Enrichment, Isolation and Characterization of Phenazine-1-Carboxylic Acid (PCA)-Degrading Bacteria Under Aerobic and Anaerobic Conditions

Home , Imidazolonepropionase, Phenazine, Tyrosinase

Enrichment, isolation and characterization of phenazine-1-carboxylic acid (PCA)-degrading bacteria under aerobic and anaerobic conditions

Miaomiao Zhang

A thesis in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Civil and Environmental Engineering

Faculty of Engineering

September, 2018

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: ZHANG

First name: Miaomiao Other name/s:

Abbreviation for degree as given in the University calendar: PhD

School: Civil and Environmental Engineering Faculty: Engineering

Title: Enrichment, isolation and characterization of phenazine-1- carboxylic acid (PCA)-degrading bacteria under aerobic and anaerobic conditions

Abstract

Phenazines are a large class of nitrogen-containing aromatic heterocyclic compounds produced and secreted by bacteria from phylogenetically diverse taxa under aerobic and anaerobic conditions. Phenazine-1-carboxylic acid (PCA) is regarded as a ‘core’ phenazine because it is transformed to other phenazine derivatives. Due to their important roles in ecological fitness, biocontrol of plant pathogens, infection in cystic fibrosis and potential in anticancer treatments, understanding the fate of phenazine compounds is prudent. Only seven bacterial species are known to degrade phenazines and all of them are aerobic. Hence, the aim of this study is to enrich, isolate and characterize additional bacteria with the ability to degrade phenazines aerobically and anaerobically.

In this study, the isolation of a PCA-degrading Rhodanobacter sp. PCA2 belonging to Grammaproteobacteria is reported. Characterization studies revealed that strain PCA2 is also capable of transforming other phenazines including phenazine, pyocyanin and 1-hydroxyphenazine. The sequencing, annotation and analysis of the genome of strain PCA2 revealed that genes (ubiD and the homolog of the MFORT_16269 gene) involved in PCA degradation were plasmid borne. Studies on abundance and expression of the homolog of MFORT_16269 gene via qPCR and RT-qPCR showed its involvement in PCA degradation by strain PCA2. Furthermore, results from LC-MS analysis together with proteomics indicated that strain PCA2 degraded PCA via decarboxylation and cleavage of aromatic and nitrogen-containing rings, potentially catalysed by UbiD, UbiX, phenylpropionate dioxygenase, biphenyl-2,3-diol 1,2-dioxygenase, amidohydrolase and nitroreductase. Three intermediates including phenazine, (4Z)-2-hydroxy-5- {[(1Z)-6-(hydroxyamino)cyclohexa -2,4-dien-1-ylidene]carbamoyl} penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine are reported.

This study also reports the isolation of the first known anaerobic PCA degrading bacterium. Enterobacteriaceae were highly enriched after anaerobic incubation with PCA as a carbon, nitrogen and energy source and iron(III) as a terminal electron acceptor. Ultimately a Morganella morganii strain was isolated and shown to oxidise PCA whilst reducing iron(III).

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).e

29/06/2018 ………………………………..…… ……………………………………..…… ……….……………………....… Signature Witness Signature Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

Abstract

In this study, the isolation of a PCA-degrading Rhodanobacter sp. PCA2 belonging to Grammaproteobacteria is reported. Characterization studies revealed that strain PCA2 is also capable of transforming other phenazines including phenazine, pyocyanin and 1- hydroxyphenazine. The sequencing, annotation and analysis of the genome of strain PCA2 revealed that genes (ubiD and the homolog of the MFORT_16269 gene) involved in PCA degradation were plasmid borne. Studies on abundance and expression of the homolog of MFORT_16269 gene via qPCR and RT-qPCR showed its involvement in PCA degradation by strain PCA2. Furthermore, results from LC-MS analysis together with proteomics indicated that strain PCA2 degraded PCA via decarboxylation and cleavage of aromatic and nitrogen-containing rings, potentially catalysed by UbiD, UbiX, phenylpropionate dioxygenase, biphenyl-2,3-diol 1,2-dioxygenase, amidohydrolase and nitroreductase. Three intermediates including phenazine, (4Z)-2- hydroxy-5-{[(1Z)-6-(hydroxyamino)cyclohexa -2,4-dien-1-ylidene]carbamoyl} penta- 2,4-dienoic acid (HCCPD) and phenylhydroxylamine are reported.

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

29/06/2018 Date …………………………………… ………......

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only).

I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.’

Signed ……………………………………………......

29/06/2018 Date ……………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

29/06/2018 Date ……………………………………………......

III

Acknowledgements

I would like to express my first and foremost thanks to my supervisor Prof. Mike Manefield for giving me the opportunity to work in his group, as well as for his great guidance and resources throughout the last few years. Thank you for providing me with the rewarding research experience to work towards and learn to be a professional scientist. Also, I must acknowledge him for allowing me the suitable degree of scientific freedom. My deepest appreciation goes out to you for your time and supervision that made my PhD study to be a professional and enjoyable journey.

I would also like to take this chance to acknowledge Dr. Sabrina Beckmann for her continued guidance as a co-supervisor during my early half and two years and offering me opportunities to work on neutral red crystal. I am grateful to have been guided by her in practical components of anaerobic microbiology study including culturing methods, gas chromatography analysis and fluorescence microscopy. Last but not least, thanks for listing me as a co-author for the publication on the high impact factor journal Energy and Environmental Science.

I am deeply thankful to Dr. Önder Kimyon for his friendship and sharing with me his knowledge in lab and beyond. Many thanks for all the help, advices and encouragement you have offered to me. Our interesting conversations on research, politics, cultures, and others are good times that put great colours in my life in UNSW. Also, thank you for ordering and keeping our lab stocked up. I wish to express my sincere appreciation to Prof. Haluk Ertan for your time and help in genomic and proteomic analyses. Thank you for sharing your knowledge and passion in science and life with me. Many thanks to Dr. Anna Yeung for her patience and huge help in my English writing. Your sunshine personality was inspiring to me. Many thanks to Dr. Russell Pickford and Lewis Adler for their assistances in mass spectrometry in BMSF- UNSW; Dr. Ling Zhong and Dr. Bat-Erdene Jugder for their guidance and help in proteomic analysis.

I must also sincerely thank everyone in the Manefield group that has made this project possible and my PhD experience colourful. These lovely people include: Dr. Sihui Tang, for being an amazing friend. Thank you for helping me settle down in the lab and making me feel much better in Sydney. Miriam Kronen, for your friendship, laughs and fun conversation. Sophie Holland, for your friendship, sunshine personality IV

and support in safety work in PC2 lab. Miao Hu, for your jokes, friendship and delicious Chinese food. Dr. Matthew Lee, for his support in gas chromatography analyses and advices in scientific research. Also many hugs and thanks all my friends I have met throughout my time in Australia including Assoc. Prof. Akifumi Hosoda, Dr. Zack Jones, Dr. Mukan Ji, Dr. Yie Kuan Wong, Dr. Ricardo Alfan, Dr. Thi Anh Thy, Dr. Hangwei Hu, Dr. Lu Yang, Dr. Tianzhe Liu, Priyanka Srivastava, James Bevington, Gan Liang and others I may have forgotten.

Special thanks to the Chinese Scholarship Council and UNSW for funding and other assistances throughout my PhD study. I would also like to acknowledge all staff in BABS and CVEN in UNSW that has supported my study in the university.

Most importantly, I would like to thank and hug my parents and brother. Without your unconditional love, care, encouragement and support I would not have made it. I would like to dedicate this thesis to my lovely family. I would also like to thank my boyfriend Fei Liu, whose infinite support and love are always beyond expectation and imagination.

Table of Contents

Abstract…...... I

ORIGINALITY STATEMENT ...... II

AUTHENTICITY STATEMENT ...... III

Acknowledgements ...... IV

List of Figures ...... XI

List of Tables ...... XVI

Chapter 1 Introduction ...... 1

1.1 Synthesis of phenazines ...... 2

1.1.1 Biosynthesis of phenazines ...... 2

1.1.2 Chemical synthesis of phenazines ...... 4

1.2 Importance of phenazines ...... 6

1.2.1 Role in ecological fitness ...... 6

1.2.2 Role in human health ...... 7

1.2.3 Phenazines as pollutants ...... 8

1.3 Removal of phenazines ...... 8

1.3.1 Non-biological removal ...... 9

1.3.2 Biological degradation ...... 10

1.4 Research objectives ...... 16

1.5 Overview of the dissertation ...... 17

References...... 19

Chapter 2 Isolation and characterization of phenazine-1-carboxylic acid (PCA)- degrading bacterium Rhodanobacter sp. PCA2 ...... 27

2.1 Introduction ...... 27

2.2 Materials and methods ...... 28

2.2.1 Inoculum source and cultivation conditions ...... 28

2.2.2 Enrichment and isolation of PCA-degrading bacterium ...... 29

2.2.3 PCA degradation with different concentration gradients of PCA ...... 29

2.2.4 Other phenazine compounds and carbon sources range study ...... 30

2.2.5 Analytical methods of phenazines ...... 30

2.2.6 Microscopic analyses ...... 30

2.2.7 Identity of isolate based on 16S rRNA sequence ...... 31

2.2.8 Proof of purity by Denaturing Gradient Gel Electrophoresis (DGGE)...... 31

2.2.9 Detection of ubiD and the homolog of MFORT_16269 gene by PCR ...... 32

2.2.10 Phylogenetic characterization ...... 32

2.2.11 Statistical analysis ...... 33

2.3 Results ...... 33

2.3.1 Enrichment, isolation and phylogenetic characterization of Rhodanobacter sp. PCA2 ...... 33

2.3.2 Growth linked degradation of PCA to Rhodanobacter sp. PCA2 ...... 35

2.3.3 Effect of PCA concentration on rate of degradation ...... 36

2.3.4 Utilization of other phenazines and C sources ...... 38

2.3.5 Detection of ubiD and the homolog of MFORT_16269 gene ...... 39

2.4 Discussion ...... 41

References...... 45

Chapter 3 Genome characterization and transcription analysis of the homolog of MFORT_16269 gene in Rhodanobacter sp. PCA2 ...... 49

3.1 Introduction ...... 49

3.2 Materials and methods ...... 50

3.2.1 Genomic DNA, RNA and plasmid extraction ...... 50

3.2.2 Genome sequencing and assembly...... 51

3.2.3 Genome annotation and analysis ...... 52

VII

3.2.4 Detection of ubiD, the homolog of MFORT_16269 and 16S rRNA genes 52

3.2.5 Quantitative PCR (qPCR) and Reverse-transcript qPCR (RT-qPCR) of the homolog of MFORT_16269 ...... 52

3.3 Results ...... 53

3.3.1 Investigation of the homolog of MFORT_16269 in strain PCA2...... 53

3.3.2 General properties of genome of strain PCA2 ...... 54

3.3.2.1 Genes encoding UbiD and UbiX...... 58

3.3.2.2 Homologs of MFORT_16269 gene encoding phenylpropionate dioxygenase ...... 58

3.3.2.3 Genes encoding amidohydrolase and pyocyanin demethylase ...... 59

3.3.3 Detection of ubiD and the homolog of MFORT_16269 from plasmid in strain PCA2 ...... 60

3.4 Discussion ...... 61

References...... 65

Chapter 4 Metabolomic and proteomic investigations of PCA degradation by Rhodanobacter sp. PCA2 ...... 69

4.1 Introduction ...... 69

4.2 Materials and methods ...... 70

4.2.1 Cultivation of Rhodanobacter sp. PCA2 ...... 70

4.2.2 Determination of PCA...... 72

4.2.3 Determination of growth by microscopy ...... 72

4.2.4 LC-MS analysis ...... 72

4.2.5 Sample preparation for mass spectrometry ...... 72

4.2.6 Nano LC-MS/MS and label-free proteomics analysis ...... 73

4.2.7 Statistical analysis ...... 74

4.3 Results ...... 74

4.3.1 Determination of intermediates and products of PCA degradation ...... 74

VIII

4.3.2 PCA degradation and cell yields in PCA-PCA and Pyr-PCA cultures ...... 78

4.3.3 Proteomics profiles in PCA and Pyr cultures ...... 79

4.4 Discussion ...... 81

Supplementary tables ...... 86

References...... 120

Chapter 5 Enrichment and isolation of phenazine-1-carboxylic acid (PCA)- degrading bacteria under anaerobic conditions ...... 123

5.1 Introduction ...... 123

5.2 Materials and methods ...... 124

5.2.1 Inocula source and cultivation conditions ...... 124

5.2.2 Enrichment of anaerobic PCA-degrading bacterium ...... 125

5.2.3 Isolation of anaerobic PCA-degrading bacterium ...... 125

5.2.4 Analytical methods...... 126

5.2.4.1 PCA analysis ...... 126

5.2.4.2 Acetate and methanol analysis ...... 126

5.2.4.3 Lactate analysis ...... 127

5.2.4.4 Iron(III) and iron(II) analysis ...... 127

5.2.4.5 Sulphate and sulphide analysis ...... 128

5.2.4.6 Nitrate and nitrite analysis ...... 128

5.2.5 Genomic DNA extraction ...... 129

5.2.6 Phylogenetic characterization ...... 129

5.2.7 Microscopic analyses ...... 129

5.2.8 Community analysis ...... 129

5.2.9 Statistical analysis ...... 130

5.3 Results ...... 130

5.3.1 PCA degradation in anaerobic sludge based microcosms with different electron acceptors and donors ...... 130

5.3.2 Enrichment of PCA-degrading bacteria with iron(III) as electron acceptor…...... 134

5.3.3 Community analysis of PCA-degrading enrichment cultures with iron(III) as electron acceptor ...... 136

5.3.4 Isolation of an anaerobic PCA-degrading bacterium ...... 137

5.3.5 Morphological and phylogenetic analysis of Morganella morganii SL11138

5.3.6 Growth and iron reduction linked PCA degradation by strain SL11 ...... 139

5.4 Discussion ...... 143

References...... 146

Chapter 6 Concluding remarks and future perspectives ...... 149

6.1 Summary of findings ...... 149

6.2 Significance and future functional characterizations of Rhodanobacter sp. PCA2...... 152

6.3 Significance and future molecular characterizations of Rhodanobacter sp. PCA2……………………………………………………………………………….. 152

6.4 Significance and future functional characterizations of Morganella morganii SL11...... 153

6.5 Concluding remarks ...... 154

References...... 155

Appendix… ...... 157

List of Figures

Figure 1.1 A collection of phenazines and the derivatives. Phenazine-1,6-dicarboxylic acid (PDC) and phenazine-1-carboxylic acid (PCA) are produced by all phenazine- producing bacteria. Phenazine, 1,6-dihydroxyphenazine and pyocyanin are synthesized by Pseudomonas, and griseolutein and PD 166,152 are produced by Streptomyces...... 1 Figure 1.2 Proposed general biosynthetic pathway of two phenazine precursors, phenazine-1-carboxylic acid (PCA) and phenazine-1,6-dicarboxylic acid (PDC)...... 3 Figure 1.3 Overview of the history of chemical synthesis of phenazines...... 5 Figure 1.4 The chemical degradation of PCA to pyrazine via oxidations of alkaline permanganate and decarboxylations...... 9 Figure 1.5 A neighbour-joining tree of near full-length 16S rRNA gene sequence of phenazine-degrading bacteria reported. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are accession numbers from NCBI or IMG. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.) ...... 11 Figure 1.6 Proposed pathway for the metabolism of PCA by Sphingomonas wittichii strain DP58 via ring cleavage. The compounds in brackets were not detected and the compounds in square dotted boxed were tentatively identified...... 12 Figure 1.7 Pattern of PCA conversion via dioxygenation catalyzed by PCA 1,2- dioxygenase (PcaA1A2A3A4) from strain Sphingomonas sp. DP58. Red, reduction; ox, oxidation...... 13 Figure 1.8 Initial steps in phenazine degradation via dioxygenation catalyzed by biphenyl 2,3-dioxygenase (BphA1fA2fA3fA4f) in strain Sphingobium yanoikuyae B1. Red, reduction; ox, oxidation...... 14 Figure 1.9 Proposed pathway of neutral red degradation via ring cleavage and deamination by white-rot fungus Perenniporia subacida...... 16 Figure 2.1 Degradation of PCA in three rounds of enrichment. The metabolic degradation rate of PCA was enhanced during enrichment, suggesting PCA-degrading bacteria was successfully enriched. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 33

Figure 2.2 Colonies on agar plates (A) amended with PCA after incubation for 3 days, and denaturing gradient gel electrophoresis (DGGE) profile (B) of the picked colony. Only one band was observed, indicating the PCA-degrading strain was purified...... 34 Figure 2.3 A neighbour-joining tree of near full-length 16S rRNA gene sequence of Rhodanobacter sp. PCA2 (in bold), along with other Rhodanobacter strains and other phenazines-degrading bacteria. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are GenBank accession numbers from NCBI or IMG. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.) ...... 35 Figure 2.4 Degradation of PCA and cell yield of Rhodanobacter sp. PCA2. Growth of strain PCA2 was linked with metabolic PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 36 Figure 2.5 PCA degradation by Rhodanobacter sp. PCA2 at different concentrations of PCA. The highest rate of PCA degradation was detected in cultures fed with 500 mg/L PCA. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 37 Figure 2.6 Cell yield of Rhodanobacter sp. PCA2 in cultures with different concentrations of PCA. The highest increase of cell concentration was detected in cultures fed with 500 mg/L PCA, which was consist with profiles of rate of PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 37 Figure 2.7 Degradation of phenazine compounds (solid line) and cell yield (dash line) of Rhodanobacter sp. PCA2 in cultures. The highest rate of PCA degradation and cell yield of strain PCA2 was observed in cultures fed with phenazine, followed by cultures fed with pyocyanin and 1-hydroxyphenazine, respectively. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 38 Figure 2.8 Cell yield of Rhodanobacter sp. PCA2 in cultures with different C sources. Strain PCA2 can utilize pyruvate, glucose, alanine and trehalose as the sole C source, with pyruvate supporting the highest cell yield. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 39

XII

Figure 3.1 Quantification of the homolog of MFORT_16269 and its transcript copies and cell yield of strain PCA2 throughout PCA degradation. Increases of gene copies and transcript copies of the homolog of MFORT_16269 were 10.6 and 28.4 fold more than the corresponding cell yield respectively, suggesting that gene transcription increased during PCA degradation and that the gene copy number was higher than the cell number suggestive of multiple copies. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 54 Figure 3.2 Agarose gel electrophoresis profiles of ubiD and the homolog of MFORT_16269 amplified from purified plasmid extract. Partial 16S rRNA gene was amplified as a control to check the purity of plasmid extract. It showed that both ubiD and the homolog of MFORT_16269 were plasmid born in strain PCA2...... 61 Figure 4.1 Schematics of cultivation of Rhodanobacter sp. PCA2 in mineral medium fed with PCA and pyruvate respectively as carbon source for investigation of inducibility (A) and proteomic assay (B)...... 71 Figure 4.2 LC-MS chromatograms of metabolites in samples PCA- 0h (A), 12h (B), 24 (C), 36h (D) and 48h (E). Phenazine, (4Z)-2-hydroxy-5-{[(1Z)-6-(hydroxyamino) cyclohexa-2,4-dien-1-ylidene]carbamoyl}penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine were detected as the intermediates of PCA degradation by strain PCA2...... 77 Figure 4.3 Proposed pathway of PCA degradation via decarboxylation and ring cleavage by Rhodanobacter sp. PCA2...... 77 Figure 4.4 Profiles of PCA degradation (solid line) and cell growth (dash line) in daughter cultures “PCA-PCA” () and “Pyr-PCA” () which were transferred from parent cultures fed with PCA and pyruvate as carbon and energy source, respectively. Compared to PCA-PCA cultures (6h), Pyr-PCA cultures experienced a longer lag phase (12h), indicating the inducibility of PCA degradation by strain PCA2. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 78 Figure 4.5 Proposed enzymatic pathway of PCA degradation via decarboxylation and cleavage of aromatic and nitrogen-containing rings by strain PCA2. The compounds in brackets were not detected in this study. Proteins and genes (in italic) involved in PCA degradation are depicted beside the arrows. *, ubiD and hacA2 (the homolog of MFORT_16269) genes are plasmid born in strain PCA2. All enzymes except amidohydrolase encoded by a gene with locus tag Ga0192503_12145 are predicted

XIII

from partial genome of strain CT6. All locus tags of genes described here are based on JGI database...... 85 Figure 5.1 PCA fate in anaerobic sludge based microcosms with different electron acceptors. Only microcosms amended with iron(III) citrate showed anaerobic degradation of PCA, indicating ferric iron was associated with PCA loss in the cultures with sludge. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 131 Figure 5.2 Reduction of electron acceptors including iron(III) (A), nitrate (B) and sulphate (C) in anaerobic sludge-based microcosms. Iron(III) was significantly reduced to iron(II). Nitrate and sulphate reduction were observed, which presumably caused by oxidation of reduced organic carbon incumbent in the sludge inoculum.Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 132 Figure 5.3 PCA fate in anaerobic sludge based microcosms with different electron donors. No PCA loss was observed in these cultures amended with electron donors. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures...... 133 Figure 5.4 Consumption of electron donors including acetate, methanol and lactate in anaerobic sludge-based microcosms. Decreases in concentrations of acetate, methanol and lactate were detected. Data points are averages of triplicate cultures and error bars represent the standard error of mean of triplicate cultures...... 134 Figure 5.5 The anaerobic degradation of PCA in the presence () or absence () of iron(III) and in the sterile cultures () as control through three rounds of enrichment. The degradation rate of PCA was enhanced during enrichment, suggesting PCA- degrading bacteria were successfully enriched. Data points are averages of triplicate cultures and error bars (0.39 ~ 24.34) represent standard error of mean of the triplicate cultures...... 135 Figure 5.6 The transformation of iron(III) to iron(II) in anaerobic microcosms through the three rounds of enrichment. Iron(III) was significantly reduced to iron(II) in the enriched PCA-degrading cultures, suggesting iron(III) was associated with PCA consumption. Data points are averages of triplicate cultures and error bars (0.05 ~4.58) represent standard error of mean of the triplicate cultures...... 135 Figure 5.7 Bacterial community composition on level of family based on 16S rRNA gene sequences in anaerobic microcosms with iron(III) added throughout three rounds

XIV

of enrichment. The Enterobacteriaceae family was dramatically enriched in the PCA- degrading cultures with iron(III) as electron acceptor. The right graph showed relative abundance of genera belonging to family Enterobacteriaceae in microcosms with iron(III) added from the third round of enrichment. Amongst the fourteen genera detected here, the Morganella genus was overwhelmingly dominant. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 137 Figure 5.8 Image of the isolated anaerobic PCA-degrading Morganella morganii strain SL11 observed after Gram staining (1000x magnification). It showed that strain Morganella morganii SL11 is a rod-shaped, Gram-negative bacterium...... 138 Figure 5.9 A neighbour-joining tree of near full-length 16S rRNA gene sequence of Morganella morganii SL11 (in bold), along with other Morganella strains and other phenazine-degrading bacteria. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are GenBank accession numbers from NCBI. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.) ...... 139 Figure 5.10 Degradation of PCA (solid line) and cell yield (dash line) in pure M. morganii strain SL11 cultures fed with PCA in presence () or absence () of iron(III) and in abiotic cultures as control (). Growth of strain SL11 was linked with metabolic PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 140 Figure 5.11 Profiles of iron(III) and iron(II) in anaerobic cultures inoculated with Morganella morganii strain SL11 and abiotic cultures as sterile control. Iron(III) was reduced to iron(II) in the cultures inoculated with PCA-degrading strain SL11. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures...... 141 Figure 5.12 Linear correlation between the amount of PCA degraded and the decreasing amount of iron(III) (A), the increasing amount of iron(II) (B) and the increasing cell yield of strain SL11 before stationary phase (C). Degradation of PCA significantly correlated with iron(III) reduction to iron(II) and the cell yield of strain SL11 in the pure cultures. The solid line represents the best fit based on linear regression of average of triplicate data points...... 142

List of Tables

Table 2.1 Sequences producing significant alignments with the ubiD gene amplified from genome of Rhodanobacter sp. PCA2 using BLAST analysis in NCBI ...... 40 Table 2.2 Sequences producing significant alignments with the MFORT_16269 gene amplified from genome of Rhodanobacter sp. PCA2 using BLAST analysis in NCBI . 40 Table 3.1 Comparison in general features of the genomes of Rhodanobacter sp. PCA2, Mycobacterium fortuitum CT6, Sphingomonas wittichii DP58 and Rhodanobacter denitrificans 2APBS1 ...... 55 Table 3.2 Comparison of genes assigned with COG functional categories in Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58, Mycobacterium fortuitum CT6 and Rhodanobacter denitrificans 2APBS1...... 56 Table 3.3 Genes encoding UbiD and UbiX identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6...... 58 Table 3.4 Phenylpropionate dioxygenases encoded by the homologs of MFORT_16269 identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6...... 59 Table 3.5 Homologs of genes encoding amidohydrolase and pyocyanin demethylase identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6...... 60 Table 4.1 Detectable proteins potentially involved in PCA degradation based on metabolic and proteomic analyses ...... 80 Table S1 The proteins expressed both in PCA and Pyr cultures ...... 86 Table S2 The proteins exclusively expressed in PCA cultures ...... 99 Table S3 The proteins exclusively expressed in Pyr cultures ...... 109

XVI

Chapter 1 Introduction

Phenazines (Figure 1.1) are a large class of nitrogen-containing aromatic heterocyclomers which can accept and donate two electrons according to the redox properties of nearby compounds (Price-Whelan et al., 2006). Phenazines have attracted scientific interest because of their colourful pigmentation and broad-spectrum antibiotic activity. Fordos (1859) reported the isolation of the first known phenazine compound, pyocyanin (Figure 1.1), from the dressings of a purulent human wound with chemical synthesis following almost 30 years later (Ris, 1886). Over the past century, more than 6,000 phenazine derivatives with substituted groups have been reported with wide- ranging bioactivities (Laursen & Nielsen, 2004). Most of them are produced and secreted by bacteria from phylogenetically diverse taxa and are extensively studied because of their effects not only on other surrounding microorganisms, but also on plants (Mavrodi et al., 2006). Additionally, a number of studies have characterized phenazines as having negative impacts on human health, acting as virulence and survival factors in some diseases such as cystic fibrosis related infections (Ran et al., 2003, Lau et al., 2004, Caldwell et al., 2009). Therefore, insights into the fate of phenazines are crucial in environmental protection as well as ecology and human health.

This chapter introduces the origins of phenazines including biological and chemical synthesis pathways. The roles of phenazines with respect to ecology and environmental and human health are then reviewed including methods for removal from the environment. Finally, the research objectives and an outline of the thesis is summarized at the end of the chapter.

1.1 Synthesis of phenazines

1.1.1 Biosynthesis of phenazines

Natural phenazines are colorful secondary metabolites predominantly synthesized and excreted by bacteria. The source was first identified as being Pseudomonas aeruginosa by Carle (1882) and since then most Pseudomonas species such as P. aeruginosa, P. aureofaciens, P. fluorescens and P. cepacia were found to produce phenazines which are mostly substituted with simple hydroxyl- and carboxyl- groups, for example, phenazine-1,6-dicarboxylic acid and 1,6-dihydroxyphenazine (Figure 1.1) (Turner & Messenger, 1986, Budzikiewicz, 1993, Laursen & Nielsen, 2004). For several years, Pseudomonas was regarded as the only bacterial genus to perform de novo synthesis of phenazine compounds. However, a complex phenazine griseolutein (Figure 1.1) was isolated from Streptomyces P37 in 1950 (Umezawa et al., 1950). It was then revealed that the Gram-positive actinomycetes Streptomyces (e.g., S. griseolutein, S. luteogriseus, S. antibioticus and S. prunicolor) synthesized not only the same simple phenazines as Pseudomonas but also some more complex phenazine compounds which often contain one or more C-substituents on the core structure of phenazines, for example, PD 116,152 (Figure 1.1) (Laursen & Nielsen, 2004). To date, phenazine production has been discovered in a range of bacteria, including Sorangium species (Hollstein & Butler, 1972), Burkholderia phenazinium (Turner & Messenger, 1986), Pelagiobacter variabilis (Imamura et al., 1997) and Erwinia herbicola (Giddens et al., 2002). The first and to date unique reported phenazine-producing archaea is Methanosarcina mazei Gӧ1 which synthesizes a lipid-like methanophenazine on cytoplasmic membranes via a significantly different pathway from the bacterial biosynthesis (Abken et al., 1998).

Phenazine biosynthesis precursors and pathways have now been studied extensively. It is generally accepted that the shikimate pathway is the primary biosynthesis pathway with chorismate as an intermediate (Figure 1.2) (Ingledew & Campbell, 1969, Hollstein et al., 1978, Laursen & Nielsen, 2004, Mavrodi et al., 2006). Phenazine-1,6- dicarboxylic acid (PDC) and phenazine-1-carboxylic acid (PCA) are common precursors which could be transformed into other more complex phenazines and derivatives via N-oxidation, decarboxylation, hydroxylation and methylation in different phenazine-producing cells (Hollstein & Marshall, 1972, Mavrodi et al., 2006, Mentel et al., 2009). Alternatively, Land et al. (1993) predicted aminodehydroquinic acid as a precursor which could generate 5-amoni-5-deoxyshikimic acid which is transformed to phenazine-1,6-dicarboxylic acid via dehydration, self-condensation and oxidation. This biosynthesis model remains to be validated.

Figure 1.2 Proposed general biosynthetic pathway of two phenazine precursors, phenazine-1-carboxylic acid (PCA) and phenazine-1,6-dicarboxylic acid (PDC) (Ingledew & Campbell, 1969, Hollstein et al., 1978, Laursen & Nielsen, 2004, Mavrodi et al., 2006).

Genes and crystal structures of the enzymes directly responsible for biosynthesis of core phenazines PCA and PDC have been characterised, resulting in a detailed synthesis pathway with underlying chemistry. Generally, a seven-gene phz-operon (containing genes phzA, phzB, phzC, phzD, phzE, phzF and phzG) is directly involved in the biosynthesis of phenazines with evidence to suggest this operon has spread through horizontal gene transfer (Pierson & Thomashow, 1992, Mavrodi et al., 1998,

Fitzpatrick, 2009, Mavrodi et al., 2010, Blankenfeldt et al., 2013). Five enzymes, namely PhzB, PhzD, PhzE, PhzF and PhzG, control the conversion from chorismate into PCA and PDC (Blankenfeldt et al., 2013). Typically, PCA- and PDC-producing bacteria can be distinguished based on the relative activities of the two enzymes, PhzB and PhzG. Larger amounts of hexahydrophenazine-1,6-dicarboxylic acid (HHPDC) could accumulate when the activity of PhzB is higher, resulting in the uncatalyzed oxidative decarboxylation to 5,10-dihydrophenazine-1-carboxylic acid (DHPCA) outcompeting the oxidation by PhzG, hence, PCA is produced. In contrast, HHPDC in cells could be oxidized immediately when the relative activity of PhzB is lower, leading to generation of PDC, together with smaller amounts of PCA (Rui et al., 2012,

Blankenfeldt et al., 2013). Additionally, phzC encoding a 3-deoxy-D- arabinoheptulosonate-7-phosphate (DAHP) synthase catalyses the first step of the shikimate pathway in chorismate biosynthesis supplying sufficient precursor for phenazine biosynthesis. phzA is a second copy of gene phzB in Pseudomonads, and PhzA shares approximately 70 % similarity with PhzB but lacks the histidine/serine diad, leading to inability to perform the second condensation step in phenazine synthesis (McDonald et al., 2001). The function of PhzA remains unknown. The phz-operon is not observed in the genome of the unique archaeal phenazine producer Methanosarcina mazei Gö1, suggesting that phenazine biosynthesis could follow a significantly different pathway in archaea (Abken et al., 1998, Deppenmeier et al., 2002).

1.1.2 Chemical synthesis of phenazines

Phenazines are also produced synthetically, including the condensation procedure (Ris, 1886), Wohl-Aue procedure (Wohl & Aue, 1901), Bamberger-Ham procedure (Bamberger & Ham, 1911), Beirut reaction (Haddadin & Issidorides, 1965), diphenylamine procedure (Vivian & Hartwell, 1953, Challand et al., 1970), palladium catalysis (Wolfe & Buchwald, 1997, Hartwig, 1998, Emoto et al., 2000), the electrochemical pathway (Hosseiny Davarani et al., 2008) and Pd-Ag binary nanocluster catalysis (Seth et al., 2016). Some common chemical approaches for phenazines synthesis are summarized in Figure 1.3. However, almost all chemical methods of phenazine synthesis suffer problems of low yield, hazardous by-products, harsh reaction conditions, and severe limitations to localization and the electrochemical character of the substituents (Laursen & Nielsen, 2004).

Figure 1.3 Overview of the history of chemical synthesis of phenazines.

1.2 Importance of phenazines

Due to their special physio-chemical properties, especially oxidation-reduction (redox) properties, phenazines have important effects on microbes and plants, leading to their roles in ecology and human health. Also, their colorimetric properties facilitate application in diverse areas such as dyes and pH indicators, which may cause environmental issues with improper disposal practices.

1.2.1 Role in ecological fitness

Aromatic structures allow phenazines to accept and donate electrons according to the redox potential of other molecules in the surroundings, and to play critical roles in ecological fitness for their producers and interactions between the producers and microbes, plants and animals.

Phenazines as antibiotics benefit their producers by inhibiting the competitors via modes of action including polynucleotide interaction, topoisomerase inhibition, radical scavenging and charge transfer (Hassan & Fridovich, 1980, Laursen & Nielsen, 2004). For instance, three phenazine derivatives phenazinolins A, B and C isolated from Streptomyces diastaticus YIM DT26 exhibited appreciable antibiotic effects and cytotoxicity to Bacillus subtilis, Staphylococcus aureus, Aspergillus niger and Botrytis cinerea, with minimum inhibitory concentration (MIC) values in the range of 12 - 27 µM (Ding et al., 2011). Moreover, the phenazine pyocyanin influences the formation of biofilms which protects bacteria within from environmental challenges, external antibiotic infiltration and attacks from the host immune system (Stewart & Franklin, 2008, Chang, 2017). Compared to the wild-type of Pseudomonas aeruginosa PA14, its pyocyanin knockout (ΔphzA‐G) mutant significantly reduce biofilm formation, which is attributed to reduced extracellular DNA (eDNA) release caused by lack of pyocyanin (Das et al., 2016). Pyocyanin degradation catalyzed by a tautomerizing demethylase (PodA) results in inhibition of biofilm formation by Pseudomonas aeruginosa (Costa et al., 2017). Additionally, 2-hydroxy-phenazine-1-carboxylic acid and PCA might enhance cellular adhesion and growth within P. aeruginosa biofilms, respectively (Maddula et al., 2008).

Phenazines also played important roles in interactions between their producers and microbial communities in the environment. It was found that pyocyanin affected

microbial communities in an oil-degrading culture, resulting in a decrease in the diversity of the microbial community and decreased degradation of polycyclic aromatic hydrocarbons (Norman et al., 2004).

Interest in interactions between phenazines as biocontrol compounds and plants first emerged in the 1980s. Phenazines produced and excreted by soil-borne bacteria aided the survival of plants via inhibition of a broad spectrum of fungal pathogens and induction of systemic resistance in plants (Chin-A-Woeng et al., 2003, Vleesschauwer et al., 2006). PCA and 1-hydroxyphenazine purified from Streptomyces griseoluteus P510 showed activity against six plant pathogens including Fusarium graminearum, Phyricularia grisea, Alternatia solani, F. oxysporium and Sclerotinia sclerotiorum and the Stevia leaf spot disease pathogen (Luo et al., 2015). Additionally, pyocyanin production by P. aeruginosa was highly correlated with a lower incidence of yeast infections (Lau et al., 2004, Cogen et al., 2008). The biocide shenqinmycin, with PCA as the main active compound, is widely applied in southern China for rice protection (Zhao et al., 2017).

1.2.2 Role in human health

Most effects of phenazines in pathogenesis are attributed primarily to their generation of reactive oxygen species (ROS) which contributes to the virulence of an infection and host invasion (Pierson & Pierson, 2010). Early research on cellular respiratory chains revealed that phenazines uncoupled oxidative phosphorylation as an electron shunt in mammalian cells (Armstrong & Stewart-Tull, 1971). Recently, it was revealed that pyocyanin produced by P. aeruginosa played key roles in pathogenesis of human lung infections (Lau et al., 2004, Rada & Leto, 2009). When P. aeruginosa grew on airway epithelial cells, pyocyanin and PCA produced could be reduced by the oxidation of glutathione and NADH, and ROS was then generated via the reaction between reduced pyocyanin and free oxygen in the lungs (Pierson & Pierson, 2010). In addition, to defend against bacterial intruders, airway epithelial cells generated lactoperoxidase and related dual oxidases which produced mild oxidants, which, are negated by pyocyanin via competing dual oxidase activity for NADPH with epithelial cells (Rada & Leto, 2009). Moreover, ROS generated by phenazines might contribute to successful host invasion and disease because of their negative impacts on a number of host cell functions, for instance, respiration, interleukin-2 release, a protease- antiprotease activity, ciliary beating, prostaglandin release, epidermal cell growth, 7

neutrophil apoptosis, Immunoglobulin G secretion and calcium homeostasis (Ran et al., 2003, Laursen & Nielsen, 2004).

On the other hand, phenazines are characterized as potential anticancer agents (Blankenfeldt et al., 2013). For example, phenazine derivative phenazinomycin produced by Streptomyces sp. WK-2057 showed antitumor activity against P388 leukemia cancer cells (Funayama et al., 1989). It was identified that phenazines interfered with topoisomerase I and II activities in eukaryotic cell. Compared with normal cells, higher levels of these two topoisomerases facilitates cancer cells more susceptible to the interference caused by phenazines (Laursen & Nielsen, 2004). Also, cells that are actively respiring, like tumor cells, appear to be more susceptible to ROS generated by phenazines.

