THE REGULATION OF TOXIN SYNTHESIS IN MICROCYSTIS AERUGINOSA
A DISSERTATION SUBMITTED
BY
Ralitza Dimitrova Alexova
B. BIOTECHNOLOGY HONS. (UTS)
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY (Ph.D.) AT THE SCHOOL OF BIOTECHNOLOGY AND BIOMOLECULAR SCIENCES THE UNIVERSITY OF NEW SOUTH WALES SYDNEY, AUSTRALIA
2010
THE ROLE AND REGULATION OF TOXIN SYNTHESIS IN MICROCYSTIS AERUGINOSA
A DISSERTATION SUBMITTED
BY
Ralitza Dimitrova Alexova
B. BIOTECHNOLOGY HONS. (UTS)
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR
THE AWARD OF
DOCTOR OF PHILOSOPHY (Ph.D.)
AT THE
SCHOOL OF BIOTECHNOLOGY AND BIOMOLECULAR SCIENCES
THE UNIVERSITY OF NEW SOUTH WALES
SYDNEY, AUSTRALIA
SUPERVISORS
PROF. BRETT A. NEILAN SUPERVISOR SCHOOL OF BIOTECHNOLOGY AND BIOMOECULAR SCIENCES THE UNIVERSITY OF NEW SOUTH WALES SYDNEY, AUSTRALIA
DR. BELINDA C. FERRARI CO-SUPERVISOR SCHOOL OF BIOTECHNOLOGY AND BIOMOLECULAR SCIENCES THE UNIVERSITY OF NEW SOUTH WALES SYDNEY, AUSTRALIA
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 …………………………
Date ……………………………
i
ii Abstract
Toxin synthesis in bloom-forming cyanobacterial communities is a dynamic process that responds to changes in the environment, such as light, nutrient availability and oxidative stress. Our limited understanding of the factors that drive bloom formation, and the increased incidence of cyanobacterial blooms worldwide present a significant challenge to the water industry. The production of the hepatotoxic non-ribosomal peptide microcystin has been proposed to be advantageous to toxic cells and appears to be regulated by nitrogen- and iron-stress responsive proteins at the transcriptional level. In this study, protein expression in the unicellular microcystin-producing Microcystis aeruginosa species was investigated in nutrient-replete and nutrient-stress conditions.
The proteome of M. aeruginosa was highly strain-specific, with protein expression changes accumulating during prolonged growth in laboratory culture. Microcystin synthesis was found to be a dynamic trait in toxigenic, but inactive M. aeruginosa strains, and the reversal of one such previously non-toxic strain to toxicity was reported. Transcription of the mcyA gene involved in microcystin synthesis was also identified in both toxic and non-toxic strains and suggests that regulation of toxin synthesis occurs at the post-transcriptional level.
In nutrient-replete conditions proteins differentially displayed between toxic and non- toxic M. aeruginosa were targets of the global nitrogen control transcriptional regulator NtcA and were involved in carbon fixation, nitrogen metabolism and redox balance. The iron starvation response in M. aeruginosa appeared to result from differential transcription of ferric uptake regulators (Fur), which affected processes involved in iron uptake and modification of the photosynthetic machinery. Iron, but not nitrogen stress, induced microcystin synthesis and prevented chlorosis in the model toxic PCC 7806 strain. In nitrogen-limited growth, the protein expression profiles of M. aeruginosa supported an active involvement of NtcA in the stress response of non-toxic strains. These results suggest that NtcA metabolic control is affected by microcystin synthesis and the nutrient status of the cells, leading to differences in the C:N ratio in toxic and non-toxic cells. This could underpin the differential response of strains of M. aeruginosa to changes in the environment. Further global expression studies of hepatotoxic cyanobacteria are required before reliable bloom control protocols can be established.
iii
iv
Acknowledgments
Thank you to my supervisors, Prof. Brett Neilan and Dr. Belinda Ferrari, for taking me on board, giving me advice when it was needed, but also allowing me the freedom to pursue my own ideas. There are many people who provided technical expertise, advice and editing of this thesis and the associated manuscripts. Prof. David Waite and Dr. Manabu Fujii, thank you for your commitment to this collaboration, for your appreciation of the complexity of biological processes and for giving me glimpses of the even more complex world of chemistry. Manabu and Cuong, for preparing the Fraquil media and growing many liters of culture. Thank you to my proteomics guides, of whom there are many: A. Prof. Paul Haynes, for his advice on nSAF in the early stages of my PhD, access to Stan, the nanoLC-MS/MS machine, and helping with the data analysis. Thanks also to Dana Pascovici and Tim Knightley for changing every trial version of the nSAF analysis tool so that it suits my data, Dr. Anne Poljak and Dr. Mark Raftery for training me on the mass spectrometers at BMSF and providing much technical assistance. Thanks to APAF and BMSF for providing access to equipment, without which this PhD would not have been possible. Prof. Elfriede Pistorius for her kind gift of anti-IdiA antibody and Debra Birch for her help with microscopy. I am also grateful for the financial support by the Environmental Biotechnology CRC and the Graduate Research School, enabling me to attend local and international conferences. There are also many people who have been a daily source of laughs, beer, coffee, smiley cupcakes and support for the past years. A great big thanks to all the groovy people who inhabited the BGGM from 2004 to 2010, especially to my benchies throughout the years who welcomed me into the lab the very first time, sparked my interest in things blue and green and shared the daily ups and downs of lab work. Thank you, Jo, for our many gossip sessions in the lab across the corridor, and Brendan, for always having the door to your new office open. Thanks also to my UTS buddies, Prue, Matt and Andrew for always being happy to procrastinate when I show up without notice. And to my Bulgarian friends who kept in touch after all these years and welcomed me home every time I went back!
v
Finally, a big thank you to my parents and granddad who believe that I can do great things, try to understand the sometimes strange world of science, got more upset than me when experiments wouldn’t work and strains changed toxicity and even contributed to the final corrections of this thesis. I love you very much!
vi Table of contents
Chapter 1...... 19 Introduction...... 19 1.1. Algal blooms and cyanotoxins ...... 3 1.2. Microcystin – mechanism of toxicity, genes and biosynthesis...... 6 1.3. Proposed roles of microcystin production ...... 10 1.3.1. Allelopathic interactions ...... 10 1.3.2. Quorum sensing and toxicity...... 11 1.3.3. Microcystin regulation and the oxidative stress response...... 12 1.3.3.1. Light and oxidative stress in cyanobacteria ...... 13 1.3.3.2. Metal availability and oxidative stress...... 15 1.3.3.3. Nitrogen limitation and oxidative stress ...... 19 1.3.3.4. Oxidative stress and transcriptional regulation of microcystin synthesis ..23 1.4. Proteomics in cyanobacteria – an emerging field with potential use in microcystin research...... 28 1.5. Scope and objectives ...... 31 Chapter 2...... 33 Differential protein expression in toxic and non-toxic strains of Microcystis aeruginosa...... 33 2.1. Introduction...... 35 2.2. Methods...... 36 2.2.1. Microcystis strains and culture growth ...... 36 2.2.2. Toxicity assays ...... 37 2.2.2.1. Microcystin extraction ...... 37 2.2.2.2. Protein Phosphatase Inhibition Assay (PPIA)...... 37 2.2.2.3. HPLC detection of microcystins...... 38 2.2.2.4. ESI Q-TOF MS/MS ...... 38 2.2.3. Proteome analysis ...... 38 2.2.3.1. Protein extraction...... 38 2.2.3.2. 1D SDS-PAGE ...... 39 2.2.3.3. In-gel digestion ...... 39 2.2.3.4. nanoLC-Tandem mass spectrometry ...... 40 2.2.3.5. Protein identification...... 40 2.2.3.6. Data analysis...... 41 2.2.4. Transcription analysis ...... 41 2.2.4.1. RNA extraction...... 41 2.2.4.2. DNase treatment ...... 42 2.2.4.3. cDNA synthesis ...... 42 2.2.4.4. Quantitative real-time PCR (qRT-PCR) ...... 42 2.2.4.5. Reverse-transcription PCR ...... 43 2.3. Results...... 43 2.3.1. The core proteome of M. aeruginosa strains...... 44 2.3.2. Functional categories of proteins expressed in M. aeruginosa strains ...... 44 2.3.3. Diversity of protein expression in M. aeruginosa strains ...... 47 2.3.4. Differentially expressed proteins in M. aeruginosa...... 49 2.3.5. Co-expression of MAE 06820 and trxM ...... 51 2.3.6. Toxicity switching in M. aeruginosa UWOCC MRC ...... 52 2.3.7. Effects of toxicity switching on protein expression ...... 54 2.4. Discussion ...... 55
vii
2.4.1. Diversity of protein expression in strains of M. aeruginosa ...... 55 2.4.2. The core proteome of M. aeruginosa...... 56 2.4.3. Differential display between toxic and non-toxic strains of M. aeruginosa 57 2.4.3.1. Suitability of the nSAF approach for comparison of bacterial strains ...... 57 2.4.3.2. Nitrogen uptake and metabolism proteins...... 58 2.4.3.3. Carboxysome proteins...... 59 2.4.3.4. Proteins involved in redox balance maintenance...... 60 2.4.4. NtcA regulation and microcystin synthesis...... 62 2.4.5. Induction of toxicity in an inactive microcystin producing strain ...... 64 2.5. Conclusions ...... 65 Chapter 3 ...... 67 Transcriptional analysis of the iron stress response in Microcystis aeruginosa ..... 67 3.1. Introduction ...... 69 3.2. Methods...... 71 3.2.1. Microcystis strains and culture growth...... 71 3.2.2. RNA extraction...... 71 3.2.3. cDNA synthesis ...... 72 3.2.4. Quantitative real-time PCR (qRT-PCR) ...... 72 3.2.5. Microcystin detection...... 73 3.2.6. Microscopy...... 73 3.3. Results...... 74 3.3.1. Growth of M. aeruginosa under iron limitation ...... 74 3.3.2. Microcystin synthesis by iron-starved M. aeruginosa PCC 7806 ...... 76 3.3.3. Transcriptional analysis of the iron stress response in M. aeruginosa ...... 76 3.3.3.1. Genes involved in oxidative stress...... 77 3.3.3.2. Iron-binding and acquisition genes...... 78 3.3.3.3. Non-ribosomal peptide synthesis genes ...... 79 3.3.3.4. Transcription of the ferric uptake regulator (fur) homologues...... 80 3.3.4. Sub-cellular localization of FutA...... 81 3.3.5. Ultrastructure of iron-starved M. aeruginosa PCC 7806 cells ...... 83 3.4. Discussion ...... 84 3.5. Conclusions ...... 91 Chapter 4 ...... 93 Proteomic analysis of the iron stress response in the bloom-forming cyanobacterium Microcystis aeruginosa ...... 93 4.1. Introduction ...... 95 4.2. Methods...... 96 4.2.1. Microcystis strains, culture growth and protein extraction ...... 96 4.2.2. In-gel digestion and nanoLC-MS/MS analysis...... 96 4.2.3. Data analysis...... 97 4.3. Results...... 98 4.3.1. Background proteome differences in iron-replete conditions ...... 99 4.3.2. Iron-induced proteomic changes...... 101 4.3.2. Protein expression in iron-starved cells of the toxic M. aeruginosa PCC 7806 ...... 105 4.3.3. Protein expression in iron-starved cells of the non-toxic M. aeruginosa PCC 7005 ...... 108 4.3.4. Protein expression in iron-starved cultures of the genetically engineered non- toxic M. aeruginosa PCC 7806 mcyH- ...... 112 ^ proteins with the same expression pattern in PCC 7806 and PCC 7806 mcyH- during iron stress ...... 115
viii 4.4. Discussion ...... 115 4.4.1. Photosynthesis and respiration ...... 115 4.4.2. Carbon-nitrogen balance and thioredoxin...... 119 4.4.3. Iron transport, iron storage and protection from oxidative stress...... 121 4.4.4. Other, hypothetical and PCC 7806-genome specific proteins ...... 123 4.4.5. Toxicity and transcriptional control of the iron stress response...... 124 4.5. Conclusions...... 125 Chapter 5...... 127 Proteomic analysis of the nitrogen stress response in the bloom-forming cyanobacteria Microcystis aeruginosa...... 127 5.1. Introduction...... 129 5.2. Methods...... 130 5.2.1. Microcystis strains, culture growth and protein extraction ...... 130 5.2.3. In-gel digestion and nanoLC-MS/MS analysis ...... 130 5.2.4. Data analysis...... 131 5.2.5. RNA Extraction ...... 132 5.2.6. DNase treatment ...... 132 5.2.7. cDNA synthesis ...... 133 5.2.8. Quantitative real-time PCR (qRT-PCR) ...... 133 5.3. Results...... 133 5.3.1. Growth and microcystin production in nitrogen-stress and nitrogen replete conditions in M. aeruginosa strains ...... 133 5.3.2. Background protein expression differences in M. aeruginosa strains during nitrogen-replete growth ...... 135 5.3.3. Nitrogen stress-induced protein expression changes in M. aeruginosa strains ...... 137 5.3.4. Protein expression in the toxic M. aeruginosa PCC 7806 in nitrogen-limited batch culture...... 138 5.3.5. Protein expression in the non-toxic M. aeruginosa PCC 7005 in nitrogen- limited batch culture...... 141 5.3.6. Protein expression in the genetically engineered non-toxic M. aeruginosa PCC 7806 mcyH- in nitrogen-limited batch culture ...... 147 ^proteins with the same expression pattern in PCC 7806 and PCC 7806 mcyH- during nitrogen starvation...... 150 5.3.4. qRT-PCR of the signal protein PII, and the transcriptional regulators NtcA and FurA ...... 150 5.4. Discussion ...... 151 5.4.1. Microcystin synthesis and growth of M. aeruginosa in nitrogen-limited conditions...... 152 5.4.1. Photosynthesis and respiration ...... 153 5.4.2. Transport of metabolites during nitrogen limitation...... 154 5.4.3. Carbon-nitrogen balance ...... 155 5.5. Conclusions...... 158 Chapter 6...... 161 Conclusions and future directions...... 161 6.1. Research motivation and objectives ...... 163 6.2. Key findings ...... 163 6.2.1. Diversity of protein expression in strains of M. aeruginosa...... 163 6.2.2. Toxin regulation and the reversal of toxicity ...... 164 6.2.3. Protein expression in toxic and non-toxic M. aeruginosa...... 165 6.2.4. Toxicity and adaptation to nutrient stress ...... 169
ix
6.2.4.1. Adaptation to iron stress...... 169 6.2.4.2. Adaptation to nitrogen stress ...... 171 6.3. Future directions...... 172 6.3.1. Microcystin production in the laboratory and during bloom formation .... 172 6.3.2. Hepatotoxins in other cyanobacterial species...... 173 6.3.3. Microcystin isoforms and other non-ribosomal peptides...... 174 6.5. Conclusion...... 175 APPENDIX A ...... 176 Primer Sequences ...... 176 APPENDIX B ...... 179 Media Composition ...... 179 APPENDIX C ...... 182 Q-TOF MS/MS of microcystin isoforms purified from M. aeruginosa UWOCC MRC and UWOCC MRD ...... 182 APPENDIX D ...... 195 Supplemental Data ...... 195 References...... 197
x Abbreviations
1D SDS-PAGE one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis 2DE two-dimensional electrophoresis ABC ATP-binding cassette ACN acetonitrile BSA bovine serum albumin CHAPS 3-[(3-cholamidopropyl)dimethylammonio]- 1-propanesulfonate DEPC diethyl pyrocarbonate DIC differential interference contrast DTT dithiothreitol ESI electrospray ionisation FDR false discovery rate Fur ferric uptake regulator HFBA heptofluorobutyric acid HPLC high performance liquid chromatography IAA iodoacetamide MCYST-LR microcystin-LR mcyS microcystin synthetase gene cluster MWCO molecular weight cut-off nanoLC/ MS-MS nano-liquid chromatography mass spectrometry NRPS nonribosomal peptide synthetase ORF open reading frame PBS phosphate buffered saline PEI polyethyleneimine PMSF phenylmethanesulphonyl fluoride pNPP p-nitrophenyl phosphate PP2a protein phosphatase type 2a PPIA protein phosphatase inhibition assay ppm parts per million PKS polyketide synthase PS photosystem qRT-PCR quantitative real-time polymerase chain reaction Q-TOF MS quadropole time of flight mass spectrometry RT-PCR reverse transcription polymerase chain reaction ROS reactive oxygen species rpoC1 RNA polymerase subunit C TEM transmission electron microscopy TFA tri-fluoroacetic acid U units v/v volume per volume w/v weight per volume WT wild type
xi
List of Publications and Proceedings
Alexova, R., Haynes, P. A., Ferrari, B. C. and B. A. Neilan. Protein expression in strains of the bloom-forming cyanobacteria Microcystis aeruginosa (in prep., intended for Moll. Cell. Proteom.).
Alexova, R., Fujii, M., Birch, D., Cheng, J., Waite, T. D., Ferrari, B. C. and B. A. Neilan. Iron uptake and toxin synthesis in Microcystis aeruginosa under iron limitation. (in prep., intended for Environ. Microbiol.).
Alexova, R., Fujii, M., Cuong, D., Raftery, M. J., Haynes, P. A., Waite, T. D., Ferrari, B. and B. A. Neilan (2010) Poster presentation. Protein expression and microcystin production in Microcystis aeruginosa. 8th International Toxic Cyanobacteria Conference, Aug 2010, Istanbul, Turkey.
Alexova, R., Fujii, M., Waite, T. D. and B. A. Neilan (2009) Oral presentation. Iron uptake and toxin synthesis in Microcystis aeruginosa under iron limitation. 13th International Symposium on Photosynthetic Prokaryotes, Aug 2009, Montreal, Canada.
Alexova, R., Cheng, J. and B. A. Neilan (2008) Oral presentation. Effect of iron stress on microcystin synthesis in the toxic cyanobacterium Microcystis aeruginosa. Australian Society for Microbiology Conference, Jul 2008, Melbourne.
xii List of Figures
Chapter 1
Figure 1.1. Blooms of Microcystis aeruginosa.
Figure 1.2. General structure of some cyanotoxins.
Figure 1.3. Organisation of the cyanobacterial microcystin synthetase cluster.
Figure 1.4. The photosynthetic electron chain in cyanobacteria.
Figure 1.5. The photosynthetic electron chain in cyanobacteria under oxidative stress.
Figure 1.6.A. Carbon-nitrogen metabolism in cyanobacteria.
Figure 1.6.B. NtcA and PII regulation of carbon nitrogen metabolism
Figure 1.7. Putative binding sites for transcriptional regulators in the central promoter region of the mcy gene cluster.
Figure 1.8. The structure of cyanopeptolin and microcystin-LR.
Figure 1.9. The structure of non-cyanobacterial peptide siderophores.
Chapter 2
Figure 2.3.1. Representative 1D SDS-PAGE gel of strains used in this chapter.
Figure 2.3.2. Functional class distribution in the core Micrococystis aeruginosa proteome.
Figure 2.3.3. Protein distribution amongst the six strains used in this study.
Figure 2.3.4. Cluster dendrogram of protein expression in strains of Microcystis aeruginosa.
Figure 2.3.5. Box-plots of nSAF values for the proteins identified by ANOVA as significantly different (p<0.05) between toxic and non-toxic strains.
Figure 2.3.6. Reverse transcription of trxM and MAE06820.
Figure 2.3.7. MS spectra showing microcystin production by M. aeruginosa UWOCC MRC (A.) and UWOCC MRD (B.)
Figure 2.4.1. Proposed model for the involvement of microcystin and NtcA regulation in the differential expression of proteins in Microcystis aeruginosa.
Chapter 3
Figure 3.3.1. Growth of Microcystis aeruginosa at different iron concentrations.
xiii
Figure 3.3.2. Microcystin production by M. aeruginosa PCC 7806 under iron limitation.
Figure 3.3.3. Transcription changes in genes involved in oxidative stress and iron uptake in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM).
Figure 3.3.4. Transcription changes in genes involved in non-ribosomal peptide synthesis in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM).
Figure 3.3.5. Transcription changes in ferric uptake regulator (fur) genes in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM).
Figure 3.3.6. Localisation of the FutA homologue in iron-starved M. aeruginosa cells by laser scanning confocal microscopy.
Figure 3.3.7. Confocal image of FutA in iron-starved M. aeruginosa PCC 7806 cells.
Figure 3.3.8. Secondary antibody control for confocal microscopy.
Figure 3.3.9. TEM ultrastructure images of representative M. aeruginosa PCC 7806 cells grown in different iron availability for 7 days (exponential growth).
Figure 3.4.1. Iron stress response model in M. aeruginosa strains.
Chapter 4
Figure 4.3.1. Representative 1D SDS-PAGE gel of strains used in this chapter.
Figure 4.3.2. Functional categories of proteins differentially expressed in iron stress in strains of M. aeruginosa.
Figure 4.3.3. Protein expression in iron-starved and iron-replete cultures of M. aeruginosa strains PCC 7806, PCC 7005 and PCC 7806 mcyH -.
Figure 4.4.3. Effect of iron stress on protein expression in M. aeruginosa strains.
Chapter 5
Figure 5.3.1. Growth of Microcystis aeruginosa strains in A. BG11 and B. modified BG110.
Figure 5.3.2. Representative 1D SDS-PAGE gel of strains used in this chapter.
Figure 5.3.3. Functional categories of proteins differentially expressed in nitrogen stress in strains of Microcystis aeruginosa.
Chapter 6
Figure 6.1. Protein expression in Microcystis aeruginosa grown in nutrient-replete conditions.
xiv Figure 6.2. Carbon metabolism in Microcystis aeruginosa grown in nutrient-replete conditions.
Appendices
Figure A.3.1. MS/MS spectra of putative microcystin isoforms in the Microcystis aeruginosa strains UWOCC MRC (A., B.) and UWOCC MRD (C., D.).
xv
List of Tables
Chapter 1
Table 1.1. Divalent metals involved in photosynthesis and maintenance of redox balance in cyanobacteria
Chapter 2
Table 2.2.1. Microcystis aeruginosa strains used in this study.
Table 2.2.2. Primers used for the quantitative real-time PCR (qRT-PCR) analysis.
Table 2.3.1. Functional categories of proteins in Microcystis aeruginosa strains.
Table 2.3.2. Proteins identified as significantly different between the Microcystis aerigonosa toxic and non-toxic group after nSAF analysis and ANOVA.
Table 2.3.3. Microcystin isoforms identified by Q-TOF MS/MS.
Table 2.3.4. Proteins identified as significantly different in expression between M. aeruginosa UWOCC MRC and UWOCC MRD after nSAF analysis and ANOVA (p<0.05).
Chapter 3
Table 3.2.1. Primers used in quantitative real-time PCR.
Table 3.3.1. Summary of transcription changes in selected genes of three M. aeruginosa strains in iron starvation (10 nM Fe) and iron limitation (100 nM Fe).
Chapter 4
Table 4.3.1. Proteins down-regulated in both non-toxic M. aeruginosa PCC 7005 and PCC 7806 mcyH-.
Table 4.3.2. Proteins up-regulated in both non-toxic M. aeruginosa PCC 7005 and PCC 7806 mcyH-.
Table 4.3.3. Proteins differentially expressed in iron-starved and iron-replete cultures of M. aeruginosa PCC 7806.
Table 4.3.4. Proteins differentially expressed in iron-starved and iron-replete cultures of M. aeruginosa PCC 7005.
Table 4.3.5. Proteins differentially expressed in iron-starved and iron-replete cultures of M. aeruginosa PCC 7806 mcyH-.
Chapter 5
Table 5.3.1. Proteins up-regulated in both non-toxic M. aeruginosa strains PCC 7005 and PCC 7806 mcyH- relative to the toxic PCC 7806.
xvi Table 5.3.2. Proteins down-regulated in both non-toxic M. aeruginosa strains PCC 7005 and PCC 7806 mcyH- relative to the toxic PCC 7806.
Table 5.3.3. Proteins differentially expressed (p<0.05, t-test) in cells of M. aeruginosa PCC 7806 grown in BG110 and BG11.
Table 5.3.4. Proteins differentially expressed (p<0.05, t-test) in cells of M. aeruginosa PCC 7005 grown in BG110 and BG11.
Table 5.3.5. Proteins differentially expressed (p<0.05, t-test) in cells of M. aeruginosa - PCC 7806 mcyH grown in BG110 and BG11.
Table 5.3.6. Transcription of selected genes in nitrogen replete and nitrogen-deficient conditions.
Appendices
Table A.1. Primers used in this study
Table A.2.1. Fraquil media composition
Table A.2.2. BG11 and nitrogen-free BG11 media composition
Table A.3.1. Assignment of ions from microcystin peaks in Q-TOF MS/MS used to identify microcystin variants.
xvii
xviii Chapter 1 Introduction
Chapter 1
2 Introduction
Cyanobacteria, also known as blue-green algae, are a diverse group of oxygenic photosynthetic prokaryotes that originated approximately 2.5 billion years ago. The ancestors of these organisms are thought to have contributed to the generation of an oxygenic atmosphere and to be the progenitors of plant chloroplasts. Today, they have adopted a variety of lifestyle strategies, such as plant-symbiotic nitrogen-fixation, the capability to cope with extreme cold in Antarctica, salt stress in tidal pools or severe nutrient limitation in the oceans. Cyanobacteria are often referred to as factories for secondary metabolites, due to the large diversity of compounds that they are able to produce. Some of these metabolites are synthesized by multi-enzyme complexes via non-ribosomal pathways similar to those for antibiotics and immunosuppressants and are being considered as potential therapeutics (Burja, 2003). Their complexity, however, precludes chemical synthesis and instead, genetic manipulation protocols for cyanobacteria have been developed in an attempt to use the existing cellular peptide synthesis machinery (Copp, et al., 2007). Despite their biotechnological potential and the fact that they are an essential part of a balanced ecosystem, cyanobacteria are also a major problem for water management of eutrophicated water bodies where they grow to high cell densities to form algal blooms.
1.1. Algal blooms and cyanotoxins Bloom development is a process affected by a range of environmental conditions, such as temperature, light, wind and availability of nutrients. The increase in human activity in conjunction with global warming has contributed to the development of conditions that favour the development and prolong the duration of algal blooms. Massive freshwater cyanobacterial blooms have now become a global phenomenon regularly affecting large water bodies such as Lake Victoria in Africa, Lake Erie in North America, Lake Taihu in China and the Baltic Sea in Europe, as well as smaller reservoirs (Paerl and Huisman, 2008) (Figure 1.1). Temperatures above 25oC facilitate the growth of cyanobacteria over diatoms and green algae (Paerl and Huisman, 2008). This is partly caused by reduced vertical mixing in the water reservoir, leading to thermal stratification, which can no longer maintain the larger and denser phytoplankton in the upper layers of the water column. Instead, buoyant cyanobacterial species become mobilized from the sediment and adjust the cellular content of gas vesicles in order to control their position in the water column (Badger, et al., 2006, Baptista and Vasconcelos, 2006, Zohary and Breen, 1989). As cyanobacteria photosynthesise near
3 Chapter 1 the water surface during the daytime, carbohydrate ballast accumulates and the bacteria sink to the nutrient-rich sediment. At night, respiration consumes stored carbohydrate and the gas vesicles are filled with CO2, causing the cyanobacteria to rise to the surface, forming a hyperscum that may be several decimeters thick and persist for several months (Brookes and Ganf, 2001, Zohary and Breen, 1989). The high photosynthetic activity in this eutrophicated surface layer causes a rise in surface temperature and dissolved oxygen, depletion of inorganic carbon and an increase of pH above 9 (Badger, et al., 2006).
The benthic population is involved in cyanobacterial overwintering and recruitment of cells during the initial stages of bloom formation (Brunberg and Blomqvist, 2002). Over the course of bloom development, several strains or species of cyanobacteria will often alternate in dominating the water body. In addition, a number of heterotrophic bacteria associate with the cyanobacterial sheath, and may enhance the growth of cyanobacteria or, in conjunction with cyanophages, cause cyanobacterial lysis and bloom senescence (Maruyama, et al., 2003, Tucker and Pollard, 2005).
Figure 1.1. Blooms of Microcystis aeruginosa. A. Swan River Estuary, Western Australia in February 2000. An estimated 900 tonnes of algal mass was removed from the river and microcystin levels reached 8 μg/L at the height of the bloom (Atkins, et al., 2001). B. Warragamba Dam in New South Wales, August 2007. The bloom spread to a surface area of approximately 26 km.
Bloom management is a significant problem for the water industry due to the changes in water odour and taste caused by the lysis of cells, and the disturbance of ecosystems caused by oxygen depletion as cyanobacterial organic matter degrades. Importantly, bloom-forming cyanobacteria are capable of producing a wide range of toxic secondary metabolites (Babica, et al., 2006). These compounds, termed cyanotoxins, are produced
4 Introduction in approximately half to 75% of blooms and contact with, or ingestion of contaminated water, can have a detrimental effect on human and animal health (Vezie, et al., 2002). Cyanotoxins are often classified according to the organ system they target and include hepatotoxins (microcystin and nodularin), neurotoxins (saxitoxin and anatoxin), cytotoxins (cylindrospermopsin) and dermatotoxins (lipopolysaccharides and lyngbyatoxin-a) (Figure 1.2) (Figueiredo, et al., 2004).
Figure 1.2. General structure of some cyanotoxins. A. microcystin-LR, B. nodularin, C. cylindrospermopsin, D. saxitoxin (R=H) and neosaxitoxin (R=OH), E. anatoxin-a.
Bloom-control strategies such as the addition of lytic agents (copper sulfate) to the water are non-discriminatory to a particular cyanobacterial species and therefore do not represent an optimal solution. Synchronized cell lysis, which occurs after algicide treatment or as a bloom senesces naturally, leads to oxygen depletion and mass release of intracellular toxins, in some cases up to 25 000 μg/ L. The toxic compounds may persist in the environment for several weeks, limiting the success of bloom control programs (Babica, et al., 2006, Edwin, et al., 2007).
The biosynthetic pathways of several cyanotoxins and their mechanisms of toxicity have been elucidated, but little is known about their contribution to the physiology of the producing cells (Mihali, et al., 2008, Mihali, et al., 2009, Moffitt and Neilan, 2004, Tillett, et al., 2000). Therefore, in order to establish effective bloom management
5 Chapter 1 strategies, cyanotoxin research has shifted towards attempts to understand the role of toxins in the formation of blooms and how toxicity is regulated by the environment.
1.2. Microcystin – mechanism of toxicity, genes and biosynthesis
One of the most commonly occurring and diverse groups of cyanotoxins are the microcystins, synthesized by members of the genera Microcystis, Nostoc, Anabaena, Anabaenopsis, Aphanizomenon, Fischerella, Planktothrix and the terrestrial Hapalosiphon (Fiore, et al., 2009, Rantala, et al., 2004, Wiedner, et al., 2003). Microcystins are cyclic heptapeptides with the general structure cyclo(-D-Ala-L-X-D- MeAsp-L-Z-Adda-D-Glu-Mdha), where Adda is (2S,3S,8S,9S)-3-amino-9-methoxy- 2,6,8-trimethyl-10-phenyl-(4E),(6E)-decadienoic acid, MeAsp is 3-methylaspartic acid and Mdha is N-methyl-dehydroalanine (Fewer, et al., 2008, Kaebernick and Neilan, 2001). Incorporation of variable amino acids at the X and Z positions, together with differences in the peptide backbone, accounts for the generation of approximately 90 microcystin isoforms that have been identified to date, each with differing polarity and toxicity (Babica, et al., 2006, Tonk, et al., 2005, Wiedner, et al., 2003). The most common and best-studied isoform is microcystin-LR (MCYST-LR) with a molecular weight of 995 Da.
Microcystins exert their hepatotoxicity in mammals by inhibiting protein phosphatases type 1 and 2a (PP1 and PP2a) in the liver. In the first step of the process, the Adda chain in microcystin binds non-covalently to a cysteine residue in the active site of the enzyme (Bagu, et al., 1997). A second time-dependent step involving Mdha forms a covalent bond between the Cys and Adda, preventing the remodelling of cytoskeletal filaments, and thus leading to the deformation of hepatocytes, necrosis and haemorrhagic shock (Bagu, et al., 1997, Fewer, et al., 2008, Soares, et al., 2006). Prolonged exposure to microcystin and the ensuing reactive oxygen species (ROS)- induced damage to DNA have been shown to be a powerful tumor promoter in rodents and an immunomodulator in vitro, but its impact as a carcinogen is yet to be established in humans (Babica, et al., 2006, Mikalsen, et al., 2003, Soares, et al., 2006). The recommended safe limit for microcystin in water has been set at 1 μg/L of MCYST-LR, with typical concentrations during bloom formation reaching 0.05-5 μg/L (Babica, et al., 2006). Microcystin is also able to cause oxidative stress damage, with associated
6 Introduction
ROS production, loss of mitochondrial potential, and apoptosis in numerous aquatic animals and plants (Babica, et al., 2006, Pflugmacher, 2002). Therefore, management of microcystin-containing blooms is important not only from a public health perspective, but also for the integrity of the ecosystem.
a) Microcystis aeruginosa PCC 7806
dnaN G F E D A B C uma1-6 mcyJIH b) Anabaena sp. 90
mcyHIF E J D G A B C atn1 c) Planktothrix agardhii CYA126
mcyT D E G H A B C J ORF1
~ 5 kb
Nonribosomal peptide synthetase (NRPS) genes Polyketide synthase (PKS) genes Tailoring genes Putative transposases Flanking genes
Figure 1.3. Organisation of the cyanobacterial microcystin synthetase cluster. The mcy open reading frames and putative transposases in the vicinity of the cluster are shown for the three microcystin-producing species in which the entire cluster has been sequenced. Figure reproduced with permission from Dr. Alexandra Roberts.
The microcystin biosynthetic pathway was first elucidated in the unicellular planktonic M. aeruginosa sp. PCC 7806 and this strain has since become the model organism to study microcystin synthesis (Dittmann, et al., 1997). Microcystins are non-ribosomally synthesized by a non-ribosomal peptide/ polyketide synthase (NRPS/PKS) enzyme complex encoded in the highly conserved mcyS gene cluster (Figure 1.3). This cluster spans 55 Kbp and encodes ten open reading frames bidirectionally transcribed from a
7 Chapter 1 central promoter. The large microcystin synthetase complex consists of three peptide synthetases (McyA to –C), a polyketide synthase, McyD, hybrid enzymes (McyE and McyG), the tailoring enzymes McyJ, McyF and McyI and a putative ABC (ATP- binding cassette) transporter McyH (Pearson, et al., 2004, Tillett, et al., 2000). The predicted transmembrane location of McyH has suggested that the entire complex may be anchored to the plasma or thylakoid membranes and the transporter may possibly be inovolved in export of the toxin (Pearson, et al., 2004). The central 732 bp promoter contains several environmental factor-responsive regulatory elements, the significance of which will be discussed in later sections of this review. In addition, alternative transcription start points have been identified for mcyA and mcyD based on environmental stimuli such as changing light and iron availability (Kaebernick, et al., 2000, Sevilla, et al., 2008).
Within each microcystin-producing genus, non-toxic strains and species have been identified. In fact, only approximately 1%-38% of reported M. aeruginosa and 0-52% of Planktothrix rubescenes cells in blooms are toxic (Christiansen, et al., 2008, Ostermaier and Kurmayer, 2009). The existence of non-toxic strains is attributed either to a progressive loss of mcy genes, with inactivation by insertion sequence elements, suggesting a redundant role for the toxin, or to a continual intrastrain horizontal gene transfer (Christiansen, et al., 2008, Ostermaier and Kurmayer, 2009, Tooming- Klunderud, et al., 2008). This mode of dissemination of mcy genes among microcystin- producing genera is supported by the lack of correlation between cyanobacterial 16S phylogeny and toxicity, as well as the transposases that are associated with the mcyS cluster (Figure 1.3.) (Mikalsen, et al., 2003). Horizontal gene transfer is not limited to toxin genes and in the past has affected the distribution of other cyanobacterial genes, such as RuBisCo subunits rbcLX, the phycocyanin operon PC-IGS, trnl(UAA) and the circadian clock kai gene cluster (Dvornyk and Nevo, 2003, Mikalsen, et al., 2003). This is made possible by the existence of a functional competence system in cyanobacteria, including microcystin producers, which would allow uptake of DNA by natural transformation, in particular at the late stages of bloom formation when massive cell lysis occurs (Nakasugi, et al., 2007). More recently, horizontal gene transfer, deletion events and recombination between mcyB and mcyC have also contributed to the mcyS cluster evolution and are regarded as the source of the extant variability within microcystin isoforms (Fewer, et al., 2008, Mikalsen, et al., 2003, Tooming-Klunderud, et al., 2008). Gene transfer has also been considered to have been involved in the
8 Introduction acquisition of the nodularin synthesis gene cluster, nda, by the filamentous Nodularia spp. This hepatotoxin is a pentapeptide (Figure 1.2B), synthesized via a pathway homologous to the one encoded by mcy, but the cluster lacks homologues for mcyA and mcyB and may have arisen from an ancestral mcy cluster (Moffitt and Neilan, 2004, Pearson, et al., 2008).
The patchy distribution of toxin genes and the fluctuations in toxicity levels during a bloom, necessitate the screening of cyanobacteria for toxicity following identification of the species involved, so that appropriate health warnings can be issued. The discovery of the mcy cluster led to the development of rapid molecular detection techniques, rather than the time-consuming traditional methods, such as high-performance liquid chromatography (HPLC), mass spectrometry and bioassays. Current quantitative PCR (qPCR) techniques require only small amounts of cellular material and allow for the detection of several cyanotoxin genes in a single reaction within a time-frame of several hours. However, one complication to the molecular approach, is the existence of inactive producers where a small percentage of the cyanobacterial population present in both natural and laboratory environments contains a complete set of toxin-encoding genes, yet are unable to express them (Kaebernick, et al., 2001, Kurmayer, et al., 2004, Schatz, et al., 2005, Via-Ordorika, et al., 2004). This means that identification of microcystin synthesis genes will not always correlate to toxicity and there are yet unidentified mechanisms of mcy gene regulation and expression. In addition, health warnings are based on MCYST-LR equivalents, but the isoforms of microcystin produced are different between blooms and even during the time course of a bloom (Briand, et al., 2008). Also, despite the rapid and reliable detection of potentially toxic bloom events, there are no established prevention and management measures once a bloom is formed. This poses a particular problem in areas where eutrophication is persistent and cyanobacterial blooms are commonplace.
Attempts to gain insight into the function of microcystin and ultimately allow the development of reliable protocols for bloom management, have been limited because of the highly efficient restriction barrier composed of an extracellular nuclease and restriction nucleases in M. aeruginosa, reducing the ease of genetic manipulation in this organism (Dittmann, et al., 1997, Takahashi, et al., 1996). Instead, research in this area has focused on comparative studies of toxic and non-toxic cyanobacteria and the regulation of microcystin by various environmental factors.
9 Chapter 1
1.3. Proposed roles of microcystin production When first discovered, microcystin was regarded as a secondary metabolite, due to the existence of non-toxic strains and thus, the non-essential function that this molecule plays in M. aeruginosa. However, later studies have shown that, unlike typical secondary metabolites, microcystin is constitutively expressed and its synthesis is coupled both to the growth rate and to total protein content (Dai, et al., 2008, Downing, et al., 2005, Kameyama, et al., 2004, Pflugmacher, 2002, Welker, et al., 2006). Toxicity, although not essential, appears to be advantageous for the producing strains under high light or nutrient limitation, and has been proposed to be involved in allelopathic interactions with other organisms or to be a component of the quorum- sensing mechanism in cyanobacteria. On a cellular level, microcystin has been considered as a scavenger for divalent metal ions or a light-responsive signaling molecule.
1.3.1. Allelopathic interactions Allelopathy is the direct or indirect effect that plant-released metabolites exert on other plants (Pflugmacher, 2002). Several studies have indicated that microcystin may have an allelopathic effect as a deterrent of competitor strains of cyanobacteria or grazing dinoflagellates. Hepatotoxic strains of Anabaena spp. have shown the ablity to outcompete neurotoxic ones, while in other competition experiments M. aeruginosa caused the lysis of Nostoc muscarum (Rapala, et al., 1997, Wiegand and Pflugmacher, 2005). Purified microcystin-RR was found to increase the transcription of antioxidant enzymes in the unicellular cyanobacterium Synechocystis sp. PCC 6803 and induce cell aggregation (Li and Sherman, 2002). In contrast, the application of culture medium of chlorotic cells of the toxic strain M. aeruginosa PCC 7806 to cells of the same strain caused pronounced bleaching, but did not have any effect on Synechocystis sp. PCC 6803. This led the authors to conclude that the effects of the media extract were species- specific (Dagnino, et al., 2006). Similarly, a toxic M. aeruginosa strain was able to inhibit the growth of a non-toxic one, but this effect could not be reproduced by the addition of 50 μg/ml of purified toxin, suggesting the involvement of additional secondary metabolites (Schatz, et al., 2005).
Microcystin-producing strains have been shown to survive in the presence of the dinoflagellate Peridinium gatunense, whereas their non-toxic counterpart lysed (Schatz, et al., 2005). Similarly, toxin synthesis increased when spent medium from Planktothrix
10 Introduction agardhii was added (Pflugmacher, 2002). In view of the conflicting findings described above, the hypothesis that the primary role of microcystin is to act as a deterrent, has limited support. The main argument against it is that the microcystin synthesis genes seem to have evolved much earlier than the metazoan lineage (Ostermaier and Kurmayer, 2009). In addition, the suggestion that microcystin production is beneficial in preventing the growth of other cyanobacteria seems to contradict the general dominance of non-toxic strains at later stages of the bloom (Edwin, et al., 2007). Nevertheless, microcystin exposure inhibits the growth of a number of aquatic plant, fish and zooplankton species and has been demonstrated to decrease plankton diversity (Valdor and Aboal, 2007).
1.3.2. Quorum sensing and toxicity Bloom formation and the coordinated cell death that occurs as the bloom senesces, are events most likely coordinated by cell-to-cell signaling induced by secondary metabolites. As cells enter stationary phase, synchronized lysis of a large proportion of the population is observed, providing the remainder of the cells with nutrients and resulting in cell number fluctuations (Schatz, et al., 2007). The requirement of a putative ABC-transporter, McyH, for toxin synthesis has suggested that microcystin is actively exported out of the cell and therefore may act as a signaling molecule (Pearson, et al., 2004). Experiments involving the addition of cell lysate from toxic cells to healthy cyanobacterial culture have shown that it leads to McyB accumulation in the surviving toxic population (Schatz, et al., 2007). However, this was only reproducible in the logarithmic growth stage, whereas stationary-phase cultures did not respond to the addition of toxin or lysate. This observation correlates with microcystin synthesis being linked to the growth rate and cell cycle of cells, thus making it a density- associated phenomenon (Braun and Bachofen, 2004, Schatz, et al., 2007). A similar study by Dagnino et al. (2006) showed that medium from starved M. aeruginosa caused rapid bleaching in a nutrient-replete culture of the same strain, a cell response that is generally associated with oxidative stress or starvation. Although the effect of pure microcystin was not investigated, the study suggested that small, apolar compounds secreted from the stressed cells were acting as growth-inhibitors.
Quorum sensing, the ability of bacteria to sense and respond to the cell density in their environment is widespread among heterotrophic bacteria (Lazazzera, 2000). Quorum sensing has been proposed to be involved in the transition of cells into stationary phase
11 Chapter 1 in order to allow gene transcription to be adequately adjusted for the nutrient limitation that occurs when cell densities are high (Lazazzera, 2000). One of the common quorum-sensing mechanisms in Gram-negative bacteria is the production of N-acyl homoserine lactones (AHL). Bacterial AHL-induced responses are varied and include biosynthesis of antibiotics, cell swarming and plasmid conjugal transfer (Braun and Bachofen, 2004). These compounds accumulate as the bacterial population increases to reach a threshold concentration during late exponential growth when they can act as transcriptional regulators for a number of genes. Microcystin producers contain the AHL-receptor LuxR gene and have been shown to respond to AHL from other bacteria, but cyanobacterial AHL production has not yet been identified (Frangeul, et al., 2008, Kaneko, et al., 2007). AHL production and toxin concentration seem to be inversely related (Braun and Bachofen, 2004), but may be reflective of a decreased growth of AHL-producing bacteria in the mucilage of cyanobacterial colonies as toxicity increases, requiring further studies of these complex communities. Further evidence in support of microcystin as a quorum-sensing compound has come from studies on an mcyB- mutant unable to produce functional multimeric microvirin, a mannan-binding lectin that networks cells within a colony, causing defective colony formation (Kehr, et al., 2006). This correlates with research showing that cells within large colonies are more likely to produce high amounts of microcystin (Kurmayer, et al., 2003). However, the lectins produced by toxic strains M. aeruginosa and M. viridis are different suggesting that the cell-attachment process may be highly strain-specific (Kehr, et al., 2006). Also, the putative involvement of microcystins as cell-signaling peptides seems to contradict observations that an increase in toxin synthesis could also be induced by exposure to two non-toxic peptides produced by M. aeruginosa, micropeptin and microginin (Schatz, et al., 2007). These are just a few examples of the non-ribosomal peptides that cyanobacteria produce and may reflect that the interactions of these chemicals, none of which currently have a defined role, are more complex than is currently appreciated (Welker, et al., 2006).
1.3.3. Microcystin regulation and the oxidative stress response Genetically engineered M. aeruginosa strains, in which mcy genes have been insertionally inactivated, usually have impaired growth, pigmentation and disorganised thylakoid membranes, as well as an increased number of gas vesicles, suggesting that the lack of microcystin may be affecting several different physiological processes (Schatz, et al., 2007). Despite the fact that, with the exception of inactive strains,
12 Introduction microcystin is produced constitutively throughout the growth cycle, most experimental evidence suggests that microcystin regulation is influenced by nutrient limitation and light intensity. These are processes, which are intimately linked to the generation of reactive oxygen species and oxidative stress in cyanobacteria.
1.3.3.1. Light and oxidative stress in cyanobacteria Oxidative stress occurs when the intracellular balance between oxidant and antioxidant molecules is disrupted, and may lead to apoptosis through caspase activation if left uncorrected (Latifi, et al., 2008, Ross, et al., 2006). The major cause of oxidative stress 1 is the formation of reactive oxygen species (ROS) – the singlet oxygen O2, the -. . superoxide ion O2 , hydrogen peroxide H2O2, and the hydroxyl radical OH . During photosynthesis, cyanobacteria constantly produce oxygen through the photosynthetic electron chain, but when light energy cannot be utilized, ROS formation occurs through incomplete oxygen reduction (Singh, et al., 2008). In turn, the photosynthetic process is regulated by light intensity, nutrient status of the cell and availability of cofactors for the photosystem proteins. These conditions change during the diurnal cycle as well as the position of the cell in the water column, and if not optimal may induce ROS formation. This requires a highly effective system for managing ROS production in cyanobacteria, so that minimal damage to the photosynthetic apparatus occurs.
NADP+ NADPH h* h*
PBS FNR Fd LHC
PQ PSII cyt b6/f PSI PQH2
cyt c6 2H2O O2
Figure 1.4. The photosynthetic electron chain in cyanobacteria. The thylakoid membrane is shown in grey and the lumen is at the bottom of the diagram. PSII, photosystem II, PBS, phycobilisome, PSI, photosystem I, LHC, light-harvesting complex, cyt b6/f, cytochrome b6/f complex, PQ, plastoquinone, cyt c6, cytochrome c6, Fd, ferroredoxin, FNR, ferredoxin NADP-oxidoreductase, PB, phycobilisome, LHC, light harvesting complex.
13 Chapter 1
In normal conditions the phycobilisome – a complex of phycobiliproteins, linker peptides and bilin on the cytoplasmic side of the thylakoids, captures light and transmits it to the chlorophylls of photosystem I (PSI) and photosystem II (PSII) (Figure 1.4) (Sarcina and Mullineaux, 2004). Light captured by the phycobilisome is first transferred to a specialised chlorophyll molecule, P680, in the PSII dimer reaction centre formed by the D1 and D2 proteins. The electron generated is moved to plastoquinone QA and after four successive reactions, an electron is extracted from the Mn2+ cluster in the PSII core. This allows the cluster to oxidize two water molecules to one molecule oxygen and four protons. Two of the protons are used in the generation of ATP by ATP synthase. The remaining two protons are used to reduce QB to QBH2 and the newly reduced molecule is released in the membrane plastoquinone pool and replaced by a new QB. The electrons pass through the cytochrome b6/f complex and cytochrome c6 to be supplied to P700+. in PSI (Guskov, et al., 2009). In the PSI trimer, light is captured by an internal antenna of chlorophyll-a and carotenoid molecules and guided to the core of the complex where it re-energizes the P700 chlorophyll dimer. The electron that is produced is passed on to ferredoxin, and finally via ferredoxin NADP+ reductase (FNR) to NADP+ (Zouni, et al., 2001).
In high light, the elevated flux of electrons through the electron transport chain is larger than the utilisation of reducing equivalents and causes ROS to accumulate and inhibit photosystem activity, in particular PSII (Allakhverdiev, et al., 2002, Aurora, et al., 2007, Sakthivel, et al., 2009). Incomplete oxidation of water in PSII and the action of superoxide dismutase (Yoshida, et al.) form hydrogen peroxide, whereas transfer of electrons to oxygen rather than plastoquinone produces superoxide that can be further converted to hydroxyl radicals via interaction with iron. The interaction of molecular oxygen with the acceptor side of PSI and excited chlorophyll generate superoxide and 1 O2, respectively (Yousef, et al., 2003). High light also causes oxygen to be used instead of ferredoxin, producing superoxide in what is known as the Mehler reaction
(Latifi, et al., 2008).This leads to inhibited repair of PSII, resulting in damage to the O2- 2+ evolving complex and release of Mn . The photosystem quinone-binding proteins D1 and D2 are degraded and synthesis of other thylakoid proteins is inhibited (Latifi, et al., 2008). Cyanobacteria usually escape this light-induced damage by having alternative energy-dissipating complexes such as high-light inducible proteins (HLIPs), the protective orange carotenoid protein (OCP) and CP43’. The D1 protein, although highly succeptible to oxidative damage and degradation, is subject to rapid turnover. However,
14 Introduction in extreme oxidative stress conditions this is insufficient to compensate for redox stress, and the translation of the psbA gene, the product of which is D1, is inhibited (Allakhverdiev, et al., 2002). This contributes to a decrease in the PSI/ PSII ratio and reduction of the antenna size in an attempt to slow down the flow of electrons through the system (Singh, et al., 2008). Recently, a component of a chaperone complex, HspA, has been identified as a protein that interacts with thylakoid membranes directly, resulting in a more increased physical order and preventing oxidative stress (Sakthivel, et al., 2009). The ability of high light to cause oxidative stress is confirmed by global expression studies, in one case, Kobayashi et al. (2004) showed both high light and methyl viologen treatment decreased photosynthetic activity, nitrate transport and carbon fixation.
1.3.3.2. Metal availability and oxidative stress Similar to light-induced damage to the photosystems, a number of divalent metal ions are also able to induce ROS production when they are in excess or insufficient, as they directly affect the photosystem structure and electron flow. In addition, degradation of the photosynthetic machinery during oxidative stress causes an increase in the concentration of intracellular free metal ions, which can lead to ROS formation, requiring the tight control of metal uptake and storage in the cell (Latifi, et al., 2008). Cyanobacterial metal requirements seem to be best adapted for the reducing conditions that have existed in the ancient sulfuric ocean where Cu, Zn and Cd were limiting and Fe, Ni, Co and Mg were abundant in the reduced form (Baptista and Vasconcelos, 2006, Cerda, et al., 2007, Koropatkin, et al., 2007). These metals are still used in many enzymes (Table 1.1) with cyanobacterial PSI alone estimated to be the largest sink for iron in the cell, containing up to one-quarter of the available iron (Ferreira and Straus, 1994, Keren, et al., 2004). However, in today’s oxygenic environment these metals become rapidly oxidized at physiological pH and become biologically unavailable, in particular in the presence of phosphate or calcium (Castielli, et al., 2009).
ROS homeostasis forms complex networks with other cellular processes, in particular iron acquisition, since iron is involved in the structure of the photosynthetic apparatus and several enzymes involved in ROS detoxification (Table 1.1, Michel and Pistorius, 2004). The importance of iron for cyanobacteria is evident from the fact that the availability of this metal is able to selectively enhance cyanobacterial growth over that of chlorophytes (Imai, et al., 1999). The involvement of this element in the
15 Chapter 1 photosynthetic process means that cyanobacteria have higher requirements for this metal compared to heterotrophs. Both excess of unbound iron and iron limitation can be detrimental to the cyanobacterial cell. This process seems to apply in particular to photosynthetic organisms, as iron starvation in E. coli does not cause the formation of ROS (Latifi, et al., 2005). In contrast, iron limitation in Anabaena spp. causes a hundred-fold increase in ROS concentration and lipid peroxidation (Latifi, et al., 2005).
In M. aeruginosa, H2O2 exposure can lead to apoptosis due to caspase activation (Ross, et al., 2006).
Table 1.1. Divalent metals involved in photosynthesis and maintenance of redox balance in cyanobacteria.
