Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Microbial and biochemical properties of the biological excess phosphate removal process Mah, Terrance Jock 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1991_A6_7 M38.pdf [ 6.12MB ]
Metadata
JSON: 831-1.0098610.json
JSON-LD: 831-1.0098610-ld.json
RDF/XML (Pretty): 831-1.0098610-rdf.xml
RDF/JSON: 831-1.0098610-rdf.json
Turtle: 831-1.0098610-turtle.txt
N-Triples: 831-1.0098610-rdf-ntriples.txt
Original Record: 831-1.0098610-source.json
Full Text
831-1.0098610-fulltext.txt
Citation
831-1.0098610.ris

Full Text

MICROBIAL AND BIOCHEMICAL PROPERTIES OF THE BIOLOGICAL EXCESS PHOSPHATE REMOVAL PROCESS by TERRANCE JOCK MAH B.Sc, The University of British Columbia, 1989 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1991 ®Terrance Jock Mah, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of \\\&^&X=i^o&H The University of British Columbia Vancouver, Canada Date (W)G^ 2^°|; [3 *U DE-6 (2/88) ABSTRACT The reasearch objectives of this thesis are to determine the behavior of specific cellular components in sludge, in response to varying environmental conditions, and determine the role of the glyoxylate cycle in anaerobic PHA storage. These objectives were addressed by idnetifying assays to determine the immediate biochemical state of the sludge biomass and applying these assay to determine typical sludge behavior then.testing a proposed biochemical model. Radioactive labelling of cells showed aerobic phosphate uptake results in storage of two major polyphosphate types, long chain granular or long chain soluble polyphosphates, that can account for up to 85 % of newly synthesized intracellular phosphate-containing compounds. Both of the polyphosphate types were greater than eight ortho-phosphate units long and susceptible to acid hydrolysis. Anaerobically, long chain soluble polyphosphates are degraded and released during polyhydroxyalkanoate (PHA) storage. PHA storage was most affected by carbon availability and not strongly influenced by the ORP or the NADH/NAD ratio. The use of a fluorometric probe for detecting intracellular NADH levels revealed several sludge responses that could be used for optimization of process control. Testing of the proposed biochemical model which accounted for two principal storage products, polyhydroxybutyric acid (PHB) and polyhydroxy valeric acid (PHV), provided evidence to suggest the glyoxylate cycle plays a central role in anaerobic PHA storage, by providing the necessary reducing power required for the storage reaction, in the normal operating U.B.C. pilot plant system. Tesing involving nutrient/inhibitor combinations demonstrated that anaerobic PHA storage is not committed exclusively to the glyoxylate cycle for a source of reducing power, since alternative sources of reducing power could also supply reducing power for the PHA storage reaction under different conditions. ii TABLE OF CONTENTS page ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES v LIST OF FIGURES vi LIST OF ABBREVIATIONS ix ACKNOWLEDGEMENTS x 1. INTRODUCTION 1 1.1 Eutophication and Phosphate Removal 1 1.2. Biological Excess Phosphate Removal and Biochemical Models 2 1.3. Thesis Objective 5 2. MATERIALS AND METHODS 2.1. BEPR Sludge Samples 6 2.2. Phosphate Accumulation, Release, and Fractionation 2.2.1. Phosphate accumulation and release 7 2.2.2. Fractionation of phosphate-containing compounds 8 2.3. Polyhydroxyalkanoate (PHA) Determination 8 2.4. Volatile Fatty Acid (VFA) Analysis 9 2.5. NADH Measurement 9 2.6. Bench Scale Batch Experiments 10 2.7. Biochemical Model Testing - - 11 3. RESULTS 3.1. Phosphate Behavior 3.1.1. Phosphate accumulation and release 12 3.1.2. Phosphate incorporation 12 3.1.3. Thin layer chromatographic analysis of polyphosphates 15 3.2. PHA Determination in BEPR Sludge 3.2.1. PHA quantification 22 3.2.2. PHA recovery and reproducibility 22 3.3. VFA Analysis 24 3.4. Bench Scale Batch Testing 24 3.4.1. Characteristic NADH and ORP responses ' 25 3.4.2. General PHA, VFA, and soluble phosphate behavior 27 3.4.3. General sludge patterns 30 3.4.4. BEPR sludge behavior during nitrate additions 39 3.4.5. Relationships of PHA, VFA, NADH, and phosphate 39 iii 3.5. Biochemical Model Testing 43 3.5.1. Acetate addition 46 3.5.2. Citrate additions 51 3.5.3. Succinate additions 53 3.5.4. Nutrient/inhibitor combinations 59 4. DISCUSSION 65 4.1. Bench Scale Batch Experiments 4.1.1. Phosphate behavior 65 4.1.2. PHA response 71 4.1.3. NADH response 75 4.2. Biochemical Model Testing and Evaluation 82 4.2.1. Glyoxylate cycle function 83 4.3. General Conclusions from Bench Scale and Biochemical Tests 86 4.4. Limitations of the Data 87 5. CONCLUSION 89 REFERENCES 90 iv LIST OF TABLES page Table I. Summary of NADH, ORP, PHA, and soluble phosphate behavior during "beginning" phase of experimental runs, without aeration or added carbon. 31 Table II. Summary of NADH, ORP, PHA, and soluble phosphate behavior during carbon addition prior to aeration. 32 Table III. Summary of NADH, ORP, PHA, and soluble phosphate behavior during aerobic carbon addition. 32 Table IV. Summary of NADH, ORP, PHA, and soluble phosphate behavior during aeration, without added carbon. 33 Table V. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anoxic carbon addition following aeration. 34 Table VI. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anoxic conditions following aeration, without added carbon. 35 Table VII. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anaerobic conditions, without added carbon. 36 Table Vlll.Summary of NADH, ORP, PHA, and soluble phosphate behavior during anaerobic carbon addition. 37 Table IX. Summary of general behavior of NADH, ORP, PHA and soluble phosphate under various environmental conditions. . 38 Table X. Comparison of net PHA accumulation and input carbon as millimoles of carbon per litre. 42 v LIST OF FIGURES pag Figure 1.1. Diagram for behavior proposed by biochemical models under anaerobic conditions. Figure 2.1. Process configuration for the University of British Columbia pilot plant. Figure 3.1. Profiles of phosphate accumulation and release. Figure 3.2. Typical cellular phosphate composition profile of BEPR sludge during phosphate uptake. Figure 3.3. Phosphate and polyphosphate accumulation profiles, without anaerobic phosphate release. Figure 3.4. Phosphate and polyphosphate accumulation profiles, with anaerobic phosphate release. Figure 3.5. Autoradiograph of thin layer chromatograph of 32p ortho-phosphoric acid and 7-32p-ATP standards. Figure 3.6. Autoradiograph of thin layer chromatograph of isolated polyphosphates. Figure 3.7. Autoradiograph of thin layer chromatograph of long chain soluble polyphosphates heated in 1 N HC1 for various times. Figure 3.8. Autoradiograph of thin layer chromatograph of long chain granular polyphosphates heated in 1 N HC1 for various times. Figure 3.9. Chromatograph of methyl ester derivatives extracted from BEPR sludge. Figure 3.10. Characteristic profile and features of NADH during typical batch run and extended unaerated run. Figure 3.11. Characteristic profile of NADH, ORP, PHA, soluble phosphate, VFA, and dissolved oxygen for a typical batch experiment. Figure 3.12. Characteristic profile of NADH, ORP, PHA, soluble phosphate, VFA, and dissolved oxygen for a typical batch run with anaerobic nutrient addition. vi Figure 3.13. NADH and ORP response during anaerobic nitrate additions. 40 Figure 3.14. Relationship between nitrate addition and length of NADH depression. 40 Figure 3.15. Experimental data from a series of anaerobic nitrate and acetate additions. 41 Figure 3.16. Proposed .biochemical model for PHA storage under anaerobic conditions. 45 Figure 3.17. Experimental data from experiment with anaerobic acetate addition. 47 Figure 3.18. Experimental data from experiment with anaerobic acetate and malonate addition. 49 Figure 3.19. Experimental data from experiment with anaerobic acetate and pyruvate addition. 50 Figure 3.20. Experimental data from experiment with anaerobic citrate addition. 52 Figure 3.21. Experimental data from experiment with anaerobic citrate, malonate, and acetate addition. 54 Figure 3.22. Experimental data from experiment with anaerobic succinate addition. 55 Figure 3.23. Experimental data from experiment with anaerobic succinate and acetate addition. 57 Figure 3.24. Experimental data from experiment with anaerobic succinate and malonate addition. 58 Figure 3.25. Experimental data from experiment with anaerobic acetate and a-ketoglutarate addition. 60 Figure 3.26. Experimental data from experiment with anaerobic acetate and monofluoroacetate addition. 62 Figure 3.27. Experimental data from experiment with sequential anaerobic addition of acetate and monofluoroacetate. 63 Figure 3.28. Experimental data from experiment with anaerobic addition of acetate, monofluoroacetate, and pyruvate. 64 vii Figure 4.1. Schematic pathways of PHB and polyphosphate metabolism. Figure 4.2. Schematic diagram of NADH/NAD cycling reactions. viii LIST OF ABBREVIATIONS ADP -adenine diphosphate AMP -adenine monophosphate ATP -adenine triphosphate BEPR -biological excess phosphate removal D.O. -dissolved oxygen ED -Entner-Duodoroff EMP -Embden-Meyerhof-Parnas NAD -nicotinamide adenine dinucleotide (oxidized form) NADH -nicotinamide adenine dinucleotide (reduced form) NFU -normalized fluorescence units nm -nanometers LCG Poly-P -long chain granular polyphosphate LCS Poly-P -long chain soluble polyphosphate ORP -oxidation-reduction potential PHB -polyhydroxybutyric acid PHV -polyhydroxyvaleric acid PHA -polyhydroxyalkanoate(s) P04 -phosphate SCAS Poly-P -short chain acid soluble polyphosphate SCFA -short chain fatty acid(s) TCA -tricarboxylic acid cycle U.B.C. -University of British Columbia VFA -volatile fatty acid(s) Note: In this thesis "normal operating system" refers to the normal operating configuration of the University of British Columbia pilot plant, in which acetate is fed continuously to the anaerobic zone. ix ACKNOWLEDGEMENT I acknowledge the following persons for their support and guidance throughout my research and study, with gratitude and respect: Dr. William D. Ramey for his supervision, patience, inspiration, and unconditional support throughout my studies. Dr. William K. Oldham, Professor and Head of the Department of Civil Engineering, for his open-minded support and encouragement. Dr. George B. Spiegelman for chairing my committee and offering valuable advice and information. FrederickA. Koch, research associate in the Environmental Engineering Group, for overseeing the operation of the University of British Columbia pilot plant, and for stimulating discussion. Dr. William B. Armiger, and Biochem Technology Inc. for providing some equipment to operate the FluroMeasure system, as well as their technical support and collaboration on this project. Angus Chu, fellow graduate student, for discussion and assistance for photographing autoradiographs. Susan Liptack and Paula Parkinson, from the Environmental Engineering Laboratory, for their invaluable technical support and advice. This work was funded, in part, by a strategic grant from the Natural Science and Engineering Research Council of Canada. x 1. INTRODUCTION 1.1. Eutrophication and Phosphate Removal The eutrophication of natural water systems represents a major surface water quality problem. Its cause has been attributed to an excessive supply of nutrients which support the excessive growth of aquatic plants and algae. Phosphate has been recognized as the nutrient responsible for the excessive eutrophication of natural waters, and eutrophication control has concentrated on controlling and limiting phosphate levels in these natural water systems (Manahan, 1984). Limitation of phosphate loading to these systems has been attempted by removing phosphate from wastewater streams discharged to the aquatic environment. Two principal methods for removal of phosphate from wastewater involve physicochemical and biological processes. Physicochemical phosphate removal involves precipitation of phosphate as the metallic complexes of calcium, aluminum, and iron (Wiechers, 1987), and requires the addition of expensive metallic salts and subsequent disposal of chemically contaminated sludge. Biological phosphate removal involves the accumulation and storage of phosphate by sludge biomass in excess of metabolic requirements and was first observed by Srinath et al. (1959). Biological excess phosphate removal (BEPR) is more cost effective than chemical precipitation (Canviro et al. ,1986; Morrison, 1988) and does not suffer from the chemical sludge disposal problems. These observations stimulated intensive research and demonstrated that enhanced phosphate removal appears to result from microbial action, and that optimization of the process would require a thorough understanding of the ecology and physiology of the microbial system (Toerien et al., 1990). 1 1.2. Biological Excess Phosphate Removal and Biochemical Models Biological excess phosphate removal in wastewater can be accomplished by modification of conventional activated sludge processes. In a continuous flow BEPR process, wastewater is fed into an anaerobic selector then an oxic reactor. BEPR systems are characterized by an increase in soluble phosphate in the anaerobic selector resulting from phosphate release by the biomass, followed by excess polyphosphate accumulation by the microorganisms in the aerobic reactor. Bioavailable substrates in the anaerobic selector, such as short chain fatty acids (SCFA), are accumulated and stored by the biomass as polyhydroxyalkanoates (PHA) through reduction and condensation reactions. These storage reactions are believed to afford certain organisms the advantage of hoarding the available energy and carbon sources before other organisms can sequester them. Over the years, several biochemical models have been developed to account for the characteristic behavior of BEPR sludge. Currently, there are two models, the Mino (Mino, Arun, Tsuzuki, and Matsuo, 1987) and Comeau/Wentzel (Comeau, Hall, Hancock, and Oldham, 1985; Wentzel, Lotter, Loewenthal and Marais, 1986) models, that describe the specific microbial processes that may be responsible for the BEPR process. A detailed review of these models is provided by Wentzel et al. (1991). Both models recognize the importance of anaerobic/oxic recycling and the role of short chain fatty acids in the anaerobic selector, but differ with respect to the source of reducing equivalents required for anaerobic PHA storage. The Mino model (Mino et al., 1987), developed using artificially fed sludge, proposes reducing equivalents are generated through carbohydrate catabolism via the Embden-Meyerhof-Panas (EMP) or Entner-Doudoroff (ED) pathways in anaerobic conditions. In most activated sludge systems, however, carbohydrate sources are minimal. In the Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) model, reducing equivalents for PHA storage are generated by metabolism of the more abundant short chain fatty acids via the glyoxylate cycle, and the storage 2 reaction is regulated by the ATP/ADP and NADH/NAD ratios. Both models describe the behavior of BEPR sludge based on the respective observations, and propose explanations for the role of stored carbon and phosphate compounds. Stored polyphosphates are believed to supply an energy source for the activation of acetyl CoA and maintenance of the membrane potential. Stored carbon, in the form of PHB, is believed to provide carbon for generation of energy, cell mass, and in the Mino model (Mino et al., 1987), carbohydrate synthesis and storage. Figure 1.1 provides diagrams outlining these models in anaerobic conditions. These modelsprovide some insight into the specific mechanisms that may be responsible for the BEPR, and describe the numerous observations, and speculation regarding the role of the NADH/NAD ratio as a regulatory mechanism further attempts to advance the understanding of the process beyond the experimental observations. These models, however, were developed primarily from general observations and assumptions regarding the regulation and behavior of biochemical pathways and cellular components, and significant parts have no direct evidence or experimental confirmation. The models may be criticized for some of their experimental techniques and limited explanations. Some of the techniques used to support these models lack the specificity and ability to test or detect the immediate biochemical state of the dynamic microbial system. The models do recognize the accumulation and storage of PHA accounting for the appearance of polyhydroxybutyric acid (PHB), but largely ignore the role of polyhydroxyvaleric acid (PHV) or the relative behavior of these two storage compounds. Similarly, the regulatory role of the NADH/NAD is speculative and based on general observation made on pure culture systems growing in quite different circumstances. Development and application of experimental methods to allow direct testing and observation of the immediate behavior and relationships of specific biochemical pathways, control mechanisms, and cellular components implicated by these models, will provide better understanding and optimization of the BEPR process. 3 A . CH COOH-outaldv ln«ide call c«ll CH COOH PHB Figure 1.1. Diagram for behavior proposed by biochemical models under anaerobic conditions (Wenzeletal., 1991). (A) Comeau/Wenzelmodel (Comeauetal., 1985; Wentzeletal., 1986), (B) Mino nodel (Mino et al., 1987). 4 1.3. Thesis Objective The general objective of this research was to gain better understanding of the microbial processes involved in the biological excess phosphate removal process. The specific objectives were to test the behavior of key cellular components of phosphate-containing compounds, PHA and NADH in varying environmental conditions, by differentiating between newly synthesized and existing phosphate-containing compounds, assessing the role and fate of these compounds, evaluating the specific role of NADH, and investigating the relative behaviors of PHB and PHV, then using this information to test the role of the glyoxylate cycle in anaerobic PHA storage. These objectives have been broken down into three specific areas of investigation. One area deals with the development and refinement of assays to allow investigation of the immediate biochemical state of key cellular components, another area deals with the determination of the behavior of these components in BEPR sludge systems, and the last area deals with specific biochemical testing. To test aspects of the Comeau/Wentzel model (Comeau et al., 1985; Wentzel et al., 1986) and resolve the importance of glyoxylate cycle, a biochemical model was derived to allow prediction and testing of specific results generated by the use of specific pathways following the addition of specific nutrients and inhibitors to a bench scale BEPR sludge system. 