1.2.3 Phenazines as pollutants

Due to their bright pigmentations and ability to change colour depending on pH and redox state, phenazines have been used as dyes and pH indicators. Neutral red is among the best known and is widely used in industry as a cationic dye, in applications such as textiles, papers, plastics, leather and cosmetic industries as well as in analytical laboratories as a biological nucleus counterstain and pH indicator over the pH range 6.8 – 8.0. However, improper disposal practices of the colorful effluents is environmentally and aesthetically unacceptable and results in reduction of light penetration and photosynthesis in aquatic environments (Gong et al., 2009). Additionally, research on ecotoxicity of neutral red dye showed that EC50 values of plants root elongation assays were 10.0 mM for Lactuca sativa, 19.9 mM for Daucus carota, 35.5 mM for Allium cepa and 1.20 mM for Themeda triandra, respectively (Kastury et al., 2015). Remarkably, the observable adverse effect on root elongation of L. sativa was observed at the concentration of 10 µM, indicating this plant was highly susceptible to neutral red (Kastury et al., 2015).

1.3 Removal of phenazines

Given their roles in ecology and environmental and human health as mentioned above, it is crucial to understand processes for turnover of phenazines as a part of their fate in the environment. Compared to our knowledge on synthesis of phenazines and

our understanding of their impacts on biological systems, information on the destruction or fate of phenazines is very limited, especially on the biological degradation of phenazines. Given the effects of phenazines on microbe, plant and human health, this represents a major knowledge gap and one that is central to this thesis. This section describes the abiotic and biotic processes of phenazine removal from the environment.

1.3.1 Non-biological removal

Adsorption of phenazines, especially neutral red as dyes from aqueous solution, under different conditions has been well-documented (Han et al., 2008, Shi et al., 2010, Yue et al., 2011, Gholivand et al., 2015). Many materials including activated carbon, nanospheres and biomass waste, are used as adsorbents for neutral red via chemisorptions, namely electrostatic attractions, and/or physisorption, namely Van der Waals interactions. The functionalities of these adsorbents are dependent on a number of parameters including pH (Zhou et al., 2011, Wang et al., 2013), absorbents dosage (Han et al., 2008), contact time (Gong et al., 2007), initial concentration of NR (Luo et al., 2010), temperature (Shi et al., 2010), particle size (Gong et al., 2009) and ionic strength (Zou et al., 2013).

To date, there has only been one study reporting the systematic chemical degradation of phenazines (Levitch, 1976). PCA degradation comprised three oxidations of alkaline permanganate and four decarboxylations of intermediates systematically, leading to the formation of a heterocyclic aromatic product, pyrazine (Figure 1.4) (Levitch, 1976). However, only 19% of PCA underwent ring cleavage and 3.45% of PCA was transferred to pyrazine.

Figure 1.4 The chemical degradation of PCA to pyrazine via oxidations of alkaline permanganate and decarboxylations (Levitch, 1976).

1.3.2 Biological degradation

Biological degradation can be metabolic or co-metabolic. When degradation occurs metabolically, it involves enzyme-catalyzed reactions that are related to carbon, nitrogen and energy conservation and microbial growth. In contrast, co-metabolic is identified as fortuitous degradation resulting from the metabolism of a separate compound, where no yield of carbon or energy benefits to bacterial cells. Phenazines have been observed to be degraded metabolically or co-metabolically by microbes including seven bacteria and one fungus, which will be reviewed in detail below. No archaea have been reported capable of degrading phenazines thus far.

Currently, only seven pure bacterial strains (Figure 1.5) including Mycobacterium fortuitum CT6 and ATCC6841 (Costa et al., 2015), M. septicum DKN1213 (Costa et al., 2015), Rhodococcus sp. JVH1 (Costa et al., 2015), Nocardia sp. LAM0056 (Costa et al., 2018), Sphingomonas wittichii DP58 (Yang et al., 2007) and Sphingobium yanoikuyae B1 (Zhao et al., 2017) are known to metabolically transform phenazines as a carbon and energy source. Moreover, dioxygenases that are extensively responsible for catalyzing ring cleavages have been observed in all the genomes of the phenazine- degrading bacteria and further identified responsible for phenazines degradation. The decarboxylase UbiD and the synthase of its cofactor UbiX catalyze decarboxylation as the first step of PCA degradation in all PCA-degrading Actinobacteria (Costa et al., 2018).This thesis expands the knowledge of biodegradation of phenazines with two additional isolates from surface soil and sewage sludge, respectively.

Figure 1.5 A neighbour-joining tree of near full-length 16S rRNA gene sequence of phenazine-degrading bacteria reported. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are accession numbers from NCBI or IMG. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.)

Sphingomonas wittichii strain DP58 was the first phenazine-degrading bacterium isolated only as recently as 2007 (Yang et al., 2007). Under aerobic conditions, this strain utilized 200 mg (0.89 mM) of PCA as a sole carbon and energy but not nitrogen source in inorganic salt medium at 6.67 mg L-1 h-1 (0.03 mM h-1) (Yang et al., 2007). Pre-culturing the strain in the presence of PCA increased degradation rates from 0.03 mM h-1 to 0.25 mM h-1, indicating PCA degradation is inducible (Zhao et al., 2017). The metabolic pathway of PCA degradation in strain Sphingomonas sp. DP58 was determined using gas chromatography (GC)–mass spectroscopy (MS) and 1H-nuclear magnetic resonance (NMR) analysis (Chen et al., 2008). Specifically, it was proposed that PCA was transformed into two products including 4-hydroxy-1-(2-carboxyphenyl) azacyclobut-2-ene-2-carbonitrile (HPAEC) and 4-hydroxy-1-(2-carboxyphenyl)-2- azetidinecarbonitrile (HPAC) via ring cleavages (Figure 1.6) (Chen et al., 2008). However, intermediates in the study were speculated rather than detected, meaning that this pathway was not essentially supported. Further research remains to be performed.

Figure 1.6 Proposed pathway for the metabolism of PCA by Sphingomonas wittichii strain DP58 via ring cleavage. The compounds in brackets were not detected and the compounds in square dotted boxed were tentatively identified (Chen et al., 2008).

More recently, by using transcriptome analysis and whole-cell transformation techniques, it was found that four genes pcaA1, pcaA2, pcaA3 and pcaA4 encoded for the initial dioxygenase system, namely, PCA 1,2-dioxygenase, for conversion of PCA to 1,2-dihydroxyphenazine (Figure 1.7) (Zhao et al., 2017). Remarkably, the pattern suggested here seems to contradict the hypothetical degradation pathway presented by Chen et al. (2008), indicating the pathway is still not well characterised and more research is required to address it. PcaA1A2 is a member of a ring-hydroxylating oxygenase family. PcaA3, a GR-type reductase associated with PcaA4 might transfer electrons to PcaA1A2 from NAD(P)H (Zhao et al., 2017). PCA 1,2-dioxygenase is a

typical type IV dioxygenase, whereby activity was only detected when all of the three components including a nonheme iron dioxygenase, a [2Fe-2S] ferredoxin, and a GR- type reductase were expressed (Zhao et al., 2017).

Figure 1.7 Pattern of PCA conversion via dioxygenation catalyzed by PCA 1,2- dioxygenase (PcaA1A2A3A4) from strain Sphingomonas sp. DP58. Red, reduction; ox, oxidation (Zhao et al., 2017).

The second isolation of phenazine degrading bacteria was carried out by Costa et al. (2015) during the course of this current project. This resulted in two more phenazines- degrading isolates, namely, Mycobacterium fortuitum CT6 and M. septicum DKN1213, from mineral medium with 2.5 mM PCA as the sole source of carbon, energy and nitrogen. The XA26_16730 gene encoding ortho-halobenzoate-1,2-dioxygenase in M. fortuitum CT6 was predicted responsible for PCA degradation by quantitative reverse transcription-PCR (qRT-PCR), and proven by allelic replacement of its homologs (gene no. MFORT_16269) in M. fortuitum ATCC6841 which was the fourth phenazine degrading bacterium to be discovered (Costa et al., 2015). This gene was also observed in the genome of Rhodococcus sp. strain JVH1 (gene GI no. 497121742) which also proved capable of PCA degradation representing the fifth known phenazine degrading bacterium (Costa et al., 2015). Moreover, during the production of this thesis, Costa et al. (2018) found that decarboxylase UbiD and the synthase of its cofactor UbiX were responsible for the decarboxylation as the first step of PCA degradation in all PCA- degrading Actinobacterial isolates including M. fortuitum CT6, M. septicum DKN1213, M. fortuitum ATCC6841, Rhodococcus sp. JVH1 and Nocardia sp. LAM0056, the latter of which represented the seventh known phenazine-degrading bacterium.

In addition to PCA degradation, M. fortuitum CT6 and M. septicum DKN1213 were also shown to utilise pyocyanin in a growth linked fashion (Costa et al., 2015). Using RNA-Seq and mutagenesis techniques, the XA26_16990 gene in M. fortuitum CT6 and its product pyocyanin demethylase (referred to as PodA30-162) was identified as obligate for pyocyanin transformation in M. fortuitum ATCC6841 (Costa et al., 2015, Costa et al., 2017). Furthermore, the activity of PodA30-162 was characterized using a heterologous expression in E. coli, suggesting that pyocyanin was converted into 1- hydrophenazine and formaldehyde via a novel demethylation pathway. PodA30-162 is the first member of a new class of tautomerizing demethylases that utilize an oxidized substrate like oxidized pyocyanin as the electron acceptor in the absence of either flavin or 2-oxoglutarate (Costa et al., 2017).

Zhao et al. (2017) reported the sixth bacterial isolate Sphingobium yanoikuyae strain B1 capable of degrading 0.2 mM phenazine as the sole carbon and energy but not nitrogen source. During the degradation of phenazine by S. yanoikuyae B1, 1,2- dihydrogen-1,2-dihydroxy phenazine was identified as one intermediate which was unstable and was further transformed into 2-hydroxyphenazine through spontaneous dehydration and an unknown dark precipitate (Figure 1.8) (Zhao et al., 2017). Furthermore, the genes bphA1f and bphA2f and the encoded product biphenyl 2,3- dioxygenase were determined to be responsible for the initial dioxygenation in phenazine conversion by strain S. yanoikuyae B1 (Figure 1.8) (Zhao et al., 2017).

Figure 1.8 Initial steps in phenazine degradation via dioxygenation catalyzed by biphenyl 2,3-dioxygenase (BphA1fA2fA3fA4f) in strain Sphingobium yanoikuyae B1. Red, reduction; ox, oxidation (Zhao et al., 2017).

Dioxygenases are ubiquitous in the biosphere, forming a functional complex with the electron transfer chain consisting of reductase and ferredoxin (Zhao et al., 2017). These enzymes are essential in the detoxification and catabolic processing of low-molecular- weight polycyclic aromatic hydrocarbons (LMW PAHs) such as naphthalene, anthracene and phenanthrene, which share structural similarities with phenazines (Mallick et al., 2011, Duarte et al., 2017, Posman et al., 2017). Ring-hydroxylating dioxygenases initiate the transformation of LMW PAHs via regio- and stereo-selective introduction of two hydroxyl groups at non-activated vicinal carbon atoms of the substrates, leading to the corresponding dihydrodiol compounds (Mallick et al., 2011). Subsequently, ring cleavage dioxygenases including meta- or ortho-cleavage dioxygenases further utilize PAH-diols, the catecholic intermediates from the upper ring-hydroxylating dioxygenases-catalysing pathways, resulting in salicylate-type or phthalate-type of intermediates (Mallick et al., 2011). The two types of intermediates are then transferred to catecholic-type structures through decarboxylative hydroxylation or ring cleavage dioxygenation (Kim et al., 2007, Mallick et al., 2007). Catechol is then introduced into the TCA cycle by meta- or ortho-cleavage dioxygenases (Mallick et al., 2007). Given the gap of knowledge on phenazines degradation and structural similarity between phenazines and LMW PAHs, the metabolic pathways of LMW PAHs may show clues to the underlying mechanism of phenazines degradation.

Aside from bacteria, a species of white-rot fungus, Perenniporia subacida, was discovered to degrade phenazines co-metabolically (Si et al., 2013). This species of fungus transformed neutral red into methylbenzene and N,N-dimethylphenylenediamine via ring cleavage and deamination (Figure 1.9) (Si et al., 2013). Compared to a control without neutral red addition, various enzymes were induced when P. subacida was incubated with neutral red, including lignin peroxidase, laccase, manganese peroxidase, tyrosinase and nicotinamide-adenine dinucleotide hydrogen-2,6-dichlorophenol indophenol reductase, suggesting their potential involvements in neutral red degradation (Si et al., 2013).

Figure 1.9 Proposed pathway of neutral red degradation via ring cleavage and deamination by white-rot fungus Perenniporia subacida (Si et al., 2013).

The efficiency of neutral red degradation by P. subacida was linked with several factors including pH, temperature, NR concentration and ionic strength (Si et al., 2013). The fungus tolerated a pH range from 2 to 10 with an optimum at pH 4, which might be owing to the electrostatic interaction between neutral red molecules and the surface of fungal cells (Si et al., 2013). Neutral red degradation occurred over a range of temperatures between 10 to 35 °C, where degradation efficiency increased with temperature (Si et al., 2013). However, due to toxicity of neutral red to P. subacida, the ability of neutral red degradation was suppressed when the concentration of neutral red was higher than 100 mg/L (0.35 mM) (Si et al., 2013). Additionally, salt at lower concentrations stimulated neutral red degradation whereas higher salt concentrations showed negative effects on the degradation (Si et al., 2013).

1.4 Research objectives

The preceding literature review has provided an overview of the significant advancements in the biosynthesis pathways, the roles of phenazine compounds in ecology and environmental and human health, and the understanding of specific bacterial and fungal strains involved in transformation of these phenazine compounds via decarboxylation and dioxygenation. Despite these advancements, the identification and studies on aerobic and anaerobic phenazine-degrading bacteria are still lacking in comparison to synthesis and importance of phenazines. More phenazine-degrading bacteria and details in pathways of phenazines biodegradation are yet to be explored. For instance, phenazines are not known to serve as a carbon source for cell yield and energy generation by any microorganisms under anaerobic conditions. Given their roles

in advantaging their producers, protecting plants as biocontrol, infecting human lungs by pathogens and potential as anticancer therapeutics, more work is required to extend knowledge of phenazines transformation.

Therefore, the central objective of this thesis is to enrich, isolate and characterize novel bacteria with the ability to degrade phenazines, namely, phenazine-1-carboxylic acid (PCA) under aerobic and anaerobic conditions. Specifically, four tasks were undertaken as part of this body of work:

1. Isolation and physiological characterization of Rhodanobacter sp. PCA2 capable of degrading PCA under aerobic conditions. 2. Analysis of abundance and activity of the homolog of the MFORT_16269 gene in PCA-degrading Rhodanobacter sp. PCA2, and genome sequencing, assembly, annotation and analysis of strain PCA2 3. Mass spectrometry, post genomics, and proteomics studies to shed light on the pathway of PCA degradation and identify proteins involved in the PCA-degrading Rhodanobacter sp. PCA2. 4. Enrichment, isolation and preliminary characterization of an anaerobic strain Morganella morganii SL11 capable of degrading PCA with iron (III) as electron acceptor.

1.5 Overview of the dissertation

Results from this investigation are organized into four chapters:

In Chapter 2, findings related to the successful isolation of a novel aerobic phenazines-degrading Rhodanobacter strain designated Rhodanobacter sp. PCA2 are presented. As mentioned previously, there are only seven PCA-degrading bacteria available in pure culture. Hence, there is a need for more pure cultures to enhance our understanding of the transformation of phenazines. In this study, the ability of the novel isolate to consume other phenazine compounds and carbon sources was tested. The effects of PCA concentration on degradation rate and cell growth were also tested. Finally, using PCR, ubiD and a homolog of the MFORT_16269 gene reported responsible for PCA transformation were detected in the novel isolate.

In Chapter 3, degradation of PCA was linked to abundance and activity of the homolog of the MFORT_16269 gene using qPCR and RT-PCR. The genome of the novel isolate from Rhodanobacter sp. PCA2 was sequenced, assembled, annotated and analysed. Based on genomic data, decarboxylase UbiD but not phenylpropionate dioxygenase was identified in strain PCA2. Sequences from strain PCA2 were compared with two available PCA-degrading genomes including M. fortuitum CT6 and Sphingomonas sp. DP58 and one Rhodanobacter genome R. denitrificans 2APBS1.

In Chapter 4, based on demonstration of inducibility of PCA degradation, the coupling of LC-MS and proteomic analysis was used to identify the PCA degradation pathway including intermediates, products and the specific enzymes involved in strain PCA2.

In Chapter 5, findings related to the successful enrichment and isolation of the first known anaerobic PCA-degrading bacterium Morganella morganii strain SL11 that utilizes PCA as its carbon, nitrogen and energy source are presented. As mentioned before, there is a lack of anaerobic bacteria capable of transforming PCA. Hence, there is a need for pure cultures to shed light on anaerobic degradation of PCA. In this study, electron acceptors and donors were tested to support PCA transformation under anaerobic conditions. Also, the bacterial composition of enrichment cultures were examined to describe the effects of PCA on community profiles.

References

Abken H-J, Tietze M, Brodersen J, Bäumer S, Beifuss U & Deppenmeier U (1998) Isolation and characterization of methanophenazine and function of phenazines in membrane-bound electron transport of Methanosarcina mazei Gö1. Journal of Bacteriology 180: 2027-2032.

Armstrong AV & Stewart-Tull DES (1971) The site of the activity of extracellular products of Pseudomonas aeruginosa in the electron-transport chain in mammalian cell respiration. Journal of Medical Microbiology 4: 263-270.

Bamberger E & Ham W (1911) The behaviour of several parasubstituted nitro benzoyl against concentrated sulphuric acid. Justus Liebigs Annalen Der Chemie 382: 82-128.

Blankenfeldt W, Sakhtah H, Price-Whelan A, et al. (2013) Microbial phenazines: Biosynthesis, agriculture and health. Springer, New York.

Budzikiewicz H (1993) Secondary metabolites from fluorescent Pseudomonads. FEMS Microbiology Reviews 10: 209-228.

Caldwell CC, Chen Y, Goetzmann HS, Hao Y, Borchers MT, Hassett DJ, Young LR, Mavrodi D, Thomashow L & Lau GW (2009) Pseudomonas aeruginosa exotoxin pyocyanin causes cystic fibrosis airway pathogenesis. American Journal of Pathology 175: 2473-2488.

Carle G (1882) Sur les colorations bleue et verte des linges à pansements. Comptes Rendus Hebdomadaires Des Seances De l'Academie Des Sciences 94: 536-568.

Challand S, Herbert R & Holliman F (1970) A new phenazine synthesis. The synthesis of griseoluteic acid, griseolutein A, and methyl diacetylgriseolutein B. Journal of the Chemical Society D: Chemical Communications 0: 1423-1425.

Chang CY (2017) Surface sensing for biofilm formation in Pseudomonas aeruginosa. Frontiers in Microbiology 8: 1-8.

Chen K, Hu H, Wang W, Zhang X & Xu Y (2008) Metabolic degradation of phenazine- 1-carboxylic acid by the strain Sphingomonas sp. DP58 the identification of two metabolites. Biodegradation 19: 659-667.

Chin-A-Woeng TFC, Bloemberg GV & Lugtenberg BJJ (2003) Phenazines and their role in biocontrol by Pseudomonas bacteria. New Phytologist 157: 503-523. 19

Cogen AL, Nizet V & Gallo RL (2008) Skin microbiota: a source of disease or defence? British Journal of Dermatology 158: 442-455.

Costa KC, Glasser N, Conway SJ & Newman DK (2017) Pyocyanin degradation by a tautomerizing demethylase inhibits Pseudomonas aeruginosa biofilms. Science 355: 170- 173.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Das T, Ibugo AI, Klare W & Manefield M (2016) Role of pyocyanin and extracellular DNA in facilitating Pseudomonas aeruginosa biofilm formation. Microbial biofilms - importance and applications,(Dhanasekaran D & Thajuddin N, eds.), p.^pp. 24-42. InTech, London.

Deppenmeier U, Johann A, Hartsch T, et al. (2002) The genome of Methanosarcina mazei: evidence for lateral gene transfer between bacteria and archaea. Journal of Molecular Microbiology and Biotechnology 4: 453-461.

Ding ZG, Li MG, Ren J, Zhao JY, Huang R, Wang QZ, Cui XL, Zhu HJ & Wen ML (2011) Phenazinolins A-E: novel diphenazines from a tin mine tailings-derived Streptomyces species. Organic and Biomolecular Chemistry 9: 2771-2776.

Duarte M, Nielsen A, Camarinha-Silva A, Vilchez-Vargas R, Bruls T, Wos-Oxley ML, Jauregui R & Pieper DH (2017) Functional soil metagenomics: elucidation of polycyclic aromatic hydrocarbon degradation potential following 12 years of in situ bioremediation. Environmental Microbiology 19: 2992-3011.

Emoto T, Kubosaki N, Yamagiwa Y & Kamikawa T (2000) A new route to phenazines. Tetrahedron Letters 41: 355-358.

Fitzpatrick DA (2009) Lines of evidence for horizontal gene transfer of a phenazine producing operon into multiple bacterial species. Journal of Molecular Evolution 68: 171-185.

Fordos J-M (1859) Recherches sur la matière colorante des suppurations bleues: pyocyanine. Rec Trav Soc d'Emul Sci Pharm 3: 30.

Funayama S, Eda S, Komiyama K, Ōmura S & Tokunaga T (1989) Structure of phenazinomycin, a novel antitumor antibiotic. Tetrahedron Letters 30: 3151-3154.

Gholivand MB, Yamini Y, Dayeni M & Seidi S (2015) Removal of methylene blue and neutral red from aqueous solutions by surfactant-modified magnetic nanoparticles as highly efficient adsorbent. Environmental Progress & Sustainable Energy 34: 1683-1693.

Giddens SR, Feng Y & Mahanty HK (2002) Characterization of a novel phenazine antibiotic gene cluster in Erwinia herbicola Eh1087. Molecular Microbiology 45: 769- 783.

Gong JL, Wang B, Zeng GM, Yang CP, Niu CG, Niu QY, Zhou WJ & Liang Y (2009) Removal of cationic dyes from aqueous solution using magnetic multi-wall carbon nanotube nanocomposite as adsorbent. Journal of Hazardous Materials 164: 1517-1522.

Gong R, Zhang X, Liu H, Sun Y & Liu B (2007) Uptake of cationic dyes from aqueous solution by biosorption onto granular kohlrabi peel. Bioresource Technology 98: 1319- 1323.

Haddadin MJ & Issidorides CH (1965) Enamines with isobenzofuroxan: a novel synthesis of quinoxaline-di-N-oxides. Tetrahedron Letters 6: 3253-3256.

Han R, Han P, Cai Z, Zhao Z & Tang M (2008) Kinetics and isotherms of neutral red adsorption on peanut husk. Journal of Environmental Sciences 20: 1035-1041.

Hartwig JF (1998) Transition metal catalyzed synthesis of arylamines and aryl ethers from aryl halides and triflates: scope and mechanism. Angewandte Chemie International Edition 37: 2046-2067.

Hassan HM & Fridovich I (1980) Mechanism of the antibiotic action of pyocyanine. Journal of Bacteriology 141: 156-163.

Hollstein U & Marshall L (1972) Biosynthesis of phenazines. Journal of Organic Chemistry 37: 3510-3514.

Hollstein U & Butler PL (1972) Inhibition of ribonucleic acid synthesis by myxin. Biochemistry 11: 1345-1350.

Hollstein U, Mock D, Sibbitt R, Roisch U & Lingens F (1978) Incorporation of shikimic acid into iodinin. Tetrahedron Letters 19: 2987-2990.

Hosseiny Davarani SS, Fakhari AR, Shaabani A, Ahmar H, Maleki A & Sheijooni Fumani N (2008) A facile electrochemical method for the synthesis of phenazine derivatives via an ECECC pathway. Tetrahedron Letters 49: 5622-5624.

Imamura N, Nishijima M, Takadera T, Adachi K, Sakai M & Sano H (1997) New anticancer antibiotics pelagiomicins, produced by a new marine bacterium Pelagiobacter variabilis. The Journal of Antibiotics 50: 8-12.

Ingledew W & Campbell J (1969) Evaluation of shikimic acid as a precursor of pyocyanine. Canadian Journal of Microbiology 15: 535-541.

Kastury F, Juhasz A, Beckmann S & Manefield M (2015) Ecotoxicity of neutral red (dye) and its environmental applications. Ecotoxicology and Environmental Safety 122: 186- 192.

Kim SJ, Kweon O, Jones RC, Freeman JP, Edmondson RD & Cerniglia CE (2007) Complete and integrated pyrene degradation pathway in Mycobacterium vanbaalenii PYR-1 based on systems biology. Journal of Bacteriology 189: 464-472.

Land CWVt, Mocek U & Floss HG (1993) Biosynthesis of the phenazine antibiotics, the saphenamycins and esmeraldins, in Streptomyces antibioticus. Journal of Organic Chemistry 58: 6576-6582.

Lau GW, Hassett DJ, Ran H & Kong F (2004) The role of pyocyanin in Pseudomonas aeruginosa infection. Trends in Molecular Medicine 10: 599-606.

Lau GW, Ran H, Kong F, Hassett DJ & Mavrodi D (2004) Pseudomonas aeruginosa pyocyanin is critical for lung infection in mice. Infection and Immunity 72: 4275-4278.

Laursen JB & Nielsen J (2004) Phenazine natural products: biosynthesis, synthetic analogues, and biological activity. Chemical Reviews 104: 1663-1685.

Levitch ME (1976) The chemical degradation of phenazine-1-carboxylic acid. Journal of the National Medical Association 68: 487-491.

Luo P, Zhao Y, Zhang B, Liu J, Yang Y & Liu J (2010) Study on the adsorption of neutral red from aqueous solution onto halloysite nanotubes. Water Research 44: 1489-1497.

Luo Q, Hu H, Peng H, Zhang X & Wang W (2015) Isolation and structural identification of two bioactive phenazines from Streptomyces griseoluteus P510. Chinese Journal of Chemical Engineering 23: 699-703.

Maddula VS, Pierson EA & Pierson LS, 3rd (2008) Altering the ratio of phenazines in Pseudomonas chlororaphis (aureofaciens) strain 30-84: effects on biofilm formation and pathogen inhibition. Journal of Bacteriology 190: 2759-2766.

Mallick S, Chatterjee S & Dutta TK (2007) A novel degradation pathway in the assimilation of phenanthrene by Staphylococcus sp. strain PN/Y via meta-cleavage of 2- hydroxy-1-naphthoic acid: formation of trans-2,3-dioxo-5-(2'-hydroxyphenyl)-pent-4- enoic acid. Microbiology 153: 2104-2115.

Mallick S, Chakraborty J & Dutta TK (2011) Role of oxygenases in guiding diverse metabolic pathways in the bacterial degradation of low-molecular-weight polycyclic aromatic hydrocarbons: a review. Critical Reviews in Microbiology 37: 64-90.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

Mavrodi DV, Ksenzenko VN, Bonsall RF, Cook RJ, Boronin AM & Thomashow LS (1998) A seven-gene locus for synthesis of phenazine-1-carboxylic acid by Pseudomonas fluorescens 2-79. Journal of Bacteriology 180: 2541-2548.

Mavrodi DV, Peever TL, Mavrodi OV, et al. (2010) Diversity and evolution of the phenazine biosynthesis pathway. Applied and Environmental Microbiology 76: 866-879.

McDonald M, Mavrodi D, Thomashow L & Floss H (2001) Phenazine biosynthesis in Pseudomonas fluorescens: branchpoint from the primary shikimate biosynthetic pathway and role of phenazine-1,6-dicarboxylic acid. Journal of the American Chemical Society 123: 9459-9460.

Mentel M, Ahuja EG, Mavrodi DV, Breinbauer R, Thomashow LS & Blankenfeldt W (2009) Of two make one: the biosynthesis of phenazines. Chembiochem 10: 2295-2304.

Norman RS, Moeller P, McDonald TJ & Morris PJ (2004) Effect of pyocyanin on a crude- oil-degrading microbial community. Applied and Environmental Microbiology 70: 4004- 4011.

Pierson LS & Thomashow LS (1992) Cloning and heterologous expression of the phenazine biosynthetic locus from Pseudomonas aureofaciens 30-84. Molecular Plant- Microbe Interactions 5: 330-339.

Pierson LS, III & Pierson EA (2010) Metabolism and function of phenazines in bacteria: impacts on the behavior of bacteria in the environment and biotechnological processes. Applied Microbiology and Biotechnology 86: 1659-1670.

Posman KM, DeRito CM & Madsen EL (2017) Benzene degradation by a Variovorax species within a coal tar-contaminated groundwater microbial community. Applied and Environmental Microbiology 83: e02658-16.

Price-Whelan A, Dietrich LE & Newman DK (2006) Rethinking 'secondary' metabolism: physiological roles for phenazine antibiotics. Nature Chemical Biology 2: 71-78.

Rada B & Leto TL (2009) Redox warfare between airway epithelial cells and Pseudomonas: dual oxidase versus pyocyanin. Journal of Immunology Research 43: 198- 209.

Ran H, Hassett D & Lau G (2003) Human targets of Pseudomonas aeruginosa pyocyanin. Proceedings of the National Academy of Sciences of the United States of America 100: 14315-14320.

Ris C (1886) Ueber das phenazin. Berichte Der Deutschen Chemischen Gesellschaft 19: 2206–2208.

Rui Z, Ye M, Wang S, Fujikawa K, Akerele B, Aung M, Floss HG, Zhang W & Yu TW (2012) Insights into a divergent phenazine biosynthetic pathway governed by a plasmid- born esmeraldin gene cluster. Chemistry and Biology 19: 1116-1125.

Seth K, Raha Roy S & Chakraborti AK (2016) Synchronous double C-N bond formation via C-H activation for a novel synthetic route to phenazine. Chemical Communications 52: 922-925.

Shi Q, Zhang J, Zhang C, Li C, Zhang B, Hu W, Xu J & Zhao R (2010) Preparation of activated carbon from cattail and its application for dyes removal. Journal of Environmental Sciences 22: 91-97.

Si J, Li X-C & Cui B-K (2013) Decolorization of heterocycle dye neutral red by white- rot fungus Perenniporia subacida. Desalination and Water Treatment 52: 5594-5604.

Stewart PS & Franklin MJ (2008) Physiological heterogeneity in biofilms. Nature Reviews Microbiology 6: 199-210.

Turner JM & Messenger AJ (1986) Occurrence, biochemistry and physiology of phenazine pigment production. Advances in Microbial Physiology 27: 211-275.

Umezawa H, Hayano S, Maeda K, Ogata Y & Okami Y (1950) On a new antibiotic, griseolutein, produced by streptomyces. Japanese Medical Journal 3: 111-117.

Vivian D & Hartwell J (1953) Phenazine syntheses. I. Halogen-substituted phenazines. Journal of Organic Chemistry 18: 1065-1069.

Vleesschauwer DD, Cornelis P & Höfte M (2006) Redox-active pyocyanin secreted by Pseudomonas aeruginosa 7NSK2 triggers systemic resistance to Magnaporthe grisea but enhances Rhizoctonia solani susceptibility in rice. Molecular Plant-Microbe Interactions 19: 1406-1419.

Wang S, Wei J, Lv S, Guo Z & Jiang F (2013) Removal of organic dyes in environmental water onto magnetic-sulfonic graphene nanocomposite. Clean - Soil, Air, Water 41: 992- 1001.

Wohl A & Aue W (1901) Ueber die einwirkung von nitrobenzol auf anilin bei gegenwart von alkali. Chemische Berichte 34: 2442-2450.

Wolfe J & Buchwald S (1997) Palladium-catalyzed amination of aryl triflates. Journal of Organic Chemistry 62: 1264-1267.

Yang ZJ, Wang W, Jin Y, Hu HB, Zhang XH & Xu YQ (2007) Isolation, identification, and degradation characteristics of phenazine-1-carboxylic acid-degrading strain Sphingomonas sp. DP58. Current Microbiology 55: 284-287.

Yue D, Jing Y, Ma J, Xia C, Yin X & Jia Y (2011) Removal of neutral red from aqueous solution by using modified hectorite. Desalination 267: 9-15.

Zhao Q, Hu HB, Wang W, Huang XQ & Zhang XH (2017) Novel three-component phenazine-1-carboxylic acid 1,2-dioxygenase in Sphingomonas wittichii DP58. Applied and Environmental Microbiology 83: e00133-00117.

Zhao Q, Bilal M, Yue S, Hu H, Wang W & Zhang X (2017) Identification of biphenyl 2, 3-dioxygenase and its catabolic role for phenazine degradation in Sphingobium yanoikuyae B1. Journal of Environmental Management 204: 494-501.

Zhou Q, Gong W, Xie C, Yang D, Ling X, Yuan X, Chen S & Liu X (2011) Removal of neutral red from aqueous solution by adsorption on spent cottonseed hull substrate. Journal of Hazardous Materials 185: 502-506.

Zou W, Gao S, Zou X & Bai H (2013) Adsorption of neutral red and malachite green onto grapefruit peel in single and binary systems. Water Environment Research 85: 466- 477.

Chapter 2 Isolation and characterization of phenazine-1-carboxylic acid (PCA)-degrading bacterium Rhodanobacter sp. PCA2

2.1 Introduction

Phenazines are a large class of nitrogen-containing aromatic heterocyclomers with broad-spectrum antibiotic and redox activities. Most phenazines are produced and secreted by bacteria from phylogenetically diverse taxa and are extensively studied for their role in protection for their producers and interactions with cohabiting microorganisms and also plants (Mavrodi et al., 2006). For instance, evidence suggests 2-hydroxy-phenazine-1-carboxylic acid and phenazine-1-carboxylic acid (PCA) enhance cellular adhesion and growth within P. aeruginosa biofilms, respectively (Maddula et al., 2008). Pyocyanin caused a decrease in the diversity of the microbial community in an oil-degrading culture and thereby a decrease in degradation of polycyclic aromatic hydrocarbons (Norman et al., 2004). PCA and 1-hydroxyphenazine purified from Streptomyces griseoluteus P510 showed activity against plant pathogens (Luo et al., 2015). Also, phenazines have been characterized as having negative impacts on human health, for example, acting as virulence and survival factors in some infectious diseases (Ran et al., 2003, Lau et al., 2004, Caldwell et al., 2009). On the other hand, it is found that phenazines are potential anticancer agents which act via the inhibition of topoisomerases, radical scavenging, interaction with polynucleotides and charge transfer (Blankenfeldt et al., 2013). Additionally, their colorimetric properties make phenazines useful as dyes, which may cause environmental issues when improperly disposed (Gong et al., 2009). Therefore, insights into turnover as part of the fate of phenazines is important in environmental protection as well as in ecological and human health.

Aerobic degradation of phenazines, especially PCA, by bacteria has attracted interest recently. To date, there are only seven pure bacterial strains capable of transformation of phenazines, including Sphingomonas wittichii DP58 (Yang et al., 2007) and Sphingobium yanoikuyae B1 (Zhao et al., 2017) belonging to Alphaproteobacteria, and

Mycobacterium fortuitum CT6 and ATCC6841 (Costa et al., 2015), M. septicum DKN1213 (Costa et al., 2015), Rhodococcus sp. JVH1 (Costa et al., 2015) and Nocardia sp. LAM0056 (Costa et al., 2018) belonging to Actinobacteria. Observation of aerobic phenazine biodegradation across two large distantly related phyla suggests it may be more common than currently appreciated.

Ortho-halobenzoate-1,2-dioxygenase encoded by the MFORT_16269 gene (annotated by NCBI) is involved in PCA degradation by M. fortuitum ATCC6841, and homologs are detected in all known PCA-degrading strains (Costa et al., 2015). Decarboxylase UbiD encoded by the MFORT_16229 gene and UbiX as the synthase of its cofactor encoded by the MFORT_16239 gene are likely responsible for the decarboxylation as the first step of PCA degradation in all PCA-degrading Actinobacterial isolates (Costa et al., 2018). Despite the identification of these cultures, research on biodegradation of phenazines is still in its infancy. In the future, more bacterial strains are expected to be isolated and studied.

In this chapter, the eighth PCA-degrading strain, Rhodanobacter sp. PCA2 within the Gammaproteobacteria, was isolated after three rounds of aerobic enrichment from soil samples obtained from the University of New South Wales (UNSW) Sydney campus. During enrichment and isolation, PCA was supplied as the sole carbon, nitrogen and energy source. The enriched cultures were diluted and transferred onto solid media to isolate the PCA-degrading bacterium. This novel isolate, Rhodanobacter sp. PCA2, was further characterized and found to be capable of degrading other phenazines. The homolog of the MFORT_16269 gene and ubiD were detected in Rhodanobacter sp. strain PCA2.

2.2 Materials and methods

2.2.1 Inoculum source and cultivation conditions

PCA-degrading strain Rhodanobacter sp. PCA2 was enriched and isolated from soil samples collected from University of New South Wales (UNSW) Sydney campus, Australia. For enrichment cultures, 1 g soil was inoculated in 20 mL minimal mineral medium containing (g/L) KH2PO4 (1.93), K2HPO4 (6.24), NaCl (2.5), MgCl2·6H2O (0.2), 1 mL trace element solution (SL 10) (1000x) and 1 mL Selenite-Wolframate

solution (1000x). PCA (500 mg/L; 2.23 mM) supplied as sole carbon, nitrogen and energy source was then dissolved into the prepared mineral medium by stirring for 30 mins. Trace element solution (SL 10) (Tschech & Pfennig, 1984) contained (g/L) HCl

(25%, w/w, 10 mL), FeCl2•4H2O (1.5), CoCl2•6H2O (0.19), MnCl2•2H2O (0.10), ZnCl2

(0.07), NiCl2•6H2O (0.024), NaMoO4•2H2O (0.036), H3BO3 (0.006) and CuCl2•2H2O (0.002). Selenite-Wolframate solution (Widdel & Pfennig, 1981) contained (g/L) NaOH

(0.4), Na2SeO3•5H2O (0.006) and Na2WO4•2H2O (0.008). The pH was adjusted to 7.0 using 1 M NaOH when required. Cultures were plugged with cotton plugs and sterilized at 121°C for 20 min; then amended with 1 mL/L of filter-sterilized vitamin solution (VS 10) (1000x) (Balch et al., 1979) containing (g/L): biotin (0.01), nicotinic acid (0.02), thiamine-dichloride (0.01), p-aminobenzoic acid (0.01), Ca-D(+)-pantothenic acid (0.005), pyridoxamine-dihydrochloride (0.05) and cyanocobalamine (0.01).