Metal Function Reference Mn Cofactor for superoxide dismutase (Keren, et al., 2002) (Yoshida, et al.), required for the holoform of the H2O-splitting complex in PSII Ca Required for the holoform of the H2O- (Shcolnick and Keren, 2006) splitting complex in PSII Cu Cofactor for plastocyanin and C-type (Baptista and Vasconcelos, cytochrome oxidase 2006) Mg Structure of the porphyrin ring of (Shcolnick and Keren, 2006) chlorophyll Zn Cofactor for carbonic anydrase and SOD, (Cavet et al., 2003) Cofactor for ferric uptake transcriptional regulator (Fur) Fe Cofactor for cytochrome c6, PSII, PSI, (Battchikova and Aro, 2007, ferredoxin, cytochrome b6/f, NADH Ferreira and Straus, 1994, Latifi, dehydrogenase, Fur and SOD, peroxidase, et al., 2008, Michel and catalase and ruberythrin Pistorius, 2004)
One cyanobacterial cell usually requires approximately 103 Fe ions that need to be incorporated into proteins. However, in the environment the Fe3+ concentration is typically 103 ions/ml, and will rapidly become limiting as the cells grow to form a bloom (Ferreira and Straus, 1994). In addition, at physiological pH, iron quickly precipitates to form hydroxides and hydroxyl-aquo complexes, becoming biologically unavailable and is rapidly scavenged by competitor organisms, thus requiring an efficient uptake mechanism. Cyanobacteria that produce mucilaginous sheaths, such as M. aeruginosa have been proposed to bind metal ions through the polysaccharides surrounding the cell, in addition to proton-active carboxyl, phosphoryl, hydroxyl and amine functional groups at the cell surface (Baptista and Vasconcelos, 2006). Several
16 Introduction mechanisms for iron acquisition involve the transformation of Fe3+ to Fe2+, either by photolysis, the production of superoxide, or enzymatically using ferric reductases (Baptista and Vasconcelos, 2006, Barbeau, et al., 2001, Fujii, et al., 2010). In addition, some cyanobacteria secrete or acquire hydroxamate-type siderophores from other bacteria in their environment. Siderophores are high affinity iron-binding compounds that assist in iron scavenging. They usually exceed the molecular size cut-off of porins and require binding to a receptor on the outer membrane, and transport via a periplasmic binding protein and an ABC transporter into the cytoplasm (Krewulak and Vogel, 2008). Once inside the cell, siderophores may be hydrolysed or their binding may facilitate photochemical reduction or local reduction of the bound iron causing a decrease in the affinity between the siderophore and the metal, releasing Fe2+ (Koropatkin, et al., 2007, Michel and Pistorius, 2004). Microcystin producers M. aeruginosa and P. aghardii have been shown to secrete weakly-binding hydroxamate siderophores, a process that results in recovery of chlorophyll-a synthesis and increase in the specific growth rate. Still, the effect of these siderophores on iron uptake appears to be small and instead cyanobacteria rely on the induction of a high-affinity iron transport system (Ferreira and Straus, 1994, Imai, et al., 1999, Nagai, et al., 2007). M. aeruginosa in particular, is a superior competitor under iron limitation (Nagai, et al., 2007). The assimilation of the newly acquired ferrous iron must be tightly controlled, in order to avoid formation of ROS from H2O2 that is induced by free intracellular iron. This occurs through the sum of the reduction of iron by superoxide ion:
. 3+ 2+ O2 + Fe O2 + Fe and the Fenton reaction, where hydrogen peroxide reacts with iron:
2+ - + 3+ - . Fe + H2O2 OH + FeO2 + H Fe + OH + OH
The net effect, known as the Haber-Weiss reaction, produces OH- and OH. and causes a superoxide attack on Fe4S4 clusters, releasing more free iron and contributing to the stress already experienced by the cell (Michel and Pistorius, 2004). This is avoided by acquisition of iron in the early growth phase, before it becomes limiting and storing it in the oxidized form in bacterioferritins or ferritins (Carrondo, 2003, Keren, et al., 2004).
Iron limitation and the resulting oxidative stress are accompanied by remodeling of the photosynthetic machinery, and a general reduction and disorganization of the thylakoid content in the cell, reducing superoxide formation (Michel and Pistorius, 2004). Prolonged iron limitation results in decreased synthesis of reaction centre proteins, a
17 Chapter 1 decrease in the amount of PSI trimers and an increase in the cyclic electron flow where electrons from ferredoxin are returned via plastoquinone and cyt b6f and cyt c6 to P700 (Nodop et al., 2008) (Figure 1.5). Cyanobacteria attempt to optimize iron transport and siderophore production, replace iron-containing proteins with iron-free alternatives and decrease their metabolic rate. These modifications allow iron-starved cells to maintain electron transport at a significant level for a prolonged period of time (Sarcina and Mullineaux, 2004).
NADP+ NADPH h* h* IsiA
Fd IdiA PBS LHC FNR PQ PSII cyt b6/f PSI PQH2
cyt c6 2H2O O2
Figure 1.5. The photosynthetic electron chain in cyanobacteria under oxidative stress. The alternative cyclic electron flow and the photosystem modifying proteins IdiA and IsiA are shown. The thylakoid membrane is shown in grey and the lumen is at the bottom of the diagram. PSII, photosystem II, PBS, phycobilisome, PSI, photosystem I, LHC, light-harvesting complex, cyt b6/f, cytochrome b6/f complex/ plastocyanin, PQ, plastoquinone, cyt c6, cytochrome c6, Fd, ferroredoxin/ flavodoxin, FNR, ferredoxin NADP-oxidoreductase, PB, phycobilisome, LHC, light harvesting complex, IsiA, iron- stress induced protein A, IdiA, iron-deficiency induced protein A.
Some of the alternative proteins expressed during iron starvation include flavodoxin (also known as iron-stress induced protein B, IsiB), an alternative for the iron- containing ferredoxin and plastocyanin, a replacement for cytochrome c6. Perhaps contradictory, increased siderophore production results in binding to Cu2+, the cofactor necessary for plastocyanin activity, and renders this metal insoluble, causing the iron- rich cytochrome c6 to still be the preferred electron carrier under iron limitation (Ferreira and Straus, 1994). As phycobilisomes are degraded, the acceptor side of PSII becomes exposed and is protected by IdiA (iron-deficiency induced protein A) (Tolle et al., 2002; Lax et al., 2007). The exact function of this protein is not clear, but it is a homologue to bacterial ferric binding proteins and has been shown to bind Fe2+ causing speculation that it may provide the thylakoid membranes with iron, or act as a sink for
18 Introduction free iron released as D1 is degraded. Expression of IdiA occurs in the early stages of iron starvation, but if the stress is prolonged, the iron-stress induced protein A IsiA, also known as CP43’, is expressed (Yousef et al., 2003, Figure 1.5). Synthesis of IsiA is also induced in high light, peroxide stress or during the stationary growth phase and is positively regulated by IdiB and the DNA protection during starvation (Dps) protein at the transcriptional level. It shares high structural homology with the CP43 subunit in PSII and has been observed to form a ring around the PSI trimer in excess light, presumably aiding in excess energy dissipation (Boekema, et al., 2001). Some IsiA also forms free rings in the thylakoid membrane where it may act as a reservoir for chlorophyll or compete with phycobilisomes for excess energy absorption as an early step in iron starvation and oxidative stress (Sarcina and Mullineaux, 2004, Yousef, et al., 2003).
IsiA was the first cyanobacterial protein that was shown to be under regulation by a ferric uptake regulator (Fur) (Ghassemian and Strauss, 1996). Members of the Fur family of transcriptional regulators are generally Zn-metalloproteins that recognize three hexameric 5’ NAT(A/T)AT 3’ repeats termed ‘iron-boxes’ in the promoter region of genes, including fur itself (Bes, et al., 2001). In the presence of ferrous iron, they dimerise via their C-termini and bind either as a dimer, or as a helical array to DNA, inhibiting RNA polymerase binding and acting as repressors of transcription (Ahmad, et al., 2009). When iron is insufficient, the Fur-DNA complex dissociates and transcription can be upregulated, as has been shown with isiA. Alternatively, Fur proteins can interact with small RNA molecules causing transcription activation (Herna´ndez, et al., 2006, Hernandez, et al., 2004, Lee and Helmann, 2007, Lopez- Gomollon, et al., 2007b, Lucarelli, et al., 2007, Parker, et al., 2005). For example, in high iron, Fur is a positive regulator for SOD and iron-storage proteins (Bes, et al., 2001, Latifi, et al., 2008). Two more Fur homologues are present in the genomes of cyanobacteria and are designated FurB, a putative DNA-protecting protein, and FurC, which does not bind metal, but is upregulated during oxidative stress (Hernandez, et al., 2004, Lopez-Gomollon, et al., 2009).
1.3.3.3. Nitrogen limitation and oxidative stress Nitrogen availability is of particular importance for the growth of cyanobacteria, as illustrated by the fact that ammonium is the most common limiting factor for the growth of these microorganisms in Australian water bodies (Brookes and Ganf, 2001). Non-
19 Chapter 1 nitrogen fixing cyanobacteria respond to nitrogen deprivation by degradation of the phycobilisome, chlorosis, and eventually a decrease in growth rate (Dai, et al., 2008). Buoyancy also decreases to allow cells to sink to more nutrient-rich waters (Brookes and Ganf, 2001).
ROS can be formed during macronutrient starvation, including nitrogen limitation, due to the existence of several cross-regulatory points between photosynthesis, redox control and nutrient acquisition, establishing a tight control over the C:N balance in the cell (Figure 1.6A) (Schwarz and Forchhammer, 2005). Reducing equivalents from photosynthesis are captured by thioredoxins, a family of small proteins with conserved cysteine (Cys) residues, which are crucial for the cellular redox balance. In their active form, the Cys in thioredoxin are reduced, and this molecule can then reduce and activate thioredoxin targets, amongst which are several proteins involved in carbon metabolism (Figure 1.6.A) (Florencio, et al., 2006). This ensures that during light periods and the corresponding active photosynthesis, photorespiration is maintained at a high level, and fixed carbon is utilized accordingly. Thioredoxin becomes oxidized at night and is unable to activate its target enzymes, but instead, a regulatory peptide, CP12, binds the GAPDH/PRK complex and inactivates it (Tamoi, et al., 2005, Trost, et al., 2006). The appropriate response of the carbon fixation machinery and downstream carbon metabolism to photosynthesis levels in cyanobacteria is established through the cellular concentration of 2-oxoglutarate (2OG). This comes about due to the lack of 2- oxoglutarate dehydrogenase in cyanobacteria, thus resulting in an open TCA cycle (Muro-Pastor, et al., 2005). During high photosynthetic activity, the accumulation of 2OG causes ammonium to become a limiting substrate for the glutamine synthetase (GlnA) enzyme (Figure 1.6.A). Therefore, in order to avoid further C:N imbalance, the cells induce the expression of nitrate transporters, as well as GlnA . On the other hand, if ammonium is abundant, the intracellular 2OG reserves become depleted and nitrogen assimilatory enzymes and nitrogen transporters are inhibited, in a process called global nitrogen control (Aldehni, et al., 2003). This process is driven by the global nitrogen regulator NtcA, a transcription factor of the cAMP receptor family. The active form of
NtcA is a dimer, which binds to a conserved GTAN8TAC sequence in the promoter region of genes (Ginn and Neilan, 2010, Olmedo-Verd, et al., 2008). This transcriptional regulator is not only nitrogen-responsive, but its activity is modulated in vitro by the presence of DTT, therefore being a general redox sensor as well. NtcA is most highly expressed in conditions of high 2OG and therefore, high C:N ratio and controls the
20 Introduction expression of GlnA, as well as isocitrate dehydrogenase, thereby integrating carbon and nitrogen metabolism (Figure 1.6.B) (Muro-Pastor, et al., 2005, Su, et al., 2005).
The responsiveness of NtcA to 2OG is facilitated by its competitive binding to PipX (PII binding protein) and either NtcA or the glnB gene product PII (Figure 1.6.B) (Aldehni, et al., 2003). PII itself, has opposing effects on nitrate transport compared to NtcA, and instead interacts with the NAGK protein involved in the synthesis of arginine, and thus nitrogen-storage cyanophycin granules during periods of high N-availability (Figure 1.6.B) (Osanai and Tanaka, 2007).
NtcA also binds both Fur and the Fur antisense RNA and therefore can regulate iron acquisition as well (López-Gomollón, et al., 2007a, Lopez-Gomollon, et al., 2007b). This is confirmed by microarray results where the redox status of the electron chain affected multiple processes such as transcription, translation, photosynthesis and nitrate transport, as well as being able to decrease the expression of genes usually activated by NtcA (Hihara, et al., 2003).
21 Chapter 1
Figure 1.6.A. Carbon-nitrogen metabolism in cyanobacteria. Regulatory points by thioredoxin (blue) and NtcA (yellow) are shown. Thioredoxin is shown in its inactive (oxidized) form. 2OG: 2-oxoglutarate; Gln: glutamine; Glu: glutamate; PHA: polyhydroxyalkanoate. 1: carbon fixation – RuBisCo large subunit and carboxysome proteins CcmM, CcmL, CcmK; 2: glyceraldehyde-3-phosphate dehydrogenase; 3: phosphoribulokinase; 4: fructose-1,6,-bisphosphatase; 5: pyruvate dehydrogenase; 6: isocitrate dehydrogenase; 7: glutamine synthetase.
22 Introduction
Figure 1.6.B. NtcA and PII regulation of carbon-nitrogen metabolism. The PipX protein shuttles between NtcA (high C:N) and PII (low C:N). 2OG:2-oxoglutarate; Gif: glutamine synthase GlnA: glutamine synthase; NAGK: N-acetyl-glutamate kinase; PhpA: PII phosphatase; PipX: PII-interacting protein; TrxA:thioredoxin A.
1.3.3.4. Oxidative stress and transcriptional regulation of microcystin synthesis Oxidative stress resulting from nutrient limitation, including macronutrients and iron, as well as high light intensity, are conditions to which cyanobacteria have to adapt daily, and seem to promote microcystin synthesis. Although numerous speculations about the involvement of toxicity in this stress response have been described, the possible genetic basis for these observations has been identified recently (Ginn and Neilan, 2010, Martin-Luna, et al., 2006b). The bidirectional mcy promoter region contains binding motifs for several transcriptional regulators, including the ferric uptake regulator Fur (involved in iron homeostasis), the global nitrogen controller NtcA and the redox-active photoreceptor RcaA (Figure 1.7) (Grossman, et al., 2001, Martin-Luna, et al., 2006b). These three elements may act together and be cross-regulated to allow fine-tuning of microcystin synthesis that matches the changing environmental conditions.
23 Chapter 1
30669 AGTGTTAGAATCGACTTGGAAAAGAATAATTATTGCGACTGACGGGGTGACAAGCAGATG 30610
30609 GAAAGTGAAACAGGGTGTAGAGTGTCGGGTTTAGGGAAAAAGCTTGAGACTTTCGCCAAA 30550
30549 AGATAACGAGGGAATTTGGTTTTTGTCTAGTAAGTCGATTAATTTGATGGATCACAGTGA 30490
30489 GGAAATTTTTCCCCCACCTCACTTAAACTTCAACCTCGTTGTCACCCCTTCAGCTATTAC 30430
30429 GACCAGACAGCTAATCGTACCTGATCAAGGTAGTAATTGTCAATAGACATCTGCAATAAA 30370
30369 CGTTTATGGGGTGTGGCATCCTAAGCTCTGCTCTCTTGGTCTCGCGCAAGCTTATCTTTA 30310
30309 AATGTCACACTTTCTGCACTTCTTAATATTTAATTAATGATTTTTACTAATTTATTGGGT 30250
30249 TCAGTGGTTTCTACAGTGAAGATTTTTTGTCAAAACATACTAGGGAATGTAAAAATATGT 30190
30189 AAAAGTATATGGAGATGTGCAGAATGTCGGTTAGTATGCTACAATGTCGAGGCTCAGAAC 30130
30129 AATTTTGGAGAAGCGACAGAAACCCTGACCTTAGCCGTAGTCGGGTTTCCTGTAGTTCAA 30070
30069 ATAGCAATAATTCCACTCGTCAGAGACCGGAATTATCGCTTTAAGGGAACTGGGAACGGG 30010
30009 GAAAAAAGCATTGTACCCCATGACTCTGAATACCGCCATCAACCACTATTTAGGGAAAAA 29950
29949 GTTAGAACAGCAATGGAAGCACATC
Figure 1.7. Putative binding sites for transcriptional regulators in the central promoter region of the mcy gene cluster. Fur box is shown in light grey, NtcA in dark grey, RcaA in black. The transcriptional start point of mcyA is underlined.
The regulation of toxicity by light may be reflective of the fact that enzymes involved in both microcystin production and photosynthesis are regulated by iron availability and the redox status of the cell (Martin-Luna, et al., 2006b). In addition, alternative transcriptional start points in the mcy cluster have been found under high light (Kaebernick, et al., 2000). The possibility that microcystin is linked to light sensing has been supported by electron microscopy studies showing the highest labeling with anti- microcystin gold conjugate to be most dense in proximity to the thylakoid membranes, and polyphosphate bodies which can bind divalent metals (Young, et al., 2005). A microcystin-thylakoid interaction can be related to the disorganized thylakoids observed in microcystin-knockout mutants and the finding that microcystin has been found to be invariant to chlorophyll-a (1:2 mol mol) (Lee, et al., 2000, Long, et al., 2001, Young, et al., 2005). This association has been proposed to be a result of the structure of microcystin, where the hydrophobic Adda moiety may interact with lipids, forming pores in the membranes and affecting its fluidity. This property of both microcystin and the structurally related nodularin, has been used to study ion flow in eukaryotic membranes (Spassova, et al., 1995, Vesterkvist and Meriluoto, 2003). Okadaic acid, another protein phosphatase inhibitor, also associates with chloroplasts of the producing dinoflagellates. However, different microcystin isoforms seem to interact differently 24 Introduction with membranes, with MCYST-LR showing no effect on the fluidity of membranes, whereas the more hydrophobic MCYST-LW can interact with lipid bilayers directly (Vesterkvist and Meriluoto, 2003). Protein fractionation has revealed that the major part of intracellular microcystin associates with the phycobilisomes. Still, the structural basis and specificity of this interaction is not clear given that the toxin can also bind proteins in a non-specific manner in vitro (Juettner and Luethi, 2008, Vela, et al., 2008).
Several studies have linked the light intensity and quality to toxicity in cyanobacteria. So far, this has been observed in both Planktothrix and Microcystis spp. in which microcystin production follows a sigmoidal curve, with little toxin being accumulated at high or low light intensity (Gerbersdorf, 2006, Tonk, et al., 2005). The maximum transcription of mcy genes occurs at 40 μmol photons m-2 s-1, a light intensity that is typical for 1 m water depth (Kadner, 2005, Kaebernick, et al., 2000, Wiedner, et al., 2003). A study with the filamentous microcystin producer, Planktothrix agardhii, showed that not only toxin concentration, but the toxin isoforms, altered with increasing light intensity (Tonk, et al., 2005). High light appeared to cause the toxin to associate with the outer layers of the cell and the cell sheath when microcystin localization in the cell was studied (Gerbersdorf, 2006). However, this may not necessarily indicate toxin export, as the majority of microcystin release is accounted for by programmed cell death (Ross, et al., 2006, Sakai, et al., 2007). It has been proposed that increased toxicity in high light may be advantageous to the cell and could explain the dominance of toxic strains in the early stages of blooms, whereas non-toxic strains are better competitors when the accumulation of a dense cell mass causes shading (Briand, et al., 2008, Edwin, et al., 2007).
The quality of light also affects toxicity, with a proteome study in M. aeruginosa concluding that under blue light the expression of MrpA was increased (Dittmann, et al., 2001). In a different experiment, red light caused toxin gene transcription to increase, whereas blue light had little effect on toxin production or decreased mcy transcription (Kaebernick, et al., 2000). The phycoerythrin-rich cells of P. rubescenes have been found to be toxic and inhabit deep-stratified waters, whereas a smaller proportion of cells of P. agardhii that flourish in shallow lakes produce microcystin (Christiansen, et al., 2008). In contrast, an immunogold microcystin localization study did not find a difference in cells treated with blue or red light (Gerbersdorf, 2006). This lack of response to blue light may be due to the high degree of attenuation of blue light in freshwater (Kaebernick, et al., 2000, Ting, et al., 2002). Such conflicting results
25 Chapter 1 regarding the effect of light quality on toxicity may be caused by culture in the laboratory, which would differ from the naturally turbid and rich in humic substances waters that microcystin producers flourish in. In this environment, blue light, for example, is highly attenuated, and the effects of blue light on toxicity will be equivalent to conditions near the water surface (Ting, et al., 2002).
Iron starvation, and the involvement of the FurA protein in toxicity is the best studied aspect in microcystin synthesis regulation. Several published reports have found an increase in toxicity in limited iron and upon chelation of iron, suggesting that toxin production may be advantageous during iron stress (Martin-Luna, et al., 2006b, Sevilla, et al., 2008). Weak binding of the toxin to divalent metals such as Cu2+, Zn2+, Fe2+ and Mg2+ has been observed previously and may provide an explanation to the increased survival of toxic strains under iron limitation (Humble, et al., 1997, Saito, et al., 2008). Microcystin shares some structural similarity with other non-ribosomal peptides acting as siderophores, as does another peptide produced by M. aeruginosa, the protease inhibitor cyanopeptolin (Humble, et al., 1997) (Figure 1.8. and 1.9). A 76 kbp cluster, encoding peptide synthestases and polyketide synthases in Anabaena sp. PCC 7120 has been implicated in siderophore production during iron limitation and oxidative stress (Jeanjean, et al., 2008). This cluster is not present in the sequenced genomes of M. aeruginosa but it cannot be excluded that microcystin has a homologous function. However, the metal-toxin interaction is of low affinity and it is still being disputed whether extracellular microcystin is the result of cell lysis and aging culture, rather than the active transport characteristic of siderophores.
Figure 1.8. The structure of (A) cyanopeptolin and (B) microcystin-LR.
Further evidence for the regulation of toxicity by available iron has come from the findings that FurA binds the mcy promoter in vitro (Martin-Luna, et al., 2006b). These
26 Introduction regulators generally act as repressors of transcription by competing with RNA polymerase for binding to DNA. Therefore, the Fur control of the mcy promoter is consistent with the increase of toxicity when iron is limited and Fur repression is alleviated (Sevilla, et al., 2008).
Figure 1.9. The structure of non-cyanobacterial peptide siderophores. A. azoverdin, B. enterobactin, C. ferrichrome (deferriferrichrome), D. alterobactin A.
NtcA is also considered as a possible regulator of the mcy gene cluster given that there is an NtcA-recognition motif in the mcyS promoter (Figure 1.7) (Ginn and Neilan, 2010). Although nutrient limitation has a profound effect on the growth of M. aeruginosa, studies on the link between toxicity and the nutrient status of cells have resulted in conflicting findings. N and P appeared to interact significantly, making N: P ratio a more important factor than the concentration of either element alone (Downing, - et al., 2005, Lee, et al., 2000, Vezie, et al., 2002). Whereas in PO4 limiting batch culture the increase of toxin was larger compared to that in nitrogen-limited culture, in a phosphate-limited chemostat a shift of microcystin production towards more toxic isoforms was observed, increasing the MCYST-LR/ MCYST-RR ratio (Dai, et al., 2008, Oh, et al., 2000). Toxic P. agardhii has been shown to predominate when placed in competition with non-toxic strains under macronutrient limitation (Briand, et al., 2008). On the other hand, others reported that P had no direct effect on toxicity but P
27 Chapter 1 limitation decreases the toxin to dry weight ratio, probably as a result of PHB accumulation and swelling of the cells (Dai, et al., 2008, Long, et al., 2001, Utkilen and Gjolme, 1995). Unfortunately, cross-study comparisons are limited by the fact that there is no standard way of reporting microcystin concentration, with various studies normalising microcystin to cell number, total protein, chlorophyll-a or dry mass.
1.4. Proteomics in cyanobacteria – an emerging field with potential use in microcystin research Due to the difficulties in genetically manipulating M. aeruginosa, the majority of the research to date on the function of microcystin has been focused on the difference between toxic and non-toxic strains, or the difference between mcy-gene knockouts of M. aeruginosa PCC 7806. This strategy has had limited success, with studies acknowledging the possible downstream effects of the antibiotic cassettes which are used to insertionally inactivate mcy genes (Kaebernick, et al., 2001). This is illustrated by the observation that cells in naturally occurring inactive toxin producers were larger than microcystin-synthesising strains, whereas the cells of the mcyB- strain were smaller than the PCC 7806 wildtype (Schatz, et al., 2005). The yet unresolved role of microcystin in toxin-producing cyanobacteria and the allelopathic effects of this molecule have also highlighted the insufficient understanding of the regulation of the mcy gene cluster on an epigenetic level. In order to investigate this, mRNA analysis has been undertaken, but has found no significant difference in the toxin-gene transcripts for microcystin producers or inactive genotypes, suggesting that toxin production may be driven by post-transcriptional regulation (Kurmayer, et al., 2004). Indeed, proteomic work in Synechocystis sp. PCC 6803 has shown that mRNA and protein levels do not always correlate and this also appears to be the case with microcystin synthesis (Suzuki, et al., 2006).
Proteomics, the study of the expressed proteins in an organism under particular conditions has been limited to the cyanobacteria for which genomes have been published, with the majority of work being carried out in the model organisms Synechocystis sp. PCC 6803 and Anabaena sp. PCC 7120 (Barrios-Llerena, et al., 2007a, Stensjo, et al., 2007). The first proteome studies in cyanobacteria were performed in the model photosynthetic prokaryote Synechocystis sp. PCC 6803 following the sequencing of its genome (Sazuka and Ohara, 1997). As the wealth of cyanobacterial genomic information increased, proteomic investigation into the
28 Introduction response of cyanobacteria under various stress conditions, the composition of the thylakoid membranes and the processes of heterocyst formation and nitrogen fixation, among others were carried out (Castielli, et al., 2009, Ow, et al., 2008, Ow, et al., 2009, Srivastava, et al., 2005). In contrast, proteome work on M. aeruginosa has been scarce so far, due to the lack of available genomic data that existed. Only one proteome study has been published on this species, where expression between the wild-type and mcyB- mutant, and found three genes to be differentially regulated (Dittmann, et al., 2001).
Proteomic investigation of cyanobacteria has not been without challenges, with the majority of studies requiring optimized protein extraction methods (Ivleva and Golden, 2007, Plominsky, et al., 2009). In particular, approximately a third of the expressed proteins in nutrient-replete cyanobacterial cells are constituents of the phycobilisome, thus obscuring many lower abundance proteins, in particular when two-dimensional electrophoresis (2DE) is used for the analysis of whole cell extracts (Plominsky, et al., 2009, Barrios-Llerena, et al., 2006). This problem is similar to the large amounts of Rubisco seen in plant green tissue extracts, or the presence of albumin in plasma samples (Haynes and Roberts, 2007). However, parallels to the commercial albumin depletion kits are not yet available for cyanobacteria, making depletion of phycobilisomes a costly and time-ineffective procedure. In addition, 2DE generally underrepresents basic, hydrophobic and membrane-spanning proteins, such as those present in the plasma and thylakoid membranes (Haynes and Roberts, 2007). This method also requires large-scale cultures to be grown, in order to obtain the necessary amount of protein, usually several hundred micrograms for a single large-scale gel.
Despite the technical difficulties outlined above, 2DE remains the separation method of choice for cyanobacterial proteomics (Burja, 2003, Barrios-Llerena et al., 2006). Shotgun multidimensional protein identification technology (MUDPIT) is an alternative, gel-free method for protein separation, which requires the protein extract to be enzymatically digested before chromatographic separation. This technique has been applied to nitrogen-fixing cyanobacteria (Anderson et al., 2006), but the authors found the need to use 450 μg protein, comparable to that required for the 2DE process. One- dimensional (1D) SDS-PAGE combined with nanoLC-MS/MS is a shotgun method for separation of proteins prior to mass spectrometric identification, but it has not been used in cyanobacterial proteome studies so far. This technique does not give information on the pI, molecular weight and isoforms of the identified proteins, but achieves results similar to MUDPIT with no requirements for specialized equipment, and small
29 Chapter 1 quantities of protein extract (up to 100 μg for a typical workflow), making it a popular choice for preliminary proteome studies (Breci et al., 2005, Haynes and Roberts, 2007).
The majority of proteomic studies aim to identify and quantify expression changes, relative to a non-treatment, or ‘normal’ control. Several protein labelling methods for relative quantiation are available, based on the different chemical properties of amino acid side chains, and are extensively reviewed in Corthals and Rose (2007). Label-free quantitation is a simple alternative that remains less-explored, mainly due to challenges in the reliability of quantiation. For example, in the MUDPIT workflow, low abundance ions can coelute with ions of greater abundance, and may not be detected, similar to high-abundance proteins masking spots from less abundant ones in 2DE (Haynes and Roberts, 2007). Spectrum counting is an analytical approach, in which the total number of MS/MS spectra matching peptides from a given protein are compared between samples. Recently developed, the normalized spectrum abundance factor (nSAF) takes into account the length and amino acid sequence composition of the protein, as small proteins have fewer peptides that can be generated after tryptic digestion (Zybailov et al., 2006). The method developed by Zybailov and colleagues (2006) also utlises natural log transformation of the spectrum abundance factor, achieving Gaussian data distribution, permitting statistical analysis and increasing the reliability of the quantitative studies.
A significant challenge in cyanobacterial protein identification is the fact that approximately one third of the theoretical open reading frames (ORFs) in the sequenced cyanobacterial genomes encode for proteins with an unknown or a hypothetical function, making the interpretation of results from differential display experiments challenging (Anderson, et al., 2006, Franguel et al., 2008, Kaneko et al., 2007).
The recent release of the M. aeruginosa genome for two toxic strains, PCC 7806 and NIES-843, allows for proteomic studies of investigations into the global response of these cyanobacteria in normal and stress-inducing conditions to be carried out (Frangeul, et al., 2008, Kaneko, et al., 2007). Both strains contain clusters for three non- ribosomal peptides – microcystin, cyanopeptolin and aeruginosin, while PCC 7806 contains additional putative NRPS/ PKS gene clusters. Comparison of the two genomes revealed that M. aeruginosa strains have a plastic genomic make-up with nearly 12% of the genome consisting of insertion sequences. In addition, many genes were unique for each strain and also unique in cyanobacteria (Frangeul, et al., 2008, Kaneko, et al.,
30 Introduction
2007). For example, the microcystin-related proteins (Mrp) from the earlier proteomic study by Dittmann and colleagues (2001) were not encoded in the NIES-843 genome, suggesting that their expression is a strain-specific response, rather than a difference caused by toxin production. This reflects the possible dangers of using a single strain, PCC 7806, which was chosen purely because it was the most suitable for transformation (Dittmann, et al., 1997), as a representative for the whole genus.
1.5. Scope and objectives Microcystin production by M. aeruginosa appears to affect many metabolic processes in cyanobacterial cells, making it a suitable candidate for global expression studies. However, due to the scarcity of sequenced cyanobacterial genomes, at present there are no proteome studies available, which focus on toxin production by bloom-forming cyanobacteria. In addition, the majority of reports comparing toxic and non-toxic strains have failed to consider background expression differences that exist during nutrient- replete conditions and this may affect the response to particular stresses.
The main objective of this thesis is to carry out a comparative proteomic investigation of toxic and inactive toxin-producing strains of M. aeruginosa in relation to microcystin synthesis. In Chapter 2, the label-free nSAF quantitation method will be used to construct a protein differential display of toxic and non-toxic strains grown in nutrient- replete conditions and the diversity of protein expression in strains of this species will be identified. The involvement of toxin production in the adaptation of these bloom- forming cyanobacteria to iron and nitrogen limitation will also be investigated, in order to assess the role of the transcriptional factors FurA and NtcA in the stress response of M. aeruginosa. The focal point of Chapter 3 and Chapter 4 will be the transcription and expression analysis of M. aeruginosa in an iron-limiting environment, in particular, with reference to iron uptake and oxidative stress. Similar analysis will be performed in Chapter 5, where cells will be grown in nitrogen-limited culture, with a focus on NtcA- regulated processes and C:N balance. The overall aim of this thesis, therefore, is to provide preliminary protein expression data in M. aeruginosa, and estimate the diversity of protein expression in a single cyanobacterial species. This will ultimately serve to identify the basis for the differential response of toxic and non-toxic strains to environmental flux and the processes that govern microcystin production.
31 Chapter 1
32
Chapter 2 Differential protein expression in toxic and non-toxic strains of Microcystis aeruginosa
Chapter 2
34 Protein expression diversity in M. aeruginosa
2.1. Introduction
Numerous studies have been published on the putative role of microcystin, suggesting the possibility of the toxin to be a molecule involved in quorum-sensing, light sensitivity or the acquisition of iron (Dittmann, et al., 2001, Humble, et al., 1997, Kaebernick, et al., 2000, Schatz, et al., 2007). However, the essential contribution of microcystin to the metabolism of cyanobacterial cells has been questioned by the existence of non-toxic strains within all species capable of microcystin production. In fact, under certain environmental conditions, non-toxic strains in both Planktothrix and Microcystis spp. are capable of dominating their environment (Briand, et al., 2008, Oh, et al., 2000, Tonk, et al., 2005, Vezie, et al., 2002).
In addition, the currently available non-toxic strains generated by insertional inactivation of mcyS genes show altered morphology, compared to the wild type toxic strain Microcystis aeruginosa PCC 7806. Some of these changes are readily observable, such as the presence of gas vesicles, changes in pigmentation and thylakoid organization, reduced cell size, as well as the tendency of mutant cells to form aggregates (Kehr, et al., 2006, Schatz, et al., 2005). Such changes have been suggested to be influenced by the association of microcystin with the thylakoid membranes in toxic cells, observed by electron microscopy and possibly, the binding of microcystin to phycobilin proteins (Juettner and Luethi, 2008, Young, 2005 #52). However, a study by Vela et al. (2008) considered this association to be non-specific.
Proteomic studies in a number of bacterial species have opened the possibility to assess the role of various metabolites and the effect of stress-inducing conditions on cellular function. On the other hand, proteomic research of toxic cyanobacteria has been limited due to the relative scarcity of sequenced genomes. In 2001, Dittmann et al. performed the only proteomic study in M. aeruginosa using 2D electrophoresis to compare the wild-type strain PCC 7806 and the mcyB- mutant strain. Using the partially sequenced genome, three proteins, showing some homology to quorum-sensing and light-regulated proteins, were identified as differentially displayed and were assigned as microcystin- related proteins (MrpA-C) (Dittmann, et al., 2001). However, these proteins were not present in the genome of Microcystis aeruginosa NIES-843, suggesting that they may not be necessarily linked to toxicity, but rather represent a strain-specific feature (Frangeul, et al., 2008).
35 Chapter 2
The recent publication of two Microcystis aeruginosa genomes has opened the possibility to study the proteome expression of this toxin-producing species in more detail (Frangeul et al., 2008, (Kaneko et al., 2007). Comparison of the two genomes has revealed that they are both highly plastic and contain regions that could be the result of recent horizontal gene transfer (Frangeul et al., 2008). It was found that 16% of the M. aeruginosa PCC 7806 genome and 28% of the NIES-843 genome coded for proteins not present in the other strain, showing many strain-specific differences, despite the high 16S rRNA gene sequence conservation reported for the M. aeruginosa species cluster (Frangeul, et al., 2008). In the past, phylogenetic comparisons have led to the belief that the Microcystis genus should be unified and strains should be defined as ecotypes adapted to a particular environment, rather than be divided into species based on morphological characterisitcs (Otsuka, et al., 2001). Although most comparative proteomic studies have been performed on the same organism grown in different conditions, there is a general lack of understanding of what constitutes the proteomic diversity within cyanobacterial species.
In this chapter, the proteome of the model microcystin-producing organism M. aeruginosa PCC 7806 and five other morphologically distinct laboratory-isolates of M. aeruginosa isolated from various geographical locations was determined under nutrient- replete conditions. The aim of these studies was to see if the genome plasticity characteristic of the Microcystis species is reflected at the protein expression level. In addition, protein expression was quantified using the label-free normalized spectral abundance factor (nSAF) method in an attempt to expand the current knowledge on the putative metabolic role of microcystin in the cyanobacterial cells. Several proteins involved in carbon-nitrogen metabolism and redox balance were found to be differentially expressed in toxic and non-toxic strains suggesting an involvement of microcystin synthesis in the regulation of the global nitrogen regulator NtcA.
2.2. Methods
2.2.1. Microcystis strains and culture growth M. aeruginosa cultures were grown in Fluka BG-11 media (Sigma, St Luis, MO) in the presence of 100 μg ml-1 cyclohexamide without shaking at 28oC under a 14:10 light/dark cycle (25 )M photons m-2 s-1) supplied by cool white fluorescent lamps.. The strains used in this chapter are listed in Table 2.2.1. For proteomic experiments, cells
36 Protein expression diversity in M. aeruginosa were filtered through a 3 μm pore-size membrane (Millipore, Bedford, MA) in order to remove contaminating bacteria.
Table 2.2.1. M. aeruginosa strains used in this chapter. Strains negative for toxicity do not produce microcystin as determined by a PPIA assay in this study, but contain a complete mcy gene cluster (Roberts and Neilan, 2010).
Isolate Strain Isolate location Toxicity Reference date (Rippka and Braakman Reservoir, The PCC 7806 1972 + Herdman, Netherlands 1992) Malpas Dam, Armidale, (Jackson, et al., UWOCC MRC 1973 + Australia 1984) Malpas Dam, Armidale, (Jackson, et al., UWOCC MRD 1973 + Australia 1984) (Tillett, et al., UWOCC CBS Lake Mendota, WI, USA Pre 1983 - 2001) (Rippka and PCC 7005 Florida, USA 1948 - Herdman, 1992) (Schwabe, et al., 1988, HUB5.3 Lake Pehlitzsee, Germany 1977 - Meissner, et al., 1996)
2.2.2. Toxicity assays
2.2.2.1. Microcystin extraction To verify the toxicity of the strains chosen for proteomic studies, 2 ml of late exponential culture were pelleted by centrifugation, the cells were resuspended in 70% methanol and lysed by four 30 s cycles of bead-beating. Cell debris were removed by centrifugation. The cell lysate was dried completely under vacuum, resuspended and mixed vigorously in 750 μl of chloroform and an equal volume of 20% methanol. Following centrifugation at 12 000 rpm for 20 min, the supernatant was collected and stored at -20o C for further analysis.
2.2.2.2. Protein Phosphatase Inhibition Assay (PPIA) Microcystin extracts were tested for inhibition of protein phosphatase type 2a (PP2a) (Promega, Madison, WI) as described in Carmichael and An (1999). Samples or standards (0-33 nM MCYST-LR (Sigma) in 20% methanol) were incubated for 5 min at o -1 37 C with 0.05 U PP2a diluted in 50 mM Tris pH 7.0, 2 mM MnCl2, 1mg ml BSA and 2 mM DTT. A reaction mixture of 25 mM p-NPP substrate in 0.2 M Tris pH 8.1, 80 -1 mM MgCl2, 0.4 mM MnCl2, 2 mg ml BSA and 4 mM DTT was added and the reaction
37 Chapter 2 was allowed to proceed at 37oC for a further 80 min. Sample absorbance was read in a Benchmark microplate reader (BioRad, Hercules, CA) at 405 nm and the percent inhibition was calculated.
2.2.2.3. HPLC detection of microcystins Samples showing inhibition of PP2a were further tested by HPLC according to the procedure described by Lawton et al. (Lawton, et al., 1994). Briefly, microcystin extracts were loaded on an Alltech 5 μm Nucleosil C18 chromatography column (250 x 4.6 mm) (Grace, Deerfield, IL) and separated using a linear gradient of 30-70% Solvent B (acetonitrile containing 0.05% trifluoroacetic acid) for 30 min. Solvent A was 0.05% trifluoroacetic acid in water with a flow rate of 1 ml min-1. Fractions eluting between 10-30 min were collected, concentrated under vaccum to 50 μl and were tested again for PP2a inhibition as described above.
2.2.2.4. ESI Q-TOF MS/MS To distinguish the isoforms of the toxins produced by Microcystis aeruginosa UWOCC MRC and UWOCC MRD, fractions collected by HPLC and positive for PP2a inhibition activity were subjected to mass spectrometry in a Q-TOF Ultima Micromass (Waters, Milford, MA) fitted with nanospray needles prepared in-house at the Biological Mass Spectrometry Facility (BMSF), UNSW. Tandem mass spectra were acquired using Ar as the collision gas at different collision energies (8-50 eV). Capillary voltage was set at 1.3 kV, cone voltage was 50 V. Microcystin-LR (Sigma) was used as a standard.
2.2.3. Proteome analysis
2.2.3.1. Protein extraction
Prior to protein extraction, cells were grown to mid-exponential phase (OD730nm 0.75- 0.85) and synchronized using the block-release method (48 h dark/ 72 h light) as described previously (Youshida, et al., 2005). The cultures were grown for a further five days with a 12:12 light:dark cycle and 100 ml of cells were harvested by centrifugation. The pellets were washed twice with sterile MilliQ water before resuspending in a final volume of 500 μl MilliQ. The cells were partially lysed by four freeze-thaw cycles in liquid nitrogen and 37oC in the presence of 15 μM PMSF (Sigma) and the lysates were treated with Benzonase endonuclease (Sigma). The cells were resuspended with an equal volume of acid extraction buffer containing 7 M urea, 2 M thiourea, 2% CHAPS, 2% sulfobetaine 3-10 (SB3-10) and 80 mM citric acid, pH 4 and
38 Protein expression diversity in M. aeruginosa proteins were extracted according to the method in Herbert and colleagues (Herbert, et al., 2006). The proteins were resuspended in 150 μl 2DE buffer (7 M urea, 2 M thiourea, 2% CHAPS, 2% SB3-10) and quantified by a Bradford assay (Sigma), as well as by serial dilutions of the sample using 1D SDS-PAGE against known BSA standards. Proteome analysis for all strains was performed with biological triplicates.
2.2.3.2. 1D SDS-PAGE One hundred micrograms of protein were buffer-exchanged by diafiltration in a 3 MWCO VivaSpin microcentrifuge tube (VivaScience, Göttingen, Germany) in 1D-SDS PAGE buffer and boiled for 5 min. A standard containing 1 μg BSA (Sigma) was included in all electrophoresis runs. Samples were loaded on a 10-20% Criterion gel (BioRad) and electrophoresis was performed at 5 mA/ gel for 15 min followed by 200 V until the dye front had reached the end of the gel. The gels were fixed in 40% methanol and 10% acetic acid and stained with Coomassie G250 overnight. The gels were destained with Milli-Q water prior to visualization.
2.2.3.3. In-gel digestion For nanoLC/MS-MS analysis, each lane of a Coomassie-stained gel was cut into 16 segments and these were sliced to approximately 1 mm3 pieces. Destaining was achieved by washing the gel pieces with 50% acetonitrile (ACN)/ 50 mM NH4HCO3 until the stain was no longer visible. After dehydrating with 100% ACN, and air-drying to remove residual acetonitrile, proteins were reduced in 10 mM DTT dissolved in 100 o mM NH4HCO3 for 1 h at 55 C and alkylated in 55 mM iodoacetamide in 100 mM
NH4HCO3 for 45 min in the dark at room temperature. The washing steps described above were repeated twice and gel pieces were dehydrated with 100% ACN.
The gel pieces were covered with 15 )g ml-1 sequencing-grade trypsin solution (Promega) and rehydrated at 4oC for 30 min. Samples were then covered with an o additional 25 μl 50 mM NH4HCO3 buffer and digested overnight at 37 C.
The digest supernatant was transferred to a clean tube and 30 )l of 50% (v/v) ACN/ 2% formic acid was added to the gel pieces, which were vortexed over 20 min to extract the remaining peptides. The supernatant was combined with the initial digest solution and the step was repeated to give a final volume of approximately 60 )l. The volume of the extracted peptides was reduced to 10 μl under vacuum and the solution was desalted using C18 tips (Eppendorf, Hamburg, Germany). The extracted peptides were spun at
39 Chapter 2
16 000 x g for 10 min to remove particulate matter and diluted to 10 μl with 1% (v/v) formic acid if necessary.
2.2.3.4. nanoLC-Tandem mass spectrometry The tryptic digest extracts from 1DE gel slices were analyzed by nanoLC-MS/MS using a LTQ-XL ion-trap mass spectrometer (Thermo-Finnigan, Waltham, MA) according to Hattrup and colleagues ( 2007). The mass spectrometry was performed by A. Prof. Paul Haynes at Macquarie University, Australia. Reversed phase columns were packed in-house to approximately 7 cm (100 m i.d.) using 100 Å, 5 mM Zorbax C18 resin (Agilent Technologies, CA, USA) in a fused silica capillary with an integrated electrospray tip. A 1.8 kV electrospray voltage was applied via a liquid junction up- stream of the C18 column. Samples were injected onto the C18 column using a Surveyor autosampler (Thermo). Each sample was loaded onto the C18 column followed by an initial wash step with buffer A (5% (v/v) ACN, 0.1% (v/v) formic acid) for 10 min at 1 L min-1. Peptides were subsequently eluted from the C18 column with 0%-50% Buffer B (95% (v/v) ACN, 0.1% (v/v) formic acid) over 58 min at 500 nL min- 1, followed by 50%-95% Buffer B over 5 min at 500 nL min-1. The column eluate was directed into a nanospray ionization source of the mass spectrometer. Spectra were scanned over the range 400–1500 amu. Automated peak recognition, dynamic exclusion, and tandem MS of the top six most intense precursor ions at 35% normalization collision energy were performed using the Xcalibur software (version 2.06) (Thermo).
2.2.3.5. Protein identification Raw files were converted to mzXML format and MS/MS spectra from all fractions were processed using the Global Proteome Machine (GPM) Tandem software (version 2007.08.29) against the proteome of M. aeruginosa NIES-843 containing 6312 ORFs (database derived from NCBI, version 31/01/08), and the common repository for abundant peptides (cRAP) (Craig and Beavis, 2003, Craig and Beavis, 2004). This proteome was chosen over the available PCC 7806, so that only proteins universal to the M. aeruginosa species rather than strain-specific proteins were identified. Reverse database searching was used for estimating false discovery frequencies (Peng, et al., 2003). To ensure the specificity of the database, mass spectra were also searched against proteome databases of known bacterial contaminants in M. aeruginosa cultures, Staphylococcus aureus and Pseudomonas aeruginosa, but no significant matches were
40 Protein expression diversity in M. aeruginosa found. Peptide identification was determined using a 0.4 Da fragment ion tolerance. Carbamidomethyl was considered as a complete modification and partial modifications, including the oxidation of methionine and threonine and deamidation of asparagine and glutamine, were also considered.
2.2.3.6. Data analysis Protein function was assigned according to the categories used in CyanoBase (http://bacteria.kazusa.or.jp/cyanobase/). Hierarchical multivariate cluster analysis with the chi-square method of cluster distance calculation was performed in SYSTAT 11. Only proteins present in all biological replicates were included in this model.
The normalized spectral abundance factor (nSAF) quantitation method developed by (Zybailov, et al., 2006) was used to quantify changes in the expression levels of differentially displayed proteins between toxic and non-toxic strains, as well as expression of proteins in UWOCC MRC and UWOCC MRD. Briefly, the method uses the equation:
This calculation takes into account the spectral count of peptides originating from a given protein (SpC) and the protein length in amino acids (L). ANOVA with no multiple adjustments was performed on the log-transformed nSAF data using the statistical package R.
2.2.4. Transcription analysis
2.2.4.1. RNA extraction Parallel to protein extraction, RNA was extracted from the same cultures of Microcystis aeruginosa. Fifty milliliters of cells were pelleted by centrifugation, washed with Milli- Q water and resuspended in 1 ml of TRIzol reagent (Invitrogen, Carlsbad, CA). The cells were snap-frozen in liquid nitrogen and all subsequent steps were performed at 4oC where possible. Cell lysis was achieved by vigorous pipetting and 400 μl chloroform was added to the lysate. After centrifugation at 13 400 x g for 15 min, the aqueous layer was collected and RNA was precipitated in ice-cold isopropanol. The
41 Chapter 2 extracts were centrifuged at 13 400 x g for 30 min and the pellets were washed twice with ethanol, before being resuspended in DEPC-treated water (Invitrogen).
2.2.4.2. DNase treatment RNA extracts were treated with 3 U of Turbo DNase (Ambion, Austin ,TX) for 4 h at 37oC. This was followed by a second extraction in TRIzol as outlined above. The success of DNA removal was assessed by PCR targetting the 16S rRNA gene using the primers 27F and 809R (Table A.1.1, Appendix A) (Jungblut, et al., 2005). The purity and quantity of RNA were estimated using a Nanodrop spectrophotometer (Nanodrop Technologies, Wilmington, DE).
2.2.4.3. cDNA synthesis The FirstStrand random cDNA synthesis kit (Marligen Biosciences, Ijamsville, MD) was used for reverse transcription of 500 ng RNA following the manufacturer’s instructions. The reaction conditions were 22oC for 5 min, 42oC for 90 min and 85oC for 5 min.
2.2.4.4. Quantitative real-time PCR (qRT-PCR)
Transcription levels of genes that were found to be differentially expressed by the nSAF proteomic analysis were quantified by qRT-PCR. Primer sequences are listed in Table 2.2.2. Both the 16S rRNA and the RNA polymerase subunit C (rpoC1) genes were considered as possible reference genes and rpoC1 was chosen for further studies after validation. Reactions were performed in a total volume of 25 μl using 1 μg cDNA, 10 pmol of forward and reverse primer and the Platinum SYBR Green qPCR supermix UDG kit (Invitrogen). Two-step cycling was performed with an initial hold of 60oC for 2 min and 95oC for 2 min, followed by 40 cycles at 95oC for 15 s and 60oC for 30 s in the Rotor-Gene 3000 system (Corbett, Valencia, CA). The efficiency of amplification for each primer set was determined using standard cDNA curves and calculated according to the equation E = 10 [-1/slope] (Pfaffl, 2001). Transcript levels were normalized to rpoC transcription and calculated relative to values for toxic strains using the 2-Ct method as described elsewhere (Pfaffl, 2001). All analyses were performed using biological and technical triplicates.
42 Protein expression diversity in M. aeruginosa
Table 2.2.2. Primers used for the quantitative real-time PCR (qRT-PCR) analysis. The rpoC gene was chosen as a reference.
Primer Sequence (5’-3’) Target name trxMF TCAGGAACTGTTGCAATCCA thioredoxin trxM trxMR AGAATCGGAGCCATCATTTG MAE27590F AATCGAACCCGATAAACCCT hypothetical protein MAE27590 MAE27590R TCACAACCGACAACAGAAGC ccmKF ACGGATCACGATTGTTGGTT carboxysome shell subunit ccmK3 ccmKR TCGGTTTTATTGACGGCTTC PIIF AAGCGATTATCCGACCCTTT nitrogen regulatory protein PII glnB PIIR ATTGACCTTTCTGACGACCG ccmLF TCTGCTTTTGCAATTCATCG carboxysome shell subunit ccmL ccmLR CAATAATCCCCACCACCATC MAE06820F AGTGGTAGCCGAAAGCGATA hypothetical protein MAE06820 MAE06820R TCCCTTCCAAGAACAAATGG nrtAF TGATGGTCGCAAAATTGAAA nitrate transporter nrtA nrtAR GGAATATAGCCCCAACGGAT ndhKF ACTACCCACAAAATGCAGGC NADH dehydrogenase subunit K ndhKR CTCTTTCGGAGGTGCTTGAC ndhK MAE06360F GCACGATCGGTTTTTGTTTT carboxymethylenebutenolidase MAE06360R ATGCGTAGATTGTCCCCTTG rpoC1F CCTCAGCGAAGATCAATGGT RNA polymerase gamma subunit rpoC1R CCGTTTTTGCCCCTTACTTT rpoC1 ntcArealF CATTTCCGTTTGCAGAATCC global nitrogen regulator ntcA ntcArealR TGTTTTTGGGGTGCTATCCT
2.2.4.5. Reverse-transcription PCR In order to assess the possibility of co-transcription of trxM and the neighbouring ORF MAE 06820, 1 μg of RNA extracted from PCC 7806 was reverse-transcribed using the Marligen random cDNA synthesis kit as described in section 2.4.3. A 470 bp fragment spanning both the trxM and MAE 06820 gene was amplified from 1 μg cDNA with the primers trxMF and MAE06820R. The initial denaturation step at 92oC for 2 min was followed by 30 cycles of denaturation at 92oC for 20 s, primer annealing at 57oC for 30 s, strand extension at 72oC for 45 s and a final extension step at 72oC for 7 min. The successful amplification was visualized after electrophoresis in a 1% agarose gel.
2.3. Results The raw mass spectrometry data associated with this chapter is available on the PRoteomics IDEntifications Database (PRIDE) webcite (www.ebi.ac.uk/pride/) under accession numbers 13291-13308 (username: review 90610, password: 2D5^4W!J).
43 Chapter 2
toxic non-toxic
MWM 7806 MRC MRD 7005 CBS HUB
Figure 2.3.1. Representative 1D SDS-PAGE gel of strains used in this chapter. For each strain, 15 μg of protein was separated by electrophoresis on a 4-20% Criterion gel (BioRad). MWM: PrecisionPlus molecular weight markers (BioRad); 7806: PCC 7806; MRC: UWOCC MRC; MRD: UWOCC MRD; 7005: PCC 7005; CBS: UWOCC CBS; HUB: HUB5.3.
2.3.1. The core proteome of M. aeruginosa strains Proteome analysis of the six strains studied in this chapter reproducibly identified 475 proteins in biological triplicates, which were therefore considered for further statistical analysis. Of the compiled proteins, only 82 were found to be expressed in all strains (Figure 2.3.2 and Table 2.3.1), constituting between 29% and 43% of the identified proteins. Ten functional categories were represented in this core proteome, the highest fraction being proteins involved in photosynthesis and respiration. Seven hypothetical proteins were identified in all strains, including MAE 06270, which shows homology to slr1623 and cce 4330, a novel NADH dehydrogenase subunit in Synechocystis sp. PCC 6803 and Cyanothece sp. ATCCC 51142, respectively. No unknown proteins were present in this core proteome.
2.3.2. Functional categories of proteins expressed in M. aeruginosa strains The total number of proteins identified for each strain varied (Table 2.3.1), with the most found in PCC 7005 (280 proteins) and the least in UWOCC CBS (192 proteins). However, in the six strains, proteins from all functional categories were identified. Similar to the results obtained for the core proteome, the photosynthetic, respiratory and
44 Protein expression diversity in M. aeruginosa hypothetical proteins, as well as proteins involved in diverse functions (the category designated ‘other’ in CyanoBase) were predominant.
photosynthesis and respiration cell envelope energy metabolism cellular processes translation transport and binding proteins amino acid biosynthesis transcription hypothetical other
Figure 2.3.2. Functional class distribution in the core M. aeruginosa proteome. Functional class categories were assigned according to CyanoBase (http://genome.kazusa.or.jp/cyanobase).
A total of 90 hypothetical proteins and 9 unknown proteins were identified reproducibly, confirming the correct annotation of these open reading frames and the expression of these proteins in nutrient-replete conditions in the exponential growth phase. Two of the unknown proteins (MAE 60260 and MAE 60250) were in adjacent open reading frames and may form part of an operon, expressed only in strain PCC 7005. Similarly, two putatively co-transcribed clusters of hypothetical proteins were found in PCC 7005 – MAE 36710 and MAE 36720, and MAE 45800 and MAE 45790, the latter pair also being expressed in HUB5.3.
