Why is there so little research into the cell biology of the secondary vascular system of trees?


  • Nigel Chaffey

    Corresponding author
    1. School of Science and the Environment, Bath Spa University College, Newton Park, Newton St Loe, Bath BA2 9BN, UK
      Author for correspondence: Nigel Chaffey Tel: +44 (0)1225 875451 Fax: +44 (0)1225 875776 Email: n.chaffey@bathspa.ac.uk
    Search for more papers by this author

Author for correspondence: Nigel Chaffey Tel: +44 (0)1225 875451 Fax: +44 (0)1225 875776 Email: n.chaffey@bathspa.ac.uk


Despite new techniques for studying the cell biology of plant development in recent years, the secondary vascular system has been neglected. Why is this? Here it is argued that some of the barriers that have prevented more widespread study of the tree secondary vascular system are no longer valid. Some of the more intriguing aspects of the secondary vascular system include the recent discovery of a putative plant muscle and identification of a cytoskeleton-facilitated three-dimensional symplasmic transport pathway that permeates the tree. There are great merits in the recently adopted model tree species, poplar, and a new model system – wood formation in Arabidopsis. The time is now right for much greater exploitation of the possibilities that exist for study of the secondary vascular system of trees.


Tremendous progress has been made in recent years in understanding many aspects of higher plant growth and development, stemming from the adoption of Arabidopsis as the ‘model plant’ (Somerville, 2000). However, the majority of that work has concerned the primary meristems of the shoot and root, and cells and tissues of the primary plant body. Although primary meristems are essential for the establishment and extension growth of all plants, on a global basis, secondary meristems, particularly the vascular cambium (‘cambium’), are probably of greater importance.

The cambium is a cylinder of meristematic tissue that encircles the stem and root of most dicotyledonous angiosperms and gymnosperms. By repeated cell division, the cambium not only maintains itself, but also generates radial files of secondary xylem (‘wood’) cells to its inside, and radial files of secondary phloem cells to its outside during secondary thickening (synonymous with secondary growth). Together, the secondary phloem, cambium and secondary xylem comprise the secondary vascular system and constitute the secondary plant body. Secondary thickening increases the girth of the plant, and is especially important to the development of trees and the production of biomass.

Why is the secondary vascular system important?

The importance of the secondary vascular system is largely due to its wood component – simply, wood continues to be of primary importance to us all. For example, it is widely used in house-building and for paper manufacture. Wood also represents a long-term sink for atmospheric carbon (Ceulemans et al., 1999), and is increasingly exploited as an environmentally cost-effective, renewable source of energy (El Bassam, 1996; Haygreen & Bowyer, 1996; Klass, 1998). The enduring importance of wood is underlined by predictions that our demand for it will increase substantially in the near future (Haygreen & Bowyer, 1996; Whiteman, 1996), not least as an alternative energy source.

With increasing pressure on a diminishing area of land that is suitable for growing trees, increased demand for wood has to be met in part by more efficient use of existing plantations, and in part by producing trees that grow faster (Bues, 1990) and are better suited to current requirements. However, although the required, harvestable volume of wood can be produced more quickly in faster-growing trees, the proportion of juvenile wood (that is made during the first 5–25 yr of tree growth) to mature wood is increased. Juvenile wood differs from adult wood and is generally perceived to be lower in quality (Haygreen & Bowyer, 1996). As the trend to use such wood seems set to continue, there is an urgent need to understand the cell biology of juvenile wood formation (Zobel & Sprague, 1998).

Nearly 50 yr ago, Bailey (1952), one of the pioneers of cell biological study of the cambium, stated that ‘the growth of forests on nonagricultural lands is so significant to the future welfare of man as potential sources of diversified organic substances and of stored solar energy that a sustained and comprehensive effort should now be made towards a better understanding of developmental processes in the formation of wood’. Unfortunately, despite those words, and the detailed ultrastructural work carried out on cambium and the secondary vascular system (Larson, 1994; Chaffey, 2001d), we still know comparatively little about the cell biology of wood formation (xylogenesis) (Box 1).