All liquid cultures were incubated in the dark at 30°C with shaking at 180 rpm. All experiments were set up in triplicate and abiotic controls were included as well.

2.2.2 Enrichment and isolation of PCA-degrading bacterium

During enrichment, parent cultures were transferred into fresh media with 1% inocula and incubated for several days. After 3 rounds of growth and enrichment, a dilution series was established in 10 mL of mineral media in 20 mL serum bottles. The first dilution (10-1) was inoculated with 1 mL of parent culture and 9 mL of media then diluted serially until 10-7. For agar series, solid media were prepared by adding 1.5% (w/v) of agar to media before autoclaving. Then a dilution series ranging from 10-3 to 10-7 of the third round of enriched cultures was streaked on to solid media. Agar plates were incubated statically at 30°C in the dark for 3 days. Visible colonies were picked and transferred to fresh liquid media to establish a pure culture.

2.2.3 PCA degradation with different concentration gradients of PCA

To test the effects of PCA concentration on degradation rate, cultures were grown in 20 mL of defined minimal mineral medium in 50 mL flasks. Gradient concentrations of PCA (100, 200, 500, 1000, 2000 mg/L) were amended into media as introduced above (Section 2.2.1). All cultures were prepared in triplicates.

2.2.4 Other phenazine compounds and carbon sources range study

To test the metabolic capacity of strain PCA2 for other phenazines and carbon sources, separate cultures were set up where 200 µM phenazine, pyocyanin (PYO), 1- hydroxyphenazine and 20 mM pyruvate, glucose, alanine, trehalose was provided as sole carbon and energy source. Specifically, the phenazine compounds were mixed thoroughly with the prepared mineral medium by stirring for 30 mins, which was then autoclaved at 121°C for 20 min to dissolve the phenazines into the media. Activity and cell yield were monitored by HPLC measurement and microscopy analyses over a period of 10 days, see Section 2.2.5 and 2.2.6.

2.2.5 Analytical methods of phenazines

To quantify phenazine compounds, 10 µL of culture was injected with an autosampler into a High-performance liquid chromatograph (HPLC; Agilent 1200 Series, Australia), equipped with a UV-Vis detector and a C18 column (Eclipse XDB- C18, 5µm, 4.6 x 150 mm; Agilent, Australia) using modified methods in Kern & Newman (2014). All phenazines were eluted in a gradient of water with 0.1% trifluoroacetic acid (TFA) (solvent A) to acetonitrile with 0.1% TFA (solvent B) at a flow rate of 0.8 mL/min as follows: linear gradient from 5% to 90% solvent B from 0 to 5 min, maintaining for 4 min, linear gradient to 5% solvent B from 9 to 10 min, then maintaining for 10 min until the end. Phenazine, 1-hydroxyphenazine, PYO and PCA were determined under wavelengths of 365 nm, 280 nm, 365 nm and 360 nm, respectively. Compounds were quantified by comparison of the peak area of the unknown to a six-point calibration curve of the corresponding standard.

2.2.6 Microscopic analyses

Growth of cells was determined by fluorescence microscopy (Leica DFC420, Leica Microsystems, Australia). SYBR○R Green I solution (Sigma, Australia) was used to stain cultures after diluted to 10-3. Mixtures of 18 µL of cultures and 2 µL of diluted SYBR○R Green I solution were stored at 4℃ and in the dark for 1 h. For microscopy, 20 µL of the samples was placed onto 2% (w/v) agarose-coated slide and covered with a glass cover slip (18 x 18 mm). Cell number was determined by imaging software Image J.

2.2.7 Identity of isolate based on 16S rRNA sequence

To identify the isolates, genomic DNA was extracted form 1 mL of PCA-degrading cultures as described in Urakawa et al. (2010). Briefly, cells were centrifuged at 8,000x g for 20 min, then lysed using 300 µL lysis buffer, bead-beat with Lysing Matrix A (MP Biomedicals, Australia), extracted with phenol: chloroform: isomylalcohol (25:24:1, v/v; Sigma, Australia) and precipitated by isopropanol overnight. Finally, the DNA was resuspended in 30 µL of molecular grade water after washing with ethanol and centrifuged at 161,000x g for 20 min. Genomic DNA concentration was measured using a Qubit instrument (Qubit○R 2.0 Fluorometer, ThermoFisher Scientific, Australia) by following the manufacturer’s protocol.

DNA extracts were amplified using PCR by targeting near-full length (1475 bp) 16S rRNA with primers 27F (5’-AGAGTTTGATCCTGGCTCAG-3’) and 1492R (5’- GGYTACCT TGTTACGACTT-3’) (Muyzer et al., 1993). PCR reaction was performed in 40 µL volumes containing 20 µL PCR Master Mix (Promega, Australia), 0.13 g/L purified BSA (New England BioLabs, USA), 0.25 µM of each primer and 1 µL of diluted template DNA (~ 5 ng). Negative control reactions without DNA template were also included in all PCR analyses. The PCR reactions were performed on a Mini-PCR (MJ MiniTM Gradient Thermal Cycler, Bio-Rad, Australia) set using the following thermal conditions: 4 min at 95°C , 30 cycles of 1 min at 94°C , 1 min at 52°C and 2 min at 72°C followed by 72°C for 10 min. A PCR purification kit (DNA Clean & ConcentratorTM-25, Zymo Research, USA) was used to clean up the product then sequenced by Sanger sequencing offered by the Ramaciotti Centre for Gene Function and Analysis, Sydney, Australia.

2.2.8 Proof of purity by Denaturing Gradient Gel Electrophoresis (DGGE)

To check the purity of PCA-degrading cultures, DNA extracts were amplified using PCR by targeting partial 16S rRNA (192 bp) with GC-clamped primers GC338F (5’- CGCCCGCGGCGCCCCCGCCCCGGCCCGCCGCCCCCGCACTCCTACGGGAGG CAC-3’) and 530R (5’-GTATTACCGCGCCTGCTG-3’) (Manefield et al., 2002). PCR reaction was performed in 40 µL volumes as above using the following thermal conditions: 2 min at 95°C , 30 cycles of 30 s at 94°C , 30 s at 61°C and 1.5 min at 72°C followed by 72°C for 10 min.

PCR products were analysed on 10% (w/v) polyacrylamide gels with a 45 ~ 60 % urea/formamide denaturing gradient (Whiteley & Bailey, 2000), which was performed on a Biorad D-Code system at 60℃ and 80 V for 16 h. Post electrophoresis, the gels were stained with SYBR gold nucleic acid gel stain (0.01% v/v) (Sigma, Australia) for 15 min in the dark and visualized by gel documentation equipment (Gel DocTM XR+, Bio-Rad, Australia). DGGE bands of interest were cut with band-cutting tips and eluted in 40 µL molecular grade water overnight at 4℃. The eluted DNA was re-amplified by PCR as above for partial 16S rRNA, the PCR products were cleaned up and sequenced as previously described.

2.2.9 Detection of ubiD and the homolog of MFORT_16269 gene by PCR

To detect ubiD gene catalysing decarboxylation in PCA degradation (Costa et al., 2018), extracted genomic DNA was amplified using primers ubiD-20F (5’- ATTGGA CATATGGGCGCTCG-3’) and ubiD-1545R (5’-CAATGTCGCCTTCCACCGAT-3’). PCR reaction was performed in 40 µL as above using the following thermal conditions: 4 min at 95°C , 30 cycles of 1 min at 94°C , 1 min at 52°C and 2 min at 72°C followed by 72°C for 10 min. The PCR product was cleaned up and sequenced as above.

To detect the homolog of MFORT_16269 gene responsible for PCA degradation as reported by Costa et al. (2015), extracted genomic DNA was amplified using primers 16269F (5’- CATCGTCAACTGGAACTGGA-3’) and 16269R (5’-GTACTGCCAGTC CGTGTCCT-3’) (Costa et al., 2015). PCR reaction was performed in 40 µL as above using the following thermal conditions: 2 min at 95°C , 30 cycles of 30 s at 94°C , 30 s at 55°C and 2 min at 72°C followed by 72°C for 10 min. The PCR product was cleaned up and sequenced as above.

2.2.10 Phylogenetic characterization

The identity of sequences of partial, near-full 16S rRNA, ubiD and the homolog of MFORT_16269 genes were obtained by searching against relative reference sequences using standard nucleotide BLAST algorithm on NCBI (National Center for Biotechnology Information, https://www.ncbi.nlm.nih.gov/) (Zhang et al., 2000).

A phylogenetic tree was generated using MEGA 6.0 to compare the phylogenetic relationship of strain PCA2 to other phenazine-degrading and the relative organisms (Tamura et al., 2013). The phylogenetic tree is a neighbour-joining tree based on near-

full length (1475 bp) sequences of 16S rRNA and the bootstrap percentages on branching points were based on a maximum parsimony tree using 1000 bootstrap values.

2.2.11 Statistical analysis

Analysis of variance were performed with one-way ANOVA followed by S-N-K (n ≥ 3) or T-tests (n = 2) using SPSS version 19.0 (IBM Co., Armonk, NY, US), respectively. P values below 0.05 were considered statistically significant.

2.3 Results

2.3.1 Enrichment, isolation and phylogenetic characterization of Rhodanobacter sp. PCA2

To obtain a PCA-degrading enrichment culture, 20 mL aerobic microcosms inoculated with 1 g soil were set up with 500 mg/L (2.23 mM) of PCA as the sole C, N and energy source. Microcosms showed metabolic degradation of PCA during enrichment, with the rate increasing from 6.7 to 13.3 µM h-1 in the first two rounds of enrichment and finally to 25.0 µM h-1 in the third round of enrichment (Figure 2.1).

Figure 2.1 Degradation of PCA in three rounds of enrichment. The metabolic degradation rate of PCA was enhanced during enrichment, suggesting PCA-degrading bacteria was successfully enriched. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

Subsequently, a dilution series from 10-3 to 10-7 was processed with the third round of enrichment, from which an active PCA-degrading culture (10-5) was further used to inoculate agar plates containing PCA. After incubation for approximately 3 days, small, round, yellow-white colonies were observed on the agar plates (Figure 2.2A). Colonies were picked and transferred to liquid mineral media amended with 500 mg/L (2.23 mM) of PCA. DGGE analysis of the isolated culture exhibited only one band (Figure 2.2B) which was identified as being related to Rhodanobacter sp. A2-60 based on sequencing of partial 16S rRNA (192 bp). Subsequently, a near full-length (1475 bp) 16S rRNA sequence of Rhodanobacter sp. PCA2 was analysed confirming that the isolate was affiliated with the Rhodanobacter genus belonging to the family Rhodanobacteraceae of order Xanthomonadales within class Gammaproteobacteria. This represents the first known phenazines degrading Gammaproteobacteria. The isolated strain here was then designated as Rhodanobacter sp. PCA2. The comparison of near full-length 16S rRNA sequence showed that strain PCA2 shared 98% identity with Rhodanobacter sp. strain A2-60 and Rhodanobacter denitrificans strain 2APBS1 (Figure 2.3).

Figure 2.3 A neighbour-joining tree of near full-length 16S rRNA gene sequence of Rhodanobacter sp. PCA2 (in bold), along with other Rhodanobacter strains and other phenazines-degrading bacteria. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are GenBank accession numbers from NCBI or IMG. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.)

2.3.2 Growth linked degradation of PCA to Rhodanobacter sp. PCA2

To confirm that the growth of Rhodanobacter sp. PCA2 is linked to degradation of PCA, fresh cultures were monitored for the decrease in PCA concentration coupled with cell yield of strain PCA2. Figure 2.4 shows the degradation of PCA by strain PCA2 in 48 h (46.2 µM h-1) while no consumption of PCA was observed in sterile cultures. Accompanying PCA degradation, cell numbers of strain PCA2 determined by fluorescence microscope increased from 7.18 × 105 to 5.46 × 107 cells/mL (P < 0.05), meaning that 1.17 × 106 cell was produced by 1 µmol of PCA (Figure 2.4). Cells were not observed in abiotic cultures during the incubation.

Figure 2.4 Degradation of PCA and cell yield of Rhodanobacter sp. PCA2. Growth of strain PCA2 was linked with metabolic PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

2.3.3 Effect of PCA concentration on rate of degradation

Different concentrations of PCA (100, 200, 500, 1000 and 2000 mg/L) were amended into cultures inoculated with strain PCA2 to determine the effect of PCA concentration on the rate of degradation. It was revealed that strain PCA2 metabolically transformed PCA up to 1,000 mg/L (4.46 mM) with concomitant yield of biomass (Figure 2.5, Figure 2.6). After incubation for 2 days, PCA concentrations ranging from 100 to 500 mg/L (from 0.45 to 2.23 mM) were depleted in PCA2 cultures whereas 1000 mg/L (4.46 mM) of PCA was completely degraded by strain PCA2 in 8 days (Figure 2.5). Cell numbers of strain PCA2 increased from 5.79 × 105 to 8.97 × 105 and from 3.78 × 105 to 1.47 × 106 cells/mL in the first 2 days before decreasing to 3.72 × 105 and 8.79 × 105 cells/mL in cultures amended with 100 and 200 mg/L (0.45 and 0.89 mM) of PCA, respectively (Figure 2.6). In cultures amended with 500 mg/L (2.23 mM) of PCA, the biomass increased significantly from 7.18 × 105 to 5.46 × 107 cells per mL (Figure 2.6). No degradation of PCA and no increases in cell concentration were observed in cultures containing 2000 mg/L (8.93 mM) of PCA (Figure 2.5, Figure 2.6).

Figure 2.5 PCA degradation by Rhodanobacter sp. PCA2 at different concentrations of PCA. The highest rate of PCA degradation was detected in cultures fed with 500 mg/L PCA. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures.

Figure 2.6 Cell yield of Rhodanobacter sp. PCA2 in cultures with different concentrations of PCA. The highest increase of cell concentration was detected in cultures fed with 500 mg/L PCA, which was consist with profiles of rate of PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures. 37

2.3.4 Utilization of other phenazines and C sources

To gain an understanding of the metabolic versatility of strain PCA2, its capability of using other phenazine compounds including phenazine, pyocyanin and 1-hydroxyphenazine was explored. Strain PCA2 was able to transform up to 200 µM of phenazine, pyocyanin and 1-hydroxyphenazine at rates of 5.67, 2.22 and 1.34 µM h-1 respectively with concomitant increase of cell numbers from 6.18 × 105 to 2.23 × 107, from 3.72 × 105 to 1.93 × 107 and from 6.40 × 105 to 1.03 × 107 cells/mL, respectively (Figure 2.7). Accordingly, strain PCA2 degraded phenazine, pyocyanin and 1- hydroxyphenazine at rates of 2.62 × 10-10, 1.17 × 10-10 and 1.39 × 10-10 µmol cell-1 h-1, respectively, which were lower than the corresponding rate of PCA degradation (8.57 × 10-10 µmol cell-1 h-1).

Figure 2.7 Degradation of phenazine compounds (solid line) and cell yield (dash line) of Rhodanobacter sp. PCA2 in cultures. The highest rate of PCA degradation and cell yield of strain PCA2 was observed in cultures fed with phenazine, followed by cultures fed with pyocyanin and 1-hydroxyphenazine, respectively. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures.

Additionally, the ability of strain PCA2 to utilize a variety of C sources including pyruvate, glucose, alanine and trehalose was also examined. Highest cell yield (3.28 × 107 cells/mL) was observed in cultures with pyruvate as sole C source, and no significant difference of cell yield was determined amongst cultures with alanine (1.37 × 107 cells/mL), glucose (3.28 × 107 cells/mL) and trehalose (6.11 × 106 cells/mL) (Figure 2.8).

Figure 2.8 Cell yield of Rhodanobacter sp. PCA2 in cultures with different C sources. Strain PCA2 can utilize pyruvate, glucose, alanine and trehalose as the sole C source, with pyruvate supporting the highest cell yield. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

2.3.5 Detection of ubiD and the homolog of MFORT_16269 gene

The ubiD and MFORT_16269 genes were previously characterized to be responsible for PCA degradation by strain Mycobacterium fortuitum ATCC6841 (Costa et al., 2015, Costa et al., 2018). In this study, ubiD (1525 bp) and the homolog of MFORT_16269 (247 bp) were successfully amplified in strain PCA2 by PCR with primer pairs ubiD- 20F/ubiD-1545R and 16269F/16269R, and were subsequently identified by Sanger sequencing. BLAST search with the NCBI database showed both of these two genes shared high homology with the corresponding genes in M. fortuitum strains ATCC6841 (95% for ubiD gene, 95% for MFORT_16269 gene) and CT6 (95% for ubiD gene, 94% 39

for MFORT_16269 gene) (Table 2.1, Table 2.2). Additionally, genes encoding 3- polyprenyl-4-hydroxybenzoate carboxy-lyase and alpha subunit ring hydroxylating dioxygenase in Mycobacterium sp. VKM Ac-1817D also shared high (95%) homology, with ubiD and the homolog of MFORT_16269 genes in strain PCA2 (Table 2.1, Table 2.2). A gene encoding phenolic acid decarboxylase subunit C in Nocardia farcinica shared 81% identify with ubiD gene identified in strain PCA2 (Table 2.1).

Table 2.1 Sequences producing significant alignments with the ubiD gene amplified from genomic DNA of Rhodanobacter sp. PCA2 using BLAST analysis in NCBI Accession Features annotated Sequence source Identity number in NCBI by NCBI Genome of Mycobacterium 95% CP011269 UbiD family decarboxylase fortuitum CT6 Genome of Mycobacterium 95% CP014258 UbiD family decarboxylase fortuitum ATCC 6841 3-polyprenyl-4- Genome of Mycobacterium 95% CP009914 hydroxybenzoate carboxy- sp. VKM Ac-1817D lylase Genome of Nocardia Phenolic acid decarboxylase 81% LN868938 farcinica subunit C

Table 2.2 Sequences producing significant alignments with the MFORT_16269 gene amplified from genomic DNA of Rhodanobacter sp. PCA2 using BLAST analysis in NCBI

Accession Features annotated Sequence source Identity number in NCBI by NCBI

Genome of Mycobacterium Photosystem I reaction 95% CP014258 fortuitum ATCC 6841 center subunit VIII Ring hydroxylating Genome of Mycobacterium dioxygenase, alpha 95% CP009914 sp. VKM Ac-1817D subunit/Rieske (2Fe-2S) protein Genome of Mycobacterium Ortho-halobenzoate 1,2- 94% CP011269 fortuitum CT6 dioxygenase, alpha subunit

2.4 Discussion

Recently, there have been reports of isolates in the Sphingomonas, Sphingobium, Mycobacterium, Rhodococcus and Nocardia genera that can couple the transformation of phenazines such as PCA, phenazine and pyocyanin with energy conservation (Yang et al., 2007)(Zhao et al., 2017)(Costa et al., 2015)(Costa et al., 2018). In this study, an enriched PCA-degrading culture was developed with PCA as the sole carbon, nitrogen and energy source. The degradation rate improved from 0.16 to 0.60 mM/day with each round of enrichment. Furthermore, the isolation and characterization of Rhodanobacter sp. PCA2 from the enriched PCA-degrading culture revealed that the strain can utilize PCA as well as phenazine, pyocyanin and 1-hydroxyphenazine as the sole carbon, nitrogen and energy sources.

Phylogenetic analysis of near full-length 16S rRNA gene of strain Rhodanobacter sp. PCA2 demonstrated that this isolate belongs to the genus Rhodanobacter in the family Rhodanobacteraceae belonging to the order Xanthomonadales within the class Gammaproteobacteria. Species in the genus Rhodanobacter show general characteristics including rod-shaped, yellow-pigmented and Gram-negative (Koh et al., 2015). Rhodanobacter species have been observed in various environments such as soil (Nalin et al., 1999, Kanaly et al., 2002), activated sludge from a wastewater treatment plant (Zhang et al., 2011) and biofilms (Lee et al., 2007) and can biodegrade a large number of compounds (Nalin et al., 1999, Kanaly et al., 2002, Zhang et al., 2011). For example, the first reported Rhodanobacter strain R. lindaniclasticus RP5557T utilizes gamma-hexachlorocyclohexane (γ-HCCH or lindane) as the sole C source (Nalin et al., 1999). Isolates R. xiangquanii BJQ-6T and Rhodanobacter sp. AYS5 are able to metabolically degrade the pesticide anilofos with 4-chloroaniline as the important building block (Zhang et al., 2011) and 1,4-dioxane via ring-cleavage (Pugazhendi et al., 2015), respectively. Due to the shared ring structures in these compounds degraded by Rhodanobacter, these studies suggest that the genus Rhodanobacter has the potential to degrade aromatic and heterocyclic compounds. Considering phenazines are a large class of aromatic heterocyclomers, it seems logical that the genus Rhodanobacter has potential to transform phenazines.

In this study, PCA was used as the sole carbon, nitrogen and energy source and its degradation was significantly and positively correlated with the growth of strain PCA2,

indicating that the transformation of PCA is a metabolic process rather than co- metabolic (Tran et al., 2013). Strain PCA2 utilized PCA at different rates according to the concentrations of PCA from 100 to 1,000 mg/L (from 0.45 to 4.46 mM), which is consistent with the pattern suggested by Edwards (1970). Specifically, essential nutrients, e.g. 100 and 200 mg/L PCA in this study, limit bacterial growth and other normal metabolic activities at low concentrations. By increasing nutrient availability, e.g. 500 mg/L PCA in this study, bacterial yields will be improved due to stimulation of the cellular metabolism. Further increasing substrate concentration, e.g. 1000 and 2000 mg/L PCA in this study, however, resulted in inhibition of cell growth via an unknown mechanism including possible reactions with component(s) of the cell, adsorption or complexing with enzymes or coenzymes, or dissociation of enzyme aggregates (Edwards, 1970). Rhodanobacter sp. PCA2 possessed an optimal degradation capability (46.2 µM h-1) at the concentration of 500 mg/L (2.23 mM) PCA. Sphingomonas sp. DP58 and M. fortuitum strain CT6 were reported to degrade PCA at rate of 29.7 µM h-1 (0.89 mM of PCA) (Yang et al., 2007) and 3.33 µM h-1 (0.20 mM of PCA) (Costa et al., 2015), respectively. The relative lower degradation rates of PCA by strains DP58 and CT6 compared with strain PCA2 might be caused by insufficient supply of the substrate.

In addition to PCA, strain PCA2 also degraded phenazine, pyocyanin and 1- hydroxyphenazine at rates of 5.67, 2.22 and 1.34 µM h-1, respectively. The ability to degrade different phenazines by PCA2 suggested that the isolate has the potential to transform the phenazine backbone consisting of nitrogen-containing and aromatic ring structures. Moreover, phenazines not only supplied carbon but also nitrogen for bacterial growth, suggesting that ring cleavage of nitrogen-containing ring must occur during the degradation of phenazines by strain PCA2. Mechanisms of PCA degradation by strain PCA2 including functional genes and proteins, intermediates and products are investigated in Chapter 3 and 4.

Compared with aromatic rings, carbon in carboxylate group is more accessible for bacterial growth. Combining gene knock-out mutation and heterologous expression revealed that decarboxylase encoded by the ubiD gene catalysed the first step of PCA degradation, which transformed PCA into phenazine in strain M. fortuitum ATCC 6841 (Costa et al., 2018). As expected, ubiD was found in strain PCA2 and shared 95% identity with that in M. fortuitum ATCC6841, suggesting that decarboxylation catalysed 42

by UbiD might be involved in PCA degradation by strain PCA2. However, the carboxylate moiety of PCA alone was insufficient for bacterial growth of strain PCA2 here. As reported by Costa et al. (2015), the carboxylic group of 120 µM PCA alone cannot support cell yield of Mycobacterium fortuitum CT6 from OD500 = 0.11 to OD500 6 7 = 0.42. According to the formula OD550 = 1 is equivalent to 2.8 × 10 - 10 Mycobacterium cells/mL (Chui et al., 2004), the cell growth of strain CT6 is 8.68 × 105- 106 cells/mL. This suggests the carboxylic group of 1 µM PCA is insufficient to support biomass yields of 7.2 × 104 cells/mL. In this study with strain PCA2, 1 µM of PCA produced a bacterial cell increase of 1.17 × 106 cells/mL, which is much higher than the case of strain CT6. Hence, the carbon source within the aromatic rings in PCA must be liberated via cleavage to support growth of strain PCA2.

Dioxygenases are well known for their activity in cleaving aromatic rings in low- molecular-weight (2-3 rings) polycyclic aromatic hydrocarbons (Mallick et al., 2011). Using allelic replacement in M. fortuitum ATCC6841, the MFORT_16269 gene annotated as an ortho-halobenzoate-1,2-dioxygenase by NCBI was identified as responsible for PCA degradation (Costa et al., 2015). Homologous sequence of the MFORT_16269 gene was also successfully amplified from strain PCA2 and shared high identity (95%) with the gene in M. fortuitum ATCC 6841. Research about these two functional genes suggested that strain PCA2 might share a similar PCA degradation pathway (e.g. ring cleavage by dioxygenases) with Mycobacterium spp.. Additionally, the different rates of degradation of different phenazine compounds such as PCA, phenazine, pyocyanin and 1-hydroxyphenazine by strain PCA2 suggests that the substituted groups of phenazines could affect affinity of responsible enzymes such as ortho-halobenzoate-1,2-dioxygenase to various phenazines.

Like other Rhodanobacter strains, PCA2 utilized different carbon sources for growth, including pyruvate, glucose, trehalose and alanine (Nalin et al., 1999, Zhang et al., 2011). Higher biomass was determined in cultures with pyruvate and alanine than the corresponding cultures with glucose and trehalose as substrates, indicating that PCA2 preferred C-chains over C-rings as sources to support growth.

In conclusion, this study isolated and purified the eighth known aerobic phenazines degrading bacterium Rhodanobacter sp. designated strain PCA2 belonging to Gammaproteobacteria from soil. Characterization of this strain also showed that strain

PCA2 degraded PCA metabolically, indicating that Rhodanobacter spp. can derive carbon, nitrogen and energy from secondary metabolites produced by cohabiting microorganisms e.g. Pseudomonas spp. and may play a role in turnover of phenazines in the environment. The ecological implications of the Rhodanobacter genus in community dynamics are not yet known. Moreover, ubiD gene and the homolog of MFORT_16269 previously characterized as responsible for PCA degradation were detected in strain PCA2 via PCR and confirmed via sequencing. The isolation of strain PCA2 presents opportunities to further understand the aerobic degradation of phenazine compounds.

References

Balch WE, Fox GE, Magrum LJ, Woese CR & Wolfe RS (1979) Methanogens: reevaluation of a unique biological group. Microbiological Reviews 43: 260-296.

Blankenfeldt W, Sakhtah H, Price-Whelan A, et al. (2013) Microbial phenazines: Biosynthesis, agriculture and health. Springer, New York.

Chui LW, King R, Lu P, Manninen K & Sim J (2004) Evaluation of four DNA extraction methods for the detection of Mycobacterium avium subsp. paratuberculosis by polymerase chain reaction. Diagnostic Microbiology and Infectious Disease 48: 39-45.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology 200: e00763-17.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Edwards VH (1970) The influence of high substrate concentrations on microbial kinetics. Biotechnology and Bioengineering 12: 679-712.

Kanaly RA, Harayama S & Watanabe K (2002) Rhodanobacter sp. strain BPC1 in a benzo[a]pyrene-mineralizing bacterial consortium. Applied and Environmental Microbiology 68: 5826-5833.

Kern SE & Newman DK (2014) Pseudomonas methods and protocols. Springer, New York.

Koh HW, Hong H, Min UG, Kang MS, Kim SG, Na JG, Rhee SK & Park SJ (2015) Rhodanobacter aciditrophus sp. nov., an acidophilic bacterium isolated from mine

wastewater. International Journal of Systematic and Evolutionary Microbiology 65: 4574-4579.

Lau GW, Ran H, Kong F, Hassett DJ & Mavrodi D (2004) Pseudomonas aeruginosa pyocyanin is critical for lung infection in mice. Infection and Immunity 72: 4275-4278.

Lee CS, Kim KK, Aslam Z & Lee ST (2007) Rhodanobacter thiooxydans sp. nov., isolated from a biofilm on sulfur particles used in an autotrophic denitrification process. International Journal of Systematic and Evolutionary Microbiology 57: 1775-1779.

Manefield M, Whiteley AS, Griffiths RI & Bailey MJ (2002) RNA stable isotope probing, a novel means of linking microbial community function to phylogeny. Applied and Environmental Microbiology 68: 5367-5373.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

Muyzer G, Ellen C. De Waal & Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59: 695-700.

Nalin R, Simonet P, Vogel TM & Normand P (1999) Rhodanobacter lindaniclasticus gen. nov., sp, nov., a lindane-degrading bacterium. International Journal of Systematic Bacteriology 49: 19-23.

Norman RS, Moeller P, McDonald TJ & Morris PJ (2004) Effect of pyocyanin on a crude- oil-degrading microbial community. Applied and Environmental Microbiology 70: 4004- 4011.

Pugazhendi A, Rajesh Banu J, Dhavamani J & Yeom IT (2015) Biodegradation of 1,4- dioxane by Rhodanobacter AYS5 and the role of additional substrates. Annals of Microbiology 65: 2201-2208.

Ran H, Hassett D & Lau G (2003) Human targets of Pseudomonas aeruginosa pyocyanin. Proceedings of the National Academy of Sciences of the United States of America 100: 14315-14320.

Si J, Li X-C & Cui B-K (2013) Decolorization of heterocycle dye neutral red by white- rot fungus Perenniporia subacida. Desalination and Water Treatment 52: 5594-5604.

Tamura K, Stecher G, Peterson D, Filipski A & Kumar S (2013) MEGA6: molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution 30: 2725- 2729.

Tran NH, Urase T, Ngo HH, Hu J & Ong SL (2013) Insight into metabolic and cometabolic activities of autotrophic and heterotrophic microorganisms in the biodegradation of emerging trace organic contaminants. Bioresource Technology 146: 721-731.

Tschech A & Pfennig N (1984) Growth yield increase linked to caffeate reduction in Acetobacterium woodii. Archives of Microbiology 137: 163-167.

Urakawa H, Martens-Habbena W & Stahl DA (2010) High abundance of ammonia- oxidizing archaea in coastal waters, determined using a modified DNA extraction method. Applied and Environmental Microbiology 76: 2129-2135.

Whiteley A & Bailey M (2000) Bacterial community structure and physiological state within an industrial phenol bioremediation system. Applied and Environmental Microbiology 66: 2400-2407.

Widdel F & Pfennig N (1981) Studies on dissimilatory sulfate-reducing bacteria that decompose fatty acids. Archives of Microbiology 129: 395-400.

Zhang J, Zheng JW, Hang BJ, Ni YY, He J & Li SP (2011) Rhodanobacter xiangquanii sp. nov., a novel anilofos-degrading bacterium isolated from a wastewater treating system. Current Microbiology 62: 645-649.

Zhang Z, Schwartz S, Wagner L & Miller W (2000) A greedy algorithm for aligning DNA sequences. Journal of Computational Biology 7: 203-214.

Chapter 3 Genome characterization and transcription analysis of the homolog of MFORT_16269 gene in Rhodanobacter sp. PCA2

3.1 Introduction

Due to their broad-spectrum antibiotic and redox activities, phenazine compounds play important roles in human health, ecological fitness of their producers and interactions with cohabiting microorganisms and plants (Ran et al., 2003, Lau et al., 2004, Mavrodi et al., 2006, Caldwell et al., 2009). Also, widespread application and improper disposal practices of phenazine dyes have resulted in pollution in aquatic environments (Gong et al., 2009). Therefore, characterising the fate of phenazines has relevance to ecological and human health as well as environmental protection.

To date only seven bacterial isolates belonging to two bacterial classes Alphaproteobacteria and Actinobacteria are known to link growth to phenazine biodegradation (Yang et al., 2007, Costa et al., 2015, Zhao et al., 2017, Costa et al., 2018). In the preceding chapter an eighth phenazine degrading bacterium, Rhodanobacter sp. PCA2 was isolated and subject to preliminary characterisation. Advances in technologies including genomics and gene deletion and expression in heterologous cells has generated additional information on phenazine-1-carboxylic acid (PCA) biodegradation (Costa et al., 2015, Zhao et al., 2017). The availability of genomic data from all phenazines degrading isolates except M. septicum DKN1213 enables a deeper understanding of their biochemistry and insight into the metabolism of PCA. Using mutants of M. fortuitum ATCC6841 and heterologous expression in E. coli, it was revealed that decarboxylation was the first step of PCA degradation, which was catalysed by the UbiD enzyme encoded by MFORT_16229, a gene previously known for involvement in ubiquinone biosynthesis (Costa et al., 2018). In the JGI database MFORT_16269 was predicted to encode a phenylpropionate dioxygenase and was ultimately identified as responsible for PCA transformation via allelic replacement in M. fortuitum ATCC6841 (Costa et al., 2015).

Rhodanobacter sp. PCA2 isolated in this study is the first phenazines-degrading Gammaproteobacteria. The genus Rhodanobacter is known for biodegradation of a large number of ring-containing compounds such as gamma-hexachlorocyclohexane (γ- HCCH or lindane) (Nalin et al., 1999), 4-chloroaniline (Zhang et al., 2011) and 1,4- dioxane (Pugazhendi et al., 2015) and are likely to cleave aromatic and heterocyclic rings which constitute the main structure of phenazines. However, no homolog of the MFORT_16269 gene is detected in currently available genomes of bacteria in the Rhodanobacter genus. Therefore, exploring genomics of strain PCA2 is important in elucidating the genetic and metabolic nature of PCA degradation.

In this chapter, the genomic sequence of Rhodanobacter sp. PCA2 was analyzed and compared to genomes of Sphingomonas wittichii DP58, Mycobacterium fortuitum CT6 and Rhodanobacter denitrificans 2APBS1. The former two genomes were selected here because they are PCA-degrading bacteria from different phylogenetic phyla. Rhodanobacter denitrificans 2APBS1 is not a PCA-degrader but is the closest relative of strain PCA2 that is physiologically well-defined. Additionally, transcriptomic analysis of the homolog of the MFORT_16269 in strain PCA2 was undertaken.

In this work, an orthologue of MFORT_16269 was found to be highly expressed throughout PCA degradation and to be located on a plasmid instead of the chromosome in strain PCA2. A gene encoding UbiD but not UbiX was found in the chromosome of strain PCA2, though a complete ubiquinone biosynthetic pathway was predicted by the genome. Additionally, based on comparative analysis of COG categories and groups, strain PCA2 was characterized with physiological properties and life style properties such as cell motility and chemotaxis, distinct from PCA-degrading Actinobacteria and Alphaproteobacteria.

3.2 Materials and methods

3.2.1 Genomic DNA, RNA and plasmid extraction

The genomic DNA of Rhodanobacter sp. PCA2 was extracted using phenol- chloroform-isomylalcohol and precipitated with isopropanol, and was then resuspended in molecular grade water as described in Section 2.2.7.

The genomic RNA of Rhodanobacter sp. PCA2 were extracted with PureLink○R RNA Mini Kit (ThermoFisher Scientific, Australia), according to the manufacturers’ protocol. Briefly, log-phase cells were harvested by centrifuge at 500x g for 5 min at 4°C and were lysed by lysozyme solution and lysis buffer. Then the cell lysate was homogenized with an 18-gauge needle and was mixed with ethanol before transferring to a Spin Cartridge with a Collection Tube. Then 30 µL RNase-free water was used to elute RNA after washing the Spin Cartridge with Wash Buffers. The extracted RNA concentration was measured with a Qubit instrument (Qubit○R 2.0 Fluorometer, ThermoFisher Scientific, Australia) according to the manufacturer’s protocol.

The bacterial plasmid was extracted using PureYieldTM Plasmid Miniprep System by following the manufacturer’s protocol (Promega, Australia). To remove sheared chromosomal DNA, plasmid extract was digested with Plasmid-SafeTM ATP-Dependent DNase (Lucigen, USA) (Jones & Marchesi, 2007), according to the manufacturer’s instruction. Genomic and plasmid DNA concentration were measured with a Qubit instrument according to the manufacturer’s protocol (Qubit, Life Technologies, US). The quality of genomic and plasmid DNA extracted were checked using agarose gel electrophoresis, described in Appendix 8.3.

3.2.2 Genome sequencing and assembly

Genomic DNA were sequenced by the Ramaciotti Cenre for Genomics (UNSW Sydney, Australia) using an Illumina Miseq v2 Micro 2x150 bp paired-end sequencer. From the sequencing of the genomic DNA, a total of 62 contigs and 4,822,346 paired- end reads with a read-length of 151 bases were generated, respectively. This provided an average of 320x coverage. Sequences analyses including quality control and assembly were performed on the platform of a computational cluster Katana offered by UNSW and all codes were described in Appendix 8.5. For quality control, only paired- reads where more than 70% bases scored higher than 20 were kept after trimming the sequence reads with the FastQC version 0.10.1 (Brown et al., 2017) and SolexaQA software package (Cox et al., 2010). Post-trimming, sequences were assembled to scaffolds with SPAdes version 3.7.0 (Bankevich et al., 2012) before ordering the scaffolds as the reference chromosomal sequences of Rhodanobacter denitrificans 2APBS1 with Mauve 2015-02-25 (Darling et al., 2004). Finally, the draft chromosomal sequences were generated with sizes of 3,912,255 base pairs.

