|Home | About | Journals | Submit | Contact Us | Français|
The timing of cambial reactivation plays an important role in the control of both the quantity and the quality of wood. The effect of localized heating on cambial reactivation in the main stem of a deciduous hardwood hybrid poplar (Populus sieboldii × P. grandidentata) was investigated.
Electric heating tape (20–22 °C) was wrapped at one side of the main stem of cloned hybrid poplar trees at breast height in winter. Small blocks were collected from both heated and non-heated control portions of the stem for sequential observations of cambial activity and for studies of the localization of storage starch around the cambium from dormancy to reactivation by light microscopy.
Cell division in phloem began earlier than cambial reactivation in locally heated portions of stems. Moreover, the cambial reactivation induced by localized heating occurred earlier than natural cambial reactivation. In heated stems, well-developed secondary xylem was produced that had almost the same structure as the natural xylem. When cambial reactivation was induced by heating, the buds of trees had not yet burst, indicating that there was no close temporal relationship between bud burst and cambial reactivation. In heated stems, the amount of storage starch decreased near the cambium upon reactivation of the cambium. After cambial reactivation, storage starch disappeared completely. Storage starch appeared again, near the cambium, during xylem differentiation in heated stems.
The results suggest that, in deciduous diffuse-porous hardwood poplar growing in a temperate zone, the temperature in the stem is a limiting factor for reactivation of phloem and cambium. An increase in temperature might induce the conversion of storage starch to sucrose for the activation of cambial cell division and secondary xylem. Localized heating in poplar stems provides a useful experimental system for studies of cambial biology.
Wood is formed by the cambial activity of trees and this activity exhibits annual periodicity in temperate zones. Cambial activity ceases in the autumn and winter seasons, and dormancy is characterized by the absence of cell division. Cambial dormancy in winter can be considered to consist of two stages: rest and quiescence (Catesson, 1994; Larson, 1994). The resting stage is maintained by conditions within the tree and it is followed by the quiescent stage, which is controlled by environmental conditions (Little and Bonga, 1974). The resting stage of dormancy is a physiological state wherein the cambium cannot divide, even under favourable conditions. By contrast, during the quiescent stage of cambial dormancy, the cambium is able to divide when exposed to appropriate environmental conditions (Savidge and Wareing, 1981; Riding and Little, 1984, 1986; Sundberg et al., 1987). Cambial activity generally resumes in the spring, with a change from the quiescent dormant state to the active state. This springtime phenomenon is known as cambial reactivation.
The timing of cambial reactivation plays an important role in the control of both the quantity and the quality of wood. Therefore, many researchers have studied the anatomical, cytological, biochemical and histochemical changes that occur during cambial reactivation (for reviews, see Catesson, 1994; Larson, 1994). For example, studies of Aesculus hippocastanum (Barnett, 1992), Robinia pseudoacacia (Farrar and Evert, 1997a,b), Populus trichocarpa (Arend and Fromm, 2003) and Pinus contorta (Rensing and Samuels, 2004) have revealed the general outline of the events that occur in cambial cells during natural reactivation. However, the physiological regulation of cambial reactivation in the spring is still not fully understood (Funada et al., 2002).
The localized heating of stems induces localized cambial reactivation in evergreen conifers, such as Pinus contorta (Savidge and Wareing, 1981), Picea sitchensis (Barnett and Miller, 1994), Cryptomeria japonica (Oribe and Kubo, 1997), Abies sachalinensis (Oribe et al., 2001, 2003) and Picea abies (Gričar et al., 2006). Such observations suggest that temperature might be a limiting factor in the onset of cambial reactivation during the quiescent dormant state. However, in the deciduous conifer Larix leptolepis, localized heating for 2 weeks was insufficient to induce the localized reactivation of the cambium (Oribe and Kubo, 1997). Thus, the cambial response to heating appears to differ between evergreen and deciduous trees and it is unclear whether localized heating can induce cambial reactivation in all species of tree. Moreover, there have been no studies, to our knowledge, of the effects of localized heating in the induction of cambial reactivation in deciduous hardwood species.