Table Box 1 .  Cell or molecular biology?
Cell biology embraces examination of the events that occur at the level of the cell and its organelles; molecular biology is concerned with phenomena at the molecular level – generally genes and the proteins they encode. In reality, of course, there is no such thing as ‘cell’ or ‘molecular’ biology, there is just biology, which can be studied at many different levels depending on the research situation.
However, with the relatively recent introduction of ‘molecular’ techniques, there has been a tendency to ignore the cellular level of study, such that there has been a ‘molecularization’ of cell biology (Albrecht-Buehler, 1990). The concomitant reduction in ‘non-molecular’ cell biological studies has resulted in a mismatch between our technical ability to probe biology at ever finer levels of resolution and our intellectual ability to interpret that information in the context of the cell, where the molecular events take place. Although this mismatch occurs in studies on the primary plant body, it is more particularly marked in the context of the secondary meristems of plants, hence the deliberate, but recognizably artificial, emphasis on ‘cell’ rather than ‘molecular’ biology in this article.
Ideally, we need to combine studies at all possible levels of enquiry to arrive at a true view of biological phenomena. Successful projects tend to be those that can co-ordinate the research expertise at all the levels necessary for the complete study of a biological problem. This more holistic approach to study of the secondary vascular system was anticipated many years ago by Bailey (1952) when he stressed that ‘To be successful such an effort [study of the biological processes in wood formation] must involve a much closer and broader integration of research in different scientific disciplines’.

Driving forces for studying the tree secondary vascular system

Encouragingly, several driving forces for an increase in cell biological study of xylogenesis have recently emerged (Chaffey, 2001a). Of those ‘drivers’, concerns of the wood-pulp/paper industry, has probably contributed most to creating the present climate where more widespread work on cambial cell biology can at last be contemplated.

One of the major preoccupations of those in the wood-pulp/paper industry is lignin. Lignin is an integral component of wood cell walls, but its presence compromises the production of high quality, nondiscolouring, white paper. Although lignin can be removed, this is not only expensive, but it generally uses environmentally harmful chemicals (O’Connell et al., 1998). Thus, development of wood with reduced lignin (so there is less to remove), or with altered lignin (so that it is easier to remove), is a highly prized goal, and has proved particularly amenable to a molecular biological approach (Christensen et al., 2000).

As the lignin experience shows, modern industrial processing would be better served by wood that is more technologically and/or environmentally friendly. And, at the other end of the supply chain, trees tailored to specific end uses would make forestry operations more effective (Dinus & Welt, 1997). However, although such considerations readily fall within the province of genetic engineering, just a decade ago the situation was that the ‘widespread application of this technology (genetic engineering of wood) awaits two developments: better methods of gene transfer, and a fundamental understanding of the developmental process of wood formation.’ (Whetten & Sederoff, 1991).

In the decade since those words were written, gene transfer to trees, although still technically challenging, has become almost commonplace (Jouanin et al., 1993; Klopfenstein et al., 1997). The results to date are encouraging, and molecular approaches to tree improvement are currently the focus of much research effort (Jung & Ni, 1998; Sederoff, 1999; Bajaj, 2000; Jain & Minocha, 2000b; Tuominen et al., 2000).

We seem therefore to have the ‘better methods of gene transfer’. But what of the fundamental understanding of wood formation? The answer to this question is not so encouraging; until very recently, the topic had been much neglected. Why is this? What are the barriers to progress?

Barriers to progress

There are three main perceived ‘barriers’ to the study of the tree secondary vascular system:

  • • Studying wood formation is too difficult;
  • • There is no model species;
  • • Tree biology is not sufficiently interesting.

How valid are these arguments against studying the tree secondary vascular system?