3.2.3 Genome annotation and analysis

Open reading frames (ORFs) and annotation for the chromosomal sequences of Rhodanobacter sp. PCA2 were determined by submitting the sequences to Joint Genome Institute (JGI, https://jgi.doe.gov/) (Huntemann et al., 2015) and ORFs of interest were searched against the COG and KEGG. The tools in IMG-ER database were used for genomic comparison of strain PCA2 with the genomes of other PCA- degrading and closely related bacteria deposited in the same database. All locus tags described in this study are based on JGI database.

3.2.4 Detection of ubiD, the homolog of MFORT_16269 and 16S rRNA genes

To check the purity, plasmid extract treated with Plasmid-SafeTM ATP-Dependent DNase were amplified using primers pair 338F/ 530R (Section 2.2.8).

To detect occupation of ubiD and the homolog of MFORT_16269 genes on plasmid, the purified plasmid extract were amplified using primers pairs ubiD-20F/ubiD-1545R (Section 2.2.9) and 16269F/16269R (Section 2.2.9), respectively.

3.2.5 Quantitative PCR (qPCR) and Reverse-transcript qPCR (RT-qPCR) of the homolog of MFORT_16269

To explore profiles of the homolog of MFORT_16269 responsible for PCA degradation, extracted genomic DNA throughout PCA degradation was used for qPCR. Negative control reactions without DNA template were also performed in all PCR analyses.

Standards curve for qPCR and RT-qPCR. The homolog of MFORT_16269 was amplified with primers pair 16269F/16269R (Section 2.2.9) from extracted genomic DNA of strain PCA2. The PCR product was checked using agarose gel electrophoresis and purified with a purification kit (DNA Clean & ConcentratorTM-25, Zymo Research, USA) before ligated into the pGEM-T Easy Vector (Promega, Australia). Then the ligation products were transformed into Escherichia coli JM109 competent cells (Promega, Australia) by following the instruction of the manufacturer. After re- amplification with the vector-specific primers T7 and SP6, the positive clones were subjected to extract plasmid DNA using PureYieldTM Plasmid Miniprep System (Promega, Australia) by following the manufacturer’s protocol. The plasmid DNA concentration was measured with a Qubit instrument (Qubit○R 2.0 Fluorometer, ThermoFisher Scientific, Australia) according to the manufacturer’s protocol and the 52

copy number of the homolog of MFORT_16269 were calculated directly from the concentration of plasmid. Ten-fold serial dilutions of the known copy number of the plasmid were used to quantitative PCR assay in triplicate as an external standard curve.

Quantitative PCR. qPCR reaction was performed in 10 µL volumes containing 5 µL SsoAdvancedTM Universal SYBR® Green Supermix (Bio-Rad, Australia), 1 µg BSA, 1 µM of each forward and reverse primers and 1 µL of diluted template DNA (~ 5 ng). The qPCR reactions were performed in triplicates on a Bio-Rad CFX96 set (Bio-Rad, Australia) using the following thermal conditions: 10 min at 50°C , 5 min at 95°C , 45 cycles of 10 s at 95°C and 30 s at 58°C , 1 min at 95°C , 1 min at 55°C , followed by a plate reading step at for 55°C 10 s.

Reverse-Transcript qPCR. RT-qPCR reaction was performed in 50 µL volumes containing 25 µL of 2X SYBR○R Green RT-PCR Reaction Mix (Bio-Rad, Australia), 1 µL of iScript Reverse Transcriptase for One-Step RT-PCR (Bio-Rad, Australia), 300 nmol of each forward and reverse primers and 1 µL of diluted template RNA (~ 5 ng). The qPCR reactions were performed in triplicates on a Bio-Rad CFX96 set (Bio-Rad, Australia) using the thermal conditions set in qPCR reaction.

3.3 Results

3.3.1 Investigation of the homolog of MFORT_16269 in strain PCA2

Given that the homologs of MFORT_16269 have previously been characterized as responsible for PCA transformation in Mycobacterium, the abundance and activity of this gene was monitored in Rhodanobacter sp. PCA2 using qPCR and RT-qPCR respectively, throughout PCA degradation described in Section 2.3.2. Figure 3.1 shows the degradation of PCA by strain PCA2 at rate of 46.2 µM h-1 and cell yield of 5.39 × 107 cells/mL, while no consumption of PCA or cell yield was observed in sterile cultures. Increases of gene copies and transcript copies of the homolog of MFORT_16269 were found from 8.12 × 106 to 5.81 × 108 and from 7.78 × 106 to 1.54 × 109 copies/mL, which were 10.6 and 28.4 fold more than the corresponding cell yield, respectively (Figure 3.1). This suggested not only that gene transcription increased during PCA degradation but curiously that the specific gene copy number was higher than the cell number suggestive of multiple copies.

3.3.2 General properties of genome of strain PCA2

Rhodanobacter sp. PCA2 is an aerobic Gammaproteobacterium that utilizes PCA and other phenazines including phenazine, pyocyanin and 1-hydroxyphenazine as the sole carbon, nitrogen and energy source (details described in Chapter 2). The genome of strain PCA2 was sequenced using Illumina technology, assembled and annotated. The draft genome had a total size of 3,909,686 base pairs with a G+C content of 68.64%, and predicted a total of 3387 protein coding genes, where 2792 (81.07%) of those genes had been annotated with functions (Table 3.1). Table 3.2 summarises protein coding genes assigned to COG categories amongst strain PCA2, two PCA-degrading bacteria Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6 and a close relative of strain PCA2, Rhodanobacter denitrificans strain 2APBS1. Generally, Rhodanobacter strains PCA2 and 2APBS differentiated with the two PCA-degrading strains CT6 and

DP58 in physiological properties and life styles. Genes related to lipid transport and metabolism (I) and secondary metabolites biosynthesis, transport and catabolism (Q) were underrepresented in the genome of Rhodanobacter group in comparison to strains CT6 and DP58 (Table 3.2). In contrast, genes functioning in cell wall/membrane/envelope biogenesis (M), cell motility (N) and posttranslational modification, protein turnover and chaperones (O) had higher relative abundance in the Rhodanobacter spp. than the phenazine degrading strains CT6 and DP58 (Table 3.2). Complete sets of genes related to flagellar assembly and pili (Type IV) were found in the genome of strain PCA2.

Table 3.1 Comparison in general features of the genomes of Rhodanobacter sp. PCA2, Mycobacterium fortuitum CT6, Sphingomonas wittichii DP58 and Rhodanobacter denitrificans 2APBS1

Rhodanobacter Rhodanobacter Mycobacterium Sphingomonas Feature denitrificans sp. PCA2 fortuitum CT6 wittichii DP58 2APBS1 Genome size 3,909,686 6,254,616 5,628,887 4,225,490 (bp) G+C content of chromosomal 68.64% 66.22% 67.80% 67.45% DNA Number of 16S 1 2 1 2 rRNA genes

Total protein 3387 6003 5669 3899 coding genes

Genes with function 2792 4581 4485 3063 prediction Genes with 2467 3844 3492 2617 COGs

Table 3.2 Comparison of genes assigned with COG functional categories in Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58, Mycobacterium fortuitum CT6 and Rhodanobacter denitrificans 2APBS1.

Gene count (percentage in genes with COGs)

Rhodanobacter COG code COG functional category Rhodanobacter sp. Mycobacterium Sphingomonas denitrificans PCA2 fortuitum CT6 wittichii DP58 2APBS1

A RNA processing and modification 1 (0.04%) 1 (0.02%) 0 (0.00%) 1 (0.03%) B Chromatin structure and dynamics 1 (0.04%) 0 (0.00%) 3 (0.07%) 1 (0.03%) C Energy production and conversion 176 (6.32%) 316 (6.98%) 280 (6.75%) 186 (6.38%) Cell cycle control, cell division, D 38 (1.36%) 33 (0.73%) 31 (0.75%) 37 (1.27%) chromosome partitioning E Amino acid transport and metabolism 226 (8.12%) 313 (6.92%) 277 (6.67%) 206 (7.06%)

F Nucleotide transport and metabolism 66 (2.37%) 89 (1.97%) 63 (1.52%) 64 (2.19%)

G Carbohydrate transport and metabolism 143 (5.14%) 203 (4.49%) 156 (3.76%) 116 (3.98%)

H Coenzyme transport and metabolism 134 (4.81%) 305 (6.74%) 208 (5.01%) 137 (4.70%)

I Lipid transport and metabolism 117 (4.20%) 475 (10.49%) 398 (9.59%) 124 (4.25%) Translation, ribosomal structure and J 208 (7.47%) 196 (4.33%) 185 (4.46%) 214 (7.34%) biogenesis K Transcription 182 (6.54%) 458 (10.12%) 327 (7.88%) 183 (6.28%)

L Replication, recombination and repair 99 (3.56%) 122 (2.70%) 117 (2.82%) 130 (4.46%)

M Cell wall/membrane/envelope biogenesis 220 (7.90%) 205 (4.53%) 192 (4.63%) 221 (7.58%)

N Cell motility 93 (3.34%) 13 (0.29%) 52 (1.25%) 100 (3.43%) Posttranslational modification, protein O 157 (5.64%) 138 (3.05%) 155 (3.73%) 164 (5.62%) turnover, chaperones

P Inorganic ion transport and metabolism 163 (5.85%) 241 (5.32%) 304 (7.33%) 162 (5.56%)

Secondary metabolites biosynthesis, Q 67 (2.41%) 326 (7.20%) 339 (8.17%) 60 (2.06%) transport and catabolism

R General function prediction only 211 (7.58%) 565 (12.48%) 504 (12.14%) 219 (7.51%)

S Function unknown 151 (5.42%) 230 (5.08%) 208 (5.01%) 193 (6.62%)

T Signal transduction mechanisms 139 (4.99%) 142 (3.14%) 134 (3.23%) 153 (5.25%) Intracellular trafficking, secretion, and U 49 (1.76%) 24 (0.53%) 70 (1.69%) 48 (1.65%) vesicular transport V Defence mechanisms 94 (3.38%) 112 (2.47%) 106 (2.55%) 111 (3.81%)

W Extracellular structures 39 (1.40%) 2 (0.04%) 24 (0.58%) 40 (1.37%)

X Mobilome: prophages, transposons 10 (0.36%) 17 (0.38%) 17 (0.41%) 45 (1.54%)

Z Cytoskeleton 0 (0.00%) 0 (0.00%) 0 (0.00%) 1 (0.03%)

3.3.2.1 Genes encoding UbiD and UbiX

The 3-Octaprenyl-4-hydroxybenzoate decarboxylase known as UbiD was recently identified as responsible for the first step of PCA transformation to phenazine (Costa et al., 2018), hence, putative homologs of the gene encoding this decarboxylase were investigated here. UbiD was predicted in the genomes of all three PCA-degrading bacteria compared though with low sequence identity (Table 3.3). In contrast, the gene homolog encoding flavin prenyltransferase (UbiX) catalysing synthesis of prenylated flavin mononucleotide (prFMN) which serves as a cofactor in the UbiD active site was only found in strains CT6 (locus tag Ga0100919_111654) and DP58 (locus tag MSEDRAFT_04478), not in the assembled chromosomal DNA of strain PCA2 (Table 3.3).

Table 3.3 Genes encoding UbiD and UbiX identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6.

Locus tag

Protein name Rhodanobacter Sphingomonas Mycobacterium sp. PCA2 wittichii DP58 fortuitum CT6 (Ga0192503_) (MSEDRAFT_) (Ga0100919_) 3-octaprenyl-4- hydroxybenzoate 11912 (29%)* 04481 (36%) 111652 (99%) decarboxylase (UbiD, EC 4.1.1.98 ) Flavin prenyltransferase — 04478 (50%) 111654 (99%) (UbiX, EC 2.5.1.129)

*, percentage in brackets means identity of UbiD and UbiX in strains PCA2, DP58 and CT6 shared with the corresponding enzymes in M. fortuitum ATCC6841.

3.3.2.2 Homologs of MFORT_16269 gene encoding phenylpropionate dioxygenase

The MFORT_16269 gene annotated as the large subunit of phenylpropionate dioxygenase in the JGI database was reported responsible for PCA degradation, hence, putative homologs of this genes were investigated here. The corresponding homologs in genomes of strain CT6 (locus tag Ga0100919_ 111660) and strain DP58 (locus tag MSEDRAFT_05237) shared 99% and 41% identity with the MFORT_16269 gene, respectively (Table 3.4). Surprisingly, the homolog of this functional gene was not observed in the assembled genome of strain PCA2, though a sequence sharing 95% identity with MFORT_16269 was successfully amplified from DNA extracts of strain PCA2 as described in Section 2.3.5, suggesting this gene has an extrachromosomal 58

location such as a plasmid in strain PCA2. Genes encoding the small subunit of phenylpropionate dioxygenase are also summarized in Table 3.4.

Table 3.4 Phenylpropionate dioxygenases encoded by the homologs of MFORT_16269 identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6.

Locus tag

Name Rhodanobacters Sphingomonas wittichii Mycobacterium p. PCA2 DP58 fortuitum CT6 (Ga0192503_) (MSEDRAFT_) (Ga0100919_ ) Phenylpropionate /trans- cinnamate dioxygenase — 05237 (41%)* 111660 (99%) (Large subunit) Phenylpropionate /trans- cinnamate dioxygenase — 05238 (36%) 111659 (100%) (Small subunit)

*, percentage in brackets means identity of phenylpropionate /trans-cinnamate dioxygenase large and small subunits in strains DP58 and CT6 shared with the corresponding enzymes in M. fortuitum ATCC6841.

3.3.2.3 Genes encoding amidohydrolase and pyocyanin demethylase

Based on the pathway of PCA degradation proposed by Chen et al. (2008), amidohydrolase might be involved in cleavage of the nitrogen-containing ring in PCA, and was hence analysed here using comparative genomics. Despite 17 genes encoding amidohydrolase being predicted from the genome of strain M. fortuitum ATCC6841, only one (locus tag Ga0124153_130528) was located in the same cluster with the MFORT_16269 gene (locus tag Ga0124153_130525). The homologs of this amidohydrolase-encoding gene (locus tag Ga0124153_130528) were detected with 99%, 35% and 26% identity in strains CT6 (locus tag Ga0100919_111662), DP58 (locus tag MSEDRAFT_00023) and PCA2 (locus tag Ga0192503_120149), respectively (Table 3.5). Pyocyanin demethylase (PodA) encoded by the MFORT_14352 gene was proven responsible for pyocyanin degradation via demethylation in M. fortuitum ATCC6841 (Costa et al., 2017). Though both strain PCA2 and CT6 were able to degrade pyocyanin, the homolog of MFORT_14352 was only detected in strain CT6 with 99% identity (locus tag Ga0100919_111685) but not in strain PCA2 or strain DP58 (Table 3.5).

Table 3.5 Homologs of genes encoding amidohydrolase and pyocyanin demethylase identified in genomes of Rhodanobacter sp. PCA2, Sphingomonas wittichii DP58 and Mycobacterium fortuitum CT6.

Locus tag Rhodanobacter Sphingomonas Mycobacterium Protein name sp. PCA2 wittichii DP58 fortuitum CT6 (Ga0192503_) (MSEDRAFT_) (Ga0100919_)

Amidohydrolase 120149 (26%)* 00023 (35%) 111662 (99%)

Pyocyanin demethylase — — 111685 (99%)

*, percentage in brackets means identity of amidohydrolase and pyocyanin demethylase in strains PCA2, DP58 and CT6 shared with the corresponding enzymes in M. fortuitum ATCC6841.

3.3.3 Detection of ubiD and the homolog of MFORT_16269 from plasmid DNA extracts in strain PCA2

Based on genomic and transcriptomic analyses, we hypothesized that ubiD and the homolog of MFORT_16269 were located on a plasmid rather than on the chromosome of strain PCA2. Therefore, an attempt was made to purify a plasmid from strain PCA2 genomic DNA with Plasmid-safeTM ATP-dependent DNase before amplifying ubiD and the homolog of MFORT_16269. As a control, an amplification the 16S rRNA gene sequence was made with primers 338F/530R to check the purity of the plasmid extract. As shown in Figure 3.2, no partial 16S rRNA gene was observed in the plasmid DNA extract digested with Plasmid-safeTM ATP-dependent DNase, suggesting the plasmid was purified and therefore suitable for further analyses. The ubiD and the homolog of MFORT_16269 were successfully amplified from the purified plasmid DNA and sequenced by Sanger sequencing which showed the same sequences as described in Section 2.3.5. Taken together, these results are consistent with the conclusion that ubiD and the homolog of MFORT_16269 are plasmid borne in contrast to the chromosomally located gene cluster found in previous PCA-degrading bacteria (Costa et al., 2015).

Figure 3.2 Agarose gel electrophoresis profiles of ubiD and the homolog of MFORT_16269 amplified from purified plasmid extract. Partial 16S rRNA gene was amplified as a control to check the purity of plasmid extract. It showed that both ubiD and the homolog of MFORT_16269 were plasmid born in strain PCA2.

3.4 Discussion

The MFORT_16269 gene from Mycobacterium fortuitum ATCC6841 is responsible for PCA degradation (Costa et al., 2015). Homologs of this gene have been observed in the genomes of all known PCA-degrading bacteria (Zhao et al., 2017, Costa et al., 2018). Curiously, the homolog of MFORT_16269 was not observed in the PCA2 genome sequenced and assembled in this study. Despite this, the homolog can be amplified from genomic DNA extracts of PCA2 and transcription of the homolog in PCA2 increased 2.68-fold more than the gene copies did throughout PCA degradation, suggesting that this gene was active during the transformation of PCA by strain PCA2. Strikingly, the gene copy number for the homolog of MFORT_16269 was 10.6 times higher than the corresponding cell yields throughout PCA degradation by strain PCA2, consistent with this active gene being located on a multi-copy plasmid rather than on the chromosome as is the case in all other PCA-degrading bacteria (Costa et al., 2015).

Additional evidence for the location of the homolog of MFORT_16269 on a plasmid was obtained by successful amplification from chromosomally depleted DNA extracts. The sequenced product shared high homology (95%) with the corresponding gene in M.

fortuitum ATCC6841, suggesting that the homolog might have been transferred to strain PCA2 from other PCA-degrading bacteria such as a Mycobacterium spp.. The fact that Mycobacterium spp. were also found in the soil where strain PCA2 was isolated (data not shown) indirectly supports the transfer hypothesis here.

Plasmid-mediated gene transfer is important not only in the spread of antibiotic resistance genes but also in dissemination of degradative pathways. Plasmids thereby support bacteria with various accessory traits and play essential roles in rapid bacterial adaption to different environmental conditions (Smalla et al., 2015). For example, strains belonging to Beta- and Gamma-proteobacteria obtained the ability to degrade 3- chlorobenzoate by adoption of plasmid pBRC60 from Alcaligenes sp. BR60 (Fulthorpe & Wyndham, 1991). Lateral gene transfer via plasmid was also observed in Rhodanobacter populations in heavy metal contaminated groundwater, which resulted in the dominance of this genus in uranium-contaminated sites (Hemme et al., 2016). Confirmation of the hypothesis that the homolog of MFORT_16269 was encoded on a plasmid in PCA2 was hindered by failure of plasmid sequencing caused by low concentration of purified plasmid extract from strain PCA2. It remains to elucidate the source of ubiD and MFORT_16269 genes in strain PCA2 with direct evidence.

MFORT_16269 gene was annotated as an ortho-halobenzoate-1,2-dioxygenase and phenylpropionate dioxygenase in NCBI and JGI, respectively (Costa et al., 2015). Given that genomic analyses including genomic comparison were processed using the JGI database, phenylpropionate dioxygenase was used as the protein product of the MFORT_16269 gene for all analyses in this study. Dioxygenases are extensively studied for their activities in cleaving rings, especially aromatic rings in polycyclic aromatic hydrocarbons (Mallick et al., 2011), of which phenylpropionate dioxygenase was widely detected in various bacteria including Mycobacterium spp. (Stingley et al., 2004, Zhang & Anderson, 2012). It was reported that phenylpropionate dioxygenase is evolutionarily related to polycyclic aromatic hydrocarbon dioxygenases (Seo et al., 2009). For instance, the small subunit of phenylpropionate dioxygenase was contained in a conserved domain of the putative protein sequence in a phthalate-degrading strain M. vanbaalenii PYR-1 (Stingley et al., 2004). However, phenylpropionate dioxygenase has never been reported to be involved in ring cleavage by Rhodanobacter spp..

Compared to aromatic rings, the carboxyl group of PCA is a preferable source for bacterial growth. UbiD known for involvement in ubiquinone biosynthesis was recently

characterized as responsible for decarboxylation as the first step of PCA transformation in all known PCA-degrading Actinobacteria (Costa et al., 2018). Prenylated flavin mononucleotide (prFMN) synthesized by the synthase UbiX is utilized as a cofactor in the active site of UbiD (Payne et al., 2015), hence, UbiX is involved in PCA degradation by Actinobacteria as well (Costa et al., 2018). Based on genomic data, a complete ubiquinone biosynthetic pathway was found in both strains PCA2 and DP58. Both ubiD and ubiX genes were detected in the Alphaproteobacterial PCA-degrading bacterium Sphingomonas sp. DP58 (Ma et al., 2012), and decarboxylation was also found during PCA degradation by strain DP58 (Zhao et al., 2017), which is consistent with the situation in PCA-degrading Actinobacteria. However, only ubiD but not ubiX was predicted from chromosome of strain PCA2. Curiously, ubiD was also successfully amplified from plasmid DNA. Considering that ubiD located next to the MFORT_16269 gene in M. fortuitum ATCC6841, it is possible that the ubiD and ubiX genes responsible for PCA degradation were transferred from Mycobacterium spp. together with the MFORT_16269 gene, while the ubiD without ubiX adjacent on the chromosome is functional in ubiquinone biosynthesis in strain PCA2. Deletion of ubiX in Escherichia coli MC4100 resulted a significantly reduction of ubiquinone biosynthesis during logarithmic growth, but had no effect on ubiquinone production in the following stationary phase, suggesting that ubiX was not indispensable for biosynthesis of ubiquinone in E. coli (Gulmezian et al., 2007). The similar mechanism of ubiquinone biosynthesis might be employed in strain PCA2. Further studies are required to verify this hypothesis.

Given that strain M. fortuitum CT6 and strain PCA2 are capable of utilizing PCA as the sole nitrogen source (Costa et al., 2015), cleavage of the nitrogen-containing ring in PCA is necessary for survival and growth of strains CT6 and PCA2. Additionally, based on the pathway of PCA degradation proposed by Chen et al. (2008), PCA was transformed into 6-(2-carboxyphenylimino)-2,4-cyclohexadien-1-one oxime via hydrolyzation of the nitrogen-containing ring by strain DP58. Therefore, amidohydrolase was proposed to be involved in PCA degradation. A gene (locus tag Ga0100919_111662) encoding amidohydrolase was found in the same cluster with the homolog of MFORT_16269 gene in strain CT6, and its homologs were detected in both strain DP58 and PCA2, suggesting that this gene might play a role in cleavage of the

nitrogen-containing ring in PCA transformation. Further studies such as heterologous expression would be required to test this hypothesis.

Pyocyanin was utilized as an alternative carbon, nitrogen and energy source by strain PCA2 and CT6. It was revealed that pyocyanin demethylase (PodA) encoded by the MFORT_14352 gene catalysed demethylation during pyocyanin degradation by M. fortuitum ATCC6841 (Costa et al., 2017). However, the homolog of this gene was not detected in strain PCA2, suggesting pyocyanin was transformed via a different pathway in this isolate and alternatively as is the case of ubiD and the homolog of MFORT_16269. Further studies such as amplification of the MFORT_14352 gene from plasmid DNA in strain PCA2 would be required to test this hypothesis.

Additionally, a number of genes involved in bacterial motility including flagella assembly and pili were identified in the genome of strain PCA2. The presence of flagella has been reported in some species in the Rhodanobacter genus such as R. koreensis THG-DD7T (Won et al., 2015) and R. fulvus Jip2T (Im et al., 2004) but not in strains R. xiangquanii BJQ-6T (Zhang et al., 2011) and Rhodanobacter sp. BPC1 (Kanaly et al., 2002). Chemotaxis enables bacteria to recognise chemical gradients (Girgis et al., 2007). It would be interesting to investigate this in strain PCA2.

In conclusion, the homolog of the MFORT_16269 gene was active in PCA degradation and was plasmid borne in strain Rhodanobacter sp. PCA2, in contrast to the chromosomally located gene cluster found in other known PCA-degrading bacteria. Likewise, ubiD was present in plasmid DNA in strain PCA2. Considering that these two functional genes are absent in other known Rhodanobacter but widely available in Mycobacterium, the ubiD and the homolog of the MFORT_16269 gene might have been transferred into Rhodanobacter sp. strain PCA2 from Mycobacterium spp. via horizontal gene transfer. Further experimentation such as knocking out the genes in vivo and expressing in vitro could be performed to directly verify the functions of ubiD and the homolog of the MFORT_16269 genes in strain PCA2. Additionally, a gene encoding amidohydrolase was likely to catalyse cleavage of the nitrogen-containing ring during PCA degradation. More research is required to confirm these hypotheses.

References

Bankevich A, Nurk S, Antipov D, et al. (2012) SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. Journal of Computational Biology 19: 455-477.

Brown J, Pirrung M & McCue LA (2017) FQC Dashboard: integrates FastQC results into a web-based, interactive, and extensible FASTQ quality control tool. Bioinformatics 33: 3137-3139.

Costa KC, Glasser N, Conway SJ & Newman DK (2017) Pyocyanin degradation by a tautomerizing demethylase inhibits Pseudomonas aeruginosa biofilms. Science 355: 170- 173.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology 200: e00763-17.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Cox MP, Peterson DA & Biggs PJ (2010) SolexaQA: At-a-glance quality assessment of Illumina second-generation sequencing data. BMC Bioinformatics 11: 485.

Darling ACE, Mau B, Blattner FR & Perna NT (2004) Mauve: multiple alignment of conserved genomic sequence with rearrangements. Genome Research 14: 1394-1403.

Fulthorpe RR & Wyndham RC (1991) Transfer and expression of the catabolic plasmid pBRC60 in wild bacterial recipients in a freshwater ecosystem. Applied and Environmental Microbiology 57: 1546-1553.

Girgis HS, Liu Y, Ryu WS & Tavazoie S (2007) A comprehensive genetic characterization of bacterial motility. PLoS Genetics 3: 1644-1660.

Gulmezian M, Hyman KR, Marbois BN, Clarke CF & Javor GT (2007) The role of UbiX in Escherichia coli coenzyme Q biosynthesis. Arch Biochem Biophys 467: 144-153.

Hemme CL, Green SJ, Rishishwar L, et al. (2016) Lateral gene transfer in a heavy metal- contaminated-groundwater microbial community. MBio 7: e02234-02215.

Huntemann M, Ivanova NN, Mavromatis K, et al. (2015) The standard operating procedure of the DOE-JGI Microbial Genome Annotation Pipeline (MGAP v.4). Standards in Genomic Sciences 10: 86.

Im W-T, Lee ST & Yokota A (2004) Rhodanobacter fulvus sp. nov., a β-galactosidase- producing gammaproteobacterium. The Journal of General and Applied Microbiology 50: 143-147.

Jones BV & Marchesi JR (2007) Transposon-aided capture (TRACA) of plasmids resident in the human gut mobile metagenome. Nature Methods 4: 55-61.

Kanaly RA, Harayama S & Watanabe K (2002) Rhodanobacter sp. strain BPC1 in a benzo[a]pyrene-mineralizing bacterial consortium. Applied and Environmental Microbiology 68: 5826-5833.

Lau GW, Ran H, Kong F, Hassett DJ & Mavrodi D (2004) Pseudomonas aeruginosa pyocyanin is critical for lung infection in mice. Infection and Immunity 72: 4275-4278.

Ma Z, Shen X, Hu H, Wang W, Peng H, Xu P & Zhang X (2012) Genome sequence of Sphingomonas wittichii DP58, the first reported phenazine-1-carboxylic acid-degrading strain. Journal of Bacteriology 194: 3535-3536.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

Nalin R, Simonet P, Vogel TM & Normand P (1999) Rhodanobacter lindaniclasticus gen. nov., sp, nov., a lindane-degrading bacterium. International Journal of Systematic Bacteriology 49: 19-23.

Payne KA, White MD, Fisher K, et al. (2015) New cofactor supports alpha,beta- unsaturated acid decarboxylation via 1,3-dipolar cycloaddition. Nature 522: 497-501.

Pugazhendi A, Rajesh Banu J, Dhavamani J & Yeom IT (2015) Biodegradation of 1,4- dioxane by Rhodanobacter AYS5 and the role of additional substrates. Annals of Microbiology 65: 2201-2208.

Ran H, Hassett D & Lau G (2003) Human targets of Pseudomonas aeruginosa pyocyanin. Proceedings of the National Academy of Sciences of the United States of America 100: 14315-14320.

Seo JS, Keum YS & Li QX (2009) Bacterial degradation of aromatic compounds. International Journal Environmental Research Public Health 6: 278-309.

Smalla K, Jechalke S & Top EM (2015) Plasmid detection, characterization and ecology. Microbiology Spectrum 3: PLAS-0038-2014.

Stingley RL, Brezna B, Khan AA & Cerniglia CE (2004) Novel organization of genes in a phthalate degradation operon of Mycobacterium vanbaalenii PYR-1. Microbiology 150: 3749-3761.

Won K, Singh H, Ngo HT, Son H, Kook M, Kim KY & Yi TH (2015) Rhodanobacter koreensis sp. nov., a bacterium isolated from tomato rhizosphere. International Journal of Systematic and Evolutionary Microbiology 65: 1180-1185.

Zhang C & Anderson AJ (2012) Multiplicity of genes for aromatic ring-hydroxylating dioxygenases in Mycobacterium isolate KMS and their regulation. Biodegradation 23: 585-596.

Chapter 4 Metabolomic and proteomic investigations of PCA degradation by Rhodanobacter sp. PCA2

4.1 Introduction

Phenazines are mainly produced and excreted into the environment as redox-active secondary metabolites by a large number of bacterial strains (Mavrodi et al., 2006, Wang et al., 2010). Owing to their roles in advantaging their producers and effects on microbial communities (Norman et al., 2004, Dietrich et al., 2008, Wang et al., 2010), fate of phenazines including degradation is of importance. To date, eight pure phenazines-degrading bacteria have been isolated (Yang et al., 2007, Costa et al., 2015, Zhao et al., 2017, Costa et al., 2018 and in this study). Phenylpropionate dioxygenase encoded by the MFORT_16269 gene was verified responsible for PCA degradation by knockout mutation of strain M. fortuitum ATCC6841 (Costa et al., 2015). Also, UbiD and UbiX were implicated in decarboxylation as the first step of PCA degradation by all PCA-degrading Actinobacteria (Costa et al., 2018). Additionally, two products of metabolic degradation of PCA, i.e. 4-hydroxy-1-(2-carboxyphenyl) azacyclobut-2-ene- 2-carbonitrile (HPAEC) and 4-hydroxy-1-(2-carboxyphenyl)-2-azetidinecarbonitrile (HPAC), were determined in pure cultures inoculated with strain Sphingomonas sp. DP58 using gas chromatography (GC)–mass spectroscopy (MS) and 1H-nuclear magnetic resonance (NMR) analysis (Chen et al., 2008). Phenazine was identified as an intermediate of PCA degradation in all Actinobacterial isolates (Costa et al., 2018). However, the underlying enzymatic metabolism of PCA is still poorly understood.

LC-MS-based metabolic analysis plays an increasingly important role in cell biology, pathophysiology and biomarker identification (Melamud et al., 2010), and label-free quantitative proteomics facilitated a high coverage of membrane proteins after FASP and nano LC-MS/MS analysis (Jugder et al., 2016). Therefore, a combination of metabolic, genomic and proteomic analyses has been widely used to ascertain the underlying mechanisms of degradation of various compounds such as pyrene (Kim et al., 2007), long-chain alkanes (Feng et al., 2007) and 2-

methylnaphthalene (Selesi et al., 2010). In this chapter, LC-MS-based metabolic analysis was employed to investigate the intermediates and products of PCA degradation by strain Rhodanobacter sp. PCA2. Moreover, given that proteomic studies have been reported comparing conditions of microbial growth (Mansfeldt et al., 2014, Jugder et al., 2016), we also undertook a label-free quantitative proteomic analysis to compare the response of enzymes in Rhodanobacter sp. PCA2 to growth on PCA and pyruvate in this study. A novel pathway of enzymatic degradation of PCA is proposed, which involves decarboxylation and cleavages of the ring structures.

4.2 Materials and methods

4.2.1 Cultivation of Rhodanobacter sp. PCA2

Rhodanobacter sp. PCA2 was cultivated aerobically in defined minimal mineral medium as described in Section 2.2.1 with two different carbon and energy sources including 2.23 mM PCA and 20 mM pyruvate, respectively. 0.5 g/L (9.35 mM) NH4Cl was supplied as nitrogen source in medium with pyruvate.

Cultures for investigation of inducibility of PCA degradation enzymes were transferred from parent cultures fed with PCA and pyruvate respectively and were grown in 20 mL of fresh mineral medium with PCA as the sole carbon, nitrogen and energy source in 50 mL flasks (Figure 4.1A). These cultures were denoted as “PCA- PCA” and “Pyr-PCA”, respectively, and were sampled for PCA degradation and cell yields analyses periodically.

Cultures for proteomic experiments were grown in 20 mL of fresh mineral medium with PCA or pyruvate as unique carbon source in 50 mL flasks for two rounds of incubation (Figure 4.1B). The second round of cultures were sampled for proteomic assays and were denoted as “PCA culture” and “Pyr culture”, respectively.

All cultures were incubated in the dark at 30°C with shaking at 180 rpm. All experiments were set up in triplicate and abiotic controls were included as well.

Figure 4.1 Schematics of cultivation of Rhodanobacter sp. PCA2 in mineral medium fed with PCA and pyruvate respectively as carbon source for investigation of inducibility (A) and proteomic assay (B).

4.2.2 Determination of PCA

Periodically, we sampled from PCA-PCA and Pyr-PCA cultures to quantitatively analysed consumption of PCA using a HPLC machine as described in Section 2.2.5.

4.2.3 Determination of growth by microscopy

Growth of Rhodanobacter sp. PCA2 was determined by fluorescence microscopy as described in Section 2.2.5.

4.2.4 LC-MS analysis

PCA cultures were sampled after incubation for 0, 12, 24, 36 and 48 h, which were denoted as “PCA-0h”, “PCA-12h”, “PCA-24h”, “PCA-36h” and “PCA-48h”, respectively. Samples were centrifuged at 16,000 x g for 10 min and the supernatant were transferred into 2 mL auto-sampler vials for LC-MS analysis. A 10 µL aliquot of each sample was injected into an Accela liquid chromatography (LC) system (Thermo Fisher Scientific, USA) equipped with a ACQUITY UPLC CSH C18 column with precolumn (1.7 – 3.5 µm, 130 Å, Waters, Australia).

A linear gradient of 0 – 100% Solvent B over 30 minutes was used to elute analytes from the column at a flow rate of 400 µL/min (where Solvent A = 0.1% formic acid in water and Solvent B = acetonitrile) for 30 min. Column eluate was directed into the heated electrospray (HESI) source region of the mass spectrometer (Orbitrap XL, Thermo Fisher Scientific, USA).

Mass spectra were obtained using a full-scan operation in the positive ion mode. The electrospray and capillary voltage were set at 4.5 kV and 18 V, respectively. The capillary and HESI vaporizer temperature at 270°C and 250°C, sheath and auxiliary gas flow rates of 25 and 5 ThermoFisher arbitrary units were used. Mass spectra were acquired from m/z 86 to 1,000 at a rate of 1s/scan throughout the LC run.

4.2.5 Sample preparation for mass spectrometry

Triplicate 50 mL PCA culture and Pyr culture of Rhodanobacter sp. PCA2 were collected and mechanically homogenised using “Lysing Matrix A” 2 mL tubes (MP Biomedicals) in Alkaline-SDS buffer (pH 8.5) containing 4% SDS, 50 mM Tris-HCl,

0.15 M NaCl, 0.1 mM EDTA, 1 mM MgCl2 and 50 mM dithiothreitol (DTT). Lysates were centrifuged at 14,000 x g for 20 min at room temperature, then supernatants were collected carefully. Protein content was determined with the Pierce BCA Protein Assay 72

Kit (Thermo Fisher Scientific, Australia) by following the manufacturer’s protocol (Smith et al., 1985). Equal amounts of total protein (~ 50 µg) from each biological triplicate of PCA culture and Pyr culture were digested by following the modified filter aided sample preparation (FASP) method (Wisniewski et al., 2009). Briefly, wash the crude protein extract with urea in a filter unit (Amicon Ultra 30KDa filter, Merck Millipore, Australia). The precipitate collected after centrifugation was mixed with DTT and incubated at 37°C for 30 min before centrifuged at 14,000 x g for 10 min. Then 0.05 M iodoacetamide (IAA) solution was added before incubation statically at room temperature for 20 min in the dark. After that, wash the filter unit with urea and

NH4HCO3 solution, respectively. Incubate the filter unit in a wet chamber at 37°C overnight after addition of trypsin. Finally, acidify with CF3COOH and desalt the filtrate.