5 2. MATERIALS AND METHODS 2.1. BEPR Sludge Samples Samples of BEPR sludge were obtained from the "A" side of the University of British Columbia pilot plant. This plant represents a modified UCT (University of Cape Town, South Africa) process and is diagrammed in Figure 2.1. One or 12 litre samples were withdrawn from the anaerobic selector and transported in sealed containers to the laboratory for experimentation. Samples were collected around the same time each experiment day, and used within one hour of collection. Figure 2.1. Process configuration of the University of British Columbia pilot plant. 6 2.2. Phosphate Accumulation, Release, and Fractionation 2.2.1. Phosphate accumulation and release The ability of BEPR sludge to accumulate and release phosphate was determined by monitoring the uptake and release of radioactive ortho-phosphate. Experiments were performed using either small 3 ml or large 4 L samples of BEPR sludge. For 3 ml sample studies, 1L of sludge was collected and dispensed into 16X150 mm tubes and either left to stand undisturbed for 2 hours or used immediately. For four litre bench scale experiments, 12 L of sludge was collected, concentrated to approximately 1.5 times the original solids concentration by decanting supernatant from the settled sludge, then transferred into a cylindrical, opaque plexiglass reactor where it was mechanically stirred. Samples of sludge were collected and dried for determination of total solids levels according to Standard Methods (A.P.H. A., 1985). To follow phosphate uptake and release, a mixture of either 0.5 /xCi or 0.7 mCi of radioactive ortho-phosphate (-^ P) (ICN Biomedicals Canada Ltd., St. Laurent, Quebec) with the desired level of non-radioactive phosphate carrier (KH2PO4) was added to respective 3 ml or 4 L sludge samples. Small samples were aerated by agitating tubes on a shaking platform. Large sludge samples were aerated by pumping air through a diffuser located near the bottom of the reactor. Samples of 150 fil or 1 ml were withdrawn at various times and transferred to 1.5 ml microfuge tubes. Samples were then microfuged for 3 minutes and 100 /A of supernatant was sampled and placed on a filter disc. The cell pellets were frozen and the filter discs dried for determination of soluble phosphate concentration. Radioactivity on the filters was determined using a Beckman LS6000IC scintillation counter. Specific activity was determined and radioactive counts used to determine soluble ortho-phosphate concentrations remaining in the supernatants. 7 2.2.2. Fractionation of phosphate-containing compounds To examine the fate of incorporated phosphate, molecular fractionation of the radioactively labelled frozen cell pellets was performed by the method of Clark et al. (1986). This techniques allows isolation of intact chains of polyphosphate, determination of short chain acid soluble polyphosphate, and estimation of the other intracellular phosphate-containing compounds. The use of 32p labelled samples allowed differentiation between newly synthesized and existing phosphate-containing materials and convenient quantitation of phosphate in each fraction. In situations where polyphosphates could not be separated from nucleotides by precipitation with magnesium, activated charcoal was used. Samples of polyphosphate classified by this method as short chain acid soluble, long chain soluble, and long chain granular were analyzed using a thin layer chromatographic solvent system described by Seiler (1969). Separation of polyphosphates with chain lengths of < 8 will migrate differently in the thin layer system, while those with chain lengths > 8 will remain at the origin. Cellulose thin layer plates were obtained from Sigma (St. Louis, MO.) and were spotted with samples of isolated polyphosphates, or long chain polyphosphates that had been heated at 100°C in the presence of 1N HC1 for various times. Heating in the presence of acid will hydrolyze the polyphosphates to shorter chains or individual ortho-phosphate units (Fuhs and Chen, 1975). Chromatography was carried out by ascending development in a closed glass chamber at room temperature and visualized by autoradiography of the dried chromatographs. Migration of the samples was compared to standards of 32p ortho-phosphate and 7^2p-ATP (NEN Research Products, Boston, MA.). 2.3. Polyhydroxyalkanoate (PHA) Determination To evaluate the extent of carbon storage in samples of BEPR sludge, PHA was isolated and analyzed. Twenty millilitres of sludge were collected in 30 ml polyallomer tubes containing 5 ml of commercial hypochlorite solution (Western Family Bleach). Hypochlorite will inhibit 8 any enzymatic activity and should enhance recovery of both polymers, which are insoluble in hypochlorite solution (Poindexter and Eley, 1983). Samples were then centrifuged and the supernatant removed. The remaining sludge pellet was frozen in an ethanol-dry ice bath. Polyhydroxyalkanoates, specifically polyhydroxybutyric acid and polyhydroxyvaleric acid, were then assayed by the technique of Comeau, et al.(1988), and quantified using standards of ot-hydroxybutyric acid sodium salt, j3-hydroxybutyric acid sodium salt, and a-hydroxyvaleric acid sodium salt (Sigma Chemical Co., St. Louis, MO.). A Perkin-Elmer Sigma 3B gas chromatograph equipped with a flame ionization detector and a Hewlett-Packard 3380A integrator were used for all gas chromatographic analyses. 2.4. Volatile Fatty Acid (VFA) Analysis A gas chromatographic technique was used to determine the concentration of volatile fatty acids in the sludge. Samples were taken, centrifuged, and a 500 fil of the supernatant was removed and acidified by adding 5 JKI of 10 % H3PO4. A I [A sample of the acidified supernatant was removed and injected into a column packed with 60/80 Carbopack C/0.3% Carbowax (Supelco Canada Ltd., Oakville, Ontario) for analysis of CI to C5 compounds. The gas chromatograph was operated isothermally with an oven temperature of 120°C, injection and detector port temperatures of 200°C, and helium carrier flow of 15 ml/min. 2.5. NADH Measurements Because of the dynamic nature of NADH, methods for rapid determination of NADH had to be identified to assess the behavior and role of NADH in the BEPR process. However, both the rapid extraction method (Wimpenney and Firth, 1972) and enzyme-linked fluorometric assay (Cartier, 1968) showed significant interference from components in the sludge. 9 Therefore, the Fluromeasure system and Fermac (Level 1) software (Biochem Technology Ltd, King of Prussia, PA.) were used to assay intracellular NADH levels. This system employs a fluorometric probe that can be attached to a reactor to irradiate cells with light of 340 nm and measures the resulting fluorescence at 460 nm. NADH produces fluorescence with a peak at 460 nm, when irradiated with light at 340 nm, allowing a nondestructive estimation of the intracellular NADH levels (Duysens and Amesz, 1957). The Fluromeasure system interfaces to a computer which collects measurements every 12 seconds, processes and records the signal on a real time basis. NADH fluorescence is reported as relative NFU, or normalized fluorescence units. 2.6. Bench Scale Batch Experiments Bench scale batch experiments were conducted using 4 L of anaerobic BEPR sludge to assess the behavior of various components in response to varying environmental conditions. Samples were placed into an opaque, cylindrical plexiglass reactor. Sludge was mechanically stirred and aeration accomplished by pumping of air through a diffuser located near the bottom of the reactor. Various nutrients were added to determine their effects on the individual components in the sludge and were added through a port located near the top of the reactor. A sampling port was located near the bottom of the reactor for sample removal. A dissolved oxygen (DO) probe (YSI Ltd., Yellow Springs, OH.) and an oxidation reduction potential (ORP) probe (Broadley-James Corp., Santa Ana, CA.) continuously monitored DO and ORP. The dissolved oxygen probe is not accurate below dissolved oxygen concentrations below 1 mg/L. The ORP probe was calibrated each experiment day using buffered quinhydrone standards. Data was recorded automatically using the Fermac software monitoring package. NADH was continu-ously measured and samples periodically withdrawn for analysis of PHA, soluble phosphate levels, intracellular phosphates, and soluble VFA. Data was compiled for determination of sludge behavior. 10 2.7. Biochemical Model Testing A biochemical model was developed to allow testing of specific areas of metabolism. The model allowed predictions regarding the behavior of cellular components to be made and experimentally tested. Testing was conducted using the bench scale reactor and adding various nutrients and inhibitors that should affect the cellular components in predictable ways. Nutrients and inhibitors were added and the sludge monitored for NADH levels, DO, ORP, and assayed for PHA, soluble phosphate and VFA levels, and intracellular phosphates. Data was compiled and compared to predictions to evaluate the role of pathways proposed by Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) and this biochemical model. 11 3. RESULTS 3.1 Phosphate Behavior 3.1.1. Phosphate accumulation and release To determine the ability of BEPR sludge to accumulate and release phosphate, the 32p assays were performed on 3 ml samples in test tubes and 4 L batch reactor samples. Over ninety samples were tested for phosphate uptake and release. Figure 3.1 shows typical patterns of phosphate uptake and release for these experiments. The majority of accumulation occurred within the first ninety minutes of aeration and ranged from 10 to 30 mg/L for the experiments. Phosphate uptake could be inhibited by the addition of 1 mM cyanide, suggesting the accumulation is an energy dependent process. Uptake typically occurred during oxic conditions, but could continue after mechanical aeration stopped (Figure 3.1.C). Phosphate release was only observed in the larger batch samples and could be induced by anaerobic nutrient additions (Figure 3.1. C). The nature and factors affecting release will be discussed later. The error associated with these measurements was estimated to be 10 % based on the error associated with measuring radioactivity. 3.1.2. Phosphorus incorporation Molecular fractionation was performed on frozen radioactively labelled cell pellets to determine the fate of incorporated phosphate. Figure 3.2 shows a typical intracellular phosphate composition profile of BEPR sludge over time. The total amount of phosphate incorporated into the sludge increased with time. Initially each fraction incorporated some of the radioactive label, but as time passed, incorporation into polyphosphate fractions was favored. The polyphosphate fraction classified as short chain acid soluble tended to be relatively constant with time, while the other polyphosphate fractions increased rapidly then stabilized. At the point of maximum phosphate accumulation, polyphosphates consistently accounted for the dominant 12 25 2 < h-< Z o x Q_ 15 B. 13 0 25 50 75 100 125 150 175 TIME (minutes) Figure 3.1. Profiles of phosphate accumulation and release. (A) 3 ml sample, (B) 4 L batch sample, (C) 4 L batch sample with phosphate release; aeration stopped at 40 minutes, acetate added at 80 minutes. 13 TIME (hours)  e - SCAS P o l y - P L C S Poly-P -© - - LCG Poly-P Nucleic Acid - © - Other Figure 3.2. Typical cellular phosphate composition profile of BEPR sludge during phosphate uptake. 14 species of phosphate-containing compounds in the sludge. On average, they represented up to 85 percent of newly synthesized phosphate-containing material, while nucleic acids, proteins and unrecovered materials accounted for 15 to 30 percent. Long chain soluble polyphosphates made up the majority of total polyphosphates, ranging from 50 to 75 percent of total polyphosphates. Long chain granular polyphosphates made up 20 to 40 percent of total polyphosphates, and short chain acid soluble polyphosphates were normally 10 percent of total polyphosphates. Phosphate uptake and polyphosphate accumulation could continue after mechanical aeration had stopped, but polyphosphate accumulation stopped when phosphate uptake quit (Figure 3.3). In conditions of phosphate release, release consistently occurred from newly synthesized long chain soluble polyphosphate pools. In approximately 50 percent of the samples where release was observed, polyphosphate release from the long chain soluble polyphosphate fraction was accompanied by a variable 1 or 2 mg/L increase in the minor long chain granular polyphosphate fraction. Figure 3.4 shows a characteristic profile of phosphate release during anaerobic nutrient addition including these features. 3.1.3. Thin layer chromatographic analysis of polyphosphates The quality of accumulated polyphosphates was analyzed by thin layer chromatography. Polyphosphates were isolated, examined, and their relative mobilities compared to standards of 32p ortho-phosphoric acid and Y ^ P - A T P . Nucleotides have an Rf of approximately 0.5 compared to an Rf of 1.0 for ortho-phosphoric acid in this solvent system (Figure 3.5). Figure 3.6 shows that long chain polyphosphate species remained at the origin of the thin layer plate, suggesting a chain length greater than 8 ortho-phosphate units long. When the long chain polyphosphate species were heated in 1 N HC1, the mobility of each species progressively increased with exposure time (Figure 3.7 and 3.8). After 7 minutes of heating in strong acid, most of the polyphosphates had been hydrolyzed and migrated with ortho-phosphoric acid. 15 10 E UJ o >-X o Q LU > o w co Q D.O. --+- po4 - - » - Acetate >*• 35 O) 30 E, 1-z 25 TA < z 20 DC LLI QL 3 W 15 z 10 O Q_ UJ H < 5 1-O < 0 0.75 1.50 2.25 3.00 3.75 35 E 30 h < (-< Q. D CO < x a. co O X a. 25 20 15 10 0 0.00 - B - SCAS Poly-P P04 — ° ~ LCS Poly-P LCG Poly-P V / *• -+--+--n 20 1 5 10 0.75 1.50 2.25 3.00 3.75 TIME (hours) co LU o < X a. co O x Q->-_ l o Figure 3.3. Phosphate and polyphosphate accumulation profiles, without anaerobic phosphate release. 16 z 111 a > x o Q UJ > _i o W W Q D.O. --+- P04 Acetate 35 30 25 20 15 10 1.40 2.10 2.80 0 50 z < < CC LU Q_ W O 0-< I-LU o < I-z < I-< z or LU Q. D W I-< X CL W o X 0-35 30 25 20 15 10 0 0.00 A -e- SCAS Poly-P - -•- - P04 — O " LCS Poly-P LCG Poly-P \ / \ • V / A 20 15 10 0.70 1.40 2.10 2.80 TIME (hours) 3.50 TO E UJ o CL W O X CL > O CL ure 3.4. Phosphate and polyphosphate accumulation profiles, with anaerobic phosphate jase. 17 Figure 3.5. Autoradiograph of thin layer chromatograph of 3 2 P ortho-phosphoric acid and 7 3 2P-ATP standards. (A) 3 2 P ortho-phosphoric acid, (B) 7 3 2P-ATP. 18 Figure 3.6. Autoradiograph of thin layer chromatograph of isolated polyphosphates. (A) 32p ortho-phosphoric acid standard, (B) long chain soluble polyphosphate, (C) long chain granular polyphosphate. 19 Figure 3.7. Autoradiograph of thin layer chromatograph of long chain soluble polyphosphates heated in 1 N HC1 for various times. (A) 32p ortho-phosphoric acid standard, (B) 1 minute heating, (C) 3 minutes heating, (D) 5 minutes heating, (E) 7 minutes heating, (F) untreated polyphosphate. 20 Figure 3.8. Autoradiograph of thin layer chromatograph of long chain granular polyphosphates heated in 1 N HC1 for various times. (A) 32p ortho-phosphoric acid standard, (B) 1 minute heating, (C) 3 minutes heating, (D) 5 minutes heating, (E) 7 minutes heating, (F) untreated polyphosphate. 21 3.2. PHA Determination in BEPR Sludge 3.2.1. PHA quantification A typical GC chromatograph obtained from a sludge sample is shown in Figure 3.9. The identity of the compounds was obtained by comparison of relative retention times with those patterns reported by Comeau (1988) and with values obtained from available standards. Hydroxybutyric acid response from the gas chromatograph was calibrated using /3-hydroxybu-tyric acid sodium salt. Standards of j8-hydroxyvaleric acid were not available for calibration of (S-hydroxyvaleric acid response. The relative response factors for a-hydroxybutyric and a-hydroxyvaleric acid should be similar to the relative response factors for jS-hydroxybutyric and i8-hydroxyvaleric acid in the GC , based on the number of carbon atoms per mass of each compound (Willard et al., 1988). Standards of a-hydroxybutyric and a-hydroxyvaleric acid were obtained, and triplicate samples of 50, 100, 150, and 200 jug/ml each of HB and HV extracted and analyzed in the gas chromatograph. The response factors for each were averaged and compared to obtain a response factor for quantification of j8-hydroxyvaleric acid. The observed response ratio for /3-hydroxyvaleric acid to /3-hydroxybutyric acid was calculated to be 1.43, and is greater than the ratio reported by Comeau (1988) of 1.21 using other standards. 