Studying wood formation is not too difficult

Study of the cell biology of the secondary vascular system of trees is difficult, problems including the large size of trees, difficulty of sampling the cells, low yields of cell components, and the presence of many interfering and naturally fluorescing compounds (Chaffey, 2002b). However, all of those problems are surmountable and techniques as demanding as cryo-fixation for transmission electron microscopy (Rensing, 2002), immunocytochemistry (Funada, 2002; Šamaj & Boudet, 2002), reporter gene histochemistry (Hawkins et al., 2002), PCR-RAPD (Magel et al., 2002), in situ hybridization (Regan & Sundberg, 2002), and even the preparation of cDNA libraries (Allona et al., 1998; Sterky et al., 1998) are now possible. Unfortunately, it is still the perception that such work is too difficult to attempt. Hopefully, recent books such as Jain & Minocha (2000a), Savidge et al. (2000) and Chaffey (2002b) will help to foster a better appreciation of what is possible.

There is a model species for wood formation

Model species are chosen so that research efforts can be focussed and critical mass achieved ensuring rapid and co-ordinated progress in a given area of biology. Thus, Arabidopsis now bears a heavy burden of responsibility as the model plant. Its value lies in our ability to use it to develop hypotheses, which can then be tested in other plants of interest. Similar arguments apply for model systems. However, it is critical to choose the right model.

Xylogenesis has long been recognized as an ideal model system to study plant cell differentiation (Roberts, 1976). However, wood formation is complex and much of the work carried out in the name of ‘xylogenesis’ has been performed with Zinnia cells in vitro. In that system, individual mesophyll cells, which have been isolated from leaves of the ornamental herbaceous annual Zinnia elegans and maintained in liquid culture, are induced to differentiate as ‘tracheary elements’ by appropriate manipulation of the culture medium (Fukuda, 1996). However, although it is a popular system, it is not necessarily a good model for all of the wood-forming process that takes place in the tree (Chaffey, 1999a).

New models: poplar

The importance of model systems in the study of xylogenesis should diminish as more use is made of trees of the genus Populus, which has been widely promoted as the model angiosperm tree (Stettler et al., 1990, 1996; Klopfenstein et al., 1997; Bradshaw et al., 2001). There are many reasons why poplar is a good choice for the study of the tree secondary vascular system (Chaffey, 1999a, 2001, 2002b), and first among these is the relatively small size of the genome (5 × 108 base pairs –Plomion et al. (2001)), comparable to that of Arabidopsis (2 × 108 base pairs –Bennett & Leitch (1997)), making it amenable to the wide range of molecular genetic approaches. Poplar also has a high growth rate, particularly such interspecific hybrids as the triploid Populus tremula×P. tremuloides (Einspahr, 1984), which means that processes related to operation of the secondary vascular system, such as formation of tension wood, can be studied in time-scales comparable to the generation time of Arabidopsis.

Approximately 10 600 expressed sequence tags (ESTs) have now been produced from poplar spp. (NCBI, 2001), and are publicly accessible. Current large-scale sequencing projects aim to increase that number significantly within the next few years (Mellerowicz et al., 2001; Plomion et al., 2001). Once the genes of interest have been identified, from EST libraries, one then has the possibility of investigating their location and action using, for example, sense (Tsai et al., 1998) and antisense (Zhong et al., 2000) technologies and in situ hybridization (Regan & Sundberg, 2002). Such EST libraries are a good starting point for identifying the genes that are expressed at various stages of cell differentiation, and their creation is a significant advance in our ability to understand the developmental cell biology of the tree secondary vascular system.

Poplar was also the first forest tree to be genetically transformed (Parsons et al., 1986; Filatti et al., 1987), and the technique is now employed almost routinely (Jouanin et al., 1993; Tzfira et al., 1996; Klopfenstein et al., 1997; Lepléet al., 2000). The ready introduction of foreign, additional or altered genes enables study of gene action in situ (Feuillet et al., 1995; Vander Mijnsbrugge et al., 1996; Fladung et al., 1997; Tuominen et al., 1997; Tsai et al., 1998; Tzfira et al., 1999; Grünwald et al., 2000; Shani et al., 2000) aiding interpretation of their role in cell development.

