NC State
BioResources
Azmul Huda, A. S. M., Koubaa, A., Cloutier, A., Hernández, R. E., and Périnet, P. (2012). "Anatomical properties of selected hybrid poplar clones grown in southern Quebec," BioRes. 7(3), 3779-3799.

Abstract

The anatomical properties of seven hybrid poplar clones grown in three sites in southern Quebec, Canada were investigated. Radial and longitudinal variations in selected anatomical properties of wood were measured by image analysis of transverse sections and by fiber quality analysis. Results indicate that all measured anatomical properties varied significantly across sites. Clonal variation was highly significant for all anatomical properties studied, and broad-sense heritability ranged from 0.10 (average vessel lumen area) to 0.76 (cell wall area percentage). Genetic gain was positive for all anatomical properties. The variation in radial pattern was characterized by a rapid increase in the first few years in fiber length, width, and proportion, wall thickness, and percent cell wall area. Ray proportion remained constant, whereas the vessel lumen area and proportion decreased with cambial age.


Download PDF

Full Article

ANATOMICAL PROPERTIES OF SELECTED HYBRID POLAR CLONES GROWN IN SOUTHERN QUEBEC

ASM Azmul Huda,Ahmed Koubaa,a,* Alain Cloutier,Roger E. Hernández,and Pierre Périnet c

The anatomical properties of seven hybrid poplar clones grown in three sites in southern Quebec, Canada were investigated. Radial and longitudinal variations in selected anatomical properties of wood were measured by image analysis of transverse sections and by fiber quality analysis. Results indicate that all measured anatomical properties varied significantly across sites. Clonal variation was highly significant for all anatomical properties studied, and broad-sense heritability ranged from 0.10 (average vessel lumen area) to 0.76 (cell wall area percentage). Genetic gain was positive for all anatomical properties. The variation in radial pattern was characterized by a rapid increase in the first few years in fiber length, width, and proportion, wall thickness, and percent cell wall area. Ray proportion remained constant, whereas the vessel lumen area and proportion decreased with cambial age.

Keywords: Hybrid poplar; Wood anatomical properties; Site; Clonal variation; Age; Heritability; Genetic gain

Contact information: a: Canada Research Chair on Wood Development Characterization and Processing, Département des sciences appliquées, Université du Québec en Abitibi-Témiscamingue, 445 Boul. de l’Université, Rouyn-Noranda, Québec, Canada J9X 5E4; b: Université Laval, Département des sciences du bois et de la forêt, 2425 Rue de la Terrasse, Québec, Canada G1V 0A6; c: Ministère des Ressources naturelles et de la Faune du Québec, 2700 Einstein, Québec, G1P 3W8, Canada.

* Corresponding author: ahmed.koubaa@uqat.ca

INTRODUCTION

Forests are one of Canada’s most important natural resources. Poplar species and hybrids are among the most widespread and fastest growing tree species in North America, and along with aspen and mixed wood stands, they make up a substantial part of the natural forest. Poplars are increasingly harvested in Canadian forest industries, and they currently account for over 50% of all hardwoods and 11% of overall timber resources in Canada (Avramidis and Mansfield 2005).