The main purpose of the present study was to investigate whether localized heating of stems could induce localized cambial reactivation in a deciduous diffuse-porous hardwood. The study examined the hybrid poplar, Populus sieboldii × P. grandidentata because of its economical importance as a fast-growing tree. In addition, hybrid poplar is considered to be a ‘model’ tree as considerable molecular and biological information about the growth and development of trees has been obtained from it (Chaffey, 1999, 2002; Bradshaw et al., 2001; Mellerowicz et al., 2001). Thus, cambial responses to localized heating in poplar stems should provide a useful model system for studies of cambial biology. The type and position of the first-dividing cambial cells, the length of time needed for cambial reactivation and the further progression of cambial derivatives in both heated portions of stems and non-heated control portions were monitored. The amounts and localization of storage starch, an important source of sucrose in trees, near the cambium in heated stems were also examined in an attempt to identify the source of energy for cambial reactivation.
Experiments were carried out on four 12-year-old specimens of a cloned hybrid poplar, Populus kitakamiensis (P. sieboldii × P. grandidentata; average height, 10 m; average diameter of stems at breast height, 13 cm) that had been developed by Nippon Paper Co. The trees were growing in the field nursery of the Tokyo University of Agriculture and Technology in Fuchu, Tokyo (35°40′N, 139°29′E). Two trees were used for the first heating, from 18 February, 2005 to 14 May, 2005, and two trees were used for the second heating, from 16 December, 2005 to 30 January, 2006. During the first heating, both of the two trees were used for sequential observations of cambial activity and for studies of the localization of storage starch around the cambium from dormancy to reactivation. During the second heating, two trees were used for sequential observations of cambial activity from dormancy to reactivation. No abnormal structure was found in the stem by artificial heating.
Electric heating tape (Silicone-Rubber Heater; O & M Heater, Nagoya, Japan), 50 cm in length and 30 cm in width, was wrapped at one side of the main stem of each tree at breast height (Oribe and Kubo, 1997; Oribe et al., 2001, 2003). An alternating current was passed through the tape at a potential of 100 V to warm the surface of the stem. The temperature between the outer bark and the heating tape was recorded with a thermometer and, at the site at which the stem was heated, the temperature was adjusted to 20–22 °C with a thermostat. Cambial temperature reached a stationary value within the selected temperature range in 4–6 h in thick-barked conifers (Oribe and Kubo, 1997).
During the first heating, samples were taken daily for the first 15 d and then at 2- to 3-d intervals from heated and non-heated control portions of stems. The opposite side of each heated area was used as the control (Fig. 1). During the second heating, samples were collected at 1-week intervals from both heated and non-heated control portions of the stem. A series of small blocks (2 × 2 × 1 cm3), containing phloem, cambium and some xylem, were removed with a disposable scalpel and chisel. Samples were removed in a zigzag fashion to eliminate effects of wounding. Each block was cut into 2-mm-thick samples immediately after removal from the tree.
The 2-mm-thick samples were fixed in 4 % glutaraldehyde in 0·1 m phosphate buffer (pH 7·3) under a vacuum for 1 h at room temperature. Fixed samples were washed in 0·1 m phosphate buffer and trimmed to 3 mm in length for subsequent fixation in 1 % osmium tetroxide in 0·1 m phosphate buffer (pH 7·3) for 2 h at room temperature. After washing in phosphate buffer, the specimens were dehydrated in a graded ethanol series and embedded in epoxy resin. Transverse and radial sections were cut at a thickness of approximately 1 µm with glass knives on an ultramicrotome (Ultracut N; Reichert, Vienna, Austria). Sections for observations of storage starch were stained with a 1 % solution of iodine-potassium iodide (IKI) in water for visualization of starch and other sections were stained with a solution of 1 % safranin in water for visualization of cambial cell division (Murakami et al., 1999; Nakaba et al., 2006). Both sets of sections were examined under a light microscope (Axioscop; Carl Zeiss, Oberkochen, Germany).
Daily average, maximum and minimum air temperatures at the experimental site in Fuchu, Tokyo, from 1 February to 30 May, 2005 (during the first heating experiment) and from 1 December, 2005 to 31 January, 2006 (during the second heating experiment) are shown in Fig. 2. From February to March and December to January, the maximum temperature was 10–15 °C and the minimum temperature was sometimes below 0 °C, indicating that our experimental site could be classified as being in a temperate zone (Oribe and Kubo, 1997).