3.2.2. PHA recovery and reproducibility Collection of sludge samples in hypochlorite solution slightly enhanced recovery. PHB and PHV recovery were enhanced only 4 % and 3 %, respectively (n=8). Collection of samples in hypochlorite solution appeared to allow better recovery and detection of other unidentified components from the extracted sludge samples. These other components included compounds with retention times of approximately 2.89, 3.94, and 5.24 minutes. Reproducibility of standards was estimated by comparison of slopes of 10 standard curves prepared from replicate 22 Figure 3.9. Chromatograph of methyl ester derivatives extracted from BEPR sludge. Peak identifications and relative retention times are as follows: (1) /S-hydroxybutyric acid, 6.66 mins., (2) /3-hydroxyvaleric acid, 8.28 mins., (3) oxo-valeric acid, 8.48 mins., (4) benzoic acid as internal standard, 9.37 mins. Unidentified peaks and their relative retention times are as follows: (A) 3.14 mins., (B) 4.45 mins., (C) 4.85 mins., (D) 5.24 mins., (E) 6.20 mins., (F) 7.62 mins., and (G) 9.21 mins. 23 samples of 50, 100, 150, and 200 /xg/ml. The observed variation in slopes was calculated to be less than 2%. Reproducibility with sludge samples was estimated by extracting replicate samples of a single sludge sample. The observed error associated with the sludge samples was calculated to be 10% of the calculated concentration of HB or HV in fig/'ml. 3.3. VFA Analysis Clarified sludge samples and acetic, propionic, and butyric acid standards were acidified and analyzed for soluble volatile fatty acid content using a sensitive GC method. Quantification was accomplished by using the external standard calibration function of the Hewlett Packard 3380A integrator. Using replicate samples of standards, the observed error associated with this GC method was calculated as 5% of the concentration in the detection range of 1 to 50 mg/L. Analyzed sludge samples appeared to contain no detectable soluble acetic, propionic, or butyric acid. Sludge pellets were tested for intracellular VFA, but none were detected. Removal of added acetic acid from BEPR sludge could be followed using this technique. 3.4. Bench Scale Batch Testing Bench scale batch testing using 4 L of BEPR sludge was conducted by the methods described previously. Twenty-five batch experiments were performed to determine the behavior of NADH, polyhydroxyalkanoates, and soluble phosphorus under aerobic, anoxic, and anaerobic conditions, with and without the presence of added nitrates or nutrients, and relative to the age of the sludge. Profiles illustrating the behavior of these components under the various conditions are provided in Figures 3.10 through 3.12. 24 3.4.1. Characteristic NADH and ORP responses Figure 3.10 illustrates some of the features of the NADH and ORP responses that were commonly observed. Typically, during the beginning phase of an experiment the observed NADH decreased at a fairly rapid rate, indicating a significant shift in the NADH/NAD ratio. Oxidation-reduction potential also decreased during this phase. Upon aeration a steep drop in observed NADH, typically from 5 to 15 NFU, was observed within the first 5 minutes, while the ORP increased (Figure 3.10.(A), feature (2)). The decrease in observed NADH was expected as the cells metabolic capabilities undergo transition to aerobic metabolism where rapid oxidation of NADH can occur. Observed NADH continued to decrease at a slower rate upon continued aeration. When aeration was stopped ORP decreased. NADH continued to decrease at the rate established immediately prior to the cessation of aeration, or stabilized until denitrification was complete. In approximately 50% of the samples tested, the NADH was stabilized after aeration stopped. When denitrification was complete, observed NADH jumped to a new level anywhere from 5 to 10 NFU higher than the pre-denitrification level (Figure 3.10(A), feature (5)). The jump was never as great as the decrease observed from aeration. A corresponding increase in the rate of oxidation-reduction potential decrease is observed (Figure 3.10(A), feature (4)) and was shown to correspond with the completion of denitrifi-cation (Koch and Oldham, 1985). These features in NADH and ORP have been termed nitrate knees'' and have been used as useful indicators to operationally define the transition between anoxic and anaerobic conditions in the sludge. In some instances the inflection in ORP slope was not detected. In these cases, the NADH knee provided an alternative indicator of the reduction state of the sludge. Once an anaerobic state was established, ORP decreased at a slower rate and eventually stabilized at a low value. NADH usually stabilized or decreased, depending on the nutrient conditions of the sample. If samples remain in an anaerobic condition for extended periods, usually greater than 15 hours, the NADH increased. The increase will continue over several hours until the observed NADH level stabilized at a high level similar 25 11.20 13.50 15.80 18.10 20.40 22.70 TIME (hours) Figure 3.10. Characteristic profile and features of NADH during typical batch run (A) and extended unaerated run (B). Some of the features illustrated by the figures include (1) "Beginning" phase of run, (2) Profile during aeration, (3) Profile during unaerated period, (4) ORP slope change referred to as "nitrate knee", (5) corresponding "nitrate knee" of NADH. 26 to that observed at the beginning of the experiment. Figure 3.10(B) shows this increase in NADH during extended anaerobic exposure. 3.4.2. General PHA, VFA, and soluble phosphate behavior Figures 3.11 and 3.12 provide examples of characteristic batch experiments with acetate additions during different phases of the run. Acetate addition prior to aeration resulted in rapid accumulation of both PHB and PHV, as soluble acetate was removed from the mixed liquor. In some of the experiments, PHA accumulation was observed during the beginning phase of the test, in the absence of added or detectable nutrient, and could be as much as 40 mg PHA/ L. In 75 % of the batch experiments, PHA accumulation continued after the detectable VFA were gone. During this accumulation period, soluble phosphate levels varied, although a significant release is expected (Comeau, 1988). A release, however, may be masked by the sudden change in specific activity of 3 2 P in the sludge, during this initial period following 32p addition. PHA and soluble phosphate levels decreased during aeration. After mechanical aeration was stopped, both PHA and soluble phosphate tended to continue to decrease until denitrification was complete. The observed rate of decrease in soluble phosphate after aeration quit was 5 to 80% slower than that observed during aeration in 70% of the samples, while the rate of PHA decrease remained relatively constant in more than 70% of the samples. Anaerobically the levels of PHA and soluble phosphate are influenced by the presence of added nutrients. Levels of PHA and soluble phosphate stabilized anaerobically if no nutrients were added (Figure 3.11). If the sample was held for extended periods anaerobically, PHA levels would be observed to decrease slowly, as much as 30 to 60%, once the NADH had begun to rise as in Figure 3.10(B). Both PHA and soluble phosphate levels increased upon anaerobic nutrient addition. Acetate addition following denitrification stimulated rapid PHA accumula-tion and phosphate release. Phosphate release stopped when soluble acetatewas removed and then remained stable (Figure 3.12). Again, both PHV and PHB accumulated rapidly and 27 28 0.00 1.75 3.50 Z LU O Q 0.00 1.75 3.50 TIME (hours) Figure 3.12. Characteristic profile of NADH, ORP, PHA, soluble phosphate, VFA, and dissolved oxygen for a typical batch run with anaerobic nutrient addition. 29 accumulation continued past the point at which soluble VFA became undetectable (Figure 3.12), and in some cases resulted in a further 50% increase in the total PHA pool. Characteristically, the rate of PHA accumulation was significantly slower after detectable VFA were consumed. 3.4.3. General sludge patterns The behaviors of NADH, ORP, PHA, and soluble phosphate were influenced strongly by the availability of nutrients and the reduction state of the sludge. The specific responses of these components under various nutrient and reduction conditions are outlined in Tables I through VIII and a summary of the general observed trends from the individual experiments provided in Table IX. From the individual runs and summary, patterns of behavior could be established for various conditions. Typically, NADH was expected to decrease during the beginning phase of an experiment, and in aerobic conditions regardless of nutrient availability. NADH response was affected by nutrient availability under anoxic and anaerobic conditions. Anoxically, nutrient addition caused a tendency for observed NADH to increase, while lack of an available nutrient resulted in a tendency for NADH to remain stable or decrease. Anaerobically, NADH tended to remain stable or occasionally decrease in the absence of added nutrient. Presence of an available nutrient stimulated a decrease in NADH. PHA accumulation was affected mostly by nutrient addition, although accumulation was observed during the beginning phase of some experiments in the absence of any added nutrients. PHA pools decreased under oxic or anoxic conditions, but accumulation occurred in some instances if nutrient was added. PHA could also decrease in situations of extended anaerobic exposure. Phosphate could be accumulated aerobically and anoxically, but was released anoxically if carbon was present. Anaerobically, phosphate release appeared most dependant on carbon availability. 30 Table I. Summary of NADH, ORP, PHA, and soluble phosphate behavior during "beginning" phase of experimental runs, without aeration or added carbon. RUN NADH ORP PHA Soluble phosphate 220 nd nd 225 nd nd 304 nd nd 312 v s nd 319 + nd 325 + nd 418 s s 430 s + 506 + s 509 nd nd 517 nd nd 524 nd nd 525 nd nd 527 nd nd 610 nd nd 611 nd nd 612 nd nd 614 nd nd 615 nd nd 616 nd nd 617 nd nd 627 nd nd Notes: "-" indicates decrease " + " indicates increase "s" indicates no significant change "v" indicates variable behavior "nd" indicates no data available 31 Table II. Summary of NADH, ORP, PHA, and soluble phosphate behavior during carbon addition prior to aeration. RUN NADH ORP PHA Soluble phosphate 304 v - + nd 312 - + nd 509 - + -517 s - + s 524 - + -525 - s s 527 - s s 610 - * + + 611 - + + 612 - + s 614 - + + 615 - + s 616 - + s 617 - + s 627 - + + Notes: "-" indicates decrease " +" indicates increase "s" indicates no significant change "v" indicates variable behavior "nd" indicates no data available Table III. Summary of NADH, ORP, PHA, and soluble phosphate behavior during aerobic carbon addition. RUN NADH ORP PHA Soluble phosphate 325 - + + nd 509 - + - -517 - + - -Notes: "-" indicates decrease " +" indicates increase "nd" indicates no data available 32 Table IV. Summary of NADH, ORP, PHA, and soluble phosphate behavior during aeration, without added carbon. RUN NADH ORP PHA Soluble phosphate 214 + nd 220a + nd 220b + nd 225 + nd 304 + nd 319 + nd 325 + nd 418 + -509 + -517 + -524 + -525 + -527 + -610 + -611 + -612 + -614 + -615 + -616 + -617 + -627 + -Notes: "-" indicates decrease " + " indicates increase "nd" indicates no data available 33 Table V. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anoxica carbon addition following aeration. RUN NADH ORP PHA Soluble phosphate 220 + . + nd 319 + - + nd 418 + - + + 325 s - s nd Notes: a - anoxic conditions defined by ORP or NADH "nitrate knee" "-" indicates decrease " +" indicates increase "s" indicates no significant change "nd" indicates no data available 34 Table VI. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anoxica conditions following aeration, without added carbon. RUN NADH ORP PHA Soluble phosphate 214 + - s nd 220 s - s nd 225 s - s nd 418 - - s s 509 - • - - -517 - - - -524 - - s s 525 - - - . s 527 s - s -610 - - - -611 - - - -612 + - s -614 s - s -615 s - - -616 - - - -617 + - s -627 s - s -Notes: a - anoxic defined by ORP arid/or NADH "nitrate knee" "-" indicates decrease " +" indicates increase "s" indicates no significant change "nd" indicates no data available 35 Table VII. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anaerobica conditions, without added carbon. RUN NADH ORP PHA Soluble phosphate 325 s - s nd 509 + - + -517 - - s s 524 s - + s 525 - - s s 527 s - s s Notes: a - anaerobic defined by ORP and/or NADH "nitrate knee" "-" indicates decrease " +" indicates increase "s" indicates no significant change "nd" indicates no data available 36 Table VIII. Summary of NADH, ORP, PHA, and soluble phosphate behavior during anaerobica carbon addition. RUN NADH ORP PHA Soluble phosphate 214 + + nd 220 s s nd 225 s + nd 319 - s nd 325 s + nd 418 s + s 517 - + + 524 s V V 525 - + + 527 - + + 610 - + + 611 - V + 612 - + s 614 - + + 615 - + s 616 - + + 617 - + + 627 - + + Notes: a - anaerobic defined by ORP and/or NADH "nitrate knee" "-" indicates decrease " +" indicates increase "s" indicates no significant change "v" indicates variable behavior "nd" indicates no data available 37 Table IX. Summary of general behavior of NADH, ORP, PHA, and soluble phosphate under various environmental conditions. Behavior of components as percent (%) of total observations Condition: NADH ORP PHA Phosphate - s + - s + - s + - s + beginning, no carbon 100 0 0 95 5 0 0 50 50 0 67 33 pre-aeration, with carbon 93 7 0 100 0 0 0 13 87 15 54 31 aerobic, with carbon 100 0 0 0 0 100 0 67 33 100 0 0 aerobic, no carbon 100 0 0 0 0 100 100 0 0 100 0 0 anoxic, no carbon 47 35 18 100 0 0 41 59 0 79 21 0 anoxic, with carbon 0 25 75 100 0 0 0 25 75 0 0 100 anaerobic, no carbon 33 50 17 100 0 0 0 67 33 20 80 0 anaerobic, with carbon 67 28 5 100 0 0 0 12 88 0 25 75 Notes: "-" indicates decrease " +" indicates increase "s" indicates no significant change Percentages based on total number of cases reporting results for each particular component under each condition 3.4.4. BEPR sludge behavior during nitrate additions Experiments designed to test the response of the sludge during anoxic conditions were performed by feeding sludge NaNC«3 at various concentrations. An abrupt decrease of 3 to 5 NFU upon nitrate addition was consistently observed. NADH remained at this lower level for various lengths of time and then abruptly rose to the pre-addition level upon completion of denitrification. This response of NADH resembled the v " knee'' effect observed during the transition between anoxic and anaerobic conditions in normal batch experiments. Figure 3.13 shows the sludge response during nitrate additions of 2.5, 6.0, and 11.5 mg/L. The length of time the NADH remained depressed after nitrate addition, was correlated to the amount of nitrate added (Figure 3.14). The NADH drops during nitrate additions were not as large as the drops observed during aeration (Figure 3.13). Nitrate addition usually caused phosphate uptake and cause a variable response in PHA. Phosphate uptake was consistently observed during simultaneous nitrate and acetate addition, but PHA response was variable (Figure 3.15). In some additions, PHA was rapidly degraded then slowly re-established. In other additions, PHA showed a net increase without observable degradation. The PHA response was highly variable in several nitrate/acetate additions and the degradation or accumulation was indepen-dent of concentration of additives or measured oxidation-reduction potential and NADH. When the NADH signal indicated completion of denitrification, phosphate release would often be observed. 3.4.5. Relationships of PHA, VFA, NADH, and phosphate PHA accumulation was dependant on carbon addition, except during the beginning phase of batch runs where PHA accumulation was observed without the addition of exogenous carbon. When added nutrient was compared to subsequent PHA accumulation, the extent of accumu-lation exceeded the predicted accumulation based on input carbon. Table X shows the ratio 39 Figure 3.13. NADH and ORP response during anaerobic nitrate additions. Concentrations and times of nitrate additions are as follows: 2.5 mg/L at 2.78 hours, 6.0 mg/L at 2.96 hours, 11.5 mg/L at 3.22 hours, and air on at 3.55 hours. Nitrate additions indicated with arrows. NITRATE ADDITION (mg/L) Figure 3.14. Relationship between nitrate addition and length of NADH depression. 40 Li-z X Q < > n_ rr O -100 > X o 0.38 i l D.O. P04 Acetate 45 00 1.27 1.90 cn O TIME (hours) Figure 3.15. Experimental data from a series of anaerobic nitrate and acetate additions (10 mg/ L each of nitrate and acetate at each addition). Additions were made at 0.26, 0.69, and 1.43 hours. 41 Table X. Comparison of net PHA accumulation and input carbon as millimoles of carbon per litre. Run Net PHA accumulation (millimoles of carbon per litre) Carbon input (millimoles of carbon per litre) Ratio of net PHA accumulation to carbon input 214 1.10 0.61 1.80 220 1.13 0.97 1.16 225 1.47 0.73 2.01 312a 1.00 0.73 1.37 312b 0.92 0.49 1.88 319 0.36 0.73 0.49 325 1.34 0.37 3.62 509 2.48 0.49 5.06 517a 1.02 0.49 2.08 517b 1.55 0.49 3.16 524 0.96 0.37 2.59 525* 1.68 2.73 0.73 527 2.63 1.69 1.56 610 4.04 1.93 2.09 611* 1.34 1.95 0.68 612 2.55 1.98 1.29 614* 0.49 1.69 0.29 615* 1.28 0.49 -2.69 616* 1.81 0.49 3.69 617* 0.16 0.49 0.33 627* 1.95 2.23 0.87 Notes: "*" indicates potential inhibitor present in input carbon 42 of net PHA (PHB and PHV) accumulated to carbon added was usually greater than 1, when compared as millimoles of carbon per litre, and can be as high as 5. With only two exceptions, the only instances where this ratio was less than 1 were those cases where potential inhibitors of cellular pathways were present. No obvious difference in the nature of the samples was noted in these exceptional cases. PHA behavior was largely independent of NADH, but some patterns during PHA accumulation resulting from anaerobic nutrient additions did exist. Nutrient additions which stimulated significant PHA storage caused either a greater extent or rate of decrease in observed NADH. Quantitative correlations between the extent or rate of PHA accumulation and NADH decrease were not found. No quantitative correlations was established between the soluble phosphate concentrations and PHA, either during anaerobic PHA accumulation and phosphate release, or during aerobic PHA decrease and phosphate uptake. Similarly, no quantitative correlations existed between acetate uptake and phosphate release. 