45 Table 2.3.1. Functional categories of proteins in M. aeruginosa strains. The number of proteins universally expressed for all strains are shown in the first column (M. aeruginosa). The number of proteins identified is in black and the % of the proteome is below in red. Functional categories were assigned according to CyanoBase (http://genome.kazusa.or.jp/cyanobase).
Protein Functional category % M. PCC UWOCC UWOCC UWOCC PCC HUB5.3 aeruginosa 7806 MRC MRD CBS 7005 4 20 13 11 12 11 17 Amino acid biosynthesis 4.8 7.6 6.0 4.1 6.2 5.2 6.1 Biosynthesis of cofactors, - 4 2 3 3 4 9 prosthetic groups and - 1.5 0.9 1.1 1.5 1.9 3.2 carriers 3 5 7 5 5 4 3 Cell envelope 3.6 1.9 3.2 1.9 2.6 1.9 1.1 5 10 10 11 7 9 10 Cellular processes 6.1 3.8 4.6 4.1 3.6 4.2 3.6 Central intermediary - 3 2 2 - 1 4 metabolism - 1.1 0.9 0.7 - 0.4 1.4 5 18 15 15 12 12 22 Energy metabolism 6.1 6.9 6.9 5.7 6.2 5.7 7.8 Fatty acid, phospholipids - - 1 1 1 1 2 and sterol metabolism - - 0.4 0.3 0.5 0.4 0.7 Photosynthesis and 32 53 54 56 41 53 52 respiration 39 20.3 25 21.3 21.3 25.2 18.5 Purines, pyrimidines, - 1 1 2 - 1 2 nucleosides and - 0.3 0.4 0.7 - 0.4 0.7 nucleotides - 2 - 5 2 2 4 Regulatory functions - 0.7 - 1.9 1.0 0.9 1.4 DNA replication, - 2 2 1 1 1 4 restriction, modification, - 0.7 0.9 0.3 0.5 0.4 1.4 recombination and repair - 4 1 2 - 2 2 Transcription - 1.5 0.4 0.7 - 0.9 0.7 7 27 17 17 10 17 40 Translation 8.5 10.3 7.9 6.4 5.2 8.1 14.3 Transport and binding 6 10 9 17 14 9 8 proteins 7.3 3.8 4.1 6.4 7.3 4.2 2.8 13 53 41 55 47 44 61 Other 15.8 20.3 18.9 20.9 24.4 20.9 21.7 7 46 37 56 35 36 37 Hypothetical 8.5 17.7 17.1 21.3 18.2 17.1 13.2 - 2 4 4 2 3 3 Unknown - 0.7 1.8 1.5 1.0 1.4 1.1 Total proteins 82 260 216 263 192 210 280
2.3.3. Diversity of protein expression in M. aeruginosa strains Genome analysis of M. aeruginosa strains has revealed that these organisms are genetically highly plastic and contain a large number of transposases and loci that could have been the result of horizontal gene transfer (Frangeul et al., 2008, Kaneko et al., 2007). We were interested to see whether this diversity was also present at the proteome level. For each strain, the major part of the proteome consisted of proteins that were present in the M. aeruginosa core proteome (Figure 2.3.3). Strain PCC 7005 was the most divergent, expressing 65 strain-specific proteins (23% of the strain proteome). On the other hand, in HUB5.3 only 7 strain-specific proteins were identified (3% of the proteome). When the number of proteins shared between strains was considered, it became evident that proteins that were present in a single strain (unique) comprised the largest category, followed by proteins in the core proteome (Figure 2.3.3). Most of the proteins expressed in a single strain were hypothetical or proteins in the ‘other’ category. This suggests that the M. aeruginosa strains differ mostly in processes involved in adaptation to a particular environment or a growth condition, rather than in essential metabolic reactions. No proteins were found to be reproducibly expressed only in toxic, but not in non-toxic strains or vice versa, which could serve as potential markers for toxicity.
Cluster analysis was also performed based on protein expression and the strains were found to be closely related with the largest distance between them being only 10% (Figure 2.3.4). Two large clades were observed, each containing both toxic and non- toxic strains. Interestingly, UWOCC MRD and UWOCC MRC, which are identical in 16S rRNA gene phylogeny and two intergenic spacer region sequences studied so far (Kaebernick, et al., 2001), were separated in different clades of the tree. UWOCC MRC, which has recently reverted to toxin production (Section 2.3.5), was grouped with two non-toxic strains, whereas the protein expression profile of UWOCC MRD was closest to that of the model toxic strain PCC 7806.
Chapter 2
A.
UWOCC CBS unique 2 strains HUB5.3 3 strains PCC 7005 4 strains 5 strains PCC 7806 all strains UWOCC MRC
UWOCC MRD
0 50 100 150 200 250 300 number of proteins
B. 200
150
100
50 number of proteins
0 unique 2 3 4 5 all number of strains
Figure 2.3.3. Protein distribution amongst the six strains in this study. A. Distribution of unique and shared proteins for each strain B. Total number of proteins involved in each combination (1 being unique and 6 being present in all strains).
48 Protein expression diversity in M. aeruginosa
PCCS_7005 7005
UWOCCS_MRD MRD
PCCS_7806 7806
UWOCCS_MRC MRC
S_HUBHUB5.3
UWOCCS_CBS CBS
00 01 02 03 04 05 06 07 08 09 10 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. Distances
Figure 2.3.4. Cluster dendrogram of protein expression in strains of M. aeruginosa. Hierarchical multivariate cluster analysis with chi-square method of cluster distance calculation. Only proteins present in all biological replicates (n=3, 475 proteins) were included in the analysis.
2.3.4. Differentially expressed proteins in M. aeruginosa nSAF values were obtained for 159 proteins, quantified by 246 652 spectral counts. The expression levels of nine proteins were found to differ significantly (p<0.05) between toxic and non-toxic strains when ANOVA (p<0.05, no multiple adjustments) was performed on the proteome data set after nSAF analysis (Table 2.3.2 and Figure 2.3.5). As can be seen on Figure 2.3.5. from the large spread of the nSAF values and standard deviation within each sample group (toxic or non-toxic), the nSAF analysis supported the inter-strain variability of protein expression presented in Section 2.3.3.
Of these, two proteins were involved in nitrogen uptake and metabolism (PII and NrtA), two comprised components of the carboxysome shell (CcmK3 and CcmL) and two are involved in maintaining the cellular redox balance (NdhK and TrxM). Two hypothetical proteins (MAE 06820 and MAE 27590) were also identified as differentially expressed, as was a carboxymethylenebutenolidase (MAE 06360). The majority of the proteins were downregulated in the non-toxic strains, with the exception of PII, NdhK and the carboxymethylenebutenolidase (Figure 2.3.5). Quantitative real-time PCR was used to observe differences at the transcript level, but the patterns predicted by the nSAF data
49 Chapter 2 were consistent with the qRT-PCR analysis only in the instance of nrtA, ndhK and ccmL (Table 2.3.2). In particular, the fold-change in expression and transcription in toxic strains relative to non-toxic M. aeruginosa was similar for ccmL and ndhK (Table 2.3.2).
Figure 2.3.5. Box-plots of nSAF values for the proteins identified by ANOVA as significantly different (p<0.05) between toxic and non-toxic strains. In each panel, expression in non-toxic strains is presented on the left and values for toxic strains are on the right. The y-axis is log nSAF values.
50 Protein expression diversity in M. aeruginosa
Table 2.3.2. Proteins identified as significantly different between the M. aeruginosa toxic and non-toxic group after nSAF analysis and ANOVA. Expression patterns determined by nSAF and transcription values from qRT-PCR are shown for non-toxic strains relative to those for the toxic strains. Transcript levels significantly different (t- test, p<0.05) from those in toxic strains are in bold. Standard deviation values are in brackets and italicized. Accession numbers are shown as they appear in CyanoBase.
Protein GI p-value Accession Function Expression Transcription gi|166085750| 0.05 MAE06360 carboxymethylenebutenolidase 3.84 -1.03 (1.71) gi|166090860| 0.02 MAE57460 nitrogen regulatory protein PII 2.87 -1.85 (0.95) (glnB) gi|166086291| 0.04 MAE11770 NADH dehydrogenase subunit 1.67 1.95 (1.31) (ndhK) gi|166086594| 0.03 MAE14800 nitrogen transport protein -1.71 -2.55 (0.93) (nrtA) gi|166087873| 0.02 MAE27590 hypothetical protein -1.76 1.06 (1.98) gi|166090653| 0.02 MAE55390 carbon concentrating -1.94 2.36 (1.08) mechanism protein (ccmK3) gi|166089906| 0.02 MAE47920 carbon concentrating -2.02 -1.94 (0.92) mechanism protein (ccmL) gi|166085796| 0.03 MAE06820 hypothetical protein -2.14 3.52 (1.36) gi|166085797| 0.01 MAE06830 thioredoxin (trxM) -2.79 1.47 (1.12)
Due to the possible involvement of the transcriptional regulator NtcA in this differential display, but the inability of the proteome analysis to detect the low expression levels of this protein, ntcA transcription was also measured and found to be 3.11(± 1.42)-fold upregulated in non-toxic strains, which was not a significant difference (p<0.05).
2.3.5. Co-expression of MAE 06820 and trxM The similar expression profiles of TrxM and the hypothetical protein MAE 06820 (Figure 2.3.5) and their location in neighbouring ORF in the genome of both M. aeruginosa NIES-843 and PCC 7806 suggested co-transcription of these genes. Reverse transcription (RT-PCR) was performed to evaluate whether the two open reading frames were indeed co-transcribed. The amplification of a fragment of the expected size (470 bp) spanning both genes was successful, confirming that they can be transcribed from a single promoter (Figure 2.3.6).
51 Chapter 2
M1 2
MAE06810 MAE06840
MAE06820 trxM 1,000 bp 500 bp
Figure 2.3.6. Reverse-transcription of trxM and MAE06820. After amplification with primers spanning a 470 bp region of the two open reading frames shown in the right panel, amplicons were electrophoresed in a 1% agarose gel. M: Gene Ruler DNA ladder (MBI Fermentas, Burlington Canada); 1: reverse-transcribed PCC 7806 RNA; 2: negative control. The binding sites of the primers used in the reverse transcription reaction are marked with black arrows and the co-transcribed region is in grey. The open reading frames are shown as their accession numbers in CyanoBase: MAE06810: phosphopantheine adenyltransferase; MAE06820: hypothetical protein; trxM: thioredoxin M; MAE06840: methionine aminopeptidase.
2.3.6. Toxicity switching in M. aeruginosa UWOCC MRC A toxicity screen of the cultures used in this chapter revealed that UWOCC MRC, a strain which was expected to be non-toxic, was producing MCYST-LR and [DAsp3] MCYST-LR (Figure 2.3.7 and Table 2.3.3). These isoforms are different from the microcystin species found in the toxic strain UWOOC MRD, which produced several aromatic isoforms of the toxin as well as other unidentified variants (Table 2.3.4). Contamination by other Microcystis spp. was ruled out by sequencing of the 16S rRNA gene and PCR of the regions of the mcy cluster characteristic for UWOCC MRC and UWOCC MRD (Roberts and Neilan, 2010). Several subcultures were tested for toxicity by PPIA over a period of two years, and the toxicity of UWOCC MRC appears to be a stable trait.
52 Protein expression diversity in M. aeruginosa
Table 2.3.3. Microcystin isoforms identified by Q-TOF MS/MS. Purified MCYST- LR was used as a standard for both HPLC and MS studies.
HPLC Molecular weight Strain Toxin isoformsa elution time Reference (Da) (min) PCC 7806 MCYST-LR 995.45 19.4-20.4 (Phelan and [DAsp3]MCYST-LR 995.50 18.7-19.4 Downing, 2007) UWOCC MRD MCYST-XZ 974.00 25.0-25.6 MCYST-XZ 1002.00 30.1-30.7 MCYST-XZ 1006.00 25.0-25.6 This study MCYST-YY 1052.30 25.0-25.6 MCYST-YM(O) 1036.50 30.1-30.7 UWOCC MRC MCYST-LR 995.50 17.2-19.7 This study [DAsp3]MCYST-LR 995.50 18.4-19.7 a X and Z, unidentified amino acids at the variable positions
]MCYST-LR 3
8.00000000
8.00000000 [DAsp RA_080617_7B 176 (3.319) Cm (45:176) TOF MS ES+ RA_080617_7C 129 (2.434) Cm (9:141) TOF MS ES+ 274.0541 3.90e4 100 100 995.4648 2.54e5
152.9900 MCYST-LR 996.4940 % %
995.5227 275.0864 509.2441 445.1060 430.9063 148.9446 520.2404 509.2607 996.5287 430.8987 981.5042 997.5236 560.8619 520.7502 0 m/z 0 m/z 100 200 300 400 500 600 700 800 900 1000 1100 200 400 600 800 1000 1200 1400 A.
8.00000000 8.00000000 RA_080617_8F 79 (1.494) Cm (25:79) TOF MS ES+ RA_080617_8D 112 (2.115) Cm (31:123) TOF MS ES+ MCYST-YY 239.1230 1.80e4 1052.4956 1.76e3 100 100
148.9491
% 445.1060 % MCYST 1053.4943 1006.4534 MCYST-YM(O)
1074.4674 1036.4753
371.0957 MCYST 446.1113
593.1549 MCYST 1075.4644 1037.4902 667.1732 1002.4938 551.7115 974.4694 1076.4738 1059.4723 0 m/z 0 m/z 600 650 700 750 800 850 900 950 1000 1050 1100 200 400 600 800 1000 1200 1400 B. Figure 2.3.7. MS spectra showing microcystin production by M. aeruginosa UWOCC MRC (A.) and UWOCC MRD (B.). Labelled peaks were identified by MS/MS spectra as microcystin isoforms. An annotated peak list and MS/MS spectra are attached in Appendix C.
53 Chapter 2
2.3.7. Effects of toxicity switching on protein expression Since UWOCC MRC, a spontaneous non-toxic mutant of UWOCC MRD was found to have reverted to toxin production, it was of interest to evaluate the extent of protein expression differences in these otherwise 16S-identical strains.
Table 2.3.4. Proteins identified as significantly different in expression between M. aeruginosa UWOCC MRC and UWOCC MRD after nSAF analysis and ANOVA (p<0.05). Expression patterns determined by nSAF are shown for UWOCC MRC relative to UWOCC MRD. Accession numbers are shown as they appear in CyanoBase.
Protein GI p-value Accession Function Expression gi|166088876| 0.02 MAE37620 gas vesicle protein GvpC 50.92 ribulose biphosphate carboxylase large gi|166089903| 0.02 MAE47890 12.21 subunit RbcL gi|166089512| 0.02 MAE43980 hypothetical protein 10.94 gi|166091382| 0.00 MAE62680 dihydropicolinate synthase 5.56 pyruvate dehydrogenase complex gi|166091078| 0.02 MAE59640 dihydrolipoamide acetyltransferase 4.76 component gi|166089740| 0.01 MAE46260 glutathione reductase 4.39 gi|166086371| 0.02 MAE12570 ferredoxin-NADP oxidoreductase 2.40 gi|166087560| 0.02 MAE24460 phycocyanin alpha subunit 1.52 phycobilisome small rod linker polypeptide gi|166086370| 0.05 MAE12560 -3.16 CpcD gi|166085178| 0.05 MAE00640 putative peptidase PatA homolog -3.67 gi|166089721| 0.04 MAE46070 10 kDa chaperonin GroES -3.84 gi|166086720| 0.05 MAE16060 aldo/keto reductase -5.84 gi|166091207| 0.02 MAE60930 bacterioferritin comigratory protein -7.08
Thirteen proteins were observed to be differentially displayed at a significant level following ANOVA analysis, with some such as GvpC, CpcD and the phycocyanin alpha subunit resulting in a readily observable phenotype (Table 2.3.4). In fact, cells of UWOCC MRC contain gas vesicles and have different pigmentation compared to the parental strain UWOCC MRD, confirming the validity of the results presented in Table 2.3.4. Several proteins were involved in carbon metabolism (dihydropicolinate synthase, RbcL and the pyruvate dehydrogenase component), as well as redox balance – glutathione reductase, bacterioferritin comigratory protein and ferredoxin NADP- oxidoreductase. No proteins involved in the cell envelope structure were found to be differentially expressed despite the ability of UWOCC MRC to form colonies in culture.
54 Protein expression diversity in M. aeruginosa
2.4. Discussion
2.4.1. Diversity of protein expression in strains of M. aeruginosa The proteome studies presented here indicate that protein expression in strains of M. aeruginosa is highly variable, with only a third of the expressed proteins in each strain constituting the core proteome of the species. This result was unexpected, given the close 16S homology shared within strains of this species. The observed proteome diversity supports previous suggestions that M. aeruginosa strains should be considered as ecotypes adapted to survival in a particular environmental niche (Otsuka, et al., 2001). It also highlights the dangers of using predominantly M. aeruginosa PCC 7806 as a model organism for the entire species. In particular, under stress-inducing conditions, such as nutrient starvation or changes in light quality and intensity, many strain-specific acclimation processes are known to take place in microcystin producing cyanobacteria (Briand et al., 2008, Edwin et al., 2007, Vezie et al., 2002). These are also reflected by the differences in protein expression observed here. It is possible that a subset of proteins was not identified in this study due to the choice of protein extraction and separation methods. This requires deeper proteome mining by subcellular proteomics and combination of several technical approaches (Breci et al., 2005), as well as the analysis of more strains (including non-toxigenic M. aeruginosa), which was beyond the scope of this study.
The results suggest that use of a single model strain is an inappropriate approach for studying toxicity and the effects of microcystin production on cyanobacterial metabolism since proteome diversity was high within the toxic and inactive toxin- producing (non-toxic) strain groups studied. This was evident from the cluster analysis (Figure 2.3.4), which failed to separate the strains based on their ability to produce microcystin, geographical origin or period of culture in the laboratory. No proteins were expressed solely in toxic or non-toxic strains, including the McyA-H proteins involved in microcystin analysis. The mixture of toxic and non-toxic strains in the same clade of the cluster tree presented a similar situation to what has been observed on a genetic level, where phylogeny-based methods have not been able to discriminate between toxic and non-toxic, but toxigenic strains (Tillett, et al., 2001). It is likely that protein expression differences between the two groups will be more pronounced when non- toxigenic strains are included in the analysis, as regulatory elements might still be acting on the remainder of the mcy cluster in the inactive producer strains, affecting
55 Chapter 2 downstream processes. However, the aim of this study was to investigate the potential differences in regulatory processes that allow the inactive strains to switch off and revert to toxicity, as was the case with UWOCC MRC,. Thus, only toxigenic strains were included in the analysis.
The existing proteomic and genomic diversity in M. aeruginosa should therefore be taken into consideration, in particular, in the study of physiological responses to changing environmental conditions. Ensuring that a number of strains are used for future comparative studies would avoid confusing conclusions as illustrated by the study of PCC 7806 and its knock-out mutant by Dittmann and colleagues (2001) and the more recent comparative genomic analysis of PCC 7806 and NIES-843 (Frangeul, et al., 2008).
An unusually large number of proteins expressed solely in PCC 7005 was also observed. Since this strain has been maintained in laboratory culture for several decades more than the remaining strains used here (Table 2.1.1), its proteome may reflect changes that occur during prolonged cultivation in the nutrient-rich environment of laboratory media. Such phenotypic changes have been observed previously in M. aeruginosa with the loss of gas vesicles due to transposition events and the dispersal of colonies after prolonged culturing (Kehr, et al., 2006, Mlouka, et al., 2004, Schatz, et al., 2005). The M. aeruginosa strain PCC 7005 also contained most of the hypothetical and unknown proteins found in adjacent reading frames, the majority of which were absent in the remaining strains of the cohort. In fact, only 9 unknown proteins, out of 1404 predicted, were identified in the proteome of the strains studied here and none of these were present in the core proteome. It is possible that the majority of proteins designated as unknown are involved in adaptation or growth-stage specific processes in M. aeruginosa and are less abundant in nutrient-replete conditions, hindering the identification of these proteins with the experimental approach used here.
2.4.2. The core proteome of M. aeruginosa The majority of proteins shared between all strains of M. aeruginosa were involved in photosynthesis and respiration processes. This distribution of protein expression is typical for cyanobacteria, where phycobilisome components are regularly found in high abundance and obscure less-abundant proteins (Anderson, et al., 2006). The number of proteins that were identified, as well as the pI and molecular range covers are also similar to other gel-based studies of total cell extracts of cyanobacteria, such as the 56 Protein expression diversity in M. aeruginosa unicellular Synechocystis sp. PCC 6803 and the filamentous nitrogen-fixing Anabaena variabilis ATCC 29413 (Barrios-Llerena, et al., 2007a, Barrios-Llerena, et al., 2007b). It is likely that other proteins remained undetected since only one genomic database was used for identification and there may be proteins expressed in certain strains but not encoded in the NIES-843 genome. However, the aim of this chapter was to find proteins shared across the M. aeruginosa species, rather than perform exhaustive proteome study, for which sub-cellular protein fraction analysis and multiple analytical approaches would be required.
The large differences in protein expression between strains of M. aeruginosa, although unexpected, are not unique for bacteria. A study comparing two strains of E. coli found 38% of the secreted proteins to be conserved and a similar fraction were strain-specific (Xia, et al., 2008). Similarly, a study of the atomic composition of strains of the marine cyanobacterial Prochlorococcus spp. found that the elemental composition of their proteomes was strongly influenced by their site of isolation and the nutritional availability at that site (Lv, et al., 2008). Nevertheless, most proteome studies comparing bacterial strains have regularly found the core proteome to consist of 70 to 90% of the identified proteins (Donaldson, et al., 2009, Dumas, et al., 2009, Patrauchan, et al., 2007, Wang, et al., 2009).
2.4.3. Differential display between toxic and non-toxic strains of M. aeruginosa
2.4.3.1. Suitability of the nSAF approach for comparison of bacterial strains The nSAF approach was initially used to compare the expression profiles of a single organism grown under different conditions (Zybailov, et al., 2006). In this chapter, this label-free quantitation method was applied to analyse protein expression in several strains of a cyanobacterial species. The results revealed a large diversity in the proteomes of M. aeruginosa strains, a trait which could potentially create problems for the statistical validation of differentially expressed proteins. The analysis of similarly diverse proteomes is not possible with image-based methods such as 2-dimensional electrophoresis (2DE), which has been recently attempted unsuccessfully in cylindrospermopsin-producing cyanobacteria (Plominsky, et al., 2009). However, the reciprocal expression profiles of PII and NrtA, the similar expression of the CcmL/ CcmK3 and TrxM/ MAE 06820 pairs, as well as the upregulation of the gas vesicle and phycocyanin proteins in UWOCC MRC, that were seen here, all suggest that
57 Chapter 2 biologically meaningful data can be extracted from even highly diverse datasets using nSAF.
2.4.3.2. Nitrogen uptake and metabolism proteins
The expression of the signal transduction protein, PII, encoded by glnB, was found to be affected to a significant extent by the presence of microcystin in M. aeruginosa cells. This protein is central to coordinating photosynthetic activity, carbon metabolism and nitrogen assimilation, due to its dependence on the availability of ATP, 2-oxoglutarate (2OG) levels and competition with the global nitrogen regulator NtcA for binding to the
PII interaction protein X (PipX) (Aldehni, et al., 2003, Espinosa, et al., 2006). This complex post-transcriptional regulation of PII may be the reason for the lack of correlation between the expression and transcription data for glnB observed here. Nevertheless, its down-regulated expression in the toxic M. aeruginosa was consistent with the observed up-regulation of the nitrate transporter NrtA in the same strains
(Table 2.3.2, Figure 2.3.5). Such reciprocal relationship between PII and NrtA is the result of the regulation of both the glnB and nrtA genes by NtcA, and has been reported previously (Takatani and Omata, 2006). The differential expression of PII between toxic and non-toxic strains is interesting, given the fact that the mcy gene cluster appears to be regulated by ntcA (Ginn and Neilan, 2010), and may suggest a regulatory process of microcystin synthesis, involving the PII-NtcA couple. The observed expression patterns of PII and NrtA are indicative of a high 2OG level in the cell as a result of a high C:N metabolic ratio, and the following activation of NtcA and up-regulation of nitrogen assimilation (Muro-Pastor, et al., 2005). Unfortunately, NtcA expression was at a level below the detection limit of the proteomic approach used here and the levels of the NtcA protein could not be determined directly, but ntcA transcription did not change significantly. Thus, the differential regulation of these proteins may be the result of increased NtcA activity, rather than levels of expression, being modulated by microcystin (Figure 2.4).
The nitrate ABC transporter, of which NrtA is a subunit has an unusual structure that is shared with members if the cyanobacterial cyanate (Cyn) and bicarbonate (Cmp) transporters (Koropatkin, et al., 2006). Both NrtA and CmpA are inhibited by high levels of nitrate, providing one of the links between carbon and nitrogen metabolism
(Koropatkin, et al., 2006). In addition, PII has been shown to regulate carbon transport and respond to the carbon fixation rate in the model cyanobacterium Synechocystis sp.
58 Protein expression diversity in M. aeruginosa
PCC 6803 (Frochhammer, 2004; Hisbergues et al., 1999). In this microorganism, PII control results in immediate decrease of nitrate transport, when CO2 fixation is inhibited or CO2 supply is limited (Frochhammer, 2004). However, a PII mutant expressed high- affinity carbon transporters, even in the presence of inorganic carbon (Hisbergues et al.,
1999). Thus, the decreased abundance of the PII protein in the toxic strains studied here, is expected to affect both nitrate and carbon metabolism, resulting in enhanced carbon fixation in microcystin-producing cells.
2.4.3.3. Carboxysome proteins Carbon fixation in cyanobacteria occurs in specialized minicompartments, the carboxysomes. In beta-cyanobacteria these are formed by a multi-faceted protein shell, comprised of repeated hexamers of the homologous carbon concentrating mechanism proteins CcmK1-4, as well as CcmL, a pentamer that forms pores at the vertices of the icosahedral shell and allows flux of metabolites in and out of the carboxysome (Kerfeld, et al., 2005, Tanaka, et al., 2008). The protein shell encases the enzymes RuBisCo and - carbonic anhydrase, which are involved in extraction of CO2 from HCO3 , and thus, the final stages of carbon fixation (Badger and Price, 2003).
From the data presented here, it appears that the non-toxic strains down-regulated the expression of two shell subunits, CcmK3 and CcmL. It is likely that this was the result of an increase in carboxysome numbers in toxic strains, rather than a change in the carboxysome size (in which case the expression levels of CcmL would remain constant). However, this was not reflected by a change in expression of the internal components of the carboxysome, such as RuBisCo and carbonic anhydrase, suggesting that at the time of harvesting the cells, the toxic strains contained a number of empty carboxysomes. In cyanobacteria, such anomalies in carboxysome structure have been observed previously by electron microscopy in high-carbon requiring mutants (Orús, et al., 2001). This hypothesis is consistent with the observed decrease of PII in toxic M. aeruginosa, causing an increase in nitrate uptake, and subsequently, inhibition of bicarbonate uptake in microcystin producers. Interestingly, of the four CcmK homologues, which were all found to be expressed in the strains studied here, only CcmK3 was differentially regulated based on toxicity. The contribution of CcmK3 to the structure of the shell has not been studied elsewhere. However, changes in the carboxysome shell composition have been proposed to influence the permeability of this microcompartment to metabolites and could be dependent on the carbon status of
59 Chapter 2 the cell (Eisenhut, et al., 2007). At the transcript level, ccmL was down-regulated in non-toxic strains, consistent with expression data, but unexpectedly, ccmK transcription appeared to be up-regulated. Such discrepancies in the transcription and expression of carboxysome genes have been reported previously in carbon-limited cells of Synechocystis sp. PCC 6803, and have been accounted for by a yet unknown mechanism of post-transcriptional regulation (Eisenhut, et al., 2007). It is possible that NtcA is involved in the expression of both ccmK and ccmL and could influence the results observed here (Su, et al., 2005).
The carboxymethylenebutenolidase (MAE 06360) that was identified in the differential protein display of toxic and non-toxic M. aeruginosa can also be involved in carbon metabolism. This enzyme, also known as diene lactone hydrolase, is part of the chlorocatechol degradation pathway, which ultimately produces 2-maleylacetate to be incorporated in the TCA cycle (Pathak and Ollis, 1990).
In terms of the relationship between carbon-fixation and microcystin synthesis, immunogold labeling experiments have shown that a proportion of the intracellular toxin is localized in proximity to carboxysomes and the possibility of the molecule acting as a inhibitor for RuBisCo in carbon-limiting conditions has been discussed (Gerbersdorf, 2006). However, the relationship between carboxysomes and microcystin is not exclusive, as the toxin is also associated with the thylakoid membranes (possibly the phycobilins), polyphosphate bodies and the nucleoplasm (Gerbersdorf, 2006, Juettner and Luethi, 2008, Young, et al., 2005).
2.4.3.4. Proteins involved in redox balance maintenance The association of microcystin with the thylakoids, in particular under high light, as well as reported observations that it binds divalent metal ions, have been a central argument to the hypothesis that this peptide is involved in the general oxidative stress response of M. aeruginosa (Gerbersdorf, 2006, Saito, et al., 2008, Utkilen and Gjolme, 1995). In the studies performed here, several proteins involved in redox balance were found to be down-regulated in the absence of microcystin, including NdhK and TrxM. NdhK is an essential subunit of the NADH dehydrogenase (NDH) complex, similar in function to the NAD(P)H:quinine oxidoreductase (NDH-1) in mitochondria and eubacteria (Prommeenate, et al., 2004). The isoforms of the NDH complex are involved in accepting electrons from reduced plastoquinone, respiration and CO2 uptake (Battchikova and Aro, 2007, Prommeenate, et al., 2004). This multitude of functions is 60 Protein expression diversity in M. aeruginosa possible due to the presence of several forms of the complex containing alternative D and F subunits, as well as an accessory carbon uptake (CUP) domain (Herranen, et al., 2004, Ogawa and Mi, 2007). The substrate specificity of the cyanobacterial NDH-1 complex is still unclear, with NADH, NADPH and reduced ferredoxin being considered as possibilities (Prommeenate, et al., 2004). The NdhK subunit is a soluble protein involved in connecting the membrane and peripheral domains of all NDH complexes and is present both in the thylakoid and plasma membranes (Battchikova and Aro, 2007). It is unclear how the changed expression of this particular NDH subunit would influence the metabolism of M. aeruginosa, but an ndhK knock-out strain of Synechocystis PCC 6803 demonstrated a decreased ability of cells to accumulate inorganic carbon (Pieulle, et al., 2000). Such a response, where NdhK expression is low in toxic strains, but CcmK and CcmL proteins are highly expressed is consistent with the expected high-carbon requiring phenotype of microcystin-producers discussed above.
Thioredoxin M expression was found to be elevated in toxic cells and further supports a difference in the mechanism for coping with alteration in the redox status of microcystin-producing and non-toxic cells. In high light conditions, NADPH levels increase as a result of active photosynthesis, leading to an increase in reduced thioredoxin and oxidation of its target proteins, which are then deactivated and thus protected from oxidative stress (Schuermann and Buchanann, 2008). The differential expression of TrxM and the TrxM-associated hypothetical protein in the toxic and non- toxic strains analysed here may be the result of altered NdhK activity and therefore, changes in photosynthetic electron flow. In Synechocystis sp. PCC 6803 thioredoxins interact with and modulate the activity of proteins involved in CO2 fixation, glycolysis, nitrogen metabolism and the oxidative stress response (Florencio, et al., 2006, Lindahl and Kieselbach, 2009). These processes are all tightly linked to photosynthetic activity, and accordingly, thioredoxin is placed under transcriptional regulation in response to light, where dark or photoinhibitory light intensities inhibit thioredoxin transcription (Navarro, et al., 2000). Although cyanobacterial thioredoxin M has not been studied in detail, the available literature supports a function of this protein in oxidative stress protection (Singh, et al., 2009). Thus, increased TrxM levels in toxic strains could be part of the protective mechanism that allows toxic M. aeruginosa to be better competitors in iron limitation or high light (Edwin, et al., 2007, Utkilen and Gjolme, 1995).
61 Chapter 2
The hypothetical protein MAE 06820 encoded by the open reading frame adjacent to thioredoxin M was found to follow the TrxM expression pattern. The two ORFs are separated by a conserved region of 34 nucleotides in both published genomes of M. aeruginosa PCC 7806 and NIES-843 and are co-transcribed, as shown on Figure 2.3.5. This apparent co-regulation at both the transcriptional and expression level could be a result of a thioredoxin or redox-related function of MAE 06820. Interestingly, none of the homologues of MAE 06820 in other cyanobacteria are encoded in open reading frames proximal to thioredoxin, suggesting that this protein may have a more divergent function in M. aeruginosa. Another hypothetical protein (MAE 27590) was also found to be down-regulated in non-toxic strains. The role of this protein has not been established yet, although it contains a putative flavin mononucleotide (FMN)-binding domain and as such, may also be involved in redox balance or photosynthesis.
2.4.4. NtcA regulation and microcystin synthesis Five of the proteins reported to be differentially expressed here, have been identified as targets of the global nitrogen regulator NtcA in cyanobacteria. These include NrtA, PII, CcmK, CcmL and TrxM (Su, et al., 2005). The expression of MAE 06820 is also likely to be influenced by NtcA activity when it is co-transcribed with TrxM. In conjunction with PII, NtcA is able to either activate or suppress gene transcription and is crucial for a balanced C:N metabolism in cyanobacteria (Jiang, et al., 1997, Su, et al., 2005). NtcA generally acts on its own promoter as a positive regulator (Ginn and Neilan, 2010, Olmedo-Verd, et al., 2008) and it was surprising to find that ntcA transcription remained unchanged in toxic strains, relative to non-toxic M. aeruginosa. However, this result is also consistent with the lack of differential expression at the protein level and could be explained by a change in the DNA-binding affinity of NtcA, rather than accumulation of this transcriptional regulator in microcystin producers. The lower levels of PII in the microcystin-producing cells would make more PipX available for binding to NtcA-DNA complexes, stimulating NtcA activity on promoters recognized by the global nitrogen regulator (Figure 2.4).
As discussed, the NtcA-binding sites present in the microcystin promoter and the differential expression of a number of NtcA-regulated genes in toxic and non-toxic strains of M. aeruginosa strongly suggest an effect of the toxin molecule on NtcA activity. The mcyS gene cluster promoter contains binding sites for two other transcription factors, the ferric uptake regulator Fur and the light-responsive Rca.
62 Protein expression diversity in M. aeruginosa
However, the results presented here did not include genes regulated by either Fur or Rca, although interaction and cross-regulation of all three transcription factors with the toxin cluster cannot be excluded. Instead, the data points to an involvement of microcystin in processes that ultimately influence the integration of photosynthesis with carbon and nitrogen metabolism. These could explain previous observations of toxic and non-toxic strains behaving differently under varying light intensities or in nutrient starvation (Dittmann, et al., 2001, Edwin, et al., 2007, Oh, et al., 2000, Sevilla, et al., 2008).
Figure 2.4.1. Proposed model for the involvement of microcystin and NtcA regulation in the differential expression of proteins in M. aeruginosa. Proteins in red are up-regulated in toxic strains relative to non-toxic strains, proteins in green are down- regulated. Boxes in grey have a putative function in this proposed model and have not been shown to be differentially expressed experimentally. Dashed lines represent processes that are down-regulated in this network. 2OG: 2-oxoglutarate; MCYST-LR: microcystin-LR.
63 Chapter 2
It is important to consider that the results obtained here reflect protein expression in the exponential growth phase and in nutrient-replete conditions. Therefore, it is highly likely that many more proteins would be differentially expressed under different nutrient and light regimes, in addition to proteins that could contribute to strain-specific adaptation in a particular environment regardless of the ability to synthesize microcystin.
2.4.5. Induction of toxicity in an inactive microcystin producing strain
Two of the strains used in this analysis, UWOCC MRC and UWOCC MRD were chosen due to their 16S rRNA sequence identity and the reported inability of UWOCC MRC to produce microcystin (Kaebernick, et al., 2001). However, when the cultures were screened for microcystin production, UWOCC MRC was found to have reverted to toxicity and was able to produce several different isoforms of microcystin compared to its parent strain (Figure 2.3.7 and Table 2.3.3). Inactive toxigenic strains such as UWOCC MRC have been reported to occur in approximately 3% of the bloom population in Microcystis and Planktothrix spp., and the reason for their inability to produce the toxin is unknown (Kurmayer, et al., 2004). Indeed, several insertions in the mcy gene cluster have been reported in M. aeruginosa strains, including UWOCC MRC and may affect the transcription of toxin genes (Roberts, 2008, Roberts and Neilan, 2010). These insertions, identified in the promoter region and mcyB, could be the result of past transposition events but do not seem to correlate with toxin production in M. aeruginosa strains. Nevertheless, it cannot be excluded that these regions are involved in a yet unknown post-transcriptional regulation of the mcyS gene cluster or in determining the toxin isoform to be produced. This is suggested by the ability of UWOCC MRC to revert back to toxicity, without any associated mutations in the mcy cluster or other phenotypic changes.
Due to the ability of UWOCC MRC to revert to toxin production, its proteome was compared to that of UWOCC MRD using the nSAF method. The majority of the proteins found to be significantly different between the two strains were involved in carbon metabolism and may result from adaptation to the ability of UWOCC MRC to form gas vesicles and colonies, as well as differences in light-harvesting reflected by the altered pigmentation and phycobilisome structure in UWOCC MRC. The observed differences between the two strains were not induced by microcystin synthesis in UWOCC MRC and therefore a possible mechanism for the induction of toxicity in
64 Protein expression diversity in M. aeruginosa inactive toxin producers such as the strains used here remains to be identified. Some of the observed differences may also be associated with the isoform of the toxin that is produced, since some authors have found that microcystin isoforms shift with changing nutrient conditions (Oh, et al., 2000). A study by Wilson et al. (2005) speculated that solid media selects for non-toxic strains, whereas liquid media favours toxin production. Since our strains are maintained in liquid BG11, prolonged culturing in these conditions may have induced the toxicity of UWOCC MRC over time, although the actual event that has caused the reversal of UWOCC MRC to toxicity remains unclear.
2.5. Conclusions Proteome analysis of M. aeruginosa revealed that the strains studied here have highly variable protein expression profiles, which may be a result of the adaptation of these organisms to the particular environmental niche that they occupy. No qualitative differences were found between toxic and non-toxic strains, and cluster analysis of protein expression did not segregate strains according to their toxicity. An interesting finding was the ability of a previously non-toxic strain UWOCC MRC to produce several isoforms of microcystin. This is the first report of a non-toxic strain of M. aeruginosa reverting to toxicity and implies that microcystin synthesis is not a static trait in UWOCC MRC and possibly other inactive toxigenic M. aeruginosa strains.
In addition, protein expression in toxic and non-toxic strains of M. aeruginosa was quantified using the label-free nSAF method. Nine proteins, involved in C-N metabolism and cellular redox balance, as well as two hypothetical proteins were found to be differentially displayed based on microcystin production. The expression of several of these proteins is regulated by the global nitrogen transcriptional regulator NtcA, which also binds the mcyS promoter and may reflect a microcystin-NtcA cross- regulatory mechanism. This model suggests that the basal metabolism of cyanobacterial cells affects hepatotoxin synthesis, and may affect their response to nutrient limitation and oxidative stress, a hypothesis which will be investigated further in following chapters of this thesis.
65 Chapter 2
66
Chapter 3 Transcriptional analysis of the iron stress response in Microcystis aeruginosa
67
68 Gene transcription during iron stress
3.1. Introduction
Iron is an essential nutrient for cyanobacteria, due to its involvement in chlorophyll-a synthesis, respiration, photosynthesis and nitrogen fixation. These processes lead to increased requirements for iron in cyanobacteria, compared to heterotrophic microorganisms, and the need for efficient uptake and storage of this element.
At circumneutral pH, the highly bioavailable ferrous iron can be rapidly oxidized to ferric iron that subsequently forms poorly soluble (oxy)hydroxides or biologically unavailable complexes with natural organic compounds. Thus, the availability of this metal is limited in the aquatic environment (Ferreira and Straus, 1994, Latifi, et al., 2008). When iron limitation in cyanobacteria becomes severe, excess light energy cannot be utilized by the modified phycobilisomes and leads to the formation of reactive oxygen species (ROS) through the Fenton reaction and oxidative stress (Latifi, et al., 2008, Michel and Pistorius, 2004).
The increased growth of the bloom-forming Microcystis aeruginosa species when supplemented with iron, and the observed changes in toxicity at different iron concentrations have led to several studies on the effect of metal availability on M. aeruginosa growth and microcystin synthesis (Imai, et al., 1999, Martin-Luna, et al., 2006b, Nagai, et al., 2007, Sevilla, et al., 2008, Utkilen and Gjolme, 1995, Xing, et al., 2007). In particular, differences have been observed in the growth of toxic and non- toxic strains of M. aeruginosa, with microcystin-producers appearing to remain viable for longer periods under conditions of iron stress. This supports the hypotheses that microcystin may allow toxic cells to take up or store iron more efficiently (Sevilla, et al., 2008, Utkilen and Gjolme, 1995).
Early studies observed weak binding of the toxin to divalent metal ions such as Cu2+, and Zn2+ in vitro (Humble, et al., 1997). This putative metal-binding property of microcystin and the increase in toxin production during iron limitation, together with the fact that iron regulates the peptide synthetase systems in other heterotrophic bacteria, led to the proposal that microcystin synthesis may also be regulated by iron availability (Drechsel and Jung, 1998, Martin-Luna, et al., 2006a). In addition, the cyclic structure of microcystin and cyanopeptolin, a non-ribosomally synthesised peptide with unknown function produced by M. aeruginosa, bear some resemblance to
69 Chapter 3 peptide siderophores produced by other organisms. However, siderophores are characterized by having strong ferric iron chelating capacity and are actively transported through the cell membrane (Drechsel and Jung, 1998). While a putative microcystin- associated ABC transporter, McyH, has been identified (Pearson, et al., 2004), a strong affinity of microcystin for iron and active transport of the toxin from the cell are two properties that have not been confirmed for this non-ribosomal peptide. A recent study in M. aeruginosa PCC 7806 revealed that in this strain superoxide-mediated reductive iron uptake, but not siderophore-mediated iron uptake is likely to play a significant role in iron transport, at least during short-term iron limitation (Fujii, et al., 2010). Non- siderophore iron transport in cyanobacteria is facilitated by an iron-binding protein, the ferric uptake ABC-transporter (Fut) (Katoh, et al., 2007). A ferrous iron transporter (FeoB) homologus to the ferrous transporters in heterotrophic bacteria has an accessory iron uptake function, and is only expressed when iron stress is severe (Katoh, et al., 2007).
The link between toxin synthesis and iron availability could be provided by the binding of the ferric uptake regulator FurA to the mcyS gene cluster, providing transcriptional control over toxin synthesis (Martin-Luna, et al., 2006a, Martin-Luna, et al., 2006b). Since FurA generally acts as a transcriptional repressor in an iron-replete environment, its control on the toxin promoter would lead to increased microcystin production in iron-deficient cells, consistent with the findings reported in the literature (Sevilla, et al., 2008, Utkilen and Gjolme, 1995).
Similar to other cyanobacteria, the M. aeruginosa genome encodes for three Fur homologues (Frangeul, et al., 2008, Kaneko, et al., 2007). Whereas FurA has been studied previously, the function of the FurB and FurC proteins and their regulatory targets in cyanobacteria remain unclear (Hernandez, et al., 2004). All three proteins are expressed in Anabaena sp. PCC 7120 and FurB was recently shown to have a DNA- protective function (Hernandez, et al., 2004, Lopez-Gomollon, et al., 2009). The studies of cyanobacterial Fur regulation have so far been limited to toxin-producing strains only, and the involvement of these transcriptional regulators in the absence of microcystin has not been investigated (Martin-Luna, et al., 2006b).
In addition, several genes involved in both iron uptake and the oxidative stress response have been identified in other freshwater cyanobacteria, but are not present in one or both of the available M. aeruginosa genomes. These include futA2, idiB, isiA, isiB and 70 Gene transcription during iron stress irpA (Badarau, et al., 2008, Michel and Pistorius, 2004, Nodop, et al., 2008). Such genomic differences may indicate that strains of the same cyanobacterial species have different strategies for adaptation to conditions of iron stress, thus affecting their ecotype and ability to dominate particular stages of bloom development. There are currently no detailed studies on the iron uptake mechanisms in different M. aeruginosa strains or the degree of involvement of microcystin in iron acquisition.
In this chapter, three M. aeruginosa strains were grown in conditions of moderate and severe iron stress. The effect of iron limitation on the growth, microcystin production and transcription of genes involved in oxidative stress, toxin synthesis and iron homeostasis in these cyanobacteria was determined. Comparison of this process in the toxic PCC 7806 strain with the iron-stress response in the wild-type non-toxic strain PCC 7005 and the genetically-engineered PCC 7806 mcyH- strain revealed that toxin production influences the remodeling of the photosynthetic machinery and iron uptake, possibly by disrupting Fur regulation in these bloom-forming cyanobacteria.
3.2. Methods
3.2.1. Microcystis strains and culture growth
Culturing was performed by Dr. Manabu Fujii (School of Civil Engineering, UNSW). For cell culturing, a modified Fraquil medium (Fraquil*), pH 8, as described by Andersen (2005) (Appendix B, Table A.2.1). The EDTA concentration was kept constant at 26 μM, while the iron concentration varied from 10 nM to 1000 nM. Briefly, triplicate batch cultures of M. aeruginosa PCC 7806, PCC 7005 and PCC 7806 mcyH- were grown in Fraquil* media containing 10 μM Fe, 1000 nM Fe, 100 nM Fe (moderate limitation) or 10 nM Fe (severe limitation). All polycarbonate culture vessels (Nalgene, Rochester, NY) were acid-washed with 5% (v/v) nitric acid prior to use. Batch cultures of M. aeruginosa were maintained at 27oC on a 14:10 light/dark cycle (90 )mol photons m-2 s-1) supplied by cool white fluorescent lamps. Cell numbers were counted using a Neubauer haemocytometer. Experiments were performed using cells harvested during day 7 (exponential growth phase) and day 14 (stationary growth phase) at densities between 3 – 20 x 106 cell ml-1.
3.2.2. RNA extraction
Twenty milliliters of culture were harvested by centrifugation at 4000 x g in the exponential and stationary growth phase and RNA was extracted immediately from cell 71 Chapter 3 pellets. Cells were lysed using TRIzol (Invitrogen) according to the manufacturer’s instructions. After chloroform extraction, RNA was collected from the aqueous layer and precipitated in isopropanol. The RNA pellet was washed twice with 75% ethanol, dissolved in 50 μl DEPC-treated water (Invitrogen), treated with 6 U Turbo DNase (Ambion) for 3 h at 37oC, and re-extracted with TRIzol as described above. The successful digestion of DNA was assessed by PCR with primers targeting 16S rRNA gene (27F and 809R) (Jungblut, et al., 2005). Total RNA concentration was measured using a Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies).
3.2.3. cDNA synthesis
Reverse transcription was performed using 100 ng total RNA with a FirstStrand cDNA kit (Marligen Bioscience) according to the manufacturer’s instructions. Reaction conditions were 22oC for 5 min, 42oC for 90 min and 85oC for 5 min. For secondary metabolite genes (mcnC, mcyA and mcyH) 1 μg of total RNA was used for the reverse transcription reaction due to their low expression levels.
3.2.4. Quantitative real-time PCR (qRT-PCR)
Specific primers were designed to amplify genes involved in the iron stress response and toxin biosynthesis (Table 3.2.1). The RNA polymerase subunit C (rpoC1) and the 16S rRNA genes were considered as housekeepers, and rpoC1 was chosen for further study as its transcription appeared more stable across the different nutrient regimens used. Transcript levels were quantified by qPCR using the Rotor-Gene 3000 system (Corbett). Reactions were performed in a total volume of 25 μl using 1 μg cDNA, 10 pmol of forward and reverse primer and the Platinum SYBR Green qPCR supermix UDG kit (Invitrogen). Two-step cycling was performed with an initial hold of 60oC for 2 min and 95oC for 2 min, followed by 40 cycles at 95oC for 15 s and 60oC for 30 s. The efficiency of amplification for each primer set was determined using standard cDNA curves and calculated according to the equation E = 10 [-1/slope] (Pfaffl, 2001). The efficiency of all primers was in the range of 90-100%. Transcript levels were normalized to rpoC transcription and calculated relative to values for cells grown in 1000 nM Fe using the 2-Ct method previously described (Pfaffl, 2001). All analyses were performed using biological and technical triplicates.
72 Gene transcription during iron stress
Table 3.2.1. Primers used in quantitative real-time PCR. The accession numbers for the respective contig in the PCC 7806 genome database (NCBI, database version 01/11/07) are given. Primer Accession Gene Sequence (5’-3’) Target Reference name number symbol Oxidative stress isiAR acgactggtgggcaggaaat Iron stress induced AM778935 isiA This study isiAF gattaaagcagcttgggcaac protein A (CP43’) Iron-binding and acquisition rtfutABCF tgtttggtcggggaaaatta Ferric uptake ABC CAO87134.1 futA This study rtfutABCR tccaccccagggtaatgata transporter feoBR ccgccgcaaatagggcataa Ferrous uptake AP009552 feoB This study feoBF gattgatgcgtttggtgggat transporter Ferric uptake regulator (Fur) homologues fur3R accctcggccaattccaact CAO87979 furA Ferric uptake regulator This study fur3F caatcatctcagtgccgaaga fur2F accttaaatcgggcagtatcc AM778941.1 furB Ferric uptake regulator This study fur2R ctgcactgcaccctgtaattt fur1F gtgactgtgggaatcgctga CAO90328.1 furC Ferric uptake regulator This study fur1R aacacctatcagcgcgagaaa Non-ribosomal peptide synthesis qmetF ttattccaagttgctcccca Microcystin synthesis (Saker, et al., CAO90227.1 mcyA qmetR ggaaatactgcacaaccgag methyltransferase 2005) rtmcyH2F ttgtcttcgctccagcctat Putative ABC CAO90232.1 mcyH This study rtmcyH2R ggccgacgaaaattcagata transporter mcnCF agctaaaacggcaaaggaca CAO90637.1 mcnC Cyanopeptolin This study mcnCR ccaattgcctccaaagttgt Transcription reference rpoC1F cctcagcgaagatcaatggt RNA polymerase (Ginn and CAO88583.1 rpoC1 rpoC1R ccgtttttgccccttacttt gamma subunit Neilan, 2010)
3.2.5. Microcystin detection
Concurrently with RNA extraction, total microcystin was extracted by diluting 1 ml of culture to a final concentration of 70% (v/v) methanol. Toxin content normalized to cell number was determined by the protein phosphatase 2A inhibition assay (PPIA) as described previously (Carmichael and An, 1999).
3.2.6. Microscopy
For laser-scanning confocal microscopy, M. aeruginosa PCC 7806 cells grown in 10 nM Fe for four days were washed twice with fresh medium and fixed in 2% paraformaldehyde in PBS for 1 h. The cells were placed on slides coated with 0.1% PEI, allowed to settle and incubated for 5 min in 0.5% (v/v) Triton-X 100 (Amresco, Solon, OH) in PBS. Blocking was performed in 10% (w/v) foetal calf serum (FCS) for 30 min, after which the cells were washed three times in PBS and incubated for 1 h in protein G-purified anti-IdiA antibody (Prof. Elfriede Pistorius, University of Bielefeld) diluted 1:100 in PBS. The cells were washed to remove unbound antibody and
73 Chapter 3 incubated for 1 h with a 1:400 dilution of anti-rabbit Alexa fluor 488 (Molecular Probes, Eugene, Oregon). A secondary antibody control with the anti-rabbit Alexa- labelled antibody, but no anti-IdiA IgG, was also included. The cells were washed and samples were imaged using an Olypmpus UplanApo 100X oil immersion objective (N.A.1.35). Excitation of Alexa 488 using the argon laser (488 nm), the signal was detected at 515 nm. All images captured with Olympus Fluoview software version 4.
For transmission electron microscopy (TEM), M. aeruginosa PCC 7806 cells were grown in 10 nM Fe or 1000 nM Fe for 7 days as described above (Section 3.2.1) and washed in 50 mM phosphate buffer, pH 7.0. For ultrastructure studies, the samples were fixed in 2.5% glutarladehyde for 4 h. The cells were washed in PBS and dehydrated with a graded ethanol series prior to infiltration with Pure-White Resin (ProSciTech, Thuringowa, QLD, Australia) over 72 h and polymerization at 60oC for a further 72 h. Ultrathin sectioning and mounting on copper TEM grids was carried out by Debra Birch (Microscopy Facility, Macquarie University). Ultrastructure observations were performed after staining with 7.7% uranyl acetate (Sigma, St Luis, MO, USA) for 30 min, followed by a Raynold’s solution of lead citrate (Sigma).
3.3. Results
3.3.1. Growth of M. aeruginosa under iron limitation
The M. aeruginosa strains used here did not exhibit any difference in growth or pigmentation when supplied with a total iron concentration of either 1000 nM or 10 μM, reaching similar cell numbers by the end of the experiment (Figure 3.3.1). Therefore, 1000 nM Fe was chosen as the highest total iron concentration for subsequent qRT-PCR analysis. In addition, for the first six days of culture, the iron concentration in the media did not affect the cell numbers in M. aeruginosa PCC 7806 and PCC 7005. For all strains, 10 nM Fe was sufficient to maintain the growth of only 106 cells ml-1 and caused rapid chlorosis and early entry to stationary phase. In contrast, in 100 nM Fraquil* alterations in the growth of the toxic M. aeruginosa PCC 7806 were observed, with a lower cell yield compared to iron-replete conditions. The non-toxic strains did not show such a pronounced difference in growth when cultured in the presence of either 100 nM or 1000 nM Fe (Figure 3.3.1B and 3.3.1C). Despite this, the cells of PCC 7005 in limited-iron culture were noticeably chlorotic by the end of culturing, whereas bleaching in PCC 7806 was less pronounced (Figure 3.3.1D).
74 Gene transcription during iron stress
Similarly, the cells of mcyH- were chlorotic when grown in Fe concentrations less than 1000 nM and had a slower growth rate throughout the culturing period (Figure 3.3.1C).