Knowledge of the anatomy of any biological system is fundamental to gain an understanding of its development. For a hardwood tree, the anatomy of the Populus secondary vascular system is relatively straightforward (Schweingruber, 1990; Telewski et al., 1996), and it has been well studied at both the light (Tuominen et al., 1997; Puech et al., 2000; Chaffey et al., 2002a) and electron (Sauter & van Cleve, 1991; Chaffey & Barlow, 2001a) microscope levels.

Immunological approaches are one of the most promising of the ‘new techniques’ of cell biology that await full exploitation in trees (Chaffey, 1999a). Over the last decade, techniques have been developed for the immunological study of a wide variety of cell components in the secondary vascular system of poplar (e.g. reserve proteins –Sauter & van Cleve (1990), enzymes of the biosynthesis of lignin –Šamaj & Boudet (2002), and lignan –Vander Mijnsbrugge et al. (2000a); enzymes of cell wall polysaccharide modification –Micheli et al. (2002); cell wall components –Guglielmino et al. (1997), Micheli et al. (2001); phenylalanine ammonia-lyase –Osakabe et al. (1996); and the cytoskeletal proteins, actin, tubulin and myosin –Chaffey & Barlow (2001a,b), Chaffey et al. (2002a)). Recently, 2-D PAGE has been exploited in identifying proteins preferentially produced in developing xylem of poplar (Vander Mijnsbrugge et al., 2000b), yielding information that is complementary to that derived from EST analysis of this tissue. At an even finer level of resolution, molecular work is already being carried out using poplar (e.g. GUS-reporter gene studies –Hawkins et al. (2002); identification of homeobox genes in wood formation –Hertzberg & Olsson (1998); and in situ hybridization –Regan & Sundberg (2002)).

The use of biomass energy (‘green energy’) as an environmentally friendly, and renewable, alternative to fossil fuels is attractive. Such issues are likely to be of increasing importance both in the light of concerns about global warming and elevated CO2 levels, and as fossil fuel reserves themselves become depleted. Poplars, and the closely related willow species, are of great interest from this point of view (Douglas, 1989; Christersson, 1996). The likely increase in the demand for such fuels increases the need for basic information on the cell biology of wood formation, particularly in view of the possibility of manipulating cell wall formation to increase sites for deposition of lignin (Chaffey, 2000), which has a calorific value almost twice that of cellulose (El Bassam, 1996).

Speeding up the process

With the current emphasis on molecular approaches to the study of plant development, ‘model species’ tend to be those which either have, or can be induced to have, a wide variety of mutants, or which have the ability to be readily genetically transformed to enable manipulation of their genome. Unfortunately, given the long generation times of trees (many years compared to a few weeks for Arabidopsis), the use of mutants is currently not so easy for this plant group. However, the transformation of poplars which exhibit precocious flowering (Weigel & Nilsson, 1995), the development of birch clones which flower within a year of rooting (Lemmetyinen et al., 1997), and of early flowering orange trees (Peña et al., 2001) suggest that even the obstacle of long generation times in trees may only be a temporary one, and identification and use of dominant mutants in trees, which can then be stored and propagated in vitro, without the need to self and raise F2 generations, provides another way around the problems of long-generation times.

Another important consideration related to the naturally long life cycle of trees is that most of the current work is performed on young trees, hence on juvenile wood. This is a particular concern for studies on transgenic trees, where saplings are generally grown for < 5 yr. It is well known that juvenile wood differs significantly from the later-formed, mature wood, and it cannot be assumed that what holds good for juvenile wood will also be the case with mature wood. Thus, there is a pressing need to extend studies from saplings to more mature trees.