No division of fusiform cambial cells and ray cambial cells was detected in samples that had been collected on 18 February, 2005 (Fig. 3A, B). The cambium was located between the secondary phloem cells and the thick-walled secondary xylem cells that had formed during the previous growing season (Fig. 3A, B). During dormancy, the cambium consisted of two or three radial layers of radially narrow and compactly arranged cells, each with a thick wall and dense cytoplasm (Fig. 3B). The radial walls of cambial cells were thicker than the tangential ones.
After 2 weeks of heating, on 4 March, 2005, new cell plates had formed in phloem cells that had been produced during the previous growing season (Fig. 4A). The first cell division occurred in phloem cells that were radially wider than cambial cells, and these phloem cells were located in the sixth to seventh layer from the previous year's xylem side. The observations indicated that cell division in phloem had started earlier than cambial cell division. In the non-heated control portion of the stem, new cell plates were evident in phloem cells on 24 March, 2005 (Fig. 4B). In the case of both heated and non-heated control portions of stems, the type and location of phloem cells that had divided prior to cambial reactivation were the same (Fig. 4A, B). The difference in terms of timing between cell division in phloem in heated and non-heated portions of stems was approximately 20 d (Table 1).
In the cambial cells of samples that had been heated for 25 d, there were no new cell plates, indicating that the cambium had remained dormant. After 28 d of heating, on 18 March, 2005, there was evidence of cambial reactivation in the heated stem (Fig. 5A). By contrast, new cell plates in the cambium of the non-heated control portion of the stem were first found on 15 April, 2005 (Fig. 5B). The difference in terms of timing between cambial reactivation in the heated and non-heated control portions of the stems was approximately 4 weeks (Table 1). The first division of cambial cells occurred in the second layer of fusiform cambial cells from the previous year's xylem side in both the heated and the non-heated control portions of stems (Fig. 5E, F). The division of fusiform cambial cells occurred 3–4 d earlier than that of ray cambial cells. As a consequence of the formation of new cell plates, the number of cell layers increased in the cambial zone.
There were no obvious differences in terms of the timing of the division of phloem cells and of cambial reactivation between the two cloned trees examined in the first heating experiment.
On 18 April, 2005, after 2 months of heating, production of secondary xylem started in the heated portion of the stem (Fig. 6A). A large volume of well-developed secondary xylem was evident in the heated portion of the stem on 25 April, 2005 (Fig. 6C). By contrast, xylem differentiation started in the non-heated control portion of the stem only on 13 May, 2005 (Fig. 6B) and active differentiation of xylem was evident on 19 May, 2005 (Fig. 6D). The difference in terms of timing between the initiation of differentiation of xylem in the heated stem and non-heated control stem was approximately 3 weeks (Table 1). In addition, the time between cambial reactivation and xylem differentiation was approximately 4 weeks and was almost the same in both the heated and the control portions of the stem (Table 1). Cambial cells differentiated into new wood fibres or vessel elements in both the heated and the control portions of the stem (Fig. 6C, D). The secondary xylem that was formed as a result of localized heating was of very similar dimensions and structure to the naturally formed secondary xylem in the non-heated control portion of stems at the light-microscope level (Fig. 6C, D).
When cambial reactivation occurred in the heated stem on 18 March, 2005, the buds were still in an unburst condition (Fig. 7A). Bud burst occurred on 15 April, 2005, when cambial reactivation was initiated in the non-heated control portion of the stem. Obvious bud flushing was visible on 25 April, 2005 (Fig. 7B). The differentiation of xylem in the heated stem started on 18 April, 2005, after bud flushing had begun.
During dormancy, starch grains accumulated mainly in the ray parenchyma cells nearest the cambium and no storage starch was evident in fusiform cambial cells (Fig. 8A). When reactivation of the phloem occurred in the heated stem, considerable storage starch was found in phloem parenchyma cells and ray parenchyma cells (Fig. 8B). In the heated stem, the amount of storage starch decreased in the phloem parenchyma cells and ray parenchyma cells nearest the cambium during cambial reactivation (Fig. 8C). After cambial reactivation, storage starch disappeared completely from phloem cells near the cambium in the heated stem (Fig. 8D).