8 A. 10 nM Fe 100 nM Fe 7 1000 nM Fe 10 000 nM Fe 6 log cell/ml log 5
4 0 2 4 6 8 10 12 14 Day
8 B. 10 nM Fe 100 nM Fe 7 1000 nM Fe 10 000 nM Fe 6 log cell/ml log 5
4 0 2 4 6 8 10 12 14 Day
8 C. 10 nM Fe 100 nM Fe 7 1000 nM Fe
6 log cell/ml log 5
4 0 2 4 6 8 10 12 14 16 18 Day
D.
Figure 3.3.1. Growth of Microcystis aeruginosa at different iron concentrations. A. PCC 7806, B. PCC 7005 and C. PCC 7806 mcyH-. D. Cells grown in Fraquil* media, from left to right: PCC 7806 in 10 nM Fe, PCC 7806 1000 nM Fe, PCC 7806 mcyH- in 10 nM Fe. For all strains, triplicate batch cultures were grown in Fraquil* media and cell counts were performed with a hemocytometer. Asterisks denote time points for RNA extraction. Figure reproduced with permission from Dr. Manabu Fujii.
75 Chapter 3
3.3.2. Microcystin synthesis by iron-starved M. aeruginosa PCC 7806
Within the first week of culturing (day 7), the total microcystin concentration (normalized to cell number) in the iron-starved cultures (10 nM Fe) increased markedly when measured by protein phosphatase inhibition assay (PPIA) (Figure 3.3.2). In the 100 nM Fe culture, there was a slight, but significant drop in microcystin synthesis relative to 1000 nM Fe. The toxin content relative to cell number decreased in stationary phase in all growth conditions analysed. As expected, no microcystin was detected by PPIA in the non-toxic PCC 7005 or the mutant strain PCC 7806 mcyH-.
0.15 exponential * stationary 0.10
0.05 * * [MC-LR] nM/cell [MC-LR]
0.00 10 100 1000 total iron concentration (nM)
Figure 3.3.2. Microcystin production by M. aeruginosa PCC 7806 under iron- limitation. The total toxin concentration (intra- and extracellular) was measured by protein phosphatase 2a inhibition assay. Asterisks denote statistically significant difference after a two-tailed t-test (p<0.05, n = 3) compared to 1000 nM Fe.
3.3.3. Transcriptional analysis of the iron stress response in M. aeruginosa
Quantitative real-time PCR was used to compare the response to iron stress in the toxic M. aeruginosa sp. PCC 7806 and the non-toxic PCC 7005. The non-toxic mutant PCC 7806 mcyH-, in which the gene encoding a putative ABC-transporter McyH has been insertionally inactivated, was used to discriminate between differences in transcription caused by strain-to-strain variation or by microcystin production. The changes in transcription are summarized in Table 3.3.1.
76 Gene transcription during iron stress
Table 3.3.1. Transcription changes in selected genes of three M. aeruginosa strains grown under conditions of iron starvation (10 nM Fe) or iron limitation (100 nM Fe). Up- and down-regulation for each gene at the exponential and stationary growth phases are reported relative to transcript levels of cells grown in 1000 nM Fe (iron- replete). Values in bold represent significant changes after a two-tailed t-test (p < 0.05, n=3), n/d: no transcript detected.
PCC 7806 PCC 7005 PCC 7806 mcyH- exponential stationary exponential stationary exponential stationary 10 100 10 100 10 100 10 100 10 100 10 100 isiA 35.92 9.55 152.74 7.63 13.67 12.09 95.00 86.42 1.18 4.94 101.99 41.37 futA 9.31 1.16 5.54 -1.79 -4.19 -1.80 2.63 -1.81 -4.26 1.17 1.64 -2.00 feoB 3.24 1.92 -14.00 -20.00 -1.78 -2.94 -5.80 -2.90 -4.42 -1.29 9.27 1.27 mcyA -1.24 -1.16 -117.36 -2.28 -4.39 -1.14 12.07 -1.52 -2.37 -6.29 -6.19 -16.67 mcyH -10.12 -2.15 -1003.61 -18.5 n/d n/d n/d n/d n/d n/d n/d n/d mcnC -7.92 -3.81 -957.64 -14.42 -3.71 -1.18 28.18 -1.77 -4.75 -1.58 1.15 1.76 furA 2.32 -1.05 -4.44 -4.60 -1.10 1.31 1.12 2.07 -1.04 -1.46 -1.32 -1.54 furB 47.51 -1.68 -3.95 -16.90 14.50 -1.57 -2.32 1.62 -18.65 -4.37 -1.35 -9.05 furC 2.73 2.19 -6.25 1.51 3.15 1.79 -2.55 2.26 -2.78 -4.75 1.32 -2.13
3.3.3.1. Genes involved in oxidative stress
Despite the lack of an isiA homologue in the M. aeruginosa NIES-843 genome, isiA transcripts were detected in all strains studied here (Table 3.3.1, Figure 3.3.3). When grown in 10 nM Fe, transcription was up-regulated in the toxic strain after a week of iron starvation (exponential growth) and continued to increase in stationary phase to reach levels 150-fold higher than in 1000 nM Fe (Table 3.3.1, Figure 3.3.3). In contrast, when grown in 100 nM Fe, the toxin-producing cells exhibited a less pronounced change in isiA transcript relative to 1000 nM Fe, a response which correlated with the lack of bleaching of these cells (Table 3.3.1, Figure 3.3.3). On the other hand, the non- toxic strain PCC 7005 appeared to be equally affected by iron limitation (100 nM) and iron starvation (10 nM), as isiA levels were similar in both growth conditions. However, the maximum levels of isiA in this strain reached only 75% of those in PCC 7806. Similarly, the mcyH- mutant strain did not respond to initial iron starvation, but isiA levels increased in an Fe-dependent manner at both 10 nM and 100 nM by the time the cells had entered in stationary phase (Table 3.3.1, Figure 3.3.3).
77 Chapter 3
A. isiA 10 nM Fe B. isiA 100 nM Fe 175 * 175 exponential 150 150 stationary 125 * * 125 100 100 * 75 75 * 50 * 50 * 25 25 * * * * fold change transcription 0 fold change transcription 0 7806 7005 mcyH- 7806 7005 mcyH-
D. C. futA 10 nM Fe futA 100 nM Fe
15 15 exponential * 10 10 stationary * 5 * 5 * * * * * 0 0
-5 -5
fold change transcription -10 fold change transcription -10 7806 7005 mcyH- 7806 7005 mcyH - E. feoB 10 nM Fe F. feoB 100 nM Fe 15 15 * exponential stationary 5 ** 5 ** *
-5 -5
-15 -15 fold change transcription fold change transcription -25 7806 7005 mcyH- -25 7806 7005 mcyH-
Figure 3.3.3. Transcription changes in genes involved in oxidative stress and iron uptake in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM). Up- and down-regulation for each gene at the exponential and stationary growth phases are reported relative to transcript levels of cells grown in 1000 nM Fe (iron-replete). Asterisks represent significant changes after a two-tailed t-test (p < 0.05, n=3).
3.3.3.2. Iron-binding and acquisition genes
When the transcription of genes involved in iron transport was investigated, the transcript levels of the putative ferric-uptake ABC transporter FutA were found to be increased in the exponential phase in toxic cells grown in 10 nM Fe (Table 3.3.1, Figure 3.3.3). However, at 10 nM Fe, both of the non-toxic strains exibited a delayed response in the up-regulation of futA transcription (Table 3.3.1, Figure 3.3.3). On the other hand, in 100 nM Fe cultures, the change in futA transcription did not reach levels significantly different to those of cells grown in 1000 nM Fe, with the exception of PCC 7806 mcyH- (Table 3.3.1, Figure 3.3.3). For the ferrous iron transporter feoB, transcription was significantly down-regulated in stationary phase cells of PCC 7806 and PCC 7005, but increased in the mcyH- strain in 10 nM Fe (Table 3.3.1, Figure 3.3.3). In contrast, feoB
78 Gene transcription during iron stress transcription decreased in all strains grown in 100 nM Fe, although to a lesser extent in the non-toxic strains compared to PCC 7806 (Table 3.3.1, Figure 3.3.3).
3.3.3.3. Non-ribosomal peptide synthesis genes
The transcription of two genes in the mcy gene cluster – mcyA, a methyltransferase involved in the first step of microcystin synthesis, and mcyH, a putative ABC- transporter essential for toxin synthesis, was also investigated (Table 3.3.1, Figure 3.3.4).
A. mcyA 10 nM Fe B. mcyA 100 nM Fe 20 20 * exponential 10 10 stationary ** * ** 0 0
-10 -10
-20 -20 fold change transcription -120 transcription change fold -120 7806 7005 mcyH- 7806 7005 mcyH-
C. mcyH 10 nM Fe D. mcyH 100 nM Fe 50 50 exponential stationary 25 25
** n/d n/d * n/d n/d 0 0
-25 -25 fold transcription change -1500.0 fold change transcription -1500 7806 7005 mcyH 7806 7005 mcyH-
E. mcnC 10 nM Fe F. mcnC 100 nM Fe 40 40 * exponential 30 30 stationary 20 20 10 10 ** * * ** * 0 0 -10 -10 -20 -20 fold change transcription fold change transcription change fold -1000 -1000 7806 7005 mcyH 7806 7005 mcyH-
Figure 3.3.4. Transcription changes in genes involved in non-ribosomal peptide synthesis in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM). Up- and down-regulation for each gene at the exponential and stationary growth phases are reported relative to transcript levels of cells grown in 1000 nM Fe (iron-replete). Asterisks represent significant changes after a two-tailed t-test (p < 0.05, n=3).
Transcription of mcyA did not follow the changes in microcystin concentration observed in the PPIA (Figure 3.3.2). Instead, there was little change in the transcription of this
79 Chapter 3 gene during exponential growth phase, but a large decrease in mcyA transcript when the iron-starved toxic cells entered stationary phase. Unexpectedly, mcyA transcript was detected in both non-toxic strains, despite their inability to synthesise microcystin (Table 3.3.1, Figure 3.3.4). In contrast to mcyA, mcyH transcription was only observed in the toxic PCC 7806, but not the non-toxic PCC 7005 or mcyH- mutant cells, providing further evidence for the essential role of this gene product in toxicity (Table 3.3.1, Figure 3.3.4).
Transcription of the mcnC gene involved in the synthesis of cyanopeptolin, another non-ribosomal peptide with a siderophore-like structure, was also quantified (Table 3.3.1, Figure 3.3.4). The transcription profile followed a similar pattern to that of mcyA for both PCC 7806 and PCC 7005. Although there was a slight fluctuation in mcnC levels in the mcyH- strain, it seems that, compared to the wild-type strain, inactivating a microcystin-synthesis gene affected the regulation of transcription of mcnC in cells in stationary growth (Table 3.3.1, Figure 3.3.4).
3.3.3.4. Transcription of the ferric uptake regulator (fur) homologues
FurA is an iron-responsive transcriptional regulator of isiA and several other genes involved in iron metabolism and, putatively, the mcy cluster (Martin-Luna, et al., 2006b, Yousef, et al., 2003). In both the toxic and toxin-mutant strains, furA transcription was reduced at 10 and 100 nM Fe by day 14, although the response was not as pronounced in the mcyH- mutant (Table 3.3.1, Figure 3.3.5). In fact, inactivating mcyH seemed to abolish the sensitivity of furA transcription for iron as there was little difference between the transcript levels observed at 10 nM and 100 nM Fe. However, the naturally occurring non-toxic strain, PCC 7005, showed the opposite response to PCC 7806 in all instances, with transcript levels increasing slightly in stationary phase.
The transcription of the other two Fur homologues encoded in the Microcystis aeruginosa genome was also analysed under conditions of iron stress (Table 3.3.1, Figure 3.3.5). At 100 nM Fe, furB, a putative sensor of oxidative stress and a DNA- protecting protein, showed reduced transcription in both PCC 7806 and mcyH-, whereas in PCC 7005 this gene did not show a significant change relative to iron-replete cultures (1000 nM Fe). At 10 nM, furB transcription was increased in both wild-type strains, but was down-regulated for mcyH-. The transcription profile of furC, a putative Zn-uptake
80 Gene transcription during iron stress regulator, was completely reversed when the wild-type PCC 7806 was compared to the mcyH-.
A. furA 10 nM Fe B. furA 100 nM Fe 8 8 exponential stationary 4 4 ** ** 0 0
-4 -4 fold change transcription
fold change transcription -8 -8 7806 7005 mcyH- 7806 7005 mcyH-
C. D. furB 100 nM Fe furB 10 nM Fe 60 60 exponential * 50 stationary 40 40 30 20 20 * 10 *** * * 0 0 -10 fold change transcription -20 fold change transcription -20 7806 7005 mcyH- 7806 7005 mcyH-
E. furC 10 nM Fe F. furC 100 nM Fe 8 8 exponential stationary 4 * 4 * ** 0 0
-4 -4 fold change transcription change fold fold changetranscription -8 -8 7806 7005 mcyH- 7806 7005 mcyH-
Figure 3.3.5. Transcription changes in ferric uptake regulator (fur) genes in three M. aeruginosa strains grown in iron starvation (10 nM) or iron limitation (100 nM). Up- and down-regulation for each gene at the exponential and stationary growth phases are reported relative to transcript levels of cells grown in 1000 nM Fe (iron- replete). Asterisks represent significant changes after a two-tailed t-test (p < 0.05, n=3).
3.3.4. Sub-cellular localization of FutA
Since transcription of the futA gene was found to differ significantly between iron- starved toxic and non-toxic cells, it was of interest to see whether the function of this protein in M. aeruginosa is similar to either a periplasmic iron transporter or a photosystem II protective protein (IdiA) that has been reported for the two FutA homologues in Synechocystis sp. PCC 6803 (Koropatkin, et al., 2007, Lax, et al., 2007).
81 Chapter 3
A.
B.
C.
Figure 3.3.6. Localisation of the FutA homologue in iron-starved M. aeruginosa cells by laser scanning confocal microscopy. All strains were grown for 7 days in Fraquil* media supplemented with 10 nM Fe. A. PCC 7806, B. PCC 7806 mcyH-, C. PCC 7005. Green fluorescence is from anti-rabbit Alexa Fluor 488 conjugate bound to anti-FutA IgG (left), red fluorescence is from autofluorescence in the 600-700 nm region (centre). In each instance, the panel on the right shows the merged DIC and fluorescence images. The scale bar is 5 μM.
Confocal laser scanning microscopy revealed that the majority of anti-IdiA binding occurred in close association with the periphery of the cells, but not the autofluorescent thylakoids (Figure 3.3.6). No labeling was observed in the mcyH- strain, which also had altered thylakoid membranes, in particular in dividing cells (Figure 3.3.6C). Not all
82 Gene transcription during iron stress cells in the PCC 7806 population labeled with anti-IdiA antibody and may not express the FutA homologue (Figure 3.3.7). The secondary antibody control did not show fluorescence originating from non-specific Alexa-Fluor 488 labeling (Figure 3.3.8).
Figure 3.3.7. Confocal image of FutA in iron-starved M. aeruginosa PCC 7806 cells. A Z-stack image of representative PCC 7806 cells grown in 10 nM Fe Fraquil* for 7 days. The DIC image is overlayed with autofluorescence signal in the 600-700 nm region (red), and fluorescence is from anti-rabbit Alexa Fluor 488 conjugate (green) bound to anti-FutA IgG. The scale bar is 10 μm.
Figure 3.3.8. Secondary antibody control for confocal microscopy. Iron-starved (10 nM Fe) M. aeruginosa cells were treated with Alexa Fluor 488 conjugate (green, left panel), autofluorescence (red, centre) and DIC images were also acquired.
3.3.5. Ultrastructure of iron-starved M. aeruginosa PCC 7806 cells
Since the cells of PCC 7806 did not show the characteristic chlorosis associated with iron starvation in cyanobacteria (Figure 3.3.1D), TEM was performed on cells grown in
83 Chapter 3
10 nM and 1000 nM Fe. The iron-starved cells exhibited an increased number of cellular inclusions, including carboxysomes, polyhydroxyalkanoate granules and polyphosphate bodies, but no disorganization of the thylakoid membranes (Figure 3.3.9).
A. B.
TM PH PH
C TM C P P
Figure 3.3.9. TEM ultrastructure images of representative M. aeruginosa PCC 7806 cells grown in different iron availability for 7 days (exponential growth). A. Cells grown in 10 nM Fe. B. Cells grown in 1000 nM Fe. Other bacteria are also visible. The scale bar is 0.5 μM. C: carboxysome; P: polyphosphate body; PH, polyhydroxyalkanoate granule; TM: thylakoid membranes.
3.4. Discussion
Microcystin synthesis has been proposed to provide a selective advantage to toxic M. aeruginosa cells under iron-limited conditions (Utkilen and Gjolme, 1995). Such properties of this cyclic peptide may contribute to the long-term survival of microcystin-producing cyanobacteria in water bodies and the prevalence of toxic strains during the early stages of bloom formation by both M. aeruginosa and Planktothrix agardhii (Briand, et al., 2008, Edwin, et al., 2007). The physiological basis behind the differential response of toxic and non-toxic strains to iron stress remains largely unknown (Briand, et al., 2008, Sevilla, et al., 2008, Utkilen and Gjolme, 1995). This is highlighted by several studies investigating toxicity in iron limitation have found conflicting results with regards to the changes in microcystin synthesis. The
84 Gene transcription during iron stress discrepancies in the literature are usually accounted for by differences in experimental design, in particular, iron concentration, types of metal-binding chelators present in the growth media and growth conditions used, making it difficult to compare the conclusions made by different groups. In our experiments, the cells were grown in Fraquil* medium, which allows for a higher level of control of iron speciation and the associated iron bioavailability. This is due to the presence of a single chemically well- defined iron chelator (EDTA), compared to the widely used BG11 that contains both EDTA and citrate as ligands, resulting in complex iron speciation and the associated bioavailability and transformations between the various species. These may be induced by the bacteria themselves and/or external factors, such as light (Barbeau, et al., 2001, Fujii, et al., 2010).
Importantly, the majority of studies attempting to link iron availability to toxin synthesis use a single strain, such as the model toxic PCC 7806, but do not discuss the natural variation that may be present in M. aeruginosa strains in the environment (Imai, et al., 1999, Kosakowska, et al., 2007, Nagai, et al., 2007, Sevilla, et al., 2008, Xing, et al., 2007). Since a large number of differences were observed in the transcription of genes involved in the iron stress response between the two wild-type strains PCC 7806 and PCC 7005 in the qRT-PCR performed here, the mutant strain PCC 7806 mcyH- was employed in order to explore the contribution of microcystin production rather than intrinsic strain-to-strain variability. The differences in the iron stress response of these strains are summarized in Figure 3.4.1 and will be discussed below.
Current research on the process of microcystin synthesis regulation by iron availability has been limited to in vitro binding studies of the transcriptional regulator Fur to the mcy promoter, and the iron-responsive transcription of a single gene in the mcy cluster, mcyD. In this chapter, iron starvation (10 nM) was found to induce toxin production to a significant level in cells grown in the EDTA-buffered Fraquil* medium, but this was not the case with the transcription of either mcyA or mcyH. The lack of correlation between mcyA transcription and microcystin content, contradicts previous observations by Sevilla et al. (2008), where mcyD transcripts increased in conjunction with toxin content. The decrease in mcyA transcription may be the result of the growth stage in which the cells were harvested. The transcription of mcyE was found to be down- regulated once the cells had passed mid-exponential growth (Rueckert and Cary, 2009), consistent with the results presented here. A post-transcriptional regulatory process
85 Chapter 3 acting on parts of, or the entire mcyS cluster is supported by the identification of mcyA transcripts in the non-toxic strains PCC 7005 and mcyH-. Similar results have been reported previously in the mcyA- and mcyB- mutants which, despite being non-toxic were still able to express McyH, albeit at a reduced level (Pearson, et al., 2004), as well as inactive microcystin producers (Mikalsen et al., 2003). In contrast, no mcyH transcript was detected by qRT-PCR in the non-toxic strains here, confirming previous observations that this protein is essential for toxicity.
Transcription of mcnC, a cyanopeptolin synthesis gene, was also studied, as this non- ribosomal cyclic peptide also has a structural resemblance to a peptide siderophore. The finding that the inactivation of a microcystin gene, mcyH, disrupted the responsiveness of a gene encoding another non-ribosomal peptide to the nutrient status and growth stage of the cells was unexpected. This suggests that the synthesis of non-ribosomal peptides in the cell is coordinated and these compounds may have an interchangeable, or complementary function as has been proposed previously (Schatz, et al., 2007). Such process may allow survival of non-toxic strains under stress conditions, despite their inability to produce microcystin (Briand, et al., 2008, Edwin, et al., 2007). This could be the reason for the increase in mcnC transcription during the stationary phase in iron- - starved cells of PCC 7005, a non-toxic environmental isolate. The mutant mcyH , would not have experienced the same selective pressure in the nutrient-rich laboratory media they are routinely maintained in.
The ability of wild-type non-toxic strains to survive during iron stress was reflected in the similar growth rate of PCC 7005 and PCC 7806 and is in agreement with theoretical calculations that 100 nM Fe is sufficient to support 107 – 108 cell ml-1 (Ferreira and Straus, 1994). These results suggest that a highly efficient system for coping with short- term iron limitation is in place in these organisms.
During iron starvation, one of the first responses of cyanobacteria is to degrade the phycobilisome leading to chlorosis and the synthesis protective proteins around photosystem I (PSI) and II (PSII) in an attempt to reduce electron flow and therefore, reduce ROS formation (Boekema, et al., 2001, Ferreira and Straus, 1994, Lax, et al., 2007). IsiA, also known as CP43’, is structurally similar to the CP43 unit of PSI, and is synthesised in response to oxidative stress and during the stationary growth phase (Boekema, et al., 2001). This chlorophyll-binding protein is thought to initially function to dissipate excess energy and later forms a ring around PSI, acting as an alternative 86 Gene transcription during iron stress antenna (Boekema, et al., 2001, Nield, et al., 2003). This gene was chosen as a marker for the oxidative stress experienced by the M. aeruginosa cells grown in different iron conditions and was found to differ between the three strains (Figure 3.4.1). The toxic strain PCC 7806 appeared to be the most responsive to iron stress, as isiA transcription was the highest in this strain, and may have allowed the cells to overcome extensive oxidative stress and retain their pigmentation, as was confirmed by the lack of thylakoid disorganization observed by TEM (Figure 3.3.5). The lower levels of isiA observed in the non-toxic strains, in particular mcyH-, revealed the inability of these strains to modify PSI to a sufficient extent during early iron stress and could have caused an attack of the photosynthetic apparatus by ROS and reduction of iron to Fe2+ in the Fenton reaction (Figure 3.4.1). Although by the end of the growth experiments, transcript levels of IsiA were high in all strains, this may be caused by both the prolonged iron stress, as well as the entry of the cells in stationary phase where IsiA expression has been observed previously (Singh and Sherman, 2006).
Despite the differences observed in isiA transcription in the strains used here, this gene is not encoded in the genome of the toxic strain M. aeruginosa NIES-843 (Kaneko, et al., 2007), suggesting that even in the absence of this protein M. aeruginosa strains are able to adjust their response via alternative pathways. In fact, when a IsiA mutant was generated in Synechocystis sp. PCC 6803, the mutant strain appeared more resistant to damage by hydrogen peroxide and the resulting oxidative stress (Singh, et al., 2005).
The three strains studied here showed a difference not only in the degree of chlorosis and modification of photosystem I by IsiA, but also in the transport of iron from the periplasm. In particular, the transcription of the ferrous uptake transporter feoB, was found to be up-regulated only in the mcyH- mutant strain. In cyanobacteria, this transport system is usually secondary to constitutively expressed siderophore- transporters and is activated only when supply of iron to the cell is insufficient (Katoh, et al., 2001). A recent study in M. aeruginosa PCC 7806 has also shown that the superoxide induced reduction of iron to Fe(II) occurs only when Fe(III) is insufficient (Fujii, et al., 2010). Thus, the decrease observed for PCC 7005 feoB may represent a situation where ferric transport is sufficient to meet the requirements of this wild-type strain, but not the mcyH- mutant. This finding is consistent with the observation that both non-toxic strains used here had a similar iron uptake rate to the periplasm in the presence of EDTA and during the exponential growth phase this was higher than the
87 Chapter 3
55Fe uptake observed for PCC 7806 (Manabu Fujii, personal communication). This result is in contrast to previous reports that toxic M. aeruginosa take up more iron with the inference that, as a result, non-microcystin producers are forced to maintain a smaller intracellular iron reserve (Utkilen and Gjolme, 1995). Due to the differences in the choice of iron transporters as illustrated by the transcription of feoB and futA, it was of interest to determine where the Fe3+-binding protein FutA is localized.
FutA in M. aeruginosa exists as a single copy in the genome whereas in Synechocystis sp. PCC 6803 two highly conserved open reading frames, slr1295 and slr0513, designated futA1 and futA2, are present (Badarau, et al., 2008, Katoh, et al., 2001). These genes were first reported to be subunits of the same ABC Fe3+ transporter (Katoh, et al., 2001, Koropatkin, et al., 2007). Subsequent studies have shown futA1 to co- purify with photosystem II and the product of this gene is now considered a homologue of the iron deficiency induced protein IdiA, whereas futA2 is found in the periplasm and cytoplasm with an unknown function (Badarau, et al., 2008, Tolle, et al., 2002, Waldron, et al., 2006). It was not clear whether the M. aeruginosa homologue studied here is a periplasmic iron transporter or IdiA. The two scenarios would have different implications with regard to whether toxicity influences uptake of Fe3+ or the remodeling of PSII, similar to the protective role that IsiA has on PSI. The confocal imaging indicated that the FutA protein was expressed in the periphery of the cell, where iron is likely to diffuse passively by a concentration gradient created between the extracellular space and the periplasm (Figure 3.3.6) (Fujii, et al., 2010). It is likely that the FutA protein homologue in M. aeruginosa is a transporter rather than an IdiA-like protein involved in photosystem protection. Importantly, not all cells were expressing this protein (Figure 3.3.7), suggesting that only some members of the population in a bloom may be responsive to iron stress.
The ferric uptake regulator FurA governs the regulation of a number of iron uptake and oxidative stress genes (including feoB, futA and isiA) and has been shown to bind to the mcy gene cluster promoter (Martin-Luna, et al., 2006b). Therefore, changes in the expression of this transcription factor may provide explanation to the link between the iron stress response and toxicity in M. aeruginosa. As expected, given the role of iron as a cofactor for FurA, furA transcription also decreased in M. aeruginosa PCC 7806 and the mcyH- strain with decreasing iron availability. However, despite the similar transcript levels of FurA in PCC 7806 at 10 nM and 100 nM Fe relative to iron-replete
88 Gene transcription during iron stress conditions, an increase in toxicity was only measured at 10 nM Fe. This is indirect evidence for an additional level of toxin synthesis regulation besides transcription repression by FurA, possibly mediated by other Fur homologues or redox-sensitive transcriptional regulators such as NtcA.
FurA regulates the transcription of iron uptake proteins and IsiA, therefore differences in photosystem modification and iron uptake that were observed in the wild-type strains studied here could be attributed to differential expression of FurA (Figure 3.4.1). The lack of a significant change in transcription in PCC 7005 furA may also explain the failure of this strain to up-regulate isiA, futA and feoB transcription to levels similar to those in PCC 7806. On the other hand, although furA was down-regulated in mcyH- as expected under iron-starved conditions, the increased transcription of Fe2+ rather than Fe3+ transporters in this strain, or the lower levels of isiA transcription, cannot be attributed to FurA regulation alone. These differences of response to iron stress in PCC 7005 and PCC 7806 mcyH- may be the result of the adaptation of naturally occurring non-toxic strains to the environmental niche that they occupy allowing their survival in a nutrient-limited environment.
The first detailed study of a cyanobacterial FurB protein, found that transcription of furB in iron starvation was influenced by oxidative stress but not iron limitation, which may only induce ROS formation if severe (Lopez-Gomollon, et al., 2009). In contrast, in the growth conditions used here FurB appeared to be iron-responsive. This may be a result of the different extent of oxidative damage experienced by the cells, when superoxide is produced in iron starvation and their ability to respond accordingly. The transcript levels of furC were also different based on the iron concentration in each strain, particularly in 10 nM Fe. The significance of these results is unclear, given that the function of FurC in cyanobacteria is unknown. The FurC protein is a homologue of members of the zinc uptake regulator (Zur) family, but in M. aeruginosa lacks the conserved Zn-binding motif that aids in dimerisation of Fur proteins (Hernandez, et al., 2004). It does not to bind haeme and is not regulated by iron starvation, but instead has been proposed to be involved in oxidative stress protection (Hernandez, et al., 2004, Lopez-Gomollon, et al., 2009). However, direct conclusions from the transcription patterns of Fur proteins are challenging because these proteins have been found to be regulated post-transcriptionally by anti-sense RNAs, as well as by binding to each
89 Chapter 3 other’s promoters, thereby attenuating or enhancing their function (Herna´ndez, et al., 2006, Hernandez, et al., 2004).
Figure 3.4.1. Iron stress response model in M aeruginosa strains. Proteins are coloured depending on observed up-regulation (red) or down-regulation (green) of transcripts in response to growth in 10 nM Fe, as assessed by qRT-PCR. A delayed toxic-like response is shown in red/green. Only significant changes (t-test, p<0.05) are shown, unchanged proteins are in grey. The abundance of phycobilisome proteins is based on the degree of chlorosis observed in cultures grown in the presence of 10 nM Fe. The possible involvement of microcystin in the processes is also indicated. IsiA: iron-stress induced protein A; FeoB: ferrous uptake transporter; FutA: ferric uptake transporter; Fur: ferric uptake regulator; MCYST: microcystin; OM: outer membrane; PM: plasma membrane; PS: photosystem; ROS: reactive oxygen species; TM: thylakoid membrane.
90 Gene transcription during iron stress
As illustrated in Figure 3.4.1, microcystin may be involved in modulating the activity of FurA and therefore the downstream response to iron starvation. A similar role for microcystin was proposed in Chapter 2, where NtcA-regulated processes were disrupted in non-toxic strains of M. aeruginosa. Rather than binding to both two transcriptional factors, a more plausible scenario would be modulation of NtcA by the toxin molecule alone and a downstream effect on FurA activity. Such cross-regulation of NtcA and FurA has been observed in the nitrogen-fixing Anabaena sp. PCC 7120 (López- Gomollón, et al., 2007a), however this process needs to be studied further in non- diazotrophic cyanobacteria.
3.5. Conclusions
In summary, studies of the transcriptional response of M. aeruginosa to iron starvation and iron limitation found the process to be highly dependent on the growth phase of the cells and the concentration of available iron. Numerous changes in transcription were observed in the PCC 7806 mcyH- mutant strain, but not in the non-toxic PCC 7005, a situation which should be considered when conclusions from studies from laboratory- generated mutants are extended to naturally-occurring non-microcystin producing strains. Toxin synthesis, but not toxin gene transcription, was increased in iron starvation and the presence of microcystin was proposed to affect the expression of FurA targets such as the photosystem protective IsiA and the iron-binding FeoB and FutA.
91 Chapter 3
92
Chapter 4 Proteomic analysis of the iron stress response in the bloom-forming cyanobacterium Microcystis aeruginosa
Chapter 4
94 Iron stress proteomics
4.1. Introduction
Proteome and transcriptome studies have been applied previously to cyanobacteria in various stress-inducing conditions, including high salinity, UV-B irradiation, temperature stress, copper and iron limitation. These high-throughput techniques allow for a global overview of the changes that occur in the cells during their adaptation to a new environment (Castielli, et al., 2009, Ehling-Schulz, et al., 2002, Fulda, et al., 2000, Gao, et al., 2009, Kurian, et al., 2006, Suzuki, et al., 2006). However, the majority of these investigations have used the model unicellular Synechocystis sp. PCC 6803 and the filamentous N2-fixing microcystin producer Anabaena sp. PCC 7120. No similar work has been performed on the bloom-forming Microcystis aeruginosa, or in other microcystin producers, with reference to toxin synthesis. Such studies may be of particular relevance, given the increased survival of toxic strains in iron-starved and high light conditions and the possibility that microcystin may contribute to this adaptation (Edwin, et al., 2007, Utkilen and Gjolme, 1995).
The previous chapter described the transcriptional response of a number of genes involved in iron starvation and the differences between toxic and non-toxic strains of M. aeruginosa. The metabolic adjustments that occurred in the cells in response to iron stress were found to be highly dynamic, depending on the growth stage at which the cells were harvested and the concentration of iron used. It appeared that overall, strains differed in their iron uptake and remodeling of the photosynthetic machinery. These differences were particularly pronounced in cells grown under conditions of severe iron stress (10 nM Fe) and could stem from a differential regulation of iron-responsive genes by the ferric uptake regulator (Fur) family of transcription factors. It was proposed that microcystin may interact with, and affect the activity of Fur directly, or via NtcA regulation of Fur. In order to investigate this hypothesis further and obtain a more global view of the response to low levels of bioavailable iron in these cyanobacteria, a proteomic approach was employed. The results provided an opportunity for the reconstruction of the metabolic changes that occur during iron starvation in these microorganisms and a more accurate assessment of the possibility that the production of microcystin influences the choice of survival strategy in M. aeruginosa.
95 Chapter 4
4.2. Methods
4.2.1. Microcystis strains, culture growth and protein extraction
For proteome studies, cells were cultured in modified Fraquil media (Fraquil*) (Andersen, 2005, Fujii, et al., 2010) (Appendix B, Table A.2.1) supplemented with 10 nM or 1000 nM Fe and harvested in the exponential growth phase. This choice of media iron concentration was based on observations in Chapter 3 that these growth conditions induced the most dramatic changes in toxicity, as well as in gene transcription. Culturing was performed by Dang The Cuong in the School of Civil and Environmental Engineering (UNSW), as described in Chapter 3. Briefly, Microcystis aeruginosa PCC 7806 (toxic), PCC 7005 (non-toxic) and PCC 7806 mcyH- (non-toxic mutant) were grown in polycarbonate culture vessels (Nalgene) that were acid-washed prior to use (Fujii, et al., 2010). Triplicate batch cultures of M. aeruginosa were maintained at 27oC -2 -1 under a 14:10 light/dark cycle (90 )mol photons m s ) supplied by cool white fluorescent lamps.
After a culturing period of 8 days, cells were collected by centrifugation and were partially lysed by three freeze/thaw cycles in liquid nitrogen and 37oC. The cells were resuspended in acid extraction buffer (Herbert, et al., 2006) and sonicated for 3 x 15 s on ice, before collecting the insoluble cellular material by centrifugation. The supernatant was buffer-exchanged with PBS and treated with Benzonaze nuclease (Sigma) and supplied with 1 mM PMSF (Sigma). Protein concentration was determined on pooled biological triplicate samples using a Bradford assay (BioRad) and running dilutions of the sample and BSA standards (Sigma) on a SDS-PAGE gel.
4.2.2. In-gel digestion and nanoLC-MS/MS analysis
For proteomic analysis, 60 μg of protein was separated by 1D SDS-PAGE in a precast 4-20% Criterion Tris-Acetate gel (BioRad) and stained with Coomassie G-250. Proteins were excised in 14 fractions from each lane and the gel pieces were destained with 50 mM NH4HCO3/50% acetonitrile (ACN). Reduction and alkylation was performed in two steps, with incubation in 15 mM DTT for 30 min at 37oC, followed by 25 mM o iodacetamide for 30 min at 37 C. The gel pieces were washed with ACN/NH4HCO3, dehydrated completely in 100% ACN and proteins were digested overnight in 80 ng sequencing-grade trypsin (Promega). Peptides were extracted in 50% ACN/1% formic
96 Iron stress proteomics acid, dried completely and resuspended in 10 μl 1% formic acid/ 0.05% heptafluorobutyric acid (HFBA) solution.
Mass spectrometry was performed in a LTQ-FT MS/MS at the Biological Mass Spectrometry Facility (BMSF), UNSW, with the assistance of Dr. Mark Raftery. Digest peptides were separated by nano-LC using an Ultimate 3000 HPLC and autosampler system (Dionex, Amsterdam, Netherlands). Samples (1 μl) were concentrated and desalted onto a micro C18 precolumn (500 μm x 2 mm, Michrom Bioresources, -1 Auburn, CA) with H2O:ACN (98:2, 0.05% HFBA) at 20 μl min . After a 4 min wash the pre-column was switched (Valco 10 port valve, Dionex) into line with a fritless nano column (75 μ x ~10 cm) containing C18 media (5 μ, 200 Å Magic, Michrom) manufactured according to Gatlin et al. (1998). Peptides were eluted using a linear gradient of H2O:ACN (98:2, 0.1% formic acid) to H2O:ACN (64:36, 0.1% formic acid) at 350 nl min-1 over 30 min. High voltage (1800 V) was applied to low volume tee (Upchurch Scientific) and the column tip positioned ~ 0.5 cm from the heated capillary (T=250°C) of a LTQ FT Ultra (Thermo Electron, Bremen, Germany) mass spectrometer. Positive ions were generated by electrospray and the LTQ FT Ultra operated in data dependent acquisition mode (DDA).
A survey scan m/z 350-1750 was acquired in the FT ICR cell (Resolution = 100,000 at m/z 400, with an accumulation target value of 1,000,000 ions). Up to the 6 most abundant ions (>3,000 counts) with charge states > +2 were sequentially isolated and fragmented within the linear ion trap using collisionally induced dissociation with an activation q = 0.25 and activation time of 30 ms at a target value of 30,000 ions. M/z ratios selected for MS/ MS were dynamically excluded for 30 seconds.
4.2.3. Data analysis
Peak lists were generated using Mascot Daemon/extract_msn (Matrix Science, London, England, Thermo) using the default parameters, and submitted to the database search program Mascot (version 2.1, Matrix Science). Search parameters were: Precursor tolerance 4 ppm and product ion tolerances ± 0.4 Da; Met(O), Acryl(C) and Carbamidomethyl(C) were specified as variable modification, enzyme specificity was trypsin, 1 missed cleavage was possible and the M. aeruginosa PCC 7806 protein database (downloaded from NCBI, version 01/11/07) searched. A decoy database was searched to calculate the false discovery rate.
97 Chapter 4
Scaffold 2 (Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK) and X! Tandem (www.thegpm.org; version 2007.01.01.1). X! Tandem and Mascot were set up to search the custom M. aeruginosa PCC 7806 proteome database containing 5161 entries, assuming tryptic digestion. Mascot and X! Tandem were searched with a fragment ion mass tolerance of 0.40 Da and a parent ion tolerance of 4.0 ppm. Oxidation of methionine and iodoacetamide derivative of cysteine were specified in X! Tandem as variable modifications. Oxidation of methionine, iodoacetamide derivative of cysteine and the acrylamide adduct of cysteine were specified in Mascot as variable modifications. Peptide identifications were accepted if they could be established at greater than 95.0% probability as specified by the Peptide Prophet algorithm (Keller, et al., 2002). Protein identifications were accepted if they could be established at greater than 99.0% probability and contained at least 1 identified peptide. Protein probabilities were assigned by the Protein Prophet algorithm (Nesvizhskii, et al., 2003). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Differentially displayed proteins were determined by a student’s t-test, with p < 0.05.
4.3. Results
PCC 7806 PCC 7005 PCC 7806 mcyH- MWM 10Fe 1000Fe 10Fe 1000Fe 10Fe 1000Fe
Figure 4.3.1. Representative 1D SDS-PAGE gel of strains used in this chapter. For each strain, 15 μg of protein was separated by electrophoresis on a 8-20% Criterion gel (BioRad). MWM: PrecisionPlus molecular weight markers (BioRad). 10Fe: 10 nM Fe; 1000Fe: 1000 nM Fe.
98 Iron stress proteomics
4.3.1. Background proteome differences in iron-replete conditions
The focus of this chapter was to investigate the difference in the iron starvation response between toxic and non-toxic M. aeruginosa strains. However, it was important to first consider the basal proteomic differences that exist when cells are in a nutrient- replete environment, as a basis for their reaction to iron limitation. In order to achieve this, the protein expression in the three different strains of M. aeruginosa cultured in 1000 nM Fe was compared. After a t-test (p-value < 0.05), 160 proteins were differentially displayed between PCC 7806 and PCC 7806 mcyH- that were grown in media containing 1000 nM Fe. The majority of these proteins were up-regulated in the mutant mcyH- strain, whereas 23 proteins were present at a higher abundance in the toxic strain PCC 7806. Numerous hypothetical proteins, and proteins encoded only in the genome of PCC 7806 were expressed differentially between the two strains, including the previously identified microcystin-related protein A (MrpA) (Dittmann, et al., 2001) and a RuBisCo-like protein (Carre-Mlouka, et al., 2006), which were both up-regulated in mcyH-. The toxic strain had an increased abundance of several transporters, nitrogen metabolism-related proteins and, importantly, a number of redox-sensitive proteins, as well as the ferric- iron binding protein FutA. On the other hand, in mcyH-, many photosynthetic, carboxysomal, ribosomal and energy metabolism proteins were increased relative to the wild-type strain PCC 7806. Unlike the PCC 7806 wild type strain, a preference for the ferrous uptake transporter (FeoA) in iron-replete growth was observed in mcyH-. When the same analysis was applied to the PCC 7005 proteome, 135 proteins were differentially displayed compared to PCC 7806 grown in 1000 nM Fe. Again, most of these proteins were over-represented in the non-toxic strain. In terms of proteins related to the iron stress response, PCC 7806 contained a higher amount of the cytochrome b6/f iron-sulfur subunit, as well as the bacterioferritin co-migratory protein and the DNA- protection protein, Dps. A summary of the proteins exhibiting the same trend in expression in both non-toxic strains, but not PCC 7806, is presented in Table 4.3.1 and Table 4.3.2. Two proteins utilising NH3 and NH4 as substrates (hydroxylamine reductase and glutamine synthase, respectively) were up-regulated in the toxic strain (Table 4.3.1).
99 Chapter 4
Table 4.3.1. Proteins down-regulated in both non-toxic M. aeruginosa PCC 7005 and PCC 7806 mcyH-. Significant t-test results (p-value < 0.05) and fold-change in expression relative to PCC 7806 are shown. ‘TO’ denotes protein present in PCC 7806 only.
PCC 7005 PCC 7806 mcyH- Accession Protein ID (NIES-843) t-test expression t-test expression gi|159030779 hypothetical protein MAE 45990 0.0008 -4.22 5.2e-4 -3.44 gi|159030833 hypothetical protein MAE 44430 0.0020 -13.65 3.2e-5 -8.49 gi|159030699 no homologue 0.0020 -6.39 0.0002 -2.43 gi|159028670 hydroxylamine reductase 0.0045 -9.77 0.0038 TO gi|159030589 bicarbonate transport system 0.0048 -1.52 0.0044 -2.42 substrate-binding protein gi|159030269 thioredoxin peroxidase 0.0085 -3.04 0.0040 -3.78 gi|159026416 membrane associated rubredoxin 0.0170 -3.86 0.0430 -2.45 gi|159030372 glutamine synthase GlnA* 0.0150 -4.52 0.0120 -5.73
* denotes thioredoxin targets identified previously in cyanobacteria (Lindahl and Kieselbach, 2009, Mata- Cabana, et al., 2007, Perez-Perez, et al., 2009)
Importantly for the iron-stress situation investigated here, rubredoxin, an iron containing enzyme, and thioredoxin-peroxidase were significantly (p-vale < 0.05) more abundant in PCC 7806. Phycobilisome polypeptides and proteins of the translation machinery (chaperons, elongation factors and 30S ribosomal subunit proteins) were increased in both of the non-toxic strains (Table 4.3.2). In addition, in the non-toxic strains, had increased levels of several proteins involved in carbon metabolism (carboxysome and Calvin cycle proteins) and storage (PHA-specific beta ketothiolase) relative to PCC 7806.
Table 4.3.2. Proteins up-regulated in both non-toxic M. aeruginosa PCC 7005 and PCC 7806 mcyH-. Significant t-test results (p-value < 0.05) and fold-change in expression relative to PCC 7806 are shown. ‘NTO’ denotes protein present only in the non-toxic strains but not found in PCC 7806.
PCC 7005 PCC 7806 mcyH- Accession Protein ID (NIES-843) t-test expression t-test expression gi|159026025 carbon concentrating mechanism 0.0024 NTO 0.0012 NTO protein CcmA gi|159026539 elongation factor Ts 0.0016 18.00 0.0002 1.66 gi|159027749 adenosylhomocysteinase 0.0072 NTO 0.0280 NTO gi|159027955 peptidyl-prolyl cis-trans isomerase 0.0410 NTO 0.0043 NTO gi|159027867 phosphoglycerate kinase* 0.0001 8.67 0.0100 6.00 gi|159027942 carbon concentrating mechanism 0.0007 44.44 9.2e-5 153.53 protein CcmM* gi|159028007 phycobilisome rod-core linker 0.0056 NTO 0.0170 NTO polypeptide CpcG* gi|159028193 30S ribosomal protein S8 5.9e-5 39.39 0.0110 18.18 gi|159028535 chaperonin GroEL (groL2) 0.0079 8.31 0.0058 10.69 gi|159030775 chaperonin GroEL (groL1) * 0.0180 4.73 0.0018 6.79 gi|159028548 cell division protein FtsZ 0.0410 8.08 0.0190 9.09
100 Iron stress proteomics
PCC 7005 PCC 7806 mcyH- Accession Protein ID (NIES-843) t-test expression t-test expression gi|159028843 3 beta hydroxysteroid 0.0043 6.02 0.0100 2.40 dehydrogenase gi|159028923 30S ribosomal protein S1 0.0094 NTO 0.0310 NTO gi|159029026 hypothetical protein MAE 32640 0.0110 NTO 0.0270 NTO gi|159029182 glyceraldehyde 3-phosphate 0.0079 15.78 0.0070 5.70 dehydrogenase* gi|159029228 elongation factor Tu* 0.0010 5.76 0.0440 3.20 gi|159029272 PHA-specific beta ketothiolase 0.0150 NTO 0.0130 NTO gi|159029342 recombinase A* 0.0024 NTO 0.0170 NTO gi|159029497 phycobilisome rod linker 5.9e-5 NTO 0.0031 NTO polypeptide gi|159029578 glutamate-1-semialdehyde 0.0012 NTO 0.0440 NTO aminotransferase gi|159029605 fructose 1,6-bisphosphate (glpX 0.0001 NTO 0.0260 NTO encoded) gi|159029660 hypothetical protein MAE 39950 0.0180 7.00 0.0006 24.00 gi|159029712 hypothetical protein MAE 59390 0.0350 NTO 0.0005 NTO gi|159029886 malic enzyme 0.0003 NTO 0.0500 NTO gi|159029969 carbamoyl phosphate synthase 0.0005 NTO 0.0034 NTO small subunit gi|159030256 glyceraldehyde 3-phosphate 0.0460 NTO 0.0070 NTO dehydrogenase* gi|159030306 hypothetical protein MAE 04620 0.0210 NTO 0.0270 NTO gi|159030620 uroporphyrinogen decarboxylase 0.0390 NTO 0.0072 NTO
* denotes thioredoxin targets identified previously in cyanobacteria (Lindahl and Kieselbach, 2009, Mata- Cabana, et al., 2007, Perez-Perez, et al., 2009)
4.3.2. Iron-induced proteomic changes
A number of proteins that were significantly different (p < 0.05) between iron-replete and iron-stressed cells in each of the three strains studied here were identified (Tables 4.3.3-5). Based on the functional categories of the differentially displayed proteins (Figure 4.3.2), these protein expression patterns revealed that in each M. aeruginosa strain analysed, a different iron stress response takes place. Several functional categories of proteins were affected by iron starvation, with the most pronounced changes being in the expression of proteins involved in photosynthesis and respiration, those assigned in the ‘other’ category, as well as a large number of hypothetical proteins (Figure 4.3.2). The response of the toxic strain PCC 7806 was limited to nine functional categories, with the majority being photosynthetic and respiratory proteins. However, inactivation of the microcystin cluster-associated ABC transporter mcyH- resulted in the iron-stress response being dominated with proteins in the ‘other’ functional category. In both non-toxic strains (PCC 7005 and mcyH-), the iron stress response included significant changes in proteins involved in amino acid synthesis and DNA replication.
101 Chapter 4
The non-toxic M. aeruginosa strains analysed also exhibited a larger number of hypothetical proteins being differentially displayed as a result of iron starvation, with the involvement of photosynthetic proteins being more restricted compared to the microcystin-producer PCC 7806 (Figure 4.3.2).
Figure 4.3.2. Functional categories of proteins differentially expressed in iron stress in strains of M. aeruginosa. Proteins significantly different after a t-test (p < 0.05) between starved (10 nM Fe) and iron-replete (1000 nM Fe) cells are categorized according to CyanoBase.
The strain-to-strain differences observed above were less pronounced when the entire proteomes of the three strains at 10 nM and 1000 nM Fe were compared (Figure 4.3.3). Under iron stress, the functional group distribution in the proteome of PCC 7806 did not change dramatically, and a very slight increase in the number of proteins involved in translation was observed. Similarly, the mcyH- knockout strain did not show many differences in the functional groups expressed in either 10 nM or 1000 nM Fe, although there was a small decrease in proteins in the purine metabolism and ‘other’ categories. When compared with its parent PCC 7806 strain, mcyH- expressed a larger number of proteins involved in DNA replication, transcription and purine synthesiss, as well as those involved in central intermediary metabolism and unknown proteins involved in both general stress and iron-replete growth. The proteome of the mutant strain also contained the smallest proportion of hypothetical proteins, while the non-toxic strain PCC 7005 showed the most obvious transition from iron-replete to iron-starved conditions, as shown by changes in the functional groups of the expressed proteome (Figure 4.3.3). Specifically, these included an increase in the percentage of proteins
102 Iron stress proteomics involved in translation, and a decrease in amino acid biosynthetic proteins, as well as proteins involved in DNA replication, purine synthesis and transcription.
103 Chapter 4
Figure 4.3.3. Protein expression in iron-starved and iron-replete cultures of M.aeruginosa strains PCC 7806, PCC 7005 and PCC 7806 mcyH-. Proteins are grouped in functional categories according to their annotation in CyanoBase.
104 Iron stress proteomics
4.3.2. Protein expression in iron-starved cells of the toxic M. aeruginosa PCC 7806
In the toxic strain PCC 7806, a total of 298 and 359 proteins were identified at 1000 and 10 nM Fe, respectively. These represented 366 unique protein IDs, of which 42 were found to be differentially expressed at a significant level between the two growth conditions investigated (p-value < 0.05) (Table 4.3.3).
Overall, PCC 7806 was the only strain capable of up-regulating the ferric-binding protein FutA (Chapter 3), significantly when grown in iron-depleted Fraquil* (Table 4.3.3, Figure 4.4.3). These changes in the periplasmic space included the down- regulation of transporters for bicarbonate and nitrate in 10 nM Fe. The majority of proteins differentially expressed by the available iron concentration in this microcystin- producing strain were involved in photosynthesis and respiration (Table 4.3.3 and Figure 4.4.3).
During iron stress, the PsaF protein involved in the core-antenna interaction with PSI was lower in abundance, as was the iron-rich PetC. Adaptation of the photosynthetic electron chain to a low-iron environment was evident from the increase in flavodoxin (IsiB), an iron-free alternative to ferredoxin. The expression of the phycobilisome chromophore-degrading protein NblB was also increased in these cultures. However, contrary to the expected degradation of the phycobilisome as a result of iron starvation, morphologically, the cells of this strain appeared to retain their pigmentation, while the proteomic data confirmed that seven proteins involved in phycobilisome structure were up-regulated in 10 nM Fe. This was paralleled by an increase in the expression of the phycobilisome-associated orange carotenoid protein (Table 4.3.3). The ATP synthase gamma subunit, AtpC, was also up-regulated, as were several proteins involved in the structure of the 50S ribosomal subunit, suggesting that both higher ATP synthesis and increased translation of proteins were taking place in the toxic M.aeruginosa PCC 7806 cells during iron stress relative to nutrient-replete cells. The changes observed in photosynthesis were also paralleled by an altered carbon fixation and assimilation process. CcmM, a carboxysome shell-associated protein, was one of the most highly up- regulated proteins under iron stress.
105 Chapter 4
Table 4.3.3. Proteins differentially expressed in iron-starved and iron-replete cultures of M. aeruginosa PCC 7806. Proteins are identified by their annotation in both the PCC 7806 and NIES 843 genomes. Only entries with a p-value < 0.05 after a t- test are presented. Expression during iron-depleted growth is given relative to 1000 nM Fe. Up: proteins expressed in 10 nM Fe only; down: proteins expressed in 1000 nM Fe only.
