N. Luthra R, Sangwan RS. 1990. Oxidative pathways and essential oil biosynthesis in the developing Cymbopogon flexuosus leaf. Plant Physiology and Biochemistry 28: 703-710. Singh N, Luthra R, Sangwan RS. 1991. Mobilization of starch and essential oil biogenesis during leaf ontogeny of lemongrass iCymbopogon fiexuosus Stapf). Plant and Cell Physiology 32: 803-811. 24 Essential Oil-Bearing Grasses: The Genus Cymbopogo Singh RS, Bhattacharyya TK, Kakti MC, Bordoloi ON. 1992. Effect of nitrogen. phosphorus and potash a essential oil production of palmarosa (Cymhopogon martinii var. motiai under rainfed condition. India Journal ofAgronomy 37: 305-308. Singh S, Ram M, Ram 0, Sharma S. Singh DV. 1997. Water requirement and productivity of palmarosa on sandy loam soil under a sub-tropical climate. Agricultural Water Management 35: 1-10. Singh S, Ram M, Ram D, Singh VP, Sharma S, Tajuddin. 2000. Response of lemongrass t Cvmbopogon jiexuo sus) under different levels of irrigation on deep sandy soils. Irrigation Science 20: 15-21. Singhsangwan N. Farooqi AHA. Sangwan RS. 1994. Effect of drought stress on growth and essential oil metab olism in lemongrasses. New Phytologist 128: 173-179. Smith BN, Brown wv. 1973. The Kranz syndrome in the Gramineae as indicated by carbon isotopic ratios. American Journal atBotany 60: 505-513. Spies 11, Troskie TH, Vandervyver E, Vanwyk SMC. 1994. Chromosome studies on African plants-II. The, tribe Andropogoneae (Poaccae, Panicoideae). Bothalia 24: 241-246. Sprengel CPJ. 1815. Cymbopogon. Plantarum Minus Cognitarum Pugillus 2: 14. Srivastava NK, Misra A. Sharma S. 1998. The substrate utilization and concentration of C-14 photosynthates in citronella under Fe deficiency. Photosynthetica 35: 391-398. Taylor PWJ, Geijskes JR, Ko HL, Fraser TA, Henry RJ, Birch RG. 1995. Sensitivity of random amplified polymorphic DNA analysis to detect genetic change in sugarcane during tissue culture. Theoretical a Applied Genetics 90: \165-1173. Trevanion SJ, Furbank RT, Ashton AR. 1997. NADP-Malate dehydrogenase in the C 4 plant Flaveria bidentis. Plant Phvsiology 113: 1153-1165. Vickery JW. 1935. The leaf anatomy and vegetative characters of the indigenous grasses of N.S. Wales. Proceedings of the Linnean Society, N.S. W 60: 340-373. Wallner E, Weising R, Rompf G, Kahl G. Kopp B. 1996. Oligonucleotide finger printing and RAPD analysis of Achillea species: characterisation and long term monitoring of micro propagated clones. Plant Cell Reports 15: 647-652. Welsh JM. McClelland M. 1990. Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Research 18: 7213-7218. Yang H, Tabei Y, Kamad H, Kayano T, Takaiwa F. 1999. Detection of somaclonal variation in cultured rice cell using digoxigenin-based random amplified polymorphic DNA. Plant Cell Reports 18: 520-526. View publication stats