3.5. Biochemical Model Testing Specific tests were designed to test the possible role of various biochemical pathways of the TCA and glyoxylate cycles in the PHA accumulation reaction. Anaerobic acetate feeding results in accumulation of both PHB and PHV (Comeau, 1988; Satoh et al, 1990). Probable biochemical pathways for production of both PHB and PHV have been established (Doi et al.,1989; Fernandez-Briera and Garrido-Pertierra, 1988)(Figure 3.16). PHB is formed by condensation of two acetyl CoA groups, while PHV formation requires condensation of acetyl CoA and propionyl CoA, a 3-carbon compound. However, the need for PHV synthesis and the source of necessary reducing power are unexplained. Acetyl CoA could enter several metabolic pathways, but under the anaerobic conditions the activity of many of these routes are impaired or produce unfavorable end-products. One proposed source of reducing power isthe metabolism of acetyl CoA to oxaloacetate in the glyoxylate cycle. Several organisms demonstrating BEPR capabilities have active glyoxylate cycle enzymes in anaerobic conditions 43 (Lotter, 1989; Matsuo, 1985; Osborn et al., 1986). Continued metabolism of acetyl CoA tin the glyoxylate cycle would produce an excess of succinate which would rapidly inhibit the pathway (Herman and Bell, 1970). This inhibition could be relieved by conversion of accumulated succinate to propionyl CoA, a source of propionyl CoA, the 3-carbon compound leading to PHV. Prolonged cycle operation necessitates a ratio of 1 NADH equivalent for each pair of acetyl CoA, or pair ofacetyl CoA and propionyl CoA condensed. Any other ratio would lead to an inhibitor imbalance. Based on this reasoning for production of reducing power and propionyl CoA, and the probable pathways assumed involved in the PHA reaction (Bell and Herman, 1967; Comeau etal., 1985; Fernandez-Briera and Garrido-Pertierra, 1988; Herman and Bell, 1970; Hodgson and McGarry, 1968; Jackson and Dawes, 1976; Morrison and Peters, 1954; Satoh et al., 1990; Weitzman, 1972; Wentzel et al., 1986;), a biochemical model for acetate utilization is presented in Figure 3.16. The model basically says that acetate is condensed directly to PHB, using the necessary reducing equivalents supplied by the glyoxylate cycle which will only function if accumulated succinate is converted to propionyl CoA and consumed in PHV production. Since changes in the ratio of PHV to PHB can be easily monitored, testing involved generating predictions from the model about the net generation of suitable reducing equivalents and relative requirements of PHB and PHV synthesis for samples treated with various nutrient and inhibitor combinations under anaerobic conditions. The predictions were tested and the observations recorded and used to obtain information about the function of various pathways under each condition. In total, twelve specific tests were designed and conducted for biochemical testing of the proposed model. Detailed explanations of the individual tests follow. 44 Acetoacetyi CoA ^ N A D H /3-Hydroxy-butyr CoA PHB P H V Acetate Acetyl CoA Acetyl CoA Inhibitors: a = a-ketoglutarate b = fluorocitrate c = anaerobiosis d = malonate Pyruvate Hydroxy-valeryl CoA N A D N A D H ^ r N A D H ^*>NAD Lactate N A D H Citrate Glyoxylate Succinate N A D Oxo-valeryl CoA j * •NAD N A D H Succinyl CoA Propionyl CoA Figure 3.16. Proposed biochemical model for PHA storage under anaerobic conditions. Points of inhibition indicated by double lines. Inhibitors indicated by lower case letters. 3.5.1. Acetate addition Based on the proposed model in Figure 3.16, acetate addition results in accumulation of both PHB and PHV. As describe earlier, acetate forms acetyl CoA and enters the TCA cycle to form citrate. Citrate is processed further to isocitrate and enters the glyoxylate shunt, since conversion of isocitrate to a-ketoglutarate is inhibited by anaerobiosis (Michal, 1982). The glyoxylate shunt produces malate and succinate. To prevent feedback inhibition from excess carbon in the cycle, succinate is converted to propionyl CoA to drain excess carbon and prevent inhibition of carbon flow in the system. The malate is cycled to oxaloacetate with the production of 1 reducing equivalent. The oxaloacetate combines with another acetyl CoA to form citrate. The reducing equivalent produced is consumed by the production of either hydroxybutyrl CoA or hydroxyvaleryl CoA from 2 acetyl CoA or 1 acetyl CoA and 1 propionyl CoA, respectively. The hydroxybutyrl CoA is condensed to produce PHB, while the hydroxyvaleryl CoA is condensed to form PHV. Because acetyl CoA is produced quickly from acetate, an excess of acetyl CoA over propionyl CoA results and causes greater synthesis of PHB. The PHA accumulation should be rapid and accompanied by a rapid release of phosphate and a decrease in NADH, based on other observations for anaerobic acetate addition. The behavior of the BEPR sludge with anaerobic acetate addition is shown in Figure 3.17. When acetate was added anaerobically, accumulation of both PHB and PHV were observed with an accumulation ratio of PHV/PHB of approximately 0.7. Typically, the ratio of PHV/PHB for other batch experiments involving anaerobic acetate addition ranged from 0.6 to 0.75. Net PHA accumulation was around 40 mg/L. The net accumulation of carbon was greater than the input carbon, with a ratio of 2.08 compared on a millimoles of carbon per litre basis, and was similar to previous observations. Rapid phosphate release, rapid PHA accumulation, and a decrease in NADH were observed. These observations supported the predictions based on the model. The appearance of PHV supported the prediction of succinate conversion to propionyl CoA. Propionyl CoA could alternatively be produced from excess malate via pyruvate and lactate. 46 E, > 10 | 35 o >-X o Q HI > -I o in jl 1 I D.O. P04 ! \ \ ! V — * - Acetate < z rr 0.00 3.50 TIME (hours) Figure 3.17. Experimental data from experiment with anaerobic acetate addition. Acetate (20 mg/L) added at 2.0 hours. 47 Succinate may produce fumarate with production of 1 reducing equivalent, rather than being converted to propionyl CoA. The fumarate is then converted to malate, resulting in malate accumulation. To prevent feedback inhibition, the excess malate is converted to propionyl CoA, with no net generation of reducing equivalents, and some malate would be cycled through to citrate as described before. The propionyl CoA, acetyl CoA, and excess reducing equivalents could be used to produce both PHB and PHV. To test whether succinate forms fumarate or propionyl CoA, an addition of acetate and malonate was made. Malonate has been reported to inhibit enzymatic conversion of succinate to fumarate (Hochster and Quastel, 1963; Michal, 1982). In the presence of malonate, either the ratio of PHV/PHB would decrease or PHA accumulation become inhibited due to accumulation of succinate, if succinate is converted to fumarate. Figure 3.18 shows the BEPR sludge response to anaerobic addition of malonate and acetate. The pattern of response for PHA and phosphate was similar to the response observed for acetate addition alone, and the PHV/PHB ratio remained approximately 0.6, again with net carbon accumulation greater than input carbon. A third experiment involving acetate and pyruvate addition was done to test the prediction that pyruvate was converted to acetyl CoA with the generation of reducing power. The addition of pyruvate with acetate should result in rapid PHA accumulation with a strong PHB response resulting from the excess available acetyl CoA. Figure 3.19 shows that pyruvate addition with acetate resulted in a slower rate of PHA accumulation than expected and was accompanied by a slow phosphate release. Both PHB and PHV were accumulated to similar extents with a PHV/ PHB ratio around 0.88. Net PHA accumulation was over 50 mg/L. The net carbon accumulation was greater than both the input carbon and the accumulation typical for a similar acetate addition, compared as millimoles of carbon per litre. The observed NADH decreased more sharply than usual. 48 49 170 150 -150 > E o 100 100 (3 > X o o w Q 1.23 1.78 2.34 TIME (hours) 2.89 3.45 < z rr UJ 0-in < i a. CO O x Figure 3.19. Experimental data from experiment with anaerobic acetate and pyruvate addition. Acetate (10 mg/L) and pyruvate (50 mg/L) added at 1.89 hours. 50 3.5.2. Citrate additions Anaerobic addition of citrate was predicted to result in accumulation of PHA composed mostly of PHB, or inhibit PHA accumulation. PHB accumulation results if citrate enters cells and produces acetyl CoA and oxaloacetate by reversal of the TCA cycle. The oxaloacetate produces malate with consumption of 1 reducing equivalent. To prevent malate accumulation, malate is converted to pyruvate which is then converted to acetyl CoA with the production of 2 reducing equivalents. A net gain of 1 reducing equivalent and 2 acetyl CoA are realized and used to produce hydroxybutyrl CoA, and subsequently PHB. Alternatively, citrate addition can inhibit PHA accumulation by causing an imbalance of reducing equivalents. Citrate may be converted to isocitrate, rather than acetyl CoA and oxaloacetate. The isocitrate will enter the glyoxylate cycle to produce malate and succinate. Excess succinate will be converted to propionyl CoA, while excess malate is converted to oxaloacetate with generation of 1 reducing equivalent. Excess citrate will inhibit further citrate production from oxaloacetate causing oxaloacetate to accumulate. Without any acetyl CoA to consume the excess reducing power through formation of PHA, the imbalance should inhibit PHA accumulation. Figure 3.20 shows transient changes in both PHB and PHV, with no significant net gain in total PHA during the observation period. The PHV/PHB ratio remained unchanged from the ratio of approximately 1 before addition. A small phosphate release was observed and NADH remained relatively stable. Assuming citrate was entering the cells, the observations support the second prediction and suggest that citrate produces some transient effect on PHA accumulation and that citrate was not converted to acetyl CoA and oxaloacetate. Alternatively, the citrate may not have entered the cells and thus, causing no effect in the response pattern. The small phosphate release and transient changes in PHA suggested, however, that citrate is likely causing some effect at the cellular level. A second test with citrate was performed with both acetate and malonate added. The 51 175 150 E 0.75 1.50 100 •=• 60 20 1.50 2.25 20 3.00 0 >-X o D O CO CO Q 0.00 0.75 1.50 TIME (hours) 2.25 3.00 CO O I Figure 3.20. Experimental data from experiment with anaerobic citrate addition. Citrate (50 mg/L) added at 1.9 hours. 52 prediction was that the addition of acetate would provide the acetyl CoA necessary to consume the excess reducing power that would normally inhibit PHA storage. Malonate will again test the importance of the succinate to fumarate conversion. When the combination of nutrients was added anaerobically, a rapid and significant increase in PHA and soluble phosphate was observed (Figure 3.21). The ratio of PHV/PHB of 0.65 was similar to those observed from acetate addition alone. The ratio of net carbon accumulation to input carbon was 0.73, expressed as millimoles of carbon per litre. NADH decreased rapidly as PHA accumulated then decreased more slowly once PHA accumulation stabilized. The observations may result solely from the effect of acetate, or may reflect the ability of the acetate to relieve the NADH imbalance caused by citrate utilization. 3.5.3. Succinate additions Based on the proposed model, succinate addition should result in inhibition of PHA accumulation. Succinate would be converted to propionyl CoA, or to fumarate with the generation of 1 reducing equivalent. Fumarate would be transformed to malate, which would accumulate. Excess malate would be drained from the cycle through conversion to pyruvate and generate a second reducing equivalent. Pyruvate would produce acetyl CoA and a third reducing equivalent. One of the reducing equivalents could be consumed in the formation of hydroxyvaleryl CoA from the resulting acetyl CoA and propionyl CoA. Unless, a source of acetyl CoA was present to consume the extra reducing equivalents, a state of imbalance would result and inhibit PHA accumulation. Anaerobic succinate addition resulted in a net increase in both PHB and PHV, after a lag period (Figure 3.22). The ratio of PHV/PHB shifted significantly to 1.4, and the net carbon accumulation was greater than the carbon input, with a ratio of 1.56. Initially, phosphate was released rapidly, then was released at a much slower rate. The NADH instantaneously dropped 4 NFU and continued to decrease another 4 NFU 53 200 150 E 2.25 3.00 100 > I O Q > TIME (hours) Figure 3.21. Experimental data from experiment with anaerobic citrate, malonate, and acetate addition. Citrate (50 mg/L), malonate (50 mg/L), and acetate (25 mg/L) added at 1.96 hours. 54 5 E a. rr O 120 120 > 0.00 3.00 x O o in 0.00 0.75 1.50 2.25 TIME (hours) in O Figure 3.22. Experimental data from experiment with anaerobic succinate addition. Succinate (50 mg/L) added at 1.67 hours. 55 over the next 1.2 hours. The sludge behaved as if some source of acetyl CoA was available to consume the excess reducing power and produce PHB. Figure 3.23 shows the results in the presence of succinate and acetate, to test if acetate, a source of acetyl CoA, could account for the previous observation. The acetate should provide a readily utilizable source of acetyl CoA for consumption of excess reducing power and production of PHB and should reproduce patterns similar to those patterns from the previous experiment. In Figure 3.23, PHV and PHB increased at a faster rate than observed for succinate addition alone, with respective increases of 400 and 100 percent. The ratio of PHV/PHB was 1.9, and the ratio of net carbon accumulation to input was 2.09. Initially, NADH dropped sharply over 7 NFU then remained stable for the remaining observation period. Phosphate levels increased at a fast rate initially, then continued to rise at a rate approximately 60 percent slower. The overall patterns observed in the two trials were similar, although the rates of change observed in phosphate and PHA were different. A third experiment designed to test the role of the succinate to fumarate conversion, during succinate utilization, was done by adding succinate and malonate. If the production of fumarate and reducing power from succinate is important, malonate should impair the PHV response compared to the response for succinate addition alone. When succinate and malonate were added anaerobically, the response pattern was similar to that observed for succinate addition alone. PHV increased slowly about 15 mg/L while PHB showed a small net decrease when malonate and succinate were added anaerobically (Figure 3.24). The PHV/PHB ratio was high at 2.1 and the carbon accumulation to carbon input ratio was only 0.29. Phosphate levels increased slightly for 1.3 hours after the succinate/malonate addition then stabilized. NADH followed a pattern similar to that observed when succinate was added alone showing an initial sharp decrease of 3 to 4 NFU followed by a further, slower decrease of 4 NFU. 56 0.70 1.40 2.10 2.80 O < TIME (hours) Figure 3.23. Experimental data from experiment with anaerobic succinate and acetate addition. Succinate (50 mg/L) and acetate (10 mg/L) added at 1.92 hours. 57 3 U-z • < 175 165 160 V 155 175 -175 > E 0.72 1.38 2.03 2.69 3.34 4.00 0.72 1.38 2.03 2.69 3.34 4.00 CD > X o o CO CO Q 0.72 2.03 2.69 TIME (hours) Figure 3.24. Experimental data from experiment with anaerobic succinate and malonate addition. Succinate (50 mg/L) and malonate (50 mg/L) added at 2.04 hours. 58 3.5.4. Nutrient/inhibitor combinations The use of compounds, other than malonate, to test the functionality of specific proposed pathways was conducted using a-ketoglutarate and monofluoroacetic acid. The entry of acetyl CoA has reportedly been blocked by a-ketoglutarate (Michal, 1982). The addition of acetate and a-ketoglutarate should impair the normal PHA accumulation response observed during acetate addition. If reducing equivalents and propionyl CoA are generated from acetyl CoA entering the glyoxylate cycle, a-ketoglutarate should prevent generation of both these components and impair the PHA response. If these components are produced by a different pathways, a typical pattern for acetate addition is expected. Figure 3.25 shows that both PHB and PHV undergo transient changes that resulted in a small net gain of only 7 mg/L total PHA. Both the ratio of PHV/PHB and net gain of PHA were atypical for acetate addition and the PHV/ PHB ratio had not changed significantly from the ratio established before the addition. An atypical phosphate release was also observed, and the rate of release was much slower than expected. NADH was shown to decrease 6 NFU after the addition. To more directly test the specific role of the glyoxylate cycle during anaerobic PHA storage, a potent inhibitor of TCA cycle activity was used. Monofluoroacetic acid can freely enter cells and is converted to fluorocitrate by functioning TCA cycles. Fluorocitrate is an inhibitor of the enzyme aconitase (Morrison and Peters, 1954) which is responsible for the conversion of citrate to isocitrate. Simultaneous additions of monofluoroacetic acid and acetate were made to BEPR sludge. If the pathways predicted for acetate utilization are correct, PHA response is expected to be severely impaired. In pure enzyme and pure culture systems, monofluoroacetate has been shown to be inhibitory at a concentrations of 0.2 fiM and 10 fj.M, respectively (Beatty, 1980; Morrison and Peters, 1954). In these experiments, additions resulted in monofluoroacetate concentration of 50 fiM and an acetate concentration of 20 mg/L. The presence of acetate may relieve the inhibition created from the fluorocitrate through competition for aconitase by 59 X D < Z 170 250 150 <r O 1.65 2.20 2.75 100 100 > i Q_ O W W a 0.00 0.55 1.10 1.65 TIME (hours) O I < Figure 3.25. Experimental data from experiment with anaerobic acetate and a-ketoglutarate addition. Acetate (10 mg/L) and a-ketoglutarate (50 mg/L) added at 1.58 hours. 