4.2.6 Nano LC-MS/MS and label-free proteomics analysis

Digested peptides prepared above were analysed by nanoLC using Utimate nano RSLC UPLC equipped with an autosampler system (Dionex, Netherlands). 2.5 µL of sample was concentrated and desalted in a micro C18 precolumn (300 µm x 5 mm, Dionex, Netherlands) with mobile phase water:acetonitrile (98:2, 0.1% trifluoroacetic acid) flowing at rate of 15 µL/min. After washing for 4 min, the precolumn was switched via Valco 10 port UPLC valve (Valco, USA) into line connected with a fritless nano-column (75 µm x 15 cm) containing C18-AQ media (1.9 µm, 120 Å Dr Maisch, Germany). The digested peptides were eluted for 30 min by a linear gradient from water:acetonitrile (98:2, 0.1% formic acid) to water:acetonitrile (64:36, 0.1% formic acid) at a flow rate of 200 nL/min. High voltage setting at 2000 V was used to low volume Titanium union (Valco, USA) with the oven of column (Sonation, Germany) heated to 45°C and the tip positioned 5 mm from capillary heated to 300 °C in a QExactive Plus mass spectrometer (Thermo Electron, Germany). A survey scan with m/z ranging from 350 to 1750 was acquired with lockmass enabled at m/z 445.12003. 10 of the most abundant ions with charge states > +2 and < +7 were progressively isolated (width m/z 2.5) and fragmented by HCD (NCE = 30) with a AGC target of 100,000 ions. Peaks generated by Mascot Distiller (Matrix Science, England) were then subjected to a database search program Mascot 2.6.0 (Matrix Science, England). Based on genomic analysis described in Chapter 3, we speculated that strain PCA2 obtained the genes encoding phenylpropionate dioxygenase, 3-Octaprenyl-4-hydroxybenzoate 73

decarboxylase (UbiD) and flavin prenyltransferase (UbiX) functional in PCA degradation via gene transfer from other PCA-degrading bacteria such as Mycobacterium spp.. Hence, the partial genome containing the functional gene cluster of M. fortuitum CT6 and the genome of strain PCA2 were combined together as a database for protein searching with Mascot. Parameters were set as following: tolerances of parent and fragment were 5 ppm and 0.05 Da, respectively; variable modification was specified as Met(O) carboxyamidomethyl-Cys; one missed cleavage was possible. XIC calculation and quantitation were performed using Mascot Distiller, whereby results generated were analysed based on spectrum counting for further relative label-free quantitation (LFQ) with Scaffold Q+ v4.8.4 (Proteome Software Inc, USA). In terms of Scaffold Q+, protein probabilities were assigned using Protein Prophet algorithm (Nesvizhskii et al., 2003), and protein identifications were accepted if their probability were higher than 95.0% and peptide(s) were identified.

4.2.7 Statistical analysis

Analyses of variance were performed with T-tests using SPSS version 19.0 (IBM Co., Armonk, USA), respectively. P values below 0.05 were considered statistically significant.

4.3 Results

4.3.1 Determination of intermediates and products of PCA degradation

Samples “PCA-0h”, “PCA-12h”, “PCA-24h”, “PCA-36h” and “PCA-48h” were subjected to LC-MS-based metabolic analysis. Compared with sample PCA-0h where only PCA was detected (Figure 4.2A), a new compound from sample PCA-12h was eluted at 13.90 min next to PCA showing a molecular ion in positive ion mode [M+H]+ at m/z 181, which was identified as phenazine (Figure 4.2B). Subsequently, the peak intensity of phenazine decreased, while another new intermediate was detected together with phenazine in sample PCA-24h. This new intermediate was then identified as (4Z)- 2-hydroxy-5-{[(1Z)-6-(hydroxyamino)cyclohexa-2,4-dien-1-ylidene]carbamoyl}penta- 2,4-dienoic acid (HCCPD) based on a m/z value at 265 (Figure 4.2C). Analysis of sample PCA-36h found another compound matching m/z at 110 being eluted at 9.42 min (Figure 4.2D). The new chemical was identified as phenylhydroxylamine. No

compound was eluted from sample PCA-48h (Figure 4.2E), suggesting the breakdown intermediates were further consumed. This data informed development of a novel pathway of PCA degradation by strain PCA2 (Figure 4.3).

A． PCA-0h

B． PCA-12h

C． PCA-24h

D． PCA-36h

E． PCA-48h

Figure 4.2 LC-MS chromatograms of metabolites in samples PCA- 0h (A), 12h (B), 24 (C), 36h (D) and 48h (E). Phenazine, (4Z)-2-hydroxy-5-{[(1Z)-6-(hydroxyamino) cyclohexa-2,4-dien-1-ylidene]carbamoyl}penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine were detected as the intermediates of PCA degradation by strain PCA2.

Figure 4.3 Proposed pathway of PCA degradation via decarboxylation and ring cleavage by Rhodanobacter sp. PCA2.

4.3.2 PCA degradation and cell yields in PCA-PCA and Pyr-PCA cultures

To figure out the inducibility of enzymes involved in PCA degradation, profiles of PCA degradation and cell growth were compared between daughter cultures, i.e. PCA- PCA and Pyr-PCA cultures, which were transferred from parent cultures fed with PCA and pyruvate as carbon and energy source, respectively. PCA degradation was immediately observed at a rate of 46.7 µM h-1 in PCA-PCA cultures, while a lag phase for 6 h occurred before transformation of PCA at a rate of 51.6 µM h-1 in Pyr-PCA cultures (Figure 4.4). In terms of cell yields, growth of strain Rhodanobacter sp. PCA2 determined by fluorescence microscopy experienced a lag phase for 6 h and then increased from 3.45 × 106 to 1.07 × 108 cells/mL, followed by a stationary phase in 24 h (Figure 4.4). In comparison, a longer lag phase for 12 h was observed before cell concentrations increased from 6.16 × 106 to 2.36 × 108 cells/mL in Pyr-PCA cultures (Figure 4.4).

Figure 4.4 Profiles of PCA degradation (solid line) and cell growth (dash line) in daughter cultures “PCA-PCA” () and “Pyr-PCA” () which were transferred from parent cultures fed with PCA and pyruvate as carbon and energy source, respectively. Compared to PCA-PCA cultures (6h), Pyr-PCA cultures experienced a longer lag phase (12h), indicating the inducibility of PCA degradation by strain PCA2. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures. 78

4.3.3 Proteomics profiles in PCA and Pyr cultures

Due to the apparent inducibility of PCA degradation, proteomic analysis of strain PCA2 via differential expression of alternate metabolic pathways, i.e. PCA and Pyr cultures, was performed to explore enzymes involved in PCA degradation. Based on genomic analysis described in Chapter 3, we speculated that strain PCA2 obtained the functional genes, e.g. ubiD, ubiX, the homolog of MFORT_16269, in PCA degradation via gene transfer from other PCA-degrading bacteria such as Mycobacterium spp.. Hence, the genomic scaffold (IMG locus tag Ga0100919_111461 - 111871) containing the cluster of these functional genes in M. fortuitum CT6 and the genome of strain PCA2 generated in this study were combined together as a database for protein searching.

A total of 680 detectable proteins were identified and quantified, of which 202 proteins were found in both PCA and Pyr cultures (Table S1) and the remaining 228 and 250 proteins were detected exclusively in the PCA and Pyr cultures, respectively (Table S2, Table S3). Of the 202 shared proteins (120 with higher abundance vs 74 with lower abundance in PCA/Pyr cultures), 45 had a significantly higher abundance in PCA cultures whilst 24 were significantly lower (P < 0.05) (Table S1). Notably, among the 680 proteins detected, 24 proteins matched with the predicted proteins of strain CT6, of which 4 proteins were found in both PCA and Pyr cultures (Table S1) and the remaining 19 and 1 proteins were detected exclusively in the PCA and Pyr cultures, respectively (Table S2,Table S3). The proteins characterized as responsible for PCA degradation including phenylpropionate dioxygenase, UbiD and UbiX were predicted from the partial genome of strain CT6. Phenylpropionate dioxygenase large subunit encoded by a gene with IMG locus tag Ga0100919_111673 was observed in both PCA and Pyr cultures, with 12-fold upregulated abundance in the former cultures, while the phenylpropionate dioxygenase large subunit encoded by the homolog of MFORT_16269 was only found in PCA cultures (Table 4.1). Likewise, phenylpropionate dioxygenase small subunit, UbiD and UbiX were detected exclusively in PCA cultures (Table 4.1). Additionally, biphenyl-2,3-diol 1,2-dioxygenase, amidohydrolase and nitroreductase encoded by the partial genome of strain PCA2 were uniquely found in cultures fed with PCA (Table 4.1). Amidohydrolase and nitroreductase encoded by the genome of strain PCA2 were exclusively observed in PCA and Pyr cultures, respectively (Table 4.1). 79

Table 4.1 Detectable proteins potentially involved in PCA degradation based on metabolic and proteomic analyses

Average number of Average percent Fold unique peptides coverage Molecular change T-Test Protein source Annotation Locus tag weight (PCA / (P value) PCA Pyr PCA Pyr Pyr) cultures cultures cultures cultures

Genome of Amidohydrolase Ga0192503_12145 41 kDa — — 2 — 6.2% — Rhodanobacter sp. PCA2 Nitroreductase Ga0192503_109212 21 kDa — — — 1 — 3.7% UbiD Ga0100919_111652 56 kDa — — 14 — 16% — UbiX Ga0100919_111654 23 kDa — — 7 — 35.7% — Phenylpropionate dioxygenase large Ga0100919_111673 49 kDa 12 < 0.0001 9 1 20.7% 1.8% subunit Partial genome Phenylpropionate (IMG locus tag dioxygenase large Ga0100919_111660 51 kDa — — 7 — 16% — Ga0100919_ subunit 111461-111871) of Phenylpropionate Mycobacterium dioxygenase small Ga0100919_111659 20 kDa — — 6 — 34.3% — fortuitum CT6 subunit Biphenyl-2,3-diol Ga0100919_111676 33 kDa — — 4 — 18.6% — 1,2-dioxygenase Amidohydrolase Ga0100919_111683 52 kDa — — 2 — 2.8% — Nitroreductase Ga0100919_111689 24 kDa — — 1 — 3.2% 1.02% a, — means not applicable.

4.4 Discussion

Degradation of phenazines including PCA was found in five Actinobacteria and two Alphaproteobacteria previously. Only phenazine and 4-hydroxy-1-(2-carboxyphenyl) azacyclobut-2-ene-2-carbonitrile (HPAEC) and 4-hydroxy-1-(2-carboxyphenyl)-2- azetidinecarbonitrile (HPAC) were verified as intermediates of PCA degradation by all PCA-degrading Actinobacteria and strain Sphingomonas wittichii DP58, respectively. Therefore, metabolic pathways of PCA degradation remain to be elucidated in more detail.

Rhodanobacter sp. PCA2 is the first Gammaproteobacterial isolate shown to be capable of degrading PCA. LC-MS-based metabolic, genomic (details described in Chapter 3) and proteomic approaches were used in this investigation to identify intermediates and products, functional genes and proteins involved, and thereby to construct a predicted pathway of PCA degradation for strain PCA2. Specifically, besides phenazine, we identified (4Z)-2-hydroxy-5-{[(1Z)-6-(hydroxyamino)cyclohexa -2,4-dien-1-ylidene]carbamoyl} penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine as intermediates and products of PCA degradation, which has never been reported before.

Furthermore, six proteins were proposed to be involved in PCA degradation by strain PCA2 using proteomic analysis, which matched the genomic analysis described in Chapter 3 well. These six proteins were 3-octaprenyl-4-hydroxybenzoate decarboxylase (UbiD) and the synthetase of its prenylated flavin mononucleotide (prFMN) cofactor named flavin prenyltransferase (UbiX), phenylpropionate dioxygenase, biphenyl-2,3- diol 1,2-dioxygenase, amidohydrolase and nitroreductase. Accordingly, analyses in the current work revealed that PCA was progressively degraded through decarboxylation and cleavages of aromatic and nitrogen-containing rings by Rhodanobacter sp. PCA2.

Initially, an experiment was performed to confirm the inducibility of PCA degradation which typically prefaces studies on differential expression of proteins. Compared to the daughter cultures transferred from PCA cultures (PCA-PCA), PCA degradation and cell growth experienced lag phases for 6 h and 12 h in the daughter cultures inoculated from pyruvate cultures (Pyr-PCA), respectively, indicating PCA transformation was indeed inducible in strain PCA2, as is the case in the PCA- degrading Sphingomonas sp. strain DP58 (Zhao et al., 2017). The inducibility of PCA

degradation allowed proteomic analysis of strain PCA2 via differential expression of alternate metabolic pathways, i.e. PCA cultures and Pyr cultures. Pyruvate was chosen as the control, because it is directly incorporated into the central carbon metabolism through the TCA cycle and hence is not related to the main (or initial) part of the PCA catabolic pathway.

In this study, using LC-MS-based metabolic, genomic and proteomic analyses, phenazine was identified as an intermediate via decarboxylation of PCA, which was consistent with the study performed with M. fortuitum CT6 very recently (Costa et al., 2018). Both UbiD and the synthetase UbiX of its cofactor prFMN were detected exclusively in PCA cultures, suggesting they are involved in PCA degradation, which was in agreement with a recent study where these two enzymes were verified responsible for the first step of PCA degradation using heterologous expression in E. coli BL21 (Costa et al., 2018). Of particular note, both UbiD and UbiX were expressed from the partial genome of strains CT6, which supported our hypothesis that gene transfer from Mycobacterium spp. facilitated degradation of PCA in strain PCA2.

Although intensity of peaks detected by LC-MS provides only semiquantitative estimates of relative abundance, it can be used to represent increasing or decreasing trends of the intermediates and products approximately (Kim et al., 2007). Based on metabolic analysis with LC-MS, a novel intermediate HCCPD was produced, along with decreasing phenazine abundance, suggesting HCCPD was transformed from phenazine. Based on the chemical structures, the degradation of phenazine to HCCPD underwent cleavages of aromatic and nitrogen-containing rings, which has been proposed previously (Chen et al., 2008, Costa et al., 2015) but this is the first time it has been observed. The opening of the aromatic ring of phenazine observed here is similar to that occurring in degradation of polycyclic aromatic hydrocarbons (PAHs) such as anthracene (Evans et al., 1965, Mallick et al., 2011), which might be attributed to the similar structure between phenazines and PAHs. For example, phenazine shares the same linearly fused tricyclic structure with anthracene, with two nitrogen atoms replacing the two carbon atoms on the middle ring. By tracking the degradation pathway of anthracene by soil Pseudomonas (Evans et al., 1965) and PCA by strain DP58 (Zhao et al., 2017), phenazine-1,2-diol was reasonably proposed as an intermediate in the transformation from phenazine to HCCPD, though it was not detected in this study. Dihydroxyphenazine was also hypothesized as a catabolic 82

intermediate of phenazine degradation by Mycobacterium spp. (Costa et al., 2018). Accordingly, strain PCA2 was proposed to hydroxylate phenazine at the C-1,2 position to phenazine-1,2-diol via ring-hydroxylating dioxygenase, and subsequently cleave the aromatic and nitrogen-containing ring of phenazine-1,2-diol to HCCPD via extradiol ring-cleavage dioxygenases and hydrolase respectively, which is in accordance with our proteomic analysis as well. Ring-hydroxylating phenylpropionate dioxygenase (large subunit) was exclusively detected in PCA cultures indicating this dioxygenase could be responsible for PCA degradation. The corresponding small subunit occurred in both PCA and Pyr cultures.

These findings were consistent with previous studies where phenylpropionate dioxygenase was characterized as responsible for PCA degradation by comparative genomic analyses and a knockout mutant of M. fortuitum ATCC6841 (Costa et al., 2015). Extradiol ring-cleavage dioxygenases were classified as one of two groups of ring-cleavage dioxygenases and facilitated cleavage of an aromatic ring by acting between a hydroxylated carbon atom and an adjacent non-hydroxylated carbon atom (Lipscomb, 2008). Biphenyl-2,3-diol 1,2-dioxygenase predicted from the partial genome of strain CT6 was the only extradiol ring-cleavage dioxygenase detected in this proteomic analysis and was extensively detected in PCA cultures, suggesting it was involved in the transformation of phenazine-1,2-diol to HCCPD. Biphenyl-2,3-diol 1,2- dioxygenase is essential for biphenyl degradation in Mycobacterium and Pseudomonas where it catalyses the transformation of 2,3-dihydroxybiphenyl to 2-hydroxy-6-oxo-6- phenylhexa-2,4-dienoic acid (HOPDA) via the meta-cleavage of the aromatic ring (Catelani et al., 1973, Moody et al., 2002). Phenazine appears to undergo a similar mechanism with the production of HCCPD in this study. Besides the aromatic rings, the nitrogen-containing ring was proposed to cleave via hydrolase during the formation of HCCPD from phenazine-1,2-diol. Moreover, PCA was utilized as the sole carbon and nitrogen source by strain PCA2 and CT6 (Costa et al., 2015), indicating that release of nitrogen from the ring via cleavage was necessary in these isolates. The first isolated PCA-degrading strain DP58 was speculated to cleave the nitrogen-containing ring via addition with a hydroxyl group during PCA degradation using GC-MS and 1H-nuclear magnetic resonance (NMR) analyses. The enzymes involved, however, were not explored (Chen et al., 2008). In the current investigation, a putative amidohydrolase acting on amide was exclusively detected in PCA cultures, suggesting its involvement

in PCA degradation via hydration of the nitrogen atom bound within the ring. However, the order of cleavages of aromatic and nitrogen-containing rings is questionable, due to the failure to detect intermediate(s) between phenazine and HCCPD.

Phenylhydroxylamine was for the first time identified as the product of PCA degradation via cleavage of a C-N bond. It was previously revealed that phenylhydroxylamine and its derivatives were the reducing products via nitroso intermediates from nitrofurazone and nitrofurantoin by nitro-reductase in Escherichia coli (Race et al., 2005). However, the corresponding nitroso compound nitrosobenzene was not observed in this study, which can be explained by rapidly enzyme-catalysed reduction of nitrosoaromatics, around 10,000-folder faster than that of nitro compounds (Race et al., 2005). Nitroreductase predicted by a gene (locus tag Ga0100919_111689) close to other PCA-degrading genes in strain CT6 was only observed in PCA cultures, though the one encoded by strain PCA2 was exclusively found in Pyr cultures. Therefore, the transformations of HCCPD into phenylhydroxylamine remain to be elucidated. Additionally, phenylhydroxylamine was completely consumed and no chemical was eluted after 48 h, suggesting another aromatic ring cleavage might occur at the end of PCA degradation, but further experiments are needed to validate this prediction.

The observations and speculations herein lead us to propose a novel pathway for PCA degradation in strain PCA2 as summarized in Figure 4.5. Additionally, despite the findings consistently produced by LC-MS-based metabolic, genomic and proteomic analyses, the intermediates and products proposed here can be further verified by nuclear magnetic resonance (NMR) spectroscopy.

In conclusion, new details on a pathway of PCA degradation is reported in this chapter by using metabolic, genomic and proteomic analyses of the isolate Rhodanobacter sp. PCA2, which contributes to a deeper understanding of the enzymatic degradation of PCA. For the first time, HCCPD and phenylhydroxylamine were observed as ring-cleavage intermediates and product by using LC-MS-based metabolic analysis. Furthermore, the exclusive detection or upregulated expression suggested roles of five proteins in PCA degradation in this study, including UbiD and UbiX, phenylpropionate dioxygenase, biphenyl-2,3-diol 1,2-dioxygenase, amidohydrolase and nitroreductase.

Figure 4.5 Proposed enzymatic pathway of PCA degradation via decarboxylation and cleavage of aromatic and nitrogen-containing rings by strain PCA2. The compounds in brackets were not detected in this study. Proteins and genes (in italic) involved in PCA degradation are depicted beside the arrows. *, ubiD and hacA2 (the homolog of MFORT_16269) genes are plasmid born in strain PCA2. All enzymes except amidohydrolase encoded by a gene with locus tag Ga0192503_12145 are predicted from partial genome of strain CT6. All locus tags of genes described here are based on JGI database.

Supplementary tables

Table S1 The proteins expressed both in PCA and Pyr cultures

Fold Average number of Average percent Locus tag in Molecular Change T-Test unique peptides coverage Protein source Annotation JGI Weight (PCA / (p-value) PCA Pyr PCA Pyr Pyr) cultures cultures cultures cultures 3-phenylpropionate/trans-cinnamate Ga0100919_ < 49 kDa 12 16 1 21% 2% dioxygenase alpha subunit 111673 0.00010 Partial genome 3-hydroxy-9,10-secoandrosta- (IMG locus tag Ga0100919_ 1,3,5(10)-triene-9,17-dione 42 kDa 44 0.000 21 1 23% 4% Ga0100919_ 111675 monooxygenase 111461-111871) Ga0100919_ of Mycobacterium FMN reductase 21 kDa 4 0.001 11 2 42% 6% 111677 fortuitum CT6 Ga0100919_ hypothetical protein 22 kDa 2.4 0.250 1 1 3% 3% 111694 Ga0192503_ acetyl-coenzyme A synthetase 100 kDa 0.6 0.063 4 4 5% 7% 10112 Ga0192503_ aldehyde dehydrogenase 55 kDa 0.5 0.011 4 6 11% 8% 10172 Ga0192503_ L-glutamine synthetase 54 kDa 1.1 0.630 5 3 9% 5% 10181 Genome of Ga0192503_ Rhodanobacter sp. methionyl-tRNA formyltransferase 33 kDa 1.5 0.720 1 1 5% 3% PCA2 10199 Genome of Ga0192503_ glucose-6-phosphate isomerase 60 kDa 1.5 0.340 3 2 7% 3% Rhodanobacter sp. 101109 PCA2 Ga0192503_ transaldolase 40 kDa 1.5 0.480 4 3 9% 8% 101110 Ga0192503_ phosphoglycerate mutase 27 kDa 2.7 0.220 1 1 6% 5% 101125 N-acetylated-alpha-linked acidic Ga0192503_ 83 kDa 0.7 0.800 1 1 1% 0% dipeptidase 101139

glucose-6-phosphate 1- Ga0192503_ 58 kDa 0.8 0.860 1 1 1% 2% dehydrogenase 10310 Ga0192503_ ATP dependent helicase, Lhr family 162 kDa 1 0.960 1 1 1% 0% 10325 Ga0192503_ putative transposase 89 kDa 0.4 0.440 1 1 1% 1% 10342 cell division checkpoint GTPase Ga0192503_ 23 kDa 0.4 0.360 1 1 3% 5% YihA 1042 Ga0192503_ chaperonin GroES 11 kDa 0.6 0.039 9 10 63% 85% 10613 Ga0192503_ chaperonin GroEL 57 kDa 0.6 0.062 2 3 2% 3% 10614 glutamate-ammonia-ligase Ga0192503_ 104 kDa 0.3 0.063 1 1 1% 1% adenylyltransferase 10618 Ga0192503_ inorganic pyrophosphatase 18 kDa 1 0.770 3 2 7% 7% 10718 Genome of Ga0192503_ Rhodanobacter sp. adenylate kinase 20 kDa 1 0.760 5 4 17% 23% 10726 PCA2 Ga0192503_ hypothetical protein 27 kDa 0.5 0.480 1 2 2% 4% 10872

nucleotide-binding universal stress Ga0192503_ 31 kDa 4.6 0.076 2 1 6% 4% protein, UspA family 10875

Ga0192503_ single-strand binding protein 17 kDa 1.7 0.046 3 1 16% 9% 10943 glyceraldehyde-3-phosphate Ga0192503_ 36 kDa 2.3 0.083 11 4 27% 12% dehydrogenase (NAD+) 109110 Ga0192503_ phosphoglycerate kinase 42 kDa 3 0.030 4 2 13% 6% 109126 mannose-1-phosphate Ga0192503_ 37 kDa 10 0.042 5 1 14% 2% guanylyltransferase 109151 electron transfer flavoprotein alpha Ga0192503_ 32 kDa 1.3 0.520 2 1 8% 7% subunit apoprotein 109170

electron transfer flavoprotein beta Ga0192503_ 28 kDa 1.8 0.130 4 2 15% 7% subunit 109171 RNA polymerase sigma-70 factor, Ga0192503_ 25 kDa 6.3 0.047 3 1 18% 5% ECF subfamily 109191 Ga0192503_ acyl-CoA dehydrogenase 42 kDa 1.3 0.850 2 1 6% 3% 109231 Ga0192503_ ribose-phosphate pyrophosphokinase 36 kDa 0.9 0.630 2 2 5% 2% 109245 translation elongation factor 1A (EF- Ga0192503_ 43 kDa 0.4 0.004 3 5 5% 8% 1A/EF-Tu) 109258 Ga0192503_ LSU ribosomal protein L1P 25 kDa 1 0.920 4 3 14% 13% 109263 Ga0192503_ LSU ribosomal protein L10P 18 kDa 0.9 0.700 4 3 22% 21% 109264 DNA-directed RNA polymerase Ga0192503_ 154 kDa 0.2 0.007 2 3 1% 2% subunit beta 109266 Genome of DNA-directed RNA polymerase Ga0192503_ Rhodanobacter sp. 154 kDa 0.3 0.005 3 7 2% 4% subunit beta 109267 PCA2 Ga0192503_ SSU ribosomal protein S12P 14 kDa 2 0.047 3 1 16% 6% 109268 Ga0192503_ SSU ribosomal protein S7P 17 kDa 1.4 0.050 8 4 34% 27% 109269 translation elongation factor 2 (EF- Ga0192503_ 76 kDa 0.008 0.000 1 26 1% 31% 2/EF-G) 109270 translation elongation factor 1A (EF- Ga0192503_ 43 kDa 0.07 0.000 1 11 2% 19% 1A/EF-Tu) 109271 Ga0192503_ SSU ribosomal protein S10P 11 kDa 1.4 0.180 10 5 55% 46% 109272 Ga0192503_ LSU ribosomal protein L3P 23 kDa 1.2 0.290 10 6 30% 27% 109273 Ga0192503_ LSU ribosomal protein L4P 23 kDa 0.04 0.004 1 7 3% 25% 109274 Ga0192503_ SSU ribosomal protein S19P 11 kDa 1.4 0.220 7 4 37% 27% 109277

Ga0192503_ SSU ribosomal protein S3P 30 kDa 1.2 0.160 10 6 23% 19% 109279 Ga0192503_ LSU ribosomal protein L16P 16 kDa 1.4 0.280 2 1 14% 8% 109280 Ga0192503_ LSU ribosomal protein L29P 9 kDa 2.9 0.005 4 1 24% 8% 109281 Ga0192503_ SSU ribosomal protein S17P 11 kDa 2.5 0.001 6 2 35% 16% 109282 Ga0192503_ LSU ribosomal protein L14P 13 kDa 1.5 0.140 6 3 31% 15% 109283 Ga0192503_ LSU ribosomal protein L5P 21 kDa 3.5 0.001 9 2 29% 8% 109285 Ga0192503_ SSU ribosomal protein S8P 14 kDa 1.1 0.410 7 5 39% 26% 109287 Ga0192503_ LSU ribosomal protein L6P 19 kDa 2 0.086 5 2 21% 9% 109288 Genome of Ga0192503_ Rhodanobacter sp. LSU ribosomal protein L18P 15 kDa 1.5 0.067 3 2 18% 10% 109289 PCA2 Ga0192503_ SSU ribosomal protein S5P 23 kDa 1.1 0.610 6 4 21% 17% 109290 Ga0192503_ SSU ribosomal protein S13P 14 kDa 1.5 0.048 9 4 45% 23% 109294 Ga0192503_ SSU ribosomal protein S11P 15 kDa 3.9 0.002 5 1 23% 8% 109295 Ga0192503_ SSU ribosomal protein S4P 23 kDa 1.9 0.006 10 4 28% 16% 109296 DNA-directed RNA polymerase Ga0192503_ 38 kDa 3.4 0.029 11 2 29% 7% subunit alpha 109297 Ga0192503_ LSU ribosomal protein L17P 20 kDa 1.1 0.760 3 2 8% 8% 109298 3-deoxy-D-arabinoheptulosonate-7- Ga0192503_ 51 kDa 3.2 0.000 4 1 9% 2% phosphate synthase 109300 succinyl-CoA synthetase (ADP- Ga0192503_ 42 kDa 0.1 0.002 1 6 2% 13% forming) beta subunit 11036

beta-barrel assembly machine Ga0192503_ 33 kDa 0.5 0.140 1 1 2% 2% subunit BamD 11041 ATP-dependent Clp protease ATP- Ga0192503_ 96 kDa 0.3 0.097 3 3 4% 2% binding subunit ClpB 11044 Ga0192503_ aspartyl-tRNA synthetase 65 kDa 0.2 0.036 1 2 1% 2% 11048 transcriptional regulator /preQ(0) Ga0192503_ 24 kDa 0.5 0.140 1 1 4% 4% biosynthesis protein QueC 11064 Ga0192503_ hypothetical protein 146 kDa 0.8 0.860 1 1 1% 1% 11084 Ga0192503_ valyl-tRNA synthetase 107 kDa 0.7 0.790 1 2 1% 1% 11113 Ga0192503_ serine hydroxymethyltransferase 52 kDa 8.1 0.008 4 1 7% 2% 11121 Ga0192503_ uroporphyrinogen decarboxylase 39 kDa 5.5 0.023 2 1 3% 2% Genome of 11217 Rhodanobacter sp. chemosensory pili system protein Ga0192503_ PCA2 ChpA (sensor histidine 220 kDa 0.8 0.860 1 1 0% 0% 11245 kinase/response regulator) 1-deoxy-D-xylulose 5-phosphate Ga0192503_ 40 kDa 0.2 0.059 1 1 2% 2% reductoisomerase 112105 Ga0192503_ superoxide dismutase, Fe-Mn family 23 kDa 2.2 0.036 3 1 12% 4% 112110 Ga0192503_ threonyl-tRNA synthetase 76 kDa 5.2 0.210 4 1 8% 2% 112123 bacterial translation initiation factor Ga0192503_ 22 kDa 1.2 0.210 6 3 24% 15% 3 (bIF-3) 112124 Ga0192503_ LSU ribosomal protein L20P 14 kDa 1.3 0.310 3 2 17% 11% 112127 D-3-phosphoglycerate Ga0192503_ 55 kDa 2.6 0.013 6 2 13% 4% dehydrogenase 112134 Protein of unknown function Ga0192503_ 82 kDa 0.3 0.063 1 1 1% 1% (DUF1631) 11341

5-(carboxyamino) imidazole Ga0192503_ 18 kDa 2.2 0.290 1 1 5% 6% ribonucleotide mutase 11343 5-(carboxyamino) imidazole Ga0192503_ 41 kDa 3.9 0.180 2 1 4% 2% ribonucleotide synthase 11344 translation elongation factor Ts (EF- Ga0192503_ 29 kDa 1.6 0.097 8 4 24% 10% Ts) 11345 Ga0192503_ SSU ribosomal protein S2P 31 kDa 1.8 0.003 14 6 32% 20% 11346 Ga0192503_ SSU ribosomal protein S6P 10 kDa 3.6 0.067 3 2 19% 18% 11551 Ga0192503_ SSU ribosomal protein S18P 10 kDa 1.3 0.210 9 5 50% 38% 11552 Ga0192503_ LSU ribosomal protein L9P 16 kDa 8.8 0.005 4 1 14% 6% 11553 Ga0192503_ SSU ribosomal protein S20P 10 kDa 1.3 0.430 4 2 13% 13% 11557 Genome of Ga0192503_ Rhodanobacter sp. glutamate dehydrogenase (NAD) 174 kDa 1.6 0.028 4 2 3% 1% 11618 PCA2 Ga0192503_ aspartate aminotransferase 44 kDa 0.1 0.032 1 2 2% 4% 117 Ga0192503_ 2-methylcitrate synthase 48 kDa 0.1 0.002 1 6 1% 10% 11715 Ga0192503_ 2-methylcitrate dehydratase 53 kDa 0.7 0.840 1 1 1% 1% 11717 Ga0192503_ hypothetical protein 76 kDa 2.4 0.250 1 1 1% 1% 11723 Ga0192503_ phosphate uptake regulator, PhoU 27 kDa 0.7 0.840 1 1 3% 5% 11751 acetyl-CoA carboxylase Ga0192503_ 31 kDa 0.8 0.870 1 1 2% 3% carboxyltransferase subunit alpha 1189

ATP phosphoribosyltransferase Ga0192503_ 30 kDa 1.9 0.570 2 1 5% 3% (homohexameric) 11819

1-(5-phosphoribosyl)-5-[(5- phosphoribosylamino) Ga0192503_ 25 kDa 4.6 0.072 2 1 10% 7% methylideneamino] imidazole-4- 11824 carboxamide isomerase Ga0192503_ DNA topoisomerase IV subunit B 69 kDa 0.6 0.760 1 1 1% 2% 11836 Ga0192503_ < enolase 45 kDa 0.07 2 10 2% 23% 11847 0.00010 Ga0192503_ trigger factor 52 kDa 1.8 0.099 4 2 7% 3% 11867 ATP-dependent Clp protease ATP- Ga0192503_ 47 kDa 0.3 0.120 1 1 2% 4% binding subunit ClpX 11869 ATP-dependent proteinase. Serine Ga0192503_ 92 kDa 0.08 0.041 1 3 1% 3% peptidase. MEROPS family S16 11870 Ga0192503_ phosphoenolpyruvate synthase 98 kDa 0.1 0.004 1 2 1% 1% 11887 Genome of Ga0192503_ Rhodanobacter sp. glutamate or tyrosine decarboxylase 71 kDa 2.4 0.250 1 1 1% 1% 11914 PCA2 Ga0192503_ nitrate reductase alpha subunit 90 kDa 1.4 0.760 1 1 1% 1% 11958 Ga0192503_ lysyl-tRNA synthetase, class II 56 kDa 5.9 0.011 3 1 4% 2% 11982 hemolysin activation/secretion Ga0192503_ 63 kDa 1.3 0.820 1 1 1% 1% protein 11986 Ga0192503_ condensin subunit Smc 129 kDa 5 0.100 2 1 1% 0% 11989 Ga0192503_ triosephosphate isomerase 27 kDa 2.8 0.003 5 1 13% 6% 12016 Ga0192503_ DNA ligase (NAD+) 87 kDa 0.06 0.140 1 4 1% 5% 12028 methenyltetrahydrofolate cyclohydrolase /5,10- Ga0192503_ 31 kDa 6.4 0.015 5 2 18% 6% methylenetetrahydrofolate 12057 dehydrogenase (NADP+)

Ga0192503_ nucleoside diphosphate kinase 15 kDa 2.1 0.170 3 2 23% 16% 12066 Ga0192503_ UDP-glucose pyrophosphorylase 32 kDa 0.4 0.370 1 1 2% 4% 12083 Ga0192503_ SSU ribosomal protein S1P 53 kDa 1.3 0.067 11 6 26% 14% 12089 Ga0192503_ cytidylate kinase 25 kDa 1.2 0.780 1 1 4% 4% 12090 Ga0192503_ transcription elongation factor GreA 18 kDa 0.8 0.750 2 2 10% 10% 12097 carbamoyl-phosphate synthase large Ga0192503_ 119 kDa 7.2 0.079 4 1 3% 1% subunit 12098 Ga0192503_ molecular chaperone DnaK 68 kDa 0.2 0.012 1 2 1% 3% 120105 3-deoxy-manno-octulosonate Ga0192503_ Genome of cytidylyltransferase (CMP-KDO 26 kDa 0.2 0.018 1 2 3% 15% 120128 Rhodanobacter sp. synthetase) PCA2 3-phosphoshikimate 1- Ga0192503_ 47 kDa 0.9 0.940 1 1 2% 2% carboxyvinyltransferase 120129 bacterial translation initiation factor Ga0192503_ 8 kDa 1.2 0.290 6 4 59% 52% 1 (bIF-1) 120134 ATP-dependent Clp protease ATP- Ga0192503_ 94 kDa 3.5 0.002 28 6 25% 8% binding subunit ClpA 120135 Ga0192503_ aconitase 102 kDa 1.5 0.068 15 8 16% 8% 120142 Ga0192503_ fumarase, class II 50 kDa 20 0.001 8 1 8% 1% 120160 Ga0192503_ adenylosuccinate lyase 47 kDa 3.6 0.000 5 1 11% 3% 120161 outer membrane protein, multidrug Ga0192503_ 56 kDa 0.3 0.330 1 2 2% 3% efflux system 12124 Ga0192503_ Histone methylation protein DOT1 30 kDa 0.3 0.330 1 1 2% 2% 12177

cell division protein DedD (protein Ga0192503_ 36 kDa 0.8 0.860 1 1 2% 2% involved in septation) 121113 3-deoxy-D-manno-octulosonic acid Ga0192503_ 26 kDa 0.4 0.110 1 1 3% 3% kinase 121123 3-oxoacyl-[acyl-carrier-protein] Ga0192503_ 27 kDa 3.7 0.370 3 1 8% 4% reductase 121146 Ga0192503_ acyl carrier protein 11 kDa 0.7 0.330 2 2 21% 24% 121147 bacterial translation initiation factor Ga0192503_ 95 kDa 4 0.005 4 1 4% 1% 2 (bIF-2) 1221 Ga0192503_ SSU ribosomal protein S15P 10 kDa 1.2 0.730 3 3 25% 20% 1224 polyribonucleotide Ga0192503_ 80 kDa 1.6 0.060 10 4 10% 6% nucleotidyltransferase 1225 Ga0192503_ HDIG domain-containing protein 31 kDa 0.3 0.280 1 2 3% 5% 1275 Genome of 4-hydroxy-3-methylbut-2-en-1-yl Ga0192503_ Rhodanobacter sp. 45 kDa 0.09 0.012 1 2 2% 6% diphosphate synthase 1278 PCA2 exodeoxyribonuclease VII large Ga0192503_ 46 kDa 0.7 0.016 1 1 2% 2% subunit 12712 cold-shock DNA-binding protein Ga0192503_ 7 kDa 1.7 0.350 2 2 29% 21% family 12731 Ga0192503_ ferredoxin 12 kDa 1.9 0.003 3 1 30% 11% 13017 Ga0192503_ polyphosphate kinase 81 kDa 7.1 0.005 3 1 5% 2% 13037 peptidyl-prolyl cis-trans isomerase A Ga0192503_ 19 kDa 2.4 0.001 3 1 16% 7% (cyclophilin A) 13040 Ga0192503_ cysteine desulfurase 45 kDa 0.7 0.830 1 1 5% 2% 13044 Ga0192503_ preprotein translocase subunit SecD 67 kDa 0.3 0.350 1 1 1% 1% 13055 Ga0192503_ RecA protein 37 kDa 0.2 0.059 1 1 4% 6% 13129

