Thesis
Role of phototrophic sulfur bacteria from the chemocline in the primary production of Lake Cadagno
STORELLI, Nicola
Abstract
Phototrophic sulfur bacteria are important for primary production in many stratified lakes. In Lake Cadagno, these bacteria greatly contribute to the total primary production with high values of CO2 fixation both in the presence and absence of light. The small-celled PSB Candidatus “Thiodictyon syntrophicum” Cad16T was the strongest CO2 assimilator and used as model organism. The draft genome sequence of strain Cad16T revealed the presence of two RuBisCO genes (cbbL and cbbM), which were deferentially expressed. 2D-DIGE analysis showed the presence of 23 protein spots up-regulated in the light, and 17 in the dark. Among the 23 protein spots that were up-regulated in the light, three are involved in the storage mechanism that produces granules of poly(3-hydroxybutyrate) from an excess of reducing power and carbon compounds. Among the 17 protein spots up-regulated in the dark, three were found to be part of the autotrophic dicarboxylate-hydroxybutyrate (DC/HB) cycle.
Reference
STORELLI, Nicola. Role of phototrophic sulfur bacteria from the chemocline in the primary production of Lake Cadagno. Thèse de doctorat : Univ. Genève, 2014, no. Sc. 4646
URN : urn:nbn:ch:unige-349153 DOI : 10.13097/archive-ouverte/unige:34915
Available at: http://archive-ouverte.unige.ch/unige:34915
Disclaimer: layout of this document may differ from the published version.
1 / 1 UNIVERSITÉ DE GENÈVE FACULTÉ DES SCIENCES Département de botanique et biologie végétale Dr. Xavier Perret Unité de microbiologie Dr. Mauro Tonolla
Role of Phototrophic Sulfur Bacteria from the Chemocline in the Primary Production of Lake Cadagno
THÈSE présentée à la Faculté des sciences de l’Université de Genève pour obtenir le grade de Docteur ès sciences, mention biologie
par
Nicola Storelli de Losone (TI)
Thèse n° 4646
GENÈVE Atelier de reprographie ReproMail 2014
ii
PUBLISHED PAPERS
†Peduzzi S., †Storelli N., Welsh A., Peduzzi R., Hahn D., Perret X., Tonolla M. (2012) Candidatus "Thiodictyon syntrophicum", sp. nov., a new purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp. Systematic and Applied Microbioogyl 35, 139-144. (DOI: 10.1016/j.syapm.2012.01.001). †: equally contributed.
Storelli N., Peduzzi S., Saad M., Frigaard N-U., Perret X., Tonolla M. (2013) CO2 assimilation in the chemocline of Lake Cadagno is dominated by a few types of phototrophic purple sulfur bacteria. FEMS Microbiology Ecology 84(2), 421-432. (DOI: 10.1111/1574-6941.12074).
Storelli N., Peduzzi S., Saad M., Frigaard N-U., Perret X., Tonolla M. (2014). Autotrophic carbon dioxide assimilation mechanism in the dark disclosed by proteome analysis of the purple sulfur bacterium Candidatus “Thiodictyon syntrophicum” strain Cad16T. EuPA Open Proteomics 2, 17-30 (DOI: http://dx.doi.org/10.1016/j.euprot.2013.11.010). iii iv
REMERCIEMENTS
En premier lieu je tiens à remercier mon directeur de thèse CC Dr. Mauro Tonolla et co- directeur MER Dr. Xavier Perret de m’avoir permis d’entreprendre ce projet à l’Université de Genève, et au Laboratoire d’écologie microbienne de Bellinzona. Merci également au Prof. Michel Goldschmidt-Clermont professeur responsable pour cette thèse. Un remerciement spécial va au Professeur associé. Niels-Ulrik Frigaard pour avoir accepté de faire partie du jury de thèse et au Dr. Maged Saad pour son soutien scientifique tout au long de ce travail. Leur aide et leur expérience ont joué un rôle fondamental dans la réussite de cette thèse.
J’aimerais remercier tous les collaboratrices et les collaborateurs de l'ICM, en particulier le groupe du Laboratoire de Biosécurité et le groupe du Laboratoire d'Ecologie Microbienne pour leur aide. Un grand merci encore au Dr. Maged Saad pour avoir partagé ses connaissances sur la protéomique, au Dr. Sandro Peduzzi et au Dr. Paola Gandolfi-Decristophoris pour l’aide au niveau de la cultivation des microorganismes anaerobiques, et au Dr. Valeria Guidi, à Francesco Danza, à Anna Mariotti-Nessurini et au Dr. Damiana Ravasi pour toutes les discussions intéressantes que nous avons eues au cours de ces années.
Ma reconnaissance s’adresse également aux personnes que j’ai connues et m’ont gentiment aidé lors de mes sejours en Danmark à l’Université de Copenhague (Biocenter Department of Biology Section for marine biology): Carina Holkenbrink, Jørgen Deiker Petersen, Chizuko Sakamoto, Bjørn Sindballe Broberg, Tonny D. Hansen; et dans les laboratoires de l’Unité de Microbiologie à Geneve: Dr. Cristina Andrés-Barrao, Dr. Antoine Huyghe, Natalia Giot, Coralie Fumeaux, Dr. Nadia Bakkou, Anissa Ravez, Vanesa Miguelez De La Torre.
Je remercie la Fondation du Centre de Biologie Alpine de Piora et son président Prof. Dr. R. Peduzzi pour le soutien et la mise à disposition des infrastructures nécessaires aux travaux sur le lac de Cadagno ainsi que la Fédération des sociétés européennes de microbiologie (FEMS) et la Societé de Microbiologie Suisse (SMS/SGM) en particulier son ancien president Professeur Dr. Dieter Haas pour le soutien financier pendant mon sejour en Danmark.
Je tiens enfin à exprimer mon immense gratitude à Alice Benzoni et ma famille pour m’avoir continuellement soutenu durant cette période. v vi
TABLE OF CONTENTS
REMERCIEMENTS ...... iv
SUMMARY ...... x
RÉSUMÉ...... xii
ABBREVIATIONS ...... xvi
LIST OF TABLES AND FIGURES ...... xx
1. INTRODUCTION ...... 2
1.1. Meromixis ...... 2 1.1.1. The crenogenic meromictic Lake Cadagno ...... 4 1.1.2. Biota of the Lake Cadagno...... 6
1.2. Phototrophic sulfur bacteria ...... 8 1.2.1. The sulfur cycle ...... 8 1.2.2. Ecological distribution of phototrophic sulfur bacteria ...... 10 1.2.3. Photosynthetic inorganic carbon fixation in PSB and GSB ...... 11 1.2.4. Phototrophic sulfur bacteria in the Lake Cadagno ...... 15
1.3. Proteomics ...... 19 1.3.1. Techniques for separating proteins ...... 19 1.3.2. Protein identification using mass spectrometry ...... 20
1.4. Aims of the PhD thesis ...... 21
2. RESEARCH PAPER 1 ...... 24
CO2 assimilation in the chemocline of Lake Cadagno is dominated by a few types of phototrophic purple sulfur bacteria ...... 24
2.1. Supporting information ...... 40 SM1. Recipe of trace elements SL10 and SL12...... 40 SM2. Dissolved inorganic carbon (DIC) and pH from dialysis bags...... 41 SM3. Genome analysis...... 42 References supporting information ...... 43 3. RESEARCH PAPER 2 ...... 48 vii
Candidatus “Thiodictyon syntrophicum”, sp. nov., a new purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp...... 48
3.1. Supplementary material S1 (Material and methods) ...... 58 Enrichment and cultivation of strain Cad16T ...... 58 Phyologenetic analysis with 16S rRNA ...... 58 MALDI-TOF MS analysis ...... 58 Pigment analysis ...... 59
3.2. Supplementary material S2 ...... 60 Description of Candidatus “Thiodictyon synthrophicum” sp. nov. strain Cad16T...... 60 References supporting information ...... 61 4. RESEARCH PAPER 3 ...... 64
Proteomic analysis of the purple sulfur bacterium Candidatus “Thiodictyon syntrophicum” strain Cad16T isolated from Lake Cadagno ...... 64
5. DISCUSSION ...... 82
5.1. Primary production by anoxygenic bacteria ...... 83 5.1.1. Contribution of phototrophic sulfur bacteria to the primary production ...... 83 5.1.2. CO2 fixation in meromictic lakes in absence of light ...... 84 5.1.3. Autotrophic inorganic carbon fixation pathways ...... 84
5.2. Phototrophic sulfur bacteria in the meromictic Lake Cadagno ...... 86 5.2.1. Composition of the phototrophic sulfur bacterial population in the chemocline 86 5.2.2. Capacity of PSB to fix carbon ...... 87 5.2.3. Candidatus “Thiodictyon syntrophicum” strain Cad16T ...... 88
5.3. Proteomic analysis of Candidatus “T. syntrophicum” strain Cad16T ...... 90 5.3.1. Metabolism of Cad16T when grown in the presence of light ...... 90 5.3.2. Metabolism of Cad16T in absence of light ...... 92
5.4. Conclusions and perspectives ...... 96
6. REFERENCES ...... 100
7. ANNEXES ...... 118
7.1. Unpublished data ...... 118 7.1.1. Detection of putative RuBisCO proteins by western blot ...... 118 7.1.2. Expression of cbbL and cbbM directly in Lake Cadagno (in situ) ...... 123
7.2. DENMARK: FEMS Research Fellowships rapport ...... 126 viii
7.2.1. Transcriptomic analyses on the purple sulfur bacterium Thiodictyon sp. Cad16 ..... 126 7.2.2. APPENDIX A: Growth conditions ...... 135 ix x
SUMMARY
Lake Cadagno is a meromictic lake located at 1921 m.a.s.l. in the southern Swiss Alps (46°33’N, 8°43’E) in the catchment area of a dolomite vein rich in gypsum (Piora-Mulde). The chemistry of the lake is directly influenced by the particular geology of this valley. The water that percolates through these erodible sedimentary rocks becomes enriched in minerals and enters the lake through underwater springs. The inflow of this denser water produces a stable stratification of the lake and establishes meromixis. The transition zone between the oxygenic mixolimnion and the anoxygenic monimolimnion that is rich in salts is known as the chemocline and is characterized by steep gradients of oxygen, sulfide, and light. Generally, the chemocline that is positioned at a depth of approximately 12 m coincides with the presence of a dense community of anaerobic phototrophic sulfur bacteria of up to 107 cells ml-1 in summer. This community includes purple sulfur bacteria (PSB; family Chromatiaceae) of the genera Chromatium, Lamprocystis, Thiocystis and Thiodictyon and green sulfur bacteria (GSB; family Chlorobiaceae) of the genus Chlorobium. These phototrophic microorganisms fix inorganic carbon (CO2) via anaerobic photosynthesis using reduction equivalents from reduced sulfur compounds and sunlight as an energy source.
Phototrophic sulfur bacteria are important for primary production in many stratified lakes. In Lake Cadagno, these bacteria greatly contribute to the total primary production with high values of CO2 fixation both in the presence and absence of light. Rates of CO2 assimilation of the most 14 abundant phototrophic sulfur bacteria of the chemocline were measured by CO2 quantitative assimilation in dialysis bags (in situ) (Chapter 2). These results indicated that the most efficient fixer was the small-celled PSB Candidatus “Thiodictyon syntrophicum” with values of fixed- 14 -1 14 -1 CO2 as high as 0.6 (± 0.1) pg of C cell in presence of light and ca. 0.4 (± 0.1) pg of C cell in the dark. In contrast, Chlorobium clathratiforme that accounted for up to 95% of all phototrophic cells, showed extremely low levels of CO2 fixation.
Candidatus “T. syntrophicum” strain Cad16T was proposed as the type strain of a new species within the genus Thiodictyon (Chapter 3), and used as a model organism for further analyses. When cultivated in vitro in light/dark cycles of 12 hours, pure cultures of strain Cad16T exhibited 14 the highest CO2 fixation during the first four hours of light. Compared to the in situ results, the 14 14 CO2 fixation in the light was similar, while CO2 fixation in the dark was found to be lowed to 14 -1 ca. 0.1 (± 0.1) instead to ca. 0.4 (± 0.1) of pg of C cell . Normally, PSB fix CO2 using the Calvin-Benson-Bassham cycle (CBB cycle), in which the key enzyme is ribulose-1,5- bisphosphate carboxylase oxygenase (RuBisCO). The draft genome sequence of strain Cad16T xi revealed the presence of cbbL and cbbM genes, which encode form I and form II of RuBisCO, respectively. Transcription analyses confirmed that, while cbbM remained poorly expressed throughout light and dark cycles, cbbL expression was modulated by light and affected by the available carbon sources (e.g., acetate). Interestingly, cbbL expression did not correlate with the highest levels of CO2 assimilation.
Two-dimensional (2D) difference gel electrophoresis (DIGE) was used to monitor the changes in the proteome of strain Cad16T grown in anoxic autotrophic conditions in presence of light or in the dark (Chapter 4). Using Melanie 7.0 software, approximately 1,000 protein spots were identified amongst which 40 where found to be up- or down-regulated when comparing the two culture conditions. Twenty-three were up-regulated in the presence of light, and 17 were up- regulated in the dark. Among the 23 protein spots that were up-regulated in the light, three are involved in the storage mechanism that produces granules of poly(3-hydroxybutyrate) from an excess of reducing power and carbon compounds. Generally, bacteria use PHB as a reserve for carbon and energy as well as a sink for reducing equivalents. Among the 17 protein spots up- regulated in the dark, three were found to be part of the autotrophic dicarboxylate- hydroxybutyrate (DC/HB) cycle, which is known to be an autotrophic pathway for CO2 assimilation in Archaea. Given all of the above, it is tempting to speculate that the observed CO2 fixation in dark requires compounds synthesized during the day (such as PHB) as energy source and reducing power.
xii
RÉSUMÉ
Le Lac Cadagno est un lac méromictique situé dans la vallée de Piora à 1921 mètres d'altitude, dans la partie sud des Alpes suisses (46° 33'N , 8° 43'E ). La vallée de Piora est très particulière car elle est traversée par une veine de dolomite riche en gypse (Piora - Mulde) qui affecte la chimie du Lac Cadagno. En effet, l'eau qui filtre à travers les roches de dolomite et entre dans le lac par des sources sous-lacustres s’enrichie en minéraux et devient très dense. L'afflux de cette eau plus dense produit une stratification stable en établissant la méromicticité du lac. Entre la couche d’eau plus dense qui se trouve au fond (monimolimnion) et la couche d’eau supérieure (mixolinmion), on trouve une partie où se vérifient des changements très prononcés de gradient d’oxygène, de soufre et de lumière (chemocline). Typiquement, la chemocline se situe à une profondeur d'environ 12 m et à son intérieur se développe une riche communauté de bactéries sulfureuses phototrophes (jusqu'à 107 cellules ml-1 en été). Cette communauté est composée par deux grandes familles: les bactéries pourpres sulfureuses (PSB; famille Chromatiaceae) des genres Chromatium, Lamprocystis, Thiocystis et Thiodictyon et les bactéries vertes sulfureuses (GSB; famille Chlorobiaceae) du genre Chlorobium. Ces micro-organismes phototrophes utilisent des composés réduits du soufre (par exemple: H2S) comme donneurs d'électrons pour la fixation du carbone inorganique (CO2) grâce à l’énergie de la lumière. Les bactéries sulfureuses phototrophes sont importantes pour la production primaire dans de nombreux lacs stratifiés (méromictiques). Dans le Lac Cadagno, ces bactéries contribuent largement à la production primaire totale avec des valeurs élevées d’assimilation du CO2 soit en présence soit en absence de lumière. Les taux d'assimilation du CO2 des 4 plus grandes populations de bactéries sulfureuses phototrophes de la chemocline ont été mesurés par 14 l’assimilation quantitative de l’isotope radioactif CO2 du carbone (Chapitre 2). Ce résultat montre qu’une population est clairement plus active dans l’assimilation du CO2 avec ou sans lumière par rapport aux autres. En effet, la population du PSB Candidatus "Thiodictyon syntrophicum", avec une assimilation d’environ 0,61 (±0,11) pg de 14C cellule-1 en présence de lumière et environ 0,41 (± 0,09) 14C cellule-1 dans l’obscurité, semble être très importante dans la production primaire de la chemocline. Cela est davantage remarquable si l’on considère que la population du GSB Chlorobium clathratiforme, qui domine la chemocline avec environ 95% des cellules phototrophes totales, affiche une assimilation du CO2 extrêmement faible.
La population plus active dans l’assimilation du CO2, le Candidatus "Thiodictyon syntrophicum", a été isolée et caractérisée comme une nouvelle espèce du genre Thiodictyon (Chapitre 3), souche-type Cad16T, et utilisée comme organisme modèle pour des analyses xiii ultérieures en laboratoire. Des cultures pures de la souche Cad16T ont été faites pousser avec des cycles de 12 heures de lumière suivis par 12 heures d’obscurité en laboratoire (in vitro) afin 14 d’analyser leurs capacités d’assimilation du CO2 pendant des tranches de 4 heures. L’activité d’assimilation résulte majeure pendant les 4 premières heures d’exposition à la lumière.
Normalement, les PSB assimilent le CO2 par le cycle de Calvin-Benson-Bassham (cycle de CBB), dans lequel l'enzyme clé est le ribulose-1,5-bisphosphate carboxylase oxygénase (RuBisCO). L’analyse de la séquence brute du génome de la souche Cad16T révèle la présence de deux gènes, cbbL et cbbM, qui codent respectivement pour la forme I et la forme II de l’enzyme RuBisCO. L’analyse de l’expression du mRNA de ces deux gènes pendant une journée (chaque 4 heures) a montré une expression constante de la forme II cbbM, tandis que l’expression de la forme I cbbL semble être influencée par la lumière et les éventuelles souches de carbone (par exemple: acétate) présentes dans le milieu. Dans un milieu autotrophe, chaque 12 heures se vérifient deux pics d’expression du gène cbbL. Ce qui est remarquable est le fait 14 que ces pics ne correspondent pas au moment d’assimilation majeure du CO2 montré dans l’expriment précédant (Chapitre 2). La technique du électrophorèse bidimensionnelle différentielle sur gel utilisant des colorant fluorescents, appelée 2D-DIGE, a été utilisée pour surveiller les changements dans le protéome de Candidatus "T. syntrophicum" souche Cad16T dans un milieu autotrophe en présence ou en absence de lumière (Chapitre 4). Pour chaque expérience de 2D-DIGE, environ 1000 spots de protéines ont été identifiés à l'aide du logiciel spécifique (Melanie 7.0). Parmi ces spots, uniquement 40 satisfont les deux règles (l’ANOVA et l’intensité d’expression minimale) pour indiquer une différence d’expression effective entre les deux situations (lumière vs obscurité). En présence de lumière, 23 protéines étaient plus nombreuses par rapport à l’obscurité, tandis que 17 protéines étaient plus nombreuses dans l’obscurité par rapport à la lumière. Parmi les 23 protéines plus nombreuses en présence de lumière, 3 semblent être impliquées dans un processus qui produit des granules de poly(3-hydroxybutyrate) (PHB), en utilisant des excès de facteurs réduits (NAD[P]H) et des composés de carbone (Acetyl-CoA). Ces composés de réserves constituent une ressource très importante de pouvoir de réduction et d’énergie, étant donné que le carbone stocké peut ensuite être oxydé par le cycle des acides tricarboxyliques (TCA) produisant ATP. Dans l’obscurité, parmi les 17 protéines plus nombreuses, 3 peuvent nous aider à mieux comprendre l’assimilation de CO2 en absence de lumière. Ces 3 protéines font partie du cycle du dicarboxylate-hydroxybutyrate (DC-HB), qui est un cycle autotrophe observé jusqu’à maintenant uniquement chez les Archaea. xiv
Le mécanisme d’assimilation du CO2 en absence de lumière semble être lié à la production de composés de réserve faite en présence de lumière, tels que les PHB ou le glycogène qui stockent le potentiel de réduction et l’énergie nécessaires pour ce mécanisme. Selon nos résultats, les deux cycles, CBB et DC – HB, sont probablement impliqués dans l’assimilation du CO2 dans l’obscurité.
Le mécanisme d’assimilation du CO2 en absence de lumière semble être lié à la production de composés de réserve faite en présence de lumière, tels que les PHB qui stockent le potentiel de réduction et l’énergie nécessaires pour ce mécanisme. xv xvi
ABBREVIATIONS
2-DE Two-dimensional gel electrophoresis 2D-DIGE Two dimensional fluorescence difference gel electrophoresis 2D-PAGE Two dimensional polyacrylamide gel electrophoresis ATP Adenosine-5'-triphosphate BChl Bacteriochlorophyll bp Base pairs °C Celsius degree Cad16T Candidatus "Thiodictyon syntrophicum" nov. sp. strain Cad16Type strain CBB Calvin-Benson-Bassham 14C Radiocarbon, radioactive isotope of carbon CHCA Alpha-cyano matrix solution DAPI 4′,6-diamidino-2-phenylindole
DCF dark CO2 fixation DC/HB cycle Dicarboxylate/4-hydroxybutyrate cycle DIC Dissolved inorganic carbon DNA Deoxyribonucleic acid DSM Deutsche Sammlung von Mikroorganismen EDTA Ethylenediaminetetraacetic acid EMBL European Molecular Biology Laboratory FISH Fluoresent in situ hybridisation GSB Green sulfur bacteria ha Hectare (10’000 m2) HP/HB cycle 3-hydroxypropionate/4-hydroxybutyrate cycle HPLC High-performance liquid chromatography MALDI TOF MS Matrix assisted laser desorption ionization - time of flight mass spectrometry m a.s.l. Meters above sea level NAD(P)+ Nicotinamide adenine dinucleotide (phosphate-oxidase) oxided NAD(P)H Nicotinamide adenine dinucleotide (phosphate-oxidase) reduced nanoSIMS Nano-scale secondary-ion mass spectrometry NJ Neighbor-joining NPP Net primary production xvii
OD Optical density ORP Oxidation reduction potential pBLAST Protein Basic Local Alignment Search Tool PCR Polymerase chain reaction PFF Peptide fragmentation fingerprinting pH Power of hydrogen PHA Polyhydroxyalkanoates PHB Polyhydroxybutyrate PMF Peptide mass fingerprinting PNSB Purple nosulfur bacteria PSB Purple sulfur bacteria qRT-PCR Reverse transcription-quantitative polymerase chain reaction rRNA Ribosomal ribonucleic acid rTCA Reverse tricarboxylic acid RuBisCO Ribulose-1,5-bisphosphate carboxylase oxygenase SARAMIS Spectral archive and microbial identification system SDS Sodium dodecyl sulfate SRB Sulfate-reducing bacteria UV Ultraviolet xviii xix xx
LIST OF TABLES AND FIGURES
1. Introduction p. 2 Table 1. Classification of water stratification regimen. p. 3 Figure 1. Typical water stratification in a meromictic lakes. p. 5 Figure 2. The crenogenic meromictic Lake Cadagno. p. 8 Figure 3. The sulfur cycle. p. 14 Figure 4. Metabolic pathways for the assimilation of the CO2 in phototrophic sulfur bacteria. p. 17 Figure 5. Phototrophic sulfur bacteria isolated from the chemocline of the Lake Cadagno.
2. Research paper 1 p. 27 Table 1. Major characteristic of strains used in this study. p. 28 Table 2. Cy3-labeled oligonucleotide probe used in this study for FISH counting. p. 30 Figure 1. Vertical profiles of oxygen, sulfide, turbidity, light, ATP, temperature, conductivity, sulfate and oxidation reduction potential (ORP) on 12 september 2007.
14 p. 31 Figure 2. Mesurements of CO2 assimilation by representative GSB and PSB in the chemocline of lake Cadagno. p. 32 Table 3. Estimated contribution to CO2 fixation by selected groups of organism in lake Cadagno.
14 T p. 32 Figure 3. Variation of CO2 assimilation in strain Cad16 under laboratory growth conditions. p. 33 Figure 4. Levels of strain Cad16T cbbL and cbbM transcripts measured by qRT-PCR. p. 44 Table SM1. Physical parameters of the Lake Cadagno (12th September 2007). p. 45 Figure SM1. Specific FISH counts of GSB (white) and PSB (black) compared to the total prokaryotic cells counted by DAPI (grey) at different depths of Lake Cadagno 14 during the day of the CO2 assimilation analysis from the cultures pre-incubated in dialysis bags (in situ, September 12, 2007) xxi
3. Research paper 2 p. 51 Figure 1. Phase contrast micrograph of strain Cad16T. p. 52 Table 1. Differentiating characteristics of species of the genus Thiodictyon and related genera. p. 53 Figure 2. Maximum likelihood tree topology from 16S rRNA gene sequences for isolate Cad16T and other closely related species of the family Chromatiaceae created using PAUP*4.0b10 and a GTR model of sequence evolution. p. 54 Figure 3. MALDI-TOF MS dendrogram of 12 strains of phototrophic sulfur bacteria, resulting from single-link clustering analysis.
4. Research paper 3 p. 68 Table 1. CyDye Labeling Scheme and Gel Setup for 2D-DIGE Analysis p. 69 Figure 1. The protein expression patterns of Candidatus “T. syntrophicum” strain Cad16T separated in a 24 cm, pH 3-10 nonlinear strip and a 12% polyacrylamide gel. p. 70-71-72 Table 2. List of proteins identified by MALDI-TOF MS/MS p. 76 Figure 2. Scheme summarizing the majors metabolic pathways suggested from 2D-DIGE analysis.
5. Discussion p. 88 Figure 6. Incubation of the four major populations of phototrophic sulfur bacteria of Lake Cadagno at 12 m depth using dialysis bags. p. 93 Figure 7. The dicarboxylate/4-hydroxybutyrate cycle described in Desulfurococcales and Thermoproteales; (B) the 3-hydroxypropionate/4-hydroxybutyrate cycle described in Sulfolobales. p. 95 Figure 8. (A) The reductive citric acid (rTCA) cycle (B) Cyclic metabolism of PHB biosynthesis and degradation in bacteria
xxii
7. Annexes
7.1. Unpublished data p. 119 Figure A1. Detection of large subunit of the RuBisCO by Western Blot. p. 120 Figure A2. Western blot against the large subunit from the RuBisCO in a 2D-GEL of proteins extracted during the light exposure. p. 124 Figure A3. Expression of cbbL and cbbM in situ.
7.2. DANMARK: FEMS Research Fellowships rapport p. 131 Table F1. Nucleic acid concentration. p. 136 Figure F1. Schematized growth conditions.
1 Chapter 1 Introduction 2
1. INTRODUCTION
1.1. Meromixis
Vertical stratification in lakes occurs because of differences in densities of water layers, which are mainly influenced by temperature and dissolved substances. In turn, stratification affects pH, dissolved oxygen, nutrient concentrations, light transmission, and the composition of planktonic and benthic organisms (Tonolli 1969; Wetzel 1983; Imboden and Wüest 1995; Pourriot and Meybeck 1995; Wetzel 2001). Lakes normally present two types of stratification: temporal or permanent. Temporary stratification results in holomictic, oligomictic, and polymictic basins that differ in the frequency and intensity of water mixing periods (Table 1). In contrast, meromictic lakes are characterized by a permanent stratification due to incomplete water circulation, resulting in basins in which the lower portion of the water mass never mixes with the rest of the lake. The current classification of such water bodies generally distinguishes three types of meromixis: biogenic, ectogenic and crenogenic (Hutchinson 1937). The biogenic meromixis is due to an intense biological activity that results in an accumulation of dissolved salts and organic material in the lower part of the lake establishment of a stratification regime. Different in the ectogenic meromixis lakes, where the meromictic condition is initiated by some external superficial events as for example the intrusion of water more reach in salts and for this reason more dense, however unless this external event re-occurs the lake change in holomictic after some period of time. In crenogenic meromictic lakes, the stratification is caused by continuous supply of denser saline water by sublacustrin springs. In all cases, the morphometry of lake basin and its topographic position can enhance the stability of the stratification.
Table 1. Classification of water stratification regimen.
Holomictic: complete vertical mixing of the water body, at least once a year; Oligomictic: irregular, infrequent complete vertical mixing of the water body; Polymictic: complete vertical mixing more than twice in a year; Meromictic: permanently stratified water body, incomplete circulation;
3 Chapter 1
The water column of meromictic lakes is generally characterized by three distinct layers (see Figure 1): an upper layer (mixolimnion) that is usually oxic and characterized by a complete circulation of the water during the year; a bottom layer (monimolimnion) that is usually anoxic and characterized by a stagnant water body; and a narrow layer called chemocline that separates the mixo- and monimolimnion and is characterized by steep physical-chemical gradients. Partial mixing events (“meros”) occur in the mixolimnion, which is exposed to the atmosphere, and for this reason, subject to thermal stratifications during summer and mixing events in the cold season.
Figure 1. Typical water stratification in a meromictic lake with the upper oxygenic mixolimnion, the intermediate chemocline and the bottom anaerobic monimolimnion. Each layer harbours specific microoorganisms performing characteristic metabolic functions such as oxygenic photosynthesis and aerobic respiration in the oxic mixolimnion, anoxigenic photosynthesis in the chemocline and anoxigenic respiration in both the chemocline and the monimolimnion.
Because of their permanent stratification, meromictic lakes are interesting objects for microbial ecology, with anoxic sediments representing biological archives of the lake’s history as a result of a stable deposition and the lack of sediment perturbation (Brown and McIntosh 1987; Birch et al. 1996; Coolen and Overmann 1998; Romero et al. 2006; Ravasi et al. 2012; Introduction 4
Wirth et al. 2012). Meromictic lakes also represent ideal models for the study of biogeochemical processes mediated by microorganisms, such as sulfur and nitrogen cycle, due to their stable anaerobic environment find in the monimolimnion (Putschew et al. 1995; Hanselmann and Hutter 1998). In addition, in the chemocline, the physico-chemical gradients and the presence of light support the development of intense blooms of phototrophic sulfur bacteria (Sorokin 1970; Parkin and Brock 1981; Van Gemerden and Mas 1995; Overmann 1997; Tonolla et al. 1998a; Hadas et al. 2001; Rodrigo et al. 2001; Koizumi et al. 2004).
1.1.1. The crenogenic meromictic Lake Cadagno
Lake Cadagno is a crenogenic meromictic lake located at 1921 m a. s. l. in the Piora valley in the southern Swiss Alps (46°33’ N, 8°43’ E). The lake is located in a geographic area where ice and snow persist for up to 6 months during the winter season. The lake is approximately 850 m long and 430 m wide, with an average surface area of 26 ha and a maximum depth of 21 m (Del Don et al. 2001). Most probably, Lake Cadagno results from glacial erosion during the last glacial period with an estimated age of 8,000 to 11,000 years (Krige 1918; Del Don et al. 2001; Wirth et al. 2012). The Piora valley is characterized by the presence of a longitudinal vein of dolomite rock rich in gypsum that influences the chemistry of the Lake Cadagno through sublacustrine springs in its southern part (see Figure 2). This vein is believed to be an ancient seabed that was pushed upward during the Alps formation.
2- -1 With sulfate (SO4 ) concentrations ranging between 100 to 200 mg l (ca. 2 mM), Lake Cadagno contains up to 10 times more sulfate than most freshwater lakes (Hanselmann and -2 Hutter 1998). Such unusual concentrations of SO4 favored the development of sulfate-reducing bacteria (SRB) in the monimolimnion and in the anoxic sediments, thus contributing to an elevated production of hydrosulfide ions (HS-1) that accumulate to levels of 0.5 to 1 mMol in the monimolimnion. In turn, sulfide (H2S) supports the proliferation of a dense community of phototrophic sulfur bacteria in the chemocline at a depth of approximately 12 m where light still reaches the sulfide front that diffuses from the anoxic layer. By metabolizing reduced sulfur 2- compound such as sulfide and thiosulfate (S2O3 ), the phototrophic sulfur bacteria protect the eukaryotic aerobic organisms found in the upper mixolimnion. In fact, phototrophic sulfur bacteria use sulfide as an electron donor for the anoxygenic photosynthesis thus producing elemental sulfur (S0) first and then sulfate. The phylogeny and the population dynamics of phototrophic sulfur bacteria established in the chemocline have been studied since 1993 with 5 Chapter 1 molecular methods based on the sequencing of 16S rRNA genes (Peduzzi et al. 1993; Tonolla et al. 1998a; Tonolla et al. 1999; Tonolla et al. 2005c). The presence of phototrophic sulfur bacteria in Lake Cadagno for as long as 10,000 years ago was confirmed by the analysis of paleo- microbial communities found in the ancient anoxic sediment layers using as molecular tools 16S rRNA genes and quantitative real-time PCR (qPCR) that specifically targeted the contemporary purple and green sulfur bacteria populations (Ravasi et al. 2012).