Protein ID Accession Protein ID (NIES-843) T-Test Expression (PCC 7806) gi|159029283 fructose-bisphosphatase class II fbaA* 0.000024 up gi|159028007 phycobilisome rod-core linker polypeptide cpcG* 0.01 up gi|159028395 50S ribosomal protein L1 rplA 0.019 up gi|159028112 no homologue isiB 0.02 up gi|159029497 phycobilisome rod linker polypeptide cpcI 0.025 up gi|159027097 phycocyanin alpha phycocyanobilin lyase related protein nblB 0.029 up gi|159029605 fructose-1,6-bisphosphatase, glpX-encoded glpX 0.031 up gi|159029093 phycobilisome core component apcF 0.034 up gi|159029996 hypothetical protein MAE 41320 unnamed protein 0.036 up gi|159027942 carbon concentrating mechanism protein ccmM ccmM * 0.000028 22.33 gi|159028535 60 kDa chaperonin GroEL2 groL2 0.0028 8.93 gi|159027971 iron transport system substrate-binding protein unnamed protein 0.0037 7.38 gi|159027867 phosphoglycerate kinase pgk* 0.0018 7.2 gi|159028500 OstA-like protein precursor unnamed protein 0.044 6.25 gi|159030775 60 kDa chaperonin GroEL1 groL1* 0.0084 6.23 gi|159027266 allophycocyanin-B unnamed protein 0.0031 5.83 gi|159029495 phycocyanin beta subunit cpcB* 0.006 5.77 gi|159028862 periplasmic protein mom72 0.0012 5.67 gi|159026581 DNA starvation/stationary phase protection protein Dps unnamed protein 0.013 5 gi|159029496 phycocyanin alpha subunit cpcA 0.0037 4.69 gi|159026883 phycobilisome core component apcE* 0.00049 4.68 gi|159027088 50S ribosomal protein L13 rplM 0.048 4.67 gi|159026585 putative peroxiredoxin unnamed protein 0.00046 4.22 gi|159028287 water-soluble carotenoid protein unnamed protein 0.022 4.14 gi|159028189 50S ribosomal protein L15 rplO 0.036 4.1 gi|159029660 hypothetical protein MAE 39950 unnamed protein 0.0023 4 gi|159029182 glyceraldehyde 3-phosphate dehydrogenase gap1* 0.027 3.71 gi|159027204 ATP synthase CF1 gamma chain atpC 0.0057 2.67 gi|159030779 hypothetical protein MAE 45990 unnamed protein 0.037 -1.51 gi|159030010 no homologue unnamed protein 0.046 -1.78 gi|159030792 hypothetical protein MAE 02890 unnamed protein 0.046 -2 gi|159028057 nitrate transporter NrtA nrtA* 0.046 -2.19 gi|159030699 no homologue unnamed protein 0.03 -2.33
106 Iron stress proteomics
Protein ID Accession Protein ID (NIES-843) T-Test Expression (PCC 7806)
gi|159030269 thioredoxin peroxidase unnamed protein 0.0094 -2.53
gi|159030833 hypothetical protein MAE 44430 unnamed protein 0.021 -2.54
gi|159030589 bicarbonate transport system substrate-binding protein unnamed protein 0.002 -2.57 gi|159028756 cytochrome b6/f complex iron-sulfur subunit petC 0.0003 -2.74
gi|159026416 membrane-associated rubredoxin unnamed protein 0.0085 -4.4
gi|159028670 hydroxylamine reductase unnamed protein 0.0042 -5.17 gi|159027523 photosystem I subunit III psaF 0.011 -8.94
gi|159027374 CAB/ELIP/HLIP superfamily protein unnamed protein 0.011 down
* denotes thioredoxin targets identified previously in cyanobacteria (Lindahl and Kieselbach, 2009, Mata- Cabana, et al., 2007, Perez-Perez, et al., 2009)
Two enzymes, phosphoglycerate kinase and glyceraldehyde 3-phosphate dehydrogenase, involved in consecutive steps of the Calvin cycle were up-regulated in 10 nM iron, in contrast to the decreased transport of bicarbonate across the cell membrane. The expression of two alternative forms of fructose-1,6,-bisphosphatase was also increased in iron-starved cells. Nitrogen metabolism was also affected, with a decrease in nitrate transport reflected in the down-regulation of hydroxylamine reductase, an enzyme which uses ammonia (NH3) as a substrate.
Several proteins involved in the cellular stress response were also differentially displayed, such as the subunits of the 60 kDa chaperone GroEL and the DNA- starvation/stationary phase protein that were both up-regulated. Peroxiredoxin, involved in H2O2 scavenging was also up-regulated, whereas the abundance of thioredoxin peroxidase and the iron-containing rubredoxin was lower in iron-starved cells. Surprisingly, the high-light inducible protein (HLIP), involved in oxidative stress was down-regulated.
Five hypothetical proteins were also found to be differentially expressed in the presence of 10 nM and 1000 nM Fe, including the homologues for MAE 41320, MAE 45990, MAE 44430 and MAE 39950, which were identified in the core Microcystis aeruginosa proteome described in Chapter 2. Another hypothetical protein, MAE 02890, not found in the core proteome of M. aeruginosa and two proteins with no homologue in the
107 Chapter 4
NIES-843 genome, were significantly different between the 10 nM and 1000 nM Fe growth conditions and may perform an iron-stress or a more general stress-related function.
4.3.3. Protein expression in iron-starved cells of the non-toxic M. aeruginosa PCC 7005
In the non-toxic strain PCC 7005, 433 proteins were observed, with 308 shared between the 10 nM and 1000 nM Fe growth conditions. After statistical analysis, 69 proteins were found to be differentially expressed to a significant level (t-test, n=3, p<0.05) between the cultures grown under the two different iron concentrations (Table 4.3.4). Unlike PCC 7806, where the majority of differentially displayed proteins were involved in phycobilisome formation and protection from ROS (Table 4.3.3), most of the significant differences in protein expression induced by a lack of iron in PCC 7005 consisted of proteins involved in energy production and translation (Table 4.3.4, Figure 4.3.1). Fifteen of the proteins in this differential display (Table 4.3.4) were not found in the PCC 7806 proteome, representing an iron-responsive protein subset specific to this strain. Of the 11 differentially displayed hypothetical proteins, two (MAE 39950 and MAE 45990) had significantly altered levels, similarly to PCC 7806, confirming a role of these proteins in the iron stress response of M. aeruginosa. However, whereas the putative DNA damage-responsive protein MAE 39950 followed the same expression pattern in both strains, this was not the case with the predicted transmembrane protein MAE 45990. Additionally, a protein encoded by gi|159030010 in the PCC 7806 proteome, but with no homologue in NIES-843 was found to have opposite expression patterns in the two strains. The fact that it is not universal to the genomes of all M. aeruginosa strains analysed so far, suggests that this is not essential for the survival of these bloom-forming cyanobacteria.
In general, the majority of differentially expressed proteins in this strain were significantly down-regulated in the iron-starved cells (Table 4.3.4). This included a number of transporters and periplasmic proteins. However, none of those affected participate in iron-scavenging or binding (Figure 4.3.3). The only potential exception was the expression of a predicted periplasmic binding protein, with an unknown substrate. The photosynthesis machinery of PCC 7005 was also affected, although the proteins that were differentially displayed were different from those in PCC 7806. Two of the photosystem I proteins involved in trimerisation of the complex, PsaF and PsaL,
108 Iron stress proteomics
decreased in abundance and could result in a decrease in the PSI/PSII ratio. The Fe4-S4 cluster-containing apocytochrome F was also down-regulated, but unlike PCC 7806, the expression of neither ferredoxin nor flavodoxin was affected. Several phycobilisome components were down-regulated consistent with the pronounced chlorosis that was evident in PCC 7005 cells cultured in 10 nM Fraquil*. However, the phycocyanin beta subunit, CpcB and the phycobilisome small rod-linker polypeptide showed increased expression. These alterations in the photosynthetic apparatus may have led to a decrease in ATP synthase subunits during iron stress and appeared to have a negative effect on translation and cell division, since both 50S and 30S ribosomal subunits and the cell division FtsH protein had significantly lower abundance in 10 nM Fe.
The decrease of metabolic function in PCC 7005, suggested by the decrease in nutrient transport and ATP synthesis was also observed in the carbon and nitrogen metabolism. In contrast to PCC 7806, this non-toxic strain appeared to have a decreased metabolite flux through the Calvin-Benson cycle, as reflected by the decreased relative abundance of 6-phosphogluconate dehydrogenase, phosphoglycerate kinase, glyceraldehyde 3- phosphate dehydrogenase and fructose-bisphosphatase class II (Table 4.3.4). The synthesis of pyruvate via gluconeogenesis was also decreased, as observed by the reduced amount of malic enzyme and phosphopyruvate hydratase in the iron-depleted cells. Two signaling proteins, LexA and CP12, which affect cyanobacterial carbon metabolism were also reduced in 10 nM Fe. The effects of iron starvation were not limited to carbon metabolism, with nitrogen assimilation also affected. A number of amino acid synthesis proteins, in particular those participating in alanine, leucine, arginine and aspartate synthesis were decreased in abundance. In this network, CarA was down-regulated, as was the glutamate--ammonia ligase, GlnA.
Similar to PCC 7806, several stress-induced proteins were differentially expressed, but their identities differed between the two strains (PCC 7806 and PCC 7005). Proteins that were up-regulated in iron stress were thioredoxin reductase, RNA binding protein and the heat shock protein 40. On the other hand, the chaperone GroEL, RecA and glutathione reductase were present at lower levels compared to cultures grown in the presence of 1000 nM Fe.
109 Chapter 4
Table 4.3.4. Proteins differentially expressed in iron-starved and iron-replete cultures of M. aeruginosa PCC 7005. Proteins are identified by their annotation in both the PCC 7806 and NIES 843 genomes. Only entries with a p-value < 0.05 after a t- test are presented. Expression is given relative to 1000 nM Fe, protein IDs in bold represent entries only found in PCC 7005, but not wild-type PCC 7806. Up: proteins expressed in 10 nM Fe only; down: proteins expressed in 1000 nM Fe only.
Protein ID Accession Protein ID (NIES-843) T-Test Expression (PCC 7806) gi|159026156 periplasmic protein unnamed protein 0.008 up gi|159026944 hypothetical protein MAE 36090 unnamed protein 0.008 up gi|159027307 heat shock protein 40 unnamed protein 0.008 up gi|159029097 hypothetical protein MAE 21680 unnamed protein 0.008 up gi|159027161 no homologue unnamed protein 0.024 4.56 gi|159028356 CP12 polypeptide unnamed protein 0.001 4.2 gi|159025877 RNA-binding protein unnamed protein 0.015 3.83 gi|159025949 phycobilisome small rod linker polypeptide unnamed protein 0.038 3.33 gi|159030973 hypothetical protein MAE 47530 unnamed protein 0.005 3.23 gi|159028192 50S ribosomal protein L6 rplF 0.022 3.12 gi|159030010 no homologue unnamed protein 0.027 3 gi|159029846 50S ribosomal protein L27 rpmA 0.039 2.5 gi|159028189 50S ribosomal protein L15 rplO 0.031 2.38 gi|159030594 hypothetical protein MAE 20050 unnamed protein 0.027 2.35 gi|159030090 hypothetical protein MAE 15680 unnamed protein 0.01 2.33 gi|159029495 phycocyanin beta subunit cpcB* 0.019 2.17 gi|159029380 thioredoxin reductase unnamed protein 0.008 2 gi|159027955 peptidyl-prolyl cis-trans isomerase B unnamed protein 0.026 1.97 gi|159030355 hypothetical protein MAE 10050 unnamed protein 0.035 1.5 gi|159028201 50S ribosomal protein L22 rplV 0.035 -1 gi|159027204 ATP synthase CF1 gamma chain atpC 0.026 -1.67 gi|159029108 hypothetical protein MAE 21730 unnamed protein 0.029 -1.67 gi|159029493 acetylglutamate kinase unnamed protein 0.004 -2 gi|159029578 glutamate-1-semialdehyde aminotransferase hemL 0.006 -2.17 gi|159029497 phycobilisome rod linker polypeptide cpcI 0.02 -2.19 gi|159026197 phosphopyruvate hydratase unnamed protein 0.008 -2.22 gi|159028256 agmatinase speB2 0.023 -2.26 gi|159028447 alanine dehydrogenase unnamed protein 0.037 -2.33
110 Iron stress proteomics
Protein ID Accession Protein ID (NIES-843) T-Test Expression (PCC 7806) gi|159029660 hypothetical protein MAE 39950 unnamed protein 0.045 -2.33 gi|159027867 phosphoglycerate kinase pgk* 0.035 -2.42 gi|159029283 fructose-bisphosphate class II fbaA* 0.004 -2.57 gi|159030988 phosphate-binding periplasmic protein pstS 0.043 -2.58 gi|159030819 hypothetical protein MAE 58860 unnamed protein 0.011 -2.8 gi|159028868 hydrophobic amino acid uptake ABC-transporter unnamed protein 0.011 -3 gi|159030390 methylenetetrahydrofolate dehydrogenase unnamed protein 0.016 -3 gi|159028837 TPR repeat-containing protein unnamed protein 0.002 -3.11 gi|159026223 extracellular solute-binding protein unnamed protein 0.003 -3.17 gi|159029182 glyceraldehyde 3-phosphate dehydrogenase gap1* 0.019 -3.24 gi|159030779 hypothetical protein MAE 45990 unnamed protein 0.012 -3.5 gi|159027109 hypothetical protein MAE 53760 unnamed protein 0.009 -3.67 gi|159029160 hemolysin secretion protein hlyD 0.025 -3.83 2.00E- gi|159029228 elongation factor Tu tuf * -3.94 04 gi|159028157 amino acid transport system substrate-binding protein unnamed protein 0.004 -3.97 gi|159030457 dihydrodipicolinate synthase dapA 0.004 -4 gi|159027052 photosystem I reaction center protein subunit XI psaL 0.031 -4 gi|159030089 hypothetical protein MAE 15650 unnamed protein 0.016 -4.22 gi|159030131 argininosuccinate synthase argG 0.008 -4.33 gi|159029341 aldo/keto reductase unnamed protein 0.039 -4.33 gi|159029969 carbamoyl phosphate synthase small subunit carA 0.004 -4.67 gi|159029694 branched-chain alpha-keto acid dehydrogenase subunit E2 unnamed protein 0.003 -4.78
2.00E- gi|159030589 bicarbonate transport system substrate-binding protein unnamed protein -4.83 04 gi|159025989 aspartate aminotransferase unnamed protein 0.002 -5.5 gi|159030372 glutamate--ammonia ligase glnA* 0.014 -5.56 5.00E- gi|159029886 malic enzyme unnamed protein -5.67 04 6.00E- gi|159028923 30S ribosomal protein S1 unnamed protein -5.89 04 7.00E- gi|159030479 periplasmic carboxyl-terminal protease unnamed protein -6 04 gi|159028626 6-phosphogluconate dehydrogenase gnd 0.001 -6 gi|159028535 60 kDa chaperonin GroEL2 groL2 0.01 -6.84 gi|159029281 periplasmic binding protein unnamed protein 0.002 -7.05 gi|159026334 SOS-response transcriptional repressor LexA unnamed protein 0.021 -7.33 5.00E- gi|159029286 leucyl aminopeptidase pepA -7.67 04
111 Chapter 4
Protein ID Accession Protein ID (NIES-843) T-Test Expression (PCC 7806)
8.00E- gi|159027880 chloroplastic outer envelope membrane protein unnamed protein -9.44 04 2.00E- gi|159027523 photosystem I subunit III psaF -15.33 04 gi|159028802 F0F1 ATP synthase subunit beta atpD* 0.024 -19.83 gi|159029290 cell division protein ftsH3* 0.004 down gi|159029342 recombinase A recA* 0.004 down gi|159030526 glutathione reductase gor 0.017 down gi|159027205 F0F1 ATP synthase subunit alpha atpA* 0.045 down gi|159028755 apocytochrome f petA 0.048 down
* denotes thioredoxin targets identified previously in cyanobacteria (Lindahl and Kieselbach, 2009, Mata- Cabana, et al., 2007, Perez-Perez, et al., 2009)
4.3.4. Protein expression in iron-starved cultures of the genetically engineered non- toxic M. aeruginosa PCC 7806 mcyH-
To determine whether differences in protein expression were due to a strain-specific, but not toxicity-associated response to iron starvation, the mcyH- strain was grown in the same conditions and its proteome was compared to both the toxic PCC 7806 and the non-toxic PCC 7005. Interestingly, many more proteins were identified in this strain, with 737 shared between the two iron concentrations investigated. Of these, only five proteins were unique for cultures in 1000 Fe, whereas 60 proteins were differentially expressed in 10 nM Fe. The proteins that were significantly affected by iron starvation, were surprisingly different from those found in both the toxic and non-toxic wild-type strains described above (Table 4.3.5).
Similar to the non-toxic PCC 7005, photosystem I proteins were down-regulated. These were represented by the reaction centre components PsaA and PsaB, and a similar response was observed for the oxidative damage-sensitive photosystem II D1 protein. ApcB1, the beta subunit of allophycocyanin was decreased, whereas ApcE was increased in both the mutated and wild-type strain. Despite these PCC 7005-like modifications of the photosynthetic machinery, ATP synthase subunits were increased in abundance as similar to the PCC 7806 wild-type (Table 4.3.3). Several proteins involved in amino acid synthesis, translation and cell division were also increased under conditions of low iron.
112 Iron stress proteomics
In contrast to the wild-type strains, few proteins involved in carbon metabolism were found to be differentially displayed to a significant level. The lack of McyH expression also seemed to affect the choice of iron imported into the cell, as the expression of the ferrous uptake protein FeoA increased, in contrast to the up-regulation of the ferric- binding protein in the toxic PCC 7806.
In terms of oxidative stress protection, IsiB expression was similar to that of wild-type PCC 7806, but the expression of a HLIP-superfamily protein was reversed. A heat shock DnaJ-like protein was also differentially expressed, but contrary to expectations, it was down-regulated when the cells were under iron stress. Here, a microcystin-related protein MrpA was observed to be down-regulated under iron starvation, consistent to the results obtained for high light in the wild-type PCC 7806, which also induces oxidative stress (Dittmann, et al., 2001). However, this protein was not found in the mcyB- mutant in the same study. Another M. aeruginosa-characteristic protein was the RuBisCo-like protein that has been proposed to be involved in the methionine salvage pathway and prevention of oxidative stress (Carre´-Mlouka, et al., 2006). This protein was also found to be down-regulated and was only detected in the mcyH- knockout strain.
Several proteins which have been found to interact with thioredoxin in cyanobacterial studies were also differentially expressed (Table 4.3.1-5). In PCC 7005, these were generally down-regulated in iron starved cells, with the exception of fructose- bisphosphate aldolase (Fba) and the phycocyanin beta subunit (CpcB). This is consistent with the decrease of thioredoxin reductase, which would cause a depletion of the reduced thioredoxin pool and cause thioredoxin targets to be reduced and activated. On the other hand, thioredoxin targets in PCC 7806 were up-regulated relative to the control, except for NrtA.
113 Chapter 4
Table 4.3.5. Proteins differentially expressed (p < 0.05, t-test) in iron-starved and iron-replete cultures of M. aeruginosa PCC 7806 mcyH-. Proteins are identified according to annotation in the PCC 7806 and NIES 843 genomes. Expression is relative to 1000 nM Fe, protein IDs in bold are entries not found in the wild-type PCC 7806. Up: proteins expressed in 10 nM Fe only; down: proteins expressed in 1000 nM Fe only. Protein ID (PCC Accession Protein ID (NIES-843) 7806) T-Test Expression ^ gi|159028112 no homologue isiB 0.0048 Up gi|159028085 no homologue unnamed protein 0.038 Up gi|159027998 hypothetical protein MAE 06000 unnamed protein 0.0005 5.8 gi|159028839 leader peptidase I unnamed protein 0.0037 5.33 gi|159029167 RNA polymerase beta subunit rpoB 0.0063 5.33 gi|159030672 no homologue unnamed protein 0.026 4.67 gi|159028082 ferrous iron transport protein A unnamed protein 0.0052 4.38 gi|159028628 reticulocyte binding protein like unnamed protein 0.022 4.22 gi|159030039 orotate phosphoribosyltransferase pyrE 0.0044 4 gi|159027682 carbamoyl phosphate synthase large subunit unnamed protein 0.023 3.88 gi|159027206 F0F1 ATP synthase subunit delta atpH 0.02 3.87 gi|159028854 RNA-binding S4 unnamed protein 0.012 2.67 gi|159027818 hypothetical protein MAE 45820 unnamed protein 0.032 2.6 gi|159030630 periplasmic solute binding protein unnamed protein 0.038 2.6 gi|159029662 pterin-4-alpha-carbinolamine dehydratase unnamed protein 0.026 2.47 gi|159028548 FtsZ cell division protein ftsZ 0.0025 2.21 gi|159029714 threonine synthase unnamed protein 0.047 2.14 gi|159027207 F0F1 ATP synthase subunit B atpF 0.041 2.02 gi|159026457 succinate dehydrogenase flavoprotein subunit frdA 0.031 2 gi|159029148 ribulose-phosphate 3-epimerase rpe 0.034 2 ^ gi|159026883 phycobilisome core component apcE* 0.045 1.9 glycine cleavage system aminomethyltransferase gi|159028726 T gcvT 0.022 1.73 gi|159029853 50S ribosomal protein L9 rplI 0.0099 1.71 gi|159027807 NAD-dependent epimerase/dehydratase unnamed protein 0.0063 1.5 rubisco-like gi|112821115 2,3-diketo-5-methylthiopentyl-1-phosphate enolase protein 0.043 -1.2 gi|159027261 no homologue mrpA 0.044 -1.24 ^ gi|159027374 CAB/ELIP/HLIP superfamily protein unnamed protein 0.028 -1.33 gi|159029241 type 2 NADH dehydrogenase unnamed protein 0.033 -1.33 gi|159027978 allophycocyanin beta subunit apcB1 0.0078 -1.4 gi|159030973 hypothetical protein MAE 47530 unnamed protein 0.039 -1.56 gi|159027821 hypothetical protein MAE 45790 unnamed protein 0.049 -1.58 gi|159027866 hypothetical protein MAE 43685 unnamed protein 0.042 -1.78 gi|159029097 hypothetical protein MAE 21680 unnamed protein 0.026 -1.89 gi|159030511 NAD-dependent epimerase/dehydratase rfbE 0.003 -2 gi|159026017 putative acyl-carrier protein reductase unnamed protein 0.003 -2 gi|159029500 D-alanyl-alanine synthetase A unnamed protein 0.0086 -2 gi|159030594 hypothetical protein MAE 20050 unnamed protein 0.0086 -2.1 gi|159026379 short-chain dehydrogenase/reductase unnamed protein 0.027 -2.11 photosystem I P700 chlorophyll a apoprotein gi|159030976 A2 psaB 0.0043 -2.22 gi|159026736 photosystem II D1 protein unnamed protein 0.047 -2.42 gi|159027224 hypothetical protein MAE 06820 unnamed protein 0.024 -2.5 gi|159027145 chromosome partitioning protein unnamed protein 0.027 -2.5 gi|159027820 hypothetical protein MAE 45800 unnamed protein 0.032 -2.67 gi|159029503 heat shock protein DnaJ-like unnamed protein 0.037 -2.67 gi|159030975 photosystem I P700 chlorophyll a apoprotein psaA 0.012 -3
114 Iron stress proteomics
Protein ID (PCC Accession Protein ID (NIES-843) 7806) T-Test Expression gi|159027960 hypothetical protein MAE 44890 unnamed protein 0.012 -3.43 gi|159028356 CP12 polypeptide unnamed protein 0.033 Down
* denotes thioredoxin targets identified previously in cyanobacteria (Lindahl and Kieselbach, 2009, Mata- Cabana, et al., 2007, Perez-Perez, et al., 2009) ^proteins with the same expression pattern in PCC 7806 and PCC 7806 mcyH- during iron stress
4.4. Discussion
In Chapter 2 the proteomes of toxic and non-toxic strains of M. aeruginosa were studied under nutrient-replete conditions and several proteins were found to be differentially expressed. These included elements of the carbon dioxide concentrating mechanism (CcmK3 and CcmL), nitrogen metabolism (NrtA and PII), redox balance (NdhK, MAE 27560, TrxM and the cotranscribed MAE 06820) and a carboxymethylenebutenolidase (Chapter 2). Differences in these processes during exponential growth in a nutrient-rich environment may affect the growth of cyanobacteria in response to starvation and stress conditions, as was observed in Chapter 3, when transcription of genes involved in the iron stress response of toxic and non-toxic strains was compared. Here, proteomic data from three strains of M. aeruginosa show that, although they are all capable of surviving iron limitation, the protein expression changes that result in this adaptation are strain-specific and, in the major part, were not influenced by the presence of microcystin.
4.4.1. Photosynthesis and respiration
In this study, the three M. aeruginosa strains differed in their photosynthetic machinery both before and during iron starvation. In iron-replete Fraquil* media, cells of both non- toxic strains expressed a larger amount of phycobilisome proteins. With the exception of CpcG, both PCC 7005 and PCC 7806 mcyH- had a different suite of proteins that were differentially expressed in comparison to PCC 7806. In particular, mcyH- had a higher abundance of several PSI and PSII proteins. These patterns of protein expression will result in more and larger phycobilisomes synthesized in the non-toxic strains. This is consistent with the up-regulation of ATP synthase subunits observed in conditions where photosynthetic proteins were found to be more abundant (non-toxic strains in
115 Chapter 4
1000 nM Fe and PCC 7806 in 10 nM Fe) and ATP synthesis is expected to be increased as a result of more active photosynthesis.
This could lead to more energy absorption and thus, increased ROS formation during the initial stages of iron starvation when the phycobilisome is modified, but not yet degraded. In addition, for mcyH-, the higher abundance of PSI and PSII proteins in 1000 nM Fe observed here, would act as a major sink for iron and decrease the intracellular reserves of this metal once iron becomes limited. Furthermore, rubredoxin was decreased in both non-toxic strains. This iron-containing protein has been shown to be involved in the synthesis of a functional PSI complex (Shen, et al., 2002a, Shen, et al., 2002b), and the observed interstrain differences in the expression of this protein may affect the electron transfer between PSII and PSI.
The photosynthetic machinery is one of the primary targets for modification during iron stress and therefore it can be expected that the background differences in photosystem protein expression in the three strains used here would lead to a difference in the iron stress response. One of the most important physiological changes that takes place during iron starvation in cyanobacteria is chlorosis, during which the pigment content decreases, the phycobilisome is degraded and a decrease in PSI/PSII occurs (Schwarz and Forchhammer, 2005). The end result is prevention of the harvesting of light energy that would otherwise not be utilised further and wold cause the generation of ROS. This is also associated with a reduction in the abundance of required iron-containing proteins (in particular subunits of the iron-rich PSI) and some recycling of sulfur and nitrogen (Dolganov and Grossman, 1999).
The differential analysis performed here revealed protein expression changes that would lead to an effective decrease in the PSI/PSII ratio. However, in each strain the mechanism by which this was achieved was different (Figure 4.4.3). For example, in PCC 7806 and PCC 7005, the observed reduction of the PSI PsaF subunit, involved in trimerisation of the complex, would lead to an increase in PSI monomers (Shen, et al., 2002b). This effect would be enhanced in PCC 7005 by the reduced abundance of PsaL, which plays a role in the interaction of PSI with the antenna (Ivanov, et al., 2006). The mcy mutant, mcyH- was the only strain in which the core subunits of PSI and the ROS- sensitive D1 subunit of PSII were affected, suggesting that this strain was under the most severe iron and oxidative stress. This may have resulted from the higher abundance of these proteins relative to the wild-type strain, when the cells were 116 Iron stress proteomics maintained in 1000 nM Fe. As a result, upon transfer to 10 nM Fe, the photosynthetic machinery would need to be rapidly degraded in order to prevent an excessive amount of reactive oxygen species being formed.
The three strains also differed dramatically in terms of phycobilisome degradation that occurred in response to iron stress. The decrease in linker polypeptides observed in PCC 7005 and the allophycocyanin core in mcyH-, could result in smaller and more functionally dissociated phycobilisomes. However, this was not the case with PCC 7806. In this toxic M. aeruginosa strain, most of the differentially displayed phycobilisome proteins increased in abundance during iron stress. This response is not typical of cyanobacteria, but reflects the lack of bleaching in the iron-stressed PCC 7806 cells observed in Chapter 3 (Figure 3.3.1). Such failure to modify the phycobilisome could result in accumulation of excess energy and oxidatative stress (Michel and Pistorius, 2004). In the case of PCC 7806, the damage to the photosynthetic apparatus could be prevented by the observed accumulation of the orange carotenoid protein relative to iron-replete cells. This protein associates with the phycobilisomes and aids in the conversion of excess energy to heat in a process called non-photochemical quenching (Bailey and Grossman, 2008). The increase in phycobilisome content in iron-stressed PCC 7806 cells also conflicted with the increase in abundance of NblB. The NblB protein is constitutively expressed and appears to maintain the appropriate phycocyanin/chlorophyll ratio during environmental flux, resulting in ordered phycobilisome degradation when co-expressed with NblA (Dolganov and Grossman, 1999). Since differential expression of NblA was not observed here, it is unlikely that the up-regulation of NblB in PCC 7806 would cause phycobilisome disassembly. The observed increase in NblB, but not NblA, in the mcyH- mutant grown in 1000 nM Fe suggests a possible yet unresolved function of this protein in phycobilisome stability.
Iron starvation is also likely to cause changes in the expression of the iron-rich cytochrome b6/f, cytochrome c6 and ferredoxin, with their corresponding replacement by the iron-free plastocyanin and flavodoxin, respectively (Latifi, et al., 2008, Michel and Pistorius, 2004). This was observed with the iron-starved wild-type strains. However, in the mcyH- strain both flavodoxin and plastocyanin were present at a higher level compared to PCC 7806 in 1000 nM Fe and suggests iron- or oxidative stress is being experienced by mcyH- at this higher iron concentration.
117 Chapter 4
118 Iron stress proteomics
4.4.2. Carbon-nitrogen balance and thioredoxin
Several proteins involved in maintaining carbon-nitrogen balance in the cell were differentially expressed when M. aeruginosa strains were grown in iron limitation. As described in Chapter 1, photosynthesis in cyanobacteria is closely linked to carbon and nitrogen metabolism. A critical step in this process is carried out by the glutamate-- ammonia ligase (glutamine synthase) GlnA, which uses ammonia, 2-oxoglutarate (2OG), and ATP generated through photosynthesis to form glutamine (Su, et al., 2005). Therefore, sensing the amount of 2OG in the cells is crucial for the integration of photosynthesis and nitrogen assimilation, with higher 2OG amounts being indicative of a higher photosynthetic rate and the need to take up more nitrogen in order to balance the C/N ratio of the cell (Su, et al., 2005). This process, including the GlnA protein, is regulated by the reciprocal action of the NtcA and PII proteins and their competitive binding to 2OG (Espinosa, et al., 2006, Osanai and Tanaka, 2007).
In the PCC7806 mcyH- strain, an alternative glutamine synthase encoded by GlnN was up-regulated relative to wild-type. This isoform of GlnA is expressed only in nitrogen- starved conditions in Synechococcus sp. PCC 7942 and Synechocystis sp. PCC 6803 and aids the recovery of chlorotic cells. The glnN promoter contains a non-canonical NtcA binding site and is inhibited by PII (Paz-Yepes, et al., 2003, Schwarz and Forchhammer, 2005). Thus, the increased GlnN levels observed here, together with the increase in subunits of the nitrate ABC transporter (NrtA and NrtC) revealed higher NtcA activity in the non-toxic mutant strain compared to wild-type PCC 7806. Still, this contradicts the low abundance of GlnA and may indicate disrupted PII-NtcA regulation in the mutant strain. The lack of response of these proteins to iron limitation in mcyH-, as was observed in the wild-type strains, may also point to a disrupted regulatory process.
Regulation of transcription by NtcA is a redox-sensitive process with many in vitro studies using DTT to enhance promoter recognition by this transcriptional regulator (Ginn and Neilan, 2010, Jiang, et al., 1997, Wisen, et al., 2004). Therefore, it is not surprising that during iron starvation and the potential increase in intracellular ROS, the expression level of many NtcA-regulated proteins was affected. In iron-limited M. aeruginosa PCC 7806, a decrease in the photosynthetic rate due to a lower PSI/ PSII ratio (Figure 4.4.3) would decrease the intracellular 2OG concentration (Su, et al., 2005). It is possible that this would result in accumulation of nitrate in the cells and
119 Chapter 4 inhibition of both nitrate and bicarbonate transport (Koropatkin, et al., 2006). The increased levels of Calvin cycle enzymes could also be a strategy to balance the C/N ratio. On the other hand, the decrease of NAGK in the iron-starved PCC 7005 would be indicative of a high C/N ratio requiring the down-regulation of carbon-fixation proteins and enhanced NtcA activity. This is in agreement with the observed decrease in the expression of Gor, one of the few targets that has been reported as being repressed by high levels of NtcA (Jiang, et al., 1997). However, rather than relying on nitrate transport, PCC 7005 could be utilizing pre-existing nitrogen reserves, as indicated by the up-regulation of enzymes involved in the synthesis of the amino acid components of cyanophycin, Asp and Arg, at 1000 nM Fe.
Further redox control of carbon-nitrogen assimilation in cyanobacteria is established via the action of the regulatory polypeptide CP12 (Tamoi, et al., 2005, Trost, et al., 2006). This is due to the binding and inactivation of CP12 to the glyceraldehyde-3-phosphate dehydrogenase/ phosphoribulokinase (GAPDH/PRK) complex in the dark. This results in the conversion of ribose-5-phosphate to xylose-5-phosphate by ribose phosphate isomerase in the oxidative pentose pathway rather than to ribulose-1,5-phosphate in carbon fixation (Tamoi, et al., 2005). The mcyH- strain appeared to follow such CP12 regulation with the levels of this protein higher in 1000 nM Fe, but low in iron-starved cells, where ROS may be produced at similar rates to cells exposed to high light. On the other hand, CP12 was up-regulated in iron-limited PCC 7005 and could therefore inactivate the GAPDH/PRK complex. This may be an attempt by the PCC 7005 cells to reduce carbon utilisation in order to match the decreased photosynthetic rate, ATP production and nitrogen metabolism. Such an effect would be enhanced by the decrease in LexA, a protein that in heterotrophic critical for cell viability in ncyanobacteria and it is negatively regulated by UV-C, heavy metal stress and H2O2 (Domain, et al., 2004). This negative regulation by oxidative stress-inducing conditions is confirmed by the proteome expression of M. aeruginosa in iron-limited culture. Previously Domain et al. carried out a microarray analysis of a LexA Synechocystis sp. PCC 6803, revealed a reduction of transcription of genes involved in carbon assimilation and uptake, as well as periplasmic proteins, iron-binding proteins and nitrate transporters, consistent with the results observed here (Domain, et al., 2004) (Figure 4.4.3).
The redox regulation of CO2 fixation is facilitated by the localization of a large proportion of the Calvin cycle enzymes in proximity to the thylakoid membranes, where
120 Iron stress proteomics they can rapidly react to changes in ATP and NADPH levels generated during photosynthesis (Agarwala, et al., 2009). These processes are mediated by thioredoxin, which targets several of the enzymes involved in carbon assimilation, including CP12, phosphoglycerate kinase, glyceraldehyde-3-phosphate dehydrogenase and fructose-1,6- bisphosphatase (Perez-Perez, et al., 2006, Trost, et al., 2006). Thioredoxin M was reported to be differentially displayed in nutrient-replete microcystin-producing and non-toxic strains of M. aeruginosa in Chapter 2, and here it appears that the expression levels of thioredoxin peroxidase were also variable depending on toxicity. Activation of thioredoxin, and therefore thioredoxin targets in the absence of thioredoxin peroxidase is consistent with the increase in photosynthetic and carboxysome components in the iron-replete non-toxic M. aeruginosa strains here. This was also the case with PCC 7806 in low-iron media, consistent with the observed lack of bleaching of the cells and the increase in phycobilisome protein levels. On the other hand, the decrease in thioredoxin reductase in the iron-starved PCC 7005 would inhibit reactivation of oxidized thioredoxin and decrease the activity of thioredoxin targets, such as those involved in carbon fixation.
4.4.3. Iron transport, iron storage and protection from oxidative stress
Peroxiredoxin is activated by thioredoxin, thus the increased expression of this enzyme was expected given the predicted increase in thioredoxin activity in this strain (Latifi, et al., 2008). Iron starvation in the non-toxic strains also resulted in the increase of heat- shock proteins, which have previously been identified as up-regulated in oxidative stress (Li, et al., 2008, Pandhal, et al., 2007) and may signify ROS accumulation in these strains. In 1000 nM Fe, both non-toxic strains had higher RecA levels relative to PCC 7806. This protein is involved in DNA repair and is negatively regulated by UV-C light and H2O2 (Domain, et al., 2004). Thus, an increase in hydrogen peroxide as a result of iron limitation may explain the lower abundance of RecA in starved PCC 7005 cells. On the other hand, peroxiredoxin expression in PCC 7806 was increased, and could effectively detoxify cells of the H2O2 generated during photosynthesis and prevent ROS damage to the photosystem.
Several differences in iron transport and storage of acquired iron were observed in the strains studied here and may be crucial for understanding their iron-stress responses. In iron-replete conditions cells of mcyH- seemed to show a preference for the uptake of ferrous iron, as illustrated by the higher abundance of the ferrous uptake transporter
121 Chapter 4 subunit FeoA relative to PCC 7806. This protein is expressed only during severe starvation in cyanobacteria, whereas the ferric uptake transporter (Feo) is usually preferred (Katoh, et al., 2001). However, Fe2+ can be generated either by photochemical reduction or enzymatically via the action of a membrane-bound oxidoreductase and by the photosynthetic generation of superoxide (Barbeau, et al., 2001, Fujii, et al., 2010). This is interesting, given the observation that in the mcyH- mutant a putative oxidoreductase was up-regulated in 1000 nM Fe, and could be involved in the reduction of iron in the periplasmic space.
The preference for Fe uptake in PCC 7806 and Fe(II) uptake in mcyH- continued when the cells were transferred to 10 nM Fe. This highlights the fact that the iron species transported is a strain-specific phenomenon rather than one determined by iron concentration in the media. These results were also consistent with the findings in Chapter 3 where FeoB transcription was significantly up-regulated only in iron-starved mcyH- cells (Table 3.3.1). The different mode of iron uptake in the toxic wild-type and mcy mutant strains was interesting given previous hypotheses that microcystin could act as a siderophore and the proposed function of McyH as a toxin-associated transporter (Humble, et al., 1997, Pearson, et al., 2004, Saito, et al., 2008, Utkilen and Gjolme, 1995). However, the microcystin molecule cannot compete for iron binding with the EDTA present in the Fraquil* media (Manabu Fujii, personal communication) and is unlikely to be an efficient siderophore.
In the PCC 7005 strain iron transport was not affected at either 10 nM or 1000 nM Fe and this strain had the lowest abundance of the iron-containing bacterioferritin co- migratory protein and the DNA-binding stationary phase protection protein at 1000 nM Fe. This result provides support for an earlier hypothesis that non-toxic strains may be forced to maintain lower concentrations of intracellular iron and are thus affected by iron starvation to a larger extent compared to microcystin producers (Utkilen and Gjolme, 1995). However, the observations on mcyH- reported here, suggest that wild- type non-toxic strains have adapted their iron requirements to a microcystin-free lifestyle. This process is not likely to have occurred in genetically engineered mcy mutant strains that have been regularly maintained in nutrient excess.
122 Iron stress proteomics
4.4.4. Other, hypothetical and PCC 7806-genome specific proteins
Several differentially expressed hypothetical and unknown proteins were and therefore may be involved in the stress response of M. aeruginosa. Importantly, differential regulation by light treatments, including the microcystin related protein A (MrpA) and the RuBisCo-like protein encoded in the M. aeruginosa PCC 7806 genome, has been reported previously and may be involved in the oxidative stress response in this organism (Carre-Mlouka, et al., 2006, Dittmann, et al., 2001). This protein was found in the wild-type strain, but not in a mcyB- mutant, and was found to be responsive to blue light (Dittmann, et al., 2001). By comparison, the MrpA protein was expressed in the mcyH- strain at higher levels compared to PCC 7806 here, suggesting that different mcy mutant strains may have different downstream effects due to the inserted chloramphenicol-resistance cassette. Nevertheless, the expression of MrpA was decreased in iron-starved cells of mcyH-, in agreement with the reported decrease in transcript of this gene following potentially ROS-inducing high light treatment (Dittmann, et al., 2001). The RuBisCo-like protein encoded in the PCC 7806 but not the NIES-843 genome, followed the same trend in expression as MrpA and gives support to the proposed involvement of this RuBisCo homologue in the oxidative stress response (Carre-Mlouka, et al., 2006).
A number of hypothetical proteins were differentially expressed between the non-toxic strains and PCC 7806, as well as within the same strain cultured in different iron concentrations. These proteins may be involved in the iron starvation/oxidative stress response in M. aeruginosa, but have not been shown to be regulated by the cellular redox status in any other cyanobacteria. Still, MAE 43685 is homologous to a universal stress protein and MAE 10050 contains a thioredoxin-like fold, whereas MAE 06820, shown to be co-transcribed with thioredoxin M in Chapter 2, was significantly decreased in iron-starved mcyH- cells and may contribute to thioredoxin activity. Several of the identified hypothetical proteins contain predicted transmembrane domains and may be located on the plasma and/or thylakoid membranes performing a role in photosynthesis or ion transport. No doubt, as more of these proteins are characterised, our knowledge of the metabolic changes occurring in bloom-forming cyanobacterial species, such as M. aeruginosa in iron limitation, will increase.
123 Chapter 4
4.4.5. Toxicity and transcriptional control of the iron stress response
In Chapter 2, the levels of the PII protein were higher in non-toxic strains relative to toxic M. aeruginosa, whereas the expression of the nitrate transporter NrtA was lower (Table 2.3.2). In the proteomes of strains grown in Fraquil* media, the levels of GlnA and NrtA, two proteins negatively regulated by PII (Espinosa, et al., 2006, Osanai and Tanaka, 2007, Su, et al., 2005), were decreased in non-toxic strains consistent with increased PII activity in non-toxic M. aeruginosa. In support of these observations, in the non-toxic PCC 7005 strain, the expression of the PII-interacting protein N-acetyl glutamate kinase (NAGK) was increased. The activation of PII and subsequent inhibition of NtcA in PCC 7005 may also have been further enhanced by the up- regulation of the transcriptional regulator ArbB, which has previously been shown to affect NtcA activity (Ishii and Hihara, 2008).
In the previous chapter, toxicity was suggested to affect the transcription and activity of the ferric uptake regulator proteins FurA-C based on observations of iron transport and photosystem modification (Figure 3.4.1). Despite the changes in Fur transcription under iron starvation described in Chapter 3, none of the Fur homologues in M. aeruginosa were found to be significantly altered at the protein level. This may be due to the low levels of expressed Fur proteins being obscured by higher-abundance proteins such as phycobiliproteins. However, many of the proteins found to be differentially displayed in limited iron were targets of Fur, including FeoB, IsiB and FutA. In addition, the potential involvement of microcystin synthesis on the activity of NtcA in M. aeruginosa strains was proposed in Chapter 2, as well as this chapter, based on proteome and redox changes affecting NtcA targets. If this is the case, it is possible that rather than the toxin itself, NtcA could be modulating Fur expression and transcriptional control in iron- stressed M. aeruginosa. Indeed, NtcA has been shown to bind to, and regulate Fur in the microcystin producing Anabaena sp. PCC 7120, while NtcA-binding sites have been found in many promoters recognised by Fur, including the mcyS promoter, suggesting cooperative control by these two transcription factors (López-Gomollón, et al., 2007a, Lopez-Gomollon, et al., 2007b). Several of the genes with promoters containing in silico predicted NtcA and Fur binding sites (including PsaL, Gor and RbcL) were recognized as differentially displayed under the iron stress conditions used here. Importantly, the trxA promoter is also a predicted target for FurA-NtcA co-regulation and as discussed earlier, thioredoxin regulated enzymes appear to be significantly
124 Iron stress proteomics affected in all the M. aeruginosa strains studied here (López-Gomollón, et al., 2007a). However, it should be noted, that few of these predicted interactions between Fur, NtcA and target promoters have been validated experimentally, or in multiple cyanobacterial species and warrants further investigation.
The multiple differences in protein expression between the three strains during iron- replete conditions, are important to consider for future studies of M. aeruginosa that attempt to relate toxicity to other protein or genetic markers. Studies of mutant mcy strains have not been successful in identifying reliable biomarkers for toxicity production other than the mcy genes themselves. Although in this chapter several mcy- mutant associated changes, including increased number of gas vesicles, and changes in the expression of MrpA and the RuBisCo-like protein confirm previous observations (Yoshida, et al., 2008), these changes were not evident in PCC 7005. The reason for this may be that PCC 7005 is likely to have been non-toxic for a long period of time in the environment and would have been exposed to changing conditions that require appropriate adaptation in order to survive selection. On the other hand, strains such as mcyH- have lost their ability to synthesise microcystin only recently and are generally maintained in nutrient-rich culture media and are unlikely to have experienced iron- stress during their lifetime. Thus, the response of a recently mutated strain may not reflect the processes that occur in natural M. aeruginosa mutants isolated from the environment.
4.5. Conclusions
The proteome studies of M. aeruginosa strains subjected to iron stress identified several trends in protein expression that could explain the differential response of toxic and non-toxic strains of this cyanobacterial species when placed in an iron-limited environment. Changes in the photosynthetic proteins in the toxic PCC 7806 appeared to involve mostly stabilisation and increase in the size of the phycobilisome, whereas in non-toxic strains iron stress caused reduced abundance of core photosystem subunits. In addition, iron uptake, carbon and nitrogen metabolism in toxic and non-toxic cells were different, possibly as a result of higher thioredoxin activity in the microcystin-producing strain. Several targets of thioredoxin, and the transcriptional regulators, NtcA and Fur were also identified in this differential display providing more circumstantial evidence that microcystin synthesis may affect the activity of these proteins.
125 Chapter 4
126
Chapter 5 Proteomic analysis of the nitrogen stress response in the bloom-forming cyanobacteria Microcystis aeruginosa
Chapter 5
128 Nitrogen stress proteomics
5.1. Introduction
Previous studies on microcystin regulation have shown that two transcriptional factors, the iron-responsive FurA and the global nitrogen regulator NtcA, bind to the mcyS promoter and may affect transcription of toxin genes under nutrient flux (Ginn and Neilan, 2010, Martin-Luna, et al., 2006b). The involvement of FurA and iron availability in relation to toxin synthesis was studied in Chapters 3 and 4 at the transcriptional and protein expression level. The increase in toxicity during growth in iron-limited conditions and the differential response of toxic and non-toxic strains to iron stress was consistent with observations previously reported in the literature (Briand, et al., 2008, Martin-Luna, et al., 2006, Sevilla, et al., 2008).
The response of M. aeruginosa to iron stress was not limited to up-regulation of iron uptake and storage, but extended to proteins involved in maintaining the cellular redox- and carbon-nitrogen balance (Chapter 4). The pattern of protein expression was suggestive of higher NtcA activity in the toxic strain PCC 7806 relative to non-toxic strains, and was hypothesised to stem from the redox sensitivity of NtcA and the cross- regulation of NtcA and FurA reported previously (López-Gomollón, et al., 2007a). In addition, a difference in NtcA activity was proposed to be involved in the differential protein display of toxic and toxigenic, but inactive-toxin producing strains of M. aeruginosa under nutrient-replete conditions (Chapter 2). Therefore, further investigation of NtcA regulation in relation to microcystin synthesis was necessary.
Nitrogen availability has been shown to have a conflicting effect on microcystin synthesis and has generally been attributed to differences in the growth rate of cells (Dai, et al., 2008, Downing, et al., 2005, Vezie, et al., 2002, Waal, et al., 2009). Given the main function of NtcA is to control nitrogen uptake and utilization, it is expected that growth in nitrogen-limited media will have the largest effect on the expression of this transcriptional regulator. In this chapter, three strains of M. aeruginosa were grown in nitrogen-replete and nitrogen-free media and the regulatory effect of nitrogen limitation on toxicity and NtcA activity in these unicellular cyanobacteria was assessed.
129 Chapter 5
5.2. Methods
5.2.1. Microcystis strains, culture growth and protein extraction
Stock cultures of Microcystis aeruginosa PCC 7806 (toxic), PCC 7005 (non-toxic) and PCC 7806 mcyH- (non-toxic mutant) were grown in BG11 broth. The media was supplemented with 100 μg ml-1 cyclohexamide for all strains, and 10 μg ml-1 chloramphenicol for the PCC 7806 mcyH- strain.
Fifty milliliters of these cells were then washed with 10 volumes of BG110 * supplemented with ferric citrate, instead of ferric ammonium citrate (BG110 ) (God, et al., 1998) (Appendix B, Table A.2.2), and used to prepare triplicate cultures in BG11 * o and BG110 of each strain. All cultures were maintained at 28 C without shaking under a 14:10 light/dark cycle (25 )M photons m-2 s-1) supplied by cool white fluorescent lamps. The cells were collected for protein extraction after 8 days of culture. Cells were partially lysed by freeze/thaw cycles in liquid nitrogen and 37oC, respectively. After resuspension in acid extraction buffer (Herbert, et al., 2006) the cells were disrupted further by sonication for 3 x 15 s on ice. Insoluble material was pelleted by centrifugation, the supernatant was buffer-exchanged with PBS, treated with Benzonase endonuclease (Sigma) and made up to 1 mM PMSF (Sigma). Protein concentration was determined on pooled biological triplicate samples using a Bradford assay (BioRad, Hercules, CA) and quantitation in parallel with BSA standards on a SDS-PAGE gel.
5.2.2. Microcystin analysis
Concurrently with protein extraction, total microcystin was extracted by diluting 1 ml of culture to a final concentration of 70% methanol. Toxin content normalized to cell number was determined by the protein phosphatase 2A inhibition assay (PPIA) as described previously (Carmichael and An, 1999).
5.2.3. In-gel digestion and nanoLC-MS/MS analysis
For proteomic analysis, 60 μg of protein was separated by 1D SDS-PAGE in a precast 4-20% Criterion Tris-Acetate gel (BioRad) and stained with Coomassie G-250. Proteins were excised in 14 fractions from each lane and the gel pieces were destained with 50 mM NH4HCO3/50% (v/v) acetonitrile (ACN). Two-step reduction and alkylation was performed, with incubation in 15 mM DTT for 30 min at 37oC, followed by 25 mM o iodacetamide for 30 min at 37 C. The gel pieces were washed with ACN/NH4HCO3,
130 Nitrogen stress proteomics dehydrated completely in 100% ACN and proteins were digested overnight in 80 ng sequencing-grade trypsin (Promega). Peptides were extracted in 50% ACN/1% (v/v) formic acid, dried completely and resuspended in 10 μl 1% formic acid/0.05% (v/v) heptafluorobutyric acid (HFBA) solution.
Mass spectrometry was performed in a LTQ-FT MS/MS at the Biological Mass Spectrometry Facility (BMSF), UNSW, with the assistance of Dr. Mark Raftery. Digest peptides were separated by nano-LC using an Ultimate 3000 HPLC and autosampler system (Dionex, Amsterdam, Netherlands). Samples (1 μl) were concentrated and desalted onto a micro C18 precolumn (500 μm x 2 mm, Michrom Bioresources, -1 Auburn, CA) with H2O:ACN (98:2, 0.05% HFBA) at 20 μl min . After a 4 min wash the pre-column was switched (Valco 10 port valve, Dionex) into line with a fritless nano column (75 μ x ~10 cm) containing C18 media (5 μ, 200 Å Magic, Michrom) manufactured according to Gatlin et al. (1998). Peptides were eluted using a linear gradient of H2O:ACN (98:2, 0.1% (v/v) formic acid) to H2O:ACN (64:36, 0.1% (v/v) formic acid) at 350 nl min-1 over 30 min. High voltage (1800 V) was applied to low volume tee (Upchurch Scientific) and the column tip positioned ~ 0.5 cm from the heated capillary (T=250°C) of a LTQ FT Ultra (Thermo Electron, Bremen, Germany) mass spectrometer. Positive ions were generated by electrospray and the LTQ FT Ultra operated in data dependent acquisition mode (DDA).
A survey scan m/z 350-1750 was acquired in the FT ICR cell (Resolution = 100,000 at m/z 400, with an accumulation target value of 1,000,000 ions). Up to the 6 most abundant ions (>3,000 counts) with charge states > +2 were sequentially isolated and fragmented within the linear ion trap using collisionally induced dissociation with an activation q = 0.25 and activation time of 30 ms at a target value of 30,000 ions. M/z ratios selected for MS/ MS were dynamically excluded for 30 seconds.
5.2.4. Data analysis
Peak lists were generated using Mascot Daemon/extract_msn (Matrix Science, London, England, Thermo) using the default parameters, and submitted to the database search program Mascot (version 2.1, Matrix Science). Search parameters were: Precursor tolerance 4 ppm and product ion tolerances ± 0.4 Da; Met(O), Acryl(C) and Carbamidomethyl(C) were specified as variable modification, enzyme specificity was trypsin, 1 missed cleavage was possible and the M. aeruginosa PCC 7806 protein
131 Chapter 5 database (downloaded from NCBI, version 01/11/07) searched. A decoy database was searched to calculate the false discovery rate. Scaffold 2 (Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK) and X! Tandem (www.thegpm.org; version 2007.01.01.1). X! Tandem and Mascot were set up to search the custom M. aeruginosa PCC 7806 proteome database containing 5161 entries, assuming tryptic digestion. Mascot and X! Tandem were searched with a fragment ion mass tolerance of 0.40 Da and a parent ion tolerance of 4.0 ppm. Oxidation of methionine and iodoacetamide derivative of cysteine were specified in X! Tandem as variable modifications. Oxidation of methionine, iodoacetamide derivative of cysteine and the acrylamide adduct of cysteine were specified in Mascot as variable modifications. Peptide identifications were accepted if they could be established at greater than 95.0% probability as specified by the Peptide Prophet algorithm (Keller, et al., 2002). Protein identifications were accepted if they could be established at greater than 99.0% probability and contained at least 1 identified peptide. Protein probabilities were assigned by the Protein Prophet algorithm (Nesvizhskii, et al., 2003). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Differentially displayed proteins were determined by a student’s t-test, with p < 0.05.
5.2.5. RNA Extraction
Parallel to protein extraction, RNA was extracted from the same cultures of Microcystis aeruginosa. Fifty milliliters of cells were pelleted by centrifugation, washed with Milli- Q water and resuspended in 1 ml of Trizol (Invitrogen). The cells were snap-frozen in liquid nitrogen and subsequent steps were performed at 4oC. Cell lysis was achieved by vigorous pipetting, and 400 μl chloroform was added to the lysate. After centrifugation at 13 000 x g for 30 min and the pellets were washed twice with ethanol, before being resuspended in DEPC-treated water (Invitrogen).
5.2.6. DNase treatment
RNA extracts were treated with 3 U of Turbo DNase (Ambion) for 4 h at 37oC. This was followed by a second extraction with Trizol as outlined above. The success of DNA removal was assessed by PCR targeting the 16S rRNA gene using the primers 27F and
132 Nitrogen stress proteomics
809R (Jungblut, et al., 2005). The purity and quantity of RNA were estimated using a Nanodrop spectrophotometer (Nanodrop Technologies).
5.2.7. cDNA synthesis
The FirstStrand random cDNA synthesis kit (Marligen Biosciences) was used for reverse transcription of 500 ng RNA following the manufacturer’s instructions. The incubation conditions were 22oC for 5 min, 42oC for 90 min and 85oC for 5 min.