Hybrid poplars are hybridizations of two or more species within the genus Populus, which, as one of the fastest growing temperate trees, has considerable commercial value (Zsuffa et al.1996). Hybrid poplars have been genetically modified through selection and crossbreeding to improve growth rate, trunk form, adaptability, and disease resistance (Hernández et al. 1998; Riemenschneider et al. 2001; Zhang et al. 2003; Pliura et al. 2007). Wood formation is a complex developmental process that includes the differentiation of vascular cambial initials into various xylem tissues, cell elongation, and secondary wall synthesis. As the secondary wall forms, the fiber and vessel cells undergo massive thickening, significantly influencing the wood quality. Wood properties are largely genetically determined (Zobel and Jett 1995) and can show growth variations in width and height depending on site conditions and age. For example, some fast growing tropical tree species consist of juvenile wood only when they are harvested at young age (Zobel and Sprague 1998). For optimum wood utilization, the age effects on wood properties must be determined, including the size and type distribution of cells (Peszlen 1994). In addition, genetic improvements designed to produce harvestable size trees at young age may also produce higher percentages of juvenile wood, which would have significant effects on wood properties and industrial processing applications (Burley and Palmer 1979; Mátyás and Peszlen 1997; Hernández et al. 1998). Accordingly, researchers have analyzed anatomical variations in wood elements within and among clones (Populus spp., Eucalyptus spp., Dalbergia spp.) in order to assess wood quality (Phelps et al. 1982; Koubaa et al. 1998; Rao et al. 2002; Pande and Singh 2005).

Panshin and de Zeeuw (1980) conducted a literature review on longitudinal and radial variations in wood anatomical properties. They found three patterns of radial variation in tracheid and fiber length: 1) a rapid increase followed by constant length from pith to bark; 2) a smooth and continuous increase from pith to bark; and 3) an increase from pith to bark up to a maximum, followed by a smooth decrease. A similar trend was reported for vessel element length and for fiber and vessel diameter, although the increase was moderate. Fiber wall thickness increases from pith to bark in some species and remains constant in others. According to Mátyás and Peszlen (1997), wood properties vary greatly within and among poplar trees. However, the findings on variation in poplar wood properties are inconclusive and in some cases contradictory. More specifically, within-tree variation in anatomical properties in hybrid poplars has not been examined, except for a few studies on fiber length (Holt and Murphey 1978; Murphey et al. 1979; Yanchuk et al. 1984; Bendtsen and Senft 1986; Koubaa et al. 1998; DeBell et al. 1998). These studies found significant clone and longitudinal variation in fiber length. DeBell et al. (1998) found that the variation in fiber length was affected by age but not by growth rate, whereas Koubaa et al. (1998) found a slightly negative correlation between fiber length and growth rate.

Peszlen (1994) found a significant effect of age on poplar anatomical features. She reported an increase in fiber and vessel lumen area from pith to bark, with a rapid increase in vessel lumen diameter in the first few years followed by constant increase toward the bark. Lei et al. (1996) found a similar variation pattern in white oak. Mátyás and Peszlen (1997) found only slight changes in the radial distribution of vessel lumen, fiber lumen, and cell walls in poplar clones. Kern et al. (2005) noted that various fiber features, such as narrower fiber lumen area, greater cell wall thickness, change in the cellulose microfibril angle, or biochemical features of the lignin in cell walls, might mitigate potential mechanical weakening caused by greater vessel lumen area. Based on an understanding of the anatomical characteristics that determine wood quality in hybrid poplar clones, better selection criteria could be developed for specific final products.

Little information is available on the genetic parameters of the wood anatomical characteristics of poplar species, such as heritability and genetic gain, except for a few studies on fiber length (Farmer and Wilcox 1968; Koubaa et al. 1998; Zhang et al. 2012) and fiber width (Wang et al. 1991). Estimating heritability and genetic correlation with traits is particularly important for predicting the genetic gain from cloned material (Foster and Shaw 1988) and for better clonal selection.

Accordingly, knowledge of the variation in fiber anatomy is essential for obtaining improved wood quality, better clonal selection, optimum rotation cycles, and better end uses of hybrid poplar clones. We therefore investigated site, clonal, and within-tree variations in selected anatomical properties of hybrid poplar clones grown in field trials in the southern part of the province of Quebec, Canada. The results will contribute to the selection of superior clones in terms of wood anatomical properties.