60 produced citrate. Simultaneous addition of monofluoroacetic acid and acetate resulted in initial transient PHA behavior followed by a slow increase to an extent much lower than normally observed for a similar acetate addition (Figure 3.26). Total PHA accumulation was less than half of that normally observed for acetate addition, and the ratio of PHV/PHB was 0.73. Phosphate release was slow, compared to the rapid release typically seen during anaerobic acetate addition. Data from the experiment with sequential addition of monofluoroacetic acid and acetate is presented in Figure 3.27. Addition of monofluoroacetic acid resulted in a decrease in PHA following a short lag. During this period, NADH decreased and a slow phosphate release was observed. NADH decreased 3 NFU until acetate was added. At this time the NADH continued to decrease at a slower rate. Phosphate release also slowed and PHA appeared to accumulate, then degrade. No significant net gain in PHA was realized over the course of the additions, and the ratio of PHV/PHB remained essentially unchanged after the additions. To test the possibility of PHA enzyme inhibition by the fluoroacetic acid, an experiment using pyruvate, monofluoroacetic acid, and acetate was conducted. From the previous experiments, it appears that reducing equivalents can be generated through the conversion of pyruvate to acetyl CoA, thus providing a source of reducing equivalents separate from the glyoxylate cycle, which should be inoperative in the presence of monofluoroacetic acid. Figure 3.28 shows the results from this experiment. Both PHB and PHV were shown to increase after the addition of pyruvate, acetate, and monofluoroacetic acid. PHB increased steadily from 31 to 56 mg/L, while PHV increased from 34 to 57 mg/L. The observed patterns in PHA and phosphate resembled those patterns observed for addition of pyruvate and acetate alone. NADH decreased slowly then stabilizes after the addition. 61 150 > E 0.74 1.48 2.22 2.96 3.70 X O Q UJ > _l O cn w Q 1.48 2.22 TIME (hours) Figure 3.26. Experimental data from experiment with anaerobic acetate and monofluoroacetate addition. Acetate (15 mg/L) and monofluoroacetate (50 JUM) added at 2.28 hours. 62 160 1S0 <r O 1.93 2.37 2.81 3.25 > x CL =d 8 \ x o o in in 1.93 TIME (hours) in O Figure 3.27. Experimental data from experiment with sequential anaerobic addition of acetate and monofluoroacetate. Monofluoroacetate (50 fiM) added at 2.32 hours; acetate (15 mg/L) added at 2.67 hours. 63 • < z 0.65 1.30 2.60 160 -75 150 3.25 rr O O >-x o Q 1.95 TIME (hours) 2.60 3.25 CL O Figure 3.28. Experimental data from experiment with anaerobic addition of acetate, monofluoroacetate, and pyruvate. Acetate (15 mg/L), monofluoroacetate (50 /uM), and pyruvate (50 mg/L) added at 1.68 hours. 64 4. DISCUSSION The experimental results from initial and bench scale batch testing will be discussed by considering the responses of phosphate, PHA, and NADH separately. The results will be compared to reported observations regarding the individual and collective behaviors of the various components, and proposed control mechanisms under various environmental condi-tions, with particular reference to the work of Comeau conducted on the University of British Columbia pilot plant. Some of the bench scale batch testing results will also be considered during discussion of the biochemical model testing experiments where the metabolic pathways will be examined. Comparison of the proposed model with existing models will be made and difficulties with the model will be described. 4.1. Bench Scale Batch Experiments 4.1.1. Phosphate behavior The response of phosphate in BEPR sludges has been extensively studied. With a few exceptions (Buchan, 1981; Hascoet et al., 1985; Hill et al., 1989; Mino et al., 1985; Suresh et al., 1985), most studies employ traditional chemical methods for detecting changes in bulk solution and intracellular phosphate levels, reporting total changes without differentiating existing and newly acquired phosphate pools. To gain a better understanding of the phosphate dynamics, techniques allowing the differentiation and fate of new phosphates and phosphate-containing materials are required. The use of radioactive phosphate allowed the response of various phosphate pools to be followed, while providing information about the fate and turnover of new phosphates. The general phosphate responses observed under the various conditions were in agreement 65 with reported observations (Comeau, 1988; Deinema et al., 1980; Fuhs and Chen, 1975; Hascoet et al., 1985; Hill et al., 1989; Iwema and Meunier, 1985; Mino et al!, 1985; Murphy and Loiter, 1986; Ohtakeetal., 1985; Suresh etal., 1985; van Groenestijn, 1985; Vasiliadis et al., 1990; Wentzel et al., 1991). The radioactive 3 2 p techniques employed in this study showed insignificant turnover or conversion within the newly synthesized polyphosphate pools, although constant turnover between various polyphosphate forms is often presumed (Fuhs and Chen, 1975; Mino et al., 1985). The size of polyphosphates had not been previously investigated, but analysis of isolated polyphosphates indicated that newly synthesized polyphosphates were greater than eight ortho-phosphate units long and susceptible to acid hydrolysis. Both the Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) and Mino (Mino et al., 1987) models postulate that aerobic consumption of PHA provides energy for cell requirements and polyphosphate acquisition. Comeau (1988), using chemical methods for phosphate determination, found correlations between the aerobic phosphate uptake and decrease in PHA, and reported molar ratios for phosphate uptake per PHA consumed of 2.6 and 4.1. No correlations were established between phosphate uptake and the PHA decrease in this study, and likely results from differences in cellular response rather than differences resulting from different methods of phosphate determination, since radioactive estimates and occasional chemical determinations were in close agreement. The energy generated from PHA consumption could be partitioned differently between cell requirements and polyphosphate storage in Comeau's studies and these tests, accounting for the lack of consistent correlation. Initial rapid phosphate uptake followed by a slower uptake rate has frequently been observed (Comeau, 1988; Nicholls et al., 1986). Biphasic uptake, however, was not observed during the batch tests. In reports of biphasic uptake, longer aeration periods were used which could account for the differences in uptake patterns. Although biphasic phosphate uptake was not observed during aeration, a slower rate of accumulation would typically be observed when mechanical aeration stopped. This rate would continue until an anaerobic condition was established and accumulation would stop, suggesting a difference in aerobic and anoxic 66 phosphate uptake. Streichan et al. (1990) studying numerous activated sludge plants demonstrating excess phosphate removal also reported lower anoxic uptake rates. Several possibilities exist to explain the change in uptake rate during anoxic conditions. PHA metabolism which is related to phosphate accumulation could slow and cause the resultant decrease in phosphate accumulation. Because degradation rates of PHA during anoxic periods are similar to those in aerobic conditions in the majority of experiments, the change in phosphate uptake rate is likely due to some change in the operation or nature of the phosphate accumulating system. The enzyme systems and control factors found to be involved in polyphosphate metabolism are numerous and have been extensively studied (Clark et al., 1986; Pepin and Wood, 1986; Rao and Torriani, 1988; Robinson and Wood, 1986; Robinson etal., 1987;Torien et al., 1990). A possibility for the decreased rate of uptake during anoxic conditions may result from the activity of phosphotransferase. The enzyme phosphotransferase has been isolated from a number of organisms demonstrating polyphosphate accumulation (Levinson et al., 1975; Lotter andDubery, 1987; Muhammed, 1961; Muhlradt, 1971; Robinson etal., 1984; Suresh etal., 1985; T'seyenetal., 1985) and appears to be a key enzyme in polyphosphate metabolism in some systems. ADP was found to be an inhibitor of this key enzyme (Romberg et al., 1956). Lower energy generating capability under anoxic conditions would cause a higher ADP/ATP ratio. Thus, the higher ADP could alter phosphotransferase activity causing slower polyphosphate accumulation and phosphate uptake. Lotter and van Der Merwe (1987) demonstrated that some enzymes associated with polyphosphate metabolism were affected by nitrates and that the extent of phosphate accumulation was proportional to enzymatic activity. Alternatively, the change in phosphate uptake rates between aerobic and anoxic conditions may result from the activity of two distinct populations of organisms. During aeration phosphate may be sequestered by polyphosphate accumulating and other organisms. Anoxically, organisms with the capability to metabolize in anoxic conditions, but that do not necessarily have polyphosphate accumulating activity, could be sequestering the phosphate for growth and metabolism. If this were true, the 67 intracellular phosphate profile would be expected to show a decreased polyphosphate pool relative to the other phosphate-containing pools when aeration stopped. Typically, this decrease is not observed, and instead, a slower accumulation of polyphosphates is observed under anoxic conditions. Anaerobically, significant phosphate release was only observed if a source of carbon was added to the sample. Comeau (1988) reports anaerobic phosphate release in the absence of added carbon, but the release may result from the use of continuous versus batch culture systems, in which a continuous and undetectable carbon source could be present. Both the Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) and Mino (Mino et al., 1987) models postulate the released phosphate results from the expulsion of phosphate used for the activation of acetate to acetyl CoA, and has been supported by observations by Doi et al. (1989). The observation that phosphate release stops when acetate is removed supports these assumptions. Comeau (1988) also suggests phosphate is also released to maintain an appropriate membrane potential, and that correlations between phosphate release and acetate uptake can be predicted from his model. Abu-ghararah and Randall (1991) believed the ratios predicted by the Comeau/Wentzel model (Comeau et al., 1985; Wentzel et al., 1986), however, are erroneous because Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) estimated the molar ratios and energy expenditure incorrectly. A range of correlations between the VFA consumed and phosphate released have been established by several groups (Arvin, 1985; Comeau et al., 1985; Fukase et al., 1982; Matsuo et al., 1984; Mino et al., 1987; Somiya et at., 1988; Wentzel et al., 1985; Wentzel et al., 1989). In this series of bench scale batch experiments, calculated molar ratios were highly variable and no correlations between phosphate release and carbon uptake could be established. The lack of correlation may result from the use of radioactive methods for phosphate determination. It is possible that non-radioactive phosphate is released from existing polyphosphate pools, rather than newly synthesized pools, causing the high variability in the calculated ratios. Because of sample 68 treatment, however, the non-radioactive polyphosphate pools are expected to be low compared to the radioactive pools. If cell lysis during anaerobiosis was causing the apparent phosphate release, increase in the phosphate levels in the bulk solution would be expected whether carbon was added or not, and would not alter the overall intracellular phosphate profile. Analysis of intracellular phosphate pools, however, shows that newly synthesized long chain soluble polyphosphates are degraded and cause an increase in extracellular soluble phosphate levels. 31p-NMR studies also demonstrated the loss of a newly synthesized polyphosphate pool. Typically, phosphate release is not observed when acetate is added to the culture near the beginning of the run before aeration. Because any existing polyphosphate pools are non-radioactive, observation of phosphate release is not expected using radioactive methods. Similar observations have been reported by Mino et al. (1985) who could detect initial phosphate changes chemically, but not radioactively following the addition of radioactive label. Phosphate release could also be observed during anoxic acetate addition, and would typically be accompanied by the rapid removal of the added VFA and accumulation of PHA. This response is consistent with others reported for acetate addition during oxic/anoxic conditions (Comeau, 1988; McLaren and Wood, 1976; Wentzel et al., 1989) and supports the prediction of Wentzel et al. (1986) that an anaerobic state is not a prerequisite to obtain phosphate release with acetate addition. Phosphate release was not observed during aerobic carbon addition, however, in bench scale batch studies. The response of phosphate to carbon addition during a typical anoxic period is much different than the phosphate response observed during simultaneous addition of nitrate and acetate to anaerobic samples. During these simultaneous additions, rapid phosphate uptake is initially observed and is usually followed by phosphate release. The rate of phosphate uptake is typically faster than that observed during aeration. Comeau (1988) reported similar initial rates of phosphate uptake when he provided anaerobic sludge with either nitrates or oxygen. 69 Hascoet et al. (1985) also reported phosphate uptake when sludge was in a condition of low carbon with nitrate levels above 10 mg/L. The difference in phosphate response with acetate addition during normal and induced anoxic conditions may result from the previous history of the sludge. Lotteretal. (1986) suggest that anaerobic/oxic cycling stimulates the polyphosphate or PHB accumulating propensities. Cells leaving an aerated state have a large polyphosphate pool and low internal carbon reserves. When they are exposed to acetate in an anoxic state, they have the necessary polyphosphate pools to activate the acetate and rebuild their carbon stores. In the normal operating system, aerobically exposed cells are typically recycled back to an anaerobic condition where acetate is available and the cells rebuild their carbon reserves. Thus, following an aerobic period, phosphate release during anoxic exposure to acetate simulates normal operating conditions. Cells in an anaerobic state, however, will have expended most of their polyphosphate during PHA storage. When these cells are exposed to nitrate and acetate they may not have the necessary polyphosphate pools required to facilitate carbon storage even though acetate is available. Comeau (1988) showed that PHA accumulation is limited by internal polyphosphate reserves. In the normal operating pilot plant system, cells are primed for phosphate uptake following anaerobiosis. Using anaerobic sludge, nitrate and acetate addition caused phosphate uptake initially and simulates the behavior expected for such sludge in the normal operating system. During some simultaneous nitrate and acetate additions, phosphate release and PHA accumulation are observed upon completion of denitrification. The brief period of phosphate uptake may provide the necessary polyphosphate stores to facilitate PHA storage if acetate is still available. From the bench scale batch studies, phosphate release seems most dependent on the availability of a usable carbon source, since phosphate release can be observed under anoxic and anaerobic conditions in the presence of acetate. Phosphate accumulation seems most dependent on the history of the sludge and the presence of oxygen or nitrate. 70 4.1.2. PHA response The anaerobic accumulation and aerobic decrease of PHA observed in the bench scale batch experiments are consistent with reported patterns (Comeau, 1988; Fukase et al., 1982; Hart andMelmed, 1982; Lotter and Dubery, 1989; Mino etal., 1987). In the bench scale batch test, decrease of cellular PHA pools aerobically and anoxically in the absence of carbon is consistent with the presumed role of PHA. Consumption of PHA is believed to supply a source of carbon and energy in conditions where readily biodegradable carbon is normally limiting, and some of this energy is used for synthesis of polyphosphate stores (Comeau et al, 1985; Wentzel et al., 1986). Anoxic decreases in PHA pools suggests that the organisms responsible for PHA metabolism can utilize nitrates as electron acceptors. Some Acinetobacter species, a group of organisms accepted as typical of BEPR organisms (Fuhs and Chen, 1975; Buchan, 1983; Chu, unpublished results; Wentzel et al., 1991), have been shown to reduce nitrate to nitrite (Lotter, 1985; Lotter et al., 1986; van Groenestijn and Deinema, 1985) and even in some cases, nitrite to nitrogen (Lotter, 1985; Lotter et al., 1986). As discussed earlier, no correlations between decrease in PHA and phosphate uptake were found in the bench scale batch testing. Storage of both PHB and PHV were observed anaerobically, anoxically, and aerobically with carbon addition. Interestingly, PHA accumulation could be observed during the early phase of experiments when no acetate was added and no VFA were detectable. Comeau (1988) also reported PHA accumulation in the absence of added or detectable nutrient. This suggests some undetectable nutrient was available and rapidly consumed for the storage of PHA. This nutrient source could be VFA produced by fermentative organisms that the BEPR organisms could use for carbon storage, or may be endogenously derived from higher fatty acids in the sludge. Wallen and Rohwedder (1974) and Schubert et al. (1988) report that the acids that comprise PHA can be produced from higher fatty acids, and Osborn et al. (1986) observed rapid uptake of some long chain fatty acids in the anaerobic zone of a full scale BEPR system. The 71 enzymes responsible for PHA metabolism have been found to be used for fatty acid degradation (Peoples and Sinskey, 1989b; Schubert et al., 1988). These findings may provide an explanation for the observation made in this study and by Comeau (1988) that the molar ratio of net accumulated carbon, calculated from PHB and PHV, to added carbon frequently exceeded 1.0. Turnover of other cellular components could provide additional carbon, resulting in the high ratio, but analysis of changes in the relative amounts of unknown peaks of materials resolved in the gas chromatograph showed no correlation with changes in PHB or PHV. Comeau (1988) reported a greater accumulation of PHV than PHB during PHA accumulation in the absence of added acetate, while PHB accumulation was favored during the beginning phase of the these batch tests, suggesting differences in the nature of the endogenously available carbon. Significant PHA accumulation was typically not observed later in the batch experiments unless carbon was added anoxically or anaerobically, and would be consistent in a system that had consumed most of the readily available carbon during a previous anaerobic and aerobic exposure. In the majority of experiments, PHB and PHV were accumulated at a ratio of 0.6 to 0.75 for acetate addition. Comeau (1988) reported ratios of PHV/PHB of 0.5 or less with acetate additions. Both PHB and PHV are expected, based on the model that was described earlier. Accumulation could continue after detectable VFA were zero implying a metabolic lag between acetate acquisition and storage as PHA, or utilization of some other endogenous nutrient after acetate had been consumed. It is also possible that the presence and metabolism of acetate increases the bioavailability of other endogenous carbon sources, not normally available for PHA storage. During extended anaerobiosis, a 30 to 60 percent decrease in the PHB pool was observed. During this period cells would be nutrient and energy limited and could utilize their carbon stores for survival. Lageveen et al. (1988) and Doi et al. (1989) reported that PHA could be degraded and utilized as a carbon source during starvation, and Tal and Okon (1985) found that cells with large PHB pools would survive better in conditions of stress than those cells without large PHB pools. 