Figure 2. Bathimetric map of the Lake Cadagno and its hydrogeological context modified from Wirth et al. (Wirth et al. 2013). Subaquatic springs are shown as white dots and surface in- and out-flows of water are represented by black arrows.
Moreover, Lake Cadagno and its anoxic layer is thought to represent living conditions similar to those found in the Precambrian era (4.5 billion to 545 million years ago) when free oxygen was probably absent from the Earth’s atmosphere. Thus, it may also represent a suitable model for studying the microbial ecology and evolution of life on Earth (Dahl et al. 2010). As meromictic lakes are relatively rare worldwide, Lake Cadagno was the object of numerous studies during the last 30 years leading to the accumulation of reliable data on its chemistry and biology (Peduzzi et al. 1991; Peduzzi et al. 1993; Fischer et al. 1996; Hanselmann and Hutter 1998; Lehmann et al. 1998; Peduzzi et al. 1998; Schanz et al. 1998; Tonolla et al. 1998a; Tonolla et al. 1998b; Tonolla et al. 1999; Bosshard et al. 2000; Lüthy et al. 2000; Camacho et al. 2001; Peduzzi et al. 2003b; Tonolla et al. 2003; Tonolla et al. 2004; Tonolla et al. 2005c; Tonolla and Introduction 6
Peduzzi 2006; Musat et al. 2008; Decristophoris et al. 2009; Gregersen et al. 2009; Habicht et al. 2011; Storelli et al. 2013a; Wirth et al. 2013).
1.1.2. Biota of the Lake Cadagno
Water stratification resulted in the formation of three distinct habitats throughout the water column of Lake Cadagno. The upper mixolimnion is approximately 12 m deep, and its water is alimented by surface runoff from of a small drainage area of approximately 2 km2 north of the lake that is composed of crystalline rocks of the Gotthard Massif. These crystalline rocks of the watershed are rather resistant to chemical weathering, resulting in a water of the mixolimnion 3- that is poor in salts but rich in oxygen. Waters until a 10 m depth have a low phosphate (PO4 ) -1 - -1 content close to the detection limit (<1 μg l ), with nitrate (NO3 ) below 50 μg l and dissolved inorganic carbon (DIC) of approximately 10 mg l-1. Phyto- and zooplanktonic communities are abundant within the first 10 m of the water column, with phytoplankton populations characterized by seasonal changes. In contrast, zooplankton communities are characterized by three major populations found unchanged throughout the year. During the summer, phytoplankton is mostly constituted by vertically and uniformly distributed Pennales and centric diatoms, whereas green algae (Echinocoleum, Sphaerocystis, and Oocystis) dominate during the autumn (Peduzzi et al. 1993; Schanz et al. 1998; Camacho et al. 2001). The chemocline located between the oxygenic mixolimnion and the anoxygenic monimolimnion is approximately 2 m thick and is characterized by steep gradients of light and chemical compounds such as oxygen, sulfide, phosphate, and ammonium (see Figure 1 of the research paper 1). A dense community of anaerobic phototrophic sulfur bacteria of up to 107 cells ml-1 in summer inhabits the chemocline, and includes photosynthetic purple sulfur bacteria (PSB; family Chromatiaceae) of the genera Chromatium, Lamprocystis, Thiocystis and Thiodictyon and green sulfur bacteria (GSB; family Chlorobiaceae) of the genus Chlorobium (Tonolla et al. 1999; Tonolla et al. 2004; Tonolla et al. 2005c) It was shown that the position of the bacterial layer in the water column is influenced by light conditions and physical displacement of water masses (Wüest 1994; Egli et al. 1998). The food chain connection with the mixolimnion is achieved by zooplankton (Eudiaptomus, Cyclops, Daphnia, Asplanchna) that grazes the bacterial communities established in the chemocline (Schanz and Stalder 1998; Camacho et al. 2001). The monimolimnion that stretches from -13 m to the bottom of the lake (-21 m) is anoxic and rich in reduced compounds because of the absence of oxygen and the activity of anaerobic bacteria, such as SRB. It is characterized by elevated concentrations of sulfate (up to 200 mg l-1), sulfide (≤30 mg l-1), nitrate 7 Chapter 1
(≤5 mg l-1), phosphate (≤0.4 mg l-1) and DIC (≤6 mg l-1). Bacteria are present at high concentrations (up to 105 cells per ml-1) throughout the monimolimnion with Desulfocapsa thiozymogenes (Desulfobulbaceae) and Desulfomonile tiedjei (Syntrophaceae) representing the most abundant species, also in the lower part of the chemocline (Tonolla et al. 1998b; Tonolla et al. 2000; Peduzzi et al. 2003b; Tonolla et al. 2005a; Garrity et al. 2007).
Sediments that cover the lake bottom above the chemocline depth are in contact with oxic water and cover a total area of 474’000 m2. In the southern part of the lake, these sediments are covered by macrophytes (Chara globularis var. globularis and Patamogeton) found between -1 to -7 m. Below this depth, the light and the sulfidogenic character of the bottom support the massive development of phototrophic bacterial mats to a depth of 11 m (Hanselmann and Hutter 1998). Below the chemocline (-12 m) sediments are permanently anoxic, and at a 21 m depth, the upper layer consists mostly of phototrophic sulfur bacteria and algae (mainly diatoms) that settled from the water column with an estimated sedimentation rate of 0.5 cm per year (Birch et al. 1996).
Introduction 8
1.2. Phototrophic sulfur bacteria
1.2.1. The sulfur cycle
Sulfur (S) is a brittle, yellow, tasteless, and odorless non-metallic element that is believed to be very abundant in the universe. Most of the Earth's sulfur is present in rocks and salts or buried deep in oceanic sediments, but can also be found in the atmosphere (Brown 1982). The emission of sulfur in the atmosphere comes from natural events such as volcanic eruptions, different bacterial processes (such as decay, respiration or photosynthesis), and evaporation from oceanic water. Furthermore, a large portion of the sulfur emissions in the atmosphere results from human activities, principally air pollution from large industrial areas coming from the combustion of organic fuels. In both cases, sulfur dioxide (SO2) and hydrogen sulfide (H2S) gases are emitted in 2- the atmosphere where they are quickly transformed into sulfate (SO4 ) by reacting with oxygen present in the air. The sulfate is then solubilized by rain water and returns to ground as acid deposition. Plants and microbes assimilate sulfate and convert it into organic forms, introducing it into the food chain. In fact, sulfur is an essential element for life, and it occurs mainly as constituents of protein (cysteine and methionine) but also in various coenzymes (e.g., coenzyme A, biotin, thiamine, etc.). Ultimately, death and subsequent decay of animals and plants reintroduce sulfur into the cycle (see Figure 3A).
Figure 3. (A) The sulfur cycle (1) and its major transformation processes (B) (Tang et al. 2009)
1 http://upload.wikimedia.org/wikipedia/commons/4/41/Sulfur_Cycle_%28Ciclo_do_Enxofre%29.png 9 Chapter 1
Of the different numbers oxidation states that exist in nature, sulfur us only found in significant amounts as sulfide (-2), elemental sulfur (0) and sulfate (+6). Transformations between these three major states result from biological as well as chemical processes, although microbial activity plays a primary role and is normally rapid (see Figure 3B).
2- 1.2.1.1. Sulfate reduction. In absence of molecular oxygen, sulfate (SO4 ) is used as a terminal electron acceptor for respiration by dissimilatory sulfate-reducing bacteria (SRB). Sulfate reduction is carried out with a variety of electrons donors such as lactate, pyruvate, malate, acetate, and others, which are oxidized to CO2 at the end of the respiration pathway (Widdel et al. 1992; Hao et al. 1996). SRB can be traced back to 3.5 billion years ago and are considered to be amongst the oldest life forms, having contributed to the sulfur cycle soon after life emerged on Earth (Schinck 1999; Barton and Fauque 2009). Other bacteria such as Proteus, Campylobacter, Pseudomonas and Salmonella also possess the ability to reduce sulfur but as facultative anaerobes they can use oxygen or other terminal electron acceptors. Normally, sulfide produced during anaerobic respiration is released into the environment thus contributing to maintain anaerobic conditions. Moreover, many bacteria can reduce small amounts of oxidized sulfur for the biosynthesis of cysteine, methionine, and other cofactors. The incorporated inorganic sulfate is first transformed in adenosine-5’-phosphosulfate (APS) by ATP sulfurylase and then further converted into 3′-phosphoadenosine-5′-phosphosulfate (PAPS) by the APS kinase, with PAPS playing a key role in sulfate transfer reactions to numerous substrates (Lin et al. 1995; Leustek et al. 2000; Negishi et al. 2001; Saito 2004).
1.2.1.2. Sulfide oxidation. Hydrogen sulfide (H2S), that is the most reduced form of inorganic sulfur, results from the activity of SRB found in sediments or in hydrothermal vents (see above 1.2.1.1. Sulfate reduction). Although sulfide is toxic to most organisms, mainly because of it interferes with aerobic respiration (Reiffenstein et al. 1992), it serves as electron donor for the energy-generating systems of photo- and chemo-lithotrophic bacteria as well as some Archaea (Brune 1995a; Van Gemerden and Mas 2004). During the autotrophic growth of these bacteria, sulfide provides electrons to a membrane-bound electron transport system in a process that ultimately leads to the production of reducing power such as NAD(P)H and reduced ferredoxin, which are required for fixation of the inorganic carbon. This transfer of electrons requires energy that is provided by light in phototrophic bacteria (Blankenship and Hartman 1998) and by dark redox reactions of inorganic substrates in chemo-lithotrophic bacteria (Kelly 1999). Introduction 10
1.2.1.3. Sulfur disproportionation. Sulfur disproportionation (dismutation or inorganic fermentation) is an energy-yielding process in the metabolism of numerous genera of SRB, which use inorganic sulfur compounds of intermediary oxidation states (elementary sulfur S0, 2- 2- thiosulfate S2O3 and sulfite SO3 ) as both donors and acceptors of electrons (Bak and Cypionka 1987; Cypionka et al. 1998). This metabolic process involved in sulfur transformation was shown to be particularly important in aquatic systems (Jörgensen 1990; Jörgensen et al. 1991).
1.2.2. Ecological distribution of phototrophic sulfur bacteria
Phototrophic sulfur bacteria are organisms commonly found in illuminated anaerobic aquatic environment containing hydrogen sulfide. In particular, these bacteria favour (1) stratified lakes where they occupy the top of the anoxic layer, and (2) the top layers of anoxic illuminated sediments on which they often establish colored microbial mats (Van Gemerden and Mas 2004). In holomictic lakes, where a thermal stratification is established transiently during summer, phototrophic sulfur bacteria are present as seasonal communities, whereas the permanent water stratification in meromictic lakes allows for the formation of stable communities throughout the year. A number of physico-chemical parameters influence the development of specific phototrophic sulfur microorganisms including the intensity and quality of light, which is affected by planktonic organisms (algae, bacteria) found in the water column, as well as the depth and chemistry of the chemocline (Abella et al. 1980; Vila and Abella 2001). Microbial mats are stratified environments that share some of the structural features of meromictic lakes, but on a smaller scale. Although these mats often harbor bacterial communities that differ significantly in composition and stability, the development of phototrophic sulfur bacteria is usually restricted to the upper 5 mm due to the rapid extinction of transmitted light into sediments. In aquatic planktonic systems, the infrared and UV part of the light spectrum are rapidly attenuated because of absorption and scattering, whereas in sediments, infrared radiation appears to penetrate deeper (Parkin and Brock 1980; Jorgensen and Des Marais 1986; Veldhuis and van Gemerden 1986).
Phototrophic organisms use a number of pigments (e.g. carotenoids) to capture light energy. In the case of purple sulfur bacteria (PSB), the most abundant pigment is okenone while spirilloxanthin, lycopenal and rhodopinal are clearly less abundant. In the case of green sulfur bacteria (GSB) pigments such as chlorobactene (green-colored species) and isorenieratene (brown-colored species) are normally observed. Carotenoids are not the only class of pigments 11 Chapter 1 found in phototrophic sulfur bacteria: the bacteriochlorophylls (BChl), in particular forms a and b for PSB and form c, d and e for GSB, were found to be essential for absorbing light and allowing a photoautotrophic growth. Pigments preserved in anoxic sediments or detected in living organisms are also used as markers to study a number of environmental conditions, evolutionary processes and biological origins (Xiong et al. 2000; Brocks and Schaeffer 2007).
1.2.3. Photosynthetic inorganic carbon fixation in PSB and GSB
Phototrophic sulfur bacteria are considered as anaerobic organisms transducing light energy into a biologically useful form without the generation of oxygen from the oxidation of water (they lack photosystem II). Instead of oxidizing water, these microorganisms use reduced forms of sulfur or possibly hydrogen gas (H2) as electrons donor for the reduction of CO2. The energy necessary for the process of CO2 fixation is generally provided by light, which is very efficiently collected by light-harvesting antenna pigments and then transferred to the photosynthetic reaction center, where electrons are displaced to specific acceptors such as NAD(P)+ and ferredoxin, creating at the same time a transmembrane potential (proton-motive force) used for the production of ATP (Overmann and Garcia-Pichel 2006).
Phototrophic sulfur bacteria are essentially divided into two major groups: the purple sulfur bacteria (PSB) and the green sulfur bacteria (GSB) (Imhoff 2004). In general, the GSB are considered to be obligate photo-autotrophs, whereas PSB are capable of both photo-autotrophy and photo-heterotrophy (Parkin and Brock 1981). GSB and PSB are also known to display different sensitivities towards molecular oxygen (O2), with GSB being obligate phototrophic anaerobes whereas some species of PSB that can may grow chemotropically in the presence of molecular oxygen, mainly under microaerophilic conditions (Trüper 1981; Kämpf and Pfennig 1986; de Wit and van Gemerden 1987). Three major structural and/or metabolic differences distinguish PSB from GSB: (1) the pigment composition (BChl and carotenoids) that affects the structure of the light-harvesting antenna and a different type of photosynthetic reaction center,
(2) the CO2 fixation pathway, and (3) the deposition of sulfur globules inside of PSB or outside of GSB the cells.
1.2.3.1. Pigments, light-harvesting antenna and photosynthetic reaction center. As stated above, different sets of pigments are observed in the PSB and GSB, in particular, the molecules of bacterio-chlorophyll (BChl) composing the light-harvesting antenna differentiate PSB from Introduction 12
GSB. As described above, the dominant pigments found in PSB are BChl a and b, while in GSB BChl c, d and e are the most abundant pigments. BChl a also occurs in GSB, but in lower quantities in photosynthetic reaction centers and not in antenna. PSB and GSB also differ in terms of the structural properties of their light-harvesting complexes. In GSB, these complexes are organized into highly structured organelles known as chlorosomes (Frigaard and Bryant 2006), allowing an highly efficient absorption of light energy. In fact, GSB appear to be better adapted than PSB to grow at low irradiance intensities, and thus are normally found deeper or just under the layer of PSB when coexisting in the same habitat. Two light-harvesting complexes (LH1 and LH2) are observed in PSB, and LH1 consists in pairs of small transmembrane polypeptides to which BChl is non-covalently attached in a 1:1 stoichiometry, forming with the reaction center a RC-LH1 ‘core’ complex. In contrast, the LH2 (or peripheral) antenna complex is only present in some purple bacteria (Law et al. 2004). The role of pigments in the light- harvesting antenna is to capture and then transfer the energy of the light to the photosynthetic reaction center. Photosynthetic reaction centers are divided into two main groups: type I, in which a ferredoxin (Fe-S) cluster acts as the terminal electron acceptor generally in GSB, and type II which in PSB and others phototrophic bacteria use a quinone as a terminal electron acceptor. Ultimately, the transit of the electrons from the photosynthetic reaction center to the cytochrome complex and the NAD(P)+ or other cofactors (e.g. ferredoxin in GSB), establishes a proton-motive force across the membrane, which is finally used to produce ATP by ATPases.
1.2.3.2. Pathway for fixing CO2. Although PSB and GSB use the reduced cofactors and ATP accumulated during photosynthesis, pathways involved in fixing CO2 are distinct (Sirevag et al.
1977; Berg 2011). PSB normally assimilate CO2 using the Calvin-Benson-Bassham cycle (CBB cycle) (see Figure 4A) in which the key enzyme is the ribulose-1,5-bisphosphate carboxylase oxygenase (RuBisCO). In contrast, GSB preferentially use the reductive reverse tricarboxylic acid (rTCA) cycle (see Figure 4B).
The CBB cycle can be divided into three major steps: (1) carbon fixation, (2) reduction of the fixed carbon, and (3) regeneration of molecules involved in the cycle. The first key reaction of carbon fixation is performed by RuBisCO that catalyzes the carboxylation of ribulose-1,5- bisphosphate (RuBP) into two molecules of 3-phosphoglyceric acid (3-PGA) using CO2 as substrate. Then 3-PGA is further metabolized and reduced using ATP and NAD(P)H in triose phosphate. The CBB cycle produces one molecule of triose phosphate (glyceraldehyde-3- phosphate, GAP) from 3 molecules of CO2. Ultimately the cycle regenerates RuBP as an 13 Chapter 1
acceptor for CO2. These reactions of regeneration in the CBB cycle are essentially catalyzed by non-specific enzymes also involved in others metabolic pathways, except for the phosphoribulokinase that regenerates RuBP from ribulose-5-phosphate (Ru-5-P) using one molecule of ATP. In total, the CBB cycle consumes nine ATP and six NADPH equivalents for the synthesis of one triose phosphate molecule.
The citric acid (TCA) cycle, also known as “Krebs cycle”, is a process utilized by aerobic organisms to generate energy (ATP) and reducing agents (NADPH) via the oxidation of acetate that is derived from carbohydrates, fats, and proteins into CO2. However, some microorganisms such as GSB favor the reverse reactions of the TCA (thus called rTCA cycle) in order to fix CO2 in triose phosphate. Several enzymes required for the rTCA are also active in the Krebs cycle, but a number of irreversible reactions need specific enzymes such as a fumarate reductase, one ferredoxin-dependent 2-oxoglutarate synthase, and an ATP-citrate lyase. These three enzymes are usually regarded as the characteristic enzymes of the rTCA cycle. As electron donors, the rTCA cycle uses reduced ferredoxin and NAD(P)H, which explains the presence in GSB of type I-photosynthetic reaction centers capable of reducing ferredoxin. Compared to the CBB cycle, the rTCA cycle seems more energy efficient since it requires only five ATPs to convert CO2 to triose phosphate. Introduction 14
Figure 4. Metabolic pathways for assimilating inorganic carbon: (A) the Calvin-Benson-Bassham cycle (CBB cycle) active in PSB, while GSB use the (B) reductive reverse tricarboxylic acid (rTCA) cycle (Berg 2011). The characteristic enzymes for each pathway are underscore.
(A) Enzymes: 1, RuBisCO; 2, 3-phosphoglycerate kinase; 3, glyceraldehyde-3-phosphate dehydrogenase; 4, triose-phosphate isomerase; 5, fructose-bisphosphate aldolase; 6, fructose-bisphosphate phosphatase; 7, transketolase; 8, sedoheptulose- bisphosphate aldolase; 9, sedoheptulose-bisphosphate phosphatase; 10, ribose- phosphate isomerase; 11, ribulose-phosphate epimerase; and 12, phosphoribulokinase.
(B) Enzymes: 1, ATP-citrate lyase; 2, malate dehydrogenase; 3, fumarate hydratase; 4, fumarate reductase; 5, succinyl-CoA synthetase; 6, ferredoxin-dependent 2-oxoglutarate synthase; 7, isocitrate dehydrogenase; 8, aconitate hydratase; 9, ferredoxin-dependent pyruvate synthase; 10, pyruvate/phosphoenolpyruvate synthase; 11, pyruvate/phosphoenolpyruvate carboxylase.
1.2.3.3. Sulfur globules. The ability to store surplus of various substances during unbalanced growth is a widespread strategy amongst microorganisms that need to cope with fluctuating environments. Phototrophic sulfur bacteria using reduced sulfur compounds during anoxygenic photosynthesis can accumulate sulfur globules, which are further oxidized to sulfate when the pool of reduced sulfur compounds becomes exhausted. In addition, it was shown that in the dark stored sulfur could be reduced to sulfide during the oxidation of glycogen because of the excess of electrons produced during the anaerobic breakdown of glucose (Van Gemerden 1968; Paschinger et al. 1974; Trüper 1978). Sulfur globules occur extracellularly in GSB as well as in the PSB members of the family Ectothiorhodospiraceae, while they are stored intracellularly in other PSB members of the Chromatiaceae (Frigaard and Dahl 2008). Irrespectively of their 15 Chapter 1 storage site, these globules consist of long sulfur chains generally terminated by organic residues similar to glutathione, most likely responsible for keeping the sulfur in a “liquid” state at ambient pressure and temperature conditions (Steudel 1996; Prange et al. 2002; Prange and Modrow 2002). Interestingly, whole-cell flotation experiments indicated that sulfur inside globules had an unexpectedly low density of approximately 1.2, compared with the common density of elemental sulfur (Guerrero et al. 1984; van Gemerden et al. 1985). This suggests a possible hydration of the sulfur to long hydrophilic chains once inside globules.
Intracellular sulfur globules by the PSB members of the Chromatiaceae appear in most cases to be separated from the cytoplasm by a unit membrane which may be continuous with the cytoplasmic membrane, depending on the organism (Pattaragulwanit et al. 1998). For example, in Allochromatium warmigii sulfur globules appeared as located at the two cell poles, whereas in Lamprocystis and Thiodictyon in the peripheral part of the cell and in Lamprobacter modestohalophilus in the “center” of the cell. Oxidation of sulfide at the outer surface of the cytoplasmic membrane, adjacent to the sulfur inclusions, may establish a proton gradient necessary for ATP synthesis and reduce the potential for sulfide toxicity within the cytoplasm. Sulfur globules are delimited by a protein envelope composed of 2 major types of “sulfur globule proteins” (SGP) of ca. 10.5 kDa (SgpA and SgpB) and 8.5 kDa (SgpC) (Brune 1995b; Pattaragulwanit et al. 1998; Prange et al. 2004; George et al. 2008).
In Ectothiorhodospira and GSB organisms, sulfur globules appear to be attached to the outer membrane of the cell wall via both spinae and capsules (Rojas et al. 1995; Pibernat and Abella 1996), thus making them difficult to be acquired by competing bacteria (van Gemerden 1986). Microscopic observations indeed confirmed that most sulfur compounds remained attached to cells, with little being free to float in the medium (Trüper 1984).
1.2.4. Phototrophic sulfur bacteria in the Lake Cadagno
The crenogenic meromictic Lake Cadagno possesses all of the characteristics needed to support sulfate-reducing and phototrophic sulfur bacteria. The permanent water stratification and the abundance of sulfur compounds brought by sublacustrine springs, allowed the formation of an environment favorable to anoxygenic sulfur bacteria (Pfennig 1975). The abundance of phototrophic sulfur bacteria is especially pronounced in the upper part of the anaerobic water layer, at the level of the chemocline, where a dense population can generally be detected, especially during the summer (Schanz et al. 1998; Del Don et al. 2001). Introduction 16
During the past 20 years, the major populations of phototrophic sulfur bacteria established in the chemocline and their role in the global economy of the lake were investigated (Peduzzi et al. 1993; Fischer et al. 1996; Tonolla et al. 1998b; Bosshard et al. 2000; Lüthy et al. 2000; Tonolla et al. 2000; Camacho et al. 2001; Peduzzi et al. 2003a; Tonolla et al. 2003; Tonolla et al. 2005b; Musat et al. 2008; Habicht et al. 2011). In particular, Tonolla et al. highlighted the presence of several new taxa of PSB (δ-Protobacteira - Chromatiaceae) and GSB (Chlorobi - Chlorobiaceae) (see Figure 5) (Tonolla et al. 1998a; Tonolla et al. 1999; Tonolla et al. 2004). Long-term monitoring of the phototrophic sulfur bacteria in the chemocline of Lake Cadagno also indicated the prevalence until 2002 of two major PSB populations: Chomatium okenii and Candidatus “Thiodictyon syntrophicum,”, the later of which was initially identified as Lamprocystis sp. population F. After the year 2000, GSBs initially represented only by Chlorobium pheobacteroides at low densities became preponderant, due to the unusual development of another new species C. clathratiforme. Such major change in community structure inside the chemocline was also accompanied by changes in the turbidity, sulfide concentration and light profiles. These changes were proposed to result from strong climatic events, such as autumnal windstorms that caused strong mixing events and a temporary disruption of the chemocline. The appearance of C. clathratiforme caused a PSB to GSB shift in dominating populations of chemocline bacteria, as well as a rise in the total number of phototrophic sulfur bacteria from ca. 106 to 107 cells per ml-1. Currently, C. clathratiforme is estimated to represent up to 95% of the phototrophic sulfur bacteria present in Lake Cadagno (Tonolla et al. 2005c; Decristophoris et al. 2009; Gregersen et al. 2009).
17 Chapter 1
Figure 5. (A) Neighbour-joining phylogenetic tree of DNA sequences of selected clones from a 16S rRNA gene library of microorganisms found in the chemocline of Lake Cadagno, together with published sequences archived in the the EMBL and GenBank databases. The distance scale indicates the expected number of changes per sequence position. Pictures from FISH experiments of the major populations of phototrophic sulfur bacteria are showed beside the corresponding clones (Tonolla et al. 2005c).
By using a series of agar-shake dilutions as in Pfenning (1978) in combination with various molecular techniques, the main populations of phototrophic sulfur bacteria could be isolated and grown separately in liquid media. Currently and with the exception of Ch. okenii, all main populations of phototrophic sulfur bacteria living in the chemocline of Lake Cadagno are being kept as pure cultures in our laboratory (see Figure 5). Among these populations, three were described as new species: Thiocystis cadagnonensis, T. chemoclinalis (Peduzzi et al. 2011), and the Candidatus “Thiodictyon syntrophicum” strain Cad16T, which has also the ability to form aggregates with the SRB Desulfocapsa thiozymogenes (Peduzzi et al. 2003a; Peduzzi et al. 2012). Introduction 18
The aforementioned previous studies on the diversity and dynamics of the population of phototrophic sulfur bacteria in the chemocline of Lake Cadagno represent an important starting point for subsequent functional and ecological studies. By fixing inorganic carbon using reductants (NAD[P]H) and energy (ATP) generated from sulfide oxidation (see paragraph 1.2.3), phototrophic sulfur bacteria prevent diffusion of toxic sulfide into the oxygen-rich mixolimnion in which fish, algae, phyto- and zooplankton as well as other aerobic bacteria thrive (Lüthy et al. 2000; Hell et al. 2008). Studies of the primary production in the lake showed that approximately half of the total CO2 assimilation of the lake occurred in the small volume of the chemocline
(Camacho et al. 2001). Moreover, the rates of dark CO2 fixation in the chemocline were even higher than rates of photo-assimilation, especially at depths where the oxygen and sulfide coexisted during part of the day (Camacho et al. 2001). 19 Chapter 1
1.3. Proteomics
Proteomics aims at following at the global level the changes that occur in protein expression, as well as quantify these changes. Organisms generally react to changes in their environment by expressing different sets of proteins (enzymes), which together constitute only a fraction of the total proteome. The term proteomics was first coined to form an analogy with genomics, the study of the genomes (complete set of genes) from an organism.
1.3.1. Techniques for separating proteins
In proteomics, one of the most important goals is to separate and distinguish as many of all of the proteins found in a given biological sample. Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) separates proteins according to their isoelectric point (pI) (the first dimension) and their molecular weight (MW) (second dimension). The pI corresponds to the pH at which the net charge of a protein is neutral. Once proteins migrated in a flat gel strip at the pH corresponding to their respective pI, they are loaded onto a polyacrylamide gel in which they will be separated by size. This procedure results in gels in which all of the proteins with the same pI and MW migrate together as one spot. Even proteins of a similar size will be resolved provided their pI is distinct, with large 2D gels resolving up to two thousand proteins. Since first described by Klose (1975), O'Farrell (1975) and Scheele (1975), this method is now widely used for separating complex mixtures of proteins. By labeling proteins with specific fluorescent dyes, the coverage and sensitivity of gel-based protein analyses was significantly improved in two- dimensional difference gel electrophoresis (2D-DIGE) (Tonge et al. 2001; Gharbi et al. 2002; Alban et al. 2003; Marouga et al. 2005). 2D-DIGE relies on the use of two mass- and charge- matched N-hydroxy succinimidyl ester derivatives of the fluorescent cyanine dyes Cy3 and Cy5, which possess distinct excitation and emission spectra. These are used to differentially label lysine residues of protein samples for a comparative analysis in a single gel. The ability to compare two samples in one gel not only limits gel-to-gel variations but also enables a more accurate and rapid quantification of differences in expression. By using an internal standard labelled with another dye (Cy2), inter-gel variations are further minimized and robustness of statistical analyses considerably improved.
Introduction 20
1.3.2. Protein identification using mass spectrometry
While two dimenional gel electrophoresis (2D-PAGE or 2D-DIGE) allows for reproducible protein separation, the identity of proteins that co-migrate in one spot remains unknown. If there are differences in spots between the proteins in a cell particularly stimulated or stressed and a normal cell, this method cannot determine the actual identity of the different proteins in the two cell samples. To identify these proteins, individual spots are first excised from 2D gels, then cleaved into shorter peptides by sequence specific proteases (e.g., trypsin), and ultimately identified via peptide mass fingerprinting (PMF) (Pappin et al. 1993; Yates et al. 1993). This is achieved by mass spectrometry using either a matrix-assisted laser desorption/ionization-time of flight mass spectrometer (MALDI-TOF MS) or electrospray ionization mass spectrometer (ESI- MS) which separate charged particles, or ions according to their respective mass to charge ratio (m/z) (Jensen et al. 1997; Fenyö et al. 1998; Andersen and Mann 2000). Analyzed samples generate complex spectra in which each individual peak represents peptides with identical m/z ratio. Since analyzed peptides were initially generated by cleaving material isolated from a single protein spot, it is possible to compare the resulting peptide mass fingerprint against a database of recorded spectra obtained by cleaving in silico all of the proteins deduced from a given genome. Dedicated softwares, such as Mascot (Matrix Science Ltd, London, UK) (Perkins et al. 1999), match the empirical against the most similar in silico spectra and ultimately provide a list of putative protein matches. Although identification of proteins via peptide mass fingerprinting is quiet accurate when the host-genome is fully sequenced, it is not necessarily a prerequisite. As many microbial genomes are now available in public databases such as the Integrated Microbial Genomes (National Center for Biotechnology Information, NCBI), it is possible to obtain a reliable protein-match even when the genome of the microbe from which the protein was isolated from is not sequenced. Thus, the rapid pace at which new genomes are currently being sequenced and added to public databases greatly reduced the impediment of missing genomic data. 21 Chapter 1
1.4. Aims of the PhD thesis
The compact chemocline of Lake Cadagno is characterised by a dense population of up to 107 phototrophic sulfur bacteria per ml, which was reported to assimilate high values of CO2 in the light and more interesting also in the dark (Camacho et al. 2001). Phototrophic sulfur bacteria of the chemocline are composed of two major groups: 1) the purple sulfur bacteria (PSB) and 2) the green sulfur bacteria (GSB) (see section 1.2.4 and Figure 5) (Tonolla et al. 2005c; Decristophoris et al. 2009). However, the ability of each single population to fix CO2 was never tested in the past. One aim of our research was to identify amongst the various PSB and GSB populations of the chemocline the bacterial strains that were the primary CO2 fixers. This lead to the identification of Candidatus “Thiodictyon syntrophicum” nov. strain Cad16T, a bacterium that forms functional aggregates with the sulfate reducing bacterium Desulfocapsa thiozymogenes (Peduzzi et al. 2003a). Because other representatives of the genus Thiodictyon (e.g. T. bacillosum and T. elegans) were not available as pure strains in reference collections of microorganisms, we could not formally confirm that strain Cad16T represented a new species, however.
