Non-structural carbohydrates (NSCs) are central to plant metabolism as the stored products of photosynthesis, building blocks for growth, and fuel for respiration. Their measurement in woody plant tissue is challenging, which has limited our ability to integrate and generalize patterns in plant C balance, particularly when synthesizing results from different studies. A recent assessment found that NSC concentrations measured for a common set of samples can vary by an order of magnitude, but sources for this variability were unclear. In order to diagnose the problem and offer a pathway for solution, we measured a standard set of nine tree tissues and two synthetically constructed samples with known sugar and starch concentrations using three methods (enzymatic digestion, ion chromatography, phenol-sulfuric acid) in six laboratories (two laboratories per method). In our study, these methods differed in which sugars were quantified; enzymatic digestion quantified only glucose, fructose, and sucrose, ion chromatography additionally quantified galactose, while the phenolic sulfuric acid method quantified a large range of mono- and oligosaccharides. We also tested whether centralizing either the extraction or quantification of NSC in one laboratory improved precision, and how variation in sample preparation and storage affected results.
Results/Conclusions
For all methods, the measured values for two artificial samples with known concentrations were no more than 11.5% different from the expected concentrations. However, for natural tree samples, differences in which sugars were quantified in each method was the largest consistent source of variation among these methods. For some natural samples, quantification using the phenol-sulfuric resulted in a doubling of sugar relative to the other methods, demonstrating that trees allocate C to a wide range of sugar molecules. Across methods for artificial samples, water extraction typically had lower accuracy than ethanol extraction, especially for the phenol-sulfuric acid method. Differences in sample handling and storage had a minimal effect on NSC, and centralizing either NSC extraction or quantification in one laboratory did not improve results beyond what was achieved by using standardized protocols. Overall, our results demonstrate that reasonably precise and accurate measurements of NSCs can be made with careful protocols. We recommend that researchers construct their own artificial samples with known NSC component concentrations as an internal quality-check procedure. With standardized protocols, the variability from different quantification methods is due to which sugars can be quantified, not imprecise methods. This will remain a challenge for comparing NSC measurements among laboratories.