Wood Identification Methods Of Two Pterocarpus Species Based On Chemical Fingerprint From DART-FTICR-MS And GC-MS

With the sharp increase in the consumption of natural forest resources around the world,wood,especially endangered and precious wood species,has quickly become the focus of attention of the international community.Correct understanding and identification of timber is an important prerequisite for the rational and effective protection,management and utilization of timber resources,as well as the technical basis for evading and resolving the risks of timber import and export trade.Traditional wood identification techniques are based on wood anatomy and are achieved through a combination of macroscopic and microscopic features of wood.The method has complicated operation process and high requirement for professional knowledge,and it is difficult to identify wood at species-level.The newly developed chemical fingerprint method helps to break the limitations of traditional technology and provide new technical support for protecting timber resources and legal trade of timber.In this study,two Pterocarpus species,P.santanlinus and P.tinctorius,which are difficult to distinguish on the market,were studied in Direct analysis in real time-Fourier transform ion cyclotron resonance mass spectrometry(DART-FTICR-MS)and gas chromatography-mass spectrometry(GC-MS)to analyze the difference in the main chemical composition of P.santanlinus and P.tinctorius.P.santalinus and P.tinctorius,subjected to eight different treatments,i.e.,air-dried wood chips,wood chips treated at 70℃,wood chips treated at 120 0C,wood powder,distilled water extract,ethanol and water extracts,ethyl acetate extract and benzene-alcohol extract,we observed distinct chemical signatures for the wood samples from the two species,enabling rapid species-level identification when multivariate statistical analysis was adopted.The goal of this study was to explore the effects of solvent types,sample physical form,and drying treatment on the chemical fingerprint of the wood by DART-FTICR-MS and GC-MS.At the same time,combined with single-dimensional and multivariate statistical analysis methods,screened out the marker components for the differentiation of P.santalinus and P.tinctorius wood.Based on DART-FTICR-MS and GC-MS fingerprint methods,the accurate identification between P.santalinus and P.tinctorius was realized,which provided a theoretical basis for the further development of chemical classification of wood species.The main results are summarized as follows:(1)The macrostructural and microstructural features of the transverse,radial and tangential sections of P.santalinus and P.tinctorius are very similar.The water extract indicated that the heartwood fluorescence colors of P.santalinus and P.tinctorius were indistinguishable.The above results show that it is difficult to distinguish the two species using the traditional wood identification technique,and further development of new methods is needed to assist it.(2)DART-FTICR-MS enables rapid detection of wood samples.Both the common peaks and their own specific peaks observed in the DART-FTICR-MS spectrum of P.santalinus and P.tinctorius.Overall,there is a significant difference in the DART-FTICR-MS profiles of the two species.After multivariate statistical analysis,it was found that samples of air-dried wood chips,wood chips treated at 70℃,wood powder and ethyl acetate extracts exhibited high prediction accuracy(100%).The classification ability of wood chips treated at 120 ℃ between these two species was the worst(only 66.67%),which indicated that the high temperature drying treatment would affect the wood classification of P.santalinus and P.tinctorius.Considering the difficulty of sample preparation and the high prediction accuracy,air-dried wood chips were selected as the optimal sample for wood identification based on DART-FTICR-MS method.Combined with the S-plot and VIP plot in the OPLS-DA model,and p value in the t-test,five differential markers were screened out from the air-dried wood chips,namely 477.3,254.2,473.3,272.2 and 258.2 m/z.The relative peak intensity of these five different variables in the P.santalinus samples was significantly higher than that in the P.tinctorius samples.(3)The chemical fingerprints of P.santalinus and P.tinctorius were established by GC-MS.There are large differences in the types and relative contents of the volatile chemical components of P.santalinus and P.tinctorius,which provide a theoretical basis for the identification of these two species based on GC-MS.According to the principal component analysis,all the samples from different treatment conditions can be clustered according to their own category except for the wood chips treated at 120 ℃.The discriminant model of wood classification was established using OPLS-DA and then the unknown samples from the test set were predicted.The results showed that air-dried wood chips,wood chips treated at 70 ℃,wood powder,ethanol and water extracts exhibited high predictive power(predictive accuracy was 100%).Similar to the results of DART-FTICR-MS,the worst classification ability was also observed in the wood chips treated at 120 ℃(accuracy rate was 77.78%).It was again confirmed that high temperature drying treatment may affect the ability to identify wood species based on chemical fingerprint methods.Considering the advantages of simple sample preparation and high classification accuracy,air-dried wood chips was recommended as the optimal samples when the GC-MS method was used to identify P.santalinus and P.tinctorius.Based on the S-plot and VIP plot in the OPLS-DA model,and p-value in the t-test,2-Naphthalenemethanol,1,2,3,4,4a,5,6,7-octahydro-a,a,4a,8-tetramethyl-,(2R-cis)-(15.69 min)and 2-Naphthalenemethanol,decahydro-α,α,4a-trimethyl-8-methylene-,[2R-(2α,4aα,8aβ)]-(15.38 min)were screened out as the marker components to distinguish the air-dried wood chips of P.santalinus and P.tinctorius.(4)The similarity network fusion algorithm was used to fuse the data of DART-FTICR-MS and GC-MS to explore whether the similarity network fusion algorithm can improve the classification accuracy.The similarity matrix diagram shows that the colors of squares in the same wood species are more similar.Compared with the results before fusion,the discrimination ability for P.santalinus and P.tinctorius was significantly improved according to the results of similarity matrix diagram.The samples from the test set were predicted based on the fused data to further verify the predictability.Four different treatment conditions samples from the test set,including high temperature drying wood chips,benzene-alcohol extract,ethanol and extract,and ethyl acetate extract,were accurately classified,and the prediction accuracy reached 100%.The predictive power was significantly improved compared to the results based on a single data type.This result indicated that the similarity network fusion analysis of DART-FTICR-MS and GC-MS data can realize the complement of two different types of data and finally improve the discriminant classification ability of P.santalinus and P.tinctorius.