ORNL/TM-2001/41

Development and Validation of Marker-Aided Selection Methods for Wood Property Traits in Loblolly Pine and Hybrid Poplar

G. Tuskan, D. West, Environmental Sciences Division, Oak Ridge National Laboratory

M. Davis, C. Elam, R. Evans, National Renewable Energy Laboratory, Golden, Colorado

H. Bradshaw, University of Washington, Seattle, Washington

D. Neale, M. Sewell, U.S. Forest Service, Berkeley, California

N. Wheeler, B. Megraw, K. Jech, Weyerhaeuser Corp., Centralia, Washington

C. Weirman, Boise Cascade Corp., Wallula, Washington

R. Dinus, Private Consultant, Bellingham, Washington

Date Published: June 2001

Prepared for the U.S. Department of Energy Office of Industrial Technologies Budget Activity Number ED 18 01 00 0 Prepared by OAK RIDGE NATIONAL LABORATORY Oak Ridge, Tennessee 37831 managed by UT-BATTELLE, LLC. for the U.S. DEPARTMENT OF ENERGY under contract DE-AC05-00OR22725

Click here for the complete report in PDF format (107 kb)
 

EXECUTIVE SUMMARY

Wood properties influence pulp and paper quality. Certainly, overall pulp yields are directly related to the cellulose content, changes in hemicellulose content are associated with changes in pulp cohesiveness, and pulping efficiency is related to lignin content. Despite the importance of wood properties on product quality, little progress has been made in improving such traits because current methods of assessing wood and fiber characteristics are time-consuming, expensive, and often imprecise. Genetic improvement of wood and fiber properties has been further hampered by the large size of trees, delayed reproductive maturity and long harvest cycles. Recent developments in molecular genetics will help overcome the physical, economic and biological constraints in assessing and improving wood properties. Genetic maps consisting of numerous molecular markers are now available for loblolly pine and hybrid poplar. Such markers/maps may be used as part of a marker-aided selection and breeding effort or to expedite the isolation and characterization of genes and/or promoters that directly control wood properties.

The objectives of this project are:

1. to apply new and rapid analytical techniques for assessing component wood properties to segregating F₂ progeny populations of loblolly pine and hybrid poplar,
2. to map quantitative trait loci and identify molecular markers associated with wood properties in each of the above species and
3. to validate marker-aided selection methods for wood properties in loblolly pine and hybrid poplar.

We have developed two rapid assays, pyrolysis molecular beam mass spectroscopy (pyMBMS) for cell wall chemical composition and computer-assisted tomography x-ray densitometry (CT scan) for wood specific gravity, annual ring boundaries and within-ring specific gravity for Populus and loblolly pine. The techniques 1) can be completed in a fraction of the time (~1-2 minutes per sample) typically required for traditional analytical methods and 2) utilize small subsamples or are non-destructive.

We have detected nine quantitative trait loci (QTLs) in loblolly pine for wood specific gravity (wsg) (explaining 34% of variance) and 5 for microfibrile angle (mfa) (44% variance). QTLs for wsg appear to be either early- or latewood specific, whereas QTLs for mfa were detected for both early- and latewood. Twelve unique QTLs influencing cell wall chemistry (cwc) were detected using the one-QTL model and most QTLs (75%) were verified by analyses of multiple peak values. Four additional cwc QTLs were detected using the two-QTL model. Significant differences in chemical contents were observed among the populations from NC vs. OK, suggesting that QTLs interact with environmental location.

We have determined lignin content (averaging 24.8% on a dry weight basis) and syringyl to guaiacyl ratios (averaging 1.9:1.0) across all 310 full-sib Populus progeny. Glucan content, including fractions from cellulose and some subcomponents of hemicellulose, averaged 45.4%. Estimates of mannan, arabinan and xylan averaged 2.7%, 0.5% and 17.0%, respectively. Mean specific gravity was 0.34 and ranged from 0.24 to 0.48. Glucan content and xylan content are negatively correlated (r = -0.91), as were arabinan content and extractive content (r = -0.76) and arabinan content and S/G ratio (r = -0.73). There were no detectable correlations between lignin content and arabinans, lignin content and extractives content, extractives content and mannans, and glucans and extractives content. The genetic analysis indicated that there are multiple alleles contributed by each grandparental genotype.