Texas A&M University introduces exopolymeric substances as agents in enhancing the self-cleansing capacity of natural waters
Most chemical substances that reach the ocean through rivers, the atmosphere, or through subterranean groundwater outflows, e.g. nutrients, trace elements, radioactive substances, trace organics, oil from oil spills, etc. are removed by sorption to particles that sweep through the ocean. Some of these are introduced by rivers, some through the atmosphere, and some are newly formed.
The net input of particulate organic carbon (POC) into the ocean is called new production and is, at steady state, equal to the POC flux, removing not only trace substances but greenhouse gases such as CO2. This is why particle flux is also called the carbon pump, as it is relevant for the removal of CO2 from the atmosphere.
The mechanisms of removing trace substances through sorption vary (e.g. surface adsorption or absorption, co-precipitation, incorporation into mineral or plankton debris) and result in oceanic residence times of decades to thousands of years for these substances, depending on the surface reactivity of these substances and the metabolic activity of photosynthetic, eukaryotic and procaryotic organisms, resulting in natural organic matter (NOM) molecules making up a variable but significant fraction of these particles.
NOM has important roles for regulating the fate of many trace substances as well as carbon as it contains both hydrophilic and hydrophobic binding sites. Often, trace element models ignore NOM and consider the main sorbents and mineral carrier phases only, which are either derived from land or freshly produced as shells to protect micro-organisms. This is because mineral matter makes up the main components that reach the bottom of the ocean, and their concentration can often be correlated to that of trace substances in sinking matter (Fig. 1). For these NOM-containing particles to be able to sink to the bottom of the ocean, however, denser mineral matter is needed.
Although we have known about the importance of NOM for removing radioactive and stable trace metals from the ocean for some time, only in recent decades has this concept been more widely accepted. However, not until recently were the main organic sorbents for some radionuclides or stable trace elements in particles identified. While the physico-chemical identity of low molecular weight metal-binding chelates is fairly well known, chelates in macromolecular organic matter are often not. Part of the problem is that NOM is heterogeneous and multi-functional, and specific moieties can be in flexible or rigid regions – some of which are also involved in electron, proton and ROS exchanges.
Nature and the roles of colloidal, macromolecular exopolymeric substances (EPS) in aquatic systems
Peter Santschi’s team showed, for the first time using Atomic Force Microscopy (AFM) and Transmission Electron Microscopy (TEM), that colloidal macromolecular NOM contains EPS and consists of macromolecules shaped as fibrils with one to three nanometres in diameter and one to six micrometres in length. Fibrils have a modern radiocarbon age when made up of pure polysaccharides (Santschi et al. 1998). Besides acidic and neutral polysaccharides, EPS also contain proteins and other macromolecules. EPS, in the form of transparent exopolymeric particles (TEP), cannot only form flocs (marine or lake snow) and play roles in colloid trace metal and radionuclide scavenging from the water, or provide templates for FeOOH, MnO2, CaCO3 and SiO2 growth, or can alter the surface characteristics of suspended particles, they can also facilitate microbial adhesion to surfaces, and thus provide the matrix components of biofilms and bind extracellular enzymes in their active forms, and are also the first line of defence against toxic substances by immobilising them.
Importance of marine snow formation for the self-cleansing capacity of aquatic systems
Santschi’s team recently showed that natural organic matter that act as templates for building silica and carbonate shells of phytoplankton can contain important binding sites for radioactive and stable trace elements. Other important chelating compounds are contained in microbially derived EPS that make up the bulk of what is called ‘marine snow’ or ‘lake snow’, and which in turn can also provide the organic ‘glue’ for mineral aggregate build-up, thus providing a vehicle for transport to the seafloor (Fig. 2).
What is surprising is how versatile NOM is in its functionalities, making them suitable not only for taking up ionic compounds or moieties, but also uncharged, more hydrophobic molecules. Many of these organic compounds are thus multifunctional and contain ionic and hydrophobic regions, and thus, are amphiphilic, i.e. surfactant-like. This makes them suitable for binding different compounds with vastly different properties. Thus, marine snow is one of the main agents or ‘actors’ in the ‘self-cleansing capacity’ of the ocean and aquatic systems in general.
Colloidal pumping of trace substances bound to EPS and other macromolecular NOM compounds
Santschi’ s interest in this self-cleansing capacity started while he was working on his PhD thesis at the Chemistry Department of the University of Bern, Switzerland, in which he studied the chemical processes in Lake Biel. Later, while in the US, he pursued sorption and removal mechanisms of trace substances through radioactive tracers in the ocean, which he found often to be associated with colloidal, nano-particulate substances. Along the way, Santschi and Honeyman pioneered insights into the kinetics of ‘colloidal pumping’ (i.e. further aggregation of colloids into particles; Honeyman and Santschi, 1989) (Fig. 3).