MATERIALS AND METHODS

Study Area

Materials for this study were collected from three hybrid poplar clonal trials established by the Direction de la recherche forestière, Ministère des Ressources naturelles et de la Faune du Québec (Department of Forestry Research at Quebec’s Ministry of Natural Resources and Wildlife) between 1991 and 1995. Trees for hybrid clones trials were planted at the Saint-Ours and Windsor sites in 1993 and in 1991 at the Pointe-Platon site (Table 1). Trees for clone DNxM-915508 were obtained from a 1995 trial at the Pointe-Platon site (Table 2). The trial sites are located in Pointe-Platon (46040’N 71051’W), Saint-Ours (45054’N 73009’W), and Windsor (45042’N 71057’W) in southern Quebec, Canada (Fig. 1). The Saint-Ours site is located in the Champlain marine deposit, where the soil consists of a silty clay deposit (40% clay). The two other sites consist of sandy loam soil (Pliura et al. 2007). All sites were originally used for agriculture, but had been abandoned for several years before the hybrid poplar clones were planted. All tree plantation trial sites had a randomized block design with ten blocks each.

Table 1. Site Characteristics of Hybrid Poplar Clonal Trials

Fig. 1. Map of sampling sites located in the southern part of the province of Quebec, Canada

Sample Collection and Preparation

Seven hybrid clones (Table 2) were selected for this study. Five trees per site were randomly sampled for each clone, for a total of 105 trees. Trees were cut from the St-Ours and Windsor sites after 15 growing seasons. Trees were cut from the Pointe-Platon site after 17 growing seasons, except for clone DNxM-915508, which was felled after 13 growing seasons. Disks 10 cm thick were collected from each tree from breast height upward at 2.5 m intervals (Fig. 2) and used for anatomical analysis. Wax was applied to the disk edges to limit drying and prevent decay and other environmental alterations. Samples were then transported to the Wood Research Centre, Université Laval, Quebec, Canada and kept frozen until test sample preparation. A 2.5 cm thick slab was cut along a diameter of each disk (bark to bark passing through the pith) and then conditioned at 20 °C and 60% relative humidity for several weeks until reaching 12% moisture content.

A series of radially oriented sample blocks sized 1 (T) x 1 (R) x 2 (L) cm was systematically cut from annual growth rings (3, 6, 9, and 12) using a precision saw and a chisel. At the annual growth ring with a cambial age of 15, there was insufficient wood to prepare samples for properties measurement by WinCELL, an image analysis system specifically designed for wood cells analysis, from Regent Instruments, Québec, Canada. Cross sections of 20 μm were cut using a rotary microtome with a disposable blade positioned at approximately 15 degrees. Sections were then bleached with sodium hypochlorite solution (80 mL water + 5 drops of bleach) for 1 minute and washed in a distilled water bath for 1 minute. Sections were then double stained with 1% safranin stain for 5 minutes and 0.1% astra blue stain for 15 minutes. Excess stain was removed by washing sections successively in 50, 80, and 100% ethanol solution. Thin sections were further dehydrated using toluene and then permanently mounted on microscope slides with coverslips using permount. Samples were left for two weeks to ensure proper drying of the permount.

Fig. 2. Schematic diagram of the sampling procedure for the analysis of hybrid poplar wood anatomical properties

Table 2. Studied Hybrid Poplar Clones

Trees for the clone DNxM-915508 at Pointe-Platon were sampled from another trial PLA16495.

Samples were photographed at ×50 magnification using a Leica compound microscope (DM 1000) equipped with a PL-A686 high resolution microscopy camera to capture black and white images (tiff electronic file format) at 1200×1600 resolution using a green filter to maximize contrast (Fig. 3). WinCELL Pro 2004a (Régent Instruments Inc. 2004) was used to measure average fiber wall thickness, fiber lumen area, vessel lumen area, fiber diameter, and vessel diameter. Average fiber diameter accounts for average fiber lumen diameter and the respective two-sided fiber wall thickness. Cell wall area (%) was estimated by subtracting the percent areas of the vessel lumen, ray, and fiber lumen from the image area (Peszlen 1994).