72 The most significant PHA accumulation was in anaerobic conditions. The variability in PHA accumulation during simultaneous nitrate and acetate additions may be due to acetate consumption for denitrification in some instances, leaving no acetate for PHA storage. Mostert et al. (1988) demonstrated that substrate utilization is partitioned between the requirements associated with denitrification and observed phosphate release, and Iwema and Meunier (1985) estimated that 4 mg of acetate would be consumed for each mg of nitrate added. If this is the case, no acetate would be available for storage as PHA in the batch tests, and storage would result from endogenous nutrients. Extensive studies with various organisms have established the PHB and PHV metabolic pathways (Doi etal., 1989; Fukuietal., 1976; Fukuietal., 1987; Haywood etal., 1988; Oeding and Schlegel, 1973; Peoples and Sinskey, 1989a; Peoples and Sinskey, 1989b; Slater et al., 1988), and a representative diagram outlining PHB metabolism is presented in Figure 4.1. Both the NADH/NAD ratio and levels of acetyl CoA have been reported to control the activity of various PHA synthesizing enzymes in in vitro studies (Jackson and Dawes, 1976; Lotter and Dubery, 1989; Peoples and Sinskey, 1989b), and Wentzel et al. (1986) proposed that PHB synthesis is triggered by a high NADH/NAD ratio. It is assumed that the high NADH/NAD ratio caused by oxygen limitation rapidly readjusts as PHB synthesis starts, and PHB assumes the role of an alternative electron acceptor, while providing beneficial carbon reserves (Jackson and Dawes, 1976; Page and Knosp, 1989). The possible NADH/NAD control, proposed by Wentzel et al. (1986) will be considered more completely during the discussion of the NADH response. In the bench scale batch tests, PHA accumulation seemed unaffected by the ranges of the NADH/NAD ratio or the external oxidation-reduction potential of the various conditions studied, and most dependent on the availability of acetate, since acetate addition would result in PHA accumulation under anaerobic, aerobic, or normal or nitrate induced anoxic conditions. The availability of nitrogen is reported to influence the accumulation of PHA (Heinzle and Lafferty, 1980; Pageand Knosp, 1989), but doesn't seem to influence accumulation in the batch 73 Acetate ^ A T V P ATP, CoA-SH — AMP, PPi (1) C02 TCA ATP (9) Acetyl-CoA PI Acetyl phosphate CoA-SH Polyphosphate (2) \ Acetoace (3) CoA-SH yl-CoA-•NADPH*H* NADP* D(-)-3-Hydroxybutyrate (5)(6) (A) D(-)-3-Hydroxybutyryl-GoA y1 P^HB CoA-SH Figure 4.1. Schematic pathways of PHB and polyphosphate metabolism (Doi et al., 1989). Enzymes are denoted as follows: (1), acetyl-CoA synthetase; (2), 3-ketothiolase; (3), NADPH-linked acetoacetyl-CoA reductase; (4), PHB synthase; (5), PHB depolymerase; (6), dimer hydrolase; (7), phosphate acetyltransferase; (8), acetate kinase; (9), polyphosphate kinase. tests since accumulation could be observed under various conditions of nitrogen availability. Lotter (1989) reported that the first enzyme of PHB synthesis responds to oxygen levels, but in these and other studies PHA accumulation could be observed under aerobic, anoxic, or anaerobic conditions with acetate addition. The extent of accumulation in each of the conditions, however, may reflect the activity of this enzyme. The fact that PHA accumulation could be observed in all three conditions during acetate addition supports the prediction of Wentzel et al. (1986) that an anaerobic state is not a pre-requisite to obtain acetate storage as PHB. 74 4.1.3. NADH response Nicotinamide nucleotides such as NAD and NADH play a central role in the oxidative reactions of all organisms, and monitoring the intracellular redox potential of these compounds should provide important information regarding patterns of cell metabolism (Chance and Williams, 1956). The levels of these nucleotides are affected by environmental conditions and can affect cellular metabolism. The models of Comeau/Wentzel (Comeau etal., 1985; Wentzel et al., 1986) and Mino (Mino et al., 1987) predict that when the NADH/NAD ratio becomes too high, cellular functions become inhibited. Based on general observations made on pure bacteria growing in quite different circumstances, or on in vitro studies of pure enzyme systems, Wentzel et al. (1986) suggested that PHA accumulation is triggered by a high NADH/NAD ratio. According to these models, the production of PHA from simple carbon sources will consume NADH resulting in a lower NADH/NAD ratio and reverse the NADH-imposed inhibition. Thus, the role of the NADH/NAD ratio becomes an important control factor in the two models. Historically, enzymatic and chemical methods were used to measure the levels of these components in cells. The methods employed extraction techniques which were destructive to the cell and too slow to give reliable results because of the rapid turnover rates of these coenzymes. Additionally, these types of assays suffered from interference in crude sludge samples. The use of a non-destructive, rapid fluorometric technique for measuring NADH provides a better alternative for measuring intracellular redox potential than the traditional methods. This technique allows measurement of the equilibrium of NADH and NAD which are being constantly recycled. Harrison and Chance (1970) applied the fluorometric technique to a cell culture and to monitored changes in the level of reduced nicotinamide nucleotides under various conditions. Since then, the technique has been refined and used to monitor intracellular changes in redox, indicated by changes in the NADH/NAD ratio, under different metabolic 75 conditions (Armiger et al., 1990, Armiger et al., 1990; Peck and Chynoweth, 1990; Srinivas and Mutharasan, 1987). During the bench scale batch tests, NADH was continuously monitored. This type of NADH measurement is affected by both the cell concentration and metabolic state of the cells (Siano and Mutharasan, 1989). Because the growth of organisms in the sludge samples are very slow, cell concentration should be relatively constant and any changes observed in the NADH signal should reflect changes in the metabolic patterns of the cells. Typically, initial NADH readings would be high, and changes in the NFU signal during aeration would be much smaller, suggesting that the range of the measurements was imposed on a high background of fluorescence. As the solids levels increased in the samples, the changes in fluorescence should be more indicative of the changes in the NADH/NAD ratio in the sludge, due to the nature of the fluorescence measurements. During the beginning phase of batch experiments, NADH consistently dropped, suggesting a decrease in the NADH/NAD ratio. This response represents a change in the nutrient state of the sludge during the transition from a continuously fed to a batch culture during sampling (unpublished results), and the activity of the cells. During the transition, acetate and endogenous nutrients become limiting and result in accumulation of NAD and thus a lower NADH/NAD ratio. Based on the diagram in Figure 4.2, NADH generation decreases as carbon availability decreases under these conditions, resulting in a lower ratio. PHA accumulation was also observed during this period. PHA accumulation is believed to be a quasi-fermentation that allows consumption of excess NADH while providing beneficial carbon stores (Jackson and Dawes, 1976). Thus, any NADH that was produced during this period would be oxidized back to NAD. A combination of these two events is believed responsible for the observed initial decrease. Certain metabolic inhibitors have been used to change the NADH/NAD ratio (Betz and Chance, 1965). Cyanide additions, resulting in the accumulation of NADH due to the 76 Figure 4.2. Schematic diagram of NADH/NAD cycling reactions. Adapted from The FluroMeasure System Users Manual, Biochem Technology Inc., 1987. blocking of NADH oxidation with oxygen (Lehninger, 1982), demonstrated that the NADH signal could be restored to the high levels initially observed, dismissing the possibility that the observed decrease was artifactual. Aerobically, cells readily oxidize NADH during energy production. During aeration, the NADH signal dropped rapidly, indicating this sudden change in theNADH/NAD ratio. Similar decreases in fluorescence have been reported during culture shifts to aerobic metabolism (Armiger et al., 1986; Harrison and Chance, 1970; Zabriskie and Humphrey, 1978), while others using traditional techniques showed a similar decrease in the NADH/NAD ratio during transition to aerobic metabolism in several microorganisms (London and Knight, 1966). The NADH signal either stabilized or decreased at a slower rate after this initial drop, demonstrating that the culture had reached a new equilibrium or was establishing one at a lower NADH/NAD ratio. As dissolved oxygen decreased and approached zero, metabolic activity, indicated by culture fluorescence, remained unchanged. Harrison and Chance (1970) reported constant metabolic activity over a range of oxygen levels in a microbial culture, using fluorometric measurements. From the cellular perspective this suggests that high metabolic activity can be maintained at relatively low levels of oxygen saturation. If fluorometric NADH measurements can be used to establish and monitor aerobic metabolism at decreasing levels of oxygen saturation, it would be useful for determining nominal levels of aeration. Typically, dissolved oxygen levels are maintained at certain levels in bioreactors, but this gives no indication of biological activity. If activity could be detected fluorometrically it could be determined if aeration was excessive or sufficient, and altered appropriately, improving efficiency. Anoxically, observed NADH tended to continue to decrease or stabilize, until denitrifica-tion was complete. At the point where nitrates were consumed, observed NADH jumped to a higher level. As oxygen and nitrates are depleted and the cells enter an anaerobic state, the ability to oxidize NADH is limited by the availability of electron acceptors, causing the step 78 increase. This increase in NADH is characteristic in other systems during transition to anaerobiosis (Armiger etal., 1986; Harrison and Chance, 1970; Betzand Chance, 1965;Siano and Mutharasan, 1989;), and corresponds to the v"nitrate knee" observed in the oxidation-reduction potential (Koch and Oldham, 1985). The magnitude of the NADH increase after denitrification was smaller than the change observed during aeration, and did not establish the NADH/NAD equilibrium at the high anaerobic level the sample initially had at the beginning of the batch test. This lower equilibrium may reflect properties of the sludge and the true anaerobic NADH/NAD equilibrium under the sludge conditions. In the pilot plant the anaerobic sludge is in a condition with a continuous carbon supply, resulting in a highly reduced environment with ORP values typically below -200 mV. The initial NADH signal is much higher than any time during a normal batch run, consistent with a more highly reduced system. After aeration in the batch system, very little carbon is available anaerobically and a more oxidized system exists. After denitrification, ORP levels were typically in the 30 to -50 mV range. As a result, the lower anaerobic NADH/NAD equilibrium could reflect the anaerobic equilibrium in the more oxidized conditions. Thus, the anaerobic NFU values established after aeration reflects the true equilibrium in the more oxidized condition and suggests that internal redox is affected by external redox conditions. If sludge is incubated anaerobically for extended periods, the NADH signal climbed back to the high level initially observed. Typically, the oxidation-reduction potential during this time was 100 to 150 mV lower than the levels initially observed during anaerobiosis. Thus, the higher level of NADH observed may reflect the new equilibrium established under the more reduced conditions, similar to those during normal plant operation. These observations also suggest that anaerobiosis can be realized by the cells over a range of oxidation-reduction potentials, and that definition of cellular anaerobiosis has a weak correlation to absolute ORP values. This is further supported by the observation that one test using highly oxidized sludge samples, resulting from plant failure, showed a typical NADH jump response, normally associated with the onset of anaerobiosis, while ORP readings were above normal levels (data not shown). In addition to the more reduced environment, the higher 79 NADH/NAD ratio observed after extended anaerobiosis could also reflect higher levels of NADH produced from the degradation of PHB in the highly stressful conditions. During anaerobic nitrate addition the observed NADH signal instantaneously dropped and remained depressed until denitrification was complete. The observed drop in NADH was not as great as the drop observed when air was introduced into the sludge. The drop may reflect the portion of the sludge population that could utilize nitrate as an electron acceptor. London and Knight (1966) and Wimpenney and Firth (1972) demonstrated that different types of organisms have different levels of nicotinamide coenzymes, and it is possible that the magnitude of change observed in the NADH signal reflects the activity of the denitrifying organisms in the sludge. The time required for the NADH signal to recover to its response prior to nitrate addition was a function of the amount of nitrate added and the solids concentration (Armiger et al., 1991 - in press), suggesting the biological nature of the NADH response. The use of these types of NADH measurements provides a rapid diagnostic procedure for estimating both the size and state of the denitrifying population. This population dependent NADH response may also affect the extent of the nitrate jump typically observed during transition from anoxic to anaerobic conditions, and may reflect the portion of the population affected by anaerobiosis. Carbon addition resulting in PHA accumulation typically caused a decrease in the observed NADH/NAD ratio. Assuming PHA accumulation drains the NADH pool, a lower NADH/ NAD ratio is expected during anaerobic PHA accumulation, and rapid and significant PHA accumulation tended to be accompanied by a faster or greater decrease in the NADH/NAD ratio. The tendency for a rapid or obvious decrease in the NADH/NAD ratio during anaerobic PHA accumulation could be a useful indicator for predicting whether PHA accumulation was occurring in a sludge sample. The models of Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) and Mino (Mino et al., 1987) predict that the NADH/NAD ratio plays a central role in the PHA storage reaction. From the bench scale batch testing it is clear that the NADH/ 80 NAD ratio is not the sole controlling factor, since PHA accumulation could be observed at a variety of NADH/NAD ratios, within a single sample. Accumulation could be observed at lower NADH/NAD ratios than those ratios observed anaerobically, where the NADH/NAD ratio should be highest according to Wentzel etal. (1986). Similarly, degradation of PHA could be realized at higher NADH/NAD ratios than those ratios where accumulation was measured. Because PHA accumulation can be observed in aerobic conditions with acetate addition (Comeau, 1988; Wentzel et al., 1989), the NADH/NAD control mechanism becomes questionable. It is possible that the NADH/NAD ratio has some effect on PHA accumulation, but it is not the primary or only factor influencing the PHA response in these batch studies. Instead, PHA accumulation seemed most affected by carbon availability in these batch tests. The application of fluorometric on-line monitoring of intracellular NADH allowed investigation into the immediate biochemical state of the sludge and overcame many of the problems associated with traditional NADH measurements. The determination of typical patterns of NADH/NAD equilibrium under various conditions provided useful information about the metabolic behavior of the cells and revealed reproducible patterns and features useful for further biochemical testing and development of on-line diagnostic and control strategies. The determination of nominal levels of oxygen, detection of cellular anaerobiosis, assessment of the denitrification capabilities of a sludge, and the trends between anaerobic PHA accumulation and observed fluorescence could all be used to develop various biologically defined operational parameters. The use of NADH as an indicator of metabolic activity for process control is not novel and has been used to control and monitor various fermentation processes, including wastewater treatment (Armiger et al., 1990, Armiger et al., 1990; Peck and Chynoweth, 1990; Srinivas and Mutharasan, 1987). 