Poplar is not the only tree

The many virtues of poplar notwithstanding, Populus is only an example of a northern temperate angiosperm (hardwood) tree, which has diffuse-porous wood and a non-storied cambium. Nevertheless, and whilst it is acknowledged that poplar does not embrace the full range of tree types (such as ring-porous, gymnosperm, tropical, or storied-cambial forms), and there remain very good economic reasons why work is carried out on other trees, such as eucalyptus (Bossinger & Leitch, 2000), walnut (Label et al., 2000), Robinia (Magel, 2002), and pine (Walter & Grace, 2000; Lev-Yadun & Sederoff, 2001), concentrating on Populus should allow tree cell biological work to be more focused. In that way it should also generate the ‘critical mass’ of interest that has made work on Arabidopsis so successful, and ultimately provide Whetten & Sederoff’s ‘fundamental understanding’ of xylogenesis.

Arabidopsis, an alternative tree?

While it has been known for several years that Arabidopsis undergoes some secondary thickening (Dolan et al., 1993; Lev-Yadun, 1997), its true extent has been appreciated only relatively recently. Preliminary work by Chaffey et al. (2002b) has established that Arabidopsis not only develops a vascular cambium, but can also be kept alive for many months, and produces substantial amounts of wood (Fig. 1a). Furthermore, both the cambial and wood cells – principally, fibres and vessel elements – of Arabidopsis are morphologically and ultrastructurally similar to those of poplar (Chaffey et al., 2002b).

Figure 1.

Aspects of the cell biology of the secondary vascular system. (a) Transverse section of 4-month-old hypocotyl of Arabidopsis showing substantial development of secondary xylem (SX) viewed with autofluorescence. Note the nonfluorescent cambial zone (CZ). SP, secondary phloem region. Bar, 100 µm. (b) Indirect immunofluorescent localizations of cytoskeletal proteins during development of bordered pits in tracheids of pine, seen in radial longitudinal sections. (i–ii) Differential interference contrast (‘Nomarski’) images of macerated tracheids showing early (i) and late (ii) stages of bordered pit development. Note the wide diameter of the pit (in i) which becomes much reduced as the pit border develops, finally leaving a small aperture in the mature tracheid (in ii). (iii–viii) Immunolocalization of myosin (iii,iv), tubulin (v,vi) and actin (vii,viii) in early (iii,v,vii) and late (iv,vi,viii) stages of bordered pit development in pine. Note the helically oriented microfilament bundles in viii. Bars, 20 µm. (c) Four panels showing indirect immunofluorescence localizations of three cytoskeletal proteins (myosin (i), actin (ii, for microfilaments) and tubulin (iii, for microtubules)), and the polysaccharide callose (iv, a marker for sites of plasmodesmata) in mature xylem ray cells, seen in radial longitudinal section of hybrid aspen wood. Antibody localizations shown in yellow-green; autofluorescence of cell walls shown in red-orange; nuclei stained red with propidium iodide. Note the longitudinal orientation of microfilaments and the similar orientation of the microtubules, which is approximately parallel to the long axis of the ray cells, and the similarity of myosin and callose localizations at pit-fields in the cell walls. T, tangential cell wall; Tr, transverse cell wall; side view. Bars, 20 µm (d) Diagrammatic representation of the putative cytoskeleton-facilitated symplasmic pathway within the long-lived ray and axial parenchyma cells of the secondary vascular system of angiosperm trees. In this instance, the pathway illustrated is for delivery of photosynthate to the cambial sink, during the active phase of the cambial seasonal cycle, and for the mobilization of reserve products from the wood. Cells are illustrated as they appear in a transverse section of tree stem or root, but are not drawn to scale. Leaf-derived photosynthate from the sieve elements or that made in situ from photosynthetic phloem cells can enter dilatation (DR) or ordinary uniseriate ray cells either directly (1) or via the long-lived axial cells of the phloem, with or without the intermediation of companion cells (2, then 3). Transport between cells within the phloem ray is facilitated by the abundance of plasmodesmata within the tangential walls, possibly with acto-myosin-assisted plasmodesmatal widening, and within cells by the net-axially oriented microtubule and microfilament bundles (4). The intercellular cytoskeletal-assisted transport of photosynthate breaks down in the cambial zone where both microtubules and microfilaments are less organized, making it available for use in cell division and growth (5). Breakdown products of storage materials held in the axial parenchyma cells of the wood are transported, via cytoskeletal-facilitated mechanisms similar to those postulated in the phloem rays, between radially adjacent axial parenchyma cells (6), thence to radially oriented xylem ray cells, where it joins material derived from mobilization of reserves held in the ray cells. Transport of materials within the xylem rays is postulated to take place as for phloem ray cells (7) until the materials are delivered to the cambial zone (8). Green, walls of living cells; brown, walls of dead cells; red, microtubules; blue, microfilaments; orange, myosin + callose + actin at pit fields; black, plasmodesmata; yellow, middle lamella; dashed line, annual ring boundary; A, axial (parenchyma) cell; BP, bordered pit; CC, companion cell; DR, cell of dilatation region of ray; F, fibre; f, fusiform cambial cell; r, ray cambial cell; SE, sieve element; VE, vessel element.

