CHEMICAL DIVERSITY OF SOIL LIPIDS REFLECTS SURROUNDING BIODIVERSITY IN A FRENCH PEATBOG - INSU - Institut national des sciences de l'Univers
Poster De Conférence Année : 2013

CHEMICAL DIVERSITY OF SOIL LIPIDS REFLECTS SURROUNDING BIODIVERSITY IN A FRENCH PEATBOG

Résumé

The aims of this study are to characterize the spatial variability of biomarker imprints in a peatland and to test to which extent this variability can be related to the surrounding vegetation. Since the mid 1970’s La Guette peatland (région Centre, France) has been colonized by vascular plants. Nowadays it is totally colonized by Molina caerulea and, to varying extent, by Pinus sylvestris and Betula pendula. In nine stations representative of distinct vegetation cover, we analysed the lipid composition of soil developed under Molinia caerulea. The lipids of the six most abundant plants encountered in the peatland were analysed as reference samples: Sphagnum rubellum, Erica tetralix, Calluna vulgaris, Molinia caerulea, Betula pendula and Pinus sylvestris. For all these samples, lipids were ultrasonically extracted with DCM and then separated into neutral, acidic and polar compounds. The neutral fraction was further separated by flash chromatography to afford (i) aliphatics and aromatics; (ii) ethers, esters, ketones and acetates and (iii) alcohols. 5α-cholestane was added prior to analysis for quantitation. Lipid fractions were then identified and quantified by gas chromatography-mass spectrometry. In order to quantify their density, trees were counted within circles of increasing radius (giving areas of 3.14; 12.6; 28.3; 50.3 and 78.5 m2) around the sampling plot. After counting, total tree density was determined as well as individual P. sylvestris and B. pendula density. For each sample (n=9) regressions were calculated between (i) lipids and (ii) the density of the tree (n=9). This calculation was reproduced for each counting surface. Then, the resulting regression coefficient was plotted against each surface of counting. Apart from ubiquist compounds such as n-alkanes and triterpenoid ketones, we found in soils several biomarkers for which the source could be constrained. Taraxeryl, friedelanyl and moretenyl acetates were detected. Friedelanyl and moretenyl acetates were found in high amounts in open vegetation plots (9.7 and 2.9 µg.g soil-1, respectively). They were also detected in E. tetralix and C. vulgaris (Ericaceae) and, probably due to contamination, in S. rubellum. Multiflorenyl and two unknown acetates, also present in low contents, may originate from undetermined, less abundant plant. Four methoxy-serratenes (Le Milbeau et al., 2013) and seven tricyclic diterpenes were detected in both soils and P. sylvestris needles. Strong correlation (r = 0.997; p ≤ 0.0001) of their respective abundances in soils attest to their common origin and similar conservation potential. The presence of betulin and related compounds in soil samples and in B. pendula leaves are in agreement with the admitted use of these biomarkers to track the former presence Betula species in soils and sediments (i.e. Otto and Simoneit, 2001). The source of these biomarkers being defined, we then assessed their spatial distribution at the nine sampling sites. Although not abundant, triterpenyl acetates, methoxy-serratenes, diterpenes and betulin derivatives exhibit a higher coefficient of variation along sampling sites than other compound such as n-alkanes or triterpenoid ketones. The correlation between methoxy-serratenes and the number of P. sylvestris in the nine sampling sites increases with the surface used for the calculation (r = 0.93, p<0.022). These values indicate that specific biomarkers of P. sylvestris in soil samples and surrounding vegetation are spatially linked. The same applies for B. pendula and betulin derivatives (r = 0.96, p<0.008). The degradability of the betulin and its relatively high solubility may induce lost and leaching out of the sampling point. Our results provide evidence that selected biomarkers in soil samples can give a reliable picture of the surrounding vegetation. It also quantifies the area of influence of trees on the soil lipid composition, over 20 m². The approach combining tree counting and quantitation of specific biomarkers that was developed in the frame of this study proved to be efficient in evidencing the spatial control of the surrounding vegetation on lipids distribution in a peatland invaded by vascular plants. REFERENCES Le Milbeau, C., Lavrieux, M., Jacob, J., Bréheret, J.G., Zocatelli, R., Disnar, J.R., 2013. Diversity of methoxy-serratenes in a soil under a conifer forest and their potential as biomarkers of Pinaceae. Organic Geochemistry 55, 45-54. Otto, A., Simoneit, B.R.T., 2001. Chemosystematics and diagenesis of terpenoids in fossil conifer species and sediment from the Eocene Zeitz formation, Saxony, Germany. Geochimica et Cosmochimica Acta 65:20, 3505–3527.
Fichier principal
Vignette du fichier
Zocatelli_IMOG_2013_1.pdf (1.67 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

insu-01295607 , version 1 (31-03-2016)

Licence

Identifiants

  • HAL Id : insu-01295607 , version 1

Citer

Renata Zocatelli, Jérémy Jacob, Sébastien Gogo, Claude Le Milbeau, Fatima Laggoun-Défarge. CHEMICAL DIVERSITY OF SOIL LIPIDS REFLECTS SURROUNDING BIODIVERSITY IN A FRENCH PEATBOG. 26th International Meeting on Organic Geochemistry, Sep 2013, Tenerife, Spain. 2013. ⟨insu-01295607⟩
436 Consultations
82 Téléchargements

Partager

More