This colloidal pumping mechanism, mainly through Brownian motion, is able to model and simulate particle concentration dependencies of values of both kinetic rate constants and (non-thermodynamic) particle-solution concentration ratios (Kd) of the radioisotope 234Th (half-life of 24 days), as well as the fact that Th-234 concentrations in different-sized colloids and particles is lowest (rather than highest) in the smaller size fractions rather than what would be expected from the relative specific surface areas and surface sites concentrations (Santschi et al. 2006).
Use of radionuclides as tracers for particle and organic carbon fluxes in the ocean
Owing to their strong particle-reactive nature, many radionuclides, such as thorium (Th), protactinium (Pa), lead (Pb), polonium (Po), and beryllium (Be), are used as tracers for oceanic particle cycling or for soil erosion and transport. The scavenging of these radionuclides can be quantified to delineate the kinetics and pathways of scavenging from the water column by the adsorption on particles and colloids, and subsequent transport the seafloor. Based on the differences in particle reactivity and disequilibria between parent and daughter radionuclides, various biogeochemical and physical processes in the ocean had been explored and quantified (Santschi and Honeyman, 1989). Examples include export fluxes of different chemical species, boundary scavenging, oceanic circulation, and paleoproductivity.
However, most of these applications are generally resting on inferences from correlations of isotopic ratios with bulk particle properties, with little knowledge on the exact carrier phase(s) of the radionuclides, nor their particle-colloid interaction mechanisms (Xu et al. 2011). The observation of enhanced scavenging of 231Pa in the Southern Ocean, which plays an important role in its CO2 uptake, and which has low particle fluxes, has provoked greater interest in the interaction and binding relationships between different radionuclides and particles or colloids in the ocean. Through both field and laboratory studies, considerable evidence has now accumulated, showing radionuclide and compound-specific scavenging and partitioning for different radionuclides, either from an inorganic or from an organic compound prospective.
Importance of molecular level techniques to study the binding of radioactive and stable trace substances to NOM compounds
Particle-reactive radionuclides will only become tracers for any particle cycling process if we know what they are attached to. Thus, it is important to decipher the exact compound and binding site, be they inorganic or organic in nature. For organic compounds, it is then necessary to use molecular-level techniques in conjunction with radiochemical techniques. Using a combination, Santschi was able to show that, at environmental levels, many radionuclides, such as those of Pu, Th, and Po, are found at significant levels in the colloidal, macromolecular fraction in soil and marine particles, tightly bound to specific chelating molecules such as hydroxamate siderophores for Pu (Xu et al. 2015), and Th, Po, Pa (Chuang et al. 2013) in sinking particles. This is surprising since siderophores are produced by micro-organisms to solubilise iron to make it available as a micronutrient.
Such a process is novel and, if confirmed, indicates that the ecosystem as a whole might benefit by transporting iron chelates to the bottom of the ocean, where they can be made accessible to benthic macro- and micro-flora. Another trace element, iodine, which also has a long-lived radioisotope (129I) that is a major risk driver for radionuclide waste disposal, has been identified as covalently bound to specific binding sites in aromatic moieties of macromolecules (Xu et al. 2012), moieties broadly similar to those of thyroxine in the human body.
Macromolecular organic carrier compounds of radioactive and other trace substances
When Santschi’s team studied the sorption of selected natural radionuclides (234Th, 233Pa, 210Po, 210Pb, and 7Be) onto inorganic (pure silica and acid-cleaned diatom frustules) and organic (diatom cells with or without silica frustules) particles in natural seawater, the role of templating biomolecules and EPS associated with the same species of diatom, Phaeodactylum tricornutum, it was revealed that the sorption of all radionuclides was much higher in the presence of templating or surrounding organic matter molecules associated with the diatoms (Chuang et al. 2015). The results strongly indicated that EPS and frustule-embedded biomolecules in diatom cells are responsible for the sorption enhancement rather than the bare silica shell itself. Isoelectric focusing results suggested that each radionuclide binds to specific biopolymeric functional groups, with the most efficient binding sites likely occurring in acid polysaccharides, iron hydroxides and proteins.
Santschi’s team also showed that Coccolithophore-associated organic biopolymers, rather than biogenic calcite, serve as the main carrier phases for Th and Pa, and that Po, Pb and Be are also incorporated into biogenic calcite to substitute for Ca2+ during coccolith formation (Lin et al. 2017), and that different patterns of fractionation between radionuclides occur for coccolithophores versus diatom-dominated systems.
Marine oil snow formation as a vehicle to remove oil spills
Most recently, Santschi became involved in studies of oil removal through ‘marine oil snow’ (MOS) formation and particle aggregation, whereby his team showed, in mesocosms that simulate critical processes in the water, that oil removal from the water is facilitated by microbial EPS production (Quigg et al. 2016), but is greatly decreased when corexit, a surfactant and emulsifier, is used, when only small amounts of ballasting minerals of terrestrial or marine origin are present.
This is relevant as during the Deep Water Horizon oil spill in 2010, which released “live oil” into the Gulf of Mexico (that is, a mixture of oil and natural gas with very high vapour pressures), significant amounts were subsequently found at the bottom of the ocean.
