par Brabant, Frédéric ;El Amri, Saïda ;Tison, Jean-Louis
Référence Limnology and oceanography, methods, 9, page (261-274)
Publication Publié, 2011
Article révisé par les pairs
Résumé : The melting of sea ice samples is acknowledged to be deleterious to sympagic microorganisms due to the hypo-osmotic shock undergone by the organism when released from high salinity brine inclusions into the sample melt. Because melting of sea ice samples was also anticipated to modify the initial proportions of dimethylsulfide (DMS), dimethylsulfoniopropionate (DMSP), and dimethylsulfoxide (DMSO), three sample treatments were tested on an Antarctic sea ice sample, with the aim of identifying an efficient procedure that could routinely be applied for the determination of DMSO in sea ice. Herein, it was demonstrated that purging the melted sample before the determination of DMSO in the sample via an enzyme-linked method produced reliable DMSO results (215.8 ± 8.9 nmol L -1, precision 4.1%). However, analysis revealed that the unintentional enzymatic cleavage of DMSP through the subsequent production of interfering DMS during melting caused an overestimation of the DMSO content in the sample by more than 59% and concurrently an underestimation of the DMSP content by approximately 9%. The sequential determination of DMSP after the DMSO determination by the enzyme-linked method was shown to be problematic. To circumvent all of these issues, we recommend an analytical procedure for the sequential determination of DMS, DMSP, and DMSO in sea ice. Ultimately, the first depth profile of DMSO at high resolution in sea ice was produced. The depth-integrated DMSO concentration in sea ice was determined to be 718 μmol m -2, which indicated that Antarctic sea ice is a potentially important source of DMSO for the Southern Ocean. © 2011, by the American Society of Limnology and Oceanography, Inc.