To verify that Candidatus “T. syntrophicum” nov. strain Cad16T played a significant role in primary CO2-assimilation in Cadagno’s chemocline, a number of in vitro and in situ analyses T were carried out. These studies confirmed that cells of Cad16 fixed elevated levels of CO2 in presence of light as well as in the dark, and that Cad16T activity represented up to 25% of the chemocline primary production (Storelli et al. 2013). To gain a deeper molecular understanding of its metabolism, the genome of strain Cad16T was analyzed using high-throughput sequencing. The draft genome of 7.2 Mp long was assembled into 1,352 contigs (D. Bryant and N. U.
Frigaard, unpublished data) and shown to code a number of genes potentially involved in CO2 fixation via the Calvin-Benson-Basham (CBB) cycle. Amongst these genes, cbbL and cbbM coded for form I and form II of RuBisCO, respectively. Since cbbM remained mostly silent and T cbbL expression did not correlate with peaks of CO2 assimilation, the metabolism of Cad16 was further probed using a proteomic approach. Two-dimensional (2D) difference gel electrophoresis (DIGE) allowed us to monitor the changes in proteome when strain Cad16T was grown in vitro in presence or absence of light. These results that are presented in Chapter 4 were recently accepted for publication in the EuPA Open Proteomics journal (Storelli et al. 2014). Introduction 22
Thus, this PhD thesis contributed to define the ecological importance and role in CO2 fixation of some of the phototrophic sulfur bacteria that are established in the chemocline of Lake Cadagno. In particular:
1. A number of bacterial populations were examined for levels of CO2-fixed during the day or at night using an in situ approach;
2. Amongst these strains, Candidatus “Thiodictyon syntrophicum” Cad16T that appeared
as the strongest CO2 fixing population in pure cultures in vitro, was further analyzed at a molecular level;
T 3. Mechanisms used by Cad16 to fix CO2 were probed using a global proteomic approach.
23 Chapter 2 Research paper 1 24
2. RESEARCH PAPER 1
CO2 assimilation in the chemocline of Lake Cadagno is dominated by a few types of phototrophic purple sulfur bacteria
Nicola Storelli1,2, Sandro Peduzzi3, Maged M. Saad1, Niels-Ulrik Frigaard4, Xavier Perret1 and Mauro Tonolla1,2, 3 *
1 University of Geneva, Sciences III, Department of Botany and Plant Biology, Microbiology Unit, 30 quai Ernest-Ansermet, CH-1211 Geneva 4, Switzerland.
2 Institute of Microbiology, via Mirasole 22a, CH-6500 Bellinzona.
3 Alpine Biology Center (ABC), Foundation Piora Valley, CH-6777 Quinto, Switzerland.
4 Section for Marine Biology, Department of Biology, University of Copenhagen, DK-3000 Helsingør, Denmark.
Published in: FEMS Microbial Ecology; volule 84, issue 2 (may 2013): 421-432.
(DOI: 10.1111/1574-6941.12074) 25 Chapter 2
RESEARCH ARTICLE
CO2 assimilation in the chemocline of Lake Cadagno is dominated by a few types of phototrophic purple sulfur bacteria Nicola Storelli1,2, Sandro Peduzzi3, Maged M. Saad1, Niels-Ulrik Frigaard4, Xavier Perret1 & Mauro Tonolla1,2,3
1Department of Botany and Plant Biology, Microbiology Unit, University of Geneva, Sciences III, Geneva, Switzerland; 2Institute of Microbiology, Bellinzona, Switzerland; 3Alpine Biology Center (ABC), Quinto, Switzerland; and 4Section for Marine Biology, Department of Biology, University of Copenhagen, Helsingør, Denmark
Correspondence: Mauro Tonolla, Via Abstract Mirasole 22a, CH-6500 Bellinzona, Switzerland. Tel.: + 4191 814 60 74; Lake Cadagno is characterized by a compact chemocline that harbors high con- fax: +4191 814 60 19; centrations of various phototrophic sulfur bacteria. Four strains representing e-mail: [email protected] the numerically most abundant populations in the chemocline were tested in
dialysis bags in situ for their ability to fix CO2. The purple sulfur bacterium Received 6 June 2012; revised 9 January T Candidatus ‘Thiodictyon syntrophicum’ strain Cad16 had the highest CO2 2013; accepted 10 January 2013. assimilation rate in the light of the four strains tested and had a high CO2 Final version published online 11 February 2013. assimilation rate even in the dark. The CO2 assimilation of the population rep- resented by strain Cad16T was estimated to be up to 25% of the total primary T DOI: 10.1111/1574-6941.12074 production in the chemocline. Pure cultures of strain Cad16 exposed to cycles of 12 h of light and 12 h of darkness exhibited the highest CO2 assimilation Editor: Gary King during the first 4 h of light. The draft genome sequence of Cad16T showed the presence of cbbL and cbbM genes, which encode form I and form II of Keywords RuBisCO, respectively. Transcription analyses confirmed that, whereas cbbM cbbL primary production; RuBisCO; and remained poorly expressed throughout light and dark exposure, cbbL expres- cbbM mRNA; Candidatus ‘Thiodictyon sion varied during the light–dark cycle and was affected by the available carbon syntrophicum’ sp. nov. strain Cad16T. sources. Interestingly, the peaks in cbbL expression did not correlate with the
peaks in CO2 assimilation.
genus Chlorobium (Tonolla et al., 1999, 2004, 2005). Introduction From 1998 to 2004 the total concentration of photo- Lake Cadagno (Switzerland) is a crenogenic meromictic trophic sulfur bacteria increased from 4.9 (Æ 0.2) to 14 À lake characterized by a narrow chemocline with high con- (Æ 1.4) 106 cells mL 1, primarily due to a remarkable centrations of sulfates; steep gradients of oxygen, sulfide, rise in the population of the GSB Chlorobium clathrati- and light; and a turbidity maximum that correlates with forme (Decristophoris et al., 2009; Gregersen et al., 2009). À large concentrations of bacteria of up to 107 cells mL 1 Previous reports have shown that despite its small volume (Tonolla et al., 1999, 2004; Luthy€ et al., 2000). Studies in (approximately 10% of the lake), the chemocline of Lake other similarly stratified lakes have confirmed the impor- Cadagno is responsible for up to 40% of the total inor- tance of phototrophic sulfur bacteria to the carbon and ganic carbon photo-assimilation (Camacho et al., 2001). sulfur cycles (Pimenov et al., 2003; Garcıa-Cantizano Interestingly, significant rates of CO2 assimilation were et al., 2005; Dimitriu et al., 2008; Casamayor et al., also found to occur during the night, indicating that 2012). In Lake Cadagno, the chemocline is typically primary production also relies on mechanisms other found at a depth of about 12 m and is characterized by than photosynthesis. Chemoautotrophic bacteria such as
MICROBIOLOGY ECOLOGY MICROBIOLOGY the presence of a dense community of anaerobic photo- those belonging to the Thiobacillus genus are often
trophic sulfur bacteria, including purple sulfur bacteria responsible for CO2 fixation in the absence of light (Mar- (PSB) of the genera Chromatium, Lamprocystis, Thiocystis tinez et al., 1983) but they have not been detected in Lake and Thiodictyon and green sulfur bacteria (GSB) of the Cadagno (Tonolla et al., 1999, 2003; Bosshard et al.,
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 422 N. Storelli et al.
2000; Gregersen et al., 2009). In general, GSB are consid- large subunit homomers in arrangements of either 2 (L2) ered to be obligate photoautotrophic (Parkin & Brock, or 5 (L5) subunits (Tabita et al., 2008). 1981), whereas PSB are capable of both photoautotrophy The aim of this study was to determine the CO2 assim- and photoheterotrophy; in addition, some PSB strains are ilation of the various populations of phototrophic sulfur capable of growing chemotrophically in the dark and bacteria that are found in the chemocline of Lake Cadag- € under microaerophilic conditions (Kampf & Pfennig, no. To do this, we compared the CO2 assimilation rates 1986; de Wit & van Gemerden, 1987). Previous nano- in pure cultures of strains isolated from Lake Cadagno scale secondary-ion mass spectrometry (nanoSIMS) has and representing the four numerically most abundant shown that the inorganic carbon uptake is highly variable phototrophic sulfur bacteria populations. The pure cul- among three populations of phototrophic sulfur bacteria tures were cultivated, and subsequently pulsed with 14 in the Lake Cadagno chemocline (Musat et al., 2008). NaH CO3, in dialysis bags positioned at a depth of 12 m Green sulfur bacteria assimilate CO2 via the reverse (described as in situ conditions). In addition, the expres- tricarboxylic acid (rTCA) cycle, whereas PSB assimilate sion of the RuBisCO cbbL (Form I) and cbbM (Form II)
CO2 via the Calvin-Benson-Bassham cycle in which ribu- genes in pure cultures of the strain Candidatus ‘Thiodict- lose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO; yon syntrophicum’ Cad16T were determined over time EC 4.1.1.39) is a key enzyme. RuBisCO is found in most during light–dark cycles under laboratory conditions autotrophic organisms, as well as in eukaryotic algae and (described as in vitro conditions). Strain Cad16T was cho- plants (Ellis, 1979; Tabita, 2005; Mulkidjanian et al., sen as the model organism because it has been shown to
2006). There are four major forms of RuBisCO (I–IV) represent one of the most prominent CO2-assimilating known in nature, distinguishable by their primary amino populations in the lake both in the light and in the dark. acid sequences (Badger & Bek, 2008; Tabita et al., 2008). Forms I, II and III of RuBisCO contain catalytically active Materials and methods amino acid residues that are necessary for carboxylation and oxygenation. In contrast, form IV of RuBisCO (also Media and growth conditions known as the RuBisCO-like protein), which is found in several non-phototrophic bacteria (e.g. Bacillus subtilis) The major characteristics of the strains used in this study and GSB, is thought to be incapable of CO2 fixation are described in Table 1. PSB were grown in Pfennig’s because it lacks several conserved residues considered medium I (Eichler & Pfennig, 1988), and GSB were essential for RuBisCO carboxylase activity (Saito et al., grown in Pfennig’s medium II (Biebl & Pfennig, 1979), À1 2009). Form I, the most abundant form of RuBisCO, is both of which contain 0.25 g of KH2PO4 L , 0.34 g of À1 À1 found in many eukaryotes and bacteria and is a multi- NH4Cl L , 0.5 g of MgSO4Á7H2OL, 0.25 g of À1 À1 meric enzyme comprising eight CbbL subunits of 50– CaCl2Á2H2OL , 0.34 g of KCl L , 1.5 g of NaH- À1 À1 55 kDa that are associated with eight CbbS subunits of CO3 L , 0.02 mg of vitamin B12 L and 0.5 mL of À 12–16 kDa (Schneider et al., 1992). Form II is found in trace element solution SL12 L 1 for PSB and SL10 for certain Proteobacteria and consists of 2–8 CbbM subunits GSB (see Supporting Information, Table S1). The media (each of 50–55 kDa) depending on the microorganism. were prepared in 2-L bottles using a flushing gas compo-
Form III is found only in some Archaea and comprises sition of 80% N2 and 20% CO2 according to Widdel &
Table 1. Major characteristics of strains used in this study
Name Characteristic physiology Reference
Candidatus ‘Thiodictyon Purple sulfur bacterium; cells are ovoid-shaped, 1.4–2.4 lm Peduzzi et al. (2012) syntrophicum’ strain Cad16T in size, Gram-negative; anaerobic or microaerophilic growth;
CO2 fixation by the Calvin-Benson-Bassham (CBB) cycle Thiocystis chemoclinalis Purple sulfur bacterium; cells are ovoid-shaped, 2.3–3.6 lm Peduzzi et al. (2011) strain CadH11T in size, Gram-negative; anaerobic or microaerophilic growth;
CO2 fixation by the Calvin-Benson-Bassham (CBB) cycle Lamprocystis purpurea Purple sulfur bacterium; cells are ovoid-shaped, 1.9–2.3 lm Eichler & Pfennig, (1988) strain CadA31 in size, Gram-negative; anaerobic or microaerophilic growth;
CO2 fixation by the Calvin-Benson-Bassham (CBB) cycle Chlorobium clathratiforme Green sulfur bacterium; cells are rod-shaped, 0.3–1.2 lm Tonolla et al. (2004)
strain Cad4DE in size, Gram-negative; strictly anaerobic growth; CO2 fixation by the reverse tricarboxylic acid (rTCA) cycle
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved T CO2 fixation by the PSB Thiodictyon syntrophicum Cad16 423
Bak (1992) and were reduced by the addition of 1.10 mM Germany) and filter set F31 (Zeiss). Twenty fields of 2 Na2S.9H2O and adjusted to a pH of approximately 7.0. All 0.01 mm were counted, and cell densities were expressed À cultures were incubated at room temperature (20–23 °C) as the mean number of cells mL 1 (ÆSE). After the incu- and subjected to a light/dark photoperiod of 12 h with a bation period of 5 weeks in the dialysis bags at a depth À À light intensity of approximately 6 lEm 2 s 1 that was of 12 m, the number of viable cells was determined using generated with incandescent 60 W bulbs. All media pro- the Live/Dead BacLight Bacterial Viability kit (Invitrogen moted autotrophic growth except for the mixotrophic – Life Technologies Europe, Zug, Switzerland). medium used for the cbbL/cbbM mRNA expression analy- sis, to which 2 mM of sodium acetate was added. Concen- In situ analysis trations of sulfide in the cultures were measured daily and adjusted to about 1 mM throughout the experiments. The Meromictic Lake Cadagno is located in the Piora Valley growth was followed by measuring the optical density of at 1921 m above sea level in the southern Swiss Alps culture aliquots at a wavelength of 650 nm (OD650) using a (46°33′N, 8°43′E). Physicochemical parameters of the UV/VIS Spectrometer Lambda 2S (Perkin-Elmer Inc, water column were determined using a YSI 6000 profiler Waltham, MA). All biochemical analyses were performed (Yellow Springs, Inc., Yellow Springs, OH) and included À on cells taken from exponentially growing cultures with an temperature (°C), conductivity (lScm 1), pH, dissolved 7 À1 OD650 of c. 0.6, which corresponds to c. 1.0 9 10 oxygen (mg L ), redox potential (mV) and turbidity À cell mL 1. (FTU, formazine turbidity unit) (Table S1). Two LI-193SA spherical quantum sensors (LI-COR Ltd, Lin- coln, NE) were used to determine the percentage of trans- Strain identification, total cell count and mission down to the chemocline of photosynthetically viability tests active radiation (%PAR-light). Sulfide concentrations Bacterial populations in the chemocline were identified were measured using Cline’s reagent of water samples using fluorescent in situ hybridization (FISH) with species- fixed with ZnCl2 (Cline, 1969). 14 specific Cy3-labeled oligonucleotides (Table 2) in 1-lL In situ CO2 assimilation analyses were performed as aliquots of paraformaldehyde-fixed water samples (n = 3) follows: each of the four selected strains was grown sepa- spotted onto gelatin-coated slides [0.1% gelatin, 0.01% rately under laboratory conditions until the cultures € KCr(SO4)2] (Glockner et al., 1996). The hybridizations reached an OD650 of 0.6. This initial period of growth were performed as described by Zarda et al. (1997). The lasted for approximately 10 days due to the relatively long slides were treated with Citifluor AF1 (Citifluor Ltd., generation times, ranging from 110 to 130 h of cultures. London, UK) and examined by epifluorescence microscopy Then, 750 mL of these pure bacterial cultures were sealed using filter sets F31 (AHF Analysentechnik, Tubingen,€ in 60-cm-long dialysis bags (inflated diameter of Germany; D360/40, 400DCLP, and D460/50 for DAPI) and 62.8 mm; Karl Roth GmbH Co. KG, Karlsruhe, Germany) F41 (AHF Analysentechnik; HQ535/50, Q565LP, and and acclimatized for 5 weeks (from 8 August to 12 HQ610/75 for Cy3). The microorganisms were counted at September 2007) in the chemocline at a depth of 12 m. a 1000-fold magnification in 40 fields of 0.01 mm2 each With pores 25–30 A in diameter, the dialysis bags allowed (Fischer et al., 1995). for the free diffusion of molecules smaller than 20 kDa Cell concentrations of pure bacterial cultures in dialysis while preventing contamination of incubated cultures by bags were determined using samples fixed with 4% form- environmental bacteria (Simek et al., 2001, 2006; Corno aldehyde (final concentration) and stained with 0.001% &Jurgens,€ 2006; Blom et al., 2010; Lindstrom€ & Ostman, (w/v) 4′,6-diamidino-2-phenylindole (DAPI) (final con- 2011). After this 5-week pre-incubation period, the cul- centration). To count cells, 10 lL of each fixed sample tures were transferred from the dialysis bags into inde- 14 was deposited onto polycarbonate filters as described in pendent 100-mL sealed bottles with 125 lL of NaH CO3 Hobbie et al. (1977) and observed at 100-fold magnifica- for primary production measurements. To limit the expo- tion using an epifluorescence microscope (Axiolab, Zeiss sure of the bacterial culture to atmospheric oxygen, the
Table 2. Cy3-labeled oligonucleotide probes used in this study for FISH counting
Probe Target Sequence (5′→3′) (formamide in hybridization buffer) Reference
S453F Candidatus ‘Thiodictyon syntrophicum’ strain Cad16T CCCTCATGGGTATTARCCACAAGGCG (40%) Tonolla et al. (1999) S453H Thiocystis chemoclinalis strain CadH11T GACGGAACGGTATTAACGCCCCGCTT (10%) Tonolla et al. (2005) Apur453 Lamprocystis purpurea strain CadA31 TCGCCCAGGGTATTATCCCAAACGAC (40%) Tonolla et al. (1999) Chlc190 Chlorobium clathratiforme strain Cad4DE GGCAGAACAACCATGCGATTGT (20%) Tonolla et al. (2005)
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 424 N. Storelli et al.
cultures were transferred in a CO2-saturated environment. RNeasy Protect Mini Kit (Qiagen) and processed accord- Due to the short exposure time, neither the pH nor the ing to the manufacturer’s instructions. Total cellular RNA dissolved inorganic carbon (DIC) in the sample were was purified with the RNeasy Mini Kit (Qiagen), resus- modified by the sublimation of CO2 (Data S2). pended in RNase-free H2O and stored at À80 °C to min- imize degradation. Measurements of radioactive inorganic carbon assimilation Reverse transcription quantitative PCR The inorganic carbon fixation was evaluated in parallel in The levels of cbbL and cbbM gene transcription were mea- one opaque (dark) and one transparent (light) sealed sured by reverse transcription-quantitative PCR (qRT- 100-mL bottle. The inorganic carbon assimilation activity PCR) analysis using a Light Cycler instrument (Roche TM was measured using radioactive 14C isotope (NEC-086S Applied Science) and the QuantiTect SYBR Green RT- 14 NaH CO3; 1 mCi; 8.40 mCi/mmol; 1-mL ampoules with PCR kit (Qiagen) according to the manufacturer’s À specific activities of 20 lCi mL 1; Perkin Elmer, Schwer- instructions. Contigs JQ780325 and JQ780326 were used 14 zenbach, Switzerland). A total of 125 lL NaH CO3 as templates to design the specific primer pairs to deter- À solution, with an activity of 0.05 lCi mL 1, was added to mine the expression of cbbL and cbbM genes. The primers each 100-mL bottle. All bottles were completely filled used were cbbL-F (5′-cttcgagttcgtgggc-3′), cbbL-R (5′- with either a bacteria-free sample from the chemocline gcacgctcgtacatct-3′) and cbbM-F (5′-caggccaagattttctctgc- (filtrated with 0.22-lm filter), a chemocline water sample 3′), cbbM-R (5′-tacctgcactaccatcgtgc-3′) for cbbL and (12 m) or with bacterial samples from the dialysis bags. cbbM, respectively. The qRT-PCR reaction mixture con- Each sample was analysed in triplicate. The filled bottles tained the following components for both primer sets: were placed at a depth of 12 m in the chemocline and 10 lL of SYBR Green buffer, 0.4 lL of both primers incubated for 4 h (from 12:00 to 16:00 h). Subsequently, (0.2 lM primer final concentration), 5 lL of RNA tem- the total radioactive carbon that was assimilated was plate, 0.2 lL of qRT-PCR mix, 0.1 lL of UNG and measured by the method described by G€achter & Mares DNAse free water (Qiagen) to a total volume of 20 lL. (1979) to ensure a total loss of unbound radioactive The qRT-PCR mix was loaded onto the LightCycler sys- carbon through the bubbling method (Schindler et al., tem, and the reaction conditions were set up as follows: 1972). At the end of the bubbling period, 10 mL of Ready (1) 20-min reverse transcription at 50 °C; (2) 15-min ini- Gel scintillation liquid (Ready GelTM; Beckman Coulter, tial activation step at 95 °C; (3) 60 reaction cycles: 0 s at Fullerton, CA) was added to each scintillation vial, and 81 °C, 15 s at 94 °C, 20 s at 63 °C and 10 s at 72 °C; the radioactivity was measured in a Beckman LS 6000 (4) melting curve: 0 s at 95 °C, 15 c at 60 °C and 0 s at Scintillation Counter (Beckman, WS-BECKLS6). 99 °C; (5) 2 min at a 40 °C cooling step. To monitor the progress of the PCR reaction, a fluorescence signal was measured at a wavelength of 530 nm at the end of each Draft genome sequencing elongation phase. A relative quantification analysis was The draft genome of Candidatus ‘Thiodictyon syntrophi- performed using the LIGHTCYCLER 4.1, which enabled a cum’ strain Cad16T was determined by pyrosequencing in comparison of the ratio of the target mRNA in each sam- the laboratory of Dr. S. C. Schuster (Z. Liu, K. Vogl, ple to the reference standard curve. The standard curve N.-U. Frigaard, L. P. Tomsho, S. C. Schuster and D. A. was calculated using serial dilutions of cbbL or cbbM Bryant, pers. commun.) at the Genomics Core Facility of mRNA amplified from cultured bacteria with the previous Pennsylvania State University. Paired-end reads from GX- sets of primers. 20 FLX Titanium chemistries were assembled into 1352 primary contigs. Two contigs predicted to code for cbbL Results and cbbM were deposited in the NCBI public database under the respective accession numbers JQ780325 and Physicochemical and biological properties of JQ780326. Lake Cadagno Sample collection and in situ experiments were carried out RNA extraction on 12 September 2007. The main physicochemical param- Cells of strain Cad16T were grown in the laboratory eters of the water column on this day, including ATP val- under previously described conditions. At specific times, ues, are shown in Fig. 1. The water column was stratified 0.5 mL of the cultures (c. 0.5 9 107 cells) were mixed into three zones: the oxic mixolimnion (from the surface with 1.0 mL of RNAprotect Bacteria Reagent from the to approximately 11 m), the anoxic chemocline
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved T CO2 fixation by the PSB Thiodictyon syntrophicum Cad16 425
Fig. 1. Vertical distribution of oxygen, sulfide, turbidity, light, ATP, temperature, conductivity, sulfate, and oxidation reduction potential (ORP) on 12 September 2007.
(at approximately 12 m), and the anoxic, sulfide-contain- CO assimilation in the chemocline of Lake ing monimolimnion (from approximately 13 m down to 2 Cadagno the sediment). The transition zone from the oxic to the anoxic water layer was also revealed by the redox potential To measure the inorganic carbon assimilated in the (ORP) by a shift from positive to negative values at chemocline and the impact of light on this phenomenon, approximately 11 m. The turbidity maximum and the a water sample was collected from a depth of 12 m. After ATP values corresponded with a dense microbial commu- 4 h of incubation (from 12:00 to 16:00 h on a sunny 7 À1 nity of c.10 cells mL that populated the chemocline day), the concentration of fixed CO2 was determined to À and consisted mostly of GSB and PSB (as determined by be 297 (Æ 52) ng of 14CmL 1 in the presence of light À FISH). The GSB C. clathratiforme was the most abundant and 231 (Æ 98) ng of 14CmL 1 in the dark (see Fig. 2a). À species with 8.3 9 106 cells mL 1 (72.9% of the total bac- In parallel, the uptake of inorganic carbon was also terial concentration) at 12 m. The PSB Candidatus ‘Thio- measured using pure cultures of the four main autoch- dictyon syntrophicum’ strain Cad16T, Lamprocystis thonous strains from Lake Cadagno (Table 1), which purpurea and Thiocystis chemoclinalis strain CadH11T were were pre-incubated for equilibration for 5 weeks at a À clearly less abundant, with 1.3 9 105 cells mL 1 (1.2% of depth of 12 m in dialysis bags as described above. After À total), 4.9 9 104 cells mL 1 (0.4% of total) and 1.9 9 104 the equilibration period, the dialysis bags were retrieved, À cells mL 1 (0.2% of total), respectively. and more than 99% of the cultures consisted of live cells.
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 426 N. Storelli et al.
(a) (b)
14 Fig. 2. Measurements of CO2 assimilation by representative GSB and PSB in the chemocline of Lake Cadagno. (a) Primary production was first established using a water sample taken directly from the chemocline. (b) In parallel, CO2 fixation by pure cultures of Chlorobium clathratiforme Cad4DE, Thiocystis chemoclinalis CadH11T, Candidatus ‘T. syntrophicum’ Cad16T or Lamprocystis purpurea CadA31 was also measured. Bacteria 14 were incubated for 4 h at a depth of 12 m and in the presence of NaH CO3, either in the presence of light (white columns) or in the dark (black columns). Activity is reported as ng of 14C fixed mLÀ1 of chemocline water (Fig. 2A) or pg of 14C cellÀ1 within pure cultures (Fig. 2b). Each data point represents the mean of three independent measurements, with standard deviations shown as error bars.
As shown in Fig. 2b, Candidatus ‘T. syntrophicum’ the chemocline (light and dark). However, the primary T Cad16 and L. purpurea CadA31 were clearly the most CO2-assimilating organism(s) within the chemocline, efficient CO2-fixing strains, with levels of 0.61 (Æ 0.11) responsible for 66.1% of the carbon fixation, still remain À and 0.39 (Æ 0.09) pg of 14C cell 1 in the light, respec- (s) unknown. tively, whereas C. clathratiforme Cad4DE and T. chemocli- nalis CadH11T showed low to negligible activities of 0.001 À CO assimilation of Candidatus (Æ 0.0002) and 0.04 (Æ 0.02) pg of 14C cell 1, respec- 2 ‘T. syntrophicum’ Cad16T grown under tively. Interestingly, both strains Cad16T and CadA31 controlled laboratory conditions showed a strong carbon uptake of 0.41 (Æ 0.09) and 0.24 14 (Æ 0.07) pg of C per cell, respectively, in opaque bot- To gain an insight into the dynamics of CO2 assimilation tles, indicating that both strains retained the ability to fix in the organism that was determined to be most efficient
CO2 for at least 4 h in the dark. at C-fixation, a time course experiment was carried out As reported in Fig. 2a, a total of 528 ng (297 light + under laboratory conditions. Cells of Cad16T were grown
231 dark) of CO2 (arbitrarily set as 100%) was fixed by in Pfennig’s Medium to c. 0.6 OD650 under light/dark 14 microorganisms found in each milliter of the chemocline photoperiods of 12 h. Thereafter, NaH CO3 was added after 4 h of incubation in the presence or absence of every 4 h, to 50-mL aliquots that were further incubated 14 light. Given the respective rates per cell of carbon assimi- for 4 h prior to quantification of CO2. As shown in lation measured for pure cultures of strains Cad4DE, Fig. 3, CO2 assimilation reached a maximum CadH11T, Cad16T and CadA31 when incubated in the (0.89 Æ 0.03 pg of 14C per cell) during the first light per- dialysis bags, we attempted to estimate the respective con- iod (from 07:00 to 11:00 h) and decreased rapidly once tributions of each population represented by the strains cells were exposed to the dark (0.14 Æ 0.03 pg of 14C per (in % of the total) to the global pool of carbon fixed in cell between 19:00 and 23:00 h), ending at a minimum À the lake. This estimation included the concentration of assimilation activity of 0.04 Æ 0.003 pg of 14C cell 1 each of the populations in the chemocline as determined between 23:00 and 03:00 h. The period between 11:00 by FISH (see Table 3). In this model, the population rep- and 15:00 h, with an in vitro assimilation of resented by Candidatus ‘T. syntrophicum’ Cad16T 0.53 Æ 0.08 pg of 14C per cell, is comparable to the in À appeared to be the most important contributor among situ activity of 0.61 Æ 0.11 pg of 14C cell 1 measured the four populations tested, because it assimilated as during the time frame from 12:00 to 16:00 h in the light. 14 14 much as 25.9% of all the hypothetical carbon fixed within Similarly, the CO2 assimilation of 0.41 Æ 0.09 pg of C
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved T CO2 fixation by the PSB Thiodictyon syntrophicum Cad16 427
Table 3. Total estimated contributions to CO2 fixation by selected groups of organisms in Lake Cadagno (phototrophic populations are represented by specific strains)
Populations in the Levels of CO2 fixation Total CO2 Strains chemocline* (104 cells mLÀ1) Light + Dark† (pg C per 104 cells) fixed (ng C mLÀ1) % of total
C. clathratiforme Cad4DE 831 8 + 4 11 2.0 T. chemoclinalis CadH11T 2 450 + 40 1 0.2 Candidatus ‘T. syntrophicum’ Cad16T 13 6100 + 4100 137 25.9 L. purpurea CadA31 5 3900 + 2400 31 5.8 Others 289 Not determined Not determined 66.1 Chemocline sample‡ 1140 528§ 100.0
*Cells for each strain count in the chemocline (12 m) by FISH the same day of the experiment showed in Fig. 2 (12 September 2007). †Values shown in Fig. 2B. ‡Total cell count in the chemocline (12 m) by DAPI the same day of the experiment showed in Fig. 2 (12 September 2007). §Experimentally determined by the sum of values shown in Fig. 2A (see text).
annotation of strain Cad16T (Data S3). To establish
which form of RuBisCO is the most important for CO2 fixation in strain Cad16T, the transcript levels of cbbL and cbbM were determined in anoxic autotrophically or anoxic mixotrophically grown cells. When cultures
reached an OD650 of 0.6, aliquots were collected at 2–4h intervals, and total RNA was prepared as described in the Materials and methods section. Figure 4 shows the levels of cbbL and cbbM transcripts measured using qRT-PCR from total cellular RNA samples. In cells growing auto- trophically, cbbL was actively transcribed, with two dis- tinct peaks of expression at 15:00 h (12.4 Æ 1.1 pg of mRNA per 106 cells) and 03:00 h (15.4 Æ 1.1 pg of mRNA per 106 cells). In contrast, under mixotrophic conditions (2 mM of acetate), the expression of cbbL 14 Fig. 3. Variations of CO2 assimilation in cells of strain Candidatus remained approximately 10-fold lower than the expres- ‘T. syntrophicum’ Cad16T grown under laboratory conditions. Fixation sion under autotrophic conditions. Although the expres- of CO2 was quantified every 4 h in pure liquid anoxic and autotrophic sion was lower, under mixotrophic conditions, a high cultures with photoperiods of 12 h, and light/dark transitions fixed at level of expression was recorded at 15:00 and 03:00 h, 07:00 and 19:00 h. Assimilation is reported as pg of 14C fixed per with 0.88 Æ 0.004 and 0.69 Æ 0.004 pg of mRNA per 106 cell, with each data point representing the mean of three independent measurements. Standard deviations are shown as error cells, respectively. Moreover, the sampling point showing bars. The concentration of H2S was measured before every the minimal expression of cbbL was at 23:00 h, with 14 Æ 6 quantification of CO2 assimilation. 0.12 0.014 pg of mRNA per 10 cells corresponding to the minimal expression under autotrophic conditions, per cell between 19:00 and 23:00 h corresponded to the with 3.21 Æ 0.036 pg of mRNA per 106 cells. 14 in situ CO2 assimilation in the dark. The level of sulfide Expression of the locus that codes for form II RuBisCO was monitored during the experiment, but no correlation remained constant and at a low level throughout the with other parameters was found. The maximum sulfide experiment for both incubation conditions. The expres- level was at 15:00 h after 8 h of light, and the minimum sion of cbbM was approximately 5-fold higher in was at 3:00 h after 8 h of dark. autotrophic compared to mixotrophic conditions. How- ever, the maximum level of expression, at 09:00 h (0.06 Æ 0.009 pg of mRNA per 106 cells) in autotrophic Expression of cbbL and cbbM genes in and at 19:00 h (0.017 Æ 0.002 pg of mRNA per 106 cells) Candidatus ‘T. syntrophicum’ Cad16T in the in mixotrophic growing conditions, and the minimum presence or absence of an organic carbon level of expression, at 07:00 h (0.022 Æ 0.006 pg of source mRNA per 106 cells) in autotrophic and at 11:00 h Two putative genes coding for the RuBisCO form I (0.004 Æ 0.001 pg of mRNA per 106 cells) in mixotrophic (cbbL) and form II (cbbM) were found in the genome growing conditions, were independent of each other.
