Combining iron affinity-based fractionation with non-targeted LC-ESI-TOFMS for the study of iron-binding molecules in dissolved organic matter

Abstract

The low solubility of inorganic iron(III) in seawater leads to very limited availability of this important micronutrient for marine organisms. Estuarine or oceanic iron is almost entirely bound to organic ligands of mainly unknown chemical structure. In this context, riverine input of iron rich, land-derived dissolved organic matter (DOM) can play an important role in coastal areas and investigation of potential Fe-ligands in DOM is of high interest. Previous studies have suggested that iron is predominantly bound to the high molecular weight fraction of DOM, but distributed over the entire size range. Logically, structural elucidation needs to start from the smallest building blocks. A model study targeting low molecular weight iron-binding constituents in Suwannee River natural organic matter (NOM) using Fe-loaded Chelex or silica for immobilized-metal affinity (IMAC)-based fractionation was undertaken. The binding strengths of different compounds could be qualitatively assessed using a differential analysis workflow. IMAC-fractionated samples were acidified and analyzed via liquid chromatography high resolution mass spectrometry (LC-HRMS) and molecular formulas were assigned using state of the art software. A total of 144 Fe-binding constituents in Suwannee River NOM were found to be of interest with the largest number observed to interact with Chelex at pH 4 (55%), and the smallest with silica at neutral pH (24%). Most binding constituents were found in the lignin- and tannin-type region of the van Krevelen plot. Results from this study support the hypothesis that very low molecular weight constituents (below 300 Da) can play a role in the iron binding mechanism of DOM and demonstrate that the employed analytical workflow is suitable for their detection.

Keywords: dissolved organic matter; high-resolution mass spectrometry; immobilized-metal affinity chromatography; iron; non-targeted analysis.

Graphical Abstract

Iron affinity chromatography supported non-targeted analysis of iron-binding molecules in natural waters.

Fig. 1

Color coded list containing all assigned compounds showing the calculated molecular formula, exact monoisotopic mass, retention time, software quality score, mass error in ppm, and the maximum and median of the chromatographic peak areas. Color coding is according to the approach described in Table 1. The red letters a–f are indicating the compounds listed in Table 2.

Fig. 2

Van Krevelen plots for annotated compounds at all experimental conditions on both resins. Each point corresponds to a single compound with a retention time. If a molecular formula was listed several times in Fig. 1, then the one showing the strongest Fe-interaction is depicted.

Fig. 3

Two examples of chromatographic separation of isobaric/isomeric compounds demonstrated to be necessary for correct discrimination of varying Fe-interaction behavior. A mass extraction window of 20 ppm for the monoisotopic ion was used.