5.2.8. Quantitative real-time PCR (qRT-PCR) qRT-PCR was used to quantify transcription levels for genes that were involved in gene transcription regulation. Primer sequences are listed in Appendix A. The RNA polymerase subunit C (rpoC1) gene was used as a reference. Transcript levels were quantified by qRT-PCR using the Rotor-Gene 3000 system (Corbett). Reactions were performed in a total volume of 25 μl using 1 μg cDNA, 10 pmol forward and reverse primer and the Platinum SYBR Green qPCR supermix UDG kit (Invitrogen). Two-step cycling was performed with an initial hold of 60oC for 2 min and 95oC for 2 min, followed by 40 cycles at 95oC for 15 s and 60oC for 30 s. The efficiency of amplification for each primer set was determined previously and is stated in the Appendix A (Table A.1.1). Transcript levels were normalized to rpoC transcription and calculated relative to values for nitrogen-replete strains using the 2-Ct method described elsewhere (Pfaffl, 2001). All analyses were performed using biological and technical triplicates.
5.3. Results
5.3.1. Growth and microcystin production in nitrogen-stress and nitrogen replete conditions in M. aeruginosa strains
When M. aeruginosa was cultured in BG11 media, all strains grew exponentially at a similar rate until day 14, when the experiment was terminated (Figure 5.3.1A). In the wild-type non-toxic strain PCC 7005, BG11 media supported a lower cell mass throughout the culturing period. Upon transition to a nitrogen-limited environment * (Figure 5.3.1B), the growth of all strains was also similar. In BG110 media the cells appeared to utilise the available nitrogen within the first 24 h of growth, indicated by an initial increase in cell density, after which they entered a dormant stationary state. In
133 Chapter 5 these nitrogen-limited cells, chlorosis became pronounced after 3 days of culturing (Figure 5.3.1C).
9 A. PCC 7806 PCC 7005 8 PCC 7806 mcyH-
7 cell number cell
10 6 log 5 0.0 2.5 5.0 7.5 10.0 12.5 15.0 Day
9 B. PCC 7806 PCC 7005 8 PCC 7806 mcyH-
7 cell number
10 6 log 5 0.0 2.5 5.0 7.5 10.0 12.5 15.0 Day
C. PCC 7005 PCC7806 mcyH-
nitrogen-replete
nitrogen-stress
Figure 5.3.1. Growth of Microcystis aeruginosa strains in (A). nitrogen-replete and (B.) N-limited conditions. Cells for each strain were grown in triplicate batch culture for 14 days and cell counts were performed using a haemocytometer. C. Chlorosis in cells grown for three days in nitrogen-replete (BG11) and nitrogen-limited (modified BG110) batch culture.
When microcystin concentration was measured by PPIA and the results were corrected for cell density, no significant difference (t-test, p-value < 0.05 n=3) between the
134 Nitrogen stress proteomics toxicity of nitrogen-replete (0.0191 nM microcystin/cell) and nitrogen-starved (0.0169 nM/cell) cultures was observed.
5.3.2. Background protein expression differences in M. aeruginosa strains during nitrogen-replete growth
PCC 7806 PCC 7806 mcyH- PCC 7005 N N0 N N0 N N0 MWM
Figure 5.3.2. Representative 1D SDS-PAGE gel of strains used in this chapter. For each strain, 15 μg of protein was separated by electrophoresis on a 4-20% Criterion gel (BioRad). MWM: PrecisionPlus molecular weight markers (BioRad); N: nitrogen- replete, N0: nitrogen-stress.
In order to investigate protein expression differences that exist in the M. aeruginosa strains during nutrient-replete growth and may have an effect on the analysis of the nitrogen-starved response, the proteomes of the toxic PCC 7806 strain and the non-toxic strains grown in BG11 were compared. The differential display revealed 380 proteins that were significantly different when the wild-type PCC 7806 and mutant PCC 7806 mcyH- strains were compared, and 98 proteins which differed significantly between PCC 7806 and PCC 7005. Of these, 34 proteins were identified in PCC 7005 and PCC 7806 mcyH- as showing the same expression pattern relative to the toxic M. aeruginosa PCC 7806 wild-type (Table 5.3.1 and Table 5.3.2).
135 Chapter 5
Table 5.3.1. Proteins up-regulated in both non-toxic M. aeruginosa strains PCC 7005 and PCC 7806 mcyH- relative to the toxic PCC 7806. Significant t-tests results (p-value<0.05) and fold-change in expression relative to PCC 7806 are shown. NTO: protein present in non-toxic strains only.
PCC 7005 PCC 7806 mcyH- Accession Protein ID (NIES-843) t-test expression t-test expression gi|159026302 photosystem II CP43 protein 0.0034 32.29 0.00084 13 gi|159026514 extracellular ligand-binding receptor 0.000065 NTO 0.0022 NTO gi|159027084 30S ribosomal protein S11 0.049 NTO 0.05 NTO gi|159027205 F0F1 ATP synthase subunit alpha 0.0043 5.68 0.00044 4.09 gi|159027379 D-3-phosphoglycerate 0.00067 33.00 0.035 22.00 gi|159027661 acetylornithine aminotransferase 0.044 4.00 0.0022 4.00 gi|159027742 hypothetical protein MAE 50360 0.032 NTO 0.0022 NTO gi|159027872 hypothetical protein MAE 43600 0.015 20.00 0.0022 4.00 gi|159028011 universal stress protein 0.045 3.60 0.021 3.20 gi|159029243 glycogen phosphorylase 0.045 NTO 0.0022 NTO gi|159029290 cell division protein FtsH3 0.0024 7.25 0.036 4.83 ATP-dependent Clp protease ATPase gi|159029700 subunit 0.011 NTO 0.046 NTO glyceraldehyde-3-phosphate gi|159030256 dehydrogenase 0.028 2.66 0.044 1.92 gi|159030393 phosphoenolpyruvate synthase 0.015 NTO 0.0022 NTO gi|159030459 trigger factor 0.015 NTO 0.021 NTO gi|159030975 PSI P700 chlorophyll-a apoprotein A1 0.013 9.95 0.00068 5.09 gi|159030976 PSI P700 chlorophyll-a apoprotein A2 0.014 11.17 0.00075 6.38 gi|159029272 PHA-specific beta ketothiolase 0.032 NTO 0.0022 NTO
The up-regulated proteins in the non-toxic strains (Table 5.3.1) were predominantly associated with photosynthesis and carbon metabolism. This included the reaction centre proteins of photosystem I and photosystem II, as well as proteins involved in carbon fixation and glycolysis. The carbon-storage polymer polyhydroxyalkanoate (PHA) also appeared to be synthesized to a more significant extent in the non-toxic strains. Two hypothetical proteins, MAE 43600 and MAE 50360 were also up-regulated only in the non-toxic strains.
The majority of proteins down-regulated in non-toxic M. aeruginosa relative to the toxic strain PCC 7806 were involved in nitrate, urea and sulfate transport, as well as assigned as hypothetical and unknown proteins (Table 5.3.2).
136 Nitrogen stress proteomics
Table 5.3.2. Proteins down-regulated in both non-toxic M. aeruginosa strains PCC 7005 and PCC 7806 mcyH- relative to the toxic PCC 7806. Significant t-tests results (p-value<0.05) and fold-change in expression relative to PCC 7806 are shown. TO: protein present in the toxic PCC 7806 strain only.
PCC 7005 PCC 7806 mcyH- Accession Protein ID (NIES-843) t-test expression t-test expression gi|159026358 porin type major outer membrane protein 0.02 -1.15 0.018 -1.97 sulfate transport system substrate binding gi|159026370 protein 0.022 TO 0.024 TO gi|159026673 hypothetical protein MAE 53760 0.04 -4.57 0.048 -64.61 gi|159027109 no homologue 0.016 TO 0.033 -6.19 gi|159027261 no homologue 0.0096 -31.50 0.03 -4.50 gi|159027314 thioredoxin A 0.00065 -10.68 0.014 -2.76 gi|159027635 no homologue 0.00099 -33.80 0.009 -4.57 ABC-type urea transport system substrate gi|159027827 binding protein 0.0078 -2.33 0.008 -2.51 gi|159027924 nitrate transport protein NrtA 0.0066 -2.10 0.009 -3.05 gi|159028057 agmatinase 0.0086 -5.43 0.003 -9.50 gi|159028256 no homologue 0.0035 TO 0.005 -13.00 sulfate transport system substrate binding gi|159028553 protein 0.043 TO 0.046 TO gi|159029008 periplasmic binding protein 0.0073 -3.40 0.004 -3.29 gi|159029281 no homologue 0.017 TO 0.019 -29.00 gi|159029460 hypothetical protein MAE 43980 0.015 -4.67 0.009 -7.00 gi|159030428 no homologue 0.0027 -4.12 0.005 -5.47
5.3.3. Nitrogen stress-induced protein expression changes in M. aeruginosa strains
When proteomic analysis was performed on M. aeruginosa cultured in nitrogen stress or in nitrogen-replete conditions, different functional groups were affected in each strain (Figure 5.3.3). In the toxic PCC 7806, the majority of proteins that exhibited a significant change after transition to a nitrogen-limited media were involved in photosynthesis and translation. On the other hand, in both non-toxic strains, PCC 7005 and PCC 7806 mcyH-, most of the significantly affected proteins were classified as ‘others’ in CyanoBase. These strains also exhibited a more pronounced change in proteins involved in amino acid biosynthesis, compared to the toxic strain. In addition, proteins for purine synthesis, transcription and central intermediary metabolism were differentially displayed in the non-toxic strains grown in altered nitrogen levels, but were not observed in the toxic M. aeruginosa PCC 7806.
137 Chapter 5
Figure 5.3.3. Functional categories of proteins differentially expressed under nitrogen stress in strains of M. aeruginosa. Proteins significantly different after a t- test (p<0.05) between nitrogen-stressed and nitrogen-replete cells are categorized according to CyanoBase.
5.3.4. Protein expression in the toxic M. aeruginosa PCC 7806 in nitrogen-limited batch culture
In the model toxic strain PCC 7806, 312 proteins were identified in the proteome of cells cultured in the two different growth conditions. Of these, 60 were differentially expressed based on nitrogen availability (Table 5.3.3). The majority of differentially expressed proteins were down-regulated in nitrogen limitation, including photosynthetic proteins, as well as those involved in carbon fixation, energy metabolism and amino acid synthesis. This was in agreement with a general decrease in protein synthesis in starved cells characterised by the down-regulation of ribosomal 50S and 30S subunits, the 60 kDa chaperone GroEL and elongation factor Tu.
138 Nitrogen stress proteomics
Table 5.3.3. Proteins differentially expressed (p<0.05, t-test) in cells of M. * aeruginosa PCC 7806 grown in BG110 and BG11. Proteins are identified according to annotation in the PCC 7806 and NIES 843 genomes. Expression is shown as fold change relative to BG11. Down: proteins expressed in BG11 only; up: proteins expressed in nitrogen limited growth (BG110) only.
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression gi|159030630 periplasmic solute binding protein unnamed protein product 0.0130 4.00 phosphate-binding periplasmic gi|159030988 protein pstS 0.0062 3.46 gi|159030269 thioredoxin peroxidase unnamed protein product 0.0120 3.29 gi|159030090 hypothetical protein MAE 15680 unnamed protein product 0.0170 3.20 gi|159027173 no homologue unnamed protein product 0.0180 2.83 gi|159027041 hypothetical protein MAE 01710 unnamed protein product 0.0170 2.50 ABC-type urea transport system gi|159027924 substrate-binding protein unnamed protein product 0.0088 2.02 photosystem I reaction center gi|159027052 protein subunit XI psaL 0.0280 1.84 gi|159030355 hypothetical protein MAE 10050 unnamed protein product 0.0240 1.50 gi|159027979 allophycocyanin alpha subunit apcA 0.0130 1.41 photosystem II reaction centre D2 gi|159026300 protein psbD 0.0260 -1.87 gi|159028755 apocytochrome f petA 0.0380 -1.93 gi|159027827 no homologue unnamed protein product 0.0170 -2.03 gi|159027390 photosystem I subunit II psaD 0.0480 -2.15 gi|159028801 ATP synthase CF1 gamma chain atpC 0.0045 -2.30 gi|159029714 threonine synthase unnamed protein product 0.0110 -2.33 gi|159029228 elongation factor Tu tuf 0.0004 -2.47 gi|159029121 50S ribosomal protein L12 unnamed protein product 0.0065 -2.55 gi|159028396 50S ribosomal protein L11 rplK 0.0092 -3.00 gi|159030372 glutamate--ammonia synthase glnA 0.0380 -3.04 gi|159029497 phycobilisome rod linker protein cpcI 0.0013 -3.21 branched-chain alpha-keto acid gi|159029694 dehydrogenase subunit E2 unnamed protein product 0.0150 -3.25 sulfate transport system substrate- gi|159026673 binding protein unnamed protein product 0.0500 -3.58 gi|159029460 no homologue unnamed protein product 0.0330 -4.33 gi|159028535 60 kDa chaperonin GroEL2 groL2 0.0057 -4.39 gi|159027867 phosphoglycerate kinase pgk 0.0002 -4.77 chloroplastic outer envelope gi|159027880 membrane protein unnamed protein product 0.0280 -4.77 gi|159029283 fructose-1,6-bisphosphatase fbaA 0.0019 -5.00 gi|159028194 50S ribosomal protein L5 rplE 0.0120 -5.11 gi|159030775 60 kDa chaperonin GroEL1 groL1 0.0069 -5.41 glyceraldehyde 3-phosphate gi|159030256 dehydrogenase unnamed protein product 0.0110 -5.67 gi|159026736 (+1) photosystem II D1 protein unnamed protein product 0.0140 -6.00 carbon dioxide concentrating gi|159027939 mechanism protein CcmK2 unnamed protein product 0.0027 -8.33 gi|159025950 ferrodoxin-NADP oxidoreductase unnamed protein product 0.0340 -8.33 gi|159029853 50S ribosomal protein L9 rplI 0.0041 -8.67 photosystem I reaction center gi|159030657 subunit IV PsaE psaE 0.0280 -9.29 putative modulator of DNA gi|159027557 gyrase pmbA 0.0004 down
139 Chapter 5
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression fructose-1,6-bisphosphatase, gi|159029605 glpX-encoded glpX 0.0005 down gi|159029916 hypothetical protein MAE 28060 unnamed protein product 0.0015 down gi|159028256 agmatinase speB2 0.0030 down carbon dioxide concentrating gi|159027942 mechanism protein CcmM ccmM 0.0031 down gi|159029286 leucine aminopeptidase pepA 0.0073 down gi|159030526 glutathione reductase gor 0.0077 down gi|159026670 hypothetical protein MAE 19160 unnamed protein product 0.0130 down gi|159029660 hypothetical protein MAE 39950 unnamed protein product 0.0140 down gi|159026700 no homologue unnamed protein product 0.0210 down 3-ketoacyl-(acyl-carrierprotein) gi|159030539 reductase fabG1 0.0240 down gi|159030620 uroporphyrinogen decarboxylase hemE 0.0240 down gi|159027064 glucokinase unnamed protein product 0.0240 down gi|159027679 dihydrolipoamide dehydrogenase lpdA 0.0240 down phosphoglucomutase/phosphoma gi|159026931 nnomutase unnamed protein product 0.0240 down gi|159027424 50S ribosomal protein L25 unnamed protein product 0.0240 down gi|159030406 hypothetical protein MAE_59000 unnamed protein product 0.0240 down gi|159029059 no homologue unnamed protein product 0.0240 down L,L-diaminopimelate gi|159030764 aminotransferase unnamed protein product 0.0250 down 3-oxoacyl-(acyl carrier protein) gi|159030853 synthase II unnamed protein product 0.0290 down gi|159028193 30S ribosomal protein S8 rpsH 0.0290 down gi|159028190 30S ribosomal protein S5 rpsE 0.0290 down ribulose 1,5-bisphosphate ribulose 1,5-bisphosphate gi|112821111 carboxylase/oxygenase large carboxylase/oxygenase large (+1) subunit subunit 0.0350 down gi|159027474 3-isopropylmalate dehydrogenase leuB 0.0400 down
Although the levels of most differentially expressed photosynthetic components were reduced under limited nitrogen, including photosystem reaction centres (D1, D2, PsaD and PsaE), the PsaL protein involved in the interaction of the photosystem I trimer showed an increased expression. Also, despite the pronounced bleaching of nitrogen- starved cells (Figure 5.3.1) and the reduced expression of the rod-linker polypeptide CpcI, the allophycocyanin alpha subunit ApcA was up-regulated in response to nitrogen stress. Thioredoxin peroxidase expression was also increased in nitrogen-limited culture, but the expression of no other oxidative stress-related proteins was altered.
Several transporters were also significantly affected by nitrogen stress, including the periplasmic solute-binding and phosphate-binding proteins and a urea-binding protein. On the other hand, a sulfate transporter was shown to have lower abundance relative to levels during nitrogen-replete growth.
140 Nitrogen stress proteomics
Approximately 11% of the differentially displayed proteins in this strain were hypothetical. Three of these, MAE 01710, MAE 10050 and MAE 15680 were up- regulated in the stressed cells, as was, a protein (gi|159027173) with no homologue in the NIES-843 genome. Such trends in the expression of these proteins may reflect a function for them in the nitrogen limitation or stress response.
5.3.5. Protein expression in the non-toxic M. aeruginosa PCC 7005 in nitrogen- limited batch culture
The non-toxic strain PCC 7005 was not only the strain with the most diverse subset of proteins affected by nitrogen concentration of the media (Figure 5.3.3), but also the one in which the largest number of proteins were affected by nitrogen stress. From the 560 proteins identified in total for this strain, 144 were differentially expressed. In contrast to PCC 7806 (Table 5.3.4), and the observed entry into stationary growth (Figure 5.3.1), the majority of proteins were up-regulated during nitrogen starvation.
A number of processes were affected by growth in nitrogen-limited conditions, in particular, carbon metabolic pathways, including carbon fixation, glycolysis, the oxidative branch of the pentose phosphate pathway and PHA synthesis. These changes were associated with an increase in amino acid synthesis, in particular, glutamate, as well as an up-regulation of the urea cycle reactions and the transport of iron, nitrogen and phosphate. In contrast to these changes, the nitrate transport system substrate binding protein NrtA was down-regulated, as was a phosphate binding protein homologue. In agreement with the observed chlorosis (Figure 5.3.1) in nitrogen-stressed PCC 7005 cells, the phycobilisome components ApcF and CpcI showed reduced abundance, although the levels of the allophycocyanin core subunits ApcE and ApcA increased.
Nitrogen starvation of PCC 7005 also induced a large number of stress and regulatory proteins, including peroxiredoxin, thioredoxin A and thioredoxin peroxidase, the universal stress protein, the transcription factor Ycf27, and the SOS response transcriptional regulator LexA. Importantly for this study, the expression of the global nitrogen regulator NtcA (Ycf28) was found to be up-regulated in nitrogen-stressed cells, but was below the detection limit under nitrogen-replete conditions.
141 Chapter 5
Twenty hypothetical proteins were differentially displayed, with only MAE 45990 was decreased in abundance after nitrogen stress. Of these, MAE 20860 was the only hypothetical protein that was also differentially displayed in PCC 7806, albeit down- regulated in the toxic strain.
142 Nitrogen stress proteomics
Table 5.3.4. Proteins differentially expressed (p<0.05, t-test) in cells of M. * aeruginosa PCC 7005 grown in BG110 and BG11. Proteins are identified according to annotation in the PCC 7806 and NIES 843 genomes. Expression is shown as fold change relative to BG11. Down: proteins expressed in BG11 only; up: proteins expressed in nitrogen limited growth (BG110) only.
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-Test Expression OmpR family two-component gi|159027454 ycf27 2.1E-05 up response regulator circadian clock protein KaiC gi|159030653 kaiC 0.0006 up homolog glucose-1- gi|159029915 glgC 0.0013 up phosphoadenyltransferase gi|159028841 GTP binding protein unnamed protein product 0.0017 up gi|159027567 serine protease unnamed protein product 0.0017 up gi|159028366 tetratricopeptide unnamed protein product 0.0019 up gi|159028277 aconitate hydratase acnB 0.0024 up 1,4-dihydroxy-2-naphtoate gi|159026916 menB 0.0025 up synthase gi|159029973 no homologue unnamed protein product 0.0031 up anthrinilate synthetase alpha- gi|159027391 unnamed protein product 0.0052 up subunit gi|159030499 probable ribonuclease D rnd 0.0066 up hypothetical protein MAE gi|159026137 unnamed protein product 0.0089 up 05380 glutamyl-tRNA gi|159027280 gatB 0.012 up amidotransferase subunit B gi|159028172 elongation factor P efp 0.012 up gi|159028447 alanine dehydrogenase unnamed protein product 0.012 up carbamoyl phosphate synthase gi|159027682 unnamed protein product 0.013 up large subunit gi|159026755 peptidase M61 unnamed protein product 0.013 up bidirectional hydrogenase gi|159027101 unnamed protein product 0.015 up hydrogenase subunit erythrocyte band 7 integral gi|159027265 unnamed protein product 0.015 up membrane protein NADH dehydrogenase subunit gi|159030165 ndhD1 0.023 up 4 bicarbonate system ATP gi|159030591 nrtC 0.023 up binding protein hypothetical protein MAE gi|159029998 unnamed protein product 0.025 up 41350 gi|159029981 no homologue unnamed protein product 0.028 up probable branched chain gi|159026756 ilvE 0.028 up amino acid aminotransferase global nitrogen regulator Ycf- gi|159029223 unnamed protein product 0.029 up 28 hypothetical protein MAE gi|159030191 unnamed protein product 0.029 up 21590 gi|159026019 aspartyl-tRNA synthase unnamed protein product 0.029 up pyridoxal phosphate gi|159028896 pdxJ 0.032 up biosynthetic protein gi|159029398 no homologue unnamed protein product 0.032 up phosphoenolpyruvate gi|159027003 capP 0.036 up carboxyalse succinate dehydrogenase gi|159026457 frdA 0.036 up flavoprotein subunit gi|159026062 dihydropicolinate reductase unnamed protein product 0.036 up
143 Chapter 5
Expression Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression gi|159028500 OstA-like protein precursor unnamed protein product 0.039 up aldehyde-alcohol gi|159030236 unnamed protein product 0.047 Up dehydrogenase gi|159028603 glutamine synthase GlnN unnamed protein product 0.0017 18.00 gi|159026837 putative methyltransferase unnamed protein product 5.7E-06 11.33 hypothetical protein MAE gi|159029027 unnamed protein product 0.002 10.33 32660 gi|159025989 aspartate aminotransferase unnamed protein product 0.0028 9.67 RNA polymerase beta prime gi|159029166 rpoC2 0.0033 8.89 subunit 1,4-alpha-glucan branching gi|159027247 glgB 0.0017 8.67 enzyme gi|159028003 glutathione-S-transferase gst1 0.0029 8.67 NADH-dependent glutamate gi|159027537 gltB 0.031 7.50 synthase large subunit gi|159028206 50S ribsomal protein L3 rplC 0.00061 6.60 gi|159027731 glycoside hydratase family S1 unnamed protein product 0.0035 6.33 6-phosphogluconate gi|159028626 gnd 0.00049 6.11 dehydrogenase gi|159028480 isocitrate dehydrogenase icd 0.00021 6.00 gi|159027969 DNA topoisomerase I topA 0.025 6.00 gi|159029975 no homologue unnamed protein product 0.00057 5.92 hypothetical protein MAE gi|159030306 unnamed protein product 0.0024 5.00 04620 gi|159030575 GMP synthetase guaA 0.0047 5.00 gi|159026747 glycine dehydrogenase unnamed protein product 0.012 4.89 gi|159026553 acetyl CoA synthetase acsA 0.011 4.83 hypothetical protein MAE gi|159028836 unnamed protein product 0.00023 4.67 63100 gi|159026585 putative peroxiredoxin unnamed protein product 0.0044 4.67 gi|159030003 transketolase, central region unnamed protein product 0.013 4.64 gi|159028816 4-alpha-glucanotransferase unnamed protein product 0.0015 4.61 putative modulator of DNA gi|159027559 unnamed protein product 0.0016 4.22 gyrase PHA-specific acetoacetyl CoA gi|159029273 fabG2 0.0082 4.00 reductase PhaB pyridine nucleotide gi|159028449 unnamed protein product 0.013 4.00 transhydrogenase beta subunit gi|159026111 6-phosphogluconatelactonase unnamed protein product 0.001 3.83 gi|159029167 RNA polymerase beta subunit rpoB 0.028 3.81 N-acetyl gamma-glutamyl gi|159026899 argC 0.019 3.67 phosphate reductase gi|159026032 sulfate adenyltransferase sat 0.025 3.67 gi|159029974 no homologue nox 0.0026 3.62 gi|159030457 dihydropicolinate synthase dapA 0.0058 3.50 gi|159026986 citrate synthase gltA 0.0031 3.44 hypothetical protein MAE gi|159028417 unnamed protein product 0.0046 3.33 15540 gi|159027069 6-phosphofructokinase pfkA1 0.012 3.33 hypothetical protein MAE gi|159027375 unnamed protein product 0.033 3.33 08240 gi|159030571 no homologue unnamed protein product 0.038 3.18 gi|159029828 peptidase unnamed protein product 0.00014 3.17 NADH-dependent glutamate gi|159026386 gltD 0.0051 3.11 synthase small subunit gi|159027871 glucose 6-P dehydrogenase zwf 0.0058 3.11
144 Nitrogen stress proteomics
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression hypothetical protein MAE gi|159029108 unnamed protein product 0.00049 3.00 21730 hypothetical protein MAE gi|159028898 unnamed protein product 0.0063 3.00 23900 gi|159026705 cysteine synthase cysK 0.005 2.91 hypothetical protein MAE gi|159028713 unnamed protein product 0.034 2.90 22790 hypothetical protein MAE gi|159026196 unnamed protein product 0.011 2.89 60140 gi|159029243 glgP 0.0097 2.87 gi|159027997 acetolactate synthase ilvB 0.0064 2.83 gi|159028833 arylsulfatase like atsD 0.029 2.80 gi|159028814 ornithine carbomoyltransferase argF 0.0084 2.78 nitrate/nitrite system ATP gi|159028059 nrtC 0.0029 2.71 binding protein transaldolase/ EF-hand gi|159027582 unnamed protein product 0.011 2.65 containing protein hypothetical protein MAE gi|159027820 unnamed protein product 0.02 2.56 45800 gi|159029272 PHA-specific beta-ketothiolase unnamed protein product 0.008 2.50 hypothetical protein MAE gi|159030804 unnamed protein product 0.022 2.40 46700 gi|159028548 FtsZ protein ftsZ 0.022 2.40 glyceraldehyde-3-phosphate gi|159029182 gap1 0.026 2.39 dehydrogenase glutamate-1-semialdehyde gi|159029578 hemL 0.0054 2.38 aminomutase gi|159026115 triose phosphate isomerase unnamed protein product 0.011 2.33 gi|159026822 ketol-acid reductoisomerase ilvC 0.04 2.28 gi|159028011 universal stress protein unnamed protein product 0.03 2.20 phosphate binding protein gi|159026174 unnamed protein product 0.019 2.17 PstA homolog gi|159029181 putative phosphoketolase unnamed protein product 0.024 2.17 gi|159026243 flavoportein unnamed protein product 0.034 2.09 (+1) extracellular solute-binding gi|159026223 unnamed protein product 0.045 2.06 protein AAA family ATPase central gi|159029921 unnamed protein product 0.0097 2.00 region gi|159027861 glucose-6-phosphate isomerase unnamed protein product 0.014 2.00 hypothetical protein MAE gi|159029916 unnamed protein product 0.028 2.00 28060 hypothetical protein MAE gi|159028813 unnamed protein product 0.029 2.00 54080 adenine gi|159026652 apt 0.032 2.00 phosphoribosyltransferase putatuve module of DNA gi|159027557 pmbA 0.033 2.00 gyrase gi|159027379 D-3-phosphoglycerate serA 0.005 1.97 hypothetical protein MAE gi|159027109 unnamed protein product 0.017 1.88 53760 gi|159030029 dihydroxyacid dehydratase unnamed protein product 0.025 1.88 phycobilisome core-membrane gi|159026883 apcE 0.0016 1.88 linker polypeptide gi|159029142 flavoprotein dfa3 0.027 1.83 hypothetical protein MAE gi|159029335 unnamed protein product 0.02 1.80 15980
145 Chapter 5
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression iron transport system gi|159027971 unnamed protein product 0.028 1.80 substrate-binding protein gi|159029229 elongation factor G fusA 0.018 1.78 gi|159026519 glucokinase unnamed protein product 0.041 1.71 SOS response transcriptional gi|159026334 unnamed protein product 0.035 1.70 regulator LexA heterodisulfide reductase gi|159027437 hdrB 0.01 1.67 subunit B gi|159027420 ferrochelatase hemH 0.048 1.67 gi|159030988 phosphate binding protein pstS 0.049 1.67 photosystem II magnesium gi|159026933 psbO 0.033 1.61 stabilising polypeptide gi|159030400 transketolase unnamed protein product 0.05 1.57 porin type major outer gi|159030354 unnamed protein product 0.012 1.48 membrane protein NADH dehydrogenase subunit gi|159026401 unnamed protein product 0.036 1.46 K pyrroline phosphate gi|159026854 proC 0.032 1.33 carboxylate reductase hypothetical protein MAE gi|159030999 unnamed protein product 0.029 1.00 07070 ATP synthase CF1 delta chain gi|159028802 atpD 0.023 -1.25 AtpD gi|159026736 photosystem II D1 protein unnamed protein product 0.025 -1.32 (+1) fructose-1,6-bisphosphatase, gi|159029605 glpX 0.038 -1.37 glpX encoded gi|159030269 thioredoxin peroxidase unnamed protein product 0.014 -1.46 Phycobilisome core gi|159029093 apcF 0.045 -1.50 component gi|159028057 nitrate transport system NrtA nrtA 0.049 -1.59 Phycobilisome rod linker gi|159029497 cpcI 0.028 -1.60 polypeptide ribulose 1,5-bisphosphate gi|112821113 RuBisCo small subunit carboxylase/oxygenase small 0.046 -1.70 (+1) subunit gi|159027979 allophycocyanin alpha subunit apcA 0.018 -1.81 gi|159028065 cell division protein unnamed protein product 0.031 -1.81 gi|159028508 cell division protein FtsH unnamed protein product 0.02 -2.00 gi|159029286 leucine aminopeptidase pepA 0.023 -2.01 gi|159029329 nucleoside diphosphate kinase ndk 0.008 -2.13 gi|159028396 50S ribosomal protein L11 rplK 0.044 -2.33 hypothetical protein MAE gi|159030779 unnamed protein product 0.011 -2.44 45990 gi|159027635 thioredoxin A unnamed protein product 0.024 -2.57 gi|159030907 no homologue unnamed protein product 0.00011 -2.68 gi|159028190 30S ribosomal protein S5 rpsE 0.0053 -2.74 gi|159028193 30S ribosomal protein S8 rpsH 0.0069 -2.85 gi|159027290 inner membrane protein OxaA unnamed protein product 0.045 -3.00 gi|159029853 50S ribosomal protein L9 rplI 0.034 -3.57 gi|159026381 putative sugar kinase unnamed protein product 0.044 -4.00
146 Nitrogen stress proteomics
5.3.6. Protein expression in the genetically engineered non-toxic M. aeruginosa PCC 7806 mcyH- in nitrogen-limited batch culture
The largest number of proteins (629) was identified in the genetically engineered mutant strain PCC 7806 mcyH-. In this strain, 90 proteins were differentially expressed in nitrogen-replete and nitrogen-limited conditions, with most changes resulting in over- expressed proteins in nitrogen-starved cultures (Table 5.3.5).
Unlike the wild-type PCC 7806, when limited for nitrogen, the mcyH- mutant strain exhibited decreased expression levels of the photosystem I protein PsaL and increased PsaD levels, increased ferredoxin-NADPH dehydrogenase and higher levels of the ATP synthase gamma subunit (AtpC). This was supported by the higher expression of proteins involved in carbon fixation, glycolysis and PHA and glycogen synthesis. Active metabolism was also expected given the increase in the RNA polymerase alpha subunit and the observed up-regulated tRNA synthetase levels for several amino acids.
Several substrate binding proteins, including a putative periplasmic metal-binding protein, a cation transporting ATPase and an ABC-transporter component were increased in expression in N-limited growth, but none of these included nitrogen- specific transporters. The periplasmic protein PilT, which is involved in the depolymerisation of type IV pili was also upregulated.
The RuBisCo-like protein characteristic of M. aeruginosa PCC 7806 and the PCC 7806-mutant strain was up-regulated in nitrogen stress. Other stress-induced proteins that were identified in mcyH- included superoxide dismutase, peroxiredoxin and a universal stress protein.
From the hypothetical proteins, MAE 32660 and MAE 23900 were up-regulated in N- starved PCC 7806 mcyH-, as was the case in PCC 7005. Eighteen other proteins were differentially expressed in both nitrogen-starved non-toxic strains, but not in PCC 7806. These included several proteins involved in carbon metabolism (PHA-specific beta ketothiolase, 6-phosphofructokinase, citrate synthase and transladolase) and the urea cycle (aspartate aminotransferase, argininosuccinate synthase and ornithine carbamoyltransferase). The phycobilisome core component, ApcE was also up- regulated in both non-toxic M. aeruginosa strains, but not in PCC 7806. In addition, a
147 Chapter 5 putative peroxiredoxin and a universal stress protein were increased in abundance in both non-toxic strains under nitrogen limitation.
Table 5.3.5. Proteins differentially expressed (p<0.05, t-test) in cells of M. - * aeruginosa PCC 7806 mcyH grown in BG110 and BG11. Proteins are identified according to annotation in the PCC 7806 and NIES 843 genomes. Expression is shown as fold change relative to BG11. Down: proteins expressed in BG11 only; up: proteins expressed in nitrogen limited growth (BG110) only.
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-Test Expression gi|159031016 putative translation factor unnamed protein product 0.017 up cobalmin synthesis cobW- gi|159030631 like protein unnamed protein product 0.017 up gi|159030498 hypothetical MAE 26040 unnamed protein product 0.017 up gi|159030489 3-dehydroquinate synthase aroB 0.017 up aspartate-semialdehyde gi|159030458 dehydrogenase asd 0.001 up inosine-5-monophosphate gi|159030105 dehydrogenase guaB 0.00059 up gi|159030071 glucokinase unnamed protein product 0.017 up diaminopimelate gi|159029858 decarboxylase unnamed protein product 0.017 up glutathione-dependent gi|159029517 formaldehyde dehydrogenase unnamed protein product 0.0026 up gi|159029342 recombinase A recA 0.0071 up poly(3-hydroxyalkanoate) gi|159029277 synthase phaC 0.017 up gi|159028893 hypothetical MAE 23950 unnamed protein product 0.017 up gi|159028858 tryptophanyl tRNA synthase trpS 0.017 up gi|159028447 alanine dehydrogenase unnamed protein product 0.0044 up ABC transporter ATP- gi|159027789 binding protein unnamed protein product 0.017 up gi|159027776 phytoene dehydrogenase pds/crtD 0.017 up cation transporting P-type gi|159027774 ATPase unnamed protein product 0.017 up gi|159027684 tyrosyl-tRNA-synthetase unnamed protein product 0.017 up GDP mannose 4,6- gi|159027653 dehydratase unnamed protein product 0.026 up HEAT-repeat containing PBS gi|159027362 lyase unnamed protein product 0.017 up N-acetyl-gamma-glutamyl gi|159026899 phosphate reductase argC 0.044 up gi|159026696 no homologue unnamed protein product 0.042 up 1-acyl-sn-glycerol-3- gi|159026188 phosphate acyltransferase unnamed protein product 0.017 up gi|159026076 no homologue unnamed protein product 0.048 up twitching motility protein gi|159026047 PilT pilT 0.014 up gi|112821115 no homologue rubisco-like protein 0.01 up gi|159027856 30S ribosomal protein S4 unnamed protein product 0.026 21.00 gi|159026712 CP12 polypeptide unnamed protein product 0.0077 18.50 hypothetical protein MAE gi|159027927 06250 unnamed protein product 0.022 13.00 ornithine gi|159028814 carbamoyltransferase argF 0.023 12.00
148 Nitrogen stress proteomics
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression 3-isopropylmalate gi|159027474 dehydrogenase leuB 0.0021 12.00 hypothetical protein MAE gi|159030229 30710 unnamed protein product 0.045 11.50 glyceraldehyde 3-P gi|159029182 dehydrogenase gap1 0.0034 11.20 PHA-specific beta 0.00000 gi|159029272 ketothiolase unnamed protein product 54 9.75 gi|159030131 argininosuccinate synthase argG 0.032 9.00 S-adenosylmethionine tRNA gi|159028722 ribosyltransferase-isomerase queA 0.019 8.75 hypothetical protein MAE gi|159029027 32660 unnamed protein product 0.00061 8.00 hypothetical protein MAE gi|159028898 23900 unnamed protein product 0.042 8.00 gi|159026585 putative peroxiredoxin unnamed protein product 0.034 7.67 hypothetical protein MAE gi|159028629 60080 unnamed protein product 0.046 7.50 gi|159025989 aspartate aminotransferase unnamed protein product 0.00032 7.40 carbamoyl phosphate gi|159029969 synthase small chain carA 0.015 7.00 gi|159026520 no homologue unnamed protein product 0.005 7.00 transaldolase/ EF-hand gi|159027582 domain-containing protein unnamed protein product 0.0017 6.71 NAD-dependent epimerase/ gi|159027807 dehydratase unnamed protein product 0.003 6.50 soluble hydrogenase 42 kDa 0.00007 gi|159027426 subunit unnamed protein product 8 6.44 gi|159026822 ketol-acid reductoisomerase ilvC 0.00036 6.22 gi|159026986 citrate synthase gltA 0.0002 6.00 gi|159028011 universal stress protein unnamed protein product 0.0044 5.66 AAA family ATPase central gi|159029921 region unnamed protein product 0.032 5.50 gi|159026288 6-phosphofructokinase unnamed protein product 0.00013 5.50 ATP dependent Clp protease gi|159029700 ATPase subunit clpC 0.042 5.29 hypothetical protein MAE gi|159027826 11600 unnamed protein product 0.023 4.83 gi|159028553 no homologue unnamed protein product 0.016 4.50 chaperone-like protein for gi|159027214 quinone binding in PSII unnamed protein product 0.042 4.50 phycobilisome core- membrane linker gi|159026883 polypeptide apcE 0.022 4.37 periplasmic carboxy-terminal gi|159030479 protease unnamed protein product 0.023 4.00 NADH dehydrogenase gi|159030038 subunit H ndhA 0.035 4.00 hypothetical protein MAE gi|159027376 08230 unnamed protein product 0.041 3.55 hypothetical protein MAE gi|159030697 62150 unnamed protein product 0.018 3.37 uncharacterised Zn-type gi|159029439 ADH-like protein unnamed protein product 0.019 3.37 periplasmic beta type gi|159031042 carbonic anhydrase unnamed protein product 0.0003 3.25 gi|159027420 Ferrochelatase hemH 0.026 2.86 ATP synthase CF1 gamma gi|159027204 chain AtpC atpC 0.039 2.63
149 Chapter 5
Accession Protein ID (NIES-843) Protein ID (PCC 7806) T-test Expression gi|159028437 no homologue unnamed protein product 0.04 2.57 gi|159026977 superoxide dismutase unnamed protein product 0.047 2.50 glutamate-1-semialdehyde gi|159029578 aminomutase hemL 0.043 2.33 phycobilisome rod-core gi|159028007 linker polypeptide cpcG 0.048 2.30 gi|159027085 RNA polymerase A rpoA 0.034 2.21 carbon concentrating gi|159026025 mechanism protein CcmA unnamed protein product 0.032 2.16 gi|159030907 no homologue unnamed protein product 0.029 2.01 glyceraldehyde-3-phosphate gi|159030256 dehydrogenase unnamed protein product 0.028 2.01 photosystem II reaction gi|159026300 centre D2 protein psbD 0.0064 1.94 gi|159028923 30S ribosomal protein S1 unnamed protein product 0.042 1.93 ferredoxin-NADP gi|159025950 oxidoreductase unnamed protein product 0.022 1.86 periplasmic solute binding gi|159030630 protein^ unnamed protein product 0.017 1.50 hypothetical protein MAE gi|159027399 23190 unnamed protein product 0.017 1.50 gi|159030372 glutamate—ammonia ligase glnA 0.04 1.29 gi|159027052 photosystem I subunit XI psaL 0.022 -1.55 gi|159027764 seryl-tRNA-synthetase unnamed protein product 0.017 -2.00 ribulose 1,5-bisphosphate carboxylase/oxygenase large gi|112821111 RuBisCo large subunit^ subunit 0.042 -2.01 carboxysome formation gi|159027942 protein CcmM^ ccmM 0.0026 -3.02 gi|159030854 acyl carrier protein unnamed protein product 0.024 -3.20 gi|159030524 glycogen synthase glgA2 0.043 -3.20 L,L-diaminopimelate gi|159030764 aminotransferase^ unnamed protein product 0.02 -3.75 hypothetical protein MAE gi|159030973 47530 unnamed protein product 0.021 -4.35 gi|159026366 RNA binding protein unnamed protein product 0.049 -5.00 gi|159029828 Peptidase unnamed protein product 0.0068 -6.00 gi|159026481 30S ribosomal protein S16 rpsP 0.048 -6.83 hypothetical protein MAE gi|159028371 53490 unnamed protein product 0.028 down proteins with the same expression pattern in PCC 7005 and PCC 7806 mcyH- during nitrogen starvation ^proteins with the same expression pattern in PCC 7806 and PCC 7806 mcyH- during nitrogen starvation
5.3.4. qRT-PCR of the signal protein PII, and the transcriptional regulators NtcA and FurA
The regulatory proteins FurA and PII which were identified as important for the starvation response of M. aeruginosa were not found to be expressed at detectable levels here. The only protein which was up-regulated significantly in conditions of
150 Nitrogen stress proteomics nitrogen stress was NtcA, and this was only observed in the non-toxic strain PCC 7005. However, given the expected change in expression of the global nitrogen transcriptional regulator NtcA during nitrogen limitation, and its cross-regulatory effect on the redox- responsive FurA and the nitrogen-responsive PII, the genes encoding these regulatory proteins were chosen for transcriptional analysis by qRT-PCR (Table 5.3.6).
Table 5.3.6. Transcription of selected genes in nitrogen replete and nitrogen- deficient conditions. Transcription in cells grown in BG110 is shown as fold-change relative to growth in BG11. Standard deviation is in brackets. The RNA polymerase gamma subunit gene rpoC was chosen as reference. Values that are significantly different from transcript levels in BG11-grown cultures (t-test, p<0.05) are shown in bold.
Gene PCC 7806 PCC 7806 mcyH- PCC 7005 furA -1.99 (0.90) 1.15 (0.45) -4.61 (0.45) ntcA -1.52 (0.72) -8.43 (0.94) -29.45 (0.64) glnB -13.17 (0.99) 1.00 (0.20) -19.00 (0.71)
After a t-test (p<0.05, n=3) no significant change in the transcription of either ntcA or furA was observed for the toxic PCC 7806. On the other hand, in both non-toxic strains, ntcA was significantly down-regulated in conditions of nitrogen starvation. The expected reciprocal relationship between glnB and ntcA was maintained in the inactivated mcyH strain, and to some extent, in the wild-type PCC 7806. The mutant mcyH- strain was the only strain in which fur transcription was up-regulated during nitrogen stress.
5.4. Discussion
Nitrogen is essential for the growth of cyanobacteria and is tightly regulated by photosynthetic activity and carbon metabolism at the transcriptional level (Flores and Herrero, 2005, Herrero, et al., 2001, Schwarz and Forchhammer, 2005). In the absence of a combined nitrogen source, nitrogen-fixing species are able to respond to nitrogen starvation by developing heterocysts and fixing atmospheric nitrogen. In non- diazotrophic cyanobacteria, such as M. aeruginosa, nitrogen starvation induces a general stress response, chlorosis, increased transport of alternative nitrogen sources and, when these are depleted, entry into a dormant state (Schwarz and Forchhammer, 2005). This general starvation response was observed here in the protein expression profiles. These will be discussed below, in particular with reference to toxin production and NtcA regulation in three M. aeruginosa strains.
151 Chapter 5
5.4.1. Microcystin synthesis and growth of M. aeruginosa in nitrogen-limited conditions
Investigations into the relationship between nutrient balance and microcystin synthesis in cyanobacteria have proposed that non-toxic strains survive better in low-nutrient environments, whereas toxic strains have an advantage in nutrient-rich culture (Vezie, et al., 2002). This effect was not observed here, as reflected by the similar growth rates obtained for all of the strains used in these experiments (Figure 5.3.1). The entry of the nitrogen-limited cells in stationary phase after 1 day of culture is consistent with results from God et al. (1998), who observed the cells of Synechococcus elongatus PCC 7942 used all available nitrogen within one cell division. The following fluctuation in cell numbers observed in the growth studies performed here could reflect a lysis of the bacterial cells, resulting in a transient increase of nutrients that become available and support growth in the remaining viable population (Schatz, et al., 2007).
Nitrogen availability affects not only the growth of M. aeruginosa, but also the synthesis of the non-ribosomal peptide microcystin in toxic cyanobacteria (Downing, et al., 2005, Waal, et al., 2009). This relationship has been accounted for by the high nitrogen content of the microcystin molecule, and the literature supports a decrease in toxin concentration in nitrogen-limited cultures (Dai, et al., 2008, Vezie, et al., 2002).
Several studies have also found a strong interaction of N:P and N:CO2 fixation ratio on growth and toxin synthesis. However, in the results presented here (Section 5.3.1) the toxin analysis did not identify a significant difference in the toxin content of M. aeruginosa cultures grown in nitrogen-limited and nitrogen-replete batch culture.
This observation is similar to the lack of significant change in MCYST-LR with changes in the N:C ratio of cells observed by Waal et al. (2009), although the rarer isoforms MCYST-RR and MCYST-YR increased during growth in high N:C ratio. In this case, although energetically expensive for the cell, the toxin contained less than 1% of the total intracellular nitrogen and was not considered to represent a significant sink for nitrogen (Waal, et al., 2009). The lack of change in toxin content observed in the present study is also interesting given the fact that the cells entered stationary growth shortly after transition into the nitrogen-limited media, which would cause a decrease in the rate of toxin synthesis (Downing, et al., 2005). It is possible that the microcystin intracellular content was already high in the cells used for inoculation and remained at a similar level for the duration of the experiment. A similar conclusion was made by
152 Nitrogen stress proteomics
Schatz et al. (2007) when exposure to toxic M. aeruginosa lysate, or purified microcystin did not induce a significant accumulation of toxin in M. aeruginosa cultures when cells in late exponential/ stationary growth were used as inoculum.
The presence of a NtcA binding motif in the mcyS promoter (Ginn and Neilan, 2010) would result in positive regulation of mcy gene transcription when NtcA levels increase in limited nitrogen. However, in the experiments performed here, no significant change in either the expression, or the transcription of the global nitrogen regulator in M. aeruginosa PCC 7806 was observed as a result of nitrogen stress. This lack of NtcA response was consistent with the protein expression changes that occur in PCC 7806 during nitrogen limitation, and correlates with the observation that toxin production in this strain was not affected by nitrogen availability. On the other hand, the protein expression profiles of the non-toxic PCC 7005 and PCC 7806 mcyH- strain appeared to follow a typical NtcA-regulated response to nitrogen stress.
5.4.1. Photosynthesis and respiration
A typical response of non-diazotrophic cyanobacteria to nitrogen starvation is to induce chlorosis (God, et al., 1998, Schwarz and Forchhammer, 2005). This allows the cells to acquire the limiting nutrients via degradation of the phycobiliproteins, which normally represent the larger part of the protein pool in these cells (Anderson, et al., 2006). Metabolic labeling has shown that dormant nitrogen-starved cells slowly turn over these photosynthetic proteins, with very few being synthesized de novo in the absence of a nitrogen source (Sauer, et al., 2001).
Unlike iron starvation, where the toxic strain PCC 7806 was found to be the only strain included in the analysis that did not bleach (Chapter 3), when placed under nitrogen stress all strains became chlorotic within three days, suggesting that the response observed in Chapter 3 was iron specific. Allophycocyanin was the only pigment that was elevated in nitrogen-starved cells in all strains relative to the cultures grown in nutrient-replete conditions. A similar finding has been observed in the marine non- nitrogen fixing Oscillatoria willei, where phycocyanin content decreased nearly 2-fold, but allophycocyanin content remained unchanged under different nitrogen regimens (Saha, et al., 2003). This could be due to the ordered degradation of the phycobilisome, a process that is normally assisted by NblA and NblB and involves removal of the pigments from the polypeptide rods first, leaving the thylakoid-associated
153 Chapter 5 allophycocyanin-rich core (Dolganov and Grossman, 1999). In the non-toxic wild-type strain PCC 7005 the expression of a two-component OmpR regulator, Ycf27, was increased in low nitrogen. This protein is involved in the coupling of energy transfer to PSI or PSII (Ashby and Mullineaux, 1999) and could play a role in the low photosynthetic activity that is typical of dormant nitrogen-starved cells (Sauer, et al., 2001).
When the proteomes of the toxic and non-toxic strains in nutrient-replete (BG11) growth were compared, a number of differences in the photosynthetic machinery, in particular the reaction centers of photosystem I (PSI) and II (PSII), as well as ATP synthase were observed (Table 5.3.1). The over-expression of these proteins in non- toxic M. aeruginosa, together with the up-regulation of proteins involved in carbon metabolism (Table 5.3.1) could lead to a high C:N ratio in these cells in nutrient-replete conditions. In nitrogen stress, all strains regulated different components of their photosystem and electron chain (Figure 5.4.1), but subunits of the NADH dehydrogenase complex and PSII were increased in both non-toxic strains studied here. This suggests that electron flow continues even in chlorotic cells of M. aeruginosa, but oxidative stress resulting from the over-reduction of electron carriers may be prevented by the overexpression of the universal stress protein and a putative peroxiredoxin in both PCC 7005 and PCC 7806 mcyH-.
An important enzyme associated with phycobilisome degradation and chorosis is alanine dehydrogenase, which normally acts to degrade alanine to ammonia and pyruvate, giving nitrogen-starved cells a source of nitrogen and carbon (Lahmi, et al., 2006). This enzyme was found to be up-regulated in the non-toxic cells, but not PCC 7806, and may be involved in maintaining appropriate levels of glutamine synthetase
(GS), PII, glnN-encoded GS and isocitrate dehydrogenase in the non-toxic strains, as it does in Synechococcus elongatus PCC 7942 (Lahmi, et al., 2006).
5.4.2. Transport of metabolites during nitrogen limitation
Under conditions of nitrogen stress, transport of nitrogenous compounds increases as part of the NtcA-regulated response, in order to counter the rise in C:N and the accumulating 2OG. This was observed in all strains analysed, with a number of periplasmic solute-binding proteins being up-regulated when the cells were grown in
BG110 (Figure 5.4.1). Several of these, including transporters for nitrate and urea, were
154 Nitrogen stress proteomics increased in BG11-grown cells of the toxic PCC 7806, relative to the non-toxic strains (Table 5.3.2). This, combined with lower photosynthetic activity (Table 5.3.1) would result in high nitrogen reserves in the toxic M. aeruginosa studied, and a lower C:N ratio than that of non-toxic strains. This is consistent with the reported increase in NrtA expression in toxic M. aeruginosa relative to non-toxic strains of the same species observed in Chapter 2, and together with the higher expression of urea transporter, a high NtcA activity in a nutrient-replete background was expected.
Interestingly, in both wild-type strains investigated, a phosphate periplasmic substrate binding protein was increased in abundance. This result may reflect the significant interaction between N and P concentration in the media and the importance of N:P ratio for the growth of cyanobacteria (Dai, et al., 2008, Downing, et al., 2005). The high expression of a sulfate transporter subunit in PCC 7806 grown in the presence of nitrogen relative to the non-toxic strains could be linked to a recent proposal that sulfur metabolism and microcystin synthesis are linked, due to the involvement of S- adenosylmethionine in the methylation reactions required for microcystin synthesis (Long, 2010).
5.4.3. Carbon-nitrogen balance
The protein expression differences in the photosynthetic and transport proteins of M. aeruginosa strains in the presence of nitrogen would lead to a difference in the cellular C:N ratio, as discussed above. This is expected to affect the response of M. aeruginosa when nitrogen becomes limiting and the C:N balance needs to be restored. The lack of a nitrogen source leads to an increase in the C:N ratio due to the accumulation of 2- oxoglutarate (2OG) and is corrected by the NtcA-induced transport of nitrogenous compounds, GS and isocitrate dehydrogenase activity (Muro-Pastor, et al., 2005, Schwarz and Forchhammer, 2005). This classical nitrogen-limitation response was reflected in the proteomic changes induced during transition to nitrogen-limiting BG110 in the PCC 7005 strain. This was also the only representative of the M. aeruginosa species studied in this chapter that had a significantly up-regulated level of NtcA.
155 Chapter 5
156 Nitrogen stress proteomics
The decrease of GS expression in PCC 7806, but not the non-toxic strains, was consistent with the lack of a significant change in either expression or transcription of NtcA. The failure of this strain to increase incorporation of 2-OG in the GS-GOGAT cycle is not likely to cause a pronounced disruption in cellular C:N, due to the already reduced carbon metabolism in this toxic strain. On the other hand, in the non-toxic strains, a different mode of suppression of the Calvin cycle was employed. This was represented by an increase of the regulatory peptide CP12 in PCC 7806 mcyH-, resulting in GAPDH/PRK inactivation (Tamoi, et al., 2005, Trost, et al., 2006). The transcriptional activator LexA has also been shown to be involved in the survival of the non-nitrogen fixing Synechocystis sp. PCC 6803 during C-starvation (Domain, et al., 2004). In the absence of LexA, several proteins were found to be decreased, including the nitrate/nitrite transport system protein NrtC, the iron transport system substrate- binding protein, a bicarbonate transporter, and NADH subunit K (Domain, et al., 2004). The proteome data showed that the LexA-regulated proteins were all increased, consistent with the observed LexA up-regulation in PCC 7005 after nitrogen starvation. LexA, similarly to NtcA, also reduces the level of the GS-inactivating factors and the combined action of these two transcriptional factors could have contributed to the rapid increase in GS-GOGAT activity found here (Domain, et al., 2004). The only enzyme which deviated from the predicted translation pattern was 6-phosphofructokinase, which was upregulated in the presence of LexA. This attempt to increase the carbon ratio appears contradictory to the already high C:N ratio expected during N-starvation. It may be a similar mode to the increase of isocitrate dehydrogenase by NtcA further increasing the 2OG ratio in the cell and thus NtcA expression (Muro-Pastor, et al., 2005). The discrepancy between the high abundance of the NtcA protein in PCC 7005 and the significant decrease in ntcA transcript levels may be a result of transcript and protein stability, as well as the auto-regulation of NtcA and its interaction with numerous co-factors (Aldehni, et al., 2003). This has been shown in microarray analysis in Synechocystis PCC 6803, where the redox status of the electron transport chain, nitrogen transport and photosynthesis were able to induce a decrease in the expression of genes that are normally activated by NtcA (Hihara, et al., 2003). Such process may aid in preventing energy to be expended on NtcA-induced nitrogen transport when N- resources are used up and prolonged nitrogen starvation sets in.