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 428 N. Storelli et al.
(Fig. 2b); moreover, these two species are also phyloge- netically closely related (Garrity et al., 2007). Musat et al. (2008) also showed that C. okenii (0.3% of the chemo- cline’s population) contributed up to 70% of the total inorganic carbon fixed during daylight. A similar estima- tion is proposed by us in Table 3, taking in account both
light and dark CO2 assimilation, in which it can be seen that Candidatus ‘T. syntrophicum’ Cad16T was the main assimilator measured, being responsible for about 25.9% of the total assimilation. However, the four most abun- dant populations of phototrophic sulfur bacteria under
study were responsible for only 33.9% of the total CO2 fixed in the chemocline, and the remaining 66.1% was not identified. According to Musat et al. (2008), we can Candidatus ’ T cbbL Fig. 4. Levels of ‘T. syntrophicum Cad16 (black) speculate that the remaining 66.1% of CO2 assimilation and cbbM (white) transcripts measured using qRT-PCR. Cells of strain in the chemocline is carried out by C. okenii. Unfortu- T Cad16 were grown under laboratory conditions in the presence nately, C. okenii could not be grown as a pure culture in (mixotrophic; squares) or absence (autotrophic; diamonds) of acetate, the laboratory and therefore the CO assimilation by this using light/dark photoperiods of 12 h. The data points are reported 2 as pg of each transcript per 106 cells and represent the means of population could not be estimated using our approach. three biological replicates, with standard deviations shown as vertical The chemocline of Lake Cadagno showed high rates of bars. carbon assimilation, not only in the presence of light of À c. 297 ng of 14CmL 1 but also in the dark at c. 231 ng 14 À1 of CmL (Fig. 2a). Similarly high CO2 assimilation Discussion rates in the dark were also reported for other stratified lakes, such as the Spanish karstic lakes (Casamayor et al., The chemocline of the meromictic Lake Cadagno harbors 2008, 2012; Casamayor, 2010), Lake Kinneret in Israel a complex microbial ecosystem in which phototrophic (Hadas et al., 2001) and Big Soda Lake in Nevada sulfur bacteria play a key role in inorganic carbon fixation (Cloern et al., 1983). In Lake Cadagno, the populations (Camacho et al., 2001). Using dialysis bags that allowed represented by Candidatus ‘T. syntrophicum’ Cad16T and bacterial cultures to remain pure while exposing and L. purpurea CadA31 were responsible for approximately equilibrating them to their natural environment (Simek one-third (28.7%) of the dark assimilation (data not et al., 2001, 2006; Corno & Jurgens,€ 2006; Blom et al., shown). Prior to the increase in the population of C. cla- 2010; Lindstrom€ & Ostman, 2011), we were able to assess thratiforme (Tonolla et al., 2005; Decristophoris et al., their CO2 assimilation in the chemocline at a depth of 2009; Gregersen et al., 2009), analysis of the CO2 assimi- 12 m (Fig. 2). Among the strains representing the four lation in the chemocline showed high rates of CO2 assim- most abundant populations of phototrophic sulfur bacte- ilation in the dark. At this time C. okenii was among the ria in the chemocline, Candidatus ‘T. syntrophicum’ dominant populations of phototrophic sulfur bacteria T Cad16 was the most effective CO2-fixing organism, in (Schanz et al., 1998; Camacho et al., 2001). This might both the presence and the absence of light. Surprisingly, indicate that PSB, and in particular C. okenii, are the the most abundant phototrophic sulfur bacterial popula- types of organisms mostly responsible for these high dark tion in the chemocline, the GSB C. clathratiforme consti- fixation rates. This speculation remains to be confirmed tuting 73% of total bacterial cells (Tonolla et al., 2005; by further analysis. There is also the possibility that a dif- Decristophoris et al., 2009; Gregersen et al., 2009) was ferent metabolic pathway (such as chemolithoautotrophy) the least efficient in terms of CO2 assimilation per cell, in unknown organisms may contribute to the dark car- with values that were approximately 100-fold lower than bon fixation activity in this ecosystem (Yngve Borsheim those measured for the Candidatus ‘T. syntrophicum’ et al., 1985; Jorgensen et al., 1991; Shively et al., 1998), to Cad16T. This finding confirms previous results from the the fermentation of stored glycogen (Gfeller & Gibbs, Lake Cadagno chemocline phototrophic community, 1984; Habicht et al., 2011), and to redox potential- which showed that C. clathratiforme was clearly a less balancing metabolisms (McKinlay & Harwood, 2010) or effective assimilator of CO2 than were Chromatium okenii other unknown metabolisms (Martinez-Garcia et al., and L. purpurea (Musat et al., 2008). In our study, 2011). To put it briefly, the chemocline of Lake Cadagno
L. purpurea CadA31 showed high rates of CO2 assimila- showed a high activity of dark CO2 assimilation that is tion similar to Candidatus ‘T. syntrophicum’ Cad16T not completely understood.
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved T CO2 fixation by the PSB Thiodictyon syntrophicum Cad16 429
Our long-term goal is to understand the microbial I cbbL and a form II cbbM (Data S3). Transcription anal- basis of CO2 fixation in Lake Cadagno in the light and in yses confirmed that cbbL expression was higher at 15:00 the dark. For this purpose, we have established laboratory and 03:00 h (Fig. 4). This suggests that a cyclic mecha- protocols to reproducibly cultivate and examine strains nism, possibly synchronized by light, regulates the expres- that we estimate represent the vast majority of the bacte- sion of cbbL in autotrophic conditions. In contrast, the ria found in Lake Cadagno’s chemocline – with the nota- presence of a suitable carbon source (e.g. acetate) in the ble exception of C. okenii. The ability to maintain these growth medium caused reduced transcription of cbbL. bacteria as pure cultures allowed us to study the molecu- Similar observations were reported for other PSBs such as lar mechanisms responsible for primary production. Can- Allochromatium vinosum (Valle et al., 1988; Kobayashi didatus ‘T. syntrophicum’ Cad16T was used as a model et al., 1991). The absence of regulation of the transcrip- organism to better understand the dynamics of carbon tion of cbbM under the tested growth conditions may fixation in the chemocline of the Lake Cadagno. While suggest that this process is not regulated by environmen- rates of CO2 assimilation in the light were comparable tal factors and therefore is constitutive. Differential between in vitro (0.53 Æ 0.08 pg of 14C per cell) and expression of forms I and II of RuBisCO was also in situ (0.61 Æ 0.11 pg of 14C per cell) conditions, this reported in other phototrophic bacteria, for which it was was not the case for rates in the dark. Indeed, the CO2 proposed that in presence of a reduced carbon source assimilated in vitro in the first 4 h of dark, at c. 0.14 pg (e.g. acetate), CbbM functions primarily as a terminal 14C per cell, was approximately threefold lower than the electron acceptor involved in maintaining the redox bal- value observed in situ in the dialysis bags c. 0.41 pg 14C ance of the cell (Wang et al., 1993; Yoshizawa et al., per cell. The absence of oxygen and organic substrates 2004; Badger & Bek, 2008; Joshi et al., 2009; Laguna (e.g. acetate) in the autotrophic Pfennig medium I sug- et al., 2010). gested two possible explanations. The absence of oxygen In conclusion, our results show that of the four strains in vitro may prevent the possibility of chemolithotrophic isolated from the chemocline of Lake Cadagno and for metabolism, in a way similar to that reported for the PSB which we established in situ and in vitro growth condi- Thiocapsa roseopersicina (de Wit & van Gemerden, 1987; tions, PSB Candidatus ‘Thiodictyon syntrophicum’ strain T Schaub & van Gemerden, 1994; Ende et al., 1996). How- Cad16 exhibited the highest CO2 assimilation activity, ever, the high rate of dark fixation normally combines both in presence of light and in the dark. Laboratory different metabolic processes more than simple chemo- experiments using pure cultures of Candidatus ‘T. syn- lithotrophy (Zopfi et al., 2001). In a highly reduced envi- trophicum’ Cad16T grown in an autotrophic medium ronment such as the chemocline of the Lake Cadagno, allowed us to detect that the maximal CO2 assimilation the ability of cells to maintain the redox balance can be rate occurred between 07:00 and 11:00 h, during the first very important. Recently, it was proposed that CO2 fixa- 4 h of light (Fig. 2). However, this maximal activity did tion is important not only as a primary production not correlate with the expression of any RuBisCO genes mechanism but also as an electron-accepting process in (cbbL and cbbM) (Fig. 3). In fact, RuBisCO form I cbbL mixotrophic organisms (Richardson et al., 1988; Hallen- showed two peaks of expression at 15:00 and 03:00 h, beck et al., 1990; Tavano et al., 2005). The Calvin cycle which occur more than 4 h before and 4 h the main in the purple non-sulfur bacterium Rhodobacter palustris assimilation period. The measured expression of RuBis- was shown to re-oxidize nearly half of the reduced cofac- COs genes suggested that form I cbbL is dependent on tors generated during the conversion of acetate to bio- environmental conditions, such as light exposure and car- mass, revealing that CO2 fixation plays a major role in bon sources, whereas the form II cbbM gene appears to cofactor recycling (McKinlay & Harwood, 2010). The dif- be constitutively expressed. ficulty in cultivating organisms in vitro highlights the problems encountered in attempting to dissect a complex Acknowledgements microbial ecosystem such as the chemocline of Lake Cad- agno and underscores the need for techniques that would Financial support for this project was provided by the enable studies in situ, such as the dialysis bags. University of Geneva, the Institute of Microbiology and
High rates of CO2 assimilation in the early hours of State of Ticino, and the Swiss National Science Founda- light exposure (Fig. 3) have also been observed in the tion (grant no. 31003A-116591). The authors are indebted diurnal cycles of cyanobacteria, and this often coincided to P. Decristophoris, A. P. Caminada, N. Ruggeri and with strong expression of RuBisCO-encoding genes C. Strambio-De-Castillia for technical and moral support. (Pichard et al., 1996; Wyman, 1999; Paul et al., 2000). In Furthermore, we are grateful to S. Bianchi, R. Bernasconi the draft genome sequence of Candidatus ‘T. syntrophi- and M. Molinari, Institute for Research in Biomedicine cum’ Cad16T, we identified two RuBisCOs genes, a form (IRB), in Bellinzona (Switzlerland), for their technical
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 430 N. Storelli et al. support. The authors would also like to thank the FEMS Dimitriu PA, Pinkart HC, Peyton BM & Mormile MR (2008) and Alpine Biology Center Foundation (ABC) for their Spatial and temporal patterns in the microbial diversity of a support. meromictic soda lake in Washington State. Appl Environ Microbiol 74: 4877–4888. Eichler B & Pfennig N (1988) A new purple sulfur bacterium References from stratified freshwater lakes. Amoebobacter purpureus sp. nov. Arch Microbiol 149: 395–400. Badger MR & Bek EJ (2008) Multiple Rubisco forms in Ellis RJ (1979) The most abundant protein in the world. proteobacteria: their functional significance in relation to T Biochem Sci 4: 241–244. CO acquisition by the CBB cycle. J Exp Bot 59: 1525–1541. 2 Ende FP, Laverman AM & van emerden H (1996) Coexistence Biebl H & Pfennig N (1979) Anaerobic CO uptake by 2 of aerobic chemotrophic and anaerobic phototrophic sulfur phototrophic bacteria. A review. Ergeb Limnol 12:48–58. bacteria under oxygen limitation. FEMS Microbiol Ecol 19: Blom JF, Hornak K, Simek K & Pernthaler J (2010) Aggregate 141–151. formation in a freshwater bacterial strain induced by growth Fischer K, Hahn D, Daniel O, Zeyer J & Amann RI (1995) state and conspecific chemical cues. Environ Microbiol 12: In situ analysis of the bacterial community in the gut of the 2486–2495. earthworm Lumbricus terrestris L. by whole-cell Bosshard PP, Santini Y, Gruter D, Stettler R & Bachofen R hybridization. Can J Microbiol 41: 666–673. (2000) Bacterial diversity and community composition in G€achter R & Mares A (1979) Comments on the acidification the chemocline of the meromictic alpine Lake Cadagno as and bubbling methods for determining phytoplankton revealed by 16S rDNA analysis. FEMS Microbiol Ecol 31: production. Oikos 33:69–73. 173–182. Garc ıa-Cantizano J, Casamayor E, Gasol J, Guerrero R & Camacho A, Erez J, Chicote A, Florin M, Squires M, Lehmann Pedros-Ali o C (2005) Partitioning of CO incorporation C & Bachofen R (2001) Microbial microstratification, 2 among planktonic microbial guilds and estimation of in situ inorganic carbon photoassimilation and dark carbon specific growth rates. Microb Ecol 50: 230–241. fixation at the chemocline of the meromictic Lake Cadagno Garrity GM, Lilburn TG, Cole JR, Harrison SH, Euz eby J & (Switzerland) and its relevance to the food web. Aquat Sci Tindall BJ (2007) Taxonomic outline of the Bacteria and 63:91–106. Archaea. Part 6–Phylum ‘Proteobacteria’: The Casamayor EO (2010) Vertical distribution of planktonic Deltaproteobacteria and Epsilonproteobacteria. Taxonomic autotrophic thiobacilli and dark CO fixation rates in lakes 2 Outline of Bacteria and Archaea Release 7.7: 148–245. with oxygen-sulfide interfaces. Aquat Microb Ecol 59: Gfeller RP & Gibbs M (1984) Fermentative metabolism of 217–228. Chlamydomonas reinhardtii I. Analysis of fermentative Casamayor EO, Garcia-Cantizano J & Pedros-Alio C (2008) products from starch in dark and light. Plant Physiol 75: Carbon dioxide fixation in the dark by photosynthetic 212–218. bacteria in sulfide-rich stratified lakes with oxic-anoxic Glockner€ FO, Amann R, Alfreider A, Pernthaler J, Psenner R, interfaces. Limnol Oceanogr 53: 1193–1203. Trebesius K & Schleifer KH (1996) An in situ hybridization Casamayor EO, Lliros M, Picazo A, Barberan A, Borrego CM protocol for detection and identification of planktonic & Camacho A (2012) Contribution of deep dark fixation bacteria. Syst Appl Microbiol 19: 403–406. processes to overall CO incorporation and large vertical 2 Gregersen LH, Habicht KS, Peduzzi S et al. (2009) Dominance changes of microbial populations in stratified karstic lakes. of a clonal green sulfur bacterial population in a stratified Aquat Sci-Res Acr Bound 74:61–75. lake. FEMS Microbiol Ecol 70:30–41. Cline JD (1969) Spectrophotometric determination of Habicht KS, Miller M, Cox RP et al. (2011) Comparative hydrogen sulfide in natural waters. Limnol Oceanogr 14: proteomics and activity of a green sulfur bacterium through 454–458. the water column of Lake Cadagno, Switzerland. Environ Cloern JE, Cole BE & Oremland RS (1983) Autotrophic Microbiol 13: 203–215. processes in meromictic Big Soda Lake, Nevada. Limnol Hadas O, Pinkas R & Erez J (2001) High chemoautotrophic Oceanogr 28: 1049–1061. primary production in Lake Kinneret, Israel: a neglected Corno G & Jurgens€ K (2006) Direct and indirect effects of link in the carbon cycle of the lake. Limnol Oceanogr 46: protist predation on population size structure of a bacterial 1968–1976. strain with high phenotypic plasticity. Appl Environ Hallenbeck PL, Lerchen R, Hessler P & Kaplan S (1990) Roles Microbiol 72:78–86. of CfxA, CfxB, and external electron acceptors in de Wit R & van Gemerden H (1987) Chemolithotrophic regulation of ribulose 1, 5-bisphosphate carboxylase/ growth of the phototrophic sulfur bacterium Thiocapsa oxygenase expression in Rhodobacter sphaeroides. J Bacteriol roseopersicina. FEMS Microbiol Lett 45: 117–126. 172: 1736–1748. Decristophoris PMA, Peduzzi S, Ruggeri-Bernardi N, Hahn D Hobbie JE, Daley RJ & Jasper S (1977) Use of nuclepore filters & Tonolla M (2009) Fine scale analysis of shifts in bacterial for counting bacteria by fluorescence microscopy. Appl community structure in the chemocline of meromictic Lake Environ Microbiol 33: 1225–1228. Cadagno, Switzerland. J Limnol 68:16–24.
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved T CO2 fixation by the PSB Thiodictyon syntrophicum Cad16 431
Jorgensen BB, Fossing H, Wirsen CO & Jannasch HW (1991) syntrophicum’, sp. nov., a new purple sulfur bacterium Sulfide oxidation in the anoxic Black Sea chemocline. Deep isolated from the chemocline of Lake Cadagno forming Sea Res I 38: S1083–S1103. aggregates and specific associations with Desulfocapsa sp. Joshi GS, Romagnoli S, VerBerkmoes NC, Hettich RL, Pelletier Syst Appl Microbiol 35: 139–144. D & Tabita FR (2009) Differential accumulation of form I Pichard SL, Campbell L, Kang JB, Tabita FR & Paul JH (1996) RubisCO in Rhodopseudomonas palustris CGA010 under Regulation of ribulose bisphosphate carboxylase gene photoheterotrophic growth conditions with reduced carbon expression in natural phytoplankton communities. I. Diel sources. J Bacteriol 191: 4243–4250. rhythms. MEPS 139: 257–265. K€ampf C & Pfennig N (1986) Chemoautotrophic growth of Pimenov NV, Rusanov II, Karnachuk OV et al. (2003) Thiocystis violacea, Chromatium gracile and C. vinosum in the Microbial processes of the carbon and sulfur cycles in Lake
dark at various O2-concentrations. JBasicMicrobiol26:517–531. Shira (Khakasia). Microbiology 72: 221–229. Kobayashi H, Viale AM, Takabe T et al. (1991) Sequence and Richardson DJ, King GF, Kelly DJ, McEwan AG, Ferguson SJ expression of genes encoding the large and small subunits of & Jackson JB (1988) The role of auxiliary oxidants in ribulose 1, 5-bisphosphate carboxylase/oxygenase from maintaining redox balance during phototrophic growth of Chromatium vinosum. Gene 97:55–62. Rhodobacter capsulatus on propionate or butyrate. Arch Laguna R, Joshi GS, Dangel AW, Luther AK & Tabita FR Microbiol 150: 131–137. (2010) Integrative control of carbon, nitrogen, hydrogen, Saito Y, Ashida H, Sakiyama T, de Marsac NT, Danchin A, and sulfur metabolism: the central role of the Calvin- Sekowska A & Yokota A (2009) Structural and functional Benson-Bassham cycle. Rec Adv Phototr Prok 675: 265–271. similarities between a ribulose-1, 5-bisphosphate Lindstrom€ ES & Ostman O (2011) The importance of carboxylase/oxygenase (RuBisCO)-like protein from Bacillus dispersal for bacterial community composition and subtilis and photosynthetic RuBisCO. J Biol Chem 284: functioning. PLoS ONE 6: e25883. 13256–13264. Luthy€ L, Fritz M & Bachofen R (2000) In situ determination Schanz F, Fischer-Romero C & Bachofen R (1998) of sulfide turnover rates in a meromictic alpine lake. Appl Photosynthetic production and photoadaptation of Environ Microbiol 66: 712. phototrophic sulfur bacteria in Lake Cadagno (Switzerland). Martinez JP, Garay E, Alcaide E & Hernandez E (1983) The Limnol Oceanogr 43: 1262–1269. genus Thiobacillus: physiology and industrial applications. Schaub BEM & van Gemerden H (1994) Simultaneous Acta Biotechnol 3:99–124. phototrophic and chemotrophic growth in the purple sulfur Martinez-Garcia M, Swan BK, Poulton NJ, Gomez ML, bacterium Thiocapsa roseopersicina M1. FEMS Microbiol Ecol Masland D, Sieracki ME & Stepanauskas R (2011) High- 13: 185–195. throughput single-cell sequencing identifies Schindler DW, Schmidt RV & Reid RA (1972) Acidification photoheterotrophs and chemoautotrophs in freshwater and bubbling as an alternative to filtration in determining bacterioplankton. ISME J 6: 113–123. phytoplankton production by the 14C method. Can J Fish McKinlay JB & Harwood CS (2010) Carbon dioxide fixation as Aquat Sci 29: 1627–1631. a central redox cofactor recycling mechanism in bacteria. Schneider G, Lindqvist Y & Branden CI (1992) RUBISCO: P Natl Acad Sci USA 107: 11669–11675. structure and mechanism. Annu Rev Bioph Biom 21: 119–143. Mulkidjanian A, Koonin E, Makarova K et al. (2006) The Shively J, Van Keulen G & Meijer W (1998) Something from cyanobacterial genome core and the origin of almost nothing: carbon dioxide fixation in photosynthesis. P Natl Acad Sci USA 103: 13126–13131. chemoautotrophs. Annu Rev Microbiol 52: 191–230. Musat N, Halm H, Winterholler B et al. (2008) A single-cell Simek K, Pernthaler J, Weinbauer MG et al. (2001) Changes view on the ecophysiology of anaerobic phototrophic in bacterial community composition and dynamics and bacteria. P Natl Acad Sci USA 105: 17861–17866. viral mortality rates associated with enhanced flagellate Parkin TB & Brock TD (1981) The role of phototrophic grazing in a mesoeutrophic reservoir. Appl Environ Microbiol bacteria in the sulfur cycle of a meromictic lake. Limnol 67: 2723–2733. Oceanogr 26: 880–890. Simek K, Hornak K, Jezbera J et al. (2006) Maximum growth Paul J, Kang JB & Tabita FR (2000) Diel patterns of regulation rates and possible life strategies of different bacterioplankton of rbcL transcription in a Cyanobacterium and a groups in relation to phosphorus availability in a freshwater Prymnesiophyte. Mar Biotechnol 2: 429–436. reservoir. Environ Microbiol 8: 1613–1624. Peduzzi S, Welsh A, Demarta A, Decristophoris P, Peduzzi R, Tabita F (2005) Research on carbon dioxide fixation in Hahn D & Tonolla M (2011) Thiocystis chemoclinalis sp. photosynthetic microorganisms (1971–present). Photosynth nov. and Thiocystis cadagnonensis sp. nov., two new motile Res 80: 315–332. purple sulfur bacteria isolated from the chemocline of Tabita FR, Satagopan S, Hanson TE, Kreel NE & Scott SS meromictic Lake Cadagno, Switzerland. Int J Syst Evol (2008) Distinct form I, II, III, and IV Rubisco proteins from Microbiol 61: 1682–1687. the three kingdoms of life provide clues about Rubisco Peduzzi S, Storelli N, Welsh A, Peduzzi R, Hahn D, Perret X evolution and structure/function relationships. J Exp Bot 59: & Tonolla M (2012) Candidatus ‘Thiodictyon 1515–1524.
FEMS Microbiol Ecol 84 (2013) 421–432 ª 2013 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 432 N. Storelli et al.
Tavano CL, Podevels AM & Donohue TJ (2005) Identification chemocline of a meromictic lake in relation to the precision of genes required for recycling reducing power during of the sampling procedure. FEMS Microbiol Lett 31: photosynthetic growth. J Bacteriol 187: 5249–5258. 337–341. Tonolla M, Demarta A, Peduzzi R & Hahn D (1999) In situ Yoshizawa Y, Toyoda K, Arai H, Ishii M & Igarashi Y (2004)
analysis of phototrophic sulfur bacteria in the chemocline of CO2-responsive expression and gene organization of three meromictic Lake Cadagno (Switzerland). Appl Environ ribulose-1, 5-bisphosphate carboxylase/oxygenase enzymes Microbiol 65: 1325–1330. and carboxysomes in Hydrogenovibrio marinus strain MH- Tonolla M, Peduzzi S, Hahn D & Peduzzi R (2003) Spatio- 110. J Bacteriol 186: 5685–5691. temporal distribution of phototrophic sulfur bacteria in the Zarda B, Hahn D, Chatzinotas A, Schonhuber€ W, Neef A, chemocline of meromictic Lake Cadagno (Switzerland). Amann RI & Zeyer J (1997) Analysis of bacterial FEMS Microbiol Ecol 43:89–98. community structure in bulk soil by in situ hybridization. Tonolla M, Peduzzi S, Demarta A, Peduzzi R & Hahn D Arch Microbiol 168: 185–192. (2004) Phototropic sulfur and sulfate-reducing bacteria in Zopfi J, Ferdelman TG, Jorgensen BB, Teske A & Thamdrup B the chemocline of meromictic Lake Cadagno, Switzerland. (2001) Influence of water column dynamics on sulfide J Limnol 63: 161–170. oxidation and other major biogeochemical processes in the Tonolla M, Peduzzi R & Hahn D (2005) Long-term chemocline of Mariager Fjord (Denmark). Mar Chem 74: population dynamics of phototrophic sulfur bacteria in the 29–51. chemocline of Lake Cadagno, Switzerland. Appl Environ Microbiol 71: 3544–3550. Valle E, Kobayashi H & Akazawa T (1988) Transcriptional Supporting Information regulation of genes for plant-type ribulose-1, 5-bisphosphate carboxylase/oxygenase in the photosynthetic bacterium, Additional Supporting Information may be found in the Chromatium vinosum. Eur J Biochem 173: 483–489. online version of this article: Wang X, Falcone DL & Tabita FR (1993) Reductive pentose Data S1. Recipe of trace elements SL10 and SL12. phosphate-independent CO fixation in Rhodobacter 2 Data S2. Dissolved inorganic carbon (DIC) and pH from sphaeroides and evidence that ribulose bisphosphate dialysis bags. carboxylase/oxygenase activity serves to maintain the redox balance of the cell. J Bacteriol 175: 3372–3379. Data S3. Genome analysis. Widdel F & Bak F (1992) Gram-negative mesophilic sulfate- Table S1. Physical parameters of the Lake Cadagno of the reducing bacteria. Prokaryotes 4: 3352–3378. 12 September 2007. Wyman M (1999) Diel rhythms in ribulose-1, 5-bisphosphate Fig. S1. Specific FISH counting of GSB (white) and PSB carboxylase/oxygenase and glutamine synthetase gene expression (black) compared to the total prokaryotic cells counted in a natural population of marine picoplanktonic cyanobacteria by DAPI (grey) at different depths of Lake Cadagno dur- – 14 (Synechococcus spp.). Appl Environ Microbiol 65: 3651 3659. ing the day of the CO2 assimilation analysis from the Yngve Borsheim K, Gijs Kuenen J, Gottschal J & Dundas I cultures pre-incubated in dialysis bags (in situ, September (1985) Microbial activities and chemical gradients in the 12, 2007).
ª 2013 Federation of European Microbiological Societies FEMS Microbiol Ecol 84 (2013) 421–432 Published by Blackwell Publishing Ltd. All rights reserved Research paper 1 38
39 Chapter 2 Research paper 1 40
2.1. Supporting information
SM1. Recipe of trace elements SL10 and SL12.
SL12 (Pfennig & Trüper, 1981):
Distilled water 1 liter
Ethylene diamine tetraacetate-Na2 3 g
FeSO4 – 7H2O 1.1 g
H3BO3 300 mg
CoCl2 – 6H2O 190 mg
MnCl2 – 4H2O 50 mg
ZnCl2 42 mg
NiCl2 – 6H2O 24 mg
Na2MoO4 – 2H2O 18 mg
CuCl2 – 2H2O 2 mg
The salts are dissolved in the order given, the pH are adjusted to 2-3 with HCl, the solution is sterilized and add to the Pfennig’s Medium 1.
41 Chapter 2
SL10 (Widdel, et al., 1983):
HCl (25% w/v) 10 ml
FeCl2 – 4H2O 1.5 g
CoCl2 – 6H2O 190 mg
MnCl2 – 4H2O 100 mg
ZnCl2 70 mg
H3BO3 6 mg
Na2MoO4 – 2H2O 36 mg
NiCl2 – 6H2O 24 mg
CuCl2 – 2H2O 2 mg
Distilled water 990 ml
The FeCl2 is dissolved in the HCl first, than distilled water is added and the other salts are sequentially dissolved.
SM2. Dissolved inorganic carbon (DIC) and pH from dialysis bags.
A sample of 100 ml distilled water was removed from a dialysis bag into a 100 ml bottle (the 14 same as used for the CO2 quantification experiment) in a CO2-saturated atmosphere obtained by the sublimation of dry ice. The dissolved inorganic carbon (DIC) and the pH, measured with the Alkalinity-Test kit (Merck KGaA, Darmstadt, Germany) and a 713 pH Meter (Metrohm, Herisau, Switzerland), respectively, were measured first in the dialysis bag and then in the bottle.
The result showed no effect of the CO2-saturated atmosphere on the distilled water sample. The values of DIC and pH remained almost the same, at 1.8 mmol l-1 1.8 mmol l-1 and 6.521 6.574, respectively.
Research paper 1 42
SM3. Genome analysis.
Analysis of the draft genome sequence of Candidatus “T. syntrophicum” strain Cad16T (1,352 contigs covering more than 7,280 kb) allowed us to identify two putative RuBisCO operons in two independent contigs: one containing the type I (cbbL, GenBank acc. JQ780325) and the other the type II (cbbM, JQ780326) RuBisCO-encoding genes. As in other microorganisms, the operon containing cbbL (JQ693373) also includes cbbS (JQ693374), as well as 6 genes putatively coding for components of a carboxysome (JQ693375-JQ693380) (Cannon, et al., 2010). In another contig, cbbM (JQ693382) is followed by cbbQ (JQ693383) and cbbO (JQ693384), while cbbR (JQ693381) is encoded by the complementary strand. The CbbR protein belongs to the LysR-type family of transcriptional regulators (LTTRs) and has been shown to regulate the expression of RuBisCO-encoding genes in various photo- and chemo-autotrophic bacteria (Dubbs & Tabita, 2003, Toyoda, et al., 2005, Maddocks & Oyston, 2008). CbbQ and cbbO are likely involved in the post-translational activation of RuBisCO (Hayashi, et al., 1997, Hayashi, et al., 1999).
43 Chapter 2
References supporting information
Cannon GC, Heinhorst S & Kerfeld CA (2010) Carboxysomal carbonic anhydrases: structure and role in microbial CO2 fixation. (BBA)-Proteins Proteom 1804, 382-392.
Dubbs J & Tabita F (2003) Interactions of the cbbII promoter-operator region with CbbR and RegA (PrrA) regulators indicate distinct mechanisms to control expression of the two cbb operons of Rhodobacter sphaeroides. J Bioll Chem 278, 16443-16450.
Hayashi N, Arai H, Kodama T & Igarashi Y (1999) The cbbQ genes, located downstream of the form I and form II RubisCO genes, affect the activity of both RubisCOs. Biochem Bioph Res Co 265, 177-183.
Hayashi NR, Arai H, Kodama T & Igarashi Y (1997) The novel genes, cbbQ and cbbO, located downstream from the RubisCO genes of Pseudomonas hydrogenothermophila, affect the conformational states and activity of RubisCO. Biochem Biophl Res Co 241, 565-569.
Maddocks SE & Oyston PCF (2008) Structure and function of the LysR-type transcriptional regulator (LTTR) family proteins. Microbiol 154, 3609-3623.
Pfennig N & Trüper HG (1981) Isolation of members of the families Chromatiaceae and Chlorobiaceae. The Prokaryotes 1, 279-289.
Toyoda K, Yoshizawa Y, Arai H, Ishii M & Igarashi Y (2005) The role of two CbbRs in the transcriptional regulation of three ribulose-1,5-bisphosphate carboxylase/oxygenase genes in Hydrogenovibrio marinus strain MH-110. Microbiol 151, 3615-3625.
