COMPARING GC×GC-TOFMS-BASED METABOLOMIC PROFILING AND WOOD ANATOMY FOR FORENSIC IDENTIFICATION OF FIVE MELIACEAE (MAHOGANY) SPECIES

Abstract

Illegal logging and associated trade have increased worldwide. Such environmental crimes represent a major threat to forest ecosystems and society, causing distortions in market prices, economic instability, ecological deterioration, and poverty. To prevent illegal imports of forest products, there is a need to develop wood identification methods for identifying tree species regulated by the Convention on International Trade in Species of Wild Fauna andFlora in Trade (CITES) and other look-alike species. In this exploratory study, we applied metabolomic profiling of five species (Swietenia mahagoni, Swietenia macrophylla, Cedrela odorata, Khaya ivorensis, and Toona ciliata) using two-dimensional gas chromatog- raphy combined with time-of-flight mass spectrometry (GC3GC-TOFMS). We also performed qualitative, quantitative (based on the measurement of vessel area, tangential vessel lumina diameter,vessel element length, ray height, and ray width), and machine-vision aided (XyloTron) wood anatomy on a subsample of wood specimens to explore thepotential and limits of each approach. Fifty dried xylaria wood specimens were ground, extracted with methanol, and subsequently analyzed by GC3GC-TOFMS. In this study, the four genera could easily be identified using qualitative wood anatomy and chemical profiling. At the spe- cies level, Swietenia macrophylla and Swietenia mahagoni specimens were found to share many major metabolites and could only be differentiated after feature selection guided by cluster resolution (FS-CR) and visualization using Principal Component Analysis (PCA). Expectedly, specimens from the two Swiete- nia spp. could not be distinguished based on qualitative wood anatomy. However, significant differences in quantitative anatomical features were obtained for these two species. Excluding T. ciliata that was not included in the reference database of end grain images at the time of testing (2021), the XyloTron could successfully identify the majority of the specimens to the right genus and 50% of the specimens to the right species. The machine-vision tool was particularly successful at identifying Cedrela odorata samples, where all samples were correctly identified. Despite the limited number of specimens available for thisstudy, our preliminary results indicate that GC3GC-TOFMS-based metabolomic profiles could be used as comple- mentary method to differentiate CITES-regulated wood specimens at the genus and species levels.