Relating algal-derived extracellular and intracellular dissolved organic nitrogen with nitrogenous disinfection by-product formation.

The dissolved organic nitrogen (DON) pool from algal-derived extracellular and intracellular organic matter (EOM and IOM) comprises proteins, peptides, free amino acids and carbohydrates, of which, proteins can contribute up to 100% of the DON. Previous reports of algal-derived DON character have focused on bulk properties including concentration, molecular weight and hydrophobicity. However, these can be similar between algal species and between the EOM and IOM even when the inherent molecular structures vary. A focus on bulk character presents challenges to the research on algal-derived nitrogenous-disinfection by-product (N-DBP) formation as N-DBP formation is sensitive to the changes in molecular structure. Hence, the main aim of this study was to characterize algal EOM and IOM-derived DON, specifically proteinaceous-DON, using a combination of bulk and molecular characterization techniques to enable a more detailed exploration of the relationship between the character of algal-derived proteins and the N-DBP formation potential. DON from the EOM and IOM of four commonly found algae and cyanobacteria in natural waters were evaluated, namely Chlorella vulgaris, Microcystis aeruginosa, Dolichospermum circinale, and Cylindrospermopsis raciborskii. It was observed that 77-96% of total DON in all EOM and IOM samples was of proteinaceous origin. In the proteins, DON concentrations were highest in the high molecular weight fraction of IOM-derived bulk proteins (0.13-0.75 mg N L-1) and low to medium molecular weight fraction of EOM-derived bulk proteins (0.15-0.63 mg N L-1) in all species. Similar observations were also made via sodium dodecyl sulphate polyacrylamide gel electrophoresis and liquid chromatography-high resolution mass spectrometry. Solid-state 15N nuclear magnetic resonance (NMR) spectroscopy of the EOM and IOM revealed the existence of common aliphatic and heterocyclic N-groups in all samples, including a dominant 2° amide peak. Species dependent variability was also observed in the spectra, particularly in the EOM; e.g. nitro signals were found only in the Cylindrospermopsis raciborskii EOM. Dichloroacetonitrile (DCAN) and N-nitrosamine concentrations from the EOM of the species evaluated in this study were lower than the guideline limits set by regulatory agencies. It is proposed that the dominant 2° amide in all samples decreased N-DBP formation upon chlorination. For chloramination, the presence of nitro groups and aliphatic and heterocyclic N-DBP precursors could cause variable N-nitrosamine formation. Compared to non-algal impacted waters, algae-laden waters are characterised by low organic carbon: organic nitrogen ratios of ∼7-14 and elevated DON and protein concentrations. Hence, relying only on bulk characterization increases the perceived risk of N-DBP formation from algae-laden waters.

Formation of algal-derived nitrogenous disinfection by-products during chlorination and chloramination.

Algal cells and algal organic matter (AOM) are a source of high dissolved organic carbon (DOC) and nitrogen (DON) concentrations. This poses a possible health risk due to their potential to form disinfection by-products (DBPs), some of which may be of health concern, after disinfection. While several studies have focussed on the formation of carbonaceous DBPs from AOM, only a few studies have focussed on the formation of nitrogen containing N-DBPs from AOM. Hence, the main aim of this study was to thoroughly investigate the N-DBP formation potential of the AOM from a species of cyanobacteria commonly found in natural waters, Microcystis aeruginosa. Three haloacetonitriles, two halonitromethanes, two haloacetamides, and eight N-nitrosamines were analysed by gas chromatography-mass spectrometry after chlorination and chloramination of the extracted AOM. To provide further insight into the influence of changing DON character on N-DBP formation potential, the AOM from three other species, Chlorella vulgaris, Dolichospermum circinale and Cylindrospermopsis raciborskii, were also tested. Dichloroacetonitrile (DCAN) was the DBP formed in the highest concentrations for both chlorination and chloramination of bulk AOM from all the species. Furthermore, during chlorination and chloramination, the high molecular weight fraction (>1 kDa) of AOM from M. aeruginosa had a greater DCAN formation potential (normalised to DOC or DON) than the AOM in the low molecular weight fraction (<1 kDa) of M. aeruginosa, regardless of growth stage. N-Nitrosamine formation from the bulk AOM of all species occurred only after chloramination. The molar concentration of N-nitrosodimethylamine (NDMA) was lower than the other N-nitrosamines detected. However, NDMA formation increased with culture age for all four species, in contrast to most other N-nitrosamines whose formation remained consistent or decreased with culture age. Overall, algal growth could result in elevated concentrations of N-DBPs due to the increasing concentrations of high molecular weight algal DON in the AOM. It is suggested that the AOM comprises precursors containing long C-chain amine (R1-NH-R2) or cyclic N-containing amine structures. Comparisons to previously measured N-DBP concentrations in drinking water suggest that the AOM from the algae and cyanobacteria examined in this study are not likely to be a major source of precursors for either DCAN or NDMA in real waters. However, AOM may present a major precursor source for other N-nitrosamines.