The excess carbon fixed in the non-toxic strains appears to be directed towards synthesis of polyhydroxyalkanoates and glycogen, which form carbon-storage granules
157 Chapter 5 in cyanobacteria (God, et al., 1998, Jau, et al., 2005). The synthesis of these compounds in non-toxic M. aeruginosa was enhanced not only in nitrogen-limited conditions, but also during nutrient-replete exponential growth in BG11. Similarly to the higher expression of photosystem core subunits in these strains supports more effective photosynthesis and carbon fixation relative to PCC 7806. The increase in carbon metabolism components during nitrogen stress is not novel, with an increase in a number of genes involved in the oxidative pentose pathway, carbon fixation and glycogen catabolism also reported as increased in Synechocystis PCC 6803, resulting from the putative regulation of SigE by NtcA (Osanai, et al., 2007).
5.5. Conclusions
In conclusion, microcystin synthesis was not affected by nitrogen starvation in M. aeruginosa and did not appear to be disadvantageous to the toxic strain during N- limited growth. Background expression differences in the three strains analysed suggested that NtcA activity is already high in nutrient-replete toxic cells as has been proposed in previous chapters of this thesis. The toxic and non-toxic strains differed in the expression of photosynthetic components, as well as enzymes involved in the Calvin cycle and glycolysis, pointing to a variation in the processes that govern C:N ratio maintenance in strains of M. aeruginosa under nutrient-replete and nitrogen-limiting conditions. Such differences may affect the long term survival of these organisms in nutrient-limited reservoirs and their growth once nitrogen becomes readily available and favours bloom formation.
158 Nitrogen stress proteomics
159 Chapter 5
160
Chapter 6 Conclusions and future directions
161 Chapter 6
162 Conclusions and future directions
6.1. Research motivation and objectives
The synthesis of the non-ribosomal toxic peptide microcystin by several species of cyanobacteria has been proposed to be advantageous for the producing cells, in particular under nutrient and light flux (Briand, et al., 2008, Edwin, et al., 2007). However, reliable protocols for the genetic manipulation of the model microcystin- producing cyanobacterium M. aeruginosa are not available, making research into the toxin synthesis process and the regulatory mechanisms governing it difficult. Proteomics has the capability to reveal global protein expression changes and thus a network of process that may be affected by a particular growth condition. Yet, so far proteomic investigations into cyanobacteria are in their infancy and have yet to be used to investigate the processes that drive toxin production in these organisms. This study presents a comprehensive proteome map of the bloom-forming hepatotoxic M. aeruginosa species in both nutrient-replete and nutrient-limited conditions. It is also the first population study targeting cyanobacteria which compares strains of the same species at the protein expression level.
In Chapter 2, inactive toxin-producing, but toxigenic strains, were used in the proteomic analysis as it was of interest to determine what differences in the regulation of the mcy cluster resulted in the inability of these strains to produce the toxin. In later chapters, the mcyH- mutant strain was used to aid in the identification of toxicity effects on protein expression rather than strain-specific differences arising from natural adaptation processes.
6.2. Key findings
6.2.1. Diversity of protein expression in strains of M. aeruginosa
When the core proteome of M. aeruginosa was assembled in Chapter 2, it became evident that despite the high 16S rDNA homology between the strains used, only one third of the proteome of each strain was representative of the species. In fact, the majority of expressed proteins were strain-specific (Chapter 2, Figure 2.3.3). The strains that were used here originated from a variety of geographical locations and have been cultured in the laboratory for various periods of time ranging between 27 and 62 years (Chapter 2, Table 2.2.1), which may be the reason for this variability. This observation was particularly highlighted in the toxic UWOCC MRC and UWOCC MRD, which
163 Chapter 6 were initially isolated as a single strain, prior to the spontaneous mutation of UWOCC MRC allowing it to switch toxin synthesis off and revert back to toxin production as was reported in Chapter 2 (Figure 2.3.7). This finding reflects that microcystin synthesis is not a static trait in at least some strains of bloom-forming cyanobacteria and is post- transcriptionally regulated by a yet unconfirmed mechanism. Additionally, in the non- toxic PCC 7005, which had been cultured in the laboratory for more than six decades, the highest number of unique proteins were identified in the soluble proteome (Chapter 2, Figure 2.3.3), revealing that cumulative changes can occur when environmental isolates are grown for a prolonged period of time in conditions of nutrient excess. These results support the hypothesis that different M. aeruginosa strains are essentially ecotypes, adapted to a particular environmental niche (Otsuka, et al., 2001). The findings also stressed the importance of using several strains when studying metabolic processes in cyanobacteria, in order to avoid incorrectly extrapolating strain-specific responses to the entire species.
6.2.2. Toxin regulation and the reversal of toxicity
In Chapter 2 the ability of a previously inactive-toxigenic strain, UWOCC MRC, to revert back to toxicity was reported (Figure 2.3.7). The spontaneous occurrence of insertion sequences disrupting active genes has been identified previously in cyanobacteria, and microcystin-encoding genes have been the subject of point mutations and inactivation (Ostermaier and Kurmayer, 2009, Schatz, et al., 2005). These processes could account for the initial inactivation of mcy genes in the parent strain UWOCC MRD, resulting in the isolate known today as UWOCC MRC (Kaebernick, et al., 2001). However, this does not explain the reversal back to microcystin production and the different isoforms of microcystin synthesized by UWOCC MRC, and suggests a post-transcriptional mechanism is in place for toxicity regulation in inactive microcystin producers. Despite the possibility that such toxicity induction and inactivation may occur continuously in the environment this process cannot be analysed using current molecular biology techniques and reports of such events are not available in the literature.
The insertions in the UWOCC MRC mcy cluster reported by Roberts and Neilan (2010) do not disrupt the binding sites of either FurA or NtcA but could affect the alternative transcriptional start points for light that have been previously reported to regulate mcy transcription (Kaebernick, et al., 2000, Kaebernick, et al., 2001, Roberts and Neilan,
164 Conclusions and future directions
2010, Sevilla, et al., 2008). In Chapter 3, transcript analysis of mcyA revealed that this gene was actively transcribed in both toxic, as well as toxigenic but inactive M. aeruginosa strains (Table 3.3.1) and provided further evidence for the post- transcriptional regulation of microcystin.
6.2.3. Protein expression in toxic and non-toxic M. aeruginosa
The protein expression of toxic and non-toxic strains of M. aeruginosa revealed a number of metabolic differences. A summary of the findings from the proteomic studies of these strains, cultured in nutrient-replete conditions (Chapters 2, 4, and 5), is presented in Figure 6.1 and Figure 6.2.
In Chapter 2, six strains were used for the differential display. In Chapter 4 and Chapter 5, the non-toxic M. aeruginosa PCC 7005 was selected for further analysis, based on the 16S and phycocyanin intergenic spacer sequence homology between PCC 7005 and the model PCC 7806. Additionally, the similarity in protein expression between these two strains observed by cluster analysis in Chapter 2 (Figure 2.3.4) highlighted the need to examine the PCC 7806 and PCC 7005 pair more closely. The results were complemented by the proteomic analysis of PCC 7806 mcyH-, a genetically-engineered non-toxic strain, which assisted in the identification of toxicity-associated protein expression patterns rather than strain-strain differences arising from the adaptation of PCC 7005 as a result of natural selection.
It is apparent that when toxic and non-toxic strains of M. aeruginosa are compared, a difference in the integration of C:N metabolism exists (Figure 6.1 and Figure 6.2). The protein expression pattern in the toxic strains was consistent with a state of high C:N ratio at the time when the cells were harvested, and would have resulted in accumulation of 2-oxoglutarate (2OG) (Figure 6.1 and Figure 6.2). This hypothesis is supported by the decrease in PII expression, gluconeogenesis and carbon storage via polyhydroxyalkanoate (PHA) synthesis (Figure 6.2), combined with an increase in nitrogen transport and incorporation of 2OG in glutamine synthesis (Figure 6.1). Many of these proteins are activated at the transcriptional level by the global nitrogen regulator NtcA (Figure 6.1) and their expression profiles were consistent with high NtcA activity resulting from an excess of 2OG, and therefore carbon in the cell. However, NtcA expression was not identified in any of the toxic strain proteomes analysed here, suggesting that rather than a fluctuation in the expression levels of NtcA,
165 Chapter 6 it is more likely that the DNA-binding affinity of this transcriptional regulator changes in the presence of microcystin.
The identification of proteins involved in C-metabolism differentially displayed between toxic and non-toxic in Chapters 4 and 5, but not Chapter 2, may be caused by the different growth stage in which the cells were harvested. In Chapter 2, the protein expression analysis was performed on cultures in late exponential growth phase.
Therefore, the differential expression of PII, NrtA, TrxM, NdhK and the carboxysome components CcmK2 and CcmL may be the result of the metabolic flow of fixed carbon to 2OG production during the early exponential growth phase (Chapter 4 and Chapter 5). This may lead to a high C:N ratio that needs to be corrected by increased nitrate uptake via NtcA regulation.
Lower expression levels of photosystem and phycobilisome components and up- regulation of thioredoxin peroxidase, inactivating thioredoxin A and its targets, appear to be characteristic of toxic M. aeruginosa (Chapter 4, Table 4.3.1 and Chapter 5, Table 5.3.2). These expression patterns could aid in the dominance of microcystin-producers in high light (Edwin, et al., 2007), where a high photosynthetic activity would result in photoinhibition and the formation of reactive oxygen species. Similarly, the high expression levels of TrxM in these toxic strains (Chapter 2, Table 2.3.2), when they are nutrient-replete may attenuate the severity of oxidative stress during iron or high light stress.
166 Conclusions and future directions
Figure 6.1. Protein expression in M. aeruginosa grown in nutrient-replete conditions. The protein expression profile is shown relative to non-toxic M. aeruginosa strains, proteins up-regulated in the toxic strains are in red, down-regulated proteins are in green. Inhibitory processs are marked with a red cross. Details of the expression of proteins involved in carbon metabolism are shown in Figure 6.2. PM: plasma membrane; TM: thylakoid membrane; Ccm: carbon concentrating mechanism; C: carbon; N: nitrogen; Gln: glutamine; Glu: glutamate; ROS: reactive oxygen species; 2OG: 2-oxoglutarate.
167 Chapter 6
Figure 6.2. Carbon metabolism in M. aeruginosa PCC 7806 grown in nutrient- replete conditions. The protein expression profile is shown relative to non-toxic M. aeruginosa PCC 7005 and PCC 7806 mcyH-, proteins up-regulated in the toxic PCC 7806 are in red, down-regulated proteins are in green. Inhibitory processes are marked with a red cross. Points in the pathway subject to thioredoxin activation are also shown. Gln: glutamine; Glu: glutamate; PHA: polyhydroxyalkanoate; 2OG: 2-oxoglutarate. Enzymes affected are: 1: fructose-1,6-bisphosphatase; 2: glyceraldehyde-3- dehydrogenase; 3: phosphoglycerate kinase; 4: phosphoenolpyruvate synthase; 5: malic enzyme; 6: PHA-specific beta-ketothiolase; 7: glutamine synthetase.
168 Conclusions and future directions
6.2.4. Toxicity and adaptation to nutrient stress
The protein expression studies performed here revealed differences between toxic and non-toxic strains of M. aeruginosa in nutrient-replete conditions, in particular with reference to carbon and nitrogen reserves in the cell, as well as redox status due to altered photosynthesis and thioredoxin activity. This affects the stress response once cyanobacteria are placed in a nutrient-limited environment as highlighted in Chapters 4 and 5.
6.2.4.1. Adaptation to iron stress
The transcriptional analysis performed in Chapter 3, revealed that the transcription of Fur-homologues in M. aeruginosa grown in iron limitation was significantly different in all strains (Table 3.3.1). In the following proteomic analysis (Chapter 4), many of the proteins found to be differentially displayed in limited iron, were targets of Fur, including FeoB, FutA and IsiB, but no differential expression of Fur proteins was observed. This may be due to the low levels of expressed Fur proteins being obscured by highly abundant phycobiliproteins in the total protein extract. The regulation of Fur in the different M. aeruginosa strains may also occur post-transcriptionaly via the activity of an anti-furA RNA molecule, such as the one observed in Anabaena sp. PCC 7120 (Hernandez, et al., 2006). This antisense RNA is able to regulate the stability of Fur transcripts and its deletion mutant has been shown to have disprganised thylakoid membranes, reduced pigment content and reduced iron storage (Hernandez, et al., 2010).
The differences in Fur transcription, combined with the expected differential regulation of NtcA, would affect the iron starvation response as was observed in Chapter 4. The most pronounced expression changes following iron limitation were the up-regulation of PCC 7806 photosynthetic proteins in 10 nM Fe and the increased expression of carbon metabolism proteins in this strain (Chapter 4, Table 4.3.3). This response could come about as a result of a lower C:N ratio and therefore, an attempt to increase the input of carbon into the system, but decrease nitrogen (such as the nitrate NrtA transporter). This result was supported further by the lack of bleaching in this strain only (Chapter 3, Figure 3.3.1), and appeared to be an iron-specific effect, as both pronounced chlorosis and a general decrease in metabolic activity (including proteins that were upregulated during iron stress) were observed in nitrogen stress (Chapter 5,
169 Chapter 6
Figure 5.3.1 and Table 5.3.3). As discussed above, the higher background levels of thioredoxin in nutrient-replete toxic M. aeruginosa cells may allow the maintenance of higher photosynthetic activity during iron limitation without the formation of excessive ROS.
The ability of microcystin-producing strains to survive iron stress over non-toxic strains of the same cyanobacterial species, combined with an increase of microcystin synthesis during iron stress has been associated with the putative binding of the toxin molecule to divalent metal ions and its function as a siderophore (Saito, et al., 2008, Utkilen and Gjolme, 1995). In this study, the toxic strain PCC 7806 did not exhibit increased 55Fe uptake into the periplasmic space relative to non-toxic strains, nor did the non-toxic M. aeruginosa appear to be disadvantaged during the iron uptake process (Manabu Fujii, personal communication). However, the toxic M. aeruginosa isolate was the only strain capable of up-regulating expression and transcription of the homologue of the Fe(III)- binding protein FutA, which was confirmed to be located in the cellular periphery by confocal microscopy (Chapter 3, Figure 3.3.6 and Chapter 4, Table 4.3.3). On the other hand, an Fe(II) transporter, FeoB, associated with severe iron stress, was increased in the PCC 7806 mcyH- mutant (Chapter 3, Table 3.3.1 and Chapter 4, Table 4.3.5). Additionally, this strain could not survive in Fraquil* media containing 1000 nM for more than two months. Thus, microcystin synthesis appears to be beneficial for the cells during iron starvation, supporting previous reports in the literature (Utkilen and Gjolme, 1995). However, as was observed from the proteomic analysis in Chapter 4, wild-type non-toxic strains such as PCC 7005 can overcome the lack of microcystin by adjusting their photosynthesis and C:N metabolism to iron-depleted conditions (Table 4.3.4). An important finding of the iron-stress investigation was the observation that only part of the cyanobacterial population within the same culture was capable of expressing FutA (Chapter 3, Figure 3.3.7). This part of the population will then be capable of adjusting to the environmental stress. Such response may have implications in water bodies inhabited by cyanobacteria, where nutrient concentration can fluctuate rapidly and only the adapted population will be able to survive. It is also similar to the quorum-sensing starvation response in heterotrophic bacteria, in which part of the population is sacrificed prior to entry in stationary growth phase, in order to allow the remaining cells to survive (Lazazzera, 2000).
170 Conclusions and future directions
6.2.4.2. Adaptation to nitrogen stress
In contrast to iron stress, nitrogen limitation did not induce a change in microcystin synthesis. In the absence of nitrogen, all analysed strains entered a stationary state and became chlorotic (Chapter 5, Section 5.3.1). Despite the similar growth of all strains, the proteome analysis revealed a contrasting nitrogen-stress response in the expression of proteins between the toxic and non-toxic M. aeruginosa (Chapter 5). In the toxic PCC 7806 the response was a general metabolic shut-down. By comparison, in the non- toxic M. aeruginosa excess carbon was stored in PHA granules, and a high metabolic flow to 2OG and GlnA was observed (Figure 5.4.1). This response is characteristic of high NtcA activity induced by high levels of 2OG during nitrogen starvation, but was not reflected in the transport of nitrogenous compounds in the cell (Chapter 5). Instead, amino acids could have come from the breakdown of the photosynthetic proteins. These results show that the presence of microcystin in the cells affected the macronutrient starvation response, as well as the responses to iron starvation inducing oxidative stress in cyanobacteria.
The proteomes of both nutrient-replete and nutrient-starved toxic and non-toxic cells of M. aeruginosa revealed differences in several NtcA-regulated processes, including 2OG and glutamate synthesis, carbon fixation, thioredoxin and Fur activity, and nitrate transport. It is possible that the toxin molecule can interact with NtcA and change its affinity for the target promoter sequence, similar to DTT and 2OG (Espinosa, et al., 2006, Wisen, et al., 2004). NtcA from M. aeruginosa has been heterologously expressed in an E. coli host (Ginn and Neilan, 2010) and could be used in isothermal calorimetry or surface plasmon resonance to investigate whether microcystin can bind to the NtcA dimer or NtcA-DNA complex. This could reflect a positive feedback mechanism in toxic strains where NtcA binds to and activates transcription in the mcy gene cluster, and microcystin then enhances the affinity of mcy-bound NtcA stimulating toxin production further. Localisation of the toxin in the cell has revealed that it is present in the nucleosome region, where NtcA would be expected to be concentrated. However, it is also associated with other cellular components, including the thylakoid membranes and carboxysomes and has been shown to bind to phycobilisomes and non- specifcally to proteins, when cells are lysed (Gerbersdorf, 2006, Juettner and Luethi, 2008, Vela, et al., 2008). Therefore, a direct interaction of microcystin with NtcA is not likely to be the only function of microcystin.
171 Chapter 6
6.3. Future directions
6.3.1. Microcystin production in the laboratory and during bloom formation
The proteome investigation carried out here concentrated on M. aeruginosa grown in batch culture. This experimental design results in a closed environment where nutrients become continuously depleted with a parallel accumulation of microcystin in the media as a result of cell lysis, thus limiting the control on the media chemical composition. It is therefore necessary to repeat these experiments in continuous culture, which may be more similar to the nutrient flux that cyanobacteria experience in the environment. Mantaining continuous culture will also aid in the prevention of self-shading resulting from the growth in nutrient-rich media, and therefore variable light availability which has a significant effect on cyanobacterial physiology. In addition, proteome studies provide a single snapshot of the processes that occur in a cell at the moment of protein extraction, but as was observed when transcription of genes involved in iron stress was investigated (Chapter 3, Table 3.3.1), the stress response to nutrient limitation is highly dynamic and does not affect all cells in the population equally (Figure 3.3.7). It is likely that more detailed investigations of nutrient stress in cells harvested in different growth stages will provide a more complete picture of the processes that occur in M. aeruginosa and identify further differences between toxic and non-toxic strains.
The results presented here revealed that not only a high diversity in protein expression in strains of M. aeruginosa exists, but the expression patterns are likely to change with prolonged culture in the laboratory. Examples of the changes that occur in laboratory- cultured Microcystis spp. are the loss of gas vesicles and the inability of these isolates to form colonies, as well as the decreased growth of strains in axenic culture when the mucilage-associated heterotrophic bacteria are removed (Kehr, et al., 2006, Mlouka, et al., 2004, Schatz, et al., 2005). Ignoring this existing diversity and the adaptation processes to which wild-type non-toxic, but not genetically-manipulated strains such as the mcy- mutant strains, would lead to conflicting results that may not be environmentally-relevant. This is illustrated by the identification of the microcystin related protein A (MrpA) (Dittmann, et al., 2001) that was confirmed in the current proteomic analysis to be specific for the PCC 7806 strain and its mcyH- mutant, but not other non-toxic strains analysed. However, the proteomic analysis of a large number of strains is currently a cumbersome and expensive process.
172 Conclusions and future directions
Metaproteomics is an emerging tool available to the study of complex relationships between bacterial populations and the environment. The term was first proposed by Wilmes and Bond (2004) and refers to the large scale proteomic analysis of environmental microorganisms in a given point in time. This technique has already been applied to fresh- and salt-water microbial communities (Kan, et al., 2005, Pierre-Alain, et al., 2007). Complemented by metagenomics, the genomic analysis of uncultured microorganisms, metaproteomic research provides a deep understanding of the metabolic potential in mixed communities (Wilmes and Bond, 2006). With the increase of genomic information on cyanobacteria and the development of optimized protocols for high-throughput techniques for cyanobacteria, metagenomic and metaproteomic studies of a bloom community would be possible.
Similarly, sub-cellular proteomics was beyond the scope of the current study, but may be useful for future proteome studies in bloom-forming cyanobacteria. This approach separates the cellular compartments so that specific metabolic processes can be studied and less abundant proteins can be identified. Protein fractionation protocols are already available for cyanobacteria, with reports on the proteome composition of the thylakoid and plasma membranes, carboxysomes and the periplasmic space (Fulda, et al., 2000, Gonzales, et al., 2005, Huang, et al., 2002, Srivastava, et al., 2005). The application of a sub-cellular proteomics approach to M. aeruginosa will allow a more focused study of the photosynthetic and transport protein expression changes that occur when these organisms are subjected to nutrient stress. It could aid in the identification of low- abundance proteins, such as transcriptional regulators, and may reveal further metabolic processes, which are affected by the presence of microcystin.
6.3.2. Hepatotoxins in other cyanobacterial species
Microcystins are produced not only by M. aeruginosa, but also several other cyanobacterial species with a diverse lifestyle, ranging from nitrogen-fixing, psychrotrophic to terrestrial lichen-symbionts (Jungblut, et al., 2006, Kaasalainen, et al., 2009, Oksanen, et al., 2004). This implies a broad range of environmental conditions during which toxicity may be beneficial for these microorganisms. Although the focus of this thesis was M. aeruginosa, future physiological studies on the remaining microcystin-producers could confirm the findings from this study and indicate a possible conserved function for microcystin. This is particularly important in diazotrophic cyanobacteria, such as Anabaena spp. due to the role of the global nitrogen
173 Chapter 6 regulator on the nitrogen fixation process. If microcystin is indeed involved in NtcA regulation as proposed here, this effect of toxicity will be expected to be further enhanced in nitrogen-fixing cyanobacteria.
Another interesting aspect is the existence of nodularin, which is thought to have arisen as a result of gene deletions in the mcy gene cluster (Moffitt and Neilan, 2004). Although this hepatotoxin is significantly less studied than microcystin, it has similar NtcA and Fur-binding motifs in its promoter (Moffitt, 2003). This, together with the proposed common origin of the two hepatotoxins, suggests that a similar regulatory mechanism is in place for these cyanobacterial metabolites. However, studies on the effects of environmental factors have found that the synthesis of nodularin is linked to the growth rate of the cell, and changes in toxicity during non-N2-fixing conditions are due to the decrease in cell mass rather than toxin regulation specifically (Jonasson, et al., 2008). The co-existence of the toxic strains Nodularia spumigena and N. harveyana PCC 7804, with the non-toxic N. harveyana and N. sphaerocarpa within the same species (Jonasson, et al., 2008) would allow for a similar quantitative proteomic approach as the one that was used in this thesis to be carried out.
6.3.3. Microcystin isoforms and other non-ribosomal peptides
Historically microcystin has been the main peptide of interest in M. aeruginosa investigations due to its toxicity and prevalence in drinking and recreational water reservoirs. This organism also produces a multitude of other peptides with unknown function such as cyanopeptolins, microginins and anaebenopeptolin (Martins, et al., 2009, Welker, et al., 2006). As was observed here, a change in cyanopeptolin gene transcription occurs when mcyH, a gene from the microcystin cluster is inactivated (Chapter 3, Table 3.3.1). These observations suggest that the non-ribosomal peptide synthesis in the cell may be coordinated and they may have an interchangeable, or a complimentary role for the cellular metabolism. In addition, a multitude of microcystin isoforms are produced, and this diversity has been established to be linked to nutrient availability, although the ecological advantage of replacing the variable amino acids in the toxin molecule is not clear (Martins, et al., 2009, Rapala, et al., 1997, Tonk, et al., 2005, Waal, et al., 2009). Therefore, studies of the peptide diversity and dynamics of peptide synthesis in blooms, in conjunction with proteomic and transcriptomic analyses are necessary to contribute to our understanding of bloom-forming cyanobacteria.
174 Conclusions and future directions
6.5. Conclusion
This thesis describes a proteomic investigation in strains of the unicellular M. aeruginosa, with particular focus on microcystin production. The major findings suggest a large diversity in protein expression in strains of this cyanobacterial species and regulation of toxicity on a post-transcriptional level. The studies of protein expression of these cyanobacteria during nutrient-replete and nutrient-limited growth indicated a possible involvement of hepatotoxin production in metabolic control by the global nitrogen regulator NtcA. Further global expression studies of M. aeruginosa and other bloom-forming cyanobacteria will aid in expanding our knowledge on the role of toxin synthesis in the environment and in developing reliable toxic algal bloom monitoring and control programs.
175
APPENDIX A Primer Sequences
176
Table A.1.1. Primers used in this study.
Gene Primer Sequence (5’-3’) Target E (%)* Chapter Reference symbol 27F gagtttgatcctggctcag Cyanobacterial Jungblut 809R gcttcggcacggctcggg 16S n/a 2-5 16S rDNA et al, 2005 tcgata rpoC1F cctcagcgaagatcaatg Ginn and RNA polymerase gt rpoC1 91.45% 2-5 Neilan, gamma subunit rpoC1R ccgtttttgccccttacttt 2010 qmetF ttattccaagttgctcccca Microcystin Saker et qmetR ggaaatactgcacaacga synthesis mcyA 96.70% 3 al., 2005 g methyltransferase rtmcyH2F ttgtcttcgctccagcctat Microcystin rtmcyHR ggccgacgaaaattcaga associated ABC mcyH 91.50% 3 This study ta transporter mcnCF agctaaaacggcaaagga ca Cyanopeptolin mcnC 92.11% 3 This study mcnCR ccaattgcctccaaagttgt isiAF gattaaagcagcttgggca ac Iron stress induced isiA 100% 3 This study isiAR acgactggtgggcaggaa protein A (CP43’) at feoBF gattgatgcgtttggtggg at Ferrous uptake feoB 90.18% 3 This study feoBR ccgccgcaaatagggcat protein aa rtfutABCF tgtttggtcggggaaaatta Ferric uptake rtfutABCR tccaccccagggtaatgat futA 100% 3 This study transporter a Fur3F caatcatctcagtgccgaa ga Ferric uptake furA 99.16% 3,5 This study Fur3R accctcggccaattccaac regulator t Fur2F accttaaatcgggcagtat cc Ferric uptake furB 100% 3 This study Fur2R ctgcactgcaccctgtaatt regulator t Fur1F gtgactgtgggaatcgctg a Ferric uptake furC 92.31% 3 This study Fur1R aacacctatcagcgcgag regulator aaa trxMF tcaggaactgttgcaatcc a Thioredoxin M trxM 100% 2 This study trxMR agaatcggagccatcattt g MAE25790F aatcgaacccgataaaccc Hypothetical t MAE protein MAE 98% 2 This study MAE25790R tcacaaccgacaacagaa 25790 25790 gc ccmKF acggatcacgattgttggtt Carboxysome ccmK3 95.15% 2 This study ccmKR tcggttttattgacggcttc shell subunit PIIF aagcgattatccgacccttt Nitrogen glnB 100% 2,5 This study PIIR attgacctttctgacgaccg regulatory protein ccmLF tctgcttttgcaattcatcg Carboxysome ccmLR caataatccccaccaccat ccmL 100% 2 This study shell subunit c MAE06820F agtggtagccgaaagcga Hypothetical ta MAE protein MAE 98.17% 2 This study MAE06820R tcccttccaagaacaaatg 06820 06820 g
177
Table A.1.1. Primers used in this study (continued).
Gene Primer Sequence (5’-3’) Target E (%)* Chapter Reference symbol nrtAF tgatggtcgcaaaattgaa Nitrate ABC nrtA 99.32% 2 This study a transporter nrtAR ggaatatagccccaacgg
at ndhKF actacccacaaaatgcag NADH gc dehydrogenase ndhK 100% 2 This study ndhKR ctctttcggaggtgcttgac subunit K MAE06360F cggtgtcggcacgatcgg 2 tt Carboxymethylene MAE 100% 2 This study MAE06360R cccgtggtgattcccgcac butenoidase 06360 2 c ntcArealF catttccgtttgcagaatcc Global nitrogen Ginn and ntcArealR tgtttttggggtgctatcct control ntcA 91.50% 2,5 Neilan transcriptional (2010) regulator
* Primer efficiency for qRT-PCR primers was determined according to Pfaffl, 2001, using the equation E=10[-1/slope], where E=2 is 100%.
178
APPENDIX B Media Composition
Table A.2.1. Modified Fraquil (Fraquil*) media composition.
Final Fraquil Stock Diluent Concentration Concentration Salt Stocks CaCl2 0.26 mM 0.26 M Milli-Q MgSO4 0.15 mM 0.15 M Milli-Q NaHCO3 0.5 mM 0.50 M Milli-Q NaNO3 1 mM 0.10 M Milli-Q K2HPO4 0.05 mM 0.05 M Milli-Q HEPES 1 mM 1.00 M Milli-Q Trace Metal Stocks . CuSO4 5H2O 0.162 μM 0.162 mM 0.01M HCl . CoCl2 6H2O 0.050 μM 0.503 mM 0.01M HCl MnCl2 0.603 μM 0.603 mM 0.01M HCl . ZnSO4 7H2O 1.2 μM 1.200 mM 0.01M HCl Na2SeO3 0.01 μM 0.100 mM 0.01M HCl . (NH4)6Mo7O24 4H2O 0.01 μM 0.100 mM 0.01M HCl Fe-Ligand Stocks . FeCl3 6H2O 10 μM – 0.01 μM 1.0 mM 0.01M HCl . Na2EDTA 2H2O 13 μM 13 mM Milli-Q Vitamin Solution Thiamine HCl 0.296 μM Add as powder Milli-Q Biotin 2.05 nM 2.05 mM Milli-Q Cyanocobalamin 0.369 nM 0.369 mM Milli-Q
Fujii, M., Rose, A. L., Omura, T., and Waite, T. D. (2010) Effect of Fe(II) and Fe(III) transformation kinetics on iron acquisition by a toxic strain of Microcystis aeruginosa. Environ. Sci. Technol. 44:1980-1986. Andersen, R. A. (2005). Algal culturing techniques. Elsevier/ Academic Press: Burlington, MA, p x, 578 p.
180
* Table A.2.2. BG11 and nitrogen-free BG11 (BG110 ) media composition.
* Stock concentration Stock concentration (BG110 ) (BG11) Stock 1 (10x) Na2MgEDTA 0.10 g/L 0.01 g/L Ferric citrate - 0.52 g/L Ferric ammonium 0.60 g/L - citrate Citric acid . 1H2O 0.60 g/L 0.60 g/L CaCl2 . 2H2O 3.60 g/L 3.60 g/L Stock 2 (10x) MgSO4 . 7H2O 7.50 g/L 7.50 g/L Stock 3 (10x) K2HPO4 . 3H2O 4.00 g/L 4.00 g/L Stock 5 (100x) H3BO3 2.86 g/L 2.86 g/L MnCl2 . 4H2O 1.81 g/L 1.81 g/L ZnSO4 . 7H2O 0.222 g/L 0.222 g/L CuSO4 . 5H2O 0.079 g/L 0.079 g/L CoCl2 . 6H2O 0.050 g/L 0.050 g/L NaMoO4 . 2H2O 0.391 g/L 0.391 g/L Add to combined stocks Na2CO3 0.02 g/L 0.02 g/L NaNO3 1.50 g/L -
Rippka R, J Deruelles, J Waterbury, M. Herdman and R. Stainer (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J. Gen. Microbiol. 111: 1-61.
181
APPENDIX C Q-TOF MS/MS of microcystin isoforms purified from M. aeruginosa UWOCC MRC and UWOCC MRD
182
Figure A.3.1. MS/MS spectra of putative microcystin isoforms in the Microcystis aeruginosa strains UWOCC MRC (A., B.) and UWOCC MRD (C., D.). Annotations of the spectra follow in Supplementary Table 1.
A. Microcystin fraction I (UWOCC MRC) 50.00000000 RA_080617_7A_MS 29 (0.554) Cm (28:137) TOF MS ES+ 100 995.5111 3.16e4 %
996.5287
375.1793 1061.4768 365.0927 981.5042 1062.4796 329.1500 553.2960 599.3363 0 m/z 200 400 600 800 1000 1200
183
Microcystin-LR (m/z 995.511) MS/MS spectrum 50.00000000 RA_080617_7A_995_MS 245 (4.622) Cm (95:297) TOF MSMS 995.50ES+ 100 70 %
247.6416
226.8670 194.8956 135.0071 190.8789 274.1087 283.0319318.0571
0 m/z 100 120 140 160 180 200 220 240 260 280 300 320 50.00000000 RA_080617_7A_995_MS 245 (4.622) Cm (95:297) TOF MSMS 995.50ES+ 494.3333 495.3194 100 30 495.3356
495.3519 %
353.1988 274.1087 496.3227 350.0110 384.1528 410.0381 494.2355 579.1150 548.2321 450.9529
0 m/z 275 300 325 350 375 400 425 450 475 500 525 550 575 600 50.00000000 RA_080617_7A_995_MS 245 (4.622) Cm (95:297) TOF MSMS 995.50ES+ 995.5342 100 36
995.5575 644.2653
996.5518 %
644.3024 645.2885 644.2095 997.5352 683.2610 995.4302 643.7725 935.6921 771.7581790.8973 994.7483 997.5583 885.7662
0 m/z 600 650 700 750 800 850 900 950 1000 1050 1100
184
B. Microcystin fraction II (UWOCC MRC) 8.00000000 RA_080617_7B 176 (3.319) Cm (45:176) TOF MS ES+ 274.0541 100 3.90e4
152.9900 %
995.5227 275.0864 445.1060 430.9063 509.2607 996.5287 560.8619
0 m/z 100 200 300 400 500 600 700 800 900 1000 1100
185
[D-Asp3] MCYST-LR (m/z 981.5) MS/MS spectrum 50.00000000 RA_080617_7C_981_MSMS 73 (1.387) Cm (65:161) TOF MSMS 981.50ES+ 174.0814 100 171
213.0563
163.0544 272.1163 % 157.0481
135.0071 268.1472
255.0960 289.1554 200.0723 292.1416 129.0329
0 m/z 60 80 100 120 140 160 180 200 220 240 260 280 300 50.00000000 RA_080617_7C_981_MSMS 73 (1.387) Cm (65:161) TOF MSMS 981.50ES+ 375.1935 110 100
375.1722
539.3048 %
289.1554 599.3363 446.2429 539.2537 368.1795 556.3162 599.3542 446.2119 456.2431 538.3099 556.3335 292.1416 385.2235 522.2729 600.3412 397.2675 509.2523
0 m/z 300 325 350 375 400 425 450 475 500 525 550 575 600 50.00000000 RA_080617_7C_981_MSMS 73 (1.387) Cm (65:161) TOF MSMS 981.50ES+ 981.5272 100 555 % 982.5031
599.3363 983.5255 953.5619 0 m/z 600 700 800 900 1000 1100
186
C. Microcystin fraction I (UWOCC MRD) 8.00000000 RA_080617_8D_1074_MSMS 184 (3.475) Cm (50:248) TOF MSMS 1074.40ES+ 100 469.1828 1074.4434 134 469.1590
1075.4524
184.0086 397.1799 1075.4763 %
470.1594 1076.4858 598.2247 628.2971 1002.4938 1076.5339
0 m/z 200 400 600 800 1000 1200 1400
187
Microcystin-YY (m/z 1052.3770) MS/MS spectrum 8.00000000 RA_080617_8D_1052_MSMS 83 (1.575) Cm (61:185) TOF MSMS 1052.40ES+ 100 213.0563 64
163.0544 265.1015
163.0638 213.0402 310.1412 258.1723 293.1069
% 136.0060 318.1487 135.0029 195.0389 292.1353 246.1386
134.9305 181.0519
0 m/z 60 80 100 120 140 160 180 200 220 240 260 280 300 320 8.00000000 RA_080617_8D_1052_MSMS 83 (1.575) Cm (61:185) TOF MSMS 1052.40ES+ 375.1793 100 94
% 544.2335 310.1412 446.2197 473.2148 544.2592 318.1487 610.2465 376.1806 609.2606 509.2854 527.2326 627.2598 376.2233 609.2153 401.1765 0 m/z 300 325 350 375 400 425 450 475 500 525 550 575 600 625 8.00000000 RA_080617_8D_1052_MSMS 83 (1.575) Cm (61:185) TOF MSMS 1052.40ES+ 100 610.2465 25
610.2738
627.2598 % 1052.3770
627.2874 918.4575 1051.4500 628.2695 1052.4718 721.2957 900.3934 1051.3430 1054.3982 1033.3992 1112.4095
0 m/z 600 700 800 900 1000 1100 1200 1300 1400
188
MCYST (m/z 1006.4301) MS/MS spectrum 30.00000000 RA_080617_8D_1006_MSMS 70 (1.330) Cm (43:147) TOF MSMS 1006.40ES+ 258.1664 100 19
328.9307 258.1781 213.0456 213.0563 134.9262 163.0591 264.0941 328.9506 219.0578 347.1789
% 297.1862 134.9177 195.0491 297.1609
134.9006
0 m/z 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 30.00000000 RA_080617_8D_1006_MSMS 70 (1.330) Cm (43:147) TOF MSMS 1006.40ES+ 375.1864 100 44
375.2077 % 427.1878 446.2197 498.1935 328.9307 446.2351 509.2689 564.1984 328.9506 581.2239 610.2828 563.2416 648.9604
0 m/z 300 325 350 375 400 425 450 475 500 525 550 575 600 625 650 30.00000000 RA_080617_8D_1006_MSMS 70 (1.330) Cm (43:147) TOF MSMS 1006.40ES+ 1006.4301 100 18
1006.4650
872.3405 872.3619 609.2968 872.3945 1007.4416 872.4269 % 873.3470 1007.4648 610.2828 873.4011 1008.4768 1006.3604 786.9458 854.4383 1009.3380 678.7963736.9055 989.3710 854.3312 1009.5591 1112.4830 1189.2830
0 m/z 700 800 900 1000 1100 1200
189
MCYST (m/z 974.4579) MS/MS spectrum 8.00000000 RA_080617_8D_974_MSMS 144 (2.722) Cm (82:238) TOF MSMS 974.40ES+ 158.9006 100 42
158.9098 % 134.9305 134.9177 213.0616 195.0337 239.1456 258.1723
293.1069 328.9307
0 m/z 60 80 100 120 140 160 180 200 220 240 260 280 300 320 8.00000000 RA_080617_8D_974_MSMS 144 (2.722) Cm (82:238) TOF MSMS 974.40ES+ 375.2006 100 30
375.1651 % 430.9063 446.2197 352.8959 446.2661 532.2581 395.1891 570.9099 328.9307 466.2354 609.2787 644.8513 512.9380
0 m/z 300 350 400 450 500 550 600 650 8.00000000 RA_080617_8D_974_MSMS 144 (2.722) Cm (82:238) TOF MSMS 974.40ES+ 609.2787 100 7 644.8513
748.7676
840.3607
% 846.9050 976.3459 1072.3788
974.4579 1119.3978
0 m/z 70080090010001100120013001400
190
D. Microcystin fraction II (UWOCC MRD) 8.00000000 RA_080617_8F 79 (1.494) Cm (25:79) TOF MS ES+ 239.1230 100 1.80e4
148.9491
% 445.1060
1036.4753 371.0957 446.1113 1037.4902 1002.4938 551.7115 1059.4723 0 m/z 200 400 600 800 1000 1200 1400
191
MCYST (m/z 1002.4938) MS/MS spectrum 8.00000000 RA_080617_8F_1002_MSMS 48 (0.917) Cm (17:85) TOF MSMS 1002.40ES+ 258.1605 36 100
258.1723 163.0544 213.0616 258.1841
213.0509
% 195.0337 260.1545 163.0450 215.1048
195.0184 215.1156 265.1313 297.1925 243.1065 318.1356 155.0330 135.0029 128.9704 0 m/z 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 8.00000000 RA_080617_8F_1002_MSMS 48 (0.917) Cm (17:85) TOF MSMS 1002.40ES+ 100 494.2437 106
375.1864
% 423.2292 423.2141 560.2633 509.2523 560.2807 446.2197 609.2877
297.1736 365.1067 376.1948 510.2367 610.3100
0 m/z 300 325 350 375 400 425 450 475 500 525 550 575 600 625 8.00000000 RA_080617_8F_1002_MSMS 48 (0.917) Cm (17:85) TOF MSMS 1002.40ES+ 868.4168 1002.4938 100 72
868.3953
1002.5403
1002.4590 851.3889 1003.5034 % 851.3568 869.4429 1003.5266 609.2877 1004.4553 870.4262 850.4058 1004.5248 610.3100 985.4451 1004.5714 743.3740 984.5253 1005.4772
0 m/z 600 700 800 900 1000 1100 1200 1300 1400
192
MCYST-YM (O) (m/z 1036.4991) MS/MS spectrum 8.00000000 RA_080617_8F_1036_MSMS 27 (0.522) Cm (20:96) TOF MSMS 1036.40ES+ 294.1366 102 100 258.1664
163.0497 213.0509 249.1086 213.0616
195.0389 % 297.1736
277.1140 318.1421 246.1214 135.9974 318.1160 164.0531 237.1436 134.9347
0 m/z 120 140 160 180 200 220 240 260 280 300 320 8.00000000 RA_080617_8F_1036_MSMS 27 (0.522) Cm (20:96) TOF MSMS 1036.40ES+ 528.2425 100 310
375.1793 % 457.2062 294.1366 446.2351 594.2439 611.2787 529.2365 509.2689 580.3057 297.1736 376.1877 458.2094 612.2664 347.1857 440.1806 566.2713 0 m/z 300 325 350 375 400 425 450 475 500 525 550 575 600 625 8.00000000 RA_080617_8F_1036_MSMS 27 (0.522) Cm (20:96) TOF MSMS 1036.40ES+ 902.4063 100 113
611.2787
885.3954 903.4082
% 1036.4991
1036.4402 612.2664 903.4742 884.4037 1037.5020 1019.4899 612.3117 743.3340 723.3043 743.3740 904.3995 1038.4935 744.3832 904.4657 0 m/z 700 800 900 1000 1100 1200 1300
193
Table A.3.1. Assignment of ions from microcystin peaks in Q-TOF MS/MS used to identify microcystin variants. Microcystins identified only by their m/z were not fully characterized.
m/z MCYST- [D-Asp3] MCYST- MCYST- MCYST MCYST MCYST Ion identity LR MCYST- YY YM(O) m/z m/z m/z 974.4 LR 1006.4 1002.5 MCYST 995.5 981.5 1052.4 1036.5 1006.4 1002.5 974.4 MCYST– H2O 1033.4 MCYST-NH3 1019.5 989.4 985.4 MCYST -CO 968.6 953.6 MCYST - 135 918.5 902.4 872.4 868.4 840.4 [L-Arg-Adda- 682.2 DGlu-Mdha + H]+ [Mdha-D-Ala-L- 610.2 Tyr-D-MeAsp-L- Tyr + H]+ [L-Arg-Adda-D- 599.4 599.3 Glu + H]+ [Mdha-D-Ala-L- 553.3 Leu-D-MeAsp-L- Arg + H]+ [Mdha-D-Ala-L- 539.3 Leu-D-Asp-L-Arg + H]+ [D-Glu-Mdha-D- 446.2 446.2 446.2 Ala-L-Tyr + H]+ [Adda-D-Glu- 375.2 375.2 375.2 375.2 375.2 375.2 375.2 Mdha + H]+ – 135 – NH3 [Mdha-D-Ala-L- 318.1 318.1 Tyr + H]+ [Tyr-D-MeAsp + 310.1 + H] + NH3 [L-Arg-D-Asp + 289.2 + NH4] [L-Arg-D-MeAsp 286.1 + H]+ [Met(O)-DMeAsp 277.1 + H]+ [L-Arg-D-Asp + 272.1 H]+ [Mdha-D-Ala-L- 268.1 Leu + H]+ [D-Glu-Mdha + 213.1 213.1 213.1 213.1 213.0 213.0 213.1 H]+ + L-Arg + NH4 174.1 174.1 + [C11H14O + H] 163.1 163.1 163.1 163.1 163.1 [Mdha-L-Ala + H]+ 155.0 [Immonium of 136.0 135.9 Tyr]+ PheCH2CHOCH3 134.9 135.0 135.0 135.0 135.0 135.0 135.0
194
APPENDIX D Supplemental Data
195
The complete results from database searches of the proteome mass spectrometry data are available as Excel and Scaffold files on the data DVD accompanying this thesis. Scaffold Viewer is available for download at:
www.proteomesoftware.com/Proteome_software_prod_Scaffold_download.html.
Listed below are the file names and the information relevant for each chapter.
Chapter 2
Protein lists for M. aeruginosa strains (Chapter2_Protein Report.xls)
Peptide lists for M. aeruginosa strains (Chapter2_Peptide Report.xls) nSAF analysis of protein expression in toxic versus non-toxic M. aeruginosa (Chapter2_nSAF.csv)
Chapter 4
Iron stress proteomics - database search and analysis results (Chapter4_Festress.sfd)
Protein list (Chapter4_Protein Report.xls)
Peptide list (Chapter4_Peptide Report.xls)
Chapter 5
Nitrogen stress proteomics – database search and analysis results (Chapter5_Nstress.sfd)
Protein list (Chapter5_Protein Report.xls)
Peptide list (Chapter5_Peptide Report.xls)
196
References
197
Agarwala, R., Ortlebb, S., Sainisa, J. K., and Melzerb, M. (2009) Immunoelectron microscopy for locating Calvin cycle enzymes in the thylakoids of Synechocystis 6803. Mol. Plant 2(1):32-42 Ahmad, R., Brandsdal, B. O., Michaud-Soret, I., and Willassen, N.-P. (2009) Ferric uptake regulator protein: Binding free energy calculations and per- residue free energy decomposition. Proteins 75:373-386. Aldehni, M. F., Sauer, J., Spielhaupter, C., Schmid, R., and Forchhammer, K. (2003) Signal transduction protein PII is required for NtcA-regulated gene expression during nitrogen deprivation in the cyanobacterium Synechococcus elongatus strain PCC 7942. J. Bacteriol. 185(8):2582-2591. Allakhverdiev, S. I., Nishiyama, Y., Miyairi, S., Yamamoto, H., Inagaki, N., Kanesaki, Y., and Murata, N. (2002) Salt stress inhibits the repair of photodamaged photosystem II by suppressing the transcription and translation of psbA genes in Synechocystis. Plant Physiol. 130:1443-1453. Andersen, R. A. (2005). Algal culturing techniques. Elsevier/ Academic Press, Burlington. Anderson, D. C., Campbell, E. L., and Meeks, J. C. (2006) A soluble 3D LC/MS/MS proteome of the filamentous cyanobacterium Nostoc punctiforme. J. Proteome Res. 5:3096-3104. Ashby, M. K., and Mullineaux, C. W. (1999) Cyanobacterial ycf27 gene products regulate energy transfer from phycobilisomes to photosystems I and II. FEMS Microbiology Letters 181:253-260. Atkins, R., Rose, T., Brown, R. S., and Robb, M. (2001) The Microcystis cyanobacteria bloom in the Swan River-February 2000. Water Sci. Technol. 43(9):107-114. Aurora, R., Hihara, Y., Singh, A. K., and Pakrasi, H. B. (2007) A network of genes regulated by light in Cyanobacteria. OMICS 11(2):166-185. Babica, P., Blaha, L., and Marsalek, B. (2006) Exploring the natural role of microcystins - a review of effects on photoautotrophic organisms J. Phycol. 42:9-20. Badarau, A., Firbank, S. J., Waldron, K. J., Yanagisawa, S., Robinson, N. J., Banfield, M. J., and Dennison, C. (2008) FutA2 is a ferric binding protein from Synechocystis PCC 6803. J. Biol. Chem. 283(18):12520-12527. Badger, M. R., and Price, G. D. (2003) CO2 concentrating mechanisms in cyanobacteria: molecular components, their diversity and evolution. J. Exp. Bot. 54(383):609-622. Badger, M. R., Price, G. D., Long, B. M., and Woodger, F. J. (2006) The environmental plasticity and ecological genomics of the cyanobacterial CO2 concentrating mechanism. J. Exp. Bot. 57(2):249-265. Bagu, J. R., Sykes, B. D., Craig, M. M., and Holmes, C. F. B. (1997) A molecular basis for different interactions of marine toxins with protein phosphatase-1. J. Biol. Chem. 272(8):5087-5097. Bailey, S., and Grossman, A. (2008) Photoprotection in Cyanobacteria: Regulation of light harvesting. Photochem. Photobiol. 84:1410-1420. Baptista, M. S., and Vasconcelos, M. T. (2006) Cyanobacteria metal interactions: Requirements, toxicity, and ecological implications. Crit. Rev. Microbiol. 32(3):127-137. Barbeau, K., Rue, E. L., Bruland, K. W., and Butler, A. (2001) Photochemical cycling of iron in the surface ocean mediated by microbial iron(III)-binding ligands. Nature 413:409-413.
Barrios-Llerena, M. E., Chong, P. K., Gan, C. S., Snijders, A. P. L., Reardon, K. F., and C.Wright, P. (2006) Shotgun proteomics of cyanobacteria applications of experimental and data-mining techniques. Brief. Funct. Genomic. Proteomic. 5(2):121-132. Barrios-Llerena, M. E., Reardon, K. F., and Wright, P. C. (2007a) 2-DE proteomic analysis of the model cyanobacterium Anabaena variabilis. Electrophoresis 28:1624-1632. Barrios-Llerena, M. E., Reardon, K. F., and Wright, P. C. (2007b) 2-DE proteomic analysis of the model cyanobacterium Anabaena variabilis. Electrophoresis 28(10):1624-1632. Battchikova, N., and Aro, E.-M. (2007) Cyanobacterial NDH-1 complexes: multiplicity in function and subunit composition. Physiol. Plant. 131:22-32. Bes, M. T., Hernandez, J. A., Peleato, M. L., and Fillat, M. F. (2001) Cloning, overexpression and interaction of recombinant Fur from the cyanobacterium Anabaena PCC 7119 with isiB and its own promoter. FEMS Microbiol. Lett. 194(2):187-92. Boekema, E. J., Hifney, A., Yakushevska, A. E., Piotrowski, M., Keegstra, W., Berry, S., Michel, K.-P., Pistorius, E. K., and Kruip, J. (2001) A giant chlorophyll-protein complex induced by iron deficiency in cyanobacteria. Nature 412:745-748. Braun, E., and Bachofen, R. (2004) Homoserine-lactones and microcystin in cyanobacterial assemblages in Swiss lakes. Hydrobiologia 522:271-280. Breci, L., Hattrup, E., Keeler, M., Letarte, J., Johnson, R. and Haynes, P. A. (2005) Comprehensive proteomics in yeast using chromatographic fractioniation, gas phase fractionation, protein gel electrophoresis, and isoelectric focusing. Proteomics 5: 2018-2028. Briand, E., Yéprémian, C., Humbert, J.-F., and Quiblier, C. (2008) Competition between microcystin- and non-microcystin-producing Planktothrix agardhii (cyanobacteria) strains under different environmental conditions. Environ. Microbiol. 10(12):3337-48. Brookes, J. D., and Ganf, G. G. (2001) Variations in the buoyancy response of Microcystis aeruginosa to nitrogen, phosphorus and light. J. Plankt. Res. 23(12):1399-1411. Brunberg, A.-K., and Blomqvist, P. (2002) Benthic overwintering of Microcystis colonies under different environmental conditions. J. Plankt. Res. 24(11):1247-1252. Burja, A. (2003) Cyanobacterial postgenomic research and systems biology. Trends Biotechnol. 21(11):504-511. Carmichael, W. W., and An, J. (1999) Using an enzyme linked immunosorbent assay (ELISA) and a protein phosphatase inhibition assay (PPIA) for the detection of microcystins and nodularins. Nat. Toxins 7(6):377- 385. Carre-Mlouka, A., Mejean, A., Quillardet, P., Ashida, H., Saito, Y., Yokota, A., Callebaut, I., Sekowska, A., Dittmann, E., Bouchier, C., and Marsac, N. T. d. (2006) A new Rubisco-like protein coexists with a photosynthetic Rubisco in the planktonic cyanobacteria Microcystis. J. Biol. Chem. 281(34):24462- 24471. Carrondo, M. A. (2003) Ferritins, iron uptake and storage from the bacterioferritin viewpoint. EMBO J. 22(9):1959-1968.
