Abstract

Genotypical effects on wood formation in aspen (Populus tremula x tremuloides) have been studied at morphological, ultrastructural and micro-distributional levels. To characterize transgenically in­duced modifications, light-, fluorescence-, UV-, scanning- and transmission electron microscopy have been applied complement­ed by immunolabeling, histochemistry, multivariate statistics and image analysis. Tension wood formation in aspen has also been studied with a focus on xyloglucan micro-distribution. To perform quantitative screeing of wood immunofluorescenc, a novel image analysis protocol was developed and applied. The protocol performs immuno-signal isolation to calculate labeling ra­tio, termed coverage. Coverage is proportional to epitope avail­ability, in relation to immunoprobe performance. The protocol was shown to perform favorably compared to operator supervised analysis and quantification. Morphological characterization of the twelve genotypes showed significant modifications only for fiber properties. Coverage char­acterization showed significant modifications for LM-10 (xylan) and JIM-5 (pectin) antibody labeling. In the cell wall architecture, structure is often as important as composition. The multivariate analysis confirms previously reported relationships between indi­vidual fiber measurements, and highlights novel immunolabeling relationships. The results show that detailed morphological and coverage characterization can be a valuable addition to the semi-high throughput screening of wood tissue. Utilizing two complementary probes the presence of xyloglucan during g-layer formation of tension wood in aspen was shown at the ultrastructural level. The observation of increased xyloglucan labeling between the g-layer and S2 strengthens the hypothesis that xyloglucan acts as conductor of tensional forces between the g-layer and the rest of the fiber wall.