Ga0192503_ 2,4-dienoyl-CoA reductase 44 kDa 0.9 0.940 1 1 2% 2% 13151 Ga0192503_ acetyl-CoA C-acetyltransferase 40 kDa 4.8 0.023 2 1 5% 2% 1332 Ga0192503_ glutamate 5-kinase 39 kDa 1.5 0.700 1 1 2% 2% 13325 Ga0192503_ argininosuccinate synthase 44 kDa 0.7 0.640 3 2 8% 4% 13330 Ga0192503_ LSU ribosomal protein L19P 13 kDa 1.5 0.170 3 2 17% 15% 13348 Ga0192503_ SSU ribosomal protein S16P 18 kDa 0.6 0.310 3 2 16% 11% 13351 Ga0192503_ LSU ribosomal protein L27P 9 kDa 1.2 0.480 5 3 43% 36% 13358 2-dehydro-3-deoxy-L-fuconate Ga0192503_ 29 kDa 4.5 0.005 6 2 15% 6% dehydrogenase 13366 Genome of enoyl-CoA hydratase/carnithine Ga0192503_ Rhodanobacter sp. 27 kDa 0.7 0.430 2 2 6% 8% racemase 13374 PCA2 thiosulfate/3-mercaptopyruvate Ga0192503_ 33 kDa 4.6 0.038 4 1 13% 3% sulfurtransferase 13376 uncharacterized conserved protein, Ga0192503_ 67 kDa 0.5 0.140 1 1 1% 1% DUF885 familyt 133134 Ga0192503_ DNA topoisomerase IV subunit A 82 kDa 2.6 0.450 2 1 2% 1% 133140 Ga0192503_ hypothetical protein 20 kDa 0.7 0.016 1 1 5% 5% 13434 Ga0192503_ Xaa-Pro dipeptidase 39 kDa 4.5 0.029 2 1 4% 2% 13657 Ga0192503_ ATP-dependent DNA helicase DinG 72 kDa 1.1 0.890 1 1 1% 1% 13673 succinate semialdehyde Ga0192503_ 55 kDa 1.7 0.640 1 1 2% 2% dehydrogenase 1374 UDP-N-acetylmuramoylalanine--D- Ga0192503_ 49 kDa 1.1 0.850 1 1 2% 2% glutamate ligase 13715

Ga0192503_ protein translocase subunit secA 85 kDa 2.4 0.017 4 1 6% 1% 14024 Ga0192503_ methionine adenosyltransferase 43 kDa 1.5 0.120 3 1 7% 3% 14055 Ga0192503_ phosphoenolpyruvate carboxylase 103 kDa 1 0.940 1 1 1% 1% 14059 Ga0192503_ adenosylhomocysteinase 53 kDa 1.1 0.770 4 3 9% 6% 14061 Cu2+-exporting ATPase Ga0192503_ 85 kDa 0.8 0.860 1 1 1% 1% [Rhodanobacter sp. PCA2] 14067 cation diffusion facilitator family Ga0192503_ 22 kDa 1.1 0.850 1 1 3% 3% transporter 140122 Ga0192503_ anthranilate synthase, component I 56 kDa 0.5 0.340 2 1 6% 1% 140142 Ga0192503_ SSU ribosomal protein S9P 17 kDa 1.8 0.057 8 3 45% 22% 140160 Genome of CRP/FNR family transcriptional Rhodanobacter sp. Ga0192503_ regulator, cyclic AMP receptor 25 kDa 2 0.053 11 4 40% 19% PCA2 140164 protein pyruvate dehydrogenase E2 Ga0192503_ component (dihydrolipoamide 57 kDa 1 0.940 4 3 7% 5% 140193 acetyltransferase) Ga0192503_ citrate synthase 36 kDa 0.05 0.001 1 5 3% 12% 14233 Ga0192503_ guanylate kinase 22 kDa 1.2 0.780 1 1 6% 3% 14246 Ga0192503_ nitrous oxidase accessory protein 46 kDa 0.4 0.410 1 1 2% 2% 14257 Ga0192503_ methylmalonyl-CoA mutase 65 kDa 4 0.095 2 1 2% 1% 14420 Ga0192503_ alpha/beta hydrolase fold 55 kDa 1.6 0.680 1 1 2% 1% 14720 Ga0192503_ hypothetical protein 16 kDa 1.6 0.680 1 1 7% 7% 14744

Ga0192503_ cytochrome c oxidase subunit 1 64 kDa 1 0.850 1 1 2% 1% 14752 Ga0192503_ coniferyl-aldehyde dehydrogenase 51 kDa 0.2 0.035 1 1 2% 2% 14756 Ga0192503_ malate synthase 65 kDa 5.5 0.026 3 1 3% 1% 147139 Ga0192503_ isocitrate lyase 85 kDa 13 0.019 5 1 6% 1% 147140 Ga0192503_ general secretion pathway protein L 42 kDa 1.4 0.520 1 1 3% 2% 1483 Ga0192503_ homoserine kinase 34 kDa 1.4 0.760 1 1 3% 2% 14837 branched-chain amino acid Ga0192503_ 40 kDa 3.3 0.190 1 1 3% 4% aminotransferase 14936 Ga0192503_ ketol-acid reductoisomerase 36 kDa 2.5 0.007 7 2 13% 4% 14941 Genome of glutamate synthase (NADPH) small Ga0192503_ Rhodanobacter sp. 50 kDa 0.8 0.860 1 1 2% 2% subunit 15131 PCA2 glutamate synthase (NADPH) large Ga0192503_ 325 kDa 1.3 0.820 1 1 0% 0% subunit 15132 Ga0192503_ DNA repair protein RadC 24 kDa 0.7 0.670 1 1 3% 3% 15313 ATP synthase F1 subcomplex beta Ga0192503_ 52 kDa 1.2 0.220 14 9 27% 22% subunit 15337 ATP synthase F1 subcomplex Ga0192503_ 33 kDa 3.2 0.038 4 2 12% 6% gamma subunit 15338 F-type H+-transporting ATPase Ga0192503_ 56 kDa 0.5 0.210 2 3 4% 4% subunit alpha 15339 Ga0192503_ bacterioferritin 18 kDa 1.5 0.440 1 1 9% 5% 15350 RNA polymerase, sigma 70 subunit, Ga0192503_ 77 kDa 0.07 0.006 1 4 2% 6% RpoD 15411 6-phosphogluconate dehydrogenase Ga0192503_ 30 kDa 4.1 0.039 2 1 5% 5% (decarboxylating) 15416

chromosome segregation DNA- Ga0192503_ 37 kDa 1.4 0.760 1 1 2% 1% binding protein 15423 Ga0192503_ type I restriction enzyme, R subunit 118 kDa 0.6 0.430 1 2 1% 1% 15454 Ga0192503_ DNA polymerase III, beta subunit 41 kDa 1.8 0.042 6 2 12% 5% 15465 tetratricopeptide repeat-containing Ga0192503_ 46 kDa 1.2 0.780 1 1 2% 2% protein 15468

pyruvate dehydrogenase E1 Ga0192503_ Genome of 103 kDa 1.8 0.018 6 2 6% 3% Rhodanobacter sp. component 15519 PCA2 malate dehydrogenase (oxaloacetate- Ga0192503_ 83 kDa 0.2 0.035 1 1 1% 1% decarboxylating)(NADP+) 15520 ATPase components of ABC Ga0192503_ transporters with duplicated ATPase 60 kDa 1.5 0.720 1 1 1% 1% 15545 domains Ga0192503_ Uracil-DNA glycosylase 25 kDa 0.3 0.340 1 1 4% 3% 15547 Ga0192503_ LSU ribosomal protein L22P 11 kDa 1 0.890 4 3 31% 25% 109278 sigma-B regulation protein RsbU Ga0192503_ 85 kDa 0.9 0.940 1 1 1% 1% (phosphoserine phosphatase) 147113

Table S2 The proteins exclusively expressed in PCA cultures

Average Average Molecular Protein source Annotation Locus tag in JGI unique percent Weight peptides coverage UbiD family decarboxylase Ga0100919_111652 56 kDa 14 16% hypothetical protein Ga0100919_111653 17 kDa 4 20% UbiX (4-hydroxy-3-polyprenylbenzoate decarboxylase) Ga0100919_111654 23 kDa 7 36% 2,3-dihydroxy-2,3-dihydrophenylpropionate dehydrogenase Ga0100919_111657 29 kDa 12 27% 3-phenylpropionate/cinnamic acid dioxygenase, small subunit Ga0100919_111659 20 kDa 10 34% phenylpropionate dioxygenase, large terminal subunit Ga0100919_111660 51 kDa 9 16% non-specific serine/threonine protein kinase Ga0100919_111661 85 kDa 3 5% 2,4-dienoyl-CoA reductase Ga0100919_111663 42 kDa 6 15% 2-hydroxymuconate-semialdehyde hydrolase Ga0100919_111669 31 kDa 4 16% Partial genome (IMG 3,4-dihydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione locus tag Ga0100919_111670 32 kDa 5 18% 4,5-dioxygenase Ga0100919_ 111461- 3-phenylpropionate/trans-cinnamate dioxygenase ferredoxin 111871) of Ga0100919_111671 45 kDa 6 13% Mycobacterium reductase subunit fortuitum CT6 3-phenylpropionate/trans-cinnamate dioxygenase ferredoxin Ga0100919_111672 12 kDa 1 12% subunit/biphenyl 2,3-dioxygenase ferredoxin subunit biphenyl 2,3-dioxygenase beta subunit Ga0100919_111674 22 kDa 14 35% biphenyl-2,3-diol 1,2-dioxygenase Ga0100919_111676 33 kDa 7 19% 2-oxo-3-hexenedioate decarboxylase Ga0100919_111682 27 kDa 2 9% amidohydrolase Ga0100919_111683 52 kDa 2 3% FAD/FMN-containing dehydrogenase Ga0100919_111684 49 kDa 2 5% hypothetical protein Ga0100919_111688 20 kDa 1 5%

nitroreductase Ga0100919_111689 24 kDa 1 3% 99

hypothetical protein Ga0192503_10126 23 kDa 1 3% TadE-like protein Ga0192503_10134 33 kDa 1 3% tetratricopeptide repeat-containing protein Ga0192503_10143 18 kDa 1 5% phospholipid-binding protein, PBP family Ga0192503_10160 22 kDa 1 3% multidrug efflux pump Ga0192503_10168 115 kDa 1 1% transcriptional regulator, LysR family Ga0192503_10170 32 kDa 2 5% glycosyl transferases group 1 Ga0192503_10196 36 kDa 1 2% peptide deformylase Ga0192503_101100 19 kDa 2 8% 3-ketoacyl-CoA thiolase Ga0192503_101121 45 kDa 1 2% putative oligopeptide transporter, OPT family Ga0192503_101146 69 kDa 1 1% MFS transporter, DHA2 family, multidrug resistance protein Ga0192503_1025 56 kDa 1 1% transcriptional regulator, MarR family Ga0192503_1028 14 kDa 1 10% Genome of outer membrane autotransporter barrel domain-containing Rhodanobacter sp. Ga0192503_10335 203 kDa 1 0% PCA2 protein dethiobiotin synthase Ga0192503_1045 23 kDa 1 6% DNA polymerase I Ga0192503_1057 98 kDa 5 8% 3-dehydroquinate dehydratase Ga0192503_1068 16 kDa 1 6% transcriptional regulator, TetR family Ga0192503_1079 25 kDa 2 6%

lactoylglutathione lyase Ga0192503_10734 18 kDa 1 4%

glutamate-1-semialdehyde 2,1-aminomutase Ga0192503_10738 44 kDa 2 4% flagellar protein FliS Ga0192503_1087 15 kDa 1 6% transcriptional regulator Ga0192503_10810 52 kDa 1 2% adenosine kinase Ga0192503_10851 34 kDa 3 7% penicillin-binding protein 6. Serine peptidase. MEROPS Ga0192503_10859 45 kDa 1 2% family S11 lipoyl(octanoyl) transferase Ga0192503_10861 27 kDa 1 4% 100

carboxyl-terminal processing protease Ga0192503_10863 84 kDa 1 1% acetate kinase Ga0192503_10876 44 kDa 1 2% tryptophanyl-tRNA synthetase Ga0192503_10916 37 kDa 1 2% Fe-S cluster biosynthesis and repair protein YggX Ga0192503_10920 10 kDa 1 10% fused signal recognition particle receptor Ga0192503_10923 44 kDa 1 2% two-component system, OmpR family, KDP operon response Ga0192503_10942 26 kDa 1 3% regulator KdpE tRNA (cytidine/uridine-2'-O-)-methyltransferase Ga0192503_147130 21 kDa 1 5% DNA-binding transcriptional regulator, XRE-family HTH Ga0192503_10968 15 kDa 1 6% domain NAD(P)-dependent dehydrogenase, short-chain alcohol Ga0192503_112112 24 kDa 1 4% dehydrogenase family cystathionine gamma-lyase Ga0192503_109156 41 kDa 2 4% Genome of UDP-N-Acetylglucosamine 2-epimerase Ga0192503_109161 41 kDa 1 3% Rhodanobacter sp. methyltransferase domain-containing protein Ga0192503_109166 26 kDa 2 8% PCA2 phosphomannomutase Ga0192503_109169 49 kDa 1 2% diguanylate cyclase (GGDEF) domain-containing protein Ga0192503_109187 109 kDa 2 2% RNA polymerase sigma-70 factor, ECF subfamily Ga0192503_109252 22 kDa 1 4% SSU ribosomal protein S14P Ga0192503_109286 7 kDa 3 36% SSU ribosomal protein S5P Ga0192503_109290 22 kDa 2 9% transcriptional regulator, LysR family Ga0192503_109320 33 kDa 1 3% alpha,alpha-trehalose phosphorylase Ga0192503_11033 111 kDa 1 1% 7-carboxy-7-deazaguanine synthase Ga0192503_11063 25 kDa 1 3% serine protease Ga0192503_11078 65 kDa 1 1% prolyl-tRNA synthetase Ga0192503_1118 62 kDa 1 1% transcriptional repressor NrdR Ga0192503_11122 17 kDa 2 12% 3-dehydroquinate synthase Ga0192503_11220 38 kDa 1 5% 101

transcriptional regulator, TetR family Ga0192503_11228 23 kDa 1 4% adenosylmethionine-8-amino-7-oxononanoate Ga0192503_11251 50 kDa 6 16% aminotransferase undecaprenyl pyrophosphate synthetase Ga0192503_112107 30 kDa 2 5% uridylate kinase Ga0192503_112109 26 kDa 3 13% alpha/beta hydrolase family protein Ga0192503_109143 30 kDa 2 4% phenylalanyl-tRNA synthetase, alpha subunit Ga0192503_112128 37 kDa 1 3% uracil phosphoribosyltransferase Ga0192503_112137 22 kDa 1 5% L-lysine 6-transaminase Ga0192503_112139 49 kDa 2 4% type IV pilus assembly protein PilA Ga0192503_11323 15 kDa 1 10% ubiquinol-cytochrome c reductase cytochrome b subunit Ga0192503_11335 62 kDa 2 3% nicotinate-nucleotide pyrophosphorylase [carboxylating] Ga0192503_11340 30 kDa 1 3% Genome of membrane protein TerC, possibly involved in tellurium Ga0192503_11525 56 kDa 1 2% Rhodanobacter sp. resistance PCA2 D-amino acid dehydrogenase small subunit Ga0192503_11541 44 kDa 1 2% alanine racemase Ga0192503_11555 38 kDa 1 3% leucine dehydrogenase Ga0192503_1161 49 kDa 1 5% DNA-binding transcriptional regulator, LysR family Ga0192503_11652 32 kDa 1 2% membrane fusion protein, cobalt-zinc-cadmium efflux system Ga0192503_11670 39 kDa 1 2% poly(3-hydroxybutyrate) depolymerase Ga0192503_11674 36 kDa 1 2% protease FtsH subunit HflC Ga0192503_11730 33 kDa 1 2% peptide-methionine (S)-S-oxide reductase Ga0192503_11733 19 kDa 1 6% chorismate synthase Ga0192503_11792 42 kDa 1 4% tryptophan synthase, alpha chain Ga0192503_1186 27 kDa 1 5% histidinol dehydrogenase Ga0192503_11820 47 kDa 2 3% histidinol phosphate aminotransferase apoenzyme Ga0192503_11821 40 kDa 1 4%

102

2-C-methyl-D-erythritol 2,4-cyclodiphosphate synthase Ga0192503_11850 34 kDa 1 3% protein-L-isoaspartate(D-aspartate) O-methyltransferase Ga0192503_11854 25 kDa 1 4% RNA-binding protein Ga0192503_11858 11 kDa 1 11% membrane protease FtsH catalytic subunit Ga0192503_11860 82 kDa 2 3% hydroxyacylglutathione hydrolase Ga0192503_11876 27 kDa 1 3% GTP-binding protein Era Ga0192503_118109 33 kDa 2 4% two-component system, OmpR family, response regulator Ga0192503_11919 26 kDa 1 4% beta-L-arabinofuranosidase, GH127 Ga0192503_11928 74 kDa 1 1% imidazolonepropionase Ga0192503_11935 48 kDa 1 2% cyclic pyranopterin phosphate synthase Ga0192503_11939 37 kDa 1 2% uroporphyrin-III C-methyltransferase / precorrin-2 Ga0192503_11954 51 kDa 1 2% dehydrogenase / sirohydrochlorin ferrochelatase Genome of RNAse E Ga0192503_11964 100 kDa 3 3% Rhodanobacter sp. [NiFe]-hydrogenase II apoprotein, ferredoxin-type subunit Ga0192503_11991 34 kDa 2 3% PCA2 hydrogenase large subunit Ga0192503_11992 59 kDa 1 2% probable Rubsico expression protein CbbX Ga0192503_119102 35 kDa 1 3% hypothetical protein Ga0192503_119119 7 kDa 1 12% phage protein Ga0192503_119124 20 kDa 1 6% putative ATP-binding cassette transporter Ga0192503_119133 71 kDa 1 1% cysteine synthase A Ga0192503_119137 34 kDa 2 5% phosphoglucosamine mutase Ga0192503_12019 46 kDa 1 2% oligoribonuclease Ga0192503_12021 24 kDa 5 22% tyrosine-protein kinase Etk/Wzc Ga0192503_12035 79 kDa 1 1% tRNA-adenosine deaminase Ga0192503_12054 40 kDa 6 16% inosine-5'-monophosphate dehydrogenase Ga0192503_12056 55 kDa 3 5% GTP-binding protein Ga0192503_12060 52 kDa 1 2% 103

dihydrodipicolinate reductase Ga0192503_120102 27 kDa 1 3% phosphoserine aminotransferase apoenzyme Ga0192503_120131 40 kDa 1 3% ATP-dependent Clp protease adaptor protein ClpS Ga0192503_120136 11 kDa 1 8% aconitase Ga0192503_120141 93 kDa 1 1% glutathione S-transferase Ga0192503_120164 23 kDa 1 4% cysteine synthase A Ga0192503_120166 33 kDa 1 2% ParB-like nuclease domain-containing protein Ga0192503_1211 33 kDa 1 3% succinate dehydrogenase subunit C Ga0192503_12131 18 kDa 2 11% succinate dehydrogenase subunit D Ga0192503_12132 16 kDa 2 16% succinate dehydrogenase subunit A Ga0192503_12133 64 kDa 8 14% amidohydrolase Ga0192503_12145 41 kDa 2 6% urocanate hydratase Ga0192503_12161 61 kDa 1 1% Genome of Rhodanobacter sp. uncharacterized zinc-type alcohol dehydrogenase-like protein Ga0192503_12175 37 kDa 3 7% PCA2 beta-lactamase class A Ga0192503_12178 31 kDa 1 2% L-alanine dehydrogenase Ga0192503_121100 39 kDa 1 4% Cu+-exporting ATPase Ga0192503_121109 90 kDa 1 1% 3-oxoacyl-[acyl-carrier-protein] synthase III Ga0192503_121140 35 kDa 1 3% [Acyl-carrier-protein] S-malonyltransferase Ga0192503_121141 30 kDa 2 9% 3-oxoacyl-[acyl-carrier-protein] synthase II Ga0192503_121148 44 kDa 2 4% NADH dehydrogenase subunit B Ga0192503_121158 20 kDa 3 11% NADH dehydrogenase subunit D Ga0192503_121160 48 kDa 3 8% ribosome maturation factor RimP Ga0192503_121172 19 kDa 1 5% curved DNA-binding protein Ga0192503_12213 33 kDa 1 2% dihydrolipoamide dehydrogenase Ga0192503_12217 49 kDa 2 2% asparagine synthase (glutamine-hydrolysing) Ga0192503_12416 70 kDa 1 1% 104

hypothetical protein Ga0192503_1273 42 kDa 1 2% leucyl/phenylalanyl-tRNA--protein transferase Ga0192503_1274 27 kDa 1 3% transcriptional regulator, RpiR family Ga0192503_12720 34 kDa 1 3% polyphosphate kinase Ga0192503_12813 82 kDa 1 1% pantothenate synthetase Ga0192503_13012 32 kDa 1 3% peroxiredoxin Q/BCP Ga0192503_13021 17 kDa 2 10% isocitrate dehydrogenase (NAD+) Ga0192503_13031 36 kDa 2 5% UDP-2,3-diacylglucosamine hydrolase Ga0192503_13041 27 kDa 1 4% preprotein translocase subunit SecD Ga0192503_13055 48 kDa 1 2% 4-aminobutyrate aminotransferase Ga0192503_13111 47 kDa 3 7% alanyl-tRNA synthetase Ga0192503_13126 83 kDa 3 4% nicotinamide-nucleotide amidase Ga0192503_13130 46 kDa 1 4% Genome of Rhodanobacter sp. ribosome-associated protein Ga0192503_1321 14 kDa 1 6% PCA2 methylglutaconyl-CoA hydratase Ga0192503_1338 34 kDa 2 4% glutamate-5-semialdehyde dehydrogenase Ga0192503_13324 45 kDa 5 10% N-acetyl-gamma-glutamyl-phosphate reductase Ga0192503_13327 35 kDa 2 4% N-acetylglutamate kinase Ga0192503_13328 31 kDa 2 6% cysteinyl-tRNA synthetase Ga0192503_13335 52 kDa 3 7% 3-demethylubiquinone-9 3-methyltransferase Ga0192503_13342 27 kDa 1 3% signal recognition particle subunit FFH/SRP54 (srp54) Ga0192503_13352 55 kDa 1 3% GTP-binding protein Ga0192503_13357 51 kDa 3 7% excinuclease ABC subunit A Ga0192503_13360 106 kDa 1 1% DNA mismatch repair protein MutL Ga0192503_13382 66 kDa 1 1% tRNA threonylcarbamoyladenosine biosynthesis protein TsaE Ga0192503_13384 16 kDa 1 10% hypothetical protein Ga0192503_13395 18 kDa 5 26% 105

ABC-2 type transport system ATP-binding protein Ga0192503_133124 36 kDa 1 3% two component transcriptional regulator, LuxR family Ga0192503_133126 23 kDa 2 8% cyclopropane-fatty-acyl-phospholipid synthase Ga0192503_133138 48 kDa 1 3% two-component system, chemotaxis family, response Ga0192503_1358 34 kDa 1 2% regulator CheV efflux transporter, outer membrane factor (OMF) lipoprotein, Ga0192503_13614 51 kDa 1 2% NodT family phosphoribosylformylglycinamidine cyclo-ligase Ga0192503_13629 37 kDa 1 3% Hpr(Ser) kinase/phosphatase Ga0192503_13645 36 kDa 1 3% hypothetical protein Ga0192503_13660 87 kDa 1 1% hypothetical protein Ga0192503_1379 60 kDa 1 1% dimethyladenosine transferase Ga0192503_14021 33 kDa 5 11% cell division protein FtsZ Ga0192503_14027 39 kDa 2 7% Genome of Rhodanobacter sp. MraZ protein Ga0192503_14040 16 kDa 3 21% PCA2 D-lactate dehydrogenase Ga0192503_14045 50 kDa 2 3% ATP-binding cassette, subfamily F, uup Ga0192503_14057 70 kDa 1 1% cytochrome c oxidase accessory protein FixG Ga0192503_14069 58 kDa 1 2% 5,10-methylenetetrahydrofolate reductase (NAD(P)) Ga0192503_14076 30 kDa 1 3% indole-3-glycerol phosphate synthase Ga0192503_14090 28 kDa 1 4% anthranilate phosphoribosyltransferase Ga0192503_14091 37 kDa 2 7% type I restriction-modification system, DNA methylase Ga0192503_140140 111 kDa 1 1% subunit integrase Ga0192503_140141 47 kDa 1 2% dihydrodipicolinate reductase Ga0192503_140189 26 kDa 3 9% SSU ribosomal protein S6P modification protein Ga0192503_14113 33 kDa 1 2% DNA-binding transcriptional regulator, MarR family Ga0192503_14218 18 kDa 1 5% Ca-activated chloride channel family protein Ga0192503_14222 36 kDa 2 5% 106

hypothetical protein Ga0192503_14232 26 kDa 1 7% TIGR00255 family protein Ga0192503_14248 32 kDa 2 5% ABC-type multidrug transport system, ATPase component Ga0192503_1432 93 kDa 2 3% membrane protease subunit, stomatin/prohibitin family, Ga0192503_14412 40 kDa 1 2% contains C-terminal Zn-ribbon domain DNA helicase/exodeoxyribonuclease V, beta subunit Ga0192503_14429 97 kDa 1 1% sec-independent protein translocase protein TatB Ga0192503_14712 15 kDa 1 5% epoxide hydrolase. Serine peptidase. MEROPS family S33 Ga0192503_14716 30 kDa 1 3% O-sialoglycoprotein endopeptidase Ga0192503_14733 35 kDa 1 4% Mg2+-importing ATPase Ga0192503_14741 96 kDa 1 1% DNA-(apurinic or apyrimidinic site) lyase Ga0192503_14783 16 kDa 1 8% iron complex outermembrane recepter protein Ga0192503_14789 106 kDa 1 1% Genome of tRNA 2-thiocytidine biosynthesis protein TtcA Ga0192503_147117 30 kDa 1 5% Rhodanobacter sp. PCA2 signal transduction histidine kinase Ga0192503_10946 104 kDa 1 1% CRISPR-associated protein, Csy4 family Ga0192503_14829 21 kDa 1 5% L-threonine synthase Ga0192503_14838 37 kDa 3 11% amino acid adenylation domain-containing protein Ga0192503_14848 231 kDa 1 0% glycosyltransferase involved in cell wall bisynthesis Ga0192503_1494 41 kDa 1 2% acetolactate synthase, large subunit Ga0192503_14943 66 kDa 1 2% TolB amino-terminal domain-containing protein Ga0192503_14949 71 kDa 1 1% hypothetical protein Ga0192503_1505 22 kDa 1 4% ATP-independent RNA helicase DbpA Ga0192503_15149 51 kDa 1 2% ATP-dependent helicase HrpB Ga0192503_15158 96 kDa 1 1% thioredoxin reductase Ga0192503_1528 34 kDa 1 4% sterol desaturase/sphingolipid hydroxylase, fatty acid Ga0192503_15228 47 kDa 1 2% hydroxylase superfamily 107

deoxyuridine 5'-triphosphate nucleotidohydrolase Ga0192503_15311 16 kDa 2 8% phosphopantothenoylcysteine decarboxylase / Ga0192503_15312 44 kDa 1 2% phosphopantothenate--cysteine ligase ATP synthase F1 subcomplex epsilon subunit Ga0192503_15336 13 kDa 1 9% hydroxymethylbilane synthase Ga0192503_15361 33 kDa 1 4% chromosome segregation ATPase Ga0192503_15424 34 kDa 1 6% Genome of DNA gyrase subunit B Ga0192503_15467 75 kDa 2 3% Rhodanobacter sp. protein of unknown function (DUF2884) Ga0192503_1552 30 kDa 1 3% PCA2 MFS transporter, putative metabolite:H+ symporter Ga0192503_15518 56 kDa 1 1% Tfp pilus assembly protein PilF Ga0192503_15529 58 kDa 1 2% protease-4 Ga0192503_15539 63 kDa 1 2% cell division ATP-binding protein FtsE Ga0192503_15550 26 kDa 2 6% NADP-dependent 3-hydroxy acid dehydrogenase YdfG Ga0192503_11531 27 kDa 2 6% hypothetical protein Ga0192503_1399 71 kDa 1 1%

108

Table S3 The proteins exclusively expressed in Pyr cultures Average Average Molecular Protein source Annotation Locus tag in JGI number of percent Weight unique peptides coverage Partial genome (IMG locus tag Ga0100919_ thioester reductase domain-containing protein Ga0100919_111637 74 kDa 1 2% 111461-111871) of Mycobacterium fortuitum CT6 dipeptidyl aminopeptidase/acylaminoacyl peptidase Ga0192503_10118 73 kDa 1 2% putative Flp pilus-assembly TadE/G-like Ga0192503_10132 58 kDa 1 2% phenylalanine 4-hydroxylase Ga0192503_10154 34 kDa 1 2% predicted PurR-regulated permease PerM Ga0192503_10165 41 kDa 1 2% alcohol dehydrogenase (cytochrome c) Ga0192503_10176 68 kDa 17 19% DNA topoisomerase I Ga0192503_101103 92 kDa 1 1% PAS domain S-box-containing protein/diguanylate cyclase Ga0192503_101119 89 kDa 1 1% (GGDEF) domain-containing protein Genome of cell division protein ZapE Ga0192503_101131 41 kDa 1 2% Rhodanobacter hypothetical protein Ga0192503_101133 25 kDa 1 4% sp. PCA2 16S rRNA m(2)G 1207 methyltransferase Ga0192503_101149 40 kDa 1 2% membrane fusion protein, multidrug efflux system Ga0192503_1026 42 kDa 1 2% thiol-disulfide isomerase or thioredoxin Ga0192503_10213 30 kDa 1 3% 6-phosphogluconolactonase Ga0192503_10311 25 kDa 1 3% 6-phosphogluconate dehydratase Ga0192503_10312 61 kDa 2 3% hypothetical protein Ga0192503_10314 21 kDa 1 4% hypothetical protein Ga0192503_10315 25 kDa 1 3% hexosaminidase Ga0192503_10323 84 kDa 1 1% 109

IMP cyclohydrolase Ga0192503_1061 57 kDa 1 2% /phosphoribosylaminoimidazolecarboxamide formyltransferase hydroxymethylpyrimidine synthase Ga0192503_10711 70 kDa 2 3% NADPH2:quinone reductase Ga0192503_10714 36 kDa 1 2% K(+)-stimulated pyrophosphate-energized sodium pump Ga0192503_10722 70 kDa 5 7% ABC-2 type transport system ATP-binding protein Ga0192503_10729 33 kDa 1 2% acetylornithine aminotransferase apoenzyme Ga0192503_10732 42 kDa 2 4% flagellar motor switch protein FliG Ga0192503_10813 37 kDa 1 2% flagellar assembly protein FliH Ga0192503_10814 23 kDa 1 4% flagellar biosynthesis protein FlhF Ga0192503_10827 46 kDa 1 3% glycerol kinase Ga0192503_10850 55 kDa 3 5% rod shape-determining protein MreB Ga0192503_10852 36 kDa 1 3% Genome of serine protease DegQ Ga0192503_10869 48 kDa 1 3% Rhodanobacter sp. PCA2 hypothetical protein Ga0192503_10873 35 kDa 1 2% acetyl-CoA synthetase Ga0192503_10898 71 kDa 9 12% pyruvate dehydrogenase E1 component beta subunit Ga0192503_1092 49 kDa 4 5% pyruvate dehydrogenase E1 component alpha subunit Ga0192503_1093 38 kDa 4 5% phosphopantetheine adenylyltransferase Ga0192503_10925 18 kDa 1 14% two-component system, OmpR family, sensor histidine kinase Ga0192503_10941 96 kDa 1 1% KdpD serine/threonine-protein kinase HipA Ga0192503_10967 46 kDa 1 4% helix-turn-helix domain-containing protein Ga0192503_10997 9 kDa 1 9% sulfide-quinone oxidoreductase Ga0192503_109113 46 kDa 1 2% pyruvate kinase Ga0192503_109119 52 kDa 2 6% fructose-bisphosphate aldolase Ga0192503_109121 36 kDa 1 3% glutamine--fructose-6-phosphate transaminase Ga0192503_109140 66 kDa 1 2%

110

GDPmannose 4,6-dehydratase Ga0192503_109167 36 kDa 3 5% glycosyltransferase, GT2 family Ga0192503_109172 75 kDa 1 1% aspartyl-tRNA(Asn)/glutamyl-tRNA(Gln) amidotransferase Ga0192503_109174 53 kDa 2 5% subunit A sulfate adenylyltransferase subunit 2 Ga0192503_109175 36 kDa 3 7% CBS domain-containing protein Ga0192503_109184 16 kDa 1 10% nitroreductase Ga0192503_109212 21 kDa 1 4% NAD(P) transhydrogenase subunit alpha Ga0192503_109214 42 kDa 1 2% NAD(P)H dehydrogenase (quinone) Ga0192503_109234 21 kDa 4 18% bacterial peptide chain release factor 1 (bRF-1) Ga0192503_109239 40 kDa 1 5% glutamyl-tRNA reductase Ga0192503_109240 47 kDa 1 2% LSU ribosomal protein L25P Ga0192503_109246 24 kDa 4 20% Genome of FecR family protein Ga0192503_109251 37 kDa 1 2% Rhodanobacter beta-aspartyl-peptidase (threonine type) Ga0192503_109253 70 kDa 1 1% sp. PCA2 LSU ribosomal protein L11P Ga0192503_109262 16 kDa 7 46% LSU ribosomal protein L12P Ga0192503_109265 13 kDa 5 25% LSU ribosomal protein L23P Ga0192503_109275 11 kDa 3 22% LSU ribosomal protein L2P Ga0192503_109276 30 kDa 9 21% LSU ribosomal protein L24P Ga0192503_109284 11 kDa 4 29% SSU ribosomal protein S14P Ga0192503_109286 12 kDa 2 16% LSU ribosomal protein L30P Ga0192503_109291 7 kDa 4 49% LSU ribosomal protein L15P Ga0192503_109292 17 kDa 8 32% 6,7-dimethyl-8-ribityllumazine synthase Ga0192503_109313 17 kDa 5 28% MFS transporter, FHS family, L-fucose permease Ga0192503_11019 46 kDa 1 2% response regulator receiver domain-containing protein Ga0192503_11024 56 kDa 1 1% Trehalose and maltose hydrolase (possible phosphorylase) Ga0192503_11029 88 kDa 1 1% 111

succinyl-CoA synthetase (ADP-forming) alpha subunit Ga0192503_11037 30 kDa 2 4% NAD+ synthase (glutamine-hydrolysing) Ga0192503_11040 60 kDa 1 1% KUP system potassium uptake protein Ga0192503_11054 71 kDa 1 1% diguanylate cyclase Ga0192503_11070 39 kDa 1 2% adenine phosphoribosyltransferase Ga0192503_11072 20 kDa 1 7% 1-deoxy-D-xylulose-5-phosphate synthase Ga0192503_11086 30 kDa 1 3% 1-deoxy-D-xylulose-5-phosphate synthase Ga0192503_11087 56 kDa 2 4% alpha-1,2-mannosidase, putative Ga0192503_11118 88 kDa 1 1% hypothetical protein Ga0192503_11235 62 kDa 1 1% aspartate carbamoyltransferase Ga0192503_11236 35 kDa 2 8% thiol-disulfide isomerase or thioredoxin Ga0192503_11270 17 kDa 1 2% thymidylate synthase Ga0192503_11284 30 kDa 2 8% Genome of microcin-processing peptidase 1. Unknown type peptidase. Rhodanobacter Ga0192503_11289 48 kDa 1 2% sp. PCA2 MEROPS family U62 acetyl-CoA carboxylase carboxyltransferase subunit alpha Ga0192503_11295 35 kDa 1 3% acyl-[acyl-carrier-protein]--UDP-N-acetylglucosamine O- Ga0192503_112100 30 kDa 1 3% acyltransferase ribosome recycling factor Ga0192503_112108 21 kDa 5 26% LSU ribosomal protein L35P Ga0192503_112126 8 kDa 3 33% uracil phosphoribosyltransferase Ga0192503_112137 23 kDa 1 4% type IV pilus assembly protein PilB Ga0192503_1134 62 kDa 1 2% phosphatidylinositol alpha-1,6-mannosyltransferase Ga0192503_1136 41 kDa 1 2% endothelin-converting enzyme Metallo peptidase. MEROPS Ga0192503_1151 74 kDa 1 1% family M13 putative ABC transport system ATP-binding protein Ga0192503_1154 26 kDa 1 3% hypothetical protein Ga0192503_11540 49 kDa 1 2% aminomuconate-semialdehyde/2-hydroxymuconate-6- Ga0192503_11545 52 kDa 1 3% semialdehyde dehydrogenase 112

isoleucyl-tRNA synthetase Ga0192503_11560 105 kDa 2 1% methionine synthase (B12-dependent) Ga0192503_11611 100 kDa 2 1% ABC-2 type transport system ATP-binding protein Ga0192503_11635 35 kDa 1 2% phospholipid N-methyltransferase Ga0192503_11657 21 kDa 1 4% Na+/proline symporter Ga0192503_1171 119 kDa 1 1% malate dehydrogenase (NAD) Ga0192503_1178 34 kDa 10 19% RNA-binding protein Hfq Ga0192503_11727 9 kDa 1 16% protease FtsH subunit HflK Ga0192503_11729 39 kDa 2 4% adenylosuccinate synthetase Ga0192503_11731 46 kDa 6 13% glycine dehydrogenase (decarboxylating) alpha subunit Ga0192503_11741 47 kDa 2 4% /glycine dehydrogenase (decarboxylating) beta subunit UPF0271 protein Ga0192503_11774 26 kDa 1 4% Genome of hypothetical protein Ga0192503_11786 53 kDa 1 2% Rhodanobacter tryptophan synthase, beta chain Ga0192503_1185 45 kDa 5 11% sp. PCA2 glutamyl-tRNA synthetase Ga0192503_11812 53 kDa 1 2% zinc uptake regulator, Fur family Ga0192503_11814 19 kDa 1 6% sensor domain CHASE2-containing protein Ga0192503_11837 43 kDa 1 2% CTP synthase Ga0192503_11845 61 kDa 1 2% 2-dehydro-3-deoxyphosphooctonate aldolase (KDO 8-P Ga0192503_11846 30 kDa 1 4% synthase) 5'-nucleotidase /3'-nucleotidase /exopolyphosphatase Ga0192503_11853 27 kDa 2 7% ATP-dependent Clp protease proteolytic subunit ClpP Ga0192503_11868 23 kDa 9 27% UDPglucose 6-dehydrogenase Ga0192503_11895 47 kDa 2 4% assimilatory nitrate reductase (NADH) beta subunit Ga0192503_11952 45 kDa 1 2% assimilatory nitrate reductase (NADH) alpha subunit Ga0192503_11953 94 kDa 1 1% apoprotein uroporphyrinogen-III synthase Ga0192503_11955 30 kDa 1 3% 113

respiratory nitrate reductase beta subunit Ga0192503_11959 34 kDa 1 4% fructose-bisphosphate aldolase Ga0192503_119106 37 kDa 1 3% oligopeptidase A Metallo peptidase. MEROPS family M03A Ga0192503_119138 75 kDa 1 1% hypothetical protein Ga0192503_12017 29 kDa 1 4% phosphoglucosamine mutase Ga0192503_12019 48 kDa 2 5% glycosyltransferase involved in cell wall bisynthesis Ga0192503_12047 42 kDa 1 2% GMP synthase (glutamine-hydrolyzing) Ga0192503_12055 58 kDa 2 3% 23S rRNA m(2)A-2503 methyltransferase Ga0192503_12065 47 kDa 2 6% 3-hydroxyacyl-CoA dehydrogenase Ga0192503_12068 32 kDa 3 6% acetyl-CoA acyltransferase Ga0192503_12069 42 kDa 2 3% transcriptional regulator, TraR/DksA family Ga0192503_12071 9 kDa 1 8%