The proportion of tissue in the different cell types was estimated from two cross-sections of each wood block. Vessel cells were distinguished from fiber and ray by analyzing 570 µm2fields (corresponding to four squares of the grid) and noting the tissue types within the field. Fiber proportion was measured using the same method. Ray proportion was obtained by subtracting from unity the vessel and fiber areas.

Fig. 3. Microscopic cross-section of hybrid poplar wood: V = vessel, R = ray, F = fiber, Scale bar = 50 µm

The same thick slabs used for wood anatomy analysis were used to analyze fiber properties from pith to bark at different tree heights using a Fiber Quality Analyzer (FQA) (OpTest Equipment Inc., LDA02, Ontario, Canada). A series of radially oriented thin tangential sections was systematically cut from annual growth rings 3, 6, 9, 12, and 15 for fiber length and fiber width measurement. Tangential sections were macerated using Franklin’s method. Sections were soaked in Franklin solution (1:1, glacial acetic acid: 30% hydrogen peroxide) and heated at 70 °C for approximately 48 hours. The solution was decanted and the remaining fibrous material was washed under vacuum with deionized water until reaching neutral pH. The distribution of fiber length and width was measured automatically using the FQA. A total of 5,000 fibers were measured for each sampled growth ring.

The SAS® statistical package, version 9.2 (SAS 2008) was used for all statistical analyses. Normality and homogeneity of variance for residuals were tested using UNIVARIATE statistics. Data transformations were not considered necessary to satisfy the assumptions of the analysis of variance and other analyses. Analyses of variance were performed using PROC GLM with Type III (partial sums of squares) estimates to assess the relative magnitude of each source of variation. Analyses were performed among and within sites and clones. The variance in anatomical properties variables among sites was analyzed using the following mixed linear model (Eq. 1),

Xijlm = μ + Si + Cj + (S x C)ij + Al + Hm+ (A x H)lm + εijlm (1)

where Xijlm is an observation on the lth age and mth height of the jth clone from the ith site; μ is the overall mean; Al, Hm, and (A x H)lm are the fixed effects and their interactions, respectively; Si is the fixed effect due to the ith site; Cj is the fixed effect due to the jth clone; (S x C)ij is the interaction between site and clone; and εijlm is the random error. Some ratios involved more than one mean square in the denominator and were tested with approximate degrees of freedom. Tree effects and the site-clone-age interaction were not considered in the analysis, as preliminary testing showed negligible contribution to the total variance. Furthermore, in many cases, the variance component for these terms could not be estimated or was not significant.

Tukey’s Studentized Range (HSD) was used to test the statistical significance (at p<0.05) of differences among means of clones for each site (PROC GLM, SAS). The variance components were estimated in the model using VARCOMP with the restricted maximum likelihood method (REML) and expressed as a percentage (VAR). The individual broad-sense heritability (H2) was calculated from the variance estimates, as follows (Eq. 2) (Becker 1984; Falconer and Mackay1996),

 (2)

where  and  are the genotypic and phenotypic variance, respectively. Phenotypic variance (  )was calculated as shown in Eq. (3).

 (3)

The genotypic coefficient of variation (CVG) and the phenotypic coefficient of variation (CVp) were calculated from Eqs. 4 and 5, respectively (Burton 1952; Henderson 1953).

 (4)

 (5)

The mathematical expression for the genetic gain (G) is expressed in Eq. 6. The potential genetic gain from individual tree selection is computed by selection differential (Eq. 7) and 10% selection intensity,

 (6)

 (7)

where  is the heritability, S is the selection differential, is the selection intensity (10%), and σis the phenotypic standard deviation.

RESULTS AND DISCUSSION

Within- and Among-Site Variation

Growth site variable showed a highly significant effect on all studied wood anatomical properties (Table 3). Table 4 presents the mean and standard error of the anatomical properties of all studied hybrid poplar clones.