Therefore, in these respects, Arabidopsis can be viewed as a ‘miniature tree’ and used to address basic questions about the cell biology of the secondary vascular system in general, and wood formation in particular. These observations, coupled with the other well documented attractions of Arabidopsis as a model system should permit its use for hypothesis development, while hypothesis testing can then be performed in the tree. Thus use of Arabidopsis (in tandem with poplar) probably represents the way forward for the immediate future of wood research (Chaffey, 1999b, 2001).

It is important to note in this context that use of Arabidopsis as a model xylogenetic system does have its limitations. For example, important characteristics of the tree secondary vascular system, such as the seasonal variation of earlywood and latewood production, the production of heartwood, and the seasonal cycle of cambial dormancy-activity, are unlikely to be features of the Arabidopsis system. In other words, Arabidopsis is unlikely to make work with trees redundant.

Tree biology is interesting

The biology of the secondary vascular system is not taught to the same extent as that of the primary meristems and the primary plant body, translating into less research and a general view that the secondary vascular system is intrinsically uninteresting. This view is misguided.

Why is the secondary vascular system so interesting?

The main reason that the secondary vascular system is so interesting is because it is so complicated. This complexity arises from the combination of the following:

  • • Wood formation takes place over six distinct zones (Bailey, 1952), from cell division in the cambium to heartwood formation;
  • • Wood is an intriguing, intimate mixture of dead and long-lived cells;
  • • There are many different wood and cambial cell types, some of which can interconvert, and with many subtle variations within different species, and embracing such characteristics as storied/non-storied cambium; softwoods/hardwoods; tropical/temperate trees; and, diffuse-porous/ring-porous wood;
  • • There is a seasonal component to cambial activity and wood-cell production, at least in temperate species;
  • • The secondary vascular system comprises two differently oriented subsystems – axial and radial;
  • • All these factors also apply, to a greater or lesser extent, to the secondary phloem.

One of the characteristic features of the secondary vascular system, which makes it ideal for studying questions concerning cell differentiation and determination, is the ordered cell files that are produced by the cambium. As an individual cambial derivative becomes displaced from the cambium, by intervening cell division, it differentiates until it becomes a mature cell within a radial file of similar cells. In principle therefore and in any given section, one can trace the stages of differentiation of an individual cell from the end-point back to the cambial starting point, or vice versa, by examining the relevant radial cell file.