Widdel F, Kohring GW & Mayer F (1983) Studies on dissimilatory sulfate-reducing bacteria that decompose fatty acids. Arch Microbiol 134, 286-294.
Research paper 1 44
Table SM1. Physical parameters of the Lake Cadagno of the 12th September 2007.
Depth Temp. Conductivity Oxygen Oxygen pH ORP Turbidity [m] [°C] [µS cm-1] % [mg l-1] [mV] [FTU] 0 13.05 110 98.6 9.77 8.16 169.8 0.1 2 12.64 109 77.1 8.62 8.27 166.1 0.2 4 12.56 121 74.9 8.47 8.35 163.7 0.2 6 12.53 120 72.9 8.22 8.35 164.4 0.2 8 11.55 121 69.6 7.89 8.31 167.2 0.2 9 8.69 123 27.8 4.4 8.25 169.7 0.1 10 7.38 150 7.5 0.86 8.01 175.4 0.2 11 6.83 161 3.3 0.4 7.9 184.6 0.3 11.5 6.09 207 1.4 0.18 7.46 -228 14.5 12 5.19 214 1.2 0.15 7.17 -277 24.4 13 4.62 220 0.9 0.12 6.91 -284 9.2 14 4.4 224 0.9 0.11 6.94 -288 5.9 15 4.31 233 0.9 0.12 6.92 -290 5.3 16 4.28 236 0.9 0.12 6.92 -290.6 5.3 17 4.27 245 0.8 0.1 6.92 -291 5.8 18 4.22 248 0.8 0.1 6.92 -291.6 5.4 19 4.21 246 0.8 0.1 6.91 -292.1 5.6 20 4.21 252 1.2 0.1 6.65 -311 9.9
45 Chapter 2
FIG. SM1. Specific FISH counting of GSB (white) and PSB (black) compared to the total prokaryotic cells counted
14 by DAPI (grey) at different depths of Lake Cadagno during the day of the CO2 assimilation analysis from the cultures pre-incubated in dialysis bags (in situ, September 12, 2007). Research paper 1 46
47 Chapter 3 Research paper 2 48
3. RESEARCH PAPER 2
Candidatus “Thiodictyon syntrophicum”, sp. nov., a new purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp.
Nicola Storelli1,5,†, Sandro Peduzzi3,†, Allana Welsh2, Raffaele Peduzzi3, Dittmar Hahn4, Xavier
Perret5 and Mauro Tonolla1,5*
1 Institute of Microbiology Canton Tessin, Via Mirasole 22A, CH-6500 Bellinzona, Switzerland
2 Swedish University of Agricultural Sciences, Uppsala BioCenter, Department of Microbiology,
Box 7025, 750 07 Uppsala, Sweden
3 Alpine Biology Centre Foundation, Piora, CH-6777 Quinto
4 Department of Biology, Texas State University, 601 University Drive, San Marcos, TX 78666,
USA
5 Microbiology Unit, Plant Biology Department, University of Geneva, 30 Quai Ernest-Ansermet
1211 Geneva 4, Switzerland
† Equally contribution.
Published in: Systematic and Applied Microbiology; volume 35, issue 3, (May 2012): 139- 144.
(DOI: 10.1016/j.syapm.2012.01.001). 49 Chapter 3
Systematic and Applied Microbiology 35 (2012) 139–144
Contents lists available at SciVerse ScienceDirect
Systematic and Applied Microbiology
journal homepage: www.elsevier.de/syapm
Candidatus “Thiodictyon syntrophicum”, sp. nov., a new purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp.ଝ
Sandro Peduzzi c,1, Nicola Storelli a,e,1, Allana Welsh b, Raffaele Peduzzi c, Dittmar Hahn d, Xavier Perret e, Mauro Tonolla a,e,∗ a Institute of Microbiology Canton Tessin, CH-6500 Bellinzona, Switzerland b Swedish University of Agricultural Sciences, Uppsala BioCenter, Department of Microbiology, 750 07 Uppsala, Sweden c Alpine Biology Centre Foundation, Piora, CH-6777 Quinto, Switzerland d Department of Biology, Texas State University, 601 University Drive, San Marcos, TX 78666, USA e Microbiology Unit, Plant Biology Department, University of Geneva, 30 Quai Ernest-Ansermet, 1211 Geneva 4, Switzerland article info a b s t r a c t
Article history: Strain Cad16T is a small-celled purple sulfur bacterium (PSB) isolated from the chemocline of crenogenic Received 22 July 2011 meromictic Lake Cadagno, Switzerland. Long term in situ observations showed that Cad16T regularly Received in revised form 11 January 2012 grows in very compact clumps of cells in association with bacteria belonging to the genus Desulfocapsa Accepted 15 January 2012 in a cell-to-cell three dimensional structure. Previously assigned to the genus Lamprocystis, Cad16T, was here reclassified and assigned to the genus Thiodictyon. Based on comparative 16S rRNA gene sequences Keywords: analysis, isolate Cad16T was closely related to Thiodictyon bacillosum DSM234T and Thiodictyon ele- Thiodictyon syntrophicum gans DSM232T with sequence similarities of 99.2% and 98.9%, respectively. Moreover, matrix-assisted Purple bacteria T Meromictic laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) analysis separated Cad16 from other PSB genera, Lamprocystis and Thiocystis. Major differences in cell morphology (oval-sphere compared to rod-shaped) and arrangement (no netlike cell aggregates), carotenoid group (presence of okenone instead of rhodopinal), chemolithotrophic growth as well as the ability to form syntrophic asso- ciations with a sulfate-reducing bacteria of the genus Desulfocapsa suggested a different species within the genus Thiodictyon. This isolate is therefore proposed and described as Candidatus “Thiodictyon syntrophicum” sp. nov., a provisionally novel species within the genus Thiodictyon. © 2012 Elsevier GmbH. All rights reserved.
Introduction with populations related to the genus Lamprocystis, i.e., Lampro- cystis purpurea and Lamprocystis roseopersicina [2], to the genus Members of the family Chromatiaceae were previously iden- Thiocystis, i.e., Thiocystis minor and Thiocystis gelatinosa [3] and tified as the most abundant phototrophic sulfur bacteria in the now to the genus Thiodictyon. Recently two previously uncultured chemocline of meromictic Lake Cadagno, Switzerland (46◦33N, Thiocystis populations, i.e. population 448 and population H, were 8◦43E) [1–4]. Molecular analyses of uncultured populations of this isolated and described as novel species: Thiocystis cadagnonensis family in the chemocline identified all large-celled members as and Thiocystis chemoclinalis [5]. Chromatium okenii, while small-celled members were more diverse The genus Thiodictyon was first described by Winogradsky in 1888 [6] and was emended in 1971 by Pfennig and Trueper [7]. Presently, this genus comprises two validly described species, Thio-
ଝ dictyon elegans and Thiodictyon bacillosum [8]. However, no recent The GenBank/EMBL/DDBJ accession number for the 16S rRNA gene sequences of studies were performed on type strains since they are not available strain Cad16 is AJ511274. The type strain Cad16T is deposited at the Japan Collection of Microorganisms (JCM 15483) and at the Korean Collection for Type Cultures (KCTC from any public culture collection. T 5955). Here, we re-evaluate the assignment of strain Cad16 , iso- ∗ Corresponding author at: Institute of Microbiology Canton Tessin, Via Mirasole lated from Lake Cadagno [9], to the genus Lamprocystis and 22A, CH-6500 Bellinzona, Switzerland. Tel.: +41 0 91 814 60 11; propose it as a provisionally novel species within the genus Thio- fax: +41 0 91 814 60 19. dictyon, Candidatus “Thiodictyon syntrophicum” sp. nov. strain E-mail address: [email protected] (M. Tonolla). T 1 Equal contribution. Cad16 .
0723-2020/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. doi:10.1016/j.syapm.2012.01.001 140 S. Peduzzi et al. / Systematic and Applied Microbiology 35 (2012) 139–144
Materials and methods
Enrichment and cultivation
Samples from the chemocline of Lake Cadagno were taken at a depth of 12.7 m corresponding to the maximum turbidity and highest bacterial density on August 28, 2001. Water samples were used to fill 0.5 L screw-cap glass bottles that were subsequently stored in the dark at 4 ◦C for 10 days. PSB accumulating at the neck of the bottle and under the screw-cap were collected with a pre- viously gassed syringe (N2) and used as inoculums for liquid and deep agar dilutions (1%, v/v) prepared by the Hungate technique [10,11]. Media for enrichments of purple sulfur bacteria were pre- pared according to Widdel and Bak (1992). For detailed information about medium content and isolation see supplementary material S1. Characteristics of isolate Cad16T, deposited at the Japan Cul- ture Collection (JCM 15483) and the Korean Collection for Type Cultures (KCTC 5955), were compared to those published for its closest cultured relatives (Table 1).
Pigment analysis Fig. 1. Phase contrast micrograph of isolate Cad16T.
The in vivo absorption spectra were determined using the alpha-cyano in 33% acetonitrile, 33% ethanol, 33% ddH O and 1% tri- sucrose method and a UV/Vis Spectrometer Lambda 2S (Perkin- 2 fluoroacetic acid). The spotted solution was air-dried for 1–2 min at Elmer, Waltham, MA) following the procedure described by room temperature, and subsequently analysed with an MALDI-TOF Pfenning [12]. Pigments were determined by ion pairing, reverse- MS AximaTM Confidence spectrometer (Shimadzu-Biotech Corp., phase HPLC [13,14]. Kyoto, Japan) as described in the supplementary material S1. 16S rRNA sequence and G+C content analysis Results Nucleic acids were extracted from 1.5 ml of a pure culture at the T exponential phase (OD650 of 0.6) using the MagNA Pure LC auto- Cad16 strain characterization mated extractor (Roche Molecular Biochemicals, Indianapolis, IN, USA) and the DNA isolation extraction kit produced by the same Cells of isolate Cad16T were spherical to oval shaped with a manufacturer. width of 1.4–2.4 m (Table 1, Fig. 1). The isolate grew in single cells 16S rRNA gene fragments were amplified, purified and as well as in compact clumps of cells in liquid media regardless of sequenced as described previously [2]. The sequence was deposited the age of the culture. Under anoxic autotrophic growth conditions, in the EMBL/GenBank databases with accession number AJ511274. the generation time was 121 h. After 2 weeks of growth, Cad16T The G+C content of genomic DNA of isolate Cad16T was deter- reached the optical density of 0.7 corresponding to 107 cell ml−1. mined at the German Collection of Microorganisms and Cell Cells of isolate Cad16T stained Gram-negative, contained gas vac- Cultures (Dr. P. Schumann) by HPLC according to Mesbah et al. [15]. uoles and had a slime capsule. Bright field microscopy revealed the presence of sulfur globules randomly distributed in the cells. The Phyologenetic analysis color of cell suspensions was purple-red, similar to Lamprocystis sp. and Thiocystis sp. but different from T. bacillosum and T. elegans The 16S rRNA sequence was aligned with related sequences that were purple-violet (Table 1). In vivo absorption spectra of pig- from Genbank and EMBL searches [16,17] using Sequencher 4.2.2 ments in cell suspensions of isolate Cad16T displayed an absorption (Gene Codes Corporation, Ann Arbor, MI), CLUSTAL X and Mac- maximum at 528 nm suggesting the presence of okenone as the pre- Clade 4.05 [18,19]. The length of all compared sequences was 1393 dominant carotenoid, while carotenoids of the rhodopinal group, bp. Phylogenetic analyses included maximum parsimony (MP), with rhodopinal and rhodopin as major pigment components, were neighbor joining (NJ) and maximum likelihood (ML) methods in reported for T. elegans or T. bacillosum, respectively [9,21] (Table 1). PAUP*4.0b10 [20]. The presence of okenone in strain Cad16T was confirmed by HPLC and genome analysis [22]. MALDI-TOF MS analysis Further physiological characterization focused on different combinations of electron donors and acceptors that were asep- Two ml of different pure cultures of phototrophic sulfur bacte- tically added to the pure cultures of Cad16T (5 mM final conc.): ria (Fig. 3), were centrifuged, the cell pellets resuspended in 15 l formate, acetate, pyruvate, propionate, butyrate, lactate, fumarate, of ddH2O, and 0.5 l of this suspension (Allochromatium vinosum succinate, malate, fructose, glucose, ethanol, propanol, and glyc- DSM180T, Chlorobium clathratiforme strain 4DE Lake Cadagno, erol. Similar to the other Thiodictyon species, Cad16T grew Chlorobium pheobacteroides strain 1VII D7 Lake Cadagno, Chloro- photolithoautotrophically under anaerobic conditions with sulfide bium tepidum WT2321T, L. purpurea DSM4197T, L. sp. Population and elemental sulfur as electron donors. Growth was also observed A strain A31 Lake Cadagno, L. sp. Population D strain D2V2 Lake with thiosulfate. Globules of sulfur were deposited inside the cells Cadagno, T. cadagnonensis JCM15111T, T. chemoclinalis JCM15112T, as intermediary oxidation products. In the presence of sulfide and T. gelatinosa DSM215T, Thiocystis violascens DSM198T and Candi- bicarbonate, photoassimilation of acetate was observed for Cad16T datus Thiodictyon syntrophicum strain Cad16T Lake Cadagno) was which corresponded to published results for the genus Thiodictyon transferred to FlexiMassTM target wells using a disposable loop, (Table 1). In the presence of carbon dioxide and sulfide, photoas- and overlaid with 1.0 l alpha-cyano matrix solution (CHCA; 40 mg similation of fructose was also observed. Growth stimulation in S. Peduzzi et al. / Systematic and Applied Microbiology 35 (2012) 139–144 141 − , Thiocapsa rosea ; 7, T Sphere 20–35 + 8 1.5–2.0 + + + + + + Pink-rose red + − − nd + Spirilloxanthin − − + 65.3 − 6.7–7.5 − + − + nd nd − DSM217 Sphere 20–35 + 7 2.0–3.0 + + + + + + Pink-rose red + − − nd + Spirilloxanthin − + + 64.3 − 6.7–7.5 − +/ − + − − − − Thiocapsa roseopersicina DSM229; 6, tolerate low concentrations of NaCl. +, substrate utilized or present; Sphere 20–35 + 6 1.2–3.0 + − + + + + Pink-rose red + − − nd +/ − Spirilloxanthin − + +/ − 63.3–66.3 + 7.3 + + + − − − − may which Lamprocystis roseopersicina ; 5, T L. rosepersicina 20–30 + 5 2.0–3.5 + + + nd − + Purple-violet + + nd + nd Rhodopinal nd − nd 63.8 nd 7.0–7.3 nd nd nd nd nd nd nd DSM4197 23–25 + 4 1.9–2.3 + + + + + + Purple-red + − + nd + Okenone − + + 63.5 − 7.0–7.3 − − − + − − − Lamprocystis purpurea ; 4, T Sphere-oval Oval-sphere Sphere 20–25 + 3 1.4–2.4 + + + + + + + − − − − Okenone − + − 67.7 − 7.0–7.2 − − + − − − − reported for taxa 1–8 except for marine strains of ; 3, strain Cad16 T was and related genera. DSM234 Rod 20–30 + 2 1.5–2.0 + + nd + + + Purple-violet Purple-red + − nd nd nd Rhodopinal nd − nd 66.3 nd 6.7–7.3 nd nd nd nd nd nd nd Thiodictyon Thiodictyon bacillosum . Data for taxa 1–2 and 4–8 are from Imhoff (2005) and Pfennig and Trueper (1989). Sulfide and elemental sulfur were utilized as photosynthetic electron donors by all strains, pyruvate T ; 2, T Rod + 1 1.5–2.0 + Netlike formation+ Irregular clumps Clumps of cells Clumps of cells Irregular aggregates Irregular aggregates Irregular aggregates Irregular aggregates nd + + + + − nd nd nd Rhodopinal nd − nd 65.3–66.3 nd 6.7–7.3 nd nd nd nd nd nd nd DSM236 DSM232 C) 20–25 ◦ in the presence of sulfide and bicarbonate Thiocapsa pendens assimilation ; 8, T Thiodictyon elegans Sulfur Sulfide Thiosulfate Acetate Pyruvate Formate Propionate Butyrate Lactate Fumarate Succinate Malate Fructose Glucose Ethanol Propanol Glycerol Characteristic Shape Size ( m) Photosynthetic electron donors (sulfur compounds) Gas vacuoles Aggregate formation Sulfur storage Substrate Slime capsule Color of cell suspension Purple-violet Motility Flagellation Carotenoid group Chemolithotrophic growth mol% G+C of DNA pH optimum Temperature optimum ( substrate not utilized or absent; +/ − , variable depending on the strain; nd, not determined. and acetate were photoassimilated by all strains. No salt requirement DSM235 Table 1 Differentiating characteristics of species of the genus Taxa: 1, 142 S. Peduzzi et al. / Systematic and Applied Microbiology 35 (2012) 139–144
Isolate Cad16T (AM086642)
100(100,100,100) Thiodictyon elegans DSM 232T (EF999973)
Thiodictyon bacillosum DSM 234T (EF999974)
Lamprocystis purpurea DSM 4197 (AJ223235) 84(91,73,99) *
Lamprocystis roseopersicina DSM 229 (AJ006063)
Thiocapsa rosea DSM 235 (AJ006062) *
Thiocapsa roseopersicina DSM 217T (Y12364)
98(100,99,100) Thiocapsa marina 5812 (Y12302)
ATCC 700894 (AJ242772) * Thiocapsa litoralis
Thiocapsa pendens DSM 236T (AJ002797)
Thiocystis violascens DSM 198T (AJ224438) *
88(98,95,100) Thiocystis gelatinosa DSM 215T (Y11317)
Thiocystis minor DSM 178T (Y12372)
Chromatium okenii DSM 169 (AJ223234)
Thiocystis violacea DSM 207T (Y11315)
Allochromatium vinosum DSM 180T (M26629) *
Allochromatium minutissimumgi DSM 1376T (Y12369)
ATCC BAA-1228T (AJ971090) * Thiorhodococcus mannitoliphagus
Thiorhodococcus minor DSM 11518 (Y11316) 96(100,92,100) Thiorhodococcus drewsii DSM 15006 (AF525306)
Halorhodospira halophila SL1 (CP000544) 0.01 substitutions/site
Fig. 2. Maximum likelihood tree topology from 16S rRNA gene sequences for isolate Cad16T and other closely related species of the family Chromatiaceae created using PAUP*4.0b10 and a GTR model of sequence evolution [20]. Numbers reflect bootstrap support (BS) measures generated in PAUP and only include those measures over 70%. Numbers in parentheses reflect BS measures from neighbor joining and maximum parsimony analyses in PAUP and Bayesian posterior probabilities (PP) created using MRBAYES version 3.0 [27], respectively. The outgroup was specified as strain H. halophila (CP000544). T: type strain. the presence of pyruvate was clearly observed at 2.5 mM, however, for each of the phylogenetic methods employed. Isolate Cad16T had growth was only slightly promoted at 5 mM and not at all at 1 mM. 100.0% and 99.9% 16S rRNA gene sequence identity (1393 bp), to Chemolithoautotrophic growth was obtained both with hydro- clone 371 (AJ006061) and clone 335 (AJ006059), respectively [2,3]. gen sulfide (0.02%) and thiosulfate (0.07%) or with sulfide alone The closest relatives of isolate Cad16T were T. bacillosum DSM234T (0.07%) in the dark, with a micro-oxic headspace atmosphere (5% with 99.2% and T. elegans DSM232T with 98.9% sequence similarity O2, 10% CO2 and 85% N2) as suggested by Kämpf and Pfennig [23] (Fig. 2). 16S rRNA gene sequence similarity between T. bacillo- in deep agar shake cultures. Unlike Cad16T both T. bacillosum and sum and T. elegans was found to be higher (99.4%). 16S rRNA gene T. elegans were not able to grow under chemolithoautotrophic con- sequences of T. elegans (EF999973) and T. bacillosum (EF999974) ditions (Table 1.) were deposited in the EMBL/GenBank databases in 2007, and were thus not included in the recent reclassification of species belonging Phylogenetic analysis to the Chromatiaceae.
A representative ML tree with BS support values for ML, NJ, G+C content and MP analyses and posterior probability values (PP) for Bayesian analysis shows that isolate Cad16T clustered with high support The genomic G+C content of isolate Cad16T was 67.7% which with representative strains of the genus Thiodictyon within the ␥- is notably different but in the same range as values reported for subdivision of Proteobacteria (Fig. 2). Tree topologies were identical T. bacillosum and T. elegans (66.3% and 65.3%, respectively), and S. Peduzzi et al. / Systematic and Applied Microbiology 35 (2012) 139–144 143
Fig. 3. MALDI-TOF MS dendrogram of 12 strains of phototrophic sulfur bacteria, resulting from single-link clustering analysis (SARAMISTM database software). Error 0.08%; range of m/y from 2000–20,000. T: type strain. more distant to values for the genera Lamprocystis (63.4–64.1%) Our data based on morphological and physiological traits as and Thiocapsa (63.3–66.3%). well as the ecological relevance of the isolated strain Cad16T support the description of a provisionally novel species within MALDI-TOF MS analysis the genus Thiodictyon. Our isolate is therefore proposed and described as Candidatus “Thiodictyon syntrophicum” sp. nov. strain All available strains of phototrophic sulfur bacteria, including Cad16T, a provisionally novel species within the genus Thiodictyon 7 strains previously isolated from Lake Cadagno, were analyzed (syn.tro’phi.cum. Gr. pref. syn, together with; Gr. adj. trophikos, nurs- by MALDI-TOF MS (Fig. 3). Unfortunately, no other Thiodictyon ing, tending or feeding; N.L. neut. adj. syntrophicum, syntrophic). strains were available. All Lamprocystis species, including L. pur- Due to the syntrophic association and cell-to-cell aggregation with purea DSM4450T and L. sp. population D strain D2V2, clustered a sulfate-reducing and sulfur disproportionating bacteria, Desulfo- together and distant from Cad16T. capsa sp., observed in mixed culture and in natural environment [3,9], see also the supplementary material S2. Discussion Acknowledgements The 16S rRNA sequence similarity of Cad16T to the genus Thio- dictyon and the clear separation from the genus Lamprocystis by The authors wish to thank N. Ruggeri, A. Caminada, C. Benagli MALDI-TOF MS analysis, as well as others molecular and physi- and S. De Respinis. This work was supported by grants from ological differences confirm the classification of strain Cad16T as the Swiss National Science Foundation (NF31-46855.96), the a member of the genus Thiodictyon, and support the description US National Science Foundation (NSF GK-12 grant 0742306), of a new Thiodictyon species, i.e. Candidatus “Thiodictyon syn- the canton of Ticino and the Alpine Biology Center Foundation trophicum” sp. nov. strain Cad16T. Moreover, our data support the (Switzerland). We acknowledge Dr. P. Schumann (DSMZ, Braun- morphology-based classification proposed by Pfennig and Trüper schweig) for the analysis of the mol% G+C content of the DNA, [8], and the phylogenetic position of the genus Thiodictyon related Dr. A. Lami (CNR Istituto Italiano di Idrobiologia) for ion pairing to the genera Lamprocystis, Thiocystis and Thiocapsa (Imhoff, 2005) reversed-phase HPLC analysis, and Dr. J. Euzéby for help on the was confirmed in this study. nomenclature. Further characterization using the criteria recommended by Imhoff and Caumette [24] showed new characteristics within this Appendix A. Supplementary data genus: chemolithoautotrophic growth under micro-oxic condi- tions in the dark and the presence of okenone was not found in Supplementary data associated with this article can be found, in T the other two Thiodictyon species from which Cad16 differed in the online version, at doi:10.1016/j.syapm.2012.01.001. cell morphology and cell arrangement. Unfortunately, only a lim- ited data set was available for the 2 described Thiodictyon species, References and thus some specific characteristics of isolate Cad16T could not be compared to those of the described species (e.g. its abil- [1] Decristophoris, P.M.A., et al. (2009) Fine scale analysis of shifts in bacterial com- ity to grow on thiosulfate which has not been assessed for the munity structure in the chemocline of meromictic Lake Cadagno, Switzerland. described species so far) (Table 1). Because T. elegans DSM232T and J. Limnol. 68 (1), 16–24. [2] Tonolla, M., et al. (1999) In situ analysis of phototrophic sulfur bacteria in the T T. bacillosum DSM234 were not available from any official culture chemocline of meromictic Lake Cadagno (Switzerland). Appl. Environ. Micro- collection, characteristics of isolate Cad16T could only be compared biol. 65 (3), 1325–1330. to data retrieved from previous publications [6,25] and to other [3] Tonolla, M., Peduzzi, R., Hahn, D. (2005) Long-term population dynamics of phototrophic sulfur bacteria in the chemocline of Lake Cadagno, Switzerland. type strains belonging to the genera Lamprocystis and Thiocapsa Appl. Environ. Microbiol. 71 (7), 3544–3550. [25,26]. Another characteristic of Cad16T is that in its natural habi- [4] Tonolla, M., et al. (1998) Microscopic and molecular in situ characterization of tat it is frequently observed in association with sulfate-reducing bacterial populations in the meromictic lake Cadagno. Doc. Istit. Ital. Idrobiol. 63, 31–44. bacteria related to Desulfocapsa thiozymogenes in a cell-to-cell con- [5] Peduzzi, S., et al. (2011) Thiocystis chemoclinalis sp. nov. and Thiocystis tact three dimensional structure [9]. cadagnonensis sp. nov., two new motile purple sulfur bacteria isolated from 144 S. Peduzzi et al. / Systematic and Applied Microbiology 35 (2012) 139–144
the chemocline of meromictic lake Cadagno, Switzerland. Int. J. Syst. Evol. [16] Altschul, S., et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of Microbiol. 61 (7), 1682–1687. protein database search programs. Nucleic Acids Res. 25 (17), 3389–3402. [6] Winogradsky, S. (1888) Beitrage zur Morphologie und Physiologie der Bakte- [17] Pearson, W.R., Lipman, D.J. (1988) Improved tools for biological sequence com- rien. Zur Morphologie and Physiologie der Schwefelbakterien. A. Felix, Leipzig. parison. Proc. Natl. Acad. Sci. USA 85 (8), 2444–2448. Microbiologie du sol. Massonet Cie, Paris, vol. 1. [18] Maddison, W.P., Maddison, D.R. 1992 MacClade: Analysis of Phylogeny and [7] Pfennig, N., Truper, H.G. (1971) Higher taxa of the phototrophic bacteria. Int. J. Character Evolution, Version 3.0, Sinauer, Sunderland, MA. Syst. Evol. Microbiol. 21 (1), 17–18. [19] Thompson, J.D., et al. (1997) The CLUSTAL X windows interface: flexible strate- [8] Pfennig, N., Trüper, H.G. 1989 Anoxygenic phototrophic bacteria Bergey’s Man- gies for multiple sequence alignment aided by quality analysis tools. Nucleic ual of Systematic Bacteriology, vol. 3, pp. 1635–1709. Acids Res. 25 (24), 4876. [9] Peduzzi, S., Tonolla, M., Hahn, D. (2003) Isolation and characterization [20] Swofford, D.L. 2003 PAUP*. Phylogenetic Analysis Using Parsimony (* and other of aggregate-forming sulfate-reducing and purple sulfur bacteria from the methods). Version 4, Sinauer Associates, Sunderland, MA. chemocline of meromictic Lake Cadagno, Switzerland. FEMS Microbiol. Ecol. [21] Pfennig, N., Markham, M.C., Liaaen-Jensen, S. (1968) Carotenoids of Thiorho- 45 (1), 29–37. daceae. Arch. Microbiol. 62 (2), 178–191. [10] Widdel, F., Bak, F. (1992) Gram-negative mesophilic sulfate-reducing bacteria. [22] Vogl, K., Bryant, D.A. (2011) Elucidation of the biosynthetic pathway for Prokaryotes 4, 3352–3378. Okenone in Thiodictyon sp. Cad16 leads to the discovery of two novel carotene [11] Pfennig, N. (1978) Rhodocyclus purpureus gen. nov. and sp. nov., a ring-shaped, ketolases. J. Biol. Chem. 286 (44), 38521–38532. vitamin B12-requiring member of the family Rhodospirillaceae. Int. J. Syst. Evol. [23] Kämpf, C., Pfennig, N. (1980) Capacity of Chromatiaceae for chemotrophic Microbiol. 28 (2), 283–288. growth. Specific respiration rates of Thiocystis violacea and Chromatium [12] Pfennig, N. (1974) Rhodopseudomonas globiformis, sp. n., a new species of the vinosum. Arch. Microbiol. 127 (2), 125–135. Rhodospirillaceae. Arch. Microbiol. 100 (1), 197–206. [24] Imhoff, J.F., Caumette, P. (2004) Recommended standards for the description [13] Mantoura, R.F.C., Llewellyn, C.A. (1983) The rapid determination of algal chloro- of new species of anoxygenic phototrophic bacteria. Int. J. Syst. Evol. Microbiol. phyll and carotenoid pigments and their breakdown products in natural waters 54 (4), 1415–1421. by reverse-phase high-performance liquid chromatography. Anal. Chim. Acta [25] Brenner, D.J., et al. 2005 Anoxygenic phototrophic purple bacteria. In: Bergey’s 151, 297–314. Manual® of Systematic Bacteriology, Springer, US, pp. 119–132. [14] Hurley, J.P. (1988) Analysis of aquatic pigments by high performance liquid [26] Eichler, B., Pfennig, N. (1988) A new purple sulfur bacterium from stratified chromatography. J. Anal. Purif. 3, 12–16. freshwater lakes. Amoebobacter purpureus sp. nov. Arch. Microbiol. 149 (5), [15] Mesbah, M., Premachandran, U., Whitman, W.B. (1989) Precise measurement 395–400. of the G+C content of deoxyribonucleic acid by high-performance liquid chro- [27] Huelsenbeck, J.P., Ronquist, F. (2001) MRBAYES: Bayesian inference of phylo- matography. Int. J. Syst. Evol. Microbiol. 39 (2), 159–167. genetic trees. Bioinformatics 17 (8), 754–755. Research paper 2 56 57 Chapter 3 Research paper 2 58
3.1. Supplementary material S1 (Material and methods)
Enrichment and cultivation of strain Cad16T
The medium contained (L-1): 0.25 g KH2PO4, 0.34 g NH4Cl, 0.5 g MgSO4 x 7H2O, 0.25 g CaCl2 x 2H2O, 0.34 g KCl, 1.5 g NaHCO3, 0.5 ml trace element solution SL10, and 0.02 mg vitamin B12 (1). The medium was reduced with 0.3 g L-1 Na2S x 9H2O (1.10 mM final conc.) and adjusted to a pH around 7.2. Acetate (2 mM) was added to pure cultures of phototrophic bacteria, as carbon source, to promote growth of isolated strains. Stock solutions of inorganic sulfur compounds, i.e. thiosulfate, sulfide and elemental sulfur, were prepared according to Janssen et al. (1996) (6). Final concentrations in culture were 10 mM for thiosulfate, 2 mM for sulfide and 20-30 mg ml-1 for elemental sulfur (2). Chemolithoautotrophic growth was tested as described by Kämpf & Pfennig (7) in uniformly inoculated deep agar shake cultures.
All cultures were incubated at room temperature (20-23°C). Dilution series of these cultures were subjected to a photoperiod (6h light/6h dark) with low light intensities generated with an incandescent 40W bulb placed at a distance of 60 cm from the cultures (1).
Phyologenetic analysis with 16S rRNA
NJ and ML analyses utilized a general-time-reversible model of sequence evolution. MP and ML analyses included a full heuristic search with 10,000 and 200 random addition sequence replicates, respectively. Confidence in tree topologies was gauged using bootstrap re-sampling methods (BS) in PAUP and only included those values over 70% with 10,000 replicates for MP and NJ and 200 replicates for ML methods (3). Additionally, Bayesian methods were used in MRBAYES v 3.0 (4) with 2 million generations sampled every 1000 trees and a 95% majority rule consensus tree generated in PAUP.
