1. Introduction
Phosphate-accumulating organisms (PAO) are aerobic or facultatively anaerobic microorganisms removing phosphorus from wastewater due to their ability to accumulate phosphorus inside their cells in amounts significantly exceeding those required for the metabolism of regular bacteria [
1,
2,
3]. According to the current concepts [
4,
5,
6], under anaerobic conditions PAO rapidly consume readily biodegradable organic compounds and store them as intracellular polymers, using the energy derived from hydrolysis of intracellular polyphosphates. Under aerobic conditions or in the presence of an alternative electron acceptor (nitrate or nitrite), they uptake orthophosphate and synthesize intracellular polyphosphates, using the intracellular carbon and energy sources accumulated under anaerobic conditions. Cell biomass growth occurs during the aerobic phase, while during the anaerobic phase the biosynthetic activity of these bacteria is insignificant (apart from accumulation of intracellular storage compounds) [
7]. The metabolic properties of PAO are used in the modern technologies for enhanced biological phosphorus removal (EBPR), which use the cyclic shifts in conditions favoring the growth of these microorganisms. The main PAO used in EBPR belong to the genera
Accumulibacter,
Tetrasphaera, and
Dechloromonas [
6]. Glycogen-accumulating organisms (GAO), the major competitors of PAO, have a similar metabolism; they are unable to cycle phosphates, but under anaerobic conditions they rapidly consume readily biodegradable organic compounds and accumulate organic storage compounds. Most PAO have not been isolated in pure cultures, we have previously suggested that they possess an obligate cyclic metabolism [
8]. However, highly enriched PAO cultures may be obtained under laboratory conditions in the feast-famine anaerobic–aerobic cycling mode.
The amount and composition of readily biodegradable organic matter supplied to the bioreactors or produced in them plays the major part in formation and functioning of the microbial community in EBPR bioreactors. Due to the differences in the spectra of utilized substrates and in their physiological properties, PAO and GAO may occupy different ecological niches and coexist in the EBPR activated sludge [
9,
10]. Most researchers consider the result of competition between GAO and PAO to be responsible for the relative abundance of these groups in the activated sludge consortia and therefore for the efficiency of biological phosphorus removal [
4].
Low-molecular weight volatile fatty acids (VFA), particularly acetate and propionate, are universally considered the preferred carbon sources for
Ca. Accumulibacter and most other PAO. The role of butyrate in
Ca. Accumulibacter metabolism and its potential for BPR remain uncleared, and the research results are contradictory. Kong et al. [
11] did not reveal butyrate consumption by the
Rhodocyclus PAO in the activated sludge of EBPR facilities under aerobic or anaerobic conditions. On the contrary, Levantasi et al. [
12] used the acetate/propionate/butyrate mixture to obtain a
Ca. Accumulibacter-enriched culture, which actively removed phosphorus. Ability to cycle phosphorus upon addition of butyrate was shown for the
Ca. Accumulibacter-enriched culture grown on propionate [
13]. Begum and Batista [
14] investigate the competition between PAO and GAO grown on butyrate as the sole carbon source. While the authors obtained the culture enriched with PAO (
Ca. Accumulibacter) and GAO (
Defluviicoccus), they noted unstable functioning of the community, as well as the low rates of butyrate consumption and phosphate release in the anaerobic phase. Wang et al. [
15] achieved EBPR in a sequencing batch reactor (SBR), in which acetate was gradually replaced by butyrate at 30°C. Although replacement of acetate with butyrate resulted in partial replacement of
Ca. Accumulibacter by
Rhodocyclaceae, phosphorus removal remained stably high.