An evaluation of measurement techniques for algal-derived organic nitrogen.

Algal-derived organic matter (AOM) from algal blooms in water supply systems contains dissolved organic nitrogen (DON) and dissolved organic carbon (DOC) among other constituents. The DON and DOC are disinfection by-product (DBP) precursor compounds, and must be well characterised to facilitate effective removal, thus minimising DBP formation during disinfection. While DOC character has been studied extensively, DON analysis suffers from inaccuracies due to sample pre-treatment and instrument sensitivities. A liquid chromatography method that combines size exclusion chromatography with highly sensitive organic carbon and nitrogen detectors (LC-OCND) has been widely adopted for DOC analysis; however, its potential for application for DON charactersation has been suggested as a viable alternative to existing DON characterisation techniquesnot been assessed despite its potential. Hence, the aim was to compare the effectiveness of conventional total dissolved N-dissolved inorganic N (TN-DIN), and LC-OCND methods for analysing DON in AOM. A suite of N-containing model compounds representative of DON and AOM extracted from Chlorella vulgaris CS-42/7 and Microcystis aeruginosa CS-555/1 were used to evaluate the techniques. The DON of both model compounds and AOM was first analysed using the conventional method and, then, via LC-OCND. It was observed that LC-OCND had a better precision for DON when TN contained more DIN. LC-OCND provided direct quantitative measurements for bulk and fractionated DON and DIN, with little interference caused by DIN. Additionally, LC-OCND provided information on MW distribution and protein content of the AOM. For example, LC-OCND results showed that M. aeruginosa AOM contained more HMW material than C. vulgaris AOM. However, as LC-OCND uses UV oxidation, it could not completely oxidise complex aromatic structures, and thus had a lower recovery for HMW model compounds and algal DON in comparison to the conventional method that used high temperature catalytic oxidation. Overall, it is advised that a combination of LC-OCND and TN analysis be used to provide a more detailed characterisation of N-containing AOM and other similar HMW aquatic NOM samples.

Impact of bromide on halogen incorporation into organic moieties in chlorinated drinking water treatment and distribution systems.

The impact of elevated bromide concentrations (399 to 750 µg/L) on the formation of halogenated disinfection by-products (DBPs), namely trihalomethanes, haloacetic acids, haloacetonitriles, and adsorbable organic halogen (AOX), in two drinking water systems was investigated. Bromine was the main halogen incorporated into all of the DBP classes and into organic carbon, even though chlorine was present in large excess to maintain a disinfectant residual. Due to the higher reactivity of bromine compared to chlorine, brominated DBPs were rapidly formed, followed by a slower increase in chlorinated DBPs. Higher bromine substitution and incorporation factors for individual DBP classes were observed for the chlorinated water from the groundwater source (lower concentration of dissolved organic carbon (DOC)), which contained a higher concentration of bromide, than for the surface water source (higher DOC). The molar distribution of adsorbable organic bromine to chlorine (AOBr/AOCl) for AOX in the groundwater distribution system was 1.5:1 and almost 1:1 for the surface water system. The measured (regulated) DBPs only accounted for 16 to 33% of the total organic halogen, demonstrating that AOX measurements are essential to provide a full understanding of the formation of halogenated DBPs in drinking waters. In addition, the study demonstrated that a significant proportion (up to 94%) of the bromide in source waters can be converted AOBr. An evaluation of AOBr and AOCl through a second groundwater treatment plant that uses conventional treatment processes for DOC removal produced 70% of AOX as AOBr, with 69% of the initial source water bromide converted to AOBr. Exposure to organobromine compounds is suspected to result in greater adverse health consequences than their chlorinated analogues. Therefore, this study highlights the need for improved methods to selectively reduce the bromide content in source waters.