Genome of molecular chaperone DnaJ Ga0192503_120103 41 kDa 2 7% Rhodanobacter RNA polymerase, sigma subunit, ECF family Ga0192503_120118 47 kDa 2 2% sp. PCA2 excinuclease ABC subunit C Ga0192503_120126 69 kDa 2 2% ADP-ribose pyrophosphatase YjhB, NUDIX family Ga0192503_120137 18 kDa 1 4% tRNA (5-methylaminomethyl-2-thiouridylate)- Ga0192503_120138 41 kDa 1 3% methyltransferase long-chain acyl-CoA synthetase Ga0192503_120143 64 kDa 1 2% sulfite reductase (NADPH) beta subunit Ga0192503_120170 67 kDa 9 15% succinate dehydrogenase subunit B Ga0192503_12134 29 kDa 1 3% NAD-dependent protein deacetylase, SIR2 family Ga0192503_12174 30 kDa 1 1% molybdate transport system regulatory protein Ga0192503_12181 27 kDa 1 3% PAS domain S-box-containing protein/diguanylate cyclase Ga0192503_12185 99 kDa 1 1% (GGDEF) domain-containing protein membrane protein required for colicin V production Ga0192503_121114 18 kDa 1 5% histidine ammonia-lyase Ga0192503_121120 68 kDa 1 1%

114

LSU ribosomal protein L32P Ga0192503_121139 7 kDa 4 34% NADH dehydrogenase subunit A Ga0192503_121157 14 kDa 1 12% NADH dehydrogenase subunit C Ga0192503_121159 24 kDa 4 19% NADH dehydrogenase subunit F Ga0192503_121162 48 kDa 2 4% NADH dehydrogenase subunit G Ga0192503_121163 82 kDa 1 1% NADH dehydrogenase subunit I Ga0192503_121165 19 kDa 5 25% methylthioribose-1-phosphate isomerase Ga0192503_1229 39 kDa 1 2% DNA gyrase subunit A Ga0192503_12210 96 kDa 1 1% hypothetical protein Ga0192503_1242 17 kDa 1 6% thermostable hemolysin Ga0192503_1248 26 kDa 1 3%

uncharacterized protein, YigZ family Ga0192503_12715 21 kDa 1 4%

Genome of CubicO group peptidase, beta-lactamase class C family Ga0192503_12718 46 kDa 1 3% Rhodanobacter sp. PCA2 CubicO group peptidase, beta-lactamase class C family Ga0192503_12723 73 kDa 1 1% enoyl-[acyl-carrier protein] reductase / trans-2-enoyl-CoA Ga0192503_12728 43 kDa 1 2% reductase (NAD+) iron-regulated ABC transporter membrane component SufB Ga0192503_13047 54 kDa 1 2% transcriptional regulator, BadM/Rrf2 family Ga0192503_13048 17 kDa 1 6% AraC-type DNA-binding protein Ga0192503_13058 34 kDa 1 2% acyl carrier protein Ga0192503_1318 10 kDa 1 10% 3-oxoacyl-[acyl-carrier-protein] synthase II Ga0192503_1319 44 kDa 1 2% DNA mismatch repair protein MutS Ga0192503_13131 94 kDa 2 1% TonB-dependent Receptor Plug Domain Ga0192503_13136 113 kDa 1 1% molecular chaperone HtpG Ga0192503_13142 67 kDa 14 24% translation elongation factor P (EF-P) Ga0192503_13318 21 kDa 5 27% argininosuccinate lyase Ga0192503_13326 49 kDa 9 18% 115

N-acetylornithine carbamoyltransferase Ga0192503_13333 34 kDa 1 4% phosphoglycolate phosphatase Ga0192503_13341 24 kDa 1 3% tRNA threonylcarbamoyladenosine biosynthesis protein TsaE Ga0192503_13384 17 kDa 1 3% protein SCO1/2 Ga0192503_133119 23 kDa 1 4% uncharacterized protein, PA2063/DUF2235 family Ga0192503_133135 40 kDa 1 2% succinyldiaminopimelate desuccinylase Ga0192503_133151 42 kDa 1 2% 2,3,4,5-tetrahydropyridine-2,6-dicarboxylate N- Ga0192503_133154 29 kDa 2 6% succinyltransferase arsenite transporter, ACR3 family Ga0192503_13427 39 kDa 1 2% UDP-N-acetylglucosamine 1-carboxyvinyltransferase Ga0192503_13634 45 kDa 6 12% 4-hydroxy-tetrahydrodipicolinate synthase Ga0192503_13655 32 kDa 4 16% inner membrane protein Ga0192503_13669 48 kDa 2 2% Genome of L-lactate dehydrogenase (cytochrome) Ga0192503_13670 41 kDa 1 2% Rhodanobacter hypothetical protein Ga0192503_13681 54 kDa 1 2% sp. PCA2 RNA helicase HrpA Ga0192503_1371 84 kDa 1 1% autotransporter secretion inner membrane protein TamB Ga0192503_13712 131 kDa 1 1% isoquinoline 1-oxidoreductase, beta subunit Ga0192503_13721 81 kDa 2 1% translation elongation factor 2 (EF-2/EF-G) Ga0192503_1383 76 kDa 25 28% uncharacterized conserved protein, DUF58 family, contains Ga0192503_1385 48 kDa 1 2% vWF domain aminopeptidase A. Metallo peptidase. MEROPS family M17 Ga0192503_1408 52 kDa 7 13% 2-octaprenyl-6-methoxyphenol hydroxylase /2-octaprenyl-3- Ga0192503_14013 51 kDa 1 2% methyl-6-methoxy-1,4-benzoquinol hydroxylase bis(5'nucleosyl)-tetraphosphatase, ApaH Ga0192503_14023 31 kDa 1 5% D-alanine--D-alanine ligase Ga0192503_14030 33 kDa 1 3% UDP-N-acetylglucosamine-N-acetylmuramylpentapeptide N- Ga0192503_14032 38 kDa 2 5% acetylglucosamine transferase adenosylhomocysteinase Ga0192503_14061 51 kDa 12 19% 116

Outer membrane protein TolC Ga0192503_140125 46 kDa 1 2% hypothetical protein Ga0192503_140146 21 kDa 1 7% hypothetical protein Ga0192503_140157 8 kDa 1 10% LSU ribosomal protein L13P Ga0192503_140161 17 kDa 5 17% DNA ligase (NAD+) Ga0192503_140172 73 kDa 1 1% hydrophobic/amphiphilic exporter-1, HAE1 family Ga0192503_14216 110 kDa 2 2% membrane fusion protein, multidrug efflux system Ga0192503_14217 39 kDa 1 3% MoxR-like ATPase Ga0192503_14225 39 kDa 3 7% type IV pilus assembly protein PilQ Ga0192503_14226 78 kDa 1 2% type IV pilus assembly protein PilO Ga0192503_14228 24 kDa 1 3% ATP-dependent DNA helicase RecG Ga0192503_14236 76 kDa 1 1% DNA-directed RNA polymerase subunit omega Ga0192503_14239 15 kDa 6 45% Genome of Rhodanobacter ABC-2 family transporter protein Ga0192503_1433 134 kDa 1 1% sp. PCA2 TonB family C-terminal domain-containing protein Ga0192503_14421 30 kDa 1 3% transcription termination factor Rho Ga0192503_1461 47 kDa 4 8% sec-independent protein translocase protein TatA Ga0192503_14713 9 kDa 5 63% outer membrane usher protein Ga0192503_14726 93 kDa 1 1% phosphate:Na+ symporter Ga0192503_14740 59 kDa 1 1% glycerol 3-phosphate dehydrogenase (NAD(P)+) Ga0192503_14772 33 kDa 1 4% LSU ribosomal protein L28P Ga0192503_14774 11 kDa 6 54% ABC-2 type transport system permease protein Ga0192503_14780 26 kDa 1 1% ATP-dependent DNA helicase Rep Ga0192503_14787 109 kDa 1 1% glycerol-3-phosphate acyltransferase Ga0192503_14793 100 kDa 1 1% 2-octaprenyl-6-methoxy-1,4-benzoquinone methylase Ga0192503_14794 27 kDa 1 5% /demethylmenaquinone methyltransferase glutathione S-transferase Ga0192503_14798 25 kDa 1 2% 117

aromatic amino acid aminotransferase apoenzyme Ga0192503_147116 43 kDa 1 2% efflux transporter, outer membrane factor (OMF) lipoprotein, Ga0192503_147121 52 kDa 1 1% NodT family predicted Zn-dependent peptidase Ga0192503_147127 103 kDa 1 1% transcriptional regulator /transcriptional regulator, LysR family Ga0192503_147138 39 kDa 1 2% magnesium chelatase family protein Ga0192503_147142 37 kDa 1 3% nitrogen regulatory protein P-II family Ga0192503_147144 12 kDa 1 10% transcriptional regulator, LysR family Ga0192503_14817 34 kDa 1 2% homoserine dehydrogenase Ga0192503_14836 42 kDa 1 2% hypothetical protein Ga0192503_14842 101 kDa 1 1% glycosyltransferase involved in cell wall bisynthesis Ga0192503_1497 44 kDa 1 2% 2-isopropylmalate synthase Ga0192503_14935 57 kDa 1 2% Genome of 3-isopropylmalate/(R)-2-methylmalate dehydratase large Ga0192503_14937 51 kDa 7 15% Rhodanobacter subunit sp. PCA2 dihydroxyacid dehydratase Ga0192503_14946 66 kDa 1 3% NAD+ synthetase Ga0192503_14951 72 kDa 1 1% release factor glutamine methyltransferase Ga0192503_14952 32 kDa 1 3% biotin synthase Ga0192503_14961 36 kDa 1 3% glycosidase Ga0192503_14970 58 kDa 1 1% glycyl-tRNA synthetase beta chain Ga0192503_1506 78 kDa 2 1% glycyl-tRNA synthetase alpha chain Ga0192503_1507 35 kDa 1 2% murein L,D-transpeptidase YcbB/YkuD Ga0192503_1512 60 kDa 1 1% galactonate dehydratase Ga0192503_1519 42 kDa 1 2% beta-xylosidase, GH43 family Ga0192503_15113 39 kDa 1 2% arylsulfatase A Ga0192503_15118 63 kDa 1 1% Zn-dependent amino- or carboxypeptidase, M28 family Ga0192503_15154 50 kDa 1 2%

118

arginine decarboxylase Ga0192503_15320 68 kDa 1 1% coproporphyrinogen oxidase Ga0192503_15327 35 kDa 1 5% ATP synthase F1 subcomplex delta subunit Ga0192503_15340 20 kDa 1 11% AmpE protein Ga0192503_15353 32 kDa 1 3% part of AAA domain-containing protein Ga0192503_1546 59 kDa 1 1% hypothetical protein Ga0192503_1548 26 kDa 1 3% Genome of dolichol-phosphate mannosyltransferase Ga0192503_15422 26 kDa 1 3% Rhodanobacter sp. PCA2 hypothetical protein Ga0192503_15445 25 kDa 1 5% ATP-dependent HslUV protease ATP-binding subunit HslU Ga0192503_15448 48 kDa 4 10% uncharacterized conserved protein Ga0192503_15453 40 kDa 1 4% protein translocase subunit yidC Ga0192503_15461 68 kDa 2 3% hypothetical protein Ga0192503_15511 17 kDa 1 5% 23S rRNA (pseudouridine1915-N3)-methyltransferase Ga0192503_11519 17 kDa 1 5% imidazole glycerol phosphate synthase subunit hisF Ga0192503_11825 11 kDa 2 22%

119

References

Catelani D, Colombi A, Sorlini C & Treccani V (1973) Metabolism of biphenyl. 2- Hydroxy-6-oxo-6-phenylhexa-2,4-dienoate- the meta-cleavage product from 2,3- dihydroxybiphenyl by Pseudomonas putida. Biochemical Journal 134: 1063-1066.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology 200: e00763-17.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Dietrich LEP, Teal TK, Price-Whelan A & Newman DK (2008) Redox-active antibiotics control gene expression and community behavior in divergent bacteria. Science 321: 1203-1206.

Evans WC, Fernley HN & Griffiths E (1965) Oxidative metabolism of phenanthrene and anthracene by soil pseudomonads. Biochemical Journal 95: 819-831.

Feng L, Wang W, Cheng J, et al. (2007) Genome and proteome of long-chain alkane degrading Geobacillus thermodenitrificans NG80-2 isolated from a deep-subsurface oil reservoir. Proceedings of the National Academy of Sciences of the United States of America 104: 5602-5607.

Jugder BE, Ertan H, Wong YK, Braidy N, Manefield M, Marquis CP & Lee M (2016) Genomic, transcriptomic and proteomic analyses of Dehalobacter UNSWDHB in response to chloroform. Environmental Microbiology Report 8: 814-824.

Lipscomb JD (2008) Mechanism of extradiol aromatic ring-cleaving dioxygenases. Current Opinion in Structural Biology 18: 644-649. 120

Mansfeldt CB, Rowe AR, Heavner GL, Zinder SH & Richardson RE (2014) Meta- analyses of Dehalococcoides mccartyi strain 195 transcriptomic profiles identify a respiration rate-related gene expression transition point and interoperon recruitment of a key oxidoreductase subunit. Applied and Environmental Microbiology 80: 6062-6072.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

Melamud E, Vastag L & Rabinowitz JD (2010) Metabolomic analysis and visualization engine for LC-MS data. Analytical Chemistry 82: 9818-9826.

Moody JD, Doerge DR, Freeman JP & Cerniglia CE (2002) Degradation of biphenyl by Mycobacterium sp. strain PYR-1. Applied Microbiology and Biotechnology 58: 364-369.

Nesvizhskii AI, Keller A, Kolker E & Aebersold R (2003) A statistical model for identifying proteins by tandem mass spectrometry. Analytical Chemistry 75: 4646-4658.

Norman RS, Moeller P, McDonald TJ & Morris PJ (2004) Effect of pyocyanin on a crude- oil-degrading microbial community. Applied and Environmental Microbiology 70: 4004- 4011.

Race PR, Lovering AL, Green RM, Ossor A, White SA, Searle PF, Wrighton CJ & Hyde EI (2005) Structural and mechanistic studies of Escherichia coli nitroreductase with the antibiotic nitrofurazone. Reversed binding orientations in different redox states of the enzyme. Journal of Biological Chemistry 280: 13256-13264.

Selesi D, Jehmlich N, von Bergen M, Schmidt F, Rattei T, Tischler P, Lueders T & Meckenstock RU (2010) Combined genomic and proteomic approaches identify gene clusters involved in anaerobic 2-methylnaphthalene degradation in the sulfate-reducing enrichment culture N47. Journal of Bacteriology 192: 295-306.

Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ & Klenk DC (1985) Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85.

121

Wang Y, Kern SE & Newman DK (2010) Endogenous phenazine antibiotics promote anaerobic survival of Pseudomonas aeruginosa via extracellular electron transfer. Journal of Bacteriology 192: 365-369.

Wisniewski JR, Zougman A, Nagaraj N & Mann M (2009) Universal sample preparation method for proteome analysis. Nature Methods 6: 359-362.

122

Chapter 5 Enrichment and isolation of phenazine-1-carboxylic acid (PCA)-degrading bacteria under anaerobic conditions

5.1 Introduction

Phenazines comprise a large group of molecules with nitrogen-containing heterocyclic ring structures, which are mainly produced by bacteria from phylogenetically diverse taxa under both aerobic and anaerobic conditions (Mavrodi et al., 2006, Wang et al., 2010). Phenazines have been attracting increasing scientific interest in anoxic and anaerobic environments. For example, phenazines functioning as extracellular electron shuttles have been reported to enhance anaerobic survival but not growth of their producer Pseudomonas aeruginosa in planktonic cultures (Wang et al., 2010, Sullivan et al., 2011) and on agar plates (Dietrich et al., 2013). Excess reducing equivalents were dissipated by phenazines to allow for ATP synthesis through facilitating the conversion of glucose and pyruvate into acetate, thereby promoting anaerobic survival of Pseudomonas aeruginosa (Glasser et al., 2014). Moreover, it has been revealed that phenazines mediate reduction from poorly soluble crystalline iron(III) to iron(II) under conditions with limited oxygen, whereby production or amendment of phenazines supported anaerobic growth of the producers, e.g. Pseudomonas chlororaphis PCL1391 and dissimilatory iron-reducing bacterium, e.g. Shewanella oneidensis MR1 (Hernandez et al., 2004). Additionally, a synthetic phenazine neutral red in a novel crystalline form has been found to increase methane production by an order of magnitude in coal and food waste fed with microbial communities and in pure Methanosarcina mazei cultures amended with acetate in anaerobic environments, through direct electron transfer to methanogens (Beckmann et al., 2016). Given the emerging importance of phenazines under oxygen-limited conditions additional information regarding the fate of phenazines in low redox environments is needed.

123

Currently, all studies on the turnover of phenazines were performed under aerobic conditions. Eight aerobic bacterial strains belonging to Alphaproteobacteria, Actinobacteria and Grammaproteobacteria were isolated and characterised capable of degrading phenazines (described in Figure 2.3) (Yang et al., 2007, Costa et al., 2015, Zhao et al., 2017, Costa et al., 2018 and in this thesis). Additionally, white-rot fungus Perenniporia subacida was able to degrade neutral red co-metabolically (Si et al., 2013). Despite the important role of phenazines under anaerobic conditions, it is still largely unknown how these compounds are transformed or degraded anaerobically.

In this chapter, phenazine-1-carboxylic acid (PCA) as a central bacterial phenazine was used as the carbon, nitrogen and energy source in this study to explore degradation of phenazines under anaerobic conditions. Iron(III) was found to serve as an electron acceptor in anaerobic PCA degrading enrichments inoculated with sewage sludge. Ultimately, Morganella morganii SL11 was isolated representing the first known anaerobic phenazine degrading bacterium.

5.2 Materials and methods

5.2.1 Inocula source and cultivation conditions

Sewage sludge from St Marys Sewage Treatment Plant in Sydney, Australia was used as the inoculum source for all enrichment and isolation cultures in this study. Unless stated otherwise, all cultures were established in 40 mL of media in 50 mL serum bottles, sealed with Teflon coated rubber septa and aluminium crimp caps. For initial start-up cultures, triplicates of sludge samples were cultivated in a defined minimal mineral medium containing the following salts (in g/L, unless stated otherwise): KH2PO4 (1.93), K2HPO4 (6.24), NaCl (2.5), MgCl2·6H2O (0.2), 1 mL of trace elements 1000x stock solution (SL 10), 1 mL of Selenite-Wolframate 1000x stock solution and vitamin 1000x stock solution (VS 10). Trace elements, Selenite- Wolframate and vitamin solutions were prepared as previously described by Tschech & Pfennig (1984), Widdel & Pfennig (1981) and Balch et al. (1979), respectively. PCA supplied at the concentration of 200 µM was then dissolved into the prepared mineral medium by stirring for 30 mins. Acetate, methanol or lactate were added at 20 mM as

124

electron donors when necessary. Sulphate, nitrate or iron(III) citrate were added at 10, 10 and 30 mM respectively as electron acceptors when necessary. The pH was adjusted to 7.0 using 1 M NaOH when necessary.

Culturing medium was dispensed into serum bottles and underwent gas exchange with N2 for 20 min. The bottles were then crimp sealed and the headspace was purged with N2/CO2 (80% :20%) for a further 5 min. The redox potential of the medium was lowered by the addition of sodium sulphide (Na2S·9H2O) solution to a final concentration of 0.1 mM and autoclaved at 121°C for 20 min. All experiments were set up in triplicates with abiotic controls and incubated statically at 30°C in the dark.

5.2.2 Enrichment of anaerobic PCA-degrading bacterium

Sludge (0.5 mL) was transferred anaerobically to six sets of triplicate culture flasks containing anaerobic mineral media with different electron acceptors (20 mM acetate, 20 mM methanol or 20 mM lactate) or donors (10 mM sulphate, 10 mM nitrate or 30 mM iron(III) citrate). Subsequent transfers were performed during the active phase of growth and before all of the PCA have been consumed. Throughout the three rounds of enrichment, parent cultures were anaerobically transferred into fresh media with 1% inocula and incubated for 2-4 months.

5.2.3 Isolation of anaerobic PCA-degrading bacterium

After three rounds of growth and enrichment, a dilution series was established in mineral media with 30 mM iron(III) citrate in 15 mL serum bottles. The first dilution (10-1) was inoculated with 1 mL of parent culture and 9 mL of media then diluted serially until 10-7. For soft agar shake series, semi solid media were prepared by adding 0.6% (w/v) of low melting agarose to media before autoclaving. After autoclaving, molten agarose was cooled to 37°C before addition of 0.15 mL of sterile vitamin solution from a 1,000x anoxic stock. Then a dilution series ranging from 10-3 to 10-8 was prepared using the 10-2 dilution (from the dilution to extinction series) as starting inoculum. The bottle of the highest dilution with clearly separated colonies was used to pick and transfer to fresh sterile mineral media to establish pure cultures. The method for colony picking was described previously (Löffler et al., 2005).

125

5.2.4 Analytical methods

5.2.4.1 PCA analysis

To determine the PCA transformation after incubation, 50 µL of cultures were sampled with a syringe then stored and exposed to the air for two days before subjected to HPLC analysis to make sure that reduced PCA in the cultures was completed re- oxidized to normal PCA. After that, samples were transferred into vials and then 10 µL of the samples were injected with an autosampler into an Agilent High-performance liquid chromatograph (HPLC; Agilent 1200 Series, Australia), equipped with a UV-Vis detector and a C18 column (Eclipse XDB-C18, 5 µm, 4.6 x 150 mm; Agilent, Australia). The mobile phases used for analysis of PCA are as described in Section 2.2.5.

5.2.4.2 Acetate and methanol analysis

Acetate was derivatized to ethyl acetate then quantified using a modified headspace analysis method as described in Lee et al. (1999). For derivatization, 0.1 mL of sample was added to a 10 mL vial containing 0.2 mL of ethanol and 0.4 mL of RO water. Then 0.2 mL of sulphuric acid was added and the vial was sealed immediately with a PTFE/butyl septum attached to an aluminium crimp cap. Samples were incubated at 60°C for 45 mins. Immediately before injection into the GC, samples were agitated at 80° for 5 min, then 0.25 mL of the headspace was withdrawn from the flask with a gas- tight glass syringe attached to an automatic sampler and directly injected into the GC. Acetate was quantified by comparison of the peak area of ethyl acetate to a six-point calibration curve of the corresponding standard. Calibration standards were prepared as for samples but adding known concentrations of sodium acetate. Headspace GC analysis were carried out on a Shimadzu Plus GC-2010 equipped with an AOC-5000 Plus autosampler and a flame ionization detector (FID). The column used was a DB- FFAP with the following dimensions 30 m x 0.32 mm x 0.25 µm (Agilent Technologies). GC conditions are described in Appendix 8.5.

Methanol was analysed using the modified headspace analysis method described for quantification of acetate. For the measurement of methanol, 0.2 mL of sample was added to a 10 mL vial containing 0.8 mL of RO water. Then the vial was sealed immediately with a PTFE/butyl septum attached to an aluminium crimp cap. Methanol 126

was quantified by comparison of the sample’s peak area to a six-point calibration curve of the corresponding standard. Calibration standards were prepared in the same method as samples but adding known concentrations of methanol. Headspace GC analysis were carried out in the same settings as mentioned above.

5.2.4.3 Lactate analysis

Lactate concentration was measured using a D-/L-Lactic Acid (Rapid) Assay Kit (Megazyme, Wicklow, Ireland) by following the manufacturer’s protocol. Sample was subjected to measurement at 340 nm by a microplate reader (BMG Labtech, Ortenberg, Germany).

5.2.4.4 Iron(III) and iron(II) analysis

Iron(III) was reduced to Iron(II) then quantified using a modified analysis method as described by (Viollier et al., 2000). For reduction of iron(III), 10 µL of sample was diluted in a 96-well microplate containing 75 µL of RO water before adding of 10 µL of ferrozine. Then 15 µL of reducing agent hydroxylamine hydrochloride (H2NOH·HCl) was added and the sample was incubated for 30 mins to complete the reducing reaction. As following, 15 µL of buffer solution was added and mixed with the sample. After incubation for 10 mins, sample was subjected to measurement at 562 nm by a microplate reader (BMG Labtech, Ortenberg, Germany). Iron(III) was quantified by comparison of the absorbance of the sample to a six-point calibration curve of the corresponding standard. Calibration standards were prepared as for samples but adding known concentrations of iron(III).

Iron(II) was analysed using the same method described for quantification of iron(III) without reducing reaction. For the measurement of iron(II), 10 µL of sample was added to a 96-well microplate containing 75 µL of RO water before mixing with 10 µL of ferrozine. Then 15 µL of buffer solution was added to the sample. After incubation for 10 mins, microplate reader analysis was carried out in the same settings as mentioned above. Iron(II) was quantified by comparison of the absorbance of the sample to a six- point calibration curve of the corresponding standard. Calibration standards were prepared in the same method as samples but adding known concentrations of iron(II).

127

5.2.4.5 Sulphate and sulphide analysis

Sulphate was quantified by turbidimetric determination as previously described by Lundquist et al. (1980). Briefly, 120 µL of sample was added to a 96-well microplate containing 40 µL of 0.5M HCl. Then 40 µL of Ba-PEG reagent was mixed with the sample and incubated for 15 mins before measuring the absorbance at 600 nm by a microplate reader (BMG Labtech, Ortenberg, Germany). Sulphate was quantified by comparison of the absorbance of the sample to a six-point calibration curve of the corresponding standard. Calibration standards were prepared as for samples but adding known concentrations of sodium sulphate.

Sulphide was quantified by colorimetric determination as described in Cline (1969). Briefly, 500 µL of sample was added to a 15 mL screw-cap tube which contained 450

µL of RO water and purged by N2 for 20 mins. Then 400 µL of sulphide reagent was anaerobically added to the sample and incubated for 1 h. 16 µL of the mixture was mixed with 184 µL of RO water in a 96-well microplate and subjected to measure the absorbance at 670 nm by a microplate reader (BMG Labtech, Ortenberg, Germany). Sulphide was quantified by comparison of the absorbance of the sample to a six-point calibration curve of the corresponding standard. Calibration standards were prepared as for samples but adding known concentrations of sodium sulphide.

5.2.4.6 Nitrate and nitrite analysis

Nitrate was reduced to nitrite then quantified with Griess reagents by following a modified analysis method as previously described by Miranda et al. (2001). For reduction of nitrate, 50 µL of sample was added to a 96-well microplate containing 50

µL of saturated VCl3 solution. Immediately, the Griess reagents comprising 50 µL of sulfanilamide (SULF) and 50 µL of N-(1-Naphthyl)ethylenediamine dihydrochloride (NEDD) was added to the sample and incubated for 30 mins before measuring the absorbance at 540 nm by a microplate reader (BMG Labtech, Ortenberg, Germany). Nitrate was quantified by comparison of the absorbance of the sample to a six-point calibration curve of the corresponding standard. Calibration standards were prepared as for samples but adding known concentrations of sodium nitrate.

Nitrite was analysed using the same method described for quantification of nitrate except that sample was only exposed to the Griess reagents. Nitrite was quantified by 128

comparison of the absorbance of the sample to a six-point calibration curve of the corresponding standard. Calibration standards were prepared in the same method as samples but adding known concentrations of sodium nitrite.

5.2.5 Genomic DNA extraction

Genomic DNA was extracted from 2 mL of PCA-degrading cultures as described in Section 2.2.7.

5.2.6 Phylogenetic characterization

The identification of 16S rRNA sequences was obtained by searching against reference 16S rRNA sequences using the BLAST algorithm in NCBI as described in Section 2.2.10.

5.2.7 Microscopic analyses

Cells in cultures was stained and counted by a fluorescence microscopy (Leica DFC420, Leica Microsystems, Australia) as described in Section 2.2.6

For Gram staining, 20 µL of the samples was placed onto 2% (w/v) agarose-coated slide and heat-fixed by a Bunsen burner. Then samples were stained using crystal violet for 1 min before mordanted by concentrated iodine for 1 min. Subsequently, alcohol iodine was used to de-stain crystal violet and concentrated iodine and was then washed thoroughly with distilled water. Samples were stained with carbol fuchsin for 5 min before viewed with light microscopy (Olympus DP71, Olympus©, Japan) at 1,000X magnification.

5.2.8 Community analysis

To analyse community composition in enriched cultures, genomic DNA was extracted from 2 mL of PCA-degrading cultures as described in Section 2.2.7, and then was used to amplify region V6-V8 of bacterial and archaeal 16S rRNA gene by a primer set 926F (AAACTYAAAKGAATTGACGG) and 1392R (ACGGGCGGTGTGTRC) (Guo et al., 2015). PCR reaction was performed in 40 µL volumes containing 20 µL PCR Master Mix (Promega, Australia), 0.13 g/L purified BSA (New England BioLabs, USA), 0.25 µM of each primer and 1 µL of diluted template DNA (~ 5 ng). Negative control reactions without DNA template were also included in all PCR analyses. The

129

PCR reactions were performed on a Mini-PCR (MJ MiniTM Gradient Thermal Cycler, Bio-Rad, Australia) set using the following thermal conditions: 3 min at 98°C, 35 cycles of 30 s at 95°C, 35 s at 62°C and 1 min at 72°C followed by 72°C for 10 min. The PCR products were subjected to clean up then sequenced by Illumina MiSeq sequencing with metagenomics library type 384 MIDs in Next Generation Sequencing Facility, Western Sydney University, Sydney, Australia. Analysis of metagenomic data was supplied by Illumina Basespace.

5.2.9 Statistical analysis

Analysis of variance were performed with one-way ANOVA followed by S-N-K using SPSS version 19.0 (IBM Co., Armonk, NY, US), respectively. P values below 0.05 were considered statistically significant.

5.3 Results

5.3.1 PCA degradation in anaerobic sludge based microcosms with different electron acceptors and donors

To examine if PCA could support bacterial growth as an electron donor or acceptor under anaerobic conditions, microcosms inoculated with sewage sludge were amended with PCA and a variety of typical bacterial electron acceptors and donors. Microcosms amended with iron(III) citrate showed anaerobic degradation of PCA at rates of 3.44 µM day-1 during the first 10 weeks of incubation and 10.24 µM day-1 during the ensuing 6 weeks (Figure 5.1). This was accompanied by a 30.3 mM decrease in iron(III) concentration and a 27.4 mM increase in iron(II) concentration (Figure 5.2A), suggesting PCA degradation might be linked to iron reduction. Cultures amended with sodium citrate showed no PCA degradation confirming ferric iron (not citrate) was associated with PCA loss (data not shown). No PCA degradation was observed in microcosms amended with nitrate or sulphate (Figure 5.1) although nitrate and sulphate reduction were observed, presumably from oxidation of reduced organic carbon incumbent in the sludge inoculum (Figure 5.2B, C)

130

Figure 5.1 PCA fate in anaerobic sludge based microcosms with different electron acceptors. Only microcosms amended with iron(III) citrate showed anaerobic degradation of PCA, indicating ferric iron was associated with PCA loss in the cultures with sludge. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures.

131

A. B.

Figure 5.2 Reduction of electron acceptors including iron(III) (A), nitrate (B) and sulphate (C) in anaerobic sludge-based microcosms. Iron(III) was significantly reduced to iron(II). Nitrate and sulphate reduction were observed, which presumably caused by oxidation of reduced organic carbon incumbent in the sludge inoculum.Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures.

132

In microcosms amended with acetate, methanol and lactate as electron donors, no PCA degradation was observed (Figure 1.1) despite consumption of 0.54 mM acetate, 0.03 mM methanol and 0.36 mM lactate, respectively (Figure 5.4). This suggests PCA could not act as an electron acceptor under the incubation conditions employed. Likewise, PCA concentrations were unaffected in cultures without electron acceptor or donor amendments or in sterile controls (Figure 5.1, Figure 1.1).

Figure 5.3 PCA fate in anaerobic sludge based microcosms with different electron donors. No PCA loss was observed in these cultures amended with electron donors. Data points are averages of triplicate cultures and error bars represent standard error of the mean of triplicate cultures.

133

Figure 5.4 Consumption of electron donors including acetate, methanol and lactate in anaerobic sludge-based microcosms. Decreases in concentrations of acetate, methanol and lactate were detected. Data points are averages of triplicate cultures and error bars represent the standard error of mean of triplicate cultures.

5.3.2 Enrichment of PCA-degrading bacteria with iron(III) as electron acceptor

Previously described microcosms with iron(III) as electron acceptor were used to inoculate (1% v/v) cultures to enrich PCA-degrading bacteria under anaerobic conditions. Transformation of PCA was observed in cultures amended with iron(III) throughout three rounds of enrichment. The rate of PCA degradation increased from 0.80 to 1.38 µM day-1 in the first two rounds of enrichment and finally to 2.43 µM day-1 in the third round of enrichment, while PCA degradation was not observed in cultures without iron(III) (Figure 5.5). Iron(III) was reduced into iron(II) at rates of 0.25, 0.54 and 0.20 µM day-1 in the three respective rounds of enrichment (Figure 5.6). No iron(II) was observed in abiotic cultures as sterile controls during enrichment, accompanying with negligible PCA degradation (Figure 5.5, Figure 5.6).

134

Figure 5.5 The anaerobic degradation of PCA in the presence () or absence () of iron(III) and in the sterile cultures () as control through three rounds of enrichment. The degradation rate of PCA were enhanced during enrichment, suggesting PCA- degrading bacteria was successfully enriched. Data points are averages of triplicate cultures and error bars represent standard error (0.39 ~ 24.34) of mean of the triplicate cultures.

Figure 5.6 The transformation of iron(III) to iron(II) in anaerobic microcosms through the three rounds of enrichment. Iron(III) was significantly reduced to iron(II) in the enriched PCA-degrading cultures, suggesting iron(III) was associated with PCA consumption. Data points are averages of triplicate cultures and error bars represent standard error (0.05 ~ 4.58) of mean of the triplicate cultures. 135

5.3.3 Community analysis of PCA-degrading enrichment cultures with iron(III) as electron acceptor

To gain insight into how the sludge community responded to enrichment in the presence of PCA and iron(III), microbial community structure was analysed by sequencing the V6-V8 region of bacterial and archaeal 16S rRNA genes from the enriched cultures with iron(III). Given that nearly all of the 16S rRNA genes sequenced were affiliated with Bacteria (99.8%), with 0.02% falling into Archaea, this study focused on community composition of Bacteria. As shown in Figure 5.7, addition of PCA had a pronounced effect on the bacterial community composition. Specifically, in microcosms from the first round of enrichment, bacterial 16S rRNA gene sequences were affiliated with eleven distinct phylogenetic lineages on the family level, with Pseudomonadaceae (38.2%) and Synergistaceae (16.3%) representing the most abundant families (Figure 5.7). Two more families, Clostridiaceae and Ruminococcaceae, were detected in microcosms from the second round of enrichment, making up to 19.1% and 7.4% of the bacterial community, respectively (Figure 5.7). Families Anaerolinaceae (5.6%), Caldilineaceae (2.1%), Commamonadaceae (6.1%) and Rhodocyclaceae (2.2%) were exclusively observed in cultures from the third round of enrichment (Figure 5.7). Significantly, the Enterobacteriaceae family was dramatically more abundant in the third round of enrichment, representing 68.0% of the bacterial community (Figure 5.7). Fourteen genera belonging to Enterobacteriaceae were detected in microcosms from the third round of enrichment but the Morganella genus was overwhelmingly dominant (93.3%) (Figure 5.7).