Ability of
Ca. Accumulibacter and other PAO to use amino acids also remains an open issue. Tian et al. [
16] observed a negative effect of glycine on EBPR and the inability of PAO- and GAO-enriched consortia to use glycine as the carbon source. At the same time, glycine stimulated phosphate release under anaerobic conditions and decreased the efficiency of phosphate uptake under aerobic conditions. Over 10 amino acids were shown to be probably involved to phosphorus cycling by the
Tetrasphaera, with glycine showing the highest efficiency [
17]. Marques et al. [
9] also confirmed the ability of T
etrasphaera to utilize glycine and other amino acids, such as aspartate and glutamate, using a combination of microautoradiography and FISH, while
Ca. Accumulibacter utilized the products of amino acid fermentation, rather than amino acids per se. On the contrary, according to Oyserman et al. [
18], identification of the genes involved an anaerobic glycine utilization by
Ca. Accumulibacter and enhanced phosphate release under anaerobic conditions indicated the ability of
Ca. Accumulibacter to use glycine as a substrate. In-depth research of amino acid metabolism in
Ca. Accumulibacter clade IIF strain SCELSE-1 was carried out by Qiu et al. [
19]. They showed that the culture enriched with
Ca. Accumulibacter clade IIF was able to metabolize 11 out of 20 α-amino acids, with aspartate, glutamate, asparagine, and glutamine being the most efficient substrates for phosphorus removal. The author suggested that aspartate was desaminated and was used to form polyhydroxyalkanoates (PHA) via core carbon metabolic pathways. Glutamate was stores as polyglutamate. Recent studies by Ziliani et al. [
20] showed that, apart from VFA, the culture highly enriched with
Ca. Accumulibacter could also use glucose.
Wastewater injected into EBPR bioreactors contains diverse organic compounds. Investigation of the spectrum of the substrates utilized by PAO and of the pathways of their metabolism is required for further elucidation of the physiological properties of these microorganisms, determination of their position in the microbial consortium, and optimization of existing EBPR technologies and development of new ones. Rapid determination of the ability of PAO to use a given substrates is needed for this purpose.
Oxygen uptake rate is among the most important, rapidly determined indicators of the catabolism rates in aerobic microorganisms. Oxygen uptake rate increase upon addition of a substrate to a starving culture usually indicates the ability to use this substrate as an energy source. However, investigation of the metabolic properties of aerobic PAO concentrates usually on the anaerobic phase, since it is considered that the unique metabolic characteristics of PAO, enabling them to compete for the substrate with other aerobic heterotrophic microorganisms, manifest themselves in the anaerobic phase [
21]. PAO aerobic metabolism has been investigated mainly as a part of the anaerobic/aerobic cycle [
22,
23], and catabolism of extracellular substrates in the presence of electron acceptors got insufficient attention.
The goal of the present work was to determine the spectrum of organic substrates used as energy sources by a Ca. Accumulibacter-enriched culture under aerobic conditions and to reveal their relationship to the phosphorus cycling.
2. Materials and Methods
2.1. The bioreactor for obtaining a Ca. Accumulibacter-enriched culture
In acute experiments, the culture grown for a long term (2 years) in SBR with acetate as the main carbon source was used. The 2-L bioreactor was equipped with a system for aeration and nitrogen supply (Eltochpribor, Russia). Supply and removal of the medium were achieved using peristaltic pumps. Automatic control of gas flows and peristaltic pumps was carried out using a LOGО universal logical module (Siemens, Chengdu, China). An IKA C-MAG MS7 magnetic stirrer (Germany) was used for agitation. Anaerobic conditions were achieved by pumping nitrogen (0.3 L/min) into the bioreactor. The temperature was maintained at 18°C; pH was measured with an Expert-01 meter (Econix, Moscow, Russia). It was not adjusted and varied during the cycle from 7.5 to 8.2.
One cycle of SBR operation (6 h) consisted of five stages: medium supply under anaerobic conditions (30 min); anaerobic phase (2 h 25 min); aerobic phase (2 h 30 min); settling stage (30 min); and supernatant removal (5 min). In each SBR cycle, 08 L of the medium was replaced, with the hydraulic retention time of 15 h. The average solid retention time was maintained at 17.5 days.