Interactions between model triacylglycerol-rich lipoproteins and high-density lipoproteins in rat, rabbit and man.

There are inverse relationships between HDL cholesterol and plasma triacylglycerol concentrations in normal and in hypertriglyceridemic individuals. To investigate the interactions between triacylglycerol-rich lipid particles and HDL, a lipid emulsion model of the triacylglycerol-rich lipoproteins was prepared. When emulsion particles were incubated with rat high-density lipoproteins (HDL) in the presence of lipid transfer activity (d greater than 1.21 g/ml fractions) from rabbit or human plasma there was a rapid bi-directional exchange of cholesteryl oleate (CO) and phospholipid (PL) labels between lighter and heavier fractions of HDL and emulsion particles. The transfers of CO and PL labels between both light and heavy fractions of HDL and the emulsion particles were increased with increasing amounts of emulsion added to the incubations. Incubation with the d greater than 1.21 g/ml fraction from rat plasma resulted in only a small exchange of CO whereas PL exchange was similar to rabbit and human plasma. Retinyl palmitate label was not transferred from emulsion particles to the HDL fractions even in the presence of lipid transfer activity from rabbit or human plasma. The present study shows that the transfer protein-mediated exchanges of surface and core lipids between HDL and the triacylglycerol-rich lipoproteins are affected by the quantity of triacylglycerol-rich particles in the system. This mechanism may contribute to the inverse relationships between plasma triacylglycerol concentrations and HDL concentrations in normal and hypertriglyceridemic individuals.

The effect of added monoacylglycerols on the removal from plasma of chylomicron-like emulsions injected intravenously in rats.

Lipid emulsions were prepared with compositions similar to the triacylglycerol-rich plasma lipoproteins, but also incorporating added small amounts of monoacylglycerols. Control emulsions without monoacylglycerol were metabolized similarly to natural chylomicrons or very-low-density lipoproteins when injected intravenously in rats. The emulsion triacylglycerols and cholesteryl esters were both removed rapidly from the bloodstream, with the removal rates of triacylglycerols faster than those of cholesteryl esters. Much of the removed cholesteryl ester was found in the liver, but only a small fraction of the triacylglycerol, consistent with hepatic uptake of the triacylglycerol-depleted remnants of the injected emulsion. Emulsions incorporating added monooleoylglycerol or stearic acid were metabolized similarly. Added 1- or 2-monostearoylglycerol had no effect on triacylglycerol removal from plasma, but the removal rate of cholesteryl esters was decreased and less cholesteryl ester was found in the liver. These effects are similar to those recently described when emulsions and chylomicrons contained triacylglycerols with a saturated acyl chain at the glycerol 2-position, suggesting that saturated monoacylglycerol produced by the action of lipoprotein lipase may cause triacylglycerol-depleted remnant particles to remain in the plasma instead of being rapidly taken up by the liver.

Regulation of the metabolism of lipid emulsion model lipoproteins by a saturated acyl chain at the 2-position of triacylglycerol.

Lipid emulsion particles were prepared by sonicating four different lipid mixtures (triacylglycerol (TAG), 70%; phospholipid, 25%; cholesteryl oleate (CO), 3%; and free cholesterol, 2%), then purified by density gradient ultracentrifugation. For three test mixtures, the TAG contained 50, 75, or 100% 1,3-dioleyl-2-stearylglycerol (OSO) with the remainder being triolein (OOO); 100% triolein in the lipid mixture was used as the control. After intravenous injection of the lipid particles into unanesthetized rats, removal of radioactive TAG fatty acid and CO from plasma was measured for 30 min, then liver and spleen uptakes were measured. When emulsions contained 75% or 100% OSO as TAG, the plasma removal rates of CO were, respectively, 60% or 30% of the rate when the TAG was 100% triolein; smaller recoveries of CO were found in the liver. The clearances of TAG fatty acid did not differ significantly and the recoveries of TAG fatty acid in the organs were not affected by the type of emulsion injected. Remnant particles were derived from donor rats in which uptake was blocked by exclusion of liver and other viscera from the circulation before injection of 100% OOO and 100% OSO emulsions. When injected into recipient intact rats, the removal of remnants from plasma was slower for remnants derived 15 min after injection of 100% OSO emulsions than from 100% OOO emulsions, showing that the slower removal of emulsion CO was due to slower remnant uptake from the plasma with OSO emulsions.(ABSTRACT TRUNCATED AT 250 WORDS)