81 4.2. Biochemical Model Testing and Evaluation Studies of BEPR have clearly shown that acetate is readily accumulated as PHB and PHV. The leading models account for the probable pathways for their synthesis and propose sources for the required NADH. The Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) model predicts that reducing power necessary for PHA storage is derived from anaerobic activity of the TCA/glyoxylate cycle. Although it is generally believed that the TCA and glyoxylate cycles do not function anaerobically, several studies have demonstrated the anaerobic activity of some of the TCA and glyoxylate cycle enzymes (Lotter, 1989; Matsuo, 1985; Osbornetal., 1986). The Mino model (Mino etal., 1987) predicts that necessary reducing power is generated by glycogen catabolism in the Embden-Meyerhof-Parnas or Entner-Duodoroff pathways. This model was based on data from a highly artificial system and Comeau (1988) found low intracellular glycogen levels in the pilot plant sludge. Therefore, the Comeau/ Wentzel model (Comeau et al., 1985; Wentzel et al., 1986) seems more useful for explaining normal pilot plant operating conditions. Neither model, however, accounts for the necessity of PHV synthesis from an even-numbered carbon VFA source or provides direct evidence for the source of necessary reducing equivalents. The Comeau/Wentzel model (Comeau et al., 1985; Wentzel et al., 1986) can be modified to account for PHV by recognizing that succinate would need to be consumed in order to avoid feedback inhibition of the glyoxylate cycle and produce a favorable ratio of NADH to available PHA precursors. In these biochemical studies, the Comeau/Wentzel model (Comeau et al., 1985; Wentzel et al., 1986) has been modified by proposing that the potential inhibitory levels of succinate are reduced by converting the succinate to propionyl CoA, which can be condensed with acetyl CoA and reduced by NADH to make hydroxyvalerate which is then accumulated as PHV. This modified model (Figure 3.16) combines several known metabolic pathways (Bell and Herman, 1967; Comeau et al., 1985; Fernandez-Briera and Garrido-Pertierra, 1988; 82 Herman and Bell, 1970; Hodgson and McGarry, 1968; Jackson and Dawes, 1976; Michal, 1982; Morrison and Peters, 1954; Satoh et al., 1990; Weitzman, 1972; Wentzel et al., 1986) to provide favorable ratios of NADH to available PHA precursors on an on-going basis during anaerobic acetate utilization. The actual pathways utilized would depend on the available combinations of nutrients, but would necessitate the use of the glyoxylate cycle in the presence of acetate. This pathway model assumes that anaerobic PHA synthesis is independent of the source of NADH and that PHA synthesis will proceed as long as there are adequate supplies of PHA precursors and NADH, in favorable proportion. For example, normal acetate fed cells produce NADH by the glyoxylate cycle and must make PHV to maintain the operating cycle, but other combinations of nutrients could use other pathways to generate appropriate levels of NADH to PHA precursors. The different routes employed by different nutrients or nutrient/ inhibitor combinations will result in corresponding types of appropriate PHA and can be readily monitored by looking at the ratio of PHV to PHB. This modified biochemical model was supported by comparing predicted effects of various nutrient and nutrient/inhibitor combina-tions with experimental observations. 4.2.1. Glyoxylate cycle function. A limited number of specific inhibitors were available to test the role of the glyoxylate cycle and various pathways as sources of NADH for anaerobic PHA storage. A number of nutrients or nutrient/inhibitor combinations were used to determine the functional probability of various pathways proposed by the model. These tests suggest that in the normal operating pilot plant system where acetate is fed anaerobically, the glyoxylate cycle does play a central role in generating reducing power for anaerobic PHA storage. Tests that involved combinations of acetate and fluoroacetate, or acetate and a-ketoglutarate impaired PHA storage and produced cellular response patterns atypical for normal anaerobic acetate feeding. Assuming the fluoroacetate and a-ketoglutarate are inhibitory, these results suggest that the principal source 83 of reducing power required by the PHA storage reaction is derived from the metabolism of acetate through the glyoxylate cycle. If the required reducing power was derived from other NADH generating reactions, the presence of these glyoxylate cycle inhibitors should not affect the PHA response pattern observed for anaerobic acetate additon. Combined addition of pyruvate and acetate produced a PHA accumulation response pattern much different that obtained for acetate feeding, and suggests that appropriate reducing power for PHA storage and PHA precursor molecules could be provided by converting pyruvate to acetyl CoA. The higher proportion of PHV relative to PHB may result from the production of propionyl CoA from the pyruvate. Simultaneous additions of acetate, pyruvate, and fluoroacetate produced a PHA response pattern similar to that obtained from addition of pyruvate with acetate. This response is consistent with an NADH generating pathway to drive PHA synthesis that does not rely on anaerobic acetate metabolism through the glyoxylate cycle. It was unclear whether the atypical PHA response pattern observed during addition of acetate and fluoroacetate resulted from inhibition of active glyoxylate cycle enzymes or PHA synthesizing enzymes. These results suggest that the fluoroacetate does not impair the function of the PHA synthesizing enzymes, since PHA accumulation is observed with combined addition of fluoroacetate, acetate and pyruvate, and further supports the role of anaerobic acetate metabolism through the gyloxylate cycle during anaerobic acetate feeding, suggested by the previous tests with combinations of acetate and fluoroacetate. The simultaneous addition of pyruvate, acetate, and fluoroacetate demonstrates that the PHA storage reaction can derive necessary reducing power from an alternate source. Both sets of tests demonstrate the ability of the cells to utilize a variety of strategies to meet the specific requirement for appropriate reducing equivalents, while maintaining a favorable ratio of reducing equivalents to PHA precursors in conditions of anaerobic PHA storage. 84 Additions of citrate or combinations of citrate, acetate, and malonate produced PHA response patterns consistent with those expected for a system dependent on the generation of reducing power through anaerobic metabolism of substrates through the glyoxylate cycle. Although citrate is a chelating agent that may reduce the concentrations of necessary cations (Sawyer and McCarty, 1978) implicated in the BEPR process (Comeau, 1988; van Groenestijn et al., 1988), possible chelating activity didn't affect the process because a good PHA response occurs in the presence of citrate and acetate. Additions of succinate or combinations of succinate and acetate, or succinate and malonate suggest that a pathway or pathways for propionyl CoA production from succinate exist in the BEPR system since there in an increase in both the amount of PHA and the ratio of PHV/PHB in the presence of succinate. This analysis is complicated since succinate should block NADH generation because of feedback inhibition. The absence of this block is presumably due to the interaction of succinate with the unknown nutrients already present in the sludge. 85 4.3. General Conclusions From Bench Scale and Biochemical Tests From the bench scale and biochemical tests, two major findings can be reported. It appears that the NADH/NAD ratio presumed by Wentzel, et al. (1986) to trigger PHA accumulation, is not the major controlling element. The study was not designed to provide information about the other numerous controls implicated, and neither the bench scale batch or biochemical tests provided evidence for these control mechanisms, although one factor affecting the PHA response did appear to be carbon availability. PHA accumulation was largely unaffected by the NADH/NAD ratios observed in these studies and PHA accumulation could be observed over a wide range of NADH/NAD ratios in oxic, anoxic, and anaerobic conditions. Experimental observations from the crude system support the central role of the glyoxylate cycle for anaerobic acetate utilization proposed by Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986). The most convincing evidence for this comes from the series of experiments using monofluoroacetic acid, which demonstrated that the early TCA/glyoxylate cycle enzymes are active, and that anaerobic acetate metabolism through the glyoxylate cycle is important for generating necessary reducing power for PHA accumulation. In addition to these major findings, the results from the study using a combination of pyruvate, acetate, and monofluoroacetic acid suggest that sludge is not committed to an NADH source for PHA accumulation through a glyoxylate cycle mechanism. Instead, it appears that any pathway that can generate the necessary reducing equivalents, while maintaining the correct carbon and NADH balance, can drive the PHA accumulation reaction. It does not have to rely exclusively on the glyoxylate cycle pathways, or the Embden-Meyerhof-Parnas and Entner-Duodoroff pathways as proposed respectively by Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986) and Mino et al. (Minoetal., 1987). Based on the normal nutrients available to the biomass, however, itis likely that the glyoxylate cycle plays a central role in the anaerobic accumulation of PHA. The various nutrient studies provided information regarding their possible anaerobic metabolism, and how combinations of nutrients can affect the sludge response. If the proposed biochemical model 86 can be more thoroughly tested, these nutrient combination studies may provide useful information that could allow optimization of BEPR systems. Possibly, by providing some nutrient that can alter the balances or the bioavailability of other compounds in the sludge, the PHA storage reaction can be optimized. The results reported here indicate several applications for on-line measurement of intracellular reduction states which may also be applied, with further study, for optimization of BEPR processes. The preliminary conclusions based on this work require further study and verification, but do provide some background and framework for more intense future study with pure culture strains , and with continuous sludge cultures. 4.4. Limitations of the Data The observations and preliminary conclusions from the tests are limited by both the nature of the samples and the assumptions made regarding the biochemical model testing. Because of the heterogeneous and complex nature of the sludge, sample consistency could not be assured. Several steps were taken to minimize some of the variability. Samples were collected around the same time each sampling day, and related experiments were done within a few days of each other, usually one a day. Because the testing was done in a batch mode, and the normal operating system is continuous, results from this study may not be applicable for the normal operating system, despite the observation that BEPR sludge responded in a similar patterns as would be expected in the continuous system. The assumptions that the major metabolic activity, observed in the biochemical model testing, was limited to the pathways outlined in the model is obviously erroneous, but it does provide a framework for testing and analysis. Similarly, suggesting that the observed responses were caused solely by organisms responsible for the BEPR process is a generalization. Instead, it is likely that the observed responses represent the gross activity of several groups of organisms affecting different aspects of the overall response. Pure culture testing with mutant isolates is required to confirm the presumptions made regarding the biochemical model testing. The effects of the interactions of the numerous microorganisms in 87 the sludge community obviously play a role in the BEPR sludge behavior, but could not be addressed because of the complexity of the problem. Analysis of storage products was limited to PHB and PHV, without regard for intracellular carbohydrates, such as glycogen, and the utilization of added nutrients, other than acetate, could not be confirmed since measurements following their uptake were not done. Finally, the proposed biochemical model is based on metabolic pathways reported in several different groups of microorganisms, and overlooks some of the assumptions of previous models, such as balancing of the membrane potential and the effect of various cations on the process. Discussion regarding the possible factors controlling or influencing the response of various components is speculative, since these tests were not designed to investigate the numerous control mechanisms implicated in the literature. Despite the many limitations of the data, however, the observations provide some evidence for the proposed model within the highly complex environment and limited capabilities of the tests. 88 5. CONCLUSION The objective of this thesis was to gain a better understanding of the microbial mechanisms involved in the BEPR process by determining the behavior of key cellular components under various environmental conditions, and testing a biochemical model to determine the role of the glyoxylate cycle in the process. Applications of novel techniques like the use of 3 2 P cell labelling and molecular fractionation, and the use of a fluorometirc NADH probe allowed investigation of the immediate biochemical state of the cell components and metabolic activity. Labelling cells with 32p demonstrated that phosphate is accumulated aerobically as two main polyphosphate forms, one of which may be involved in activation of acetate anaerobically. The anaerobic PHA storage reaction seemed most dependent on carbon availability, and less affected by the NADH/NAD ratio, contrary to previous assumptions. The development and testing of a biochemical model which accounted for the production of reducing power and the two principal carbon storage products, PHB and PHV, provided experimental evidence addressing the unresolved question regarding the source of reducing power required for anaerobic PHA storage. Testing the biochemical model suggested that necessary reducing equivalents required for anaerobic PHA storage can come from different sources depending on the nutrient composition and availability, but in the normal operating U.B.C. pilot plant system the anaerobic acetate metabolism through the gyloxylate cycle provides this source, as proposed by Comeau/Wentzel (Comeau et al., 1985; Wentzel et al., 1986). Further study with pure culture and continuous sludge systems is required and will advance current understanding by providing more conclusive evidence for the microbial mechanisms responsible for the biological excess phosphate removal phenomenon. 89 REFERENCES Abu-ghararah, Z.H., and C.W. Randall. 1991. The effect of organic compounds on biological phosphorus removal. Wat. Sci. Tech. 23, 585-594. American Public Health Association (A.P.H.A.), American Water Works Association (A.W.W.A.), and Water Pollution Control Federation (W.P.C.F.). 1985. Standard methods for the exzmination of water and wastewater, 16th ed. A.P.H.A., Washington, D.C. Armiger W.B., J.F. Forro, L.M. Montalvo, and J.F. Lee. 1986. The interpretation of on-line process measurements of intracellular NADH in fermentation processes. Chem. Eng. Commun. 45, 197-206. Armiger, W.B., J.F. Lee, W. Roshong, and J. Doutrich. 1990. Keeping a constant eye on plant performance. Operations Forum 7, 30-32. Armiger, W.B., G. Lee, K. Bordacs, and K.D. Tracy. 1990. Automated on-line strategy for the optimization of nutrient removal processes, in Instrumentation, control, and automation of water and wastewater treatment and transport systems, Proceedings of the fifth IAWPRC workshop in Yokohama and Kyoto, Japan, 26 July - 3 August, ed. R. Briggs, Pergammon Press, Oxford, 187-194. Armiger, W.B., J.F. Lee, S. Maneshin, andX. Yang. 1991. Control of process performance by on-line monitoring of biological activity (in press, Proc. WPCF 64th conf.,Toronto, Canada). Arvin, E. 1985. Biological removal of phosphorus from waste water. CRC Critical Rev. Environmental Control 15, 24-64. Beatty, J.T. 1980. Biosynthetic and bioenergetic functions of citric acid cycle reactions in photosynthetic bacteria. Ph.D. dissertation, Indiana University, U.S.A. Bell, E.J., and N. J. Herman. 1967. Effect of succinate on isocitrate lyase synthesis in Mima polymorpha. J. Bacteriol. 93, 2020-2021. Betz, A., and B. Chance. 1965. Influence of inhibitors and temperature on the oscillation of reduced pyridine nucleotides in yeast cells. Arch. Biocem. Biophys. 109, 579-584. Biochem Technology Inc. 1987. FluroMesaure system users manual. Biochem Technology Inc., Malvern, PA, USA. Buchan, L. 1981. The location and nature of accumulated phosphorus in seven sludges from activated sludge plants which exhibited enhanced phosphorus removal. Water SA 7, 1-7. 90 Buchan, L. 1983. Possible biological mechanism of phosphorus removal. Wat. Sci. Tech. 15, 87-103. Canviro Consultants Ltd., N.W. Schmidtke and Associates Ltd., and D.I. Jenkins and Associates Inc. 1986. Retrofitting municipal wastewater treatment plants for enhanced biological phosphorus removal. Report Environmental Protection Service, 3/UP/3. Cartier, P.H. 1968. Dosage des pyridine nucleotides oxydes et reduits dans le sang et les tissus animaux. Eur. J. Biochem. 4, 247-255. Chance, B., and G.R. Williams. 1956. The resipiratory chain and oxidative phosphorylation. Advan. Enzymol. 