During xylem differentiation, storage starch appeared again in phloem parenchyma cells and ray parenchyma cells in the heated stem (Fig. 8E). Throughout the sampling period, no storage starch was detected in fusiform cambial cells in the heated portion of the stem.
No cell division was observed in fusiform cambial cells and ray cambial cells in samples that had been collected on 16 December, 2005. After 3 weeks of heating, no new cell plates were found in phloem cells. After 4 weeks of heating, cell division started in the phloem (Fig. 4C). The positions and nature of the phloem cells that divided earlier than the cambial cells were the same as observed during the first heating (Fig. 4A, C). During the second heating, approximately 4 weeks were required to produce new cell plates in phloem cells, which was twice as long as during the first heating. At the same time, the non-heated control portion of the stem had remained dormant (Fig. 4D).
New cell plates were evident in the fusiform cambial cells after 6 weeks of heating, which was again longer than in the first heating experiment (Fig. 5C). The locations and type of first-dividing cambial cells were the same as during the first heating (Fig. 5A, C). At the same time, cambium in the non-heated control portion of the stem remained dormant (Fig. 5D).
As observed during the first heating, there were no obvious differences in terms of the timing of the division of phloem cells and of cambial reactivation between the two cloned trees.
Localized heating of the stem during the dormant season induced the division of phloem and cambial cells in a deciduous diffuse-porous hardwood poplar. The results reveal that an increase in temperature of the stem is one of the most important factors in cambial reactivation in deciduous trees. Division of phloem cells started prior to cambial reactivation in both heated and non-heated control portions of stems. Heat treatment for at least 2 weeks was necessary to produce new cell plates in the phloem. New cell plates were formed in phloem cells that had been produced during the previous growing season both in heated and in non-heated control portions of stems. Therefore, the type and locations of first-dividing cells were the same in heated and control portions of stems. This observation suggests that phloem cells might respond to heating more rapidly than cambial cells in hybrid poplar.
Localized heating has been reported to induce cambial reactivation earlier than natural cambial reactivation in several evergreen conifers, such as Pinus contorta (Savidge and Wareing, 1981), Picea sitchensis (Barnett and Miller, 1994), Cryptomeria japonica (Oribe and Kubo, 1997), Abies sachalinensis (Oribe et al., 2001, 2003) and Picea abies (Gričar et al., 2006). Oribe et al. (2001, 2003) observed, in the evergreen conifer Abies sachalinensis, that heating for 5–6 d was necessary for cambial reactivation. By contrast, in the diffuse-porous hardwood poplar studied here, approximately 4 weeks was required for cambial reactivation. Therefore, we suggest that, in poplar, the state of dormancy is deep and that lengthy heating is needed to break cambial dormancy. In the deciduous conifer Larix leptolepis, 2 weeks of heating failed to reactivate the division of cambial cells (Oribe and Kubo, 1997). Longer heat treatment of deciduous trees, as compared with evergreen conifers, might be needed for cambial reactivation from the quiescent dormant state.
The present results also showed that, during the second heating treatment in December and January, more time was required for cambial reactivation than during the first heating treatment in February and March. Previous studies have shown that resting cells in the cambium are unable to divide, even when exposed to favourable conditions, while cambium in the quiescent state retains the potential to generate new cells (Savidge and Wareing, 1981; Barnett and Miller, 1994; Oribe and Kubo, 1997; Oribe et al., 2001, 2003; Rensing and Samuels, 2004). Therefore, it is likely that the cambium of poplar in December and January might be in a transitional state, between resting and quiescence, at our experimental site and, therefore, no rapid response to heating was possible.
Barnett (1992) found, in the deciduous hardwood Aesculus hippocastanum, that the first cambial cell division occurred on the phloem side of cambial cells. Similarly, Oribe et al. (2003) and Gričar et al. (2006) reported that, in the evergreen conifers Abies sachalinensis and Picea abies, the first cambial cell division started on the phloem side of cambial cells. By contrast, here the first cell division in the cambial zone occurred in the second layer of fusiform cambial cells from the previous year's xylem side. This observation is in agreement with the site of first cambial cell division in the deciduous hardwoods Fraxinus excelsior (Funada and Catesson, 1991) and Robinia pseudoacacia (Farrar and Evert, 1997a). Therefore, the pattern of cambial reactivation appears to depend on species.