The medium contained the following (g/L tap water): CH3COONa·3H2O, 0.67; (NH4)2SO4, 0.139; КH2PO4, 0.109; yeast extract, 0.009; MgSO4·7H2O, 0.150; and trace elements solution, 1 mL. Trace elements solution contained the following (g/L distilled water): Na-EDTA, 10; FeCl3·6H2O, 1.5; H3BO3, 0.15; CuSO4·5H2O, 0.03; MnCl2·4H2O, 0.12; Na2MoO4·2H2O, 0.06; ZnSO4·7H2O, 0.12; KI, 0.18; and CoCl2·H2O, 0.15. To suppress nitrification, thiourea was added to the final concentration of 2.5 mg/L. For pH adjustment, 0.5 M HCl was added to the medium (7.5 mL/L).
2.2. Determination of the oxygen uptake rates
At the end of the aerobic phase, a 800-mL sample of the culture was collected and incubated in a refrigerator for 20 h in order to obtain a culture in the standard physiological state with predominance of endogenous metabolism. The culture was then aerated for 2 h at 20°C, with the concentration of dissolved oxygen maintained at 6‒8 mg O2/L. A 100-mL aliquot was diluted 5-fold with the salt buffer (the medium without acetate and yeast extract) and supplemented with 1 mL of phosphate buffer to the final pH of 7.1. The samples thus obtained were dispensed into glass vials and hermetically sealed without the gas phase. Dynamics of endogenous O2 consumption was then determined, followed by O2 consumption after addition of a carbon and energy source. For all substrates, except for Bacto-tryptone, the initial concentration was 200 mg COD/L. Bacto-tryptone was added to the final concentration of 2 g/L. The tested substrates were VFA salts (acetate, propionate, butyrate, and formate), amino acids (glutamic acid, alanine, serine, threonine, aspartic acid, lysine, and glycine), glucose, ethanol, succinate, pyruvate, glycerol, and tryptone. All reagents were produced by Diaem (Moscow, Russia).
The experiments were carried out at 20±0.2°C. Oxygen concentration was measured electrochemically using an inoLab® Oxi 7310 oxygen meter with a StirrOx G sensor (WTW, Germany). The rate of oxygen uptake was calculated at the linear part of the curve of dissolved O2 concentration decrease with subsequent calculation of the specific oxygen uptake rate. If not stated otherwise, the specific oxygen uptake rate (OUR) was used:
OUR = OURsum – OURend, where
OURsum is the total specific oxygen uptake rate and
OURend is the endogenous specific oxygen uptake rate.
2.3. Determination of dynamics of substrate and phosphate in batch experiments
A 200-mL culture sample from the SBR was transferred into a 500-mL vial. The vial was purged with nitrogen (0.3 L/min) for 10 min in order to establish anaerobic conditions, and the substrate was added to the final concentration of 200 mg COD/L (succinate concentration used was 100 mg COD/L). COD was calculated according to the chemical reaction of complete oxidation of an organic substrate with oxygen. After 2 h, air was supplied to the culture (2 L/min), which resulted in oxygen concentrations of at least 2.0 mg O2/L. The culture to which no substrate was added was used as the control. Dynamics of the concentrations of phosphate and the substrate in the medium was monitored during the experiment. The temperature was maintained at 20±0.2°.
2.4. Molecular techniques
The composition of microbial communities was determined based on analysis of the results of the 16S rRNA gene sequencing, as described previously [
24]. The technique used for metagenome analysis was described previously [
25].
2.5. Analytical techniques
Phosphate ion concentration, TSS, and VSS were determined as described previously [
24]. The concentrations of VFA, succinate, and pyruvate we determined by HPLC using Staier (Akvilon, Russia).
2.6. Statistical data processing
The concentrations of phosphate ions and organic compounds were carried out in triplicate. Average values and standard deviation were determined for each experimental point. The deviation was within the limits of experimental data scattering and did not exceed 5%. It is shown on the graphs as data scattering.