17, 65-134. Chu, A. 1990. A microbial study of the enhanced biological phosphorus removal process. M.Sc. thesis, University of British Columbia, Canada. Clark, J.E., H. Beegen, andH.G. Wood. 1986. Isolation of in tact chains of polyphosphate from Propionibacterium shermanii grown on glucose or lactate. J. Bacteriol. 168, 1212-1219. Comeau, Y., K.J. Hall, R.E.W. Hancock, and W.K. Oldham. 1985. Biochemical model for enhanced biological phosphorus removal. Proc. B.C. Conference on new methods and research and waste treatment residuals management, June, Vancouver, Canada. [Pub. in Water Research, 20, 1511-1521 (1986)]. Comeau, Y. 1988. The role of carbon storage in phosphate removal from wastewater. Ph.D. thesis, University of British Columbia, Canada. Deinema, M.H., L. Habets, J. Scholten, E. Turkstra, and H. Webers. 1980. The accumulation of polyphosphate in Acinetobacter spp. FEMS Microbiol. Lett. 9, 275-279. Doi, Y., M. Kunioka, Y. Nakamura, and K. Soga. 1987. Biosynthesis of copolyesters in Alcaligenes eutrophus H16 from 13c_iaDei;e(i acetate and propionate. Macromolecules 20, 2988-2991. Doi, Y., Y. Kawaguchi, Y. Nakamura, and M. Kunioka. 1989. Nuclear magnetic resonance studies of poly-(3)-hydroxybutyrate and polyphosphate metabolism in Alcaligenes eutrophus. Appl. Environ. Microbiol. 55, 2932-2938. Donald, L.J., G.F. Molgat, and H.W. Duckworth. 1989. Cloning, sequencing, and expression of the gene for NADH-sensitive citrate synthase of Pseudomonas aeruginosa. J. Bacteriol. 171, 5542-5550. Duysens, L.N.M.,andJ. Amez. 1957. Fluorescence spectrometry of reduced phosphopyridine nucleotide in intact cells in the near-ultraviolet and visible region. Biochem. Biophys. Acta 24, 19-26. 91 Fernandez-Briera, A., and A. Garrido-Pertierra. 1988. A degradation pathway of propionate in Salmonella typhimurium LT-2. Biochimie 70, 757-768. Fuhs, G.W., and M. Chen. 1975. Microbial basis of phosphate removal in the activated sludge process for the treatment of wastewater. Microbial Ecol. 2, 119-138. Fukase, T., M. Shibata, and X. Mijayi. 1982. Studies on the mechansism of biological phosphorus removal. Japan J. Water Pollut. Res. 5, 309-317. Fukui, T., A. Yoshimoto, M. Matsumoto, S. Hosokawa, T. Saito, H. Nishikawa, and K. Tomita. 1976. Enzymatic synthesis of poly-jS-hydroxybutyrate in Zoogloea ramigera. Arch. Microbiol. 110, 149-156. Fukui, T., M. Ito, T. Saito, and K. Tomita. 1987. Purification and characterization of NADP-linked acetoacetyl-CoA reductase from Zoogloea ramigera I-16-M. Biochim. Biophys. Acta 917, 365-371. Groenestijn, J. W. van., and M.H. Deinema. 1985. Effects of cultural conditions on phosphate accumulation and release by Acinetobacter strain 210A. Proc. Int. Conf. Management Strategies for Phosphorus in the Environment, Lisbon, Portugal, 405-410. Groenestijn, J.W. van., G.J.F.M. Vlekke, D.M.E. Anink, M.H. Deinema, and A.J.B. Zehnder. 1988. Role of cations in accumulation and release of phosphate by Acinetobacter strain 210A. Appl. Environ. Microbiol. 54, 2894-2901. Groenestijn, J.W. van., M.M. Bentvelsen, M.H. Deinema, and A.J.B. Zehnder. 1989. Polyphosphate-degrading enzymes in Acinetobacter spp. and activated sludge. Appl. Environ. Microbiol. 55, 219-223. Harrison, D.E.F., and B. Chance. 1970. Fluorimetric technique for monitoring changes in the level of reduced nicotinamide nucleotides in continuous cultures of microorgansims. Appl. Microbiol. 19, 446-450. Hart, M.A., and L.N. Melmed. 1982. Microbiology of nutrient removing activated sludge. Wat. Sci. Tech. 14, 1501-1502. Hascoet, M.C., andM. Florentz. 1985. Influence of nitrates on biological phosphorus removal from wastewater. Water SA 11, 1-8. Hascoet, M.C., M. Florentz, and P. Granger. 1985. Biochemical aspects of enhanced biological phosphorus removal from wastewater. Wat. Sci. Tech. 17, 23-41. Haywood, G.W., A.J. Anderson, L.Chu, and E.A. Dawes. 1988. Characterization of two 3-ketothiolases possessing differing substrate specificities in thepolyhydroxyalkanoate sythesizing organism Alcaligenes eutrophus. FEMS Microbiol. Lett. 52, 91-96. 92 Heinzle, E., and R.M. Lafferty. 1980. A kinetic model for growth and synthesis of poly-/3-hydroxybutyric acid (PHB) in Alcaligenes eutrophus H16. Eur. J. Appl. Microbiol. Biotechnol. 11, 8-16. Herman, N.J., andEJ. Bell. 1970. Metabolic control in Acinetobacter sp. I. Effect of C4 versus C2 and C3 substrates on isocitrate lyase synthesis. Can. J. Microbiol. 16, 769-774. Hill, W.E., L.D. Benefield, and S.R. Jing. 1989. 31p-NMR spectroscopy characterization of polyphosphates in activated sludge exhibiting enhanced phosphorus removal. Wat. Res. 23, 1177-1181. Hochster, R.M., and J.H. Quastel. 1963. Metabolic inhibitors-a comprehensive treatise. Vol. I. Academic Press, New York. Hodgson, B., and J.D. McGarry. 1968. A direct pathway for the metabolism of propionate in cell extracts from Moraxella Iwoffi. Biochem. J. 107, 19-28. Iwema, A., and A. Meunier. 1985. Influence of nitrate on acetic acid induced biological phosphate removal. Wat. Sci. Tech. 17, 289-294. Jackson, F. A., and E. A. Dawes. 1976. Regulation of the tricarboxylic acid cycle and poly-/3-hydroxybutyrate metabolism in Azotobacter beijerinckii grown under nitrogen or oxygen limitation. J. Gen. Microbiol. 97, 303-312. Jones, C.W. 1988. Membrane-associated energy conservation in bacteria: a general introduc-tion, in Bacterial energy transduction, ed. C. Anthony. Academic Press, London. Koch, F.A., and W.K. Oldham. 1985. Oxidation-reduction potential - a tool for monitoring, control and optimization of biological nutrient removal systems. Wat. Sci. Tech. 17,259-281. Kornberg, A., S.R. Kornberg, and E.S. Simms. 1956. Metaphosphate synthesis by an enzyme from Escherichia coli. Biochem. Biophys. Acta 20, 215-227. Lageveen, R.G., G.W. Huisman, H. Preusting, P. Ketelaar, G. Eggink, andB. Witholt. 1988. Formation of polyesters by Pseudomonas oleovorans: Effect of substrates on formation and composition of poly-0R)-3-hydroxyalkanoates and poly-(R)-3-hydroxyalkenoates. Appl. Environ. Microbiol. 54, 2924-2932. Lehninger, A.L. 1982. Principles of biochemistry. Worth Publishers Inc., New York. Levinson, S.L., L.H. Jacobs, T.A. Krulwich, andH.C. Li. 1975. Purification and character-ization of a polyphosphate kinase from Arthrobacter atrocyaneus. J. Gen. Microbiol. 88, 65-74. 93 London, J., and M. Knight. 1966. Concentrations of nicotinamide nucleotide coenzymes in micro-organisms. J. Gen. Microbiol. 44, 241-254. Lotter, L.H. 1985. The roleof bacterial phosphate metabolism in enhanced phosphorus removal from the activated sludge process. Wat. Sci. Tech. 17, 127-138. Lotter, L.H., M.C. Wentzel, R.E. Loewenthal, G.A. Ekama, and G.v.R. Marais. 1986. A study of selected characteristics of Acinetobacter spp. isolated from activated sludge in anaerobic/anoxic/aerobic and aerobic systems. Water SA 12, 203-208. Lotter, L.H., M.C. Wentzel, G.A. Ekama, and G.v.R. Marais. 1986. An investigation into the heterotrophic bacterial population of various activated sludge plants. Water SA (manuscript in preparation). Lotter, L.H., 1987. Preliminary observations on polyhydroxybutyrate metabolism in the activated sludge process. Water SA 13, 189-191. Lotter, L.H., and LA. Dubery. 1987. Metabolic control in polyphosphate accumulating bacteria and its role in enhanced biological phosphate removal, in Advances Water Pollution Control. Biological Phosphate Removal from Wastewaters, ed. R. Ramadori, Pergamon Press, Oxford. Lotter, L.H., and E.H.M. van Der Merwe. 1987. The activities of some fermentation enzymes in activated sludge and their relationship to enhanced phosphorus removal. Wat. Res. 21,1307-1310. Lotter, L.H., and LA. Dubery. 1989. Metabolic regulation of /3-hydroxybutyrate dehydroge-nase in Acinetobacter calcoaceticus var. Iwoffi. Water SA 15 , 65-70. Lotter, L.H. 1989. Two year study on the enhancement of biological phosphorus removal by altering process feed composition (metabolic control mechanisms). Report to the Water Research Commission of South Africa, WRC 137/3/89. Manahan, S.E. 1984. Environmental Chemistry, 4th ed., Brooks/Cole Publishing Co. Monterey, CA., USA. Matsuo, Y., A. Miya, and A. Watanabe. 1984. Experimental studies on the anaerobic reaction of the activated sludge by the anaerobic aerobic process. J. Jap. Sewage Wks. Ass. 21,11-22. Matsuo, Y. 1985. Functioning of the TCA cycle under anaerobic conditions in the anaerobic-aerobic acclimatized activated sludge. Proc. 40th Annu. Conf. Jap. Soc. Civ. Engrs. 40, 989. McLaren, A.R., and R.J. Wood. 1976. Effective phosphorus removal from sewage by biological means. Water SA 2, 47-50. 94 Michal, G. 1982. Biochemical pathways. Boehringer Mannheim GmbH - Biochemica. W. Germany. Mino, T., T. Kawakami, and T. Matsuo. 1985. Behavior of intracellular polyphosphate in biological removal process. Wat. Sci. Tech. 17, 11-21. Mino, T., V. Arun, Y. Tsuzuki, and T. Matsuo. 1987. Effect of phosphorus accumulation on acetate metabolism in the biological phosphorus removal process, in Advances Water Pollution Control. Biological Phosphate Removal from Wastewaters. Ed. R. Ramadori, Pergammon Press, Oxford. Morrison, J.F., and R. A. Peters. 1954. Biochemistry of fluoroacetate poisoning: the effect of fluorocitrate and purified aconitase. Biochem J. 58, 473-479. Morrison, K. 1988. An assessment of the feasibility of biological phosphorus removal in Canadian wastewater treatment plants. M.A.Sc. thesis. University of British Columbia, Canada. Morse, D., and H.W. Duckworth. 1980. A comparison of the citrate synthases of Escherichia coli and Acinetobacter anitratum. Can. J. Biochem. 58, 696-706. Mostert, E.S., A. Gerber, and C.J.J. Van Riet. 1988. Fatty acid utilisation by sludge from full-scale nutrient removal plants, with special reference to the role of nitrate. Water SA 14, 179-184. Muhammed, A. 1961. Studies on biosynthesis of polymetaphosphate by an enzyme from Corynebacterium xerosis. Biochem. Biophys. Acta 54, 121-132. Muhlradt, P.F. 1971. Synthesis of high molecular weight polyphosphate with a partially purified enzyme from Salmonella. J. Gen. Microbiol. 68, 115-122. Murphy, M., and L.H. Lotter. 1986. The effect of acetate on polyphosphate formation and degradation in activated sludge with particular reference to Acinetobacter calcoaceticus: A microscopic study. Water SA 12, 63-66. Nicholls, H.A., D.W. Osborn, and A.R. Pitman. 1986. Biological phosphorus removal at the Johannesburg Northern and Goudkoppies wastewater purification plants. Water SA 12, 13-18. Oeding, V.,andH.G. Schlegel. 1973. 0-ketothiolase from Hydrogenomonas eutropharl\6 and its significance in the regulation of poly-/3-hydroxybutyrate metabolism. Biochim. J. 134,239-248. Ohtake, H., K. Takahashi, Y. Tsuzuki, and K. Toda. 1985. Uptake and release of phosphate by a pure culture of Acinetobacter calcoaceticus. Wat. Res. 19, 1587-1594. 95 Osborn, D.W., L.H. Lotter, A.R. Pitman, and H.A. Nicholls. 1986. Enhancement of biological phosphate removal by altering process feed composition. Report to the Water Research Commission of South Africa: WRC 137/1/86. Page, W.J., and O. Knosp. 1989. Hyperproduction of poly-/3-hydroxybutyrate during exponential growth of Azotobacter vinelandii UWD. Appl. Environ. Microbiol. 55, 1334-1339. Peck, M.W., and D.P. Chynoweth. 1990. On-line monitoring of the methanogenic fermen-tation by measurement of culture fluorescence. Biotechnol. Lett. 12, 17-22. Peoples, O.P., and A.J. Sinskey. 1989b. Poly-j3-hydroxybutyrate biosynthesis in Alcaligenes eutrophus HI6: Characterization of the genes encoding /3-ketothiolase and acetoacetyl-CoA reductase. J. Biol. Chem. 264, 15293-15297. Peoples, O.P., and A.J. Sinskey. 1989a. Pory-/?-hydroxybutyrate (PHB) biosynthesis in Alcaligenes eutrophus HI6; Identification and characterization of the PHB polymerase gene (phbC). J. Biol. Chem. 264, 15298-15303. Pepin, C.A., and H.G. Wood. 1986. Polyphosphate glucokinase from Propionibacterium shermanii: Kinetics and demonstration that the mechanisms involves both processive and nonprocessive type reactions. J. Biol. Chem. 261, 4476-4480. Poindexter, J.S., and L.F. Eley. 1983. Combined procedure for assays of poly-/3-hydroxybu-tyric acid and inorganic polyphosphate. J. Microbial Methods 1, 1-17. Ramos, S., and H.R. Kaback. 1977. The relationship between the electrochemical proton gradient and active transport in Escherichia co/z'membrane vesicles. Biochem. 16, 854-859. Ramsay, B.A., J.A. Ramsay, amdD.G. Cooper. 1989. Production of Poly-jS-hydroxyalkanoic acid by Pseudomaonas cepacia. Appl. Environ. Microbiol. 55, 584-589. Rao, N.N., and A. Torriani. 1988. Utilization by Escherichia coli of high-molecular-weight linear polyphosphate: Roles of polyphosphatases and pore proteins. J. Bacteriol. 170, 5216-5223. Robinson, N.A., N.H. Goss, and H.G. Wood. 1984. Polyphosphate kinase from Propionibacterium shermanii: Formation of an enzymatically active insoluble complex with basic proteins and characterisation of synthesized polyphosphate. Biochem. Int. 8, 757-769. Robinson, N.A., and H.G. Wood. 1986. Polyphosphate kinase from Propionibacterium shermanii: Demonstration that the synthesis and utilization of polyphosphate is by a processive mechanism. J. Biol. Chem. 261, 4481-4485. 96 Robinson, N.A., J.E. Clark, and H.G. Wood. 1987. Polyphosphate kinase from Propionibacterium shermanii: Demonstration that polyphosphates are primers and determina-tion of the size of the synthesized polyphosphate. J. Biol. Chem. 262, 5216-5222. Rosen, B.P., and S. Silver. 1987. Ion transport in prokaryotes. Academic Press Inc., New York. Satoh, H., T. Mino, and T. Matsuo. 1990. Biochemical mechanisms of anaerobic glutamate uptake in the enhanced biological phosphorus removal process. Poster session at the 15th Biennial Conference of the International Association on Water Pollution Reasearch and Control, held in Kyoto, Japan, 29 July-3 August 1990. Sawyer, C.N., and P.L. McCarty. 1978. Chemistry for environmental engineering, 3rd ed. McGraw Hill Book Co., New York. Schubert, P., A. Steinbuchel, and H.G. Schlegel. 1988. Cloning of the Alcaligenes eutrophus genes for synthesis of poly-/3-hydroxybutyric acid (PHB) and synthesis of PHB in Escherichia coli. J. Bacteriol. 170, 5837-5847. Seiler, J. 1969. Inorganic ions, in Thin Layer Chromatography - a laboratory handbook, 2nd ed. Ed. E. Stahl, Springer-Verlag, New York. Senior, P.J., G.A. Beech, G.A.F. Ritchie, and E.A. Dawes. 1972. The role of oxygen limitation in the formation of poly-/3-hydroxybutyrate during batch and continuous culture of Azotobacterbeijerinckii. Biochem. J. 128, 1193-1201. Siano, S.A., and R. Mutharasan. 1989. NADH and flavin fluorescence responses of starved yeast cultures to substrate additions. Biotechnol. Bioeng. 34, 660-670. Slater, S., W.H. Voige, and D.E. Dennis. 1988. Cloning and expression in Escherichia coli of int Alcaligenes eutrophus H16 poly-j8-hydroxybutyrate biosynthetic pathway. J. Bacteriol. 170,4431-4436. Somiya, I., H. Tsuno, and M. Matsumoto. 1988. Phosphorus release-storage reaction and organic substrate behavior in biological phosphorus removal. Wat. Res. 22, 49-58. Srinath, E.G., C.A. Sastry, and S.C. Pillai. 1959. Rapid removal of phosphorus from sewage by activated sludge. Experimentia 15, 339-340. Srinivas, S.P., andR. Mutharasan. 1987. Culture fluorescence characteristics and its metabolic significance in batch cultures of Clostridium acetobutylicum. Biotechnol. Lett. 9, 139-142. Streichan, M., J.R. Golecki, and G. Schon. 1990. Polyphosphate-accumulatingbacteria from sewage plants with different processes for biological phosphorus removal. FEMS Microbiol. Ecol. 73, 113-124. 97 Suresh, N., R. Warburg, M. Timmerman, J. Wells, M. Coccia, M.F. Roberts, and H.O. Halvorson. 1985. New strategies for the isolation of microorganisms responsible for phosphate accumulation. Wat. Sci. Tech. 17, 99-111. Tal, S., and Y. Okon. 1985. Production of the reserve material poly-/3-hydroxybutyrate and its function in Azospirillum brasilense Cd. Can. J. Microbiol.31, 608-613. Toerien, D.F., A. Gerber, L.H. Lotter, and T.E. Cloete. 1990. Enhanced biological phosphorus removal in activated sludge systems, in Advances in Microbial Ecology, Vol. II, ed. K.C. Marshall. Plenum Press, New York. T'seyen, J., D. Malnou, J.C. Block, and G. Faup. 1985. Polyphosphate kinase activity during phosphate uptake by bacteria. Wat. Sci. Tech. 17, 43-56. Vasiliadis, G., A. Duncan, R.C. Bayly, and J.W. May. 1990. Polyphosphate production by strains of Acinetobacter. FEMS Microbiol. Lett. 70, 37-40. Wallen, L.L. , and W.K. Rohwedder. 1974. Poly-/3-hydroxyalkanoate from activated sludge. Environ. Sci. Technol. 8, 576-579. Weitzman, P.D.J. 1972. Regulation of a-ketoglutarate dehydrogenase activity in Acinetobacter. FEBS Lett. 22, 323-326. Wentzel, M.C., P.L. Dold, G.A. Ekama, and G.v.R. Marais. 1985. Kinetics of biological release. Wat. Sci. Tech. 17, 57-71. Wentzel, M.C., L.H. Lotter, R.E. Loewenthal, and G.v.R. Marais. 1986. Metabolic behavior of Acinetobacterspp. in enhanced biological phosphorus removal - a biochemical model. Water SA 12, 209-224. Wentzel, M.C., P.L. Dold, R.E. Loewenthal, G.A. Ekama, and G.v.R. Marais. 1989. Enhanced polyphosphate organism cultures in activated sludge systems. Part II: Experimental behavior. Water SA 15, 71-88. Wentzel, M.C., L.H. Lotter, G.A. Ekama, R.E. Loewenthal, and G.v.R. Marais. 1991. Evalutaion of biochemical models for biological excess phosphorus removal. Wat. Sci. Tech. 23, 567-576. Wentzel, M.C., G.A. Ekama, and G.v.R. Marais. 1991. Kinetics of nitrification denitrification biological excess phosphorus removal systems - a review. Wat. Sci. Tech. 23, 555-565. Wiechers, H.N.S. 1987. Guidelines for chemical phosphate removal from municipal waste waters. Water Research Commission, Pretoria, S.A. 98 Willard, H.H., L.L. Merritt, J.A. Dean, and F.A. Settle, Jr. 1988. Instrumental methods of analysis, 7th ed. Wadaworth Publishing Co., Belmont, CA., USA. Wimpenney, J.W.T., and A. Firth. 1972. Levels of nicotinamide adenine dinucleotide and reduced nicotinamide adenine dinucleotide in facultative bacteria and the effect of oxygen. J. Bacteriol. I l l , 24-32. Zabriskie, D.W., and A.E. Humphrey. 1978. Estimation of fermentation biomass concentra-tion by measuring culture fluorescence. Appl. Environ. Microbiol. 35, 337-343. 99 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098610/manifest

Comment

Related Items