In heated poplar stems, xylem differentiation started after 2 months of heating and well-developed xylem cells were produced within 1 week. The structure of the newly formed secondary xylem cells was more or less similar to that of the normally formed xylem cells in hybrid poplar stems. Oribe and Kubo (1997) and Gričar et al. (2006) observed, in the evergreen conifers Cryptomeria japonica and Picea abies, that localized heating induced the formation of earlywood tracheids with a large radial diameter and secondary walls during cambial dormancy in a temperate zone. By contrast, Oribe et al. (2003) observed, in heated Abies sachalinensis, that only minimal differentiation of xylem occurred after cambial reactivation in a cool temperate zone. Their results are different from those of the present investigation of heated poplar stems. The results here show clearly that heating can induce the formation of large volumes of normal secondary xylem according to the normal pattern found in the temperate zone. The timing and extent of induction of xylem differentiation by heating might depend on climatic conditions.
It has been reported that bud burst and the development of new leaves are related to cambial reactivation and xylem differentiation (Aloni, 1991; Suzuki et al., 1996). Barnett (1992) showed, in Aesculus hippocastanum, that the first division of cells in the cambium started when buds started to swell. Here bud burst and cambial reactivation in non-heated portions of stems started simultaneously during our first heating experiment. However, cambial reactivation in heated portions of stems started without bud burst, indicating that no close relationship exists between the timing of bud burst and cambial reactivation. Therefore, we suggest that, in deciduous hardwood hybrid poplar, bud burst is not a prerequisite for cambial reactivation from the quiescent state. Xylem differentiation in heated portions of stems started after bud flushing, suggesting that some factor(s) from expanding new leaves might be required for xylem differentiation.
Heat treatment changed the amount and localization of storage starch in the phloem and cambium. The amount of storage starch increased near the cambium, as compared with that during the dormant state, when phloem cell division occurred in heated portions of stems. By contrast, during cambial reactivation, a small amount of storage starch was found mainly in phloem parenchyma cells and ray parenchyma cells. After cambial reactivation, storage starch disappeared completely and continuous cell division occurred in the cambium in heated portions of stems. It seems that storage starch, which is a source of sugars, is needed for the division of phloem cells and cambial reactivation. Cambial reactivation occurred in heated portions of the stem of a deciduous hardwood before the development of new leaves, indicating that there was no need for a supply of photosynthate from leaves. Therefore, the utilization of storage starch might be needed for new cell division in the phloem and cambium.
Oribe et al. (2003) demonstrated, in Abies sachalinensis, that cells of heat-reactivated cambium stopped dividing with the disappearance of starch from the storage tissues around the cambium and little or no new xylem was formed. Therefore, they concluded that continued division of cells in the cambium, after cambial reactivation, requires a continuous supply of sucrose. In the current study, starch grains reappeared in the phloem parenchyma cells and ray parenchyma cells near the cambium after cambial reactivation, although the source of this additional starch was not identified. Consequently, a large volume of secondary xylem cells was produced. We can postulate that the differentiation of secondary xylem cells needs considerable energy, which can be obtained from stored carbohydrate. Increases in temperature might induce the enzyme-mediated conversion of storage starch to sucrose for the continuous division of cells in the cambium and the production of new xylem cells in deciduous hardwood poplar.
It is clear that an increase in the temperature of the stem can directly induce the breaking of cambial dormancy in deciduous hardwood hybrid poplar, a model tree. The pattern of cambial reactivation and secondary xylem differentiation induced by heating was the same as that in non-heated control stems. Therefore, cambial reactivation by heat treatment in winter might be a good model system with which to investigate the dynamics of cambial biology because it is easy to follow the process from the division of phloem and cambial cells to secondary xylem development. Using this system, it should be possible to identify the endogenous factors, such as specific genes and proteins and plant growth regulators, that are involved in reactivation of the cambium.
This work was supported, in part, by Grants-in-Aid for Scientific Research (nos. 17580137 and 19580183) and for the ‘Research Revolution 2002 (RR2002) Project for Sustainable Coexistence of Humans, Nature and the Earth; Parameterization of the Terrestrial Ecosystem for Integrated Global Modeling’ from the Ministry of Education, Science and Culture, Japan.