Aspects of cambial cell division challenge some of our established ideas of this important process, derived from study of primary meristems. For example, periclinal (longitudinal) division of fusiform cells, which can be up to 3 mm in conifers, defies Errera’s Law (Cutter, 1978), which states that cells divide by a wall of minimal surface area. Additionally, fusiform cell division appears to take place in the absence of a preprophase band (ppb) (Farrar & Evert, 1997; Oribe et al., 2001). The ppb, a band of microtubules which encircles the cell and appears to define both the plane of cell division and the site where the division wall joins up with the parental cell wall (Wick, 1991), as identified in cells of the primary plant body, is generally considered to be essential to ordered cell division. The fact that fusiform cambial cells divide in an ordered manner in the absence of a ppb not only demands further study, but is likely to cause us to re-evaluate our views of the role of the ppb (Chaffey, 2002d). Finally, it is not too farfetched to suggest that study of the remarkable longevity of the cambium, which can remain alive for over 5000 yr in the case of bristlecone pine (Thomas, 2000), may provide some insight into human lifespan.

As intriguing as these observations are, recent work on the tree secondary vascular system has provided some unexpected insights into the cell biology of this tissue system.

Identification of a putative plant muscle in bordered pit development

My particular interest in the tree secondary vascular system is in the role of the cytoskeleton during secondary vascular differentiation (Chaffey, 2000). The cytoskeleton is a diverse assemblage of proteins found in the cytoplasm of all eukaryotic cells, but mainly represented by two components in plant cells, microtubules (MTs) and microfilaments (MFs) (Lloyd, 1991; Staiger et al., 2000). The MTs, and MFs to a lesser extent, are intimately involved in cell division, for example as the ppb, in the mitotic and meiotic spindles, which helps to separate the daughter genomes, and in the phragmoplast, which participates in assembly of the cell plate (Otegui & Staehelin, 2000). In nondividing, growing cells, the MTs are generally acknowledged to have a role in orienting the cellulose microfibrils of the developing cell walls (Giddings & Staehelin, 1991; Baskin, 2001). Since the orientation of the relatively inflexible microfibrils is in large part responsible for anisotropic cell expansion (Brett, 2000), MTs appear to play an important part in cellular morphogenesis. In view of the necessity of highly ordered cell division in the cambium, and the role that both microfibril orientation (Butterfield, 1998) and wood-cell morphology have in determining the physical characteristics of wood, the case for an interest in the cytoskeleton of the tree secondary vascular system is clear. Furthermore, the interest in cell walls as ‘biotechnological targets’ (Chapple & Carpita, 1998) and of manipulation of MTs to achieve modifications of plant biology (Nick, 2000) serve to underline the importance of cytoskeletal study of xylogenesis.

Using immunofluorescent localization of α-tubulin (a component of MTs), a ring of MTs has been identified at the periphery of bordered pits during early stages of their development, and at their aperture at late stages of development (Chaffey et al., 1997). A similar immunolocalization of actin (the major component of MFs) (Chaffey et al., 2000) prompted the suggestion that the MT-MF ring at the periphery of the bordered pit contracted during development of the border to become the aperture-associated MT-MF ring seen at late stages of pit development. Recently, myosin has been immunolocalized at both the periphery and aperture of developing bordered pits (Chaffey & Barlow, 2001b) in vessel elements of the hardwood trees, Populus tremula×P. tremuloides (hybrid aspen) and Aesculus hippocastanum (horse-chestnut), and in the tracheids of the gymnosperm, Pinus pinea (stone pine) (Fig. 1b).

These immunolocalizations provide strong circumstantial evidence that an acto-myosin contractile system may operate during bordered pit formation, in a manner akin to the contraction of mammalian muscle. Although the exact nature of this putative ‘plant muscle’ has yet to be confirmed, it offers the exciting possibility that we may be able to manipulate its degree of contraction to give bordered pits of a range of diameters. In that way, but subject to causing no adverse effects in the tree, it might be possible to increase pit diameter, facilitating better infiltration of preservatives into the timber, or to reduce pit diameter, thereby hindering the cell–cell spread of pathogenic microorganisms throughout the living tree.

A cytoskeleton-facilitated three-dimensional transport pathway permeates the tree

Bordered pit formation is a good example of cytoskeletal interactions over very short distances. Work on the long-lived parenchyma cells of the secondary xylem and phloem suggests cytoskeletal interactions over a much greater range (Chaffey & Barlow, 2001a,b), and is summarized here.