136

Figure 5.7 Bacterial community composition on level of family based on 16S rRNA gene sequences in anaerobic microcosms with iron(III) added throughout three rounds of enrichment. The Enterobacteriaceae family was dramatically enriched in the PCA- degrading cultures with iron(III) as electron acceptor. The right graph showed relative abundance of genera belonging to family Enterobacteriaceae in microcosms with iron(III) added from the third round of enrichment. Amongst the fourteen genera detected here, the Morganella genus was overwhelmingly dominant. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

5.3.4 Isolation of an anaerobic PCA-degrading bacterium

A dilution-to-extinction series (from 10-3 to 10-8) was performed on the microcosms amended with iron(III) from the third round of enrichment, and used to inoculate soft agar shakes containing PCA as the carbon and nitrogen source and electron donor, and iron(III) as an electron acceptor. Approximately 8 weeks after inoculation, small round colonies were observed in soft agar shakes and the solid medium turned green,

137

indicating iron(III) was reduced to iron(II). Colonies were transferred to minimal mineral medium amended with PCA and iron(III). After incubation for four weeks, approximately 110 M PCA was degraded and bacterial DNA was subsequently extracted for sequencing of partial 16S rRNA gene (192 bp). This revealed that the lineage selected here was pure and most closely related to a facultative anaerobe Morganella morganii strain M567 and thereby was designated Morganella morganii strain SL11 (the 11th colony picked from cultures inoculated with sludge sample).

5.3.5 Morphological and phylogenetic analysis of Morganella morganii SL11

Light microscopy of the pure culture showed that strain Morganella morganii SL11 is a rod-shaped, Gram-negative bacterium (Figure 5.8). Subsequently, a near full-length 16S rRNA gene sequence of Morganella morganii strain SL11 was analysed, which showed that it was affiliated with the Morganella genus belonging to family Enterobacteriaceae, order Enterobacteriales and class Gammaproteobacteria within the phylum Proteobacteria. The comparison of near full-length 16S rRNA gene sequence showed that Morganella morganii strain SL11 shared 100% identity with Morganella morganii strain M567 and 99% identity with Morganella morganii strain CU-BS1 and E8-9 (Figure 5.9).

Figure 5.8 Image of the isolated anaerobic PCA-degrading Morganella morganii strain SL11 observed after Gram staining (1,000x magnification). It showed that strain Morganella morganii SL11 is a rod-shaped, Gram-negative bacterium.

138

Figure 5.9 A neighbour-joining tree of near full-length 16S rRNA gene sequence of Morganella morganii SL11 (in bold), along with other Morganella strains and other phenazine-degrading bacteria. Numbers beside branches show bootstrap values using 1000 replicates > 50%. Scale bar shows 2% nucleotide base difference. Numbers in front of bacterial name are GenBank accession numbers from NCBI. (Given unavailable sequences of Mycobacterium septicum DKN1213, it was showed using the type strain M. septicum DSM44393.)

5.3.6 Growth and iron reduction linked PCA degradation by strain SL11

To confirm that growth of M. morganii strain SL11 is linked to the degradation of PCA and reduction of iron(III), fresh cultures were prepared and monitored for the decrease in PCA and iron(III) concentrations coupled with growth of strain SL11. Figure 5.10 shows the degradation of PCA by strain SL11 at the rate of 3.69 µM day-1 in cultures amended with PCA and iron(III), while no obvious consumption of PCA was observed in cultures amended with PCA only or in sterile cultures. The growth of strain SL11 was concomitant with the anaerobic transformation of PCA with iron(III) supplied as electron acceptor. Using fluorescence microscopy, cell numbers of strain SL11 increased from 6.23 × 105 to 1.79 × 107 cells/mL in cultures amended with PCA and iron(III) (P < 0.05), while the corresponding cell numbers of strain SL11 decreased from 7.54 × 105 to 7.17 × 103 cells/mL in cultures amended with PCA only (Figure 5.10). Cells were not observed in abiotic cultures during the incubation (Figure 5.10).

139

Figure 5.10 Degradation of PCA (solid line) and cell yield (dash line) in pure M. morganii strain SL11 cultures fed with PCA in presence () or absence () of iron(III) and in abiotic cultures as control (). Growth of strain SL11 was linked with metabolic PCA degradation. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

Accompanying PCA degradation and growth, the concentration of iron(III) decreased from 31.71 to 3.87 mM, and the concentration of iron(II) increased from 0.84 to 26.08 mM in cultures inoculated with strain SL11 and amended with PCA and iron(III) after incubation for 50 days (Figure 5.11). No evidence of iron(III) reduction was observed in abiotic cultures amended with PCA and iron(III) (P < 0.05), and PCA consumption was not observed (Figure 5.10, Figure 5.11).

140

Figure 5.11 Profiles of iron(III) and iron(II) in anaerobic cultures inoculated with Morganella morganii strain SL11 and abiotic cultures as sterile control. Iron(III) was reduced to iron(II) in the cultures inoculated with PCA-degrading strain SL11. Data points are averages of triplicate cultures and error bars represent standard error of mean of the triplicate cultures.

Furthermore, correlation analysis revealed that the amount of PCA degraded correlated with the decreasing amount of iron(III) (r2 = 0.90, n = 7, P < 0.05) (Figure 5.12A) and the increasing amount of iron(II) (r2 = 0.95, n = 7, P < 0.05) (Figure 5.12B). Moreover, the cell yield of strain SL11 before stationary phase also significantly correlated with the amount of PCA transformed (r2 = 0.96, n = 5, P < 0.05) (Figure 5.12C).

141

A. B.

Figure 5.12 Linear correlation between the amount of PCA degraded and the decreasing amount of iron(III) (A), the increasing amount of iron(II) (B) and the increasing cell yield of strain SL11 before stationary phase (C). Degradation of PCA significantly correlated with iron(III) reduction to iron(II) and the cell yield of strain SL11 in the pure cultures. The solid line represents the best fit based on linear regression of average of triplicate data points.

142

5.4 Discussion

Successful attempts have been made to demonstrate the degradation of phenazine compounds by cultivating pure organisms in microcosms under aerobic conditions (Yang et al., 2007, Costa et al., 2015, Costa et al., 2017, Costa et al., 2018). However, nothing is known about the anaerobic degradation of phenazines. This chapter describes the enrichment and isolation of a PCA-degrading strain SL11 which anaerobically utilizes PCA as its organic carbon, nitrogen and energy source under the conditions with iron(III) as electron acceptor. Throughout the enrichment period, community composition of bacteria changed significantly in response to the addition of PCA and iron(III), resulting in significant enrichment of family Enterobacteriaceae which mainly encompassed the facultatively anaerobic genus Morganella. From the enriched cultures, PCA-degrading strain SL11 was isolated, which is affiliated with the Morganella genus based on the near full-length 16S rRNA gene sequence and is thus designated Morganella morganii strain SL11.

Enterobacteriaceae can survive in different niches, because of their rapid generation times and ability to exchange genetic information (Grbic-Galic, 1986). Isolates belonging to this lineage have been harnessed in areas of bioremediation, and degradation of herbicides and other toxic products including compounds with ring structures (Pileggi et al., 2012). In this current work, Enterobacteriaceae mainly encompassing the facultatively anaerobic Morganella genus was significantly enriched after three rounds of enrichment. It is reasonable to suggest that this family especially Morganella is capable of anaerobic PCA degradation using iron(III) as terminal electron acceptor. Strain SL11 belonging to facultatively anaerobic species Morganella morganii utilized PCA as its carbon, nitrogen and energy source under anaerobic conditions with iron(III) as the terminal electron acceptor, representing the first known anaerobic phenazine-degrading bacterial isolate. M. morganii is a Gram-negative, facultative anaerobic rod species widely found in the open environments such as soil and lakes and in the intestinal tracts of mammals and humans (Eleuterio & Batista, 2010, Hamada et al., 2015). M. morganii strains are able to degrade complex organic matter, for example, nitramine explosives 2,4,6-trinitrotoluene (TNT) (Kitts et al., 2000), carbaryl (Hamada et al., 2015), hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) (Kitts et al., 2000) and microcystin-LR (Eleuterio & Batista, 2010). Despite the ring structures shared in these compounds, no ring cleavage was thus far reported. In this study, however, strain SL11

143

must break down aromatic and nitrogen-containing rings in phenazines. First, the carboxylate moiety of PCA alone was insufficient for the level of growth of strain SL11 observed. As reported by Costa et al. (2015), the carboxylic group of 120 µM PCA alone cannot support cell yield of Mycobacterium fortuitum CT6 from OD500 = 0.11 to 6 OD500 = 0.42. According to the formula OD550 = 1 is equivalent to 2.8 × 10 - 107 Mycobacterium cells/mL (Chui et al., 2004), the cell growth of strain CT6 is 8.68 × 105-106 cells/mL. The carboxylic group of 1 µM PCA is insufficient to support biomass yields of 7.2 × 103 - 104 cells/mL, suggesting that ring cleavage must occurs in PCA degradation by strain CT6 (Costa et al., 2015). In this study, consumption of 184 µM PCA resulted in growth of strain SL11 from 6.23 × 105 to 1.79 × 107 cells/mL, meaning 1 µM PCA supports bacterial increase of 9.39 × 104 cells/mL, which is slightly higher than the case of strain CT6. Therefore, the carbon source bound within the aromatic rings in PCA must be liberated via cleavage to support growth of strain SL11 as is case for PCA degradation by strain CT6. Second, the fact that strain SL11 can utilize PCA as the sole nitrogen source, indicates that cleavage of nitrogen-containing ring must occur to liberate nitrogen from PCA. Attempts to PCR amplify homolog(s) of the MFORT_16269 gene encoding phenylpropionate dioxygenase (Costa et al., 2015) using DNA from Morganella morganii SL11 were unsuccessful (data not shown). This enzyme was found in only one, i.e. M. morganii GM1DA1, of sixteen Morganella genomes available in the JGI database. It seems logical that ring cleavages in anaerobic PCA degradation are catalysed by enzymes different from that in aerobic PCA degradation. In contrast, ubiD and ubiX genes responsible for decarboxylation of aerobic PCA degradation (Costa et al., 2018) were found in all Morganella morganii genomes except M. morganii SA36, suggesting UbiD and UbiX enzymes might be involved in anaerobic PCA degradation as is the case for aerobic PCA transformation. Further studies are required to elucidate the underlying metabolic pathway of PCA degradation under anaerobic conditions.

Iron(III) is an abundant electron acceptor in subsurface environments, and consequently has the potential to play a crucial role in the anaerobic oxidation of organic compounds (Lovley & Coates, 2000). For example, dissimilatory iron-reducing bacteria were commonly found in polycyclic aromatic hydrocarbons (PAHs)- contaminated subsurface environments and supported anaerobic transformation of PAHs which have structural similarities to phenazines in terms of ring structure and

144

redox active functional groups (Farkas et al., 2017). In this study, iron(III) reduction occurred in parallel with PCA degradation by strain SL11, suggesting the role of iron(III) reduction in anaerobic PCA degradation. Around 25.2 mM iron(II) was produced accompanying with degradation of 184 µM PCA, indicating 1 µM of PCA transformation resulted in 137 µM of iron(III) reduction. Assuming that PCA was completely transformed into CO2, 50 electrons can be donated by 1 molecule of PCA. Therefore, a maximum of 9.2 mM electron can be served by 184 µM PCA degraded in this study. Because one electron is required for iron(III) reduction to iron(II), 25.2 mM electron was needed to produce 25.2 mM iron(II) here, which is higher than the available electrons served by PCA degradation in M. morganii SL11 cultures. The underlying explanation for the gap in the electron balance here is unknown and required further experiments to be conducted.

Strain SL11 is the first known iron reducing Morganella lineage. Genomic data from M. morganii strain AR0057 (NCBI accession number: CP027177) shows a ferric reductase like transmembrane component family protein (NCBI accession number: AVK38034) which is responsible for catalysing the reductive uptake of iron(III)-salts and iron(III) bound to catecholate or hydroxamate siderophores. A ferric iron reductase protein FhuF (locus tag Ga0038529_101265) involved in iron transport is also predicted by the genome of M. morganii strain GM1DA1. To identify genes relating to iron(III) reduction, it will be interesting to sequence and analyse the genome of M. morganii strain SL11.

In conclusion, this chapter reports on the anaerobic enrichment and isolation of a strain Morganella morganii SL11 that is capable of degrading PCA with iron(III) as terminal electron acceptor. To our knowledge, this is the first pure culture described capable of PCA degradation under anaerobic conditions. Results of this study may lead to a better understanding of the fate of PCA in anaerobic zones.

145

References

Balch WE, Fox GE, Magrum LJ, Woese CR & Wolfe RS (1979) Methanogens: reevaluation of a unique biological group. Microbiological Reviews 43: 260-296.

Beckmann S, Welte C, Li X, et al. (2016) Novel phenazine crystals enable direct electron transfer to methanogens in anaerobic digestion by redox potential modulation. Energy and Environmental Science 9: 644-655.

Cline JD (1969) Spectrophotometric determination of hydrogen sulfide in natural waters. Limnology and Oceanography 14: 454–458.

Costa KC, Glasser N, Conway SJ & Newman DK (2017) Pyocyanin degradation by a tautomerizing demethylase inhibits Pseudomonas aeruginosa biofilms. Science 355: 170- 173.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Dietrich LE, Okegbe C, Price-Whelan A, Sakhtah H, Hunter RC & Newman DK (2013) Bacterial community morphogenesis is intimately linked to the intracellular redox state. Journal of Bacteriology 195: 1371-1380.

Eleuterio L & Batista JR (2010) Biodegradation studies and sequencing of microcystin- LR degrading bacteria isolated from a drinking water biofilter and a fresh water lake. Toxicon 55: 1434-1442.

Farkas M, Szoboszlay S, Benedek T, Revesz F, Veres PG, Kriszt B & Tancsics A (2017) Enrichment of dissimilatory Fe(III)-reducing bacteria from groundwater of the Siklos BTEX-contaminated site (Hungary). Folia Microbiologica 62: 63-71.

146

Glasser NR, Kern SE & Newman DK (2014) Phenazine redox cycling enhances anaerobic survival in Pseudomonas aeruginosa by facilitating generation of ATP and a proton-motive force. Molecular Microbiology 92: 399-412.

Grbic-Galic D (1986) O-Demethylation, dehydroxylation, ring-reduction and cleavage of aromatic substrates by Enterobacteriaceae under anaerobic conditions. Journal of Applied Bcteriology 61: 491-497.

Guo J, Cole JR, Zhang Q, Brown CT & Tiedje JM (2015) Microbial community analysis with ribosomal gene fragments from shotgun metagenomes. Applied and Environmental Microbiology 82: 157-166.

Hamada M, Matar A & Bashir A (2015) Carbaryl degradation by bacterial isolates from a soil ecosystem of the Gaza Strip. Brazilian Journal of Microbiology 46: 1087-1091.

Hernandez ME, Kappler A & Newman DK (2004) Phenazines and other redox-active antibiotics promote microbial mineral reduction. Applied and Environmental Microbiology 70: 921-928.

Kitts CL, Green CE, Otley RA, Alvarez MA & Unkefer PJ (2000) Type I nitroreductases in soil enterobacteria reduce TNT (2,4,6-trinitrotoluene) and RDX (hexahydro-1,3,5- trinitro-1,3,5-triazine). Canadian Journal of Microbiology 46: 278-282.

Lee X-P, Kumazawa T, Kondo K, Sato K & Suzuki O (1999) Analysis of methanol or formic acid in body fluids by headspace solid-phase microextraction and capillary gas chromatography. Journal of Chromatography B: Biomedical Sciences and Applications 734: 155-162.

Löffler FE, Sanford RA & Ritalahti KM (2005) Enrichment, cultivation, and detection of reductively dechlorinating bacteria. Methods in Enzymology, 397: 77-111.

Lovley DR & Coates JD (2000) Novel forms of anaerobic respiration of environmental relevance. Current opinion in Microbiology 3: 252-256.

Lundquist P, Mårtensson J, Sörbo B & Ohman S (1980) Turbidimetry of inorganic sulfate, ester sulfate and total sulfurin urine. Clinical Chemistry 26: 1178-1181.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

147

Miranda KM, Espey MG & Wink DA (2001) A rapid, simple spectrophotometric method for simultaneous detection of nitrate and nitrite. Nitric Oxide 5: 62-71.

Pileggi M, Pileggi SA, Olchanheski LR, da Silva PA, Munoz Gonzalez AM, Koskinen WC, Barber B & Sadowsky MJ (2012) Isolation of mesotrione-degrading bacteria from aquatic environments in Brazil. Chemosphere 86: 1127-1132.

Si J, Li X-C & Cui B-K (2013) Decolorization of heterocycle dye neutral red by white- rot fungus Perenniporia subacida. Desalination and Water Treatment 52: 5594-5604.

Sullivan NL, Tzeranis DS, Wang Y, So PT & Newman D (2011) Quantifying the dynamics of bacterial secondary metabolites by spectral multiphoton microscopy. ACS Chemical Biology 6: 893-899.

Tschech A & Pfennig N (1984) Growth yield increase linked to caffeate reduction in Acetobacterium woodii. Archives of Microbiology 137: 163-167.

Viollier E, Inglett PW, Hunter K, Roychoudhury AN & Cappellen PV (2000) The ferrozine method revisited: Fe(II)/Fe(III) determination in natural waters. Applied Geochemistry 15: 785-790.

Wang Y, Kern SE & Newman DK (2010) Endogenous phenazine antibiotics promote anaerobic survival of Pseudomonas aeruginosa via extracellular electron transfer. Journal of Bacteriology 192: 365-369.

Widdel F & Pfennig N (1981) Studies on dissimilatory sulfate-reducing bacteria that decompose fatty acids. Archives of Microbiology 129: 395-400.

148

Chapter 6 Concluding remarks and future perspectives

6.1 Summary of findings

Phenazine compounds as active metabolites are secreted by phylogenetically diverse bacteria including genera of medical and ecological significance such as Pseudomonas and Streptomyces (Umezawa et al., 1950, Turner & Messenger, 1986, Budzikiewicz, 1993, Laursen & Nielsen, 2004). Due to their broad-spectrum antibiotic and redox activities, phenazine compounds play important role in ecological fitness of their producers, interactions with cohabiting microorganisms and plants and human health under aerobic and anaerobic conditions (Ran et al., 2003, Lau et al., 2004, Mavrodi et al., 2006, Caldwell et al., 2009). For example, addition of pyocyanin caused decreases in both the degradation rate of polycyclic aromatic hydrocarbons in an aerobic oil- degrading culture and the diversity of the microbial community (Norman et al., 2004). Phenazines including PCA, pyocyanin and 1-hydroxyphenazine enhanced anaerobic survival of their producer Pseudomonas aeruginosa in planktonic cultures (Wang et al., 2010; Sullivan et al., 2011) and on agar plates (Dietrich et al., 2013). Also, widespread application and improper disposal practices of phenazine dyes such as neutral red have resulted in pollution in aquatic environments (Gong et al., 2009). Despite the ecological and environmental importance, the fate, especially degradation, of phenazines in the environment is poorly understood.

Prior to this study, seven phenazines-degrading bacteria had been isolated under aerobic conditions, including two Alphaproteobacteria (Yang et al., 2007, Zhao et al., 2017) and five Actinobacteria (Costa et al., 2015, Costa et al., 2018). There have been recent advances in the characterization of genes and enzymes involved in aerobic PCA degradation. Using comparative genomic analyses, knockout mutations and heterologous expression, the UbiD enzyme encoded by the MFORT_16229 gene and phenylpropionate dioxygenase encoded by the MFORT_16269 gene were shown to be responsible for PCA degradation in pure M. fortuitum ATCC 6841 cultures (Costa et al., 2015, Costa et al., 2018). Given that UbiX synthesizes the cofactor prenylated flavin mononucleotide (prFMN) in the active site of UbiD, UbiX was also proposed to be

149

involved in PCA degradation (Costa et al., 2018). The homologs of genes encoding these three enzymes are found in all known PCA-degrading isolates (Ma et al., 2012, Costa et al., 2015, Costa et al., 2018). Pyocyanin demethylase encoded by the MFORT_14352 gene was proven responsible for pyocyanin degradation via demethylation in M. fortuitum ATCC6841 (Costa et al., 2017).

Until now, no study has explored anaerobic degradation of phenazines and more phenazines-degrading bacteria belonging to phylogenetically diverse taxa remain to be isolated under aerobic and anaerobic conditions. Additionally, phenazine is the only intermediate of PCA degradation identified to date, and this lack of other intermediates and products has limited characterisation of the metabolism of phenazines.

The objective of this study was to enrich, isolate and characterize aerobic and anaerobic PCA-degrading cultures. In Chapter 2 of this study, the eighth PCA- degrading strain designated Rhodanobacter sp. PCA2 was isolated from soil. It also utilized phenazine, pyocyanin and 1-hydroxyphenazine as sole carbon, nitrogen and energy sources. In order to gain a deeper understanding of the genetics and biochemistry of PCA metabolism, the genome of this new isolate Rhodanobacter sp. PCA2 was sequenced, annotated and analyzed as described in Chapter 3. Transcript analysis of the homolog of MFORT_16269 gene responsible for PCA degradation was also performed. In Chapter 4, mass spectrometry, inducibility experiment and proteomic analysis were employed to identify and characterize intermediates, products and enzymes responsible for the PCA-degrading capabilities of strain PCA2, resulting in a proposed enzymatic pathway of PCA degradation under aerobic conditions. In Chapter 5, the enrichment, isolation and characterization of the first anaerobic PCA-degrading bacteria designated Morganella morganii SL11 was described. The major findings in this study are detailed below:

1) In this study, a new and only known Gammaproteobacterial isolate from soil designated Rhodanobacter sp. PCA2 was shown to utilize PCA, phenazine and pyocyanin as the sole carbon, nitrogen and energy sources to support bacterial growth. 2) Strain PCA2 is the first isolate capable of degrading 1-hydroxyphenazine. 3) Strain PCA2 displayed the highest degradation rate of PCA at the concentration of 500 mg/L (2.23 mM).

150

4) The homolog of the MFORT_16269 gene in strain PCA2 shared 95% sequence identity with the corresponding gene in M. fortuitum ATCC6841. 5) Transcriptomic analysis found the homolog of the MFORT_16269 gene was significantly expressed during PCA degradation by strain PCA2. The MFORT_16269 gene homologue copy number also increased 10.6-fold more than the corresponding cell yields throughout PCA degradation by strain PCA2. 6) Genomic analysis indicated that only ubiD was found in the chromosomal DNA of strain PCA2, whereas the homologs of MFORT_16269 and ubiX were not observed. Gene homologues encoding pyocyanin demethylase responsible for pyocyanin transformation were not found in the chromosome of strain PCA2. 7) Unlike all known PCA-degrading Actinobacteria and Alphaproteobacteria, genes involved in the metabolism of PCA were identified on the plasmid instead of the chromosome in Rhodanobacter sp. PCA2. It was hypothesized that strain PCA2 obtained the ability to degrade PCA via horizontal gene transfer. 8) Mass spectrometry data identified phenazine, (4Z)-2-hydroxy-5-{[(1Z)-6- (hydroxyamino)cyclohexa -2,4-dien-1-ylidene]carbamoyl} penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine as intermediates of PCA degradation via decarboxylation and ring cleavages. 9) PCA utilization was inducible in strain PCA2. 10) Using proteomic analysis, it was revealed that UbiD, UbiX and phenylpropionate dioxygenase previously known for PCA degradation were only detected in PCA2 cultures when exposed to PCA. Likewise, biphenyl-2,3-diol 1,2-dioxygenase, amidohydrolase and nitroreductase were expressed extensively in strain PCA2 on exposure to PCA, suggesting their potential involvement in PCA degradation. 11) By combining genomic and proteomic data with mass spectrometry, an enzymatic pathway of PCA degradation by strain PCA2 was proposed. 12) The first anaerobic PCA-degrading bacterium designated Morganella morganii SL11 was enriched and isolated from sewage sludge. 13) Preliminary characterization of strain SL11 indicated that iron(III) as electron acceptor was necessary for PCA degradation under anaerobic condition.

151

6.2 Significance and future functional characterizations of Rhodanobacter sp. PCA2

The isolation of Rhodanobacter sp. PCA2 has enabled an understanding of its metabolism and characteristics. The molecular and physiological characterization improves our knowledge on PCA-degrading bacteria belonging to the Gammaproteobacteria. Besides degradation of PCA, phenazine and pyocyanin, strain PCA2 is also characterized as capable of degrading 1-hydroxyphenazine. An interesting avenue for future research, therefore, is to determine whether the metabolisms of 1- hydroxyphenazine and other phenazine compounds converge on dihydroxyphenazine as a catabolic intermediate as hypothesized by Costa et al. (2018) and this study. LC-MS- based metabolic analysis will be able to address this hypothesis. Despite the fact that intermediates of PCA degradation were identified, the final product(s) are still unknown. Nuclear magnetic resonance (NMR) spectroscopy could be performed to further verify the intermediates identified by LC-MS analyses in this study and explore the final products of PCA degradation. Additionally, given that different degradation rates are observed in phenazine compounds with different substituted groups, it would be interesting to explore the effects of substituted groups on transformation and the underlying mechanisms.

6.3 Significance and future molecular characterizations of Rhodanobacter sp. PCA2

In this study, the chromosome of strain PCA2 was sequenced, annotated and analysed. Its addition into the limited genome pool of phenazine degrading bacteria will enhance our understanding of the evolution of differences and similarities of various PCA-degrading strains. Currently, there are six genomes of phenazines degraders available publicly, including S. wittichii DP58 (Yang et al., 2007), S. yanoikuyae B1 (Zhao et al., 2017), M. fortuitum CT6 and ATCC6841 (Costa et al., 2015), Rhodococcus sp. JVH1 (Costa et al., 2015) and Nocardia sp. LAM0056 (Costa et al., 2018).

Genomic and PCR analyses in this study suggest that genes responsible for PCA degradation, i.e. ubiD and a homolog of the MFORT_16269, are located on a plasmid in strain PCA2 and are possibly derived from Mycobacterium spp. However, the molecular foundation facilitating strain PCA2 to degrade pyocyanin is not understood. Pyocyanin-

152

degrading gene was not observed in the assembled chromosomal genome of strain PCA2, which might be explained as the case of PCA degradation in strain PCA2. Sequencing plasmid DNA in strain PCA2 will undoubtedly help to verify this hypothesis. Given the lack of a reference sequence for plasmids in the Rhodanobacter genus, PacBio long-read sequencing is proposed to obtain the plasmid sequence in strain PCA2. The ubiD gene transforming PCA into phenazine was not as well studied as the homolog of MFORT_16269 in this study, because only recently it was discovered to be responsible for the first step of PCA degradation (Costa et al., 2018). Transcriptomic analysis of this gene is suggested to identify its responsiveness to PCA in strain PCA2.

Proteomic analysis coupled with LC-MS-based metabolic studies revealed that phenazine was one of the intermediates of PCA degradation, which is consistent with research reported by Costa et al. (2018). For the first time, two additional intermediates including (4Z)-2-hydroxy-5-{[(1Z)-6-(hydroxyamino)cyclohexa -2,4-dien-1- ylidene]carbamoyl} penta-2,4-dienoic acid (HCCPD) and phenylhydroxylamine were reported in this study. They are produced from phenazine by ring cleavages which are potentially catalysed by phenylpropionate dioxygenase, biphenyl-2,3-diol 1,2- dioxygenase and amidohydrolase. Determining functions of these enzymes in PCA degradation remains a promising research direction, ideally through gene knock-out mutation and heterologous expression.

6.4 Significance and future functional characterizations of Morganella morganii SL11

The isolation of the first anaerobic PCA-degrading bacterium Morganella morganii strain SL11 provides insights into PCA transformation under anaerobic condition. It was found that iron(III) reduction was linked with anaerobic PCA degradation. With this novel isolate, it will be useful for filling in knowledge gaps regarding anaerobic transformation of phenazines and may serve as a model bacterium for other anaerobic phenazines degraders. The ability to utilize other phenazine compounds is worth exploring in strain SL11. The lack of PCA-degrading isolates has impeded studies on metabolism of PCA under anaerobic conditions. The use of 13C-labeled PCA (if

153

available in the future) can be employed to identify functional bacteria in PCA- degrading communities, which will be useful for isolation of PCA degraders.

Several intermediates and products of anaerobic PCA transformation remain unknown, but can be addressed by nuclear magnetic resonance (NMR) spectroscopy and LC-MS analyses. Unlike aerobic PCA degraders, the homolog of MFORT_16269 was not found in the genome of strain SL11. To gain a better understanding of anaerobic PCA degradation, more functional and molecular characterization is needed. Genomic analysis enables a deeper understanding of physiology and biochemistry of PCA degraders and metabolism of PCA, however, there is no genome available for anaerobic PCA degrading bacteria. An approach such as PacBio or Illumina sequencing is proposed to fill this gap. By combining transcriptomic and genomic analyses, potential genes involved in anaerobic PCA degradation are expected to be proposed. These can be further identified using mutants and heterologous expression.

6.5 Concluding remarks

This research thesis sheds light on aerobic degradation of PCA and provides insights into anaerobic transformation of PCA for the first time. The isolation of Rhodanobacter sp. PCA2 that utilizes PCA, phenazine, pyocyanin and 1-hydroxyphenazine as a sole carbon, nitrogen and energy source, enhances our knowledge on phylogenetic diversity and metabolic versatility of bacteria capable of degrading PCA. A homolog of the MFORT_16269 gene active throughout transformation of PCA was identified on a plasmid in strain PCA2. The genomic and proteomic analyses of strain PCA2 deepens our understanding of genes and proteins involved in PCA degradation. Metabolic analysis based on LC-MS made a significant contribution to knowledge on intermediates and products of PCA biodegradation. Additionally, the isolation of Morganella morganii SL11 represents a new opportunity to explore PCA transformation under anaerobic condition. In conclusion, this study provides advances in degradation of phenazines especially PCA under aerobic and anaerobic conditions.

154

References

Budzikiewicz H (1993) Secondary metabolites from fluorescent Pseudomonads. FEMS Microbiology Reviews 10: 209-228.

Costa KC, Glasser N, Conway SJ & Newman DK (2017) Pyocyanin degradation by a tautomerizing demethylase inhibits Pseudomonas aeruginosa biofilms. Science 355: 170- 173.

Costa KC, Moskatel LS, Meirelles LA & Newman DK (2018) PhdA catalyzes the first step of phenazine-1-carboxylic acid degradation in Mycobacterium fortuitum. Journal of Bacteriology 200: e00763-17.

Costa KC, Bergkessel M, Saunders S, Korlach J & Newman DK (2015) Enzymatic degradation of phenazines can generate energy and protect sensitive organisms from toxicity. MBio 6: e01520-01515.

Lau GW, Ran H, Kong F, Hassett DJ & Mavrodi D (2004) Pseudomonas aeruginosa pyocyanin is critical for lung infection in mice. Infection and Immunity 72: 4275-4278.

Laursen JB & Nielsen J (2004) Phenazine natural products: biosynthesis, synthetic analogues, and biological activity. Chemical Reviews 104: 1663-1685.

Mavrodi DV, Blankenfeldt W & Thomashow LS (2006) Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annual Review of Phytopathology 44: 417-445.

155

Norman RS, Moeller P, McDonald TJ & Morris PJ (2004) Effect of pyocyanin on a crude- oil-degrading microbial community. Applied and Environmental Microbiology 70: 4004- 4011.

Ran H, Hassett D & Lau G (2003) Human targets of Pseudomonas aeruginosa pyocyanin. Proceedings of the National Academy of Sciences of the United States of America 100: 14315-14320.

Turner JM & Messenger AJ (1986) Occurrence, biochemistry and physiology of phenazine pigment production. Advances in Microbial Physiology 27: 211-275.

Umezawa H, Hayano S, Maeda K, Ogata Y & Okami Y (1950) On a new antibiotic, griseolutein, produced by Streptomyces. Japanese Medical Journal 3: 111-117.

156

Appendix

8.1 DNA extraction

1. Lysis buffer Component Concentration Tris-HCl (pH 3.5) 0.10 M NaCl 0.30 M Ethylenediaminetetraacetic acid (EDTA) 0.02 M Sodium dodecyl sulphate (SDS) 2% w/v 2-Mercaptoethanol 2% v/v sterilize by filtration. 2. Ammonium acetate buffer Component Concentration Ammonium acetate 7.5 M Adjust pH of buffer solution to 8.0 and sterilize by filtration.

Procedures Using a 2 mL microcentrifuge tube with screw cap, centrifuge 2 mL of culture at 8,000 g for 20 min. Then, remove 1.6 mL of supernatant and add lysis beads (Matrix A from FastPrep) and 400 µL of lysis buffer before mix thoroughly. Lyse cells using a tissue lyser machine (Mo-Bio) at 30 Hz for 5 min. after that, add equivolume (~ 800 µL) of cold (4-8°C) phenol:chloroform:isomylalcohol (25:24:1, Sigma) in a fumehood. Mix tube by inversion and centrifuge tubes at 16,000 g for 1 min at room temperature. Transfer around 500 µL of upper aqueous phase into a clean 2 mL microcentrifuge tube and add equivolume (500 µL) of 7.5 M ammonium acetate buffer. Precipitate DNA with 0.6 volumes of ice-cold isopropanol and leave it overnight at room temperature. Then, centrifuge at 16,000 g for 20 min. Discard supernatant immediately and add 300 µL of 80% ethanol to remove remaining phenol:chloroform. Centrifuge the tube at 16,000 g for 10 min, then discard the supernatant immediately and carefully with pipette and leave the DNA extract for air dry for 20 min in biosafety cabinet. Dissolve dried DNA extract with 30 µL of Molecular Biology Water (Invitrogen).

157

8.2 Denaturing gradient gel electrophoresis (DGGE) buffer

1. Gel solutions

Stacking gel Low denaturant High denaturant Components solution (0%) gel solution (45%) gel solution (60%)

40% Acrylamide/Bis solution (37.5:1) 25 mL 25 mL 25 mL (Bio-Rad) 50x TAE buffer 2 mL 2 mL 2 mL (Invitrogen) Deionised formamide N/A 18 mL 24 mL (Bio-Rad) Urea N/A 18.9 g 25.2 g ddH2O to 100 mL to 100 mL to 100 mL 2. Ammonium persulphate buffer Component Concentration Ammonium persulphate 10% w/v

ddH2O 1.0 mL

8.3 Agarose gel electrophoresis

Prepare 1% (w/v) agarose and pour into a gel mould (Bio-Rad). Insert comb carefully to avoid any bubbles and dry the gel for 20 – 30 min at room temperature. Remove comb and place gel into a horizontal electrophoresis cell and load samples (mixture of 5 µL of DNA or PCR product and 1 µL of 6x loading dye) into lanes. Run electrophoresis at 100 V for 25 min.

8.4 Sanger sequencing protocol

1. Sequencing reaction Components Volume (µL) Big Dye terminator v3.1 1 5x Sequencing buffer 3.5 Primer (10 µM) 0.32 PCR product (20 – 50 ng/µL) 1 Molecular Biology water up to 20 µL

158

2. PCR conditions 30 cycles of 96°C for 10 s, 50°C for 5 s and 60°C for 4 min. 3. Purification using ethanol/EDTA Transfer all product of sequencing reaction into a clear 1.5 mL sterile microcentrifuge tube. Add 5 µL of 125 mM EDTA and 60 µL of 100% ethanol then vortex briefly. Leave the tube to precipitate at room temperature for more than 30 min. After that, centrifuge the tube at 15,000 g for 20 min. Discard supernatant carefully with pipette. Add 250 µL of 70% ethanol and vortex briefly. Centrifuge the tube at 16,000 g for 10 min at 4°C. discard supernatant carefully without disturb the precipitate and dry the extract in a heat block set at 90°C for 1 min.

8.5 Codes for genomic analysis

1. Quality control (1) FastQC #!/bin/bash #PBS -N NR-SL12 #PBS -l nodes=1:ppn=1 #PBS -l vmem=5gb #PBS -l walltime=1:00:00 module load fastx/0.0.14 module load fastqc/0.10.1 cd /srv/scratch/z5010401/PCA2 fastx_clipper -a CTGTCTCTTATACACATCT -l 25 -i SL12_R1.fastq -o PCA2_R1_trimmed.fastq fastx_clipper -a CTGTCTCTTATACACATCT -l 25 -i SL12_R2.fastq -o PCA2_R2_trimmed.fastq fastq_quality_filter -q 20 -p 70 -i SL12_R1_trimmed.fastq -o PCA2_R1_filtered.fastq fastq_quality_filter -q 20 -p 70 -i SL12_R2_trimmed.fastq -o PCA2_R2_filtered.fastq (2) SolexaQA ./SolexaQA++ lengthsort “PCA2_R1_filtered” “PCA2_R2_filtered” –d “PCA2_R1_QCed” “PCA2_R2_QCed” -c 2. Assembly #!/bin/bash #PBS -N assembly #PBS -l nodes=1:ppn=1 #PBS -l vmem=10gb

159

#PBS -l walltime=10:00:00 module load spades/3.7.0 module load python/2.7.12 module load perl/5.20.1 module load quast/4.3 cd /srv/scratch/z5010401/PCA2 spades.py --pe1-1 PCA2_R1_QCed.fastq --pe1-2 PCA2_R2_QCed.fastq -o PCA2_assemblied python filter_coverage.py 5 PCA2_assemblied/scaffolds.fasta python filter_contigs.py 700 PCA2_assemblied/scaffolds.filter5cov.fasta quast.py PCA2_assemblied/PCA2_scaffolds.fasta 3. Order scaffolds as reference #!/bin/bash #PBS -N move_contigs #PBS -l nodes=1:ppn=1 #PBS -l vmem=10gb #PBS -l walltime=5:00:00 module load java/8u91 java -Xmx500m -cp /srv/scratch/z5010401/mauve_snapshot_2015/Mauve.jar org.gel.mauve.contigs.ContigOrderer -output /srv/scratch/z5010401/PCA2/move_dir -ref /srv/scratch/z5010401/PCA2/R_denitrificans.gbk -draft /srv/scratch/z5010401/PCA2/PCA2_assembled/PCA2_scaffolds.fasta

8.6 Gas chromatography (Shimadzu GC-2010 Plus) with flame ionization detector (GC-FID)

Quantification of methanol and ethyl acetate:

Inlet temperature: 250°C

Split ratio: 1:10

Flow rate: Helium; 2 mL/min

GC oven: 30°C isothermal for 5 min

160