MALDI-TOF MS analysis
MS analyses were performed in positive linear mode in the range of 2,000– 20,000 mass-to- charge ratio (m/z) with delayed, positive ion extraction (delay time: 104 ns with a scale factor of 800) and an acceleration voltage of 20 kV. For every sample, 2×50 averaged profile spectra were stored and used for analysis. All spectra were processed by the MALDI MS Launchpad 2.8 59 Chapter 3 software (Shimadzu Biotech) with baseline correction, peak filtering and smoothing. A minimum of 20 laser shots per sample were used to generate each ion spectrum. For each bacterial sample, 50 protein mass fingerprints were averaged and processed. Spectra were analyzed using SARAMISTM (Spectral Archive and Microbial Identification System, AnagnosTecGmbH) at default settings. Dendrograms were based on the peak patterns of all analyzed strains submitted to single-link clustering analysis using SARAMISTM with an error of 0.08% and a m/z range of 2000 to 20,000 Daltons.
Pigment analysis
Specific pigments were determined by ion pairing, reverse-phase HPLC (modified from Mauntoura & Llewellyn, 1983 (8), and Hurley, 1988 (5)). The ion pairing (tetrabutyl ammonium phosphate 10-3 M) allows for greater resolution of the dephytolated acidic chloropigments (Chl c, chlorophyllide a, and pheophorbide a). The equipment employed consisted of a gradient pumping system and dual channel variable wavelength UV-VIS detector (set at 460 nm and 665 nm for carotenoids and chloropigments, respectively) controlled by a computer (Summit, Dionex). An auto-sampler for sample injection was connected through a precolumn to a reverse- phase C18 ODS column (OMNISPHERE C18: 5 µm particle size; 250 mm x 4.6 mm i.d.). After sample injection (100 µl), a gradient program that ramped from 85% mobile-phase A (80:20, by vol. methanol: aqueous solution of 0.001 M ion-pairing and 0.001 M propionic acid) to 100% mobile-phase B (60:40, acetone: methanol) in 30 min with a hold for 20 min provided sufficient resolution of all pigments of interest. Flow rates from 1 ml min-1 to 2 ml min-1. The column was re-equilibrated between samples by linear ramping to 85% mobile-phase A for 5 min and maintenance for 10 min before sample injection.
Research paper 2 60
3.2. Supplementary material S2
Description of Candidatus “Thiodictyon synthrophicum” sp. nov. strain Cad16T.
Thiodictyon syntrophicum (syn.tro'phi.cum. Gr. pref. syn, together with; Gr. adj. trophikos, nursing, tending or feeding; N.L. neut. adj. syntrophicum, syntrophic. Due to the syntrophic association and cell-to-cell aggregation with a sulfate-reducing and disproportionating bacteria, Desulfocapsa sp., observed in mixed culture and in natural environment).
Cells are oval to spherical, measuring about 1.4. to 2.4 μm, non-motile, Gram negative and divide by binary fission. Occur as single cells and can form irregular aggregates of variable size with up to about 100 cells. Contain gas vacuoles and develop a slime capsule. Salt is not required for growth. Color of cell suspension is purple-red, in liquid cultures cells accumulate at the interface between liquid and gas headspace and on vessel glass-walls, forming a typical ring- shaped purple-red mat. Photosynthetic pigments are bacteriochlorophyll a and carotenoids of the okenone group are present.
Photolithoautotrophic growth occurs under anaerobic conditions with hydrogen sulfide, thiosulfate and elemental sulfur as electron donors. Globules of sulfur are deposited inside the cells as intermediary oxidation products. In the presence of carbon dioxide and sulfide, acetate, pyruvate and fructose are photo-assimilated. Chemolithoautotrophicgrowth with hydrogen sulfide or thiosulfate under micro-oxic conditions in the dark.
Growth occurs in a temperature range from 5-25°C with an optimum at 20-23°C, and a pH range from 6.8 -7.5.
Habitat: freshwater stratified sulfidic environments such as meromictic lakes, thrives at the boundary layer, chemocline, between the oxic upper layer and the anoxic and sulfidic layer. Along physicochemical gradients can show sharp stratification. Can form very compact clump of cells, of variable size, in association with a Desulfocapsa sp. in cell to cell contact [2, 9].
The G+C content of the genomic DNA is 67.7 mol%.
The type strain is Cad16T (=JCM 15483 = KCTC 5955), isolated from the chemocline of the meromictic alpine Lake Cadagno in the Southern Swiss Alps at 1923 m above sea level (46°33'N, 8°43'E). 16S rRNA gene sequence deposited in the EMBL/GenBank databases with accession number AJ511274.
61 Chapter 3
References supporting information
1. Eichler, B., and N. Pfennig. (1988). A new purple sulfur bacterium from stratified freshwater lakes, Amoebobacter purpureus sp. nov. Arch of Microbiol 149, 395-400.
2. Finster, K., W. Liesack, and B. Thamdrup. (1998). Elemental sulfur and thiosulfate disproportionation by Desulfocapsa sulfoexigens sp. nov., a new anaerobic bacterium isolated from marine surface sediment. Appl Environm Microbiol 64, 119-125.
3. Hillis, D. M., and J. J. Bull. (1993). An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst Biol 42, 182-192.
4. Huelsenbeck, J. P., and F. Ronquist. (2001). MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754-755.
5. Hurley, J. P. (1988). Analysis of aquatic pigments by high performance liquid chromatography. J. Anal. Purif 3, 12-16.
6. Janssen, P. H., A. Schuhmann, F. Bak, and W. Liesack. (1996). Disproportionation of inorganic sulfur compounds by the sulfate-reducing bacterium Desulfocapsa thiozymogenes gen. nov., sp. nov. Arch of Microbil 166, 184-192.
7. Kämpf, C., and N. Pfennig. (1980). Capacity of Chromatiaceae for chemotrophic growth. specific respiration rates of Thiocystis violacea and Chromatium vinosum. Arch of Microbiol 127, 125-135.
8. Mantoura, R. F. C., and C. A. Llewellyn. (1983). The rapid determination of algal chlorophyll and carotenoid pigments and their breakdown products in natural waters by reverse- phase high-performance liquid chromatography. Analytica Chimica Acta 151, 297-314. Research paper 2 62
63 Chapter 4 Research paper 3 64
4. RESEARCH PAPER 3
Proteomic analysis of the purple sulfur bacterium Candidatus “Thiodictyon syntrophicum” strain Cad16T isolated from Lake Cadagno
Nicola Storelli1,2, Maged M. Saad1, Niels-Ulrik Frigaard3, Xavier Perret1 and Mauro Tonolla1,2,4 *
1 University of Geneva, Sciences III, Department of Botany and Plant Biology, Microbiology Unit, 30 quai Ernest-Ansermet, CH-1211 Geneva 4, Switzerland.
2 Institute of Microbiology, via Mirasole 22a, CH-6500 Bellinzona.
3 Section for Marine Biology, Department of Biology, University of Copenhagen, DK-3000 Helsingør, Denmark.
4 Alpine Biology Center (ABC), Foundation Piora Valley, CH-6777 Quinto, Switzerland.
Published in: EuPA Open Proteomics; Volume 2 (March 2014), pages 17-30.
DOI: http://dx.doi.org/10.1016/j.euprot.2013.11.010. 65 Chapter 4
e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
Available online at www.sciencedirect.com ScienceDirect
journal homepage: http://www.elsevier.com/locate/euprot
Proteomic analysis of the purple sulfur bacterium Candidatus “Thiodictyon syntrophicum” strain Cad16T isolated from Lake Cadagnoଝ
Nicola Storelli a,b,∗, Maged M. Saad a, Niels-Ulrik Frigaard c, Xavier Perret a, Mauro Tonolla a,b,d a University of Geneva, Sciences III, Department of Botany and Plant Biology, Microbiology Unit, 30 quai Ernest-Ansermet, CH-1211 Geneva 4, Switzerland b Laboratory of Applied Microbiology, via Mirasole 22a, CH-6500 Bellinzona, Switzerland c Section for Marine Biology, Department of Biology, University of Copenhagen, DK-3000 Helsingør, Denmark d Alpine Biology Center (ABC), Foundation Piora Valley, CH-6777 Quinto, Switzerland article info a b s t r a c t
Article history: Lake Cadagno is characterised by a compact chemocline with high concentrations of purple Received 14 August 2013 sulfur bacteria (PSB). 2D-DIGE was used to monitor the global changes in the proteome of Received in revised form Candidatus “Thiodictyon syntrophicum” strain Cad16T both in the presence and absence
23 October 2013 of light. This study aimed to disclose details regarding the dark CO2 assimilation of the Accepted 15 November 2013 PSB, as this mechanism is often observed but is not yet sufficiently understood. Our results showed the presence of 17 protein spots that were more abundant in the dark, including Keywords: three enzymes that could be part of the autotrophic dicarboxylate/4-hydroxybutyrate cycle, Phototrophic sulfur bacteria normally observed in archaea.
Dark CO2 assimilation © 2013 The Authors. Published by Elsevier B.V. on behalf of European Proteomics Granules of poly(3-hydroxybutyrate) Association (EuPA). All rights reserved. (PHB) Dicarboxylate/4-hydroxybutyrate (DC/HB) cycle
oxygen, sulfide, and light; and a turbidity maximum that 1. Introduction correlates with a dense community of phototrophic sulfur bacteria (107 cells ml−1) [1–3]. This community is composed Lake Cadagno is a crenogenic meromictic lake located in the of species belonging to two distinct groups: the purple sul- Piora valley at 1921 m above sea level in the southern Swiss fur bacteria (PSB; family Chromatiaceae) and the green sulfur ◦ ◦ Alps (46 33 N, 8 43 E). This lake is characterised by a nar- bacteria (GSB; family Chlorobiaceae) [4]. Although both groups row chemocline found at a depth of approximately 12 m that oxidise sulfur compounds for anoxygenic photosynthesis, contains high concentrations of sulfates; steep gradients of they also exhibit three major structural and/or metabolic
ଝ This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. ∗ Corresponding author at: Via Mirasole 22a, CH-6500 Bellinzona, Switzerland. Tel.: +41 91 814 60 12; fax: +41 91 814 60 39. E-mail addresses: [email protected], [email protected] (N. Storelli). 2212-9685/$ – see front matter © 2013 The Authors. Published by Elsevier B.V. on behalf of European Proteomics Association (EuPA). All rights reserved. http://dx.doi.org/10.1016/j.euprot.2013.11.010 18 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
differences [5]. First (1) the pigments composition (BChl and followed by MALDI-TOF mass spectrometry identification carotenoids) that allows a different structure of the light- using the 7,3-Mbp draft genome of strain Cad16T as database. harvesting antenna and a different type of photosynthetic reaction centre. In GSB pigments are organised into high 2. Materials and methods structured organelles known as chlorosomes [6,7], while two light-harvesting complexes (LH1 and LH2) are observed in PSB 2.1. Media and growth conditions [8]. The second main difference (2) is in the CO2 fixation path- way, PSB normally fix CO2 using the Calvin–Benson–Bassham Candidatus “T. syntrophicum” strain Cad16T [18] was grown cycle (CBB cycle) while all GSB use the reductive reverse tri- in Pfennig’s Medium I [30]: 0.25 g of KH PO l−1, 0.34 g of carboxylic acid (rTCA) cycle [9,10]. The last but not the least 2 4 NH Cl l−1, 0.5 g of MgSO ·7H O l−1, 0.25 g of CaCl ·2H O l−1, difference (3) is in the deposition of sulfur globules inside or 4 4 2 2 2 0.34 g of KCl l−1, 1.5 g of NaHCO l−1, 0.02 mg of vitamin B l−1 outside the cell, for PSB or GSB respectively [11]. 3 12 and 0.5 ml of trace element solution SL12 l−1. The medium was Phototrophic sulfur bacteria are important for the primary prepared in 2 l bottles using a flushing gas composed of 80% production in many stratified lakes and have been observed N2 and 20% CO2 according to Widdel and Bak [31] and was to contribute with value as high as 80% of the total CO2 fix- reduced by the addition of 1.10 mM Na S·9H O and adjusted ation in some meromictic lakes [12]. In Lake Cadagno, the 2 2 to a pH of approximately 7.0. Cultures were incubated at room chemocline’s contribution to the total primary production temperature (20–23 ◦C) and subjected to cycles of 12 h of light has been estimated to reach 40% despite its small volume followed from 12 h of dark until the cultures reach the density (approximately 10% of the total lake volume). Interestingly, of approximately 107 cells per ml−1 (about 10 days). The light high rates of CO2 assimilation have also been recorded in the intensity was set up at 6 E m−2 s−1 generated with incandes- dark [13]. However, the mechanism of dark CO2 fixation, espe- cent 60 W bulbs. Bacterial growth was followed by measuring cially from PSB, remains largely unknown [14,15]. Recently, the the optical density of culture aliquots at 650 nm using a UV/VIS rates of CO2 assimilation of the most abundant phototrophic Spectrometer Lambda 2S (Perkin-Elmer Inc, Waltham, MA, sulfur bacteria from the chemocline in Lake Cadagno were USA). Concentrations of sulfide in the cultures were measured measured using both nano-scale secondary-ion mass spec- 14 daily and adjusted to about 1.00 mM throughout the experi- trometry (nanoSIMS) [16] and CO2 quantitative assimilation ments. in dialysis bags [17]. According to these studies, the strongest The differential protein expressions of Candidatus “T. syn- CO2 assimilators in the light and in the dark are populations trophicum” strain Cad16T at autotrophic growth conditions of the large-celled PSB Chromatium okenii and the small-celled in the light and in the dark were investigated. How say PSB Candidatus “Thiodictyon syntrophicum” respectively [18]. above, bacterial cells were exposed to a photoperiod of 12 h Moreover, the PSB Candidatus “T. syntrophicum” population is of light (07:00–19:00) followed by 12 h of dark (19:00–07:00) for also a strong CO2 assimilator in the presence of light. Although approximately 10 days until they reached an optical density it only represents approximately 2% of the total chemocline’s of approximately OD = 0.6 corresponding to approximately bacterial population, Candidatus “T. syntrophicum” appears to 650 107 cells per ml−1. Then, the total proteins were extracted after play a key role in CO2 fixation in Lake Cadagno, both in the 4 h of light (at 11:00) and again 12 h later after 4 h of dark (at presence and absence of light [17]. 23:00) always following the cycles of light and dark of 12 h each. The elucidation of cellular metabolisms in response to Three independent cultures were used to ensure the biologi- different environmental conditions requires the use of a com- cal reproducibility of the experiment, and prior to the 2D-DIGE bination of different techniques that can record metabolic experiment, each protein extract was previously analysed on adaptations under different environmental conditions. Com- silver-stained 2-DE gels in triplicate. parative proteomics allows a global analysis of differentially expressed functional and regulatory protein [19–23]. Tw o - dimensional polyacrylamide gel electrophoresis (2D-PAGE) is 2.2. Total cell count generally applied for separating large numbers of proteins and measuring their differential expression levels by com- Cell concentrations of pure bacterial cultures were deter- paring spot intensities [24,25]. Two-dimensional difference gel mined using samples fixed with 4% formaldehyde (final electrophoresis (2D-DIGE) has been implemented as a pow- concentration) and stained with 0.001% (w/v) 4 ,6-diamidino- erful alternative to conventional 2D-PAGE for quantitative 2-phenylindole (DAPI) (final concentration). Ten l of each expression analysis [26,27]. Proteins from 2D-PAGE or 2D- fixed and stained sample were deposited onto polycarbonate DIGE are commonly identified by peptide mass fingerprinting filters as described in Hobbie et al. [32] and observed at 100-fold (PMF) using matrix-assisted laser desorption ionisation-time- magnification using an epifluorescence microscope (Axiolab, of-flight mass spectrometry (MALDI-TOF MS) [28,29]. Zeiss Germany) and the filter set F31 (Zeiss, Germany). Twenty fields of 0.01 mm2 were counted, and cell densities were The major aim of this study was to elucidate the CO2 −1 ± assimilation processes of the PSB Candidatus “Thiodictyon expressed as the mean number of cells ml ( standard error). syntrophicum” strain Cad16T in light and especially in the dark by proteomic analysis. The total proteins were extracted 2.3. Protein extraction in the light and in the dark from bacterial suspensions that were incubated with a photoperiod of 12 h of light fol- The total proteins were extracted from cells from expo- lowed by 12 h of dark during approximately 10 days. Sample nentially growing cultures with an OD650 of ca. 0.6, which − extracts from the light and dark were compared using 2D-DIGE corresponds to approximately 1.0 × 107 cells per ml 1. The e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 19
Each combined labelled sample (150 g of protein) was Table 1 – CyDye labelling scheme and gel setup for 2D-DiGE analysis. resuspended in solubilisation buffer (SB: 7 M urea, 2 M thiourea, 30 mM Tris–HCl pH 8.5, 4% (w/v) CHAPS, 40 mM DTT, Gel # CyDye Sample type 1% (v/v) 3–10 NL IPG buffer, 0.002% (w/v) bromophenol blue) 1 Cy3 Light 1 (50 g) prior to isoelectric focusing (IEF). IEF separation, was per- Cy5 Dark 1 (50 g) formed in 24 cm DryStrips with a nonlinear pH 3–10 gradient Cy2 Internal standard (50 g) (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). The Plu- 2 Cy3 Dark 1 (50 g) sOne DryStrip Cover fluid was used to fill the 24 cm ceramic Cy5 Light 1 (50 g) holders (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) Cy2 Internal standard (50 g) containing the DryStrips, which were then introduced into 3 Cy3 Light 2 (50 g) an Ettan®IPGphor 3 (GE Healthcare Bio-Sciences AB, Upp- Cy5 Dark 2 (50 g) sala, Sweden). After 18 h of strip rehydration at 18 ◦C, the Cy2 Internal standard (50 g) ◦ IEF programme was performed at 18 C under the follow- 4 Cy3 Dark 2 (50 g) ing steps: 300 V for 4 h, 300–1000 V gradient for 70 min, Cy5 Light 2 (50 g) 3000 V for 4 h, 3000–6000 V gradient for 2 h, 6000 V for 1 h, Cy2 Internal standard (50 g) 6000–8000 V gradient for 2 h, and 8000 V until a total volt- 5 Cy3 Light 3 (50 g) age of 120,000 V h was reached. The isoelectrofocused strips Cy5 Dark 3 (50 g) were incubated in equilibration solution (ES: 0.5 M Tris–HCl, Cy2 Internal standard (50 g) pH 8.5, 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS) for 15 min 6 Cy3 Dark 3 (50 g) with the addition of 2.0% (w/v) DTT, then for 10 min in Cy5 Light 3 (50 g) ES with the addition of 2.5% (w/v) iodoacetamide. The sec- Cy2 Internal standard (50 g) ond dimension of separation, SDS–PAGE, was performed in an EttanTM DALTsix Electrophoresis System (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) following the manufac- liquid bacterial cultures (500 ml) were centrifuged for 15 min turer’s instructions for overnight migration (80 V, 10 mA per × at 15,000 g, and the pellets were washed thrice with PBS gel). The isoelectrofocused strips were applied onto a 12% × 1 and twice with 0.5 M Tris–HCl, pH 6.8. The washed pel- polyacrylamide gel casted in low-fluorescence glass plates lets were resuspended in 2 ml of lysis buffer (7 M urea, 2 M for EttanTM DALT using a DALTsix Gel Caster (GE Health- thiourea, 30 mM Tris–HCl, pH 8.6, 4% (w/v) CHAPS) and were care Bio-Sciences AB, Uppsala, Sweden). Immediately after − ◦ incubated overnight at 20 C. The samples were sonica- electrophoresis, the gels were scanned in an EttanTM DIGE ted (SONOPULS HD 2070, Bandelin Electronics, Germany) for Imager (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) 5 cycles of 15 s each at 25% of the maximal power, with a following the manufacturer’s instructions and default param- pause of 5 min on ice between each cycle, and then cen- eters. The differences in the protein expression levels were × ◦ trifuged for 15 min at 20,000 g at 4 C. The supernatants were analysed using the Melanie© 7.0 software [Geneva Bioin- − ◦ washed twice with acetone for 60 min at 20 C to eliminate formatics (GeneBio) SA, Geneva, Switzerland]. The Cy3/Cy5 all pigments that could interfere with the dyes during the 2D- 2D-DIGE image overlays were obtained for bi-fluorescence by × DIGE experiment. After centrifugation for 10 min at 15,000 g, processing the individual scanned images with ImageJ soft- the protein concentrations present in the pigment-free super- ware (http://rsbweb.nih.gov/ij/). natants were measured by Bradford assays (Bio-Rad, Reinach BL, Switzerland) according to the manufacturer’s instructions 2.5. Proteins identification [33]. The samples were then aliquoted into 50 g portions and − ◦ stored at 80 C. For protein characterisation, the 2D-DIGE gels were stained with Coomassie brilliant blue R-250 after the fluorescence 2.4. 2D-DIGE analysis image acquisition [35]. The differentially expressed protein spots were picked manually using a scalpel and digested The protein extracts were labelled with Cy2, Cy3 and Cy5 using trypsin gold enzymes following the manufacturer’s CyDye DIGE Fluor minimal dyes (GE Healthcare Bio-Sciences instructions (Promega AG, Dübendorf, Switzerland). A vol- AB, Uppsala, Sweden) for 30 min in the dark according to the ume of 0.5 l of each sample was loaded onto a FlexiMassTM manufacturer’s instructions for the labelling process (8 pmol target well, which was overlaid with 0.5 l of a saturated per 50 g-1). In brief, 50 g of protein from light or dark condi- ␣-cyano-4-hydroxycinnamic acid solution in 50% (v/v) ace-
tions of three independent biological replica were individually tonitrile, 50% (v/v) dH2O and 0.1% (v/v) trifluoroacetic acid labelled using 400 pmol of either Cy3 or Cy5. A mixed (equal (TFA) and crystallised by air drying. The peptide mix PepMix1 pool of all samples, 3× light and 3× dark) internal standard (LaserBio Labs, Valbonne, France) was used as a standard for labelled with 400 pmol of Cy2 was included for spot normal- calibration. Peptide mass fingerprints (PMFs) of tryptic pep- isation and to allow comparison across all gels within the tides were collected by an AXIMA Confidence matrix-assisted analysis as published previously [27,34]. On completion of the laser desorption/ionisation time-of-flight tandem mass spec- labelling reaction, the Cy3 and Cy5 labelled samples were com- trometry (MALDI-TOF-MS) (Shimadzu Biotech, Manchester, bined pair wise and 50 g of Cy2-labelled internal standard UK). Carbamidomethylation of cysteine (Cys) and oxida- was added to each gel (see Table 1), as recommended by the tion of methionine (M) as fixed and variable modifications, manufacturer. respectively, were taken into account for database searching. 20 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
The following search parameters were used in all MASCOT searches: tolerance of one missed cleavage and a maximum error tolerance of 0.3 Da in the MS data and 0.8 Da in the MS/MS data. All peptide PMFs and peptide fragmentation fin- gerprints (PFFs) data were searched using MASCOT search engine (http://www.matrixscience.com; [36]) against an in- house database containing the draft genome of strain Cad16T in addition to public databases (NCBInr and Swissprot). Iden- tifications were accepted based on the significant MASCOT Mowse score and direct correlation between the identified pro- tein and its estimated molecular mass and pI determined from the 2D-gel. A MASCOT score of 64 corresponds to p < 0.05 for mass fingerprint experiments, while a MASCOT score of 37 corresponds to p < 0.05 for MS/MS sequencing; these thresh- olds were chosen as the cutoff for a significant hit.
2.6. Draft genome of strain Cad16T Fig. 1 – The protein expression patterns of Candidatus “T. The draft genome sequence of Candidatus ‘Thiodictyon syn- syntrophicum” strain Cad16T separated in a 24 cm, pH 3–10 T trophicum’ strain Cad16 was determined by pyrosequencing nonlinear strip and a 12% polyacrylamide gel. The proteins in the laboratory of Dr. S.C. Schuster (Z. Liu, K. Vogl, N.- that were extracted in the presence of light were labelled U. Frigaard, L.P. Tomsho, S.C. Schuster and D.A. Bryant, with Cy3 (green), and those extracted during the dark unpubl. data) at the Genomics Core Facility of Pennsylvania phase were labelled with Cy5 (red), both of which were State University. Paired-end reads from GX-20 FLX Titanium submitted to 2D-DIGE analysis. Proteins more abundant in chemistries were assembled into 1352 primary contigs of the light are characterised by an “L” before the number, and a total of 7.3 Mbp. A total of 7063 ORFs were detected by those in the dark are characterised by a “D”. Non-regulated annotation using a pipeline based on FGENESB software proteins that were present in the same quantity in both (Softberry, Inc., USA), Artemis (Sanger Institution, UK), and conditions are characterised by an “N” before the number. custom-made Perl scripts (ActivePerl; ActiveStateInc., Van- (For interpretation of the references to color in this text, the couver, BC, USA). The genome sequencing project has been reader is referred to the web version of the article.) assigned the bioproject number PRJNA32527 in GenBank (http://www.ncbi.nlm.nih.gov/bioproject).
3. Results and discussion
2.7. Analysis of the total mRNA during autotrophic 3.1. Differentially expressed proteome analysis growing conditions (2D-DIGE)
To determine a relationship between the protein spots that A representative 2D-DIGE image (Table 1, gel 1) showing the were analysed and their actual gene expression, the total fluorescent levels of the total proteins extracted in the light T mRNA of Candidatus “T. syntrophicum” strain Cad16 was (Cy3, green) and in the dark (Cy5, red) is shown in Fig. 1. More extracted and converted to cDNA prior to Solexa paired-end than 1400 protein spots were detected during the analysis of sequencing [37]. Transcriptome analysis was performed from the 2D-DIGE gel using Melanie 7.0 software. Among them, 56 T the cells of strain Cad16 that were incubated under the same protein spots had an ANOVA p-value <0.05. These protein spots conditions of growth that were applied for the proteomics were selected as showing a differential expression that was analysis. The total mRNA was extracted after 4 h of light (at statistically significant. Considering 1.5-fold to be the mini- T 11:00). The total RNA of strain Cad16 was isolated using Trizol mal level of differential expression, the expression levels of reagent (Invitrogen, Zug, Switzerland) and enriched for mRNA 40 protein spots were modified when Candidatus “T. syntroph- using the MicrobExpress kit (Ambion, Zug, Switzerland) fol- icum” strain Cad16T was exposed to the light compared to lowing the manufacturer’s instructions. Conversion to cDNA their exposure to 4 h of darkness. Sixteen spots from the 56 was performed using the MessageAmp IIBacteria kit (Ambion, protein spots showing an ANOVA p-value < 0.05 were excluded Zug, Switzerland) and the SuperScript ds-cDNA Synthesis from the analysis because they were below the differential kit (Invitrogen, Zug, Switzerland) according to the manu- expression threshold ratio of 1.5 measured through the flo- facturers’ instructions. The cDNA was subjected to Solexa rescence intensity of the signal. Among the 40 differentially × paired-end sequencing (2 75 bp) of a 200-bp insert library expressed spots satisfying the ANOVA and intensity param- (Beijing Genome Institute). Paired-end sequencing was used eters, 23 spots were more abundant in the presence of the rather than single-end sequencing to increase the number of light, and 17 spots were more abundant in the dark. The dif- bases available for the analysis of the transcriptome. ferentially expressed proteins were identified by peptide mass e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 21 *** 53 x 661 119 178 462 912 666 to Gene (mRNA) 4340 1636 4340 2157 1038 1376 3755 2094 Number of Reads Mapping * ve light; 3.84 1.92 1.69 1.53 5.10 4.91 5.66 2.68 5.73 3.10 2.83 1.59 1.96 1.58 2.30 2.44 Fold ative dark) Exprestion (positi neg P a 0.0003 0.0090 0.0006 0.0070 0.0004 0.0017 0.0003 0.0005 0.0008 0.0007 0.0065 0.0074 0.0230 0.0197 0.0135 0.0500 Anov match 34 36 23 24 38 36 43 23 25 11 42 25 19 37 29 34 erage of the PMF % of sequence cov hes 9 7 7 6 6 4 9 ** 12 20 10 10 15 26 20 15 searc ptides used matching for Mascot Number of pe (pI) 6.02 7.86 5.41 6.76 6.02 5.19 5.86 6.61 8.63 3 (PFF) 5.37 4.77 6.32 9.35 4.86 4.76 6.85 Point Isoelectric by 9907 kDa 69,523 40,707 55,095 45,458 69,523 19,442 86,955 26,330 60,153 17,976 33,657 21,660 17,738 19,319 eight (Mw) Molecular inducted w Experimental the protein (no violascens protein (no in bacteria ve domain) ve domain) ve domain) by otective antigen system, ati ed ed ati ati otein eductases ystis ate-binding protein and related in Cad16T y Pr synthase subunit genome MALDI-TOF MS/MS. otein/pr ansporter xidor hain best matching by encoded Outer membrane pr Uncharacterised conserv Outer membrane protein (porin) Poly(R)-hydroxyalkanoic acid synthase subunit PhaE alpha ATP NADPH-dependent glutamate synthase beta c Elongation factor G (EF-G) famil Transcription termination factor Rho gene Outer membrane protein (porin) Uncharacterised protein conserv Hypothetical protein (no conserv conserv 50S ribosomal protein L28 58,149 ABC-type molybdate transport tr Ribosome recycling factor (RRF) Hypothetical conserv OMA87 periplasmic component (Thioc (tellurite res. TerB) DSM 198) substr o ed Locus Cad16T genome | 390,952,282 Amino acid ABC Names of the best match of Order Thd2499 Thd4312 Thd3787 Thd6580 Thd3932 Thd6617 Thd0032 Thd5879 Thd3787 Thd5760 Thd5501 Thd3346 Thd5116 gi Thd2504 Thd3751 on the images gel given L6 L9 L5 L8 L4 L2 L15 L7 L3 L12 L13 L10 L14 L16 Table 2 – List of proteins identified Spot number as L1 L11 2-D 22 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 84 96 49 67 67 118 488 546 575 201 694 to Gene (mRNA) 1095 3019 2163 2163 1357 1645 1645 Number of Reads Mapping * ve light; 2.47 2.49 4.42 1.83 2.77 2.39 2.01 Fold − 2.05 − 4.22 − 1.88 − 1.80 − 2.24 − 2.20 − 3.34 − 1.85 − 3.88 − 1.85 − 2.81 ative dark) Exprestion (positi neg P a 0.0229 0.0450 0.0006 0.0205 0.0162 0.0090 0.0128 0.0007 0.0138 0.0144 0.0350 0.0070 0.0009 0.0396 0.0110 0.0027 0.0256 0.0004 Anov match 63 73 47 84 71 38 63 58 39 38 38 58 40 54 11 27 72 72 erage of the PMF % of sequence cov hes 5 9 9 ** 12 12 10 16 18 15 15 12 12 11 12 13 12 13 searc ptides used matching for Mascot Number of pe 1 (PFF) (pI) 5.15 6.18 5.34 5.16 5.15 7.71 5.20 5.20 5.74 7.02 7.10 6.45 8.37 6.75 5.13 5.91 5.31 5.31 Point Isoelectric by 8373 kDa 15,774 17,812 10,520 15,774 20,832 27,831 27,831 42,270 40,614 35,171 26,932 33,930 28,025 12,680 53,964 19,781 19,781 eight (Mw) Molecular inducted w Experimental (MalE) ase the protein ve domain) ve domain) ve domain) by (desulfoviridin), subunit ati ati ati ansfer usE) ogenases ogenase otein yl-prolyl isomerase yl-prolyl isomerase yl-prolyl isomerase yl-prolyl isomerase in Cad16T Pr genome ydr ydr ptid ptid ptid ptid oteins/domains amma hemotaxis eductase best matching encoded Dissimilatory sulfite r Parvulin-like pe Methyl-accepting c phasin (PhaP) hypothetical protein (SpoVT/AbrB like domain) Acetyl-CoA acetyltr Parvulin-like pe Molecular chaperone (small heat shock protein) Hsp20 Parvulin-like pe Hypothetical protein (no conserv Malate/lactate deh Hypothetical protein (no conserv Hypothetical protein (no conserv gene Co-chaperonin GroES (HSP10) phasin (PhaP) Maltose-binding periplasmic pr Parvulin-like pe 3-hydroxyacyl-CoA deh (DsrC/T g ed Locus Cad16T genome Names of the best match of Order Thd2631 Thd2748 Thd6562 Thd6554 Thd4286 Thd6555 Thd0849 Thd4906 Thd2748 Thd1984 Thd5260 Thd6937 Thd6937 Thd4540 Thd6554 Thd4019 Thd1849 Thd0844 on the images gel given – Table 2 ( Continued ) L20 D1 L22 L18 L19 D5 D9 L23 D2 D3 D6 D7 D10 L21 D4 D8 D11 Spot number as L17 2-D e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 23 34 44 28 74 278 519 522 106 422 167 408 285 411 575 171 301 124 to Gene (mRNA) 6534 1966 Number of Reads Mapping * selected >1.5-fold with was ve light; 0.51 0.60 0.90 0.93 0.96 1.01 Fold − 0.56 − 0.33 − 0.59 − 0.47 − 0.55 − − 2.85 − 2.75 − 2.12 − 0.74 − 0.18 − 2.07 − 2.13 − 3.34 ative dark) Exprestion (positi neg P a 0.5560 0.1247 0.1635 0.1278 0.2339 0.1741 0.1500 0.4588 0.7305 0.7871 0.0007 0.0036 0.0010 0.2910 0.9825 0.1757 0.0201 0.0067 0.0040 Anov match 4 41 54 48 50 58 45 23 43 23 24 43 56 53 28 44 43 19 70 erage of the PMF % of sequence cov hes 5 8 9 8 7 ** ** 12 15 22 16 11 15 13 11 11 19 16 searc ptides used matching (PFF) for Mascot Number of ** pe proteins from minimal autotrophic medium (Pfennig’s medium) after 4 h of light (11:00 = L1-22) T (pI) 5.89 4.67 5.89 4.80 5.05 9.10 5.14 5.83 5.22 6.07 5.48 5.28 5.50 6.51 5.87 5.81 3 (PFF) 6.47 4 4.85 1 (PFF) 7.98 Point 12 h of dark (7:00–19:00) light/dark mean ANOVA < 0.05 and 1.5 fold expration. Isoelectric by by proteins (N). kDa 33,289 33,289 45,200 17,358 15,334 27,023 24,285 38,924 38,472 59,079 36,171 36,268 39,817 51,271 21,228 22,142 25,972 22,294 stable eight (Mw) Molecular inducted w Experimental “Thiodictyon syntrophicum” strain Cad16 pendent pendent the , large subunit in bacteria gluconate ve domain) by ed ati Candidatus otein yl-prolyl isomerase in Cad16T anese-de anese-de xylase Pr genome ydrogenase otein best matching encoded Superoxide dismutase, mang Phosphoribulokinase Ribulose 1,5-bisphosphate carbo Predicted 6-phospho Ribosomal protein L7/L12 12,979 Phosphoribulokinase Spermidine/putrescine- binding periplasmic pr Enolase Peroxiredoxin Ribosomal protein L9 Peroxiredoxin Peroxiredoxin Uncharacterised protein conserv gene Gluconolactonase Fructose/tagatose bisphosphate aldolase Superoxide dismutase, mang Hypothetical protein (no conserv Parvulin-like peptid Periplasmic glucans biosynthesis protein deh (CbbM) (tellurite res. TerB) MALDI-TOF MS/MS ed Locus by Cad16T genome PFF with one or more independent peaks of the PMF spectrum. by Names of the best match of Order Thd0660 Thd1255 Thd5907 Thd2730 Thd6339 Thd1255 Thd6763 Thd0697 Thd2537 Thd5951 Thd1338 Thd0914 Thd2858 Thd4648 Thd6175 Thd2672 Thd4647 Thd3915 Thd5863 on the images gel value < 0.05. Proteins that have fold expression 01.5 are considered as given P Average ratio of the protein fold expression (abundance) between total cellular proteinsProteins isolated identified from Light (L) vs.Proteins Dark not (D), presents The in minimal the level trascritome of analysis differential report. expression – Table 2 ( Continued ) N9 N8 N3 N11 N13 N7 N5 N2 N10 N12 D13 D14 D15 N6 N4 D16 D17 N1 and after 4 h of dark (23:00 = D1-18) in a continuos photoperiod of 12 h of light fallowed Spot number as D12 List of proteins identified 2-D ∗ ∗∗ ∗∗∗ 24 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
fingerprinting (PMF) or peptide fragmentation fingerprinting photosynthesis (see Section 3.2.3 below) suggested a cell (PFF) (Table 2). The reliability of the protein identification was state characterised by an active metabolism in the presence assessed on the basis of Mascot scores (p-value < 0.05) using of a relatively high content of nutrients and energy, which the draft genome of strain Cad16T as a reference database could be used for protein biosynthesis during cell growth. (in-house database). The increased expression of the enzyme NADPH-dependent The transcriptome isolated from strain Cad16T in the light glutamate synthase beta chain (Table 2, spot L2), which is was used to confirm the expression of the analysed protein responsible for the production of amino acids, confirmed the spots (Table 2; right column: Number of Reads Mapping to Gene high energetic state of the cells by promoting the turnover of (mRNA)). NAD(P)H that originated from photosynthesis.