On the grounds of their longevity and the abundance of plasmodesmata within their tangential walls, cells of the rays, which extend from the phloem through the cambial zone and into the xylem, are considered to be the major radial symplasmic transport pathway within the tree (Sauter, 2000). These cells, along with the long-lived axial parenchyma cells of the secondary xylem, are also important sites of deposition of reserve materials, such as proteins, starch and lipids (Höll, 2000), which are generally laid down as winter-dormancy approaches.

Callose localization has often been used as a marker for sites of plasmodesmata in plant cell walls, and in the tree secondary vascular system its pattern agrees closely with the known distribution of plasmodesmata in the pit-fields of ray cell walls. Thus, the colocalization of callose (Fig. 1c-iv) and myosin (Fig. 1c-i) at the pit-fields constitutes strong evidence that myosin itself is a component of the plasmodesmata. Together with actin localization at the pit fields (Chaffey & Barlow, 2001a) and the bundles of both MFs (Fig. 1c-ii) and MTs (Fig. 1c-iii), which often appear to be similarly oriented, we have proposed the existence of a cytoskeleton-facilitated symplasmic pathway of solute transport throughout the tree (Fig. 1d) (Chaffey & Barlow, 2001a). We hypothesise that MTs and MFs are involved in transport within the ray and axial parenchyma cells, and, in concert with the ‘gate-keeping’ role of a contractile acto-myosin system which could reversibly close or open the plasmodesmata at pit fields, are also important in intercellular transport. Thus, the proposed symplasmic coupling between ray cells, between axial parenchyma cells, and between axial parenchyma and ray cells represents an extensive 3-dimensional communication pathway which permeates the tree both radially, from the phloem through the cambium into the wood, and circumferentially, via the axial xylem parenchyma, particularly at the growth ring boundary where such cells are a prominent feature. Furthermore, coupling this radial-circumferential pathway with the long-distance symplasmic transport within the axially oriented sieve tubes/cells of the phloem (which also appears to have a cytoskeletal dimension –Chaffey & Barlow (2001b)) creates a ‘super-symplasmic continuum’ that permeates the whole tree.

We suggest that this cytoskeletal pathway has an important role throughout the seasonal cycle of the tree: in delivery of photosynthate, and mobilized reserves, to the actively dividing cambium, and in the movement of materials to sites of reserve deposition, principally within the wood, prior to dormancy. The possibility of acto-myosin ‘gating’ of plasmodesmata at pit-fields allows the suggestion that the cell–cell transport of solutes could be controlled, thereby creating symplasmic domains, which could lead to the co-ordination of developmental processes throughout the tree.

The role(s) that this three-dimensional network might play remain to be explored, but the possibilities that it represents for manipulation of tree physiology, particularly biomass production, are sufficiently ‘interesting’ to warrant more research in this area of whole-tree physiology.


Research into the tree secondary vascular system is generally very poorly funded, and this lack of support is in part due to the disinclination of would-be researchers and those who allocate the funds because of a belief in the barriers discussed here. Thus, there is a need to ‘educate’ both these groups so that the biology of secondary meristems is given the importance it deserves.

Notwithstanding the availability of techniques for study of the tree secondary vascular system, because it has been the poor relation to the primary plant body for such a long time, there is generally a lack of basic understanding of its biology. Thus, a great deal of fundamental work is required to describe the system at the level of the cell before it is possible to fully exploit the possibilities that might exist for its manipulation at the molecular genetic level. In the light of such global issues as concerns over rising CO2 levels and dwindling fossil fuel reserves, there has never been a better time to promote the relevance and importance of work on the tree secondary vascular system.


I am grateful to the Biotechnology and Biological Sciences Research Council of the UK, the Kempe Foundation (Sweden), the Swedish University of Agricultural Sciences (Umeå, Sweden), and the European Union for past funding of my work on the cell biology of the tree secondary vascular system. I also thank the anonymous referees for their kind comments and suggestions.