Based on radiocarbon and 13C NMR results, Santschi’s group showed that the presence of dispersants drastically decreased petrocarbon incorporation into sinking MOS, but increased it to dispersed, non-sinking colloidal EPS containing a relatively low mineral content. The formation of MOS and subsequent sinking proceeded in two stages: first a faster removal via terrestrial-derived detritus containing humics; and, subsequently, slower removal via freshly produced material, such as EPS sequestering the oil (Fig. 4) (Xu et al. 2017a).
Santschi’s team also showed that the presence of the water accommodated fraction (WAF) of oil stimulated extracellular polysaccharide production by phytoplankton, whereas corexit, an emulsifier and detergent, promoted protein production by mostly bacteria. Corexit also promoted the association between oil and EPS, thus retaining both more oil and EPS in suspended particles and aggregates. Likely, microbially mediated extracellular polysaccharides are the key compound class that anchors mineral ballast until the aggregates can sink or degrade. Therefore, it is the interactions between corexit and EPS that partially regulate the partitioning of petroleum hydrocarbon between the water column and the sinking MOS, and their subsequent removal from the water column (Xu et al. 2017b).
The importance of ROS-mediated protein polymerisation reactions of EPS for MS and MOS formation
While more evidence has now accumulated for the importance of microbially produced substances in ocean and land, much more remains to be learned about their direct and indirect roles in the cycling and removal of trace substances.
Sunlight is known to inhibit or disrupt the aggregation process of marine colloids via the cleavage of high molecular weight compounds into smaller, less stable fragments. In contrast, what has not been well known is that some biomolecules, such as proteins excreted from bacteria, can form aggregates via random cross-linking due to photo-oxidation, as was recently shown by Santschi’s team. When irradiation experiments were conducted on a well-characterised protein-containing exopolymeric substance from the marine bacterium Sagitulla stellata, after one hour of sunlight irradiation, more aggregates formed, as shown by increased turbidity levels, mass concentrations and flow cytometry, in contrast to a non-protein-containing EPS, or in contrast to dark conditions. Reactive oxygen species, e.g. the hydroxyl radical and peroxide, played critical roles in this photo-oxidation process. This new light-induced aggregation provides new insights into polymer assembly, marine snow formation, and the fate/transport of organic carbon and nitrogen in the ocean (Sun et al. 2017).
Autopoietic systems and their importance in initiating and maintaining aggregation and sedimentation processes
The microbial release of EPS in aquatic environments may be thought of as part of an autopoietic system, that is, a self-sustaining community response. If this is indeed the case, then EPS production is of even greater ecological importance. The processes, mechanisms and participants in the cycling of EPS in the aquatic environment, however, are often poorly known and require further study.
The importance of teamwork in this research
Just like the ecological teamwork of autopoietic systems, successful research in environmental systems requires not only self-replication, i.e. new students that practise the science, but also synergistic teamwork. Thus, much of what is described above was accomplished by members of a closely interacting team, whereby each member takes the lead at times and, at other times, takes a supporting role. Environmental research is increasingly becoming teamwork rather than single investigator-driven research. Alliances are formed within groups, nationally or internationally by academics, or by university-private partnerships. Santschi’s research is an example of that.
PHS acknowledges support from NSF (OCE-0851191), DOE (DE-SC0014152), and GOMRI (doi:10.7266/N7C53JBS)
Alldredge, AL, Gottschalk, CC. 1989. Deep-Sea Res. 36, 159-171
Chuang, C-Y. et al. 2015. J. Geophys. Res. Biogeosci. 120, 9, 1858-1869
Chuang, C-Y. et al. 2013. Mar. Chem. 157, 131-143
Honeyman, BD, and Santschi, PH. 1989. J. Mar. Res. 47/4, 950-995
Lin, P. et al. 2017. J. Geophys. Res.: Biogeosciences, 122, doi:10.1002/2017JG003779
Quigg, A. et al. 2016. L&O Letters, 1, 2016, 3-26
Santschi, PH and Honeyman, BD. 1989. Int. J. Radiat. Appl. Instrum. Part C, Radiat. Phys. Chem. 34, 213-240.
Santschi, PH. et al. 1998. Limnol. Oceanography, 43(5), 896-908
Santschi, PH. et al. 2003. Geophysical Res. Lett. 30(2), Art. No. 1044, dio 10.1029/2002GL016046
Santschi, PH. et al. 2006. Mar. Chem. 100, 250-268
Sun, L. et al. 2017. Chemosphere 181, 675-681
Varela, H.R., et al. 1974. In: Biosystems, Vol.5, No.4, pp.187-196.
Xu, C., et al. 2017b. Environmental Pollution, submitted
Xu, C. et al. 2011. Mar. Chem. 123, 111-126
Xu, C. et al. 2017a. Scientific Report, in revision
Xu, C. et al. 2015. Environ. Sci. Technol. 49, 11458-11467
Xu, C. et al. 2012. Geochim. Cosmochim. Acta. 97, 166-182