199
Castielli, O., Cerda, B. D. l., Navarro, J. A., Hervas, M., and Rosa, M. A. D. l. (2009) Proteomic analyses of the the response of cyanobacteria to different stress conditions. FEBS Lett. 583(11):1753-1758. Cerda, B. D. l., Castielli, O., V.Duran, R., Navarro, J. A., Hervas, M., and Rosa, M. A. D. l. (2007) A proteomic approach to iron and copper homeostasis in cyanobacteria. Brief. Funct. Genomic. Proteomic. 6(4):322-9. Christiansen, G., Molitor, C., Philmus, B., and Kurmayer, R. (2008) Non- toxic strains of cyanobacteria are the result of major gene deletion events induced by a transposable element. Mol. Biol. Evol. 25(8):1695-1704. Copp, J. N., Roberts, A. A., Marahiel, M. A., and Neilan, B. A. (2007) Characterization of PPTNs, a cyanobacterial phosphopantetheinyl transferase from Nodularia spumigena NSOR10. J. Bacteriol. 189(8):3133-3139. Corthals, G. L. and Rose, K. (2007) Quantitation in proteomics. In: Proteome Research - Concepts, Technology and Application. 2nd edition, editors Wilkins, M., Appel, R. D., Williams, K. L. and Hochstrasser, D. F., Springer Verlag Berlin Heidelberg, pp. 69-93. Craig, R., and Beavis, R. C. (2003) A method for reducing the time required to match protein sequences with tandem mass spectra. Rapid Commun. Mass Spectrom. 17:2310-2316. Craig, R., and Beavis, R. C. (2004) TANDEM: matching proteins with tandem mass spectra. Bioinformatics 20(9):1466-1467. Dagnino, D., Meireles, D. d. A., and Almeida, J. C. d. A. (2006) Growth of nutrient-replete Microcystis PCC 7806 cultures is inhibited by an extracellular signal produced by chlorotic cultures. Environ. Microbiol. 8(1):30-36. Dai, R., Liu, H., Qu, J., Zhao, X., Ru, J., and Hou, Y. (2008) Relationship of energy charge and toxin content of Microcystis aeruginosa in nitrogen-limited or phosphorous-limited cultures. Toxicon 51(4):649-58. Dittmann, E., Erhard, M., Kaebernick, M., Scheler, C., Neilan, B. A., Dohren, H. v., and Borner, T. (2001) Altered expression of two light- dependent genes in a microcystin-lacking mutant of Microcystis aeruginosa PCC 7806. Microbiology 147:3113-3119. Dittmann, E., Neilan, B. A., Erhard, M., Dohren, H. v., and Borner, T. (1997) Insertional mutagenesis of a peptide synthetase gene that is responsible for hepatotoxin production in the cyanobacterium Microcystis aeruginosa. Mol. Microbiol. 26(4):779-787. Dolganov, N., and Grossman, A. R. (1999) A polypeptide with similarity to phycocyanin a-subunit phycocyanobilin lyase involved in degradation of phycobilisomes. J. Bacteriol. 181(2):610-617. Domain, F., Houot, L., Chauvat, F., and Cassier-Chauvat, C. (2004) Function and regulation of the cyanobacterial genes lexA, recA and ruvB: LexA is critical to the survival of cells facing inorganic carbon starvation. Mol. Microbiol. 53(1):65-80. Downing, T. G., Sember, C. S., Gehringer, M. M., and Leukes, W. (2005) Medium N:P ratios and specific growth rate comodulate microcystin and protein content in Microcystis aeruginosa PCC7806 and M. aeruginosa UV027. Microb. Ecol. 49:468-473. Drechsel, H., and Jung, G. (1998) Peptide Siderophores. J. Peptide. Sci. 4:147- 181. Dvornyk, V., and Nevo, E. (2003) Genetic polymorphism of cyanobacteria under permanent natural stress: A lesson from the “Evolution Canyons”. Res. Microbiol. 154:79-84.
200
Edwin, W., Kardinaal, A., Tonk, L., Janse, I., Hol, S., Slot, P., Huisman, J., and Visser, P. M. (2007) Competition for light between toxic and non-toxic strains of the harmful cyanobacterium Microcystis. Appl. Environ. Microbiol. 73:2939-2946. Ehling-Schulz, M., Schulz, S., Wait, R., Görg, A., and Scherer, S. (2002) The UV-B stimulon of the terrestrial cyanobacterium Nostoc commune comprises early shock proteins and late acclimation proteins. Mol. Microbiol. 46(3):827- 843. Eisenhut, M., Wobeser, E. A. v., Jonas, L., Schubert, H., Ibelings, B. W., Bauwe, H., Matthijs, H. C. P., and Hagemann, M. (2007) Long-term response toward inorganic carbon limitation in wild type and glycolate turnover mutants of the cyanobacterium Synechocystis sp. strain PCC 6803. Plant Physiol. 144:1946-1959. Espinosa, H., Frochammer, K., Burillo, S., and Contreras, A. (2006) Interaction network in cyanobacterial nitrogen regulation: PipX, a protein that interacts in a 2-oxoglutarate dependent manner with PII and NtcA. Mol. Microbiol. 61(2):457-469. Ferreira, F., and Straus, N. A. (1994) Iron deprivation in cyanobacteria. J. Appl. Phycol. 6:199-210. Fewer, D. P., Tooming-Klunderud, A., Jokela, J., Wahlsten, M., Rouhiainen, L., Kristensen, T., Rohrlack, T., Jakobsen, K. S., and Sivonen, K. (2008) Natural occurrence of microcystin synthetase deletion mutants capable of producing microcystins in strains of the genus Anabaena (Cyanobacteria). Microbiology 154(Pt 4):1007-14. Figueiredo, D. d., Azeiteiro, U., Esteves, S., Gonçalves, F., and Pereira, M. (2004) Microcystin-producing blooms-a serious global public health issue. Ecotoxicol. Environ. Saf. 59(2):151-163. Fiore, M. F., Genuario, D. B., Silva, C. S. P. d., Shishido, T. K., Moraes, L. A. B., Neto, R. C., and Silva-Stenico, M. E. (2009) Microcystin production by a freshwater spring cyanobacterium of the genus Fischerella. Toxicon 53(7- 8):754-761. Florencio, F. J., Perez-Perez, M. E., Lopez-Maury, L., Mata-Cabana, A., and Lindahl, M. (2006) The diversity and complexity of the cyanobacterial thioredoxin systems. Photosynth. Res. 89:157-171. Flores, E., and Herrero, A. (2005) Nitrogen assimilation and nitrogen control in cyanobacteria. Biochem. Soc. Trans. 33(1):164-167. Frangeul, L., Quillardet, P., Castets, A.-M., Humbert, J.-F., Matthijs, H. C., Cortez, D., Tolonen, A., Zhang, C.-C., Gribaldo, S., Kehr, J.-C., Zilliges, Y., Ziemert, N., Becker, S., Talla, E., Latifi, A., Billault, A., Lepelletier, A., Dittmann, E., Bouchier, C., and Marsac, N. T. d. (2008) Highly plastic genome of Microcystis aeruginosa PCC 7806, a ubiquitous toxic freshwater cyanobacterium. BMC Genomics 9:274-295. Frochhammer, K. (2004) Global carbon/nitrogen control by PII signal transduction in cyanobacteria: from signals to targets. FEMS Microbiol. Rev.28:319-333 Fujii, M., Rose, A. L., Omura, T., and Waite, T. D. (2010) Effect of Fe(II) and Fe(III) transformation kinetics on iron acquisition by a toxic strain of Microcystis aeruginosa. Environ. Sci. Technol. 44:1980-1986. Fulda, S., Huang, F., Nilsson, F., Hagemann, M., and Norling, B. (2000) Proteomics of Synechocystis sp. strain PCC 6803 Identification of periplasmic
201
proteins in cells grown at low and high salt concentrations. Eur. J. Biochem. 267:5900-5907. Gao, Y., Xiong, W., Li, X.-b., Gao, C.-F., Zhang, Y.-l., Li, H., and Wu, Q.-y. (2009) Identification of the proteomic changes in Synechocystis sp. PCC 6803 following prolonged UV-B irradiation. J. Exp. Bot. 60(4):1141-1154. Gatlin, C. L., Kleemann, G. R., Hays, L. G., Link, A. J., and Yates, J. R. (1998) Protein identification at the low femtomole level from silver-stained gels using a new fritless electrospray interface for liquid chromatography-microspray and nanospray mass spectrometry. Anal. Biochem. 263(1):93-101. Gerbersdorf, S. U. (2006) An advanced technique for immuno-labelling of microcystins in cryosectioned cells of Microcystis aeruginosa PCC 7806 (cyanobacteria): Implementations of an experiment with varying light scenarios and culture densities. Toxicon 47:218-228. Ghassemian, M., and Strauss, N. A. (1996) Fur regulates the expression of iron-stress genes in the cyanobacterium Synechococcus sp. strain PCC 7942. Microbiology 142:1469-1476. Ginn, H. P., and Neilan, B. A. (2010) NtcA from Microcystis aeruginosa PCC 7806 is autoregulatory and binds to the microcystin promoter. Appl. Environ. Microbiol. accepted. God, M., Sauer, J., Baier, T., and Forchhammer, K. (1998) Nitrogen- starvation-induced chlorosis in Synechococcus PCC 7942 : adaptation to long- term survival. Microbiology 144:2449-2458. Gonzales, A. D., Light, Y. K., Zhang, Z., Iqbal, T., Lane, T. W., and Martino, A. (2005) Proteomic analysis of the CO2-concentrating mechanism in the open-ocean cyanobacterium Synechococcus WH8102. Can. J. Bot. 83:735- 745. Grossman, A. R., Bhaya, D., and He, Q. (2001) Tracking the light environment by cyanobacteria and the dynamic nature of light harvesting. J. Biol. Chem. 276(15):11449-11452. Guskov, A., Kern, J., Gabdulkhakov, A., Broser, M., Zouni, A., and Saenger, W. (2009) Cyanobacterial photosystem II at 2.9-A° resolution and the role of quinones, lipids, channels and chloride Nat. Struct. Mol. Biol. 16(3):334- 342. Hattrup, E., Neilson, K. A., Breci, L., and Haynes, P. A. ( 2007) Proteomic Analysis of Shade-Avoidance Response in Tomato Leaves. J. Agric. Food Chem. 55:8310-8318. Haynes, P. and Roberts, T. H. (2007) Subcellular shotgun proteomics in plants: Looking beyond the usual suspects. Proteomics 7: 2963-2975. Herbert, B. R., Grinyer, J., McCarthy, J. T., Isaacs, M., Harry, E. J., Nevalainen, H., Traini, M. D., Hunt, S., Schulz, B., Laver, M., Goodall, A. R., Packer, J., Harry, J. L., and Williams, K. L. (2006) Improved 2-DE of microorganisms after acidic extraction. Electrophoresis 27:1630-1640. Herna´ndez, J. A., Muro-Pastor, A. M., Flores, E., Bes, M. T., Peleato, M. L., and Fillat, M. a. F. (2006) Identification of a furA cis Antisense RNA in the Cyanobacterium Anabaena sp. PCC 7120. J. Mol. Biol. 355:325-334. Hernandez, J. A., Alonso, I., Pellicer, S., Peleato, M. L., Cases, R., Strasser, R., Barja, F., and Fillat, M. F. (2010) Mutants of Anabaena sp. PCC7120 lacking alr1690 and anti-furA antisense RNA show a pleiotropic phenotype and altered photosynthetic machinery. J. Plant Physiol. 167(6):430-7.
202
Hernandez, J. A., Lopez-Gomollon, S., Bes, M. T., Fillat, M. F., and Peleato, M. L. (2004) Three fur homologues from Anabaena sp. PCC7120: exploring reciprocal protein-promoter recognition. FEMS Microbiol. Lett. 236:275-282. Hernandez, J. A., Muro-Pastor, A. M., Flores, E., Bes, M. T., Peleato, M. L., and Fillat, M. F. (2006) Identification of a furA cis antisense RNA in the cyanobacterium Anabaena sp. PCC 7120. J. Mol. Biol. 355:325-334. Herranen, M., Battchikova, N., Zhang, P., Graf, A., Sirpio, S., Paakkarinen, V., and Aro, E.-M. (2004) Towards functional proteomics of membrane protein complexes in Synechocystis sp. PCC 6803 Plant Physiol. 134:470-481. Herrero, A., Muro-Pastor, A. M., and Flores, E. (2001) Nitrogen Control in Cyanobacteria. Journal of Bacteriology 183(2):411-425. Hihara, Y., Sonoike, K., Kanehisa, M., and Ikeuchi, M. (2003) DNA microarray analysis of redox-responsive genes in the genome of the cyanobacterium Synechocystis sp. strain PCC 6803. J. Bacteriol. 185(5):1719- 1725. Hisbergues, M., JeanJean, R., Joset, F., Tandeau de Marsac, N. and Bedu, S. (1999) Protein PII regulates both inorganic carbon and nitrate uptake and is modified by redox signal in Synechocystis PCC6803. FEBS Letters 463: 216- 220. Huang, F., Parmryd, I., Nilsson, F., Persson, A. L., Pakrasi, H. B., Andersson, B., and Norling, B. (2002) Proteomics of Synechocystis 6803, Identification of plasma membrane proteins. Mol. Cell. Proteomics 1(12):956- 966. Humble, A. V., Gadd, G. M., and Codd, G. A. (1997) Binding of copper and zinc to three cyanobacterial microcystins quantified by differential pulse polarography. Water Res. 31(7):1679-1686. Imai, A., Fukushima, T., and Matsushige, K. (1999) Effects of iron limitation and aquatic humic substances on the growth of Microcystis aeruginosa. Can. J. Fish. Aquat. Sci. 56:1929-1937. Ishii, A., and Hihara, Y. (2008) An AbrB-like transcriptional regulator, sll0822, is essential for the activation of nitrogen-regulated genes in Synechocystis sp. PCC 6803. Plant Physiol. 148:660-670. Ivanov, A. G., Krol, M., Sveshnikov, D., Selstam, E., Sandström, S., Koochek, M., Park, Y.-I., Vasilev, S., Bruce, D., Öquist, G., and Huner, N. P. A. (2006) Iron deficiency in cyanobacteria causes monomerization of photosystem I trimers and reduces the capacity for state transitions and the effective absorption cross section of PSI in vivo. Plant Physiol. 141(4):1436-45. Ivleva, N., and Golden, S. 2007. Protein extraction, fractioniation, and purification from cyanobacteria, p. 365-373. In E. Rosato (ed.), Methods in Molecular Biology - Circadian Rythms: Methods and Protocols, vol. 362. Humana Press Inc, Tototwa, NJ. Jackson, A. R., McInnes, A., Falconer, I. R., and Runnegar, M. T. (1984) Clinical and pathological changes in sheep experimentally poisoned by the blue- green alga Microcystis aeruginosa. Vet. Pathol. 21:102-113. Jeanjean, R., Talla, E., Latifi, A., Havaux, M., Janicki, A., and Zhang, C.-C. (2008) A large gene cluster encoding peptide synthetases and polyketide synthases is involved in production of siderophores and oxidative stress response in the cyanobacterium Anabaena sp. strain PCC 7120. Environ. Microbiol. 10(10):2574-2585. Jiang, F., Mannervik, B., and Bergman, B. (1997) Evidence for redox regulation of the transcription factor NtcA, acting both as an activator and a
203
repressor, in the cyanobacterium Anabaena PCC 7120. Biochem. J. 327:513- 517. Jonasson, S., Vintila, S., Sivonen, K., and El-Shehawy, R. (2008) Expression of the nodularin synthetase genes in the Baltic Sea bloom-former cyanobacterium Nodularia spumigena strain AV1. FEMS Microbiol. Ecol. 65:31-39. Juettner, F., and Luethi, H. (2008) Topology and enhanced toxicity of bound microcystins in Microcystis PCC 7806. Toxicon 51:388-397. Jungblut, A. D., Hawes, I., Mountfort, D., Dietrich, D. R., Burns, B. P., and Neilan, B. A. (2005) Diversity within cyanobacterial mat communities in variable salinity meltwater ponds of McMurdo ice shelf, Antarctica. Enivron. Microbiol. 7(4):519-529. Jungblut, A. D., Hoeger, S. J., Mountfort, D., Hitzfeld, B. C., Dietrich, D. R., and Neilan, B. A. (2006) Characterization of microcystin production in an Antarctic cyanobacterial mat community. Toxicon 47(3):271-8. Kaasalainen, U., Jokela, J., Fewer, D. P., Sivonen, K., and Rikkinen, J. (2009) Microcystin production in the tripartite cyanolichen Peltigera leucophlebia. Mol. Plant Microbe Interact. 22(6):695-702. Kadner, R. J. (2005) Regulation by iron: RNA rules the rust. J. Bacteriol. 187(20):6870-6873. Kaebernick, M., and Neilan, B. A. (2001) Ecological and molecular investigations of cyanotoxin production. FEMS Microbiol. Ecol. 35:1-9. Kaebernick, M., Neilan, B. A., Borner, T., and Dittmann, E. (2000) Light and the transcriptional response of the microcystin biosynthesis gene cluster. Appl. Environ. Microbiol. 66(8):3387-3392. Kaebernick, M., Rohrlack, T., Christoffersen, K., and Neilan, B. A. (2001) A spontaneous mutant of microcystin biosynthesis: genetic characterisation and effect on Daphnia. Environ. Microbiol. 3(11):669-679. Kameyama, K., Sugiura, N., Inamori, Y., and Maekawa, T. (2004) Characteristics of microcystin production in the cell cycle of Microcystis viridis. Environ. Toxicol. 19:20-25. Kan, J., Hanson, T. E., Ginter, J. M., Wang, K., and Chen, F. (2005) Metaproteomic analysis of Chesapeake Bay microbial communities. Saline Systems 1(7):doi:10.1186/1746-1448-1-7. Kaneko, T., Nakajima, N., Okamoto, S., Suzuki, I., Yuuhiko Tanabe, Tamaoki, M., Nakamura, Y., Kasai, F., Watanabe, A., Kawashima, K., Kishida, Y., Ono, A., Shimizu, Y., Takahashi, C., Minami, C., Fujishiro, T., Kohara, M., Katoh, M., Nakazaki, N., Nakayama, S., Yamada, M., Tabata, S., and Watanabe, M. (2007) Complete genomic structure of the bloom- forming toxic cyanobacterium Microcystis aeruginosa NIES-843. DNA Res. 14:247-256. Katoh, H., Hagino, N., Grossman, A. R., and Ogawa, T. (2001) Genes essential to iron transport in the cyanobacterium Synechocystis sp. strain PCC 6803. J. Bacteriol. 183(9):2779-2784. Kehr, J.-C., Zilliges, Y., Springer, A., Disney, M. D., Ratner, D. D., Bouchier, C., Seeberger, P. H., Marsac, N. T. d., and Dittmann, E. (2006) A mannan binding lectin is involved in cell–cell attachment in a toxic strain of Microcystis aeruginosa. Mol. Microbiol. 59(3):893-906. Keller, A., Nesvizhskii, A. I., Kolker, E., and Aebersold, R. (2002) Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Annal. Chem. 74(20):5383-92.
204
Keren, N., Aurora, R., and Pakrasi, H. B. (2004) Critical roles of bacterioferritins in iron storage and proliferation of Cyanobacteria. Plant Physiol. 135:1666-1673. Keren, N., Kidd, M. J., Penner-Hahn, J. E., and Pakrasi, H. B. (2002) A light-dependent mechanism for massive accumulation of manganese in the photosynthetic bacterium Synechocystis sp. PCC 6803. Biochemistry 41:15085- 15092. Kerfeld, C. A., Sawaya, M. R., Tanaka, S., Nguyen, C. V., Phillips, M., Beeby, M., and Yeates, T. O. (2005) Protein structures forming the shell of primitive bacterial organelles Science 309:936-938. Kobayashi, M., Ishizuka, T., Katayama, M., Kanehisa, M., Bhattacharyya- Pakrasi, M., Pakrasi, H. B., and Ikeuchi, M. (2004) Response to oxidative stress involves a novel peroxiredoxin gene in the unicellular cyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiol. 45(3):290-299 Koropatkin, N., Randich, A. M., Bhattacharyya-Pakrasi, M., Pakrasi, H. B., and Smith, T. J. (2007) The structure of the iron-binding protein, FutA1, from Synechocystis 6803. J. Biol. Chem. 282(37):27468-27477. Koropatkin, N. M., Pakrasi, H. B., and Smith, T. J. (2006) Atomic structure of a nitrate-binding protein crucial for photosynthetic productivity. Proc. Natl. Acad. Sci. U.S.A. 103(26):9820-9825. Kosakowska, A., Nedzi, M., and Pempkowiak, J. (2007) Responses of the toxic cyanobacterium Microcystis aeruginosa to iron and humic substances. Plant Physiol. Biochem. 45(5):365-270. Krewulak, K. D., and Vogel, H. J. (2008) Structural biology of bacterial iron uptake. Biochim. Biophys. Acta 1778:1781-1804. Kurian, D., Phadwal, K., and Menp, P. (2006) Proteomic characterization of acid stress response in Synechocystis sp. PCC 6803. Proteomics 6: 3614- 3624. Kurmayer, R., Christiansen, G., and Chorus, I. (2003) The abundance of microcystin-producing genotypes correlates positively with colony size in Microcystis sp. and determines its microcystin net production in Lake Wannsee. Appl. Environ. Microbiol. 69(2):787-795. Kurmayer, R., Christiansen, G., Fastner, J., and Börner, T. (2004) Abundance of active and inactive microcystin genotypes in populations of the toxic cyanobacterium Planktothrix spp. Environ. Microbiol. 6(8):831-841. Lahmi, R., Sendersky, E., Perelman, A., Hagemann, M., Forchhammer, K., and Schwarz, R. (2006) Alanine dehydrogenase activity is required for adequate progression of phycobilisome degradation during nitrogen starvation in Synechococcus elongatus PCC 7942. J. Bacteriol. 188(14):5258-5265. Latifi, A., Jeanjean, R., Lemeille, S., Havaux, M., and Zhang, C.-C. (2005) Iron starvation leads to oxidative stress in Anabaena sp. strain PCC 7120. J. Bacteriol. 187(18):6596-6598. Latifi, A., Ruiz, M., and Zhang, C.-C. (2008) Oxidative stress in cyanobacteria. FEMS Microbiol. Rev. 33(2):258-78. Lawton, L. A., Edwards, C., and Codd, G. A. (1994) Extraction and high- performance liquid chromatographic method for the determination of microcystins in raw and treated waters. Analyst 119:1525-1530. Lax, J. E.-M., Arteni, A. A., Boekema, E. J., Pistorius, E. K., Michel, K.-P., and Rögner, M. (2007) Structural response of Photosystem 2 to iron deficiency: Characterization of a new Photosystem 2–IdiA complex from the
205
cyanobacterium Thermosynechococcus elongatus BP-1. Biochim. Biophys. Acta 1767:528-534. Lazazzera, B. A. (2000) Quorum sensing and starvation: signals for entry into stationary phase. Curr. Opin. Microbiol. 3:177-182. Lee, J.-W., and Helmann, J. D. (2007) Functional specialization within the Fur family of metalloregulators. Biometals 20:485-499. Lee, S. J., Jang, M.-H., Kim, H.-S., Yoon, B.-D., and Oh, H.-M. (2000) Variation of microcystin content of Microcystis aeruginosa relative to medium N:P ratio and growth stage. J. Appl. Microbiol. 89:323-329. Li, H., and Sherman, L. A. (2002) Characterization of Synechocystis sp. strain PCC 6803 and nbl mutants under nitrogen-deficient conditions. Arch. Microbiol. 178: 256-266. Li, H., Xie, P., Zhang, D., and Chen, J. (2008) The first study on the effects of microcystin-RR on gene expression profiles of antioxidant enzymes and heat shock protein-70 in Synechocystis sp. PCC6803. Toxicon 53(6):595-601. Lindahl, M., and Kieselbach, T. (2009) Disulphide proteomes and interactions with thioredoxin on the track towards understanding redox regulation in chloroplasts and cyanobacteria. J. Proteomics 72(3):416-438. Long, B. M. (2010) Evidence that sulfur metabolism plays a role in microcystin production by Microcystis aeruginosa. Harmful Algae 9:74-81. Long, B. M., Jones, G. J., and Orr, P. T. (2001) Cellular microcystin content in N-limited Microcystis aeruginosa can be predicted from growth rate. Appl. Environ. Microbiol. 67(1):278-283. López-Gomollón, S., Hernández, J. A., Pellicer, S., Angarica, V. E., Peleato, M. L., and Fillat, M. F. (2007a) Cross-talk between iron and nitrogen regulatory networks in Anabaena (Nostoc) sp. PCC 7120: Identification of overlapping genes in FurA and NtcA regulons. J. Mol. Biol. 374:267-281. Lopez-Gomollon, S., Hernandez, J. A., Wolk, C. P., Peleato, M. L., and Fillat, M. F. (2007b) Expression of furA is modulated by NtcA and strongly enhanced in heterocysts of Anabaena sp. PCC 7120. Microbiol. 153(Pt 1):42- 50. Lopez-Gomollon, S., Sevilla, E., Bes, M. T., Peleato, M. L., and Fillat, M. F. (2009) New insights into the role of Fur proteins: FurB (All2473) from Anabaena protects DNA and increases cell survival under oxidative stress. Biochem. J. 418:201-207. Lucarelli, D., Russo, S., Garman, E., Milano, A., Meyer-Klaucke, W., and Pohl, E. (2007) Crystal structure and function of the zinc uptake regulator FurB from Mycobacterium tuberculosis. J. Biol. Chem. 282(13):9914-9922. Lv, J., Li, N., and Niu, D.-K. (2008) Association between the availability of environmental resources and the atomic composition of organismal proteomes: Evidence from Prochlorococcus strains living at different depths. Biochem. Biophys. Res. Commun. 375:241-246. Martin-Luna, B., Hernandez, J. A., Bes, M. T., Fillat, M. F., and Peleato, M. L. (2006a) Identification of a ferric uptake regulator from Microcystis aeruginosa PCC7806. FEMS Microbiol. Lett. 254:63-70. Martin-Luna, B., Sevilla, E., Hernandez, J. A., Bes, M. T., Fillat, M. F., and Peleato, M. L. (2006b) Fur from Microcystis aeruginosa binds in vitro promoter regions of the microcystin biosynthesis gene cluster. Phytochemistry 67:876- 881. Martins, J., Saker, M. L., Moreira, C., Welker, M., Fastner, J., and Vasconcelos, V. M. (2009) Peptide diversity in strains of the cyanobacterium
206
Microcystis aeruginosa isolated from Portuguese water supplies. Appl. Microbiol. Biotechnol. 82(5):951-61. Maruyama, T., Kato, K., Yokoyama, A., Tanaka, T., Hiraishi, A., and Park, H.-D. (2003) Dynamics of microcystin-degrading bacteria in mucilage of Microcystis. Microb. Ecol. 46:279-288. Mata-Cabana, A., Florencio, F. J., and Lindahl, M. (2007) Membrane proteins from the cyanobacterium Synechocystis sp. PCC 6803 interacting with thioredoxin. Proteomics 7:3953–3963. Meissner, K., Dittmann, E. and Börner, T. (1996) Toxic and non-toxic strains of the cyanobacterium Microcystis aeruginosa contain sequences homologous to peptide synthetase genes. FEMS Microbiol. Lett. 135: 295-303. Michel, K.-P., and Pistorius, E. K. (2004) Adaptation of the photosynthetic electron transport chain in cyanobacteria to iron deficiency: The function of IdiA and IsiA. Physiol. Plant. 120:36-50. Mihali, T. K., Kellmann, R., Muenchhoff, J., Barrow, K. D., and Neilan, B. A. (2008) Characterisation of the gene cluster responsible for cylindrospermopsin biosynthesis. Appl. Environ. Microbiol. 74(3):716-22. Mihali, T. K., Kellmann, R., and Neilan, B. A. (2009) Characterisation of the paralytic shellfish toxin biosynthesis gene clusters in Anabaena circinalis AWQC121C and Aphanizomenon sp. NH-5. BMC Biochem. 30(10):8. Mikalsen, B., Boison, G., Skulberg, O. M., Fastner, J., Davies, W., Gabrielsen, T. M., Rudi, K., and Jakobsen, K. S. (2003) Natural variation in the microcystin synthetase operon mcyABC and impact on microcystin production in Microcystis strains. J. Bacteriol. 185(9):2774-2785. Mlouka, A., Comte, K., Castets, A. M., Bouchier, C., and Marsac, N. T. d. (2004) The gas vesicle gene cluster from Microcystis aeruginosa and DNA rearrangements that lead to the loss of cell buoyancy J. Bacteriol. 186(8):2355- 2365. Moffitt, M. C. (2003). Non-ribosomal biosynthesis of the cyanobacterial toxin nodularin. University of New South Wales, Sydney. Moffitt, M. C., and Neilan, B. A. (2004) Characterisation of the nodularin synthetase gene cluster and proposed theory of the evolution of cyanobacterial hepatotoxins. Appl. Environ. Microbiol. 70(11):6353-62. Muro-Pastor, M. I., Reyes, J. C., and Florencio, F. J. (2005) Ammonium assimilation in cyanobacteria. Photosyn. Res. 83:135-150. Nagai, T., Imai, A., Matsushige, K., and Fukushima, T. (2007) Growth characteristics and growth modeling of Microcystis aeruginosa and Planktothrix agardhii under iron limitation. Limnology 8:261-270. Nakasugi, K., Alexova, R., Svenson, C. J., and Neilan, B. A. (2007) Functional analysis of PilT from the toxic cyanobacterium Microcystis aeruginosa PCC 7806. J. Bacteriol. 189(5):1689-1697. Navarro, F., Martın-Figueroa, E., and Florencio, F. J. (2000) Electron transport controls transcription of the thioredoxin gene (trxA) in the cyanobacterium Synechocystis sp. PCC 6803. Plant Mol. Biol. 43:23-32. Nesvizhskii, A. I., Keller, A., Kolker, E., and Aebersold, R. (2003) A statistical method for identifying proteins by mass tandem spectrometry. Anal. Chem. 75(17):4646-58. Nield, J., Morris, E. P., Bibby, T. S., and Barber, J. (2003) Structural Analysis of the Photosystem I Supercomplex of Cyanobacteria Induced by Iron Deficiency Biochemistry 42:3180-3188.
207
Nodop, A., Pietsch, D., Hocker, R., Becker, A., Pistorius, E. K., Forchhammer, K., and Michel, K.-P. (2008) Transcript profiling reveals new insights into the acclimation of the mesophilic fresh-water cyanobacterium Synechococcus elongatus PCC 7942 to iron starvation. Plant Physiol. 147:747- 763. Ogawa, T., and Mi, H. (2007) Cyanobacterial NADPH dehydrogenase complexes. Photosynth. Res. 93:69-77. Oh, H.-M., Lee, S. J., Jang, M.-H., and Yoon, B.-D. (2000) Microcystin production by Microcystis aeruginosa in a phosphorus-limited chemostat. Appl. Environ. Microbiol. 66(1):176-179. Oksanen, I., Jokela, J., Fewer, D. P., Wahlsten, M., Rikkinen, J., and Sivonen, K. (2004) Discovery of rare and highly toxic microcystins from lichen-associated cyanobacterium Nostoc sp. strain IO-102-I. Appl. Environ. Microbiol. 70(10):5756-63. Olmedo-Verd, E., Valladares, A., Flores, E., Herrero, A., and Muro-Pastor, A. M. (2008) Role of two NtcA-binding sites in the complex ntcA gene promoter of the heterocyst-forming cyanobacterium Anabaena sp. strain PCC 7120. J. Bacteriol. 190(22):7584-7590. Orús, M. I., Rodríguez-Buey, M. L., Marco, E., and Fernández-Valiente, E. (2001) Changes in carboxysome structure and grouping and in photosynthetic affinity for inorganic carbon in Anabaena strain PCC 7119 (Cyanophyta) in + response to modification of CO2 and Na supply. Plant Cell Physiol. 42(1):46- 53. Osanai, T., Azuma, M., and Tanaka, K. (2007) Sugar catabolism regulated by light- and nitrogen-status in the cyanobacterium Synechocystis sp. PCC 6803. Photochem.Photobiol. Sci. 6:508-514. Osanai, T., and Tanaka, K. (2007) Keeping in touch with PII: PII-interacting proteins in unicellular cyanobacteria. Plant Cell Physiol. 48(7):908-914. Ostermaier, V., and Kurmayer, R. (2009) Distribution and abundance of nontoxic mutants of cyanobacteria in lakes of the Alps. Microb. Ecol. 58(2):323-33. Otsuka, S., Suda, S., Shibata, S., Oyaizu, H., Matsumoto, S., and Watanabe, M. M. (2001) A proposal for the unification of five species of the cyanobacterial genus Microcystis Kutzing ex Lemmermann 1907 under the rules of the Bacteriological Code. Int. J. Syst. Evol. Microbiol. 51(Pt 3):873-879. Ow, S. Y., Cardona, T., Taton, A., Magnuson, A., Lindblad, P., Stensjö, K., and Wright, P. C. (2008) Quantitative shotgun proteomics of enriched heterocysts from Nostoc sp. PCC7120 using 8-plex isobaric peptide tags. J. Proteome Res. 7(4):1615-1628. Ow, S. Y., Noirel, J., Cardona, T., Taton, A., Lindblad, P., Stensjö, K., and Wright, P. C. (2009) Quantiative overview of N2 fixation in Nostoc punctiforme ATCC 29133 through cellular enrichments and iTRAQ shotgun proteomics. . J. Proteome Res. 8(1):187-98. Paerl, H. W., and Huisman, J. (2008) Blooms like it hot. Science 320:57-58. Pandhal, J., Wright, P. C., and Biggs, C. A. (2007) A quantitative proteomic analysis of light adaptation in a globally significant marine cyanobacterium Prochlorococcus marinus MED4. J. Proteome Res. 6:996-1005. Parker, D., Kennan, R. M., Myers, G. S., Paulsen, I. T., and Rood, J. I. (2005) Identification of a Dichelobacter nodosus ferric uptake regulator and determination of its regulatory targets. J. Bacteriol. 187(1):366-375.
208
Pathak, D., and Ollis, D. (1990) Refined structure of dienelactone hydrolase at 1.8 A. J. Mol. Biol. 214(2):497-525. Paz-Yepes, J., Flores, E., and Herrero, A. (2003) Transcriptional effects of the signal transduction protein PII (glnB gene product) on NtcA-dependent genes in Synechococcus sp. PCC 7942. FEBS Lett. 543:42-46. Pearson, L. A., Hisbergues, M., Borner, T., Dittmann, E., and Neilan, B. A. (2004) Inactivation of an ABC transporter gene, mcyH, results in loss of microcystin production in the cyanobacterium Microcystis aeruginosa PCC 7806. Appl. Environ. Microbiol. 70(11):6370-6378. Pearson, L. A., Moffitt, M. C., Ginn, H. P., and Neilan, B. A. (2008) The molecular genetics and regulation of cyanobacterial peptide hepatotoxin biosynthesis. Crit. Rev. Toxicol. 38:847-856. Peng, J., Elias, J. E., Thoreen, C. C., Licklider, L. J., and Gygi, S. P. (2003) Evaluation of multidimensional chromatography coupled with tandem mass spectrometry (LC/LC-MS/MS) for large-scale protein analysis: The yeast proteome. J. Proteome Res. 2:43-50. Perez-Perez, M. E., Florencio, F. J., and Lindahl, M. (2006) Selecting thioredoxins for disulphide proteomics: Target proteomes of three thioredoxins from the cyanobacterium Synechocystis sp. PCC 6803. Proteomics 6:S186– S195. Perez-Perez, M. E., Martın-Figueroa, E., and Florencio, F. J. (2009) Photosynthetic regulation of the cyanobacterium Synechocystis sp. PCC 6803 thioredoxin system and functional analysis of TrxB (Trx x) and TrxQ (Trx y) thioredoxins. Mol. Plant 2(2):270-283. Pfaffl, M. W. (2001) A new mathematical model for relative quantification in real time RT-PCR. Nucleic Acids Res. 29(900):2002-2007. Pflugmacher, S. (2002) Possible allelopathic effects of cyanotoxins, with reference to microcystin-LR, in aquatic ecosystems. Environ. Toxicol. 17:407- 413. Phelan, R. R., and Downing, T. G. (2007) Optimisation of laboratory scale production and purification of microcystin-LR from pure cultures of Microcystis aeruginosa. Afr. J. Biotechnol. 6(21):2451-2457. Pierre-Alain, M., Christophe, M., Severine, S., Houria, A., Philippe, L., and Lionel, R. (2007) Protein extraction and fingerprinting optimization of bacterial communities in natural environment. Microb. Ecol. 53:426-434 Pieulle, L., Guedeney, G., Cassier-Chauvat, C., Jeanjean, R., Chauvat, F., and Peltier, G. (2000) The gene encoding the NdhH subunit of type 1 NAD(P)H dehydrogenase is essential to survival of Synechocystis PCC6803. FEBS Lett. 487:272-276. Plominsky, A. M., Soto-Liebe, K., and Vasquez, M. (2009) Optimization of 2D-PAGE protocols for proteomic analysis of two nonaxenic toxin-producing freshwater cyanobacteria: Cylindrospermopsis raciborskii and Raphidiopsis sp. Lett. Appl. Microbiol. 49(3):332-337. Prommeenate, P., Lennon, A. M., Markert, C., Hippler, M., and Nixon, P. J. (2004) Subunit composition of NDH-1 complexes of Synechocystis sp. PCC 6803 Identification of two new ndh gene products with nuclear-encoded homologues in the chloroplast Ndh complex. J. Biol. Chem. 279(27):28163- 28173. Rantala, A., Fewer, D. P., Hisbergues, M., Rouhiainen, L., Vaitomaa, J., rner, T. B., and Sivonen, K. (2004) Phylogenetic evidence for the early
209
evolution of microcystin synthesis. Proc. Natl. Acad. Sci. U.S.A. 101(2):568- 573. Rapala, J., Sivonen, K., Lyra, C., and Niemela, S. I. (1997) Variation of microcystins, cyanobacterial hepatotoxins, in Anabaena spp. as a function of growth stimuli. Appl. Environ. Microbiol. 63():2206-2212. Rippka, R., and Herdman, M. (1992). Pastuer culture collection (PCC) of cyanobacterial strains in axenic culture, vol. 1. Institut Pasteur, Paris. Roberts, A. A. (2008). Unnatural production of natural products: Heterologous expression and combinatorial biosynthesis of cyanobacterial-derived compounds. PhD Thesis. The University of New South Wales. Roberts, A. A., and Neilan, B. A. (2010) Evolution and transcriptional regulation of the microcystin synthetase gene clusters in Microcystis spp.: The role of transposases and UV-C light. in preparation. Ross, C., Santiago-Vazquez, L., and Paul, V. (2006) Toxin release in response to oxidative stress and programmed cell death in the cyanobacterium Microcystis aeruginosa. Aquat. Toxicol. 78:66–73. Rueckert, A., and Cary, S. C. (2009) Use of an armored RNA standard to measure microcystin synthetase E gene expression in toxic Microcystis sp. by reverse transcription QPCR. Limnol. Oceanogr.: Methods 7:509-520. Saha, S. K., Uma, L., and Subramanian, G. (2003) Nitrogen stress induced changes in the marine cyanobacterium Oscillatoria willei BDU 130511. FEMS Microbiol. Ecol. 45:263-272. Saito, K., Sei, Y., Miki, S., and Yamaguchi, K. (2008) Detection of microcystin- metal complexes by using cryospray ionization-Fourier transform ion cyclotron resonance mass spectrometry. Toxicon 51:1496-1498. Sakai, H., Oguma, K., Katayama, H., and Ohgaki, S. (2007) Effects of low or medium-pressure UV irradiation on the release of intracellular microcystin. Water Res. 41:3458-3464. Saker, M. L., Jungblut, A.-D., Neilan, B. A., Rawn, D. F. K., and Vasconcelos, V. M. (2005) Detection of microcystin synthetase genes in health food supplements containing the freshwater cyanobacterium Aphanizomenon flos-aquae. Toxicon 46(5):555-562. Sakthivel, K., Watanabe, T., and Nakamoto, H. (2009) A small heat-shock protein confers stress tolerance and stabilizes thylakoid membrane proteins in cyanobacteria under oxidative stress. Arch. Microbiol. 191(4):319-28. Sarcina, M., and Mullineaux, C. W. (2004) Mobility of the IsiA chlorophyll- binding protein in cyanobacterial thylakoid membranes. J. Biol. Chem. 279(35):36514-36518. Sauer, J. r., Schreiber, U., Schmid, R., lker, U. V., and Forchhammer, K. (2001) Nitrogen Starvation-Induced Chlorosis in Synechococcus PCC 7942. Low-Level Photosynthesis As a Mechanism of Long-Term Survival. Plant Physiology 126:233-243. Sazuka, T., and Ohara, O. (1997) Towards a proteome project of cyanobacterium Synechocystis sp. strain PCC6803: linking 130 protein spots with their respective genes. . Electrophoresis 18(8):1252-1258. Schatz, D., Keren, Y., Hadas, O., Carmeli, S., Sukenik, A., and Kaplan, A. (2005) Ecological implications of the emergence of non-toxic subcultures from toxic Microcystis strains. Environ. Microbiol. 7(6):798-805. Schatz, D., Keren, Y., Vardi, A., Sukenik, A., Carmeli, S., Börner, T., Dittmann, E., and Kaplan, A. (2007) Towards clarification of the biological
210
role of microcystins, a family of cyanobacterial toxins. Environ. Microbiol. 9(4):965-970. Schuermann, P., and Buchanann, B. B. (2008) The ferredoxin/thioredoxin system of oxygenic photosynthesis. Antioxid. Redox Signal. 10(7):1235-1273. Schwabe, W., Weihe, A., Boerner, T., Henning, M., and Kohl, J.-G. (1988) Plasmids in toxic and nontoxic strains of the cyanobacterium Microcystis aeruginosa. Curr. Microbiol. 17:133-137. Schwarz, R., and Forchhammer, K. (2005) Acclimation of unicellular cyanobacteria to macronutrient deficiency: emergence of a complex network of cellular responses. Microbiol. 151:2503-2514. Sevilla, E., Martin-Luna, B., Vela, L., Bes, M. T., Fillat, M. F., and Peleato, M. L. (2008) Iron availability affects mcyD expression and microcystin-LR synthesis in Microcystis aeruginosa PCC7806. Environ. Microbiol. 10(10):2476-83. Shen, G., Antonkine, M. L., Est, A. v. d., Vassiliev, I. R., Brettel, K., Bittl, R., Zech, S. G., Zhao, J., Stehlik, D., Bryant, D. A., and Golbeck, J. H. (2002a) Assembly of Photosystem I II. Rubredoxin is required for the in vivo assembly of Fx in Synechococcus sp. PCC 7002 as shown by optical and EPR spectroscopy. J. Biol. Chem. 277(23):20355-20366. Shen, G., Zhao, J., Reimer, S. K., Antonkine, M. L., Cai, Q., Weiland, S. M., Golbeck, J. H., and Bryant, D. A. (2002b) Assembly of Photosystem I: I. Inactivation of the rubA gene encoding a membrane-associated rubredoxin in the cyanobacterium Synechococcus sp. PCC 7002 causes a loss of photosystem I activity. J. Biol. Chem. 277(23):20343-20354. Singh, A. K., Bhattacharyya-Pakrasil, M., Elvitigala, T., Ghosh, B., Aurora, R., and Pakrasi, H. B. (2009) A systems level analysis of the effects of light quality on the metabolism of a cyanobacterium. Plant Physiol. 151(3):1596-608. Singh, A. K., Elvitigala, T., Bhattacharyya-Pakrasi, M., Aurora, R., Ghosh, B., and Pakrasi, H. B. (2008) Integration of carbon and nitrogen metabolism with energy production is crucial to light acclimation in the cyanobacterium Synechocystis. Plant Physiol. 148:467–478. Singh, A. K., Li, H., Bono, L., and Sherman, L. A. (2005) Novel adaptive responses revealed by transcription profiling of a Synechocystis sp. PCC 6803 DisiA mutant in the presence and absence of hydrogen peroxide. Photosyn. Res. 84:65-70. Singh, A. K., and Sherman, L. A. (2006) Iron-independent dynamics of IsiA production during the transition to stationary phase in the cyanobacterium Synechocystis sp. PCC6803. FEMS Microbiol. Lett. 256:159-164. Soares, R. M., Yuan, M., Servaites, J. C., Delgado, A., Magalhaes, V. F., Hilborn, E. D., Carmichael, W. W., and Azevedo, S. M. F. O. (2006) Sublethal exposure from microcystins to renal insufficiency patients in Rio de Janeiro, Brazil. Environ. Toxicol. 21:95-103. Spassova, M., Mellor, I. R., Petrov, A. G., Beattie, K. A., Codd, G. A., Vais, H., and Usherwood, P. N. R. (1995) Pores formed in lipid bilayers and in native membranes by nodularin, a cyanobacterial toxin. Eur. Biophys. J. 24:69- 76. Srivastava, R., Pisareva, T., and Norling, B. (2005) Proteomic studies of the thylakoid membrane of Synechocystis sp. PCC 6803. Proteomics 5:4905-4916. Stensjo, K., Ow, S. Y., Barrios-Llerena, M. E., Lindblad, P., and Wright, P. C. (2007) An iTRAQ-based quantitative analysis to elaborate the proteomic
211
response of Nostoc sp. PCC 7120 under N2 fixing conditions. J. Proteome Res. 6:621-635. Su, Z., Olman, V., Mao, F., and Xu, Y. (2005) Comparative genomics analysis of NtcA regulons in cyanobacteria: regulation of nitrogen assimilation and its coupling to photosynthesis. Nucleic Acids Res. 33(16):5156-5171. Suzuki, I., Simon, W. J., and Slabas, A. R. (2006) The heat shock response of Synechocystis sp. PCC 6803 analysed by transcriptomics and proteomics. J. Exp. Bot. 57(7):1573-1578. Takahashi, I., Hayano, D., Asayama, M., Masahiro, F., Watahiki, M., and Shirai, M. (1996) Restriction barrier composed of an extracellular nuclease and restriction nuclease in the unicellular cyanobacterium Microcystis. FEMS Microbiol. Lett. 145:107-111. Takatani, N., and Omata, T. (2006) Effects of PII deficiency on expression of the genes involved in ammonium utilization in the cyanobacterium Synechocystis sp. strain PCC 6803. Plant Cell Physiol. 47(6):679-688. Tamoi, M., Miyazaki, T., Fukamizo, T., and Shigeoka, S. (2005) The Calvin cycle in cyanobacteria is regulated by CP12 via the NAD(H)/NADP(H) ratio under light/dark conditions Plant J. 42: 504-513. Tanaka, S., Kerfeld, C. A., Sawaya, M. R., Cai, F., Heinhorst, S., Cannon, G. C., and Yeates, T. O. (2008) Atomic-level models of the bacterial carboxysome shell. Science 318:1083-1086. Tillett, D., Dittmann, E., Erhard, M., Dohren, H. v., Borner, T., and Neilan, B. A. (2000) Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide-polyketide synthetase system. Chem. Biol. 7:753-764. Tillett, D., Parker, D. L., and Neilan, B. A. (2001) Detection of toxigenicity by a probe for the microcystin synthetase a gene (mcyA) of the cyanobacterial genus Microcystis. Comparison of toxicities with 16S rRNA and phycocyanin operon (phycocyanin intergenic spacer) phylogenies. Appl. Environ. Microbiol. 67(6):2810-2818. Ting, C. S., Rocap, G., King, J., and Chisholm, S. W. (2002) Cyanobacterial photosynthesis in the oceans: the origins and significance of divergent light- harvesting strategies. Trends Microbiol. 10(3):134-142. Tolle, J., Michel, K.-P., Kruip, J., Kahmann, U., Preisfeld, A., and Pistorius, E. K. (2002) Localization and function of the IdiA homologue Slr1295 in the cyanobacterium Synechocystis sp. strain PCC 6803. Microbiology 148:3293- 3305. Tonk, L., Visser, P. M., Christiansen, G., Dittmann, E., Snelder, E. O. F. M., Wiedner, C., Mur, L. R., and Huisman, J. (2005) The microcystin composition of the cyanobacterium Planktothrix agardhii changes toward a more toxic variant with increasing light intensity. Appl. Environ. Microbiol. 71(2):629-635. Tooming-Klunderud, A., Mikalsen, B., Kristensen, T., and Jakobsen, K. S. (2008) The mosaic structure of the mcyABC operon in Microcystis. Microbiology 154:1886-1899. Trost, P., Fermani, S., Marri, L., Zaffagnini, M., Falini, G., Scagliarini, S., Pupillo, P., and Sparla, F. (2006) Thioredoxin-dependent regulation of photosynthetic glyceraldehyde-3-phosphate dehydrogenase: autonomous vs. CP12-dependent mechanisms. Photosynth. Res. 89:263-275.
212
Tucker, S., and Pollard, P. (2005) Identification of cyanophage Ma-LBP and infection of the cyanobacterium Microcystis aeruginosa from an Australian subtropical lake by the virus. Appl. Environ. Microbiol. 71(2):629-635. Utkilen, H., and Gjolme, N. (1995) Iron-stimulated toxin production in Microcystis aeruginosa. Appl. Environ. Microbiol. 61(2):797-800. Valdor, R., and Aboal, M. (2007) Effects of living cyanobacteria, cyanobacterial extracts and pure microcystins on growth and ultrastructure of microalgae and bacteria. Toxicon 49(6):769-79. Vela, L., Sevilla, E., Gonzalez, C., Bes, M. T., Fillat, M. F., and Peleato, M. L. (2008) Exploring the interaction of microcystin-LR with proteins and DNA. Toxicol. In Vitro 22:1714-1718. Vesterkvist, P. S. M., and Meriluoto, J. A. O. (2003) Interaction between microcystins of different hydrophobicities and lipid monolayers. Toxicon 41:349-355. Vezie, C., Rapala, J., Vaitomaa, J., Seitsonen, J., and Sivonen, K. (2002) Effect of nitrogen and phosphorus on growth of toxic and nontoxic Microcystis strains and on intracelular microcystin concentrations. Microb. Ecol. 43:443- 454. Via-Ordorika, L., Fastner, J., Kurmayer, R., Hisbergues, M., Dittmann, E., Komarek, J., Erhard, M., and Chorus, I. (2004) Distribution of microcystin- producing and non-microcystin-producing Microcystis sp. in European freshwater bodies: Detection of microcystins and microcystin genes in individual colonies. Syst. Appl. Microbiol. 27:592-602. Waal, D. B. V. d., Verspagen, J. M. H., rling, M. L., Donk, E. V., Visser, P. M., and Huisman, J. (2009) The ecological stoichiometry of toxins produced by harmful cyanobacteria: an experimental test of the carbon-nutrient balance hypothesis. Ecology Letters 12:1-10. Waldron, K. J., Tottey, S., Yanagisawa, S., Dennison, C., and Robinson, N. J. (2006) A Periplasmic Iron-binding Protein Contributes toward Inward Copper Supply. J. Biol. Chem. 282(6):3837-3846. Welker, M., Marsalek, B., Sejnohova, L., and Dohren, H. v. (2006) Detection and identification of oligopeptides in Microcystis (cyanobacteria) colonies: Toward an understanding of metabolic diversity. Peptides 27:2090-2103. Wiedner, C., Visser, P. M., Fastner, J., Metcalf, J. S., Codd, G. A., and Mur, L. R. (2003) Effects of light on the microcystin content of Microcystis strain PCC 7806. Appl. Environ. Microbiol. 69(3):1475-1481. Wiegand, C., and Pflugmacher, S. (2005) Ecotoxicological effects of selected cyanobacterial secondary metabolites a short review. Toxicol. Appl. Pharmacol. 203:201-218. Wilmes, P., and Bond, P. L. (2004) The application of two-dimensional polyacrylamide gel electrophoresis and downstream analyses to a mixed community of prokaryotic microorganisms. Environ. Microbiol. 6:911-920. Wilmes, P., and Bond, P. L. (2006) Metaproteomics: studying functional gene expression in microbial ecosystems. Trends Microbiol. 14(2):92-97. Wilson, A. E., Sarnelle, O., Neilan, B. A., Salmon, T. P., Gehringer, M. M., and Hay, M. E. (2005) Genetic variation of the bloom-forming cyanobacterium Microcystis aeruginosa within and among lakes: Implications for harmful algal blooms Appl. Environ. Microbiol. 71(10):6126–6133 Wisen, S., Bergman, B., and Mannervika, B. (2004) Mutagenesis of the cysteine residues in the transcription factor NtcA from Anabaena PCC 7120 and its effects on DNA binding in vitro. Biochim. Biophys. Acta 1679:156- 163.
213
Xia, X.-X., Han, M.-J., Lee, S. Y., and Yoo, J.-S. (2008) Comparison of the extracellular proteomes of Escherichia coli B and K-12 strains during high cell density cultivation. Proteomics 8:2089-2103. Xing, W., Huang, W.-m., Li, D.-h., and Liu, Y.-d. (2007) Effects of Iron on Growth, Pigment Content, Photosystem II Efficiency, and Siderophores Production of Microcystis aeruginosa and Microcystis wesenbergii. Curr. Microbiol. 55:94-98. Yoshida, M., Yoshida, T., Satomi, M., Takashima, Y., Hosoda, N., and Hiroishi, S. (2008) Intra-specific phenotypic and genotypic variation in toxic cyanobacterial Microcystis strains. J. Appl. Microbiol. 105(2):407-15. Young, F. M., Thomson, C., Metcalf, J. S., Lucocq, J. M., and Codd, G. A. (2005) Immunogold localisation of microcystins in cryosectioned cells of Microcystis. J. Struct. Biol. 151:208-214. Yousef, N., Pistorius, E. K., and Michel, K.-P. (2003) Comparative analysis of idiA and isiA transcription under iron starvation and oxidative stress in Synechococcus elongatus PCC 7942 wild-type and selected mutants. Arch. Microbiol. 180:471-483. Youshida, T., Maki, M., Okamoto, H., and Hiroshi, S. (2005) Coordination of DNA replication and cell division in cyanobacteria M. aeruginosa. FEMS Microbiol. Lett. 251:149-154. Zohary, T., and Breen, C. M. (1989) Environmental factors favouring the formation of Microcystis aeruginosa hyperscums in a hypertrophic lake. Hydrobiologia 178:179-192. Zouni, A., Witt, H.-T., Kern, J., Fromme, P., Krauss, N., Saenger, W., and Orth, P. (2001) Crystal structure of photosystem II from Synechococcus elongatus at 3.8A resolution. Nature 409:739-743. Zybailov, B., Mosley, A. L., Sardiu, M. E., Coleman, M. K., Florens, L., and Washburn, M. P. (2006) Statistical analysis of membrane proteome expression changes in Saccharomyces cerevisiae. J. Proteome Res. 5:2339-2347.
214
2