Trees from the Saint-Ours site showed the highest fiber length (0.99 ± 0.17 mm), and trees from Pointe-Platon showed the lowest (0.90 ± 0.19 mm), with a 10% difference between the lowest and highest length (Table 4). Highest fiber diameter, vessel lumen area, and vessel diameter were also found in trees from Saint-Ours, with the lowest from Pointe-Platon.

For fiber width, the Windsor site showed the highest average and Pointe-Platon the lowest average (Table 4). Pointe-Platon trees showed higher fiber wall thickness than trees from the two other sites, with a 12.4% difference between the highest and lowest average (Table 4).

Pointe-Platon trees showed the lowest average fiber lumen area, fiber diameter, vessel lumen area, and vessel diameter (Table 4). Pointe-Platon trees showed the highest fiber proportion and cell wall area, and the lowest average vessel lumen area and diameter and ray proportion (Table 4).

The significant site effect concurs with previous studies (Murphey et al. 1979; Phelps et al. 1982; Yanchuk et al. 1984; Bendtsen and Senft 1986; DeBell et al. 1998; Chauhan et al. 1999; Zhang et al. 2003; Pliura et al. 2007). Several factors may explain this significant site effect on the anatomical properties, including edaphic and climatic conditions (Peszlen 1994; Pliura et al. 2007). Trees from the Saint-Ours site showed higher fiber length, vessel proportion, and vessel dimensions. This could be explained by higher moisture availability and better drainage conditions as well as the soil surface deposition at the Saint-Ours site compared to the two other sites.

Clonal Variation in Fiber Anatomical Properties

The clone effect on the examined wood anatomical properties was highly significant (Table 3). The variance component analysis indicated that the clone effect varied among the studied properties, ranging from 0.2% to 47.9%. The clone effect on fiber width, although significant at the 0.05 probability level, was low (1.8%). Similarly, the clonal variance component of the vessel lumen area was also negligible (0.2%). In contrast, the variance component for the cell wall area percentage was the highest (47.8%) among all the anatomical properties (Table 3).

Differences among clones were also significant for wood element proportions (fiber, vessel, and ray), as shown in Table 3. The highest interclonal variation was observed for vessel proportion, whereas ray proportion showed a significant clone effect but a low variance component (Table 3).

The clone effect was reflected in the differences in means among clones for all studied properties. For example, clone DxN-4813 showed the highest fiber wall thickness, whereas clone DxN-3565 showed the highest fiber lumen area and average fiber diameter. Clone DxN-131 showed the highest vessel lumen area, and clone DxN-3570 showed the highest average vessel diameter (Table 4).

Table 3. Results of the Analysis of Variance of Wood Anatomical Characteristics of Hybrid Poplar Clones (F values and variance components of 11 wood anatomical properties: fiber length (FL), fiber width (FW), fiber wall thickness (FWT), average fiber lumen area (AFLA), average fiber diameter (AFD), average vessel lumen area (AVLA), average vessel diameter (AVD), fiber proportion (FP), vessel proportion (VP), ray proportion (RP), and cell wall area (CWA)).

*Significant at P<0.05 probability level.

**Significant at P<0.01 probability level.

ns Non-significant at P<0.05 probability level.

Variance component as a percentage of the total variance.

Table 4. Least Squares Means of Clones at Different Sites and Multiple Comparison Tests of Hybrid Poplar Clones (fiber length (FL), fiber width (FW), fiber wall thickness (FWT), average fiber lumen area (AFLA), average fiber diameter (AFD), average vessel lumen area (AVLA), average vessel diameter (AVD), fiber proportion (FP), vessel proportion (VP), ray proportion (RP), and cell wall area (CWA)).

Means within a column followed by the same letter are not statistically different at p=0.05 for each site separately.

The highest fiber proportion was found for clone DxN-3565 and the lowest for clone DNxM-915508, for a 17.4% difference (Table 4). The highest vessel proportion percentage was found for clone DxN-3