3.2. Protein spots more abundant in the presence of light 3.2.2. Storage biosynthesis Interestingly, 3 proteins (Table 2, spots L8, L17 and L18) Among the 23 spots that were more abundant in the pres- involved in a storage mechanism that produces intra- ence of light, we found proteins that were predominantly cellular granules of polyhydroxyalkanoates (PHAs) were involved with (a) protein biosynthesis, (b) storage biosynthe- over-expressed under the light conditions. This mechanism sis, (c) sulfur compound oxidation and photosynthesis (d), has been observed in a wide variety of microbes that trans-membrane transport and (e) other metabolic processes. accumulate PHAs as a carbon and energy storage reserve when essential nutrients, such as nitrogen or phospho- 3.2.1. Protein biosynthesis rus, are limited but carbon sources are in excess [44,45]. During the light phase, strain Cad16T over-expressed many A typical PHA that is accumulated in microbial cells is proteins involved in protein biosynthesis, such as ribosomal poly(3-hydroxybutyrate) (PHB), which consists of 1000–30,000 proteins and enzymes needed for gene transcription (Table 2, hydroxy fatty acid monomers, forming a granule in the cyto- spots L1, L7, L14, L15, L21 and L23). The transcription of genes, plasm. PHB synthesis in Alcaligenes eutrophus is stimulated by performed by RNA polymerase, is ended by the transcrip- both high intracellular concentrations of NAD(P)H and high + tion terminator factor Rho (Table 2, spot L7), which binds ratios of NAD(P)H/NAD(P) which also inhibit citrate synthase to the transcription terminator pause site and stops mRNA activity and thereby facilitating the metabolic flux of acetyl- transcription [38]. The ribosome links the amino acids and CoA to the PHB synthetic pathway [46]. The most important translates the genes found on the mRNA into proteins. In enzyme in the biosynthesis of this storage substance is the prokaryotes, the ribosome is composed of two subunits of 50S PHA synthase. PHA synthases currently are divided into four and 30S. The large subunit catalyses peptide bond formation classes depending on their subunit composition and sub- [39]. The ribosome of Escherichia coli contains 22 proteins in strate specificity. It has been shown in other species of PSB, the small subunit, labelled S1–S22, and 36 proteins in the large such as Allochromatium vinosum and Thiocystis violacea [47,48], subunit, labelled L1–L36 [40]. In this study, L28 showed an up- that the presence of a Class III PHA synthase consists of regulation during the light period (Table 2, spot L1). During two subunits (PhaC and PhaE). This enzyme is responsible the translational step, the ribosome interacts with different for the final step of the synthesis, in which PHB is syn- GTPase factors as well as initiation factors (IFs), elongation fac- thesised from (R)-3-hydroxybutyryl-CoA molecules and free tors (EFs) and release factors (RFs) via its conserved C-terminal CoA is released. However, the process typically starts from domain [41]. Initiation factors IF1, IF2 and IF3 are involved in 2 acetyl-CoA molecules that are coupled together to form the initiation of protein synthesis to form the 70S initiation acetoacetyl-CoA in a condensation reaction that is catalysed complex. Elongation factors EF-Tu, EF-Ts and EF-G (Table 2, by 3-ketothiolase (PhaA), and this product is subsequently spot L15) are needed for the extension of protein synthesis. stereoselectively reduced to (R)-3-hydroxybutyryl-CoA in a Release factors RF1, RF2 and RF3 regulate the steps during the reaction catalysed by NADPH-dependent acetoacetyl-CoA termination phase of translation. Lastly, ribosome recycling reductase (PhaB). However, PHB can also be synthesised from factor (RRF) (Table 2, spot L14), along with EF-G, catalyses the an intermediate of the beta-oxidation/bio-synthesis of fatty recycling of the ribosomal subunits [42]. acids. The molecule of trans-2-enoyl-CoA is converted to (R)-3- The efficient folding of many newly synthesised proteins hydroxyacyl-CoA via (R)-specific hydration, which is catalysed depends on assistance from molecular chaperones. Molecu- by (R)-specific enoyl-CoA hydratase (PhaJ) [49–51]. The storage lar chaperones, such as the heat shock proteins (HSPs), are granules of PHA require other associated proteins to protect among the more abundant cytosolic proteins found in cells their hydrophobic core from the cytoplasm, all of which play from the three kingdoms: eukaryotes, bacteria and archaea. a role in regulating the number and sizes of the granules pro- Their role is to recognise and bind nascent polypeptide chains duced. The most abundant granule-associated protein is an and partially folded protein intermediates, preventing their amphiphilic protein known as phasin (PhaP) [52]. aggregation and misfolding, both under normal conditions The over-expression of proteins involved in this storage T and when cells are exposed to stresses such as high tem- mechanism in strain Cad16 , such as the PhaE subunit of PHA perature [43]. Protein spots L21 and L23 (Table 2), which were synthase and phasins (PhaP), suggested the presence of suffi- more abundant under the light conditions, corresponded to cient amounts of both carbon sources and reducing factors in chaperones HSP10 and HSP20, respectively. the cells during the light period, likely originating from photo- The large number of proteins involved in protein biosyn- synthesis [53]. This storage mechanism is commonly observed thesis, storage mechanisms (see Section 3.2.2 below), and in phototrophic and chemotrophic bacteria [54]. e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 25
3.2.3. Sulfur compound oxidation and photosynthesis expression between the light and the dark condition of these Purple sulfur bacteria can grow photo-autotrophically using proteins. inorganically reduced sulfur compounds (i.e. S−2,S0 and 2− S2O3 ) as electron donors for CO2 fixation via the CBB 3.2.4. Transmembrane transport proteins cycle [11]. The photochemical reaction centre complex uses Additional proteins that showed strong expression in Can- light energy to transfer electrons from reduced sulfur com- didatus “T. syntrophicum” strain Cad16T in the presence of pound to NAD(P)+ [55]. Dissimilatory sulfite reductase (DsrAB, light were those involved in transmembrane transport, such a tetramer composed of two alpha and two beta subunits) as unspecific porins (Table 2, spots L3, L5 and L6) and more is responsible for the oxidation of the sulfide (S−2) to sul- specifically, ABC transporter (Table 2, spots L10 and L11). Porins 2− fite (SO3 ) to generate electrons for photosynthetic CO2 are proteins that cross cellular membranes in Gram-negative reduction. This enzyme is also required for the oxidation bacteria and act like a pore, through which different types of intracellularly stored sulfur granules, especially in sulfur- of molecules up to 600 Da can diffuse. Their expression is storing PSB [56]. DsrAB has been reported to bind another regulated by the environmental conditions of growth. Porins subunit, called gamma subunit DsrC [57,58]. This gamma sub- allow cells to fine tune the uptake of appropriate nutrients unit (DsrC) was more abundant in our organism during the and to protect themselves optimally against external fac- light period (Table 2, spot L20), suggesting an over-expression tors, such as osmotic pressure and temperature [64]. ABC of the dissimilatory sulfite reductase enzyme. transporters are transmembrane proteins that utilise energy The photosynthetic reaction centre (PSI) is also respon- from the hydrolysis of ATP to transport various substrates sible for generating the proton motive force necessary for across the membrane. Proteins are classified as ABC trans- ATP generation by the ATP synthase. The ATP synthase sub- porters based on the sequence and organisation of their unit alpha (spot L4) was more abundant in the light. This ATP-binding cassette (ABC) domain(s), and they are specific ubiquitous enzyme is present in the plasma membrane of for the uptake of a large variety of nutrients, such as sug- bacteria, the thylakoid membrane of chloroplasts, and the ars, amino acids, peptides, inorganic ions and vitamins [65,66]. inner mitochondrial membrane. ATP synthase is a transmem- Moreover, it was shown that ABC transporters are important brane protein, and for this reason it is composed of two in the antibiotic resistance in different bacteria [67–69]. Many distinct regions: the hydrophobic F0, which is embedded in secondary metabolites (e.g. antibiotics and toxins) are toxic to the membrane and allows the transit of protons, and the the microorganisms that produce them, so ABC transporters hydrophilic F1, a complex composed of five types of subunits: would avoid inhibition of growth of the producing strain by ␣3 (Table 2, spot L4), 3, ␥1, ␦1 and 1 [59]. At low activity of preventing the drug from going back to the cell, acting as a the proton motive force, the flow is reversed, and the enzyme one-way in-out pump. functions as a proton-pumping ATPase. In many bacterial The increased expression of the transmembrane transport species (mostly anaerobic), the reverse reaction is essential proteins suggested a strong interaction between the cells and for life, when neither the respiratory chain nor the photosyn- their environment for the uptake of nutriments and the excre- thetic proteins are able to generate the proton motive force. tion of metabolites. In this way, many important cellular functions, such as flag- ellar motility or ion\nutrient transmembrane transport are 3.2.5. Other metabolic processes supported. Other cellular metabolic processes were also highlighted while The photosynthetic lifestyle of the PSB from Lake Cadagno analysing the total proteins that were extracted from Can- and especially of Candidatus “T. syntrophicum” strain Cad16T, didatus “T. syntrophicum” strain Cad16T in the presence of was previously investigated [13,14,16–18]. We were there- light. The methyl-accepting chemotaxis protein (MCP) (Table 2, fore surprised to find a relatively low number of expressed spot L22), a component of the chemotactic response system proteins related to photosynthesis, such as RuBisCO. It is in bacteria and archaea, governs the migration of bacteria known that many bacteria have microcompartments that towards chemical attractants and away from repellents by concentrate the metabolically related enzymes to increase translating temporal changes in the levels of chemical stimuli their efficiency and to control the flow and concentra- into a modulation of the cell’s swimming direction [70]. MCP tions of substrates and intermediates [60,61]. During the is a transmembrane sensor protein in bacteria, allowing the
CO2-fixing process, different phototrophic bacteria showed detection of concentrations of molecules in the extracellu- the presence of carboxysome-like microcompartments that lar matrix [71]. Candidatus “T. syntrophicum” strain Cad16T is encapsulate RuBisCO and carbonic anhydrase and thereby highly metabolically active in the light and use this MCP to enhance carbon fixation by increasing the levels of CO2 in search for nutrients. It could be that the expression of this the vicinity of RuBisCO [62,63]. All the genes that are involved gene is an automatic reply to external stimuli, e.g. the need of in the formation of a carboxysome-like microcompartment substances involved in some metabolic pathways [72]. are present immediately after the type I RuBisCO cbbL and Moreover, five other spots more abundant in the presence cbbS genes in the draft genome of strain Cad16T (Genbank: of light (Table 2, spots L9, L12, L13, L16 and L19) were identified JQ693375–JQ693380). In our analysis, these large, complex car- as “hypothetical proteins” or “Uncharacterised conserved pro- boxysome shell structures could be lost during the extraction tein”. However, conserved domains were recognised in two of of the total proteins or during the protein separation process, this spots (Table 2, spots L12 and L19) using BLASTp analysis and for this reason, we did not observe the type I RuBisCO (http://blast.ncbi.nlm.nih.gov/). Spot L12 showed a tellurium (CbbL or CbbS) in our gels. However, by our results the sim- cassette component in its sequence. In the past, prior to the ply explanation was the absence of relevant difference in the development of antibiotics, tellurite compounds were used 26 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
to treat conditions such as leprosy, tuberculosis, dermatitis, as components involved in different pathways of CO2 fixa- cystitis and eye infections. It has been proposed that tel- tion [9]. The first enzyme, the malate dehydrogenase (MDH) lurite toxicity is due to its strong oxidising ability and its (Table 2, spot D6), is an enzyme part of the reverse tricar- ability to replace sulphur in various cellular functions [73]. boxylic acid (rTCA) cycle found especially in anaerobic bacteria An analysis of the protein spot L19 highlighted the presence (e.g. GSB) and of the dicarboxylate/4-hydroxybutyrate (DC/HB) of a SpoVT/AbrB-like domain. In Bacillus subtilis, this domain cycle found at the present day only in some archaeal species appears to be involved in the transcription activation of the [77,78]. This enzyme reversibly reduces the oxaloacetate in repression of genes expressed during the transition state malate using the oxidation of NADH to NAD+. Interestingly, between the exponential and the stationary phase. Antibi- also the others two enzymes up-regulated in the dark could be otic resistance protein B (AbrB) is a representative of a large part of the DC/HB cycle, catalysing the 2 last reactions of the superfamily of known and putative transcription factors that cycle: the oxidation of 3-hydroxybutyryl-CoA in acetoacetyl- includes transition-state regulators, putative regulators of cell CoA using NAD(P)+ as an electron acceptor catalyses by the wall biosynthesis, regulators of phosphate uptake, and a large enzyme 3-hydroxyacyl-CoA dehydrogenase (Table 2, spot D11), number of proteins of as yet unknown activity [74]. However, and the oxidisation of acetoacetyl-CoA in 2 acetyl-CoA cataly- we must be cautious and use this identification by BLASTp ses by the enzyme acetyl-CoA acetyltransferase (Table 2, spot as a strictly hypothetical identification. Unfortunately, the D5). However, these two enzymes could also be part of the last hypothetical proteins of spots L9, L13 and L16 did not show step of the beta-oxidation of fatty acid and PHB granules. The any conserved domains in their gene sequence, and for this degradation of the PHB granules produced during the day in reason, their function remain unknown. These uncharac- the presence of light (see the Section 3.2.2 before) might pro- terised proteins and those showing a conserved domain were duce acetyl-CoA and reducing power in the form of NAD(P)H. likely proteins with unknown functions, or they were not yet Both substrates are potentially used in the CO2 fixing process annotated in our draft genome (e.g. between 2 contigs). (rTCA or/and DC/HB cycle or others) during the dark period. Therefore, the rTCA or the DC/HB cycle have to be investigated 3.3. Protein spots more abundant in the dark more in the detail in strain Cad16T.
Among the 17 protein spots that were more abundant in the 3.3.3. Stress metabolism dark, we found proteins that were mainly involved with (a) Proteins involved in the oxidative stress response (superoxide isomerases, (b) carbon dioxide fixation, (c) stress metabolism and hydrogen peroxide) were also detected (Table 2, spots D13, and (d) other metabolic processes. D14 and D16). Superoxide dismutase is an enzyme that catal- yses the dismutation of superoxide into oxygen and hydrogen 3.3.1. Isomerases peroxide. It plays an important antioxidative role in cells that The main proteins expressed by Candidatus “T. syntrophicum” are exposed to oxygen. Peroxiredoxins are a ubiquitous family strain Cad16T in the dark belonged to the family of isomerases, of antioxidant enzymes that use thioredoxin (Trx) to detox- especially to the peptidyl-prolyl isomerases (PPIs) and protein ify hydrogen peroxide. Moreover, these types of proteins can disulphide isomerase (PDI) (Table 2, spots D1, D2, D8, D9 and also act as molecular chaperones and, most importantly, as D12). PPIs promote proper protein folding by increasing the regulators in the 24 h internal circadian clock of many orga- rate of transition of proline residues between the cis and trans nisms [79–81]. Candidatus “T. syntrophicum” strain Cad16T was states, and they also possess a chaperone-like activity. Pro- grown in autotrophic Pfennig’s medium (see Section 2) without teins with prolyl isomerase activity include cyclophilin, FKBPs, the presence of oxygen in the gas phase; therefore, the up- and parvulin, although larger proteins can also contain pro- regulation of these proteins, especially peroxiredoxin, could lyl isomerase domains [75]. PDI is an enzyme that catalyses be linked to the circadian rhythm mechanism. Indeed, the cel- the formation and breakage of disulphide bonds via its four lular ROS balance is important for robust 24 h circadian clock thioredoxin-like domains. PDI can act also as a chaperone by function, as was recently described in Neurospora [82]. assisting in the reactivation of denatured proteins that do not Another possible explanation for these anti-stress contain cysteine residues. In E. coli, 2 prologue proteins of PDI enzymes comes from one of the key enzymes of the DC/HB (DsbC and DsbG) are located in the periplasmic space outside cycle. In fact, the 4-hydroxybutyryl-CoA dehydratase is con- the cytoplasm [76]. sidered a “radical enzyme” that uses radicals as intermediates It is difficult to interpret the considerable number of iso- and enables the metabolism of otherwise refractory com- merases that were more abundant in the dark; however, they pounds. Because they are highly reactive towards dioxygen, could play a certain chaperone role through the use of the radicals are often found during catalysis by enzymes from stored substances produced during the phototrophic activity anaerobic microorganisms [83–85]. An hypothetical enhanced in the light. activity of 4-hydroxybutyryl-CoA dehydratase in the dark most likely produced more dangerous free radicals, which 3.3.2. Carbon dioxide fixation were limited by the simultaneous increased expression of The ability of Candidatus “T. syntrophicum” strain Cad16T to antioxidant enzymes.
fix CO2 in the dark was previously observed in several oth- ers experimental analyses; however, the mechanism involved 3.3.4. Other metabolisms in the dark assimilation is not yet understood [15,17]. We Similar to what was observed from proteins that were more found three protein spots that were more abundant in the abundant in the light, we also found a membrane transport dark (Table 2, spots D5, D6 and D11) and were identified mechanism that was highlighted in the dark. A maltose ABC e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 27
Fig. 2 – Scheme summarising the majors metabolic pathways suggested from 2D-DIGE analysis, (A) in presence of light and (B) in the dark.
transporter periplasmic protein (Table 2, spot D4) was found dark. While the mechanism of CO2 fixation in the presence of to be more abundant during the dark phase. light provided by the CBB cycle is relatively well known [86], the Moreover, 5 protein spots that were more abundant in the assimilation of inorganic carbon in the dark remains poorly dark (Table 2, spots D3, D7, D10, D15 and D17) were identified as understood. In this study, we showed that 3 enzymes that “hypothetical proteins”. Unfortunately, no conserved domains were more abundant in the dark could be part of the anaerobic were found after online pBLAST analysis for four of the five dicarboxylate/4-hydroxybutyrate (DC/HB) cycle, which repre- spots. A specific domain was assigned only to spot D15, which sents an autotrophic CO2 fixation pathway that is found in identified tellurite resistance from the domain, similar to the some archaea [9]. The substrates needed for this process protein spot L12 that was also more abundant in the light. such as reducing power and energy could be provided by the degradation of the storage globules of poly(3-hydroxybutyrate) 3.4. Proteins with unvaried abundance (PHB), which synthesis was shown to be more abundant during the light conditions. In conclusion, the abundant presence of Protein spots with similar expression level between the enzymes potentially involved in the autotrophic DC/HB cycle light/dark conditions were also analyses. From the 13 selected in the dark suggests to us a possible explanation for the high T spots (see Table 2; N1–N13), we found important housekeeping capacity shown by PSB, and especially in strain Cad16 , to proteins, such as ribosomal proteins and periplasmic glucan fix CO2 in the absence of light. In the future we will try to T biosynthesis proteins (Table 2, spots N1, N12 and N13). Inter- learn more about the effective capacity of strain Cad16 to fix esting proteins potentially involved in the stress metabolism CO2 via this cycle. The scheme of Fig. 2 summarise the majors (Table 2, spots N9 and N10) or implicated in other cellular metabolic reactions in presence of light (Fig. 2A) or in the dark mechanisms such as the circadian rhythm (see Section 3.3.3) (Fig. 2B). were also found. Furthermore, the presence of type II RuBisCO protein (Table 2, spot N3) in the unchanging expressed protein Acknowledgments confirmed recent results that showed a constant expression of the cbbM gene throughout a day in strain Cad16T [17]. Financial support for this project was provided by the Uni- versity of Geneva, the Institute of Microbiology and State of 4. Conclusion Ticino, and the Swiss National Science Foundation (grant no. 31003A-116591). The authors are indebted to Cinzia Benagli, We present here the first analysis of the expression changes Sophie De Respinis and Natalia Giot for technical and moral in the proteome between the light and dark phases of a PSB support. Furthermore, we are grateful to D. Burri and M. Wit- grown in anoxic autotrophic conditions. Our goal was to inves- twer from SPIEZ Laboratory, Federal Office for Civil Protection tigate the mechanisms behind the high capacity of these (FOCP), Spiez (Switzlerland), for their technical support. The bacteria, and especially of the PSB Candidatus “T. syntroph- authors would also like to thank the FEMS and Alpine Biology icum” strain Cad16T, to assimilate inorganic carbon in the Center Foundation (ABC) for their support. 28 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
r e f e r e n c e s purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp. Systematic and Applied Microbiology 2012;35:139–44. [1] Tonolla M, Demarta A, Peduzzi R, Hahn D. In situ analysis of [19] Arioli S, Roncada P, Salzano AM, Deriu F, Corona S, phototrophic sulfur bacteria in the chemocline of Guglielmetti S, et al. The relevance of carbon dioxide meromictic Lake Cadagno (Switzerland). Applied and metabolism in Streptococcus thermophilus. Microbiology Environmental Microbiology 1999;65:1325–30. 2009;155:1953–65. [2] Decristophoris PMA, Peduzzi S, Ruggeri-Bernardi N, Hahn D, [20] Kluge S, Hoffmann M, Benndorf D, Rapp E, Reichl U. Tonolla M. Fine scale analysis of shifts in bacterial Proteomic tracking and analysis of a bacterial mixed culture. community structure in the chemocline of meromictic Lake Proteomics 2012;12:1893–901. Cadagno, Switzerland. Journal of Limnology 2009;68:16–24. [21] Tannu NS, Hemby SE. Two-dimensional fluorescence [3] Tonolla M, Peduzzi R, Hahn D. Long-term population difference gel electrophoresis for comparative proteomics dynamics of phototrophic sulfur bacteria in the chemocline profiling. Nature Protocols 2006;1:1732–42. of Lake Cadagno, Switzerland. Applied and Environmental [22] Tomanek L. Environmental proteomics: changes in the Microbiology 2005;71:3544–50. proteome of marine organisms in response to [4] Imhoff JF. Bergey’s manual of systematic bacteriology. environmental stress, pollutants, infection, symbiosis, and Springer; 2005. development. Annual Review of Marine Science [5] Van Gemerden H, Mas J. Ecology of phototrophic sulfur 2011;3:373–99. bacteria. Boston: Kluwer Academic Publishers (Springer); [23] Yonemori H, Kubota D, Taniguchi H, Tsuda H, Fujita S, 2004. Murakami Y, et al. Laser microdissection and [6] Frigaard N-U, Bryant D. Chlorosomes: antenna organelles in two-dimensional difference gel electrophoresis with photosynthetic green bacteria. Complex Intracellular alkaline isoelectric point immobiline gel reveals proteomic Structures in Prokaryotes 2006;7:9–114. intra-tumor heterogeneity in colorectal cancer. EuPA Open [7] Tang JK-H, Saikin SK, Pingali SV, Enriquez MM, Huh J, Frank Proteomics 2013;1:17–29. HA, et al. Temperature and carbon assimilation regulate the [24] Görg A, Weiss W, Dunn MJ. Current two-dimensional chlorosome biogenesis in green sulfur bacteria. Biophysical electrophoresis technology for proteomics. Proteomics Journal 2013;105:1346–56. 2004;4:3665–85. [8] Law CJ, Roszak AW, Southall J, Gardiner AT, Isaacs NW, [25] O’Farrell PH. High resolution two-dimensional Cogdell RJ. The structure and function of bacterial electrophoresis of proteins. Journal of Biological Chemistry light-harvesting complexes (review). Molecular Membrane 1975;250:4007–21. Biology 2004;21:183–91. [26] Marouga R, David S, Hawkins E. The development of the [9] Berg IA. Ecological aspects of the distribution of different DIGE system: 2D fluorescence difference gel analysis autotrophic CO fixation pathways. Applied and 2 technology. Analytical and Bioanalytical Chemistry Environmental Microbiology 2011;77:1925–36. 2005;382:669–78. [10] Sirevag R, Buchanan BB, Berry JA, Troughton JH. Mechanisms [27] Alban A, David SO, Bjorkesten L, Andersson C, Sloge E, Lewis of CO fixation in bacterial photosynthesis studied by the 2 S, et al. A novel experimental design for comparative carbon isotope fractionation technique. Archives of two-dimensional gel analysis: two-dimensional difference Microbiology 1977;112:35–8. gel electrophoresis incorporating a pooled internal standard. [11] Frigaard NU, Dahl C. Sulfur metabolism in phototrophic Proteomics 2003;3:36–44. sulfur bacteria. London: Academic Press; 2009. [28] Pappin DJ, Hojrup P, Bleasby AJ. Rapid identification of [12] Overmann J. Sulfur metabolism in phototrophic organisms: proteins by peptide-mass fingerprinting. Current Biology: CB ecology of phototrophic sulfur bacteria. Urbana: Springer; 1993;3:327–32. 2008. [29] Henzel WJ, Billeci TM, Stults JT, Wong SC, Grimley C, [13] Camacho A, Erez J, Chicote A, Florin M, Squires M, Lehmann Watanabe C. Identifying proteins from two-dimensional gels C, et al. Microbial microstratification, inorganic carbon by molecular mass searching of peptide fragments in photoassimilation and dark carbon fixation at the protein sequence databases. Proceedings of the National chemocline of the meromictic Lake Cadagno (Switzerland) Academy of Sciences 1993;90:5011–5. and its relevance to the food web. Aquatic Sciences [30] Eichler B, Pfennig N. A new purple sulfur bacterium from 2001;63:91–106. stratified freshwater lakes, Amoebobacter [14] Casamayor EO, Garcia-Cantizano J, Pedros-Alio C. Carbon purpureus sp. nov. Archives of Microbiology 1988;149: dioxide fixation in the dark by photosynthetic bacteria in 395–400. sulfide-rich stratified lakes with oxic-anoxic interfaces. [31] Widdel F, Bak F. Gram-negative mesophilic sulfate-reducing Limnology and Oceanography 2008;53:1193–203. bacteria. The Prokaryotes 1992;4:3352–78. [15] Casamayor EO, Lliros M, Picazo A, Barberan A, Borrego CM, [32] Hobbie JE, Daley RJ, Jasper S. Use of nuclepore filters for Camacho A. Contribution of deep dark fixation processes to counting bacteria by fluorescence microscopy. Applied and overall CO incorporation and large vertical changes of 2 Environmental Microbiology 1977;33:1225–8. microbial populations in stratified Karstic lakes. Aquatic [33] Bradford MM. A rapid and sensitive method for the Sciences – Research Across Boundaries 2012;74:61–75. quantitation of microgram quantities of protein utilizing the [16] Musat N, Halm H, Winterholler B, Hoppe P, Peduzzi S, Hillion principle of protein-dye binding. Analytical Biochemistry F, et al. A single-cell view on the ecophysiology of anaerobic 1976;72:248–54. phototrophic bacteria. Proceedings of the National Academy [34] Karp NA, Lilley KS. Maximising sensitivity for detecting of Sciences 2008;105:17861–6. changes in protein expression: experimental design using [17] Storelli N, Peduzzi S, Saad MM, Frigaard NU, Perret X, Tonolla minimal CyDyes. Proteomics 2005;5:3105–15. M. CO assimilation in the chemocline of Lake Cadagno is 2 [35] Syrovy I, Hodny Z. Staining and quantification of proteins dominated by a few types of phototrophic purple sulfur separated by polyacrylamide gel electrophoresis. Journal of bacteria. FEMS Microbiology Ecology 2013;84:421–32. Chromatography B: Biomedical Sciences and Applications [18] Peduzzi S, Storelli N, Welsh A, Peduzzi R, Hahn D, Perret X, 1991;569:175–96. et al. Candidatus Thiodictyon syntrophicum, sp. nov., a new e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30 29
[36] Perkins DN, Pappin DJ, Creasy DM, Cottrell JS. [55] Brune D. Sulfur compounds as photosynthetic electron Probability-based protein identification by searching donors. Dordrecht/The Netherlands: Kluwer Academic sequence databases using mass spectrometry data. Publishers; 1995. Electrophoresis 1999;20:3551–67. [56] Pott AS, Dahl C. Sirohaem sulfite reductase and other [37] Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool proteins encoded by genes at the dsr locus of Chromatium for transcriptomics. Nature Reviews Genetics 2009;10:57–63. vinosum are involved in the oxidation of intracellular sulfur. [38] Richardson JP. Rho-dependent termination and ATPases in Microbiology 1998;144:1881–94. transcript termination. Biochimica et Biophysica Acta-Gene [57] Cort JR, Selan U, Schulte A, Grimm F, Kennedy MA, Dahl C. Structure and Expression 2002;1577:251–60. Allochromatium vinosum DsrC: solution-state NMR structure, [39] Marintchev A, Wagner G. Translation initiation: structures, redox properties and interaction with DsrEFH, a protein mechanisms and evolution. Quarterly Reviews of Biophysics essential for purple sulfur bacterial sulfur oxidation. Journal 2004;37:197–284. of Molecular Biology 2008;382:692–707. [40] Nissen P, Hansen J, Ban N, Moore PB, Steitz TA. The [58] Pierik AJ, Duyvis MG, Helvoort JMLM, Wolbert RBG, Hagen structural basis of ribosome activity in peptide bond WR. The third subunit of desulfoviridin-type dissimilatory synthesis. Science 2000;289:920–30. sulfite reductases. European Journal of Biochemistry [41] Helgstrand M, Mandava CS, Mulder FAA, Liljas A, Sanyal S, 2005;205:111–5. Akke M. The ribosomal stalk binds to translation factors IF2, [59] Feniouk BA, Suzuki T, Yoshida M. Regulatory interplay EF-Tu, EF-G and RF3 via a conserved region of the L12 between proton motive force, ADP, phosphate, and subunit C-terminal domain. Journal of Molecular Biology Ï in bacterial ATP synthase. Journal of Biological Chemistry 2007;365:468–79. 2007;282:764–72. [42] Ganoza MC, Kiel MC, Aoki H. Evolutionary conservation of [60] Kerfeld CA, Sawaya MR, Tanaka S, Nguyen CV, Phillips M, reactions in translation. Microbiology and Molecular Biology Beeby M, et al. Protein structures forming the shell of Reviews 2002;66:460–85. primitive bacterial organelles. Science 2005;309:936–8. [43] Hartl FU, Hayer-Hartl M. Molecular chaperones in the [61] Yeates TO, Crowley CS, Tanaka S. Bacterial cytosol: from nascent chain to folded protein. Science microcompartment organelles: protein shell structure and 2002;295:1852–8. evolution. Annual Review of Biophysics 2010;39:185. [44] Sudesh K, Abe H, Doi Y. Synthesis, structure and properties [62] Cannon GC, Heinhorst S, Kerfeld CA. Carboxysomal carbonic
of polyhydroxyalkanoates: biological polyesters. Progress in anhydrases: structure and role in microbial CO2 fixation. Polymer Science 2000;25:1503–55. Biochimica et Biophysica Acta (BBA)-Proteins & Proteomics [45] Jendrossek D. Polyhydroxyalkanoate granules are complex 2010;1804:382–92. subcellular organelles (carbonosomes). Journal of [63] Yeates TO, Kerfeld CA, Heinhorst S, Cannon GC, Shively JM. Bacteriology 2009;191:3195–202. Protein-based organelles in bacteria: carboxysomes and [46] Haywood GW, Anderson AJ, Chu L, Dawes EA. The role of related microcompartments. Nature Reviews Microbiology NADH-and NADPH-linked acetoacetyl-CoA reductases in the 2008;6:681–91. poly-3-hydroxybutyrate synthesizing organism Alcaligenes [64] Schulz GE. Bacterial porins: structure and function. Current eutrophus. FEMS Microbiology Letters 1988;52:259–64. Opinion in Cell Biology 1993;5:701–7. [47] Liebergesell M, Steinbüchel A. Cloning and nucleotide [65] Davidson AL, Dassa E, Orelle C, Chen J. Structure, function, sequences of genes relevant for biosynthesis of poly and evolution of bacterial ATP-binding cassette systems. (3-hydroxybutyric acid) in Chromatium vinosum strain D. Microbiology and Molecular Biology Reviews 2008;72:317–64. European Journal of Biochemistry 1992;209:135–50. [66] Moussatova A, Kandt C, O’Mara ML, Tieleman DP. [48] Liebergesell M, Steinbüchel A. Cloning and molecular ATP-binding cassette transporters in Escherichia coli. analysis of the poly (3-hydroxybutyric acid) biosynthetic Biochimica et Biophysica Acta (BBA)-Biomembranes genes of Thiocystis violacea. Applied Microbiology and 2008;1778:1757–71. Biotechnology 1993;38:493–501. [67] Méndez C, Salas JA. The role of ABC transporters in [49] Fukui T, Shiomi N, Doi Y. Expression and characterization of antibiotic-producing organisms: drug secretion and (R)-specific enoyl coenzyme A hydratase involved in resistance mechanisms. Research in Microbiology polyhydroxyalkanoate biosynthesis by Aeromonas caviae. 2001;152:341–50. Journal of Bacteriology 1998;180:667–73. [68] Otto M, Götz F. ABC transporters of staphylococci. Research [50] Tsuge T, Fukui T, Matsusaki H, Taguchi S, Kobayashi G, in Microbiology 2001;152:351–6. Ishizaki A. Molecular cloning of two (R)-specific enoyl-CoA [69] Martin JF, Casqueiro J, Liras P. Secretion systems for hydratase genes from Pseudomonas aeruginosa and their use secondary metabolites: how producer cells send out for polyhydroxyalkanoate synthesis. FEMS Microbiology messages of intercellular communication. Current Opinion Letters 2006;184:193–8. in Microbiology 2005;8:282–93. [51] Vo MT, Lee K-W, Jung Y-M, Lee Y-H. Comparative effect of [70] Alon U, Surette MG, Barkai N, Leibler S. Robustness in overexpressed phaJ and fabG genes supplementing bacterial chemotaxis. Nature 1999;397:168–71. (R)-3-hydroxyalkanoate monomer units on biosynthesis of [71] Gestwicki JE, Lamanna AC, Harshey RM, McCarter LL, mcl-polyhydroxyalkanoate in Pseudomonas putida KCTC1639. Kiessling LL, Adler J. Evolutionary conservation of Journal of Bioscience and Bioengineering 2008;106:95–8. methyl-accepting chemotaxis protein location in bacteria [52] Steinbüchel A, Aerts K, Liebergesell M, Wieczorek R, Babel and archaea. Journal of Bacteriology 2000;182:6499–502. W, Föllner C, et al. Considerations on the structure and [72] Martinez-Antonio A, Collado-Vides J. Identifying global biochemistry of bacterial polyhydroxyalkanoic acid regulators in transcriptional regulatory networks in bacteria. inclusions. Canadian Journal of Microbiology 1995;41:94–105. Current Opinion in Microbiology 2003;6:482–9. [53] Overmann J, Garcia-Pichel F. Prokaryotes the phototrophic [73] Taylor DE. Bacterial tellurite resistance. Trends in way of life. Springer; 2006. Microbiology 1999;7:111–5. [54] Hazer DB, Kilic¸ay E, Poly Hazer B. (3-Hydroxyalkanoate)s: [74] Phillips ZEV, Strauch MA. Bacillus subtilis sporulation and diversification and biomedical applications: a state of the art stationary phase gene expression. Cellular and Molecular review. Materials Science and Engineering: C 2012;32:637–47. Life Sciences 2002;59:392–402. 30 e u p a o p e n p r o t e o m i c s 2 ( 2 0 1 4 ) 17–30
[75] Fanghänel J, Fischer G. Insights into the catalytic [80] Knoops B, Loumaye E, Eecken V. Peroxiredoxin systems mechanism of peptidyl prolyl cis/trans isomerases. Frontiers evolution of the peroxiredoxins. Netherlands: Springer; 2007. in Bioscience 2004;9:3453–78. [81] Edgar RS, Green EW, Zhao Y, van Ooijen G, Olmedo M, Qin X, [76] Kadokura H, Katzen F, Beckwith J. Protein disulfide bond et al. Peroxiredoxins are conserved markers of circadian formation in prokaryotes. Annual Review of Biochemistry rhythms. Nature 2012;485:459–64. 2003;72:111–35. [82] Yoshida Y, Iigusa H, Wang N, Hasunuma K. Cross-talk [77] Huber H, Gallenberger M, Jahn U, Eylert E, Berg IA, between the cellular redox state and the circadian system in Kockelkorn D, et al. A dicarboxylate/4-hydroxybutyrate Neurospora. PloS one 2011;6:e28227. autotrophic carbon assimilation cycle in the [83] Buckel W, Golding BT. Radical species in the catalytic hyperthermophilic Archaeum Ignicoccus hospitalis. pathways of enzymes from anaerobes. FEMS Microbiology Proceedings of the National Academy of Sciences Reviews 1998;22:523–41. 2008;105:7851–6. [84] Buckel W, Golding BT. Radical enzymes in anaerobes. [78] Tang K-H, Tang YJ, Blankenship RE. Carbon metabolic Annual Review of Microbiology 2006;60:27–49. pathways in phototrophic bacteria and their broader [85] Matias PM, Pereira IAC, Soares CM, Carrondo MA. Sulphate evolutionary implications. Frontiers in Microbiology respiration from hydrogen in Desulfovibrio bacteria: a 2011;2:1–23. structural biology overview. Progress in Biophysics and [79] Jang HH, Lee KO, Chi YH, Jung BG, Park SK, Park JH, et al. Tw o Molecular Biology 2005;89:292–329. enzymes in one: two yeast peroxiredoxins display oxidative [86] Tabita FR, Hanson TE, Li H, Satagopan S, Singh J, Chan S. stress-dependent switching from a peroxidase to a Function, structure, and evolution of the RubisCO-like molecular chaperone function. Cell 2004;117: proteins and their RubisCO homologs. Microbiology and 625–35. Molecular Biology Reviews 2007;71:576. Research paper 3 80
81 Chapter 5
Discussion 82
5. DISCUSSION
Concentration of CO2 in the atmosphere increased significantly over the last century, and much of this change results from human activities such as deforestation and the widespread use of fossil fuels (Heimann and Reichstein 2008; Frank et al. 2010). In turn, increased CO2 concentrations is thought to have contributed to global warming, resulting in a decrease in the ice cover at the North and South poles as well as in mountain ranges, and ultimately in a rise of sea levels (Bindoff et al. 2007; Jones et al. 2007). The net primary production (NPP) of the planet, which includes the production of organic compounds from atmospheric and dissolved-aquatic 17 -1 CO2, was recently estimated to be approximately 10 g C year , up to 40% of which is produced in the oceans (Geider et al. 2001; Ito 2011). This estimate takes into account CO2 consumption due to oxygenic photosynthesis but does not consider the activity of the so-called “hidden world” of anaerobic autotrophic microorganisms. In ecosystems such as stratified lakes and some marine environments, CO2 assimilation does not only occur via oxygenic photosynthesis, but also through the activity of anoxygenic phototrophic bacteria living in oxic-anoxic boundary zones (Overmann and Garcia-Pichel 2006). The pronounced ability of these organisms to anaerobically fix CO2 in anaerobic conditions suggests that the total NPP on Earth was likely to be underestimated. So far, this “hidden world” remained confined and difficult to study and was therefore omitted from global analyses. In this study, we took the meromictic Lake Cadagno, which harbors a dense community of anoxygenic phototrophic sulfur bacteria, as a model to better understand anoxygenic photoautotrophic organisms. 83 Chapter 5
5.1. Primary production by anoxygenic bacteria
Primary production is defined as the synthesis of organic compounds from atmospheric (CO2) - or aquatic inorganic carbon (HCO3 ). As described previously (see paragraphs 1.2.3), this process can occur through photosynthesis (using light as an energy source) or chemosynthesis by oxidation or reduction of various compounds. Organisms that contribute to primary production are known as autotrophs, and together they form the base of the food chain: photo- and chemo- trophic bacteria, phytoplankton, and plants. Unlike organisms that benefit from oxygenic photosynthesis, anoxygenic phototrophic bacteria may use instead of H2O as electron donor a +2 number of organic substances as well as sulfide (H2S), hydrogen (H2), or ferrous iron (Fe ). Photoautotrophic anaerobic organisms are mainly found in redox transition zones of the freshwater and marine environments as well as in the part of sediments where light can penetrate. As the intensity of light in these environments is generally low, bacteria have evolved highly efficient photosynthetic pigments, which give cells a distinct color ranging from red- purple to green-brownish (see sections 1.2.2. and 1.2.3.1.).
5.1.1. Contribution of phototrophic sulfur bacteria to the primary production
Phototrophic sulfur bacteria represent one of the most important group of anoxygenic photosynthetic organisms and play a key role in the sulfur cycle (Imhoff 2004; Van Gemerden and Mas 2004). Several studies in different meromictic lakes have highlighted the importance of the dense populations of phototrophic sulfur bacteria in the CO2 economy of meromictic lakes, as well as their role as detoxifying agents when using toxic H2S as an electron donor during anoxygenic photosynthesis. Thus, in the presence of phototrophic sulfur bacteria, H2S concentrations in the upper layer of the chemocline tend to decrease during the day but increase again during the night (van Gemerden 1967; Takahashi and Ichimura 1968; Sorokin 1970; Sorokin and Donato 1975; Parkin and Brock 1981; Guerrero et al. 1985; Overmann et al. 1991).
The overnight increase in H2S was interpreted as respiration by phototrophic sulfur bacteria, and for this reason CO2 assimilation was presumed to occur only in the presence of light. In fact, any
CO2 fixation that occurred in the dark was mostly ignored or subtracted from the carbon incorporation that occurred in the presence of light. However, measurable and, in some cases, high levels of CO2 fixation in the dark were reported in a number of meromictic lakes and marine environments (Jorgensen et al. 1979; Cloern et al. 1983; Pedros-Alio and Guerrero 1991; Kuuppo-Leinikki and Salonen 1992; Sorokin et al. 1995), including: Lake Kinneret (Hadas et al.
Discussion 84
2001), lakes from a karstic region in Spain (García-Cantizano et al. 2005; Casamayor et al. 2008), or in Arctic seawater (Alonso-Saez et al. 2010). Incidentally, these studies reported high rates of CO2 fixation in the dark at depths where dense populations of purple sulfur bacteria (PSB) occurred. Under laboratory conditions, various PSB strains showed a pronounced ability to incorporate CO2 in the dark (Van Gemerden and Mas 1995; Van Gemerden and Mas 2004).
However, to date, no significant correlation was made between CO2 fixation in absence of light, nutrient distribution and composition of microbial communities (Casamayor 2010). Yet, overall contribution of phototrophic sulfur bacteria to primary production in stratified lakes was estimated at ca. 28.7% (Overmann 2008).
5.1.2. CO2 fixation in meromictic lakes in absence of light
Autotrophic processes other than photosynthesis were also described in various non-sulfurous freshwater environments (Gorlenko et al. 1983), it seems that alternative processes of CO2 incorporation are more widespread than initially thought. Since dark CO2 fixation (herein referred to as “DCF”) was also reported to contribute significantly to overall inorganic carbon fixation in open marine waters (Prakash et al. 1991), DCF appears to be an important process in marine as well as in freshwater ecosystems. In meromictic lakes, DCF was mostly attributed to chemoautotrophs such as bacteria belonging to the genus Thiobacillus, which oxidize reduced inorganic compounds to obtain the energy to fix CO2 in absence of photosynthesis. Although chemoautotrophic activities carried out by non-photosynthetic prokaryotes were reported for many habitats (Shively et al. 1998a), no potential chemo-autotrophe was detected in Lake Cadagno (Martinez et al. 1983; Tonolla et al. 2003; Gregersen et al. 2009). Yet, the ability of phototrophic sulfur bacteria to fix CO2 in the dark was initially shown using the Wood-Werkman reaction (Wood and Stjernholm 1962), and reported to occur at depths corresponding to dense populations of PSB (Cohen et al. 1977; Camacho et al. 2001; Casamayor et al. 2008). In this respect, our results confirmed that PSB, and in particular Candidatus “T. syntrophicum” strain Cad16T and L. purpurea strain CadA31, exhibited important rates of DCF (Chapter 2).
5.1.3. Autotrophic inorganic carbon fixation pathways
Phototrophic sulfur bacteria are considered extremely versatile and capable of utilizing a multiplicity of metabolic pathways. In general, PSB are capable of both photo-autotrophy and 85 Chapter 5 photo-heterotrophy while green sulfur bacteria (GSB) are considered to be obligate photo- autotrophs (Parkin and Brock 1981; Van Gemerden and Mas 2004). In addition, some PSB strains are capable of chemotrophic growth in absence of light and under microaerophilic conditions (Kämpf and Pfennig 1986; de Wit and van Gemerden 1987). The mechanism of anoxygenic photosynthesis carried out by phototrophic sulfur bacteria has been studied extensively and is now well understood (see section 1.2.3.2.), with H2S being the most common electron donor for PSB and GSB in the process of CO2 assimilation process. In contrast, the chemo-autotrophic mechanisms by which these bacteria fix CO2 in the absence of light remain unclear. PSB are known to produce storage products in presence of light, including sulfur globules, polyhydroxyalkanoates (PHAs) granules, polyphosphate and glycogen (Van Gemerden and Mas 2004). These macromolecules may later serve as energy supplies for DCF processes, as suggested by the substantial levels of fixation retained by photosynthetic organisms once light disappears (García-Cantizano et al. 2005; Casamayor et al. 2008). To date, six distinct mechanisms of CO2 fixation were described: (i) photo- or chemo-autotrophy using the Calvin- Benson-Bassham cycle (CBB cycle), (ii) the reductive citric acid cycle (rTCA cycle), (iii) the reductive acetyl-CoA pathway (Wood-Ljungdahl pathway), (iv) the 3-hydroxypropionate cycle, (v) the 3-hydroxypropionate/4-hydroxybutyrate cycle (HP/HB cycle) and (vi) the dicarboxylate/4-hydroxybutyrate cycle (DC/HB cycle) mechanism (Rothschild 2008; Berg 2011; Hanson et al. 2012). While increase in biomass remains the main purpose for these autotrophic pathways, an alternative function appears to be the disposal of excess reducing power as was observed in purple non-sulfur bacteria (Wang et al. 1993; Joshi et al. 2009). However, none of these six fixation mechanisms has been shown to be involved in DCF in phototrophic sulfur bacteria.
Discussion 86
5.2. Phototrophic sulfur bacteria in the meromictic Lake Cadagno
The dolomite and gypsum found in the bedrock of the valley determine the chemistry and consequently the meromictic state of Lake Cadagno (see sections 1.1.1.). Presence of a purple- red stratum in the upper section of the anaerobic water layers of the neighboring lakes Ritom and Cadagno was first described in the early 20th century (Düggeli 1919; 1924), yet detailed biological analyses begun only about 25 years ago (Peduzzi et al. 1991).
5.2.1. Composition of the phototrophic sulfur bacterial population in the chemocline
Changes in the composition of bacterial populations in Cadagno’s chemocline were analyzed since 1993 with the help of a number of molecular methods (Peduzzi et al. 1993; Tonolla et al. 1999; Bosshard et al. 2000; Peduzzi et al. 2003b; Tonolla et al. 2003; Tonolla et al. 2004; Tonolla et al. 2005). More recently these analyses allowed for the characterization of three new PSB species, two of which belong to the genus Thiocystis (Peduzzi et al. 2011) while the third was determined as a member of the genus Thiodictyon (Peduzzi et al. 2012). Currently the chemocline harbors a dense population of phototrophic sulfur bacteria (see sections 1.1.2.) that is composed of heterogeneous PSB from the Chromatium, Lamprocystis, Thiocystis and Thiodictyon genera, as well as one proposed “clonal population” of Chlorobium clathratiforme GSB, which increased by a factor three since year 2000 (Decristophoris et al. 2009; Gregersen et al. 2009). Although 95% of bacteria found in the chemocline are C. clathratiforme, these cells that are characterized by a cellular volume of ca. 1.2 µm3 represent together only 15% of the total chemocline biovolume (Chapter 2). In contrast, the less abundant (ca. 2.0 105 cells per ml-1, counting in september 2007) PSB Ch. okenii occupies up to 83% of the total biovolume in the chemocline because of a more important cellular volume estimated at 270 µm3. Interestingly, the total number of PSB cells did not vary significantly throughout the years, suggesting that the recent blooming of C. clathratiforme did not occur to the expense of Ch. Okenii and the other PSB populations. Either GSB and PSB did not compete for the same nutrients, or those were abundant enough to sustain an increased population C. clathratiforme. Incidentally, the concomitant presence of both types of phototrophic sulfur bacteria in similar water layers was also observed in other meromictic lakes, yet with a distinct spatial distribution of GSB just below PSB because of a more effective light-harvesting system (van Gemerden and Beeftink 1981; Van Gemerden and Mas 2004; Overmann 2008).
87 Chapter 5
5.2.2. Capacity of PSB to fix carbon
Major populations of phototrophic sulfur bacteria found in the chemocline of Lake Cadagno were analyzed for their ability to fix CO2 using nano-scale secondary-ion mass spectrometry (nanoSIMS)(Musat et al. 2008). Not unexpectedly, metabolic rates of individual bacteria belonging to the same species varied significantly, indicating that microbial populations in the environment include individuals with clearly distinct physiological status. This analysis showed also that the abundant GSB C. clathratiforme assimilated little CO2 when compared to + Lamprocystis purpurea and Ch. okenii and only small amounts of NH4 (Musat et al. 2008; Halm et al. 2009). Later metaproteomic studies further confirmed that fixation of CO2 by C. clathratiforme was restricted to day time and that at most 50% of the population fixed CO2 (Habicht et al. 2011).
As reported in Chapter 3, we compared in situ the fixation rates of pure cultures of four of the most abundant strains of phototrophic sulfur bacteria found in the chemocline: C. clathratiforme strain 4DE, Candidatus “T. syntrophicum” strain Cad16T, Thiocystis chemoclinalis strain CadH11T and Lamprocystis purpurea strain CadA31. First grown in laboratory conditions, cells were then incubated in dialysis bags for five weeks at a depth of 12 m in Lake Cadagno (see Figure 1). At the time of the experiment, C. clathratiforme represented 95.4% (15.3% of the biovolume) of all the phototrophic sulfur bacteria established at this depth. The PSB Candidatus “T. syntrophicum”, L. purpurea and T. chemoclinalis were clearly less abundant, representing respectively 2.0% (1.0% of the biovolume), 0.6% (0.4% of the biovolume) and 0.2% (0.2% of the biovolume) of chemocline bacteria. Our results agree with those of Musat (Musat et al. 2008) and Habicht (Habicht et al. 2011), showing low levels CO2 fixation by C. clathratiforme. Moreover, our results demonstrate the ability of Candidatus “T. syntrophicum” Cad16T and L. purpurea strain CadA31 to fix CO2 at high rates, especially in the dark.
Discussion 88
Figure 6. Experimental setup for incubating in situ and inside dialysis bags cultures phototrophic sulfur bacteria (Storelli et al. 2013). From left to right: Candidatus “T. syntrophicum” strain Cad16T, L. purpurea strain CadA31, Ch. clathratiforme strain 4DE and T. chemoclinalis strain CadH11T.
5.2.3. Candidatus “Thiodictyon syntrophicum” strain Cad16T
Based upon 16S rDNA sequence and MALDI-TOF MS analyses we have recently proposed to reassign strain Cad16T from the genus Lamprocystis to the genus Thiodictyon (Chapter 3). In fact, the 16S rRNA gene of Cad16T was found to be identical to clone 371 (Genbank accession number AJ006061), which characterized the previously uncultured population F of the chemocline in Lake Cadagno (Tonolla et al. 1999). Initially, cells belonging to population F were found to share morphological and physiological characteristics with members of the genus Lamprocystis. Yet, despite these similarities the 16S sequence of clone 371 was only 95.3% and 95.4% similar to corresponding genes of L. purpurea and L. roseopersicina, respectively (Peduzzi et al. 2003a), while it shared higher identity with Thiodictyon bacillosum DSM234 (99.2%) and T. elegans DSM232 (98.9%). The high sequence similarity of 16S rRNA genes within the same genus is a consistent trend in the Chromatiaceae, where the similarity between T. bacillosum and T. elegans is 99.4%; between Lamprocystis purpurea and L. roseopersicina is 99.2%; between Thiocapsa rosea and T. roseopersicina is 98.6%; and between Allochromatium 89 Chapter 5 vinosum and A. minutissimum is 99.0%. However, Cad16T possessed several physiological and cellular features distinct from Thiodictyon species, including the presence of okenone in pigments of the light harvesting antenna and a capacity to grow as a chemo-litho-autotrophe under microaerophilic conditions and in the absence of light. Detailed comparisons were unfortunately not possible since T. bacillosum and T. elegans were not available in public collections of microorganisms.
Interestingly, strain Cad16T forms aggregates in situ with the sulfate-reducing bacterium (SRB) Desulfocapsa thiozymogenes (Peduzzi et al. 2003b). Bacteria in their natural habitats are rarely uniformly distributed. Instead, they often accumulate around substrate sources (Krembs et al. 1998; Fenchel 2002), and sometimes form mutualistic aggregates composed of several species (Boetius et al. 2000; Overmann and Gemerden 2000; Schramm et al. 2000; Jorgensen 2001). The association between D. thiozymogenes and strain Cad16T that can be observed in the chemocline of Lake Cadagno is likely to provide an ecological advantage to one or both of the species and could be centered on the exchange of sulfur compounds. For example, D. T thiozymogenes could produce H2S, which is utilized by the PSB Cad16 during photosynthesis 2- yielding SO4 that is in turn used by the SRB associate. In such mutualistic association, the PSB T Cad16 could stay as close as possible to the light while having access to sufficient H2S to perform an efficient photosynthesis and protecting the SRB from oxygen and elevated H2S concentrations (Peduzzi et al. 2003a). So far, none of the other two members of the genus Thiodictyon were reported to form such close associations with SRB or other bacteria.
Discussion 90
5.3. Proteomic analysis of Candidatus “T. syntrophicum” strain Cad16T
To better understand the molecular basis of the process of CO2 fixation in absence of light, the proteome of Candidatus “T. syntrophicum” strain Cad16T was examined using two-dimensional difference gel electrophoresis (2D-DIGE) (Chapter 4). To this end, cells of Cad16T were grown for several days as anaerobic autotrophic cultures with a light/dark photoperiod of 12 h. Once populations reached 107 cells ml-1, cells were harvested a first time after being exposed for 4 h to light, and also 12 h later once bacteria had been exposed to dark for 4 h. Total cellular proteins were extracted and separated using 2D-DIGE. Approximately 1000 spots were identified using Melanie 7.0 software. Out of the 40 spots that met the statistical ANOVA and intensity thresholds, 23 were found to be up-regulated in presence of light while 17 were more abundant in the dark. Using mass spectrometry, as much as 30 spots were matched against the draft genome of strain Cad16T that was composed of 1352 contigs. Only few genomes of PSB have been completely sequenced (and only partly annotated) including those of Allochromatium vinosum DSM180 (GenBank Assembly ID: GCA_000025485.1), Lamprocystis purpurea DSM4197 (GenBank Assembly ID: GCA_000379525.1), Thiocapsa marina DSM5653 (GenBank Assembly ID: GCA_000223985.2) and Thiocystis violascens DSM198 GenBank Assembly ID: GCA_000227745.3). For this reason, several protein spots were linked to “hypothetical genes” or “genes of unknown function”. pBLAST analysis allowed us to identify conserved regions in a number of these proteins which enabled us to predict their function.
5.3.1. Metabolism of Cad16T when grown in the presence of light
During the light phase, Candidatus “T. syntrophicum” strain Cad16T many proteins responsible for cellular growth accumulated, including ribosomal proteins and enzymes needed for gene transcription and amino acid synthesis. In fact, in the presence of light, Cad16T cells appeared to be metabolically active and to produce carbon compounds (triose phosphate), energy (ATP) and reducing equivalents (NADH/NADPH or reduced ferredoxin) by anaerobic photosynthesis (Overmann and Garcia-Pichel 2006). PSB often take advantage of an excess of photosynthates to synthesize storage polymers which nature varies according to the type substrate that is found in excess. While CO2 fixation normally leads to glycogen formation, excess of acetate or acetyl-CoA leads to the production of polyhydroxyalkanoate (PHA) granules (Mas and Van Gemerden 2004). Formation of storage granules was shown to occur in the PSB C. okenii grown under laboratory conditions, with up to 60% of the bacterial dry mass consisting 91 Chapter 5 in energy reserves (Del Don et al. 1994). In strain Cad16T, three proteins potentially involved in the synthesis of polyhydroxybutyrate (PHB) were found to be up-regulated in presence of light. Formation PHB granules is commonly observed in phototrophic and chemotrophic bacteria (Hazer et al. 2012). PHB that is one of the most common PHAs, consists of 1,000 to 30,000 hydroxy fatty acid monomers that form a granule in the cytoplasm of many microbial cells. Generally, PHAs represent storage compounds for carbon and energy sources and are synthesized under unbalanced growth conditions, i.e. when a carbon source is found in excess while concentration of another nutrient is growth-limiting. PHB synthesis in Alcaligenes eutrophus was shown to be stimulated when intracellular concentrations of NAD(P)H and acetyl-CoA are unusually high. Citrate synthase activity was also inhibited during the synthesis of PHAs, facilitating the metabolic flux of acetyl-CoA to the PHB synthetic pathway (Haywood et al. 1988). A recent study showed that inhibiting the PHA-cycle in Pseudomonas putida KT2442 caused an increase in acetyl-CoA in cells grown under low nitrogen conditions. The excess acetyl-CoA stimulated the activity of the TCA cycle and the glyoxylate cycle, leading to
CO2 production instead of biomass generation. This suggested that the ability of P. putida to synthesize PHA was probably required to channel excess energy and maintain the energy balance (Escapa et al. 2012).
Intringuigly, none of the proteins that appeared as more abundant in Cad16T cells exposed to light matched form I of RuBisCO (CbbL), which was expected because of the high rate of CO2 fixation in these growth conditions (see Chapter 2). Instead, several protein spots of 52 kDa that remained stable throughout the conditions were identified as isoforms of the large subunit of RuBisCO form II (CbbM). Previously, we showed by qRT-PCR that expression of cbbM remained constant and at a basal level throughout the day (Storelli et al. 2013), suggesting a constitutive expression that was not modulated by environmental stimuli. Differences in the expression of RuBisCO form I and form II were also reported for other phototrophic bacteria, suggesting that in the presence of a reduced carbon source CbbM functions primarily as a terminal electron acceptor to maintain the redox balance of the cell (Wang et al. 1993; Yoshizawa et al. 2004; Badger and Bek 2008; Joshi et al. 2009; Laguna et al. 2010). Thus, the 2D-DIGE results we obtained concur with the hypothesis that form II of RuBisCO is not necessarily directly involved in CO2 fixation during photosynthesis.
Enzymes required for the CBB cycle are generally found in the cytoplasm of most autotrophic bacteria. In some autotrophs, RuBisCO accumulates specifically within polyhedral inclusion bodies known as carboxysomes, however (Shively et al. 1998b; Yeates et al. 2008; Cannon et al.
Discussion 92
2010). In this respect, genes potentially involved in the formation of carboxysome-like micro- compartments were identified downstream of cbbL and cbbS in the draft genome of strain Cad16T (GenBank No. JQ693375-JQ693380). The apparent absence of form I RuBisCO (CbbL) in our dataset may be due to technical difficulties during protein purification and/or separation because of a possible sequestration of RuBisCO in carboxysome-like micro-compartments.
5.3.2. Metabolism of Cad16T in absence of light
One of the aims of the proteomic analysis was to elucidate the metabolic process that allowed T strain Cad16 to fix CO2 in the dark. Among the 17 protein spots that were significantly up- regulated in the dark, three in particular are of interest due to their potential role in the autotrophic dicarboxylate-hydroxybutyrate (DC/HB) cycle (e.g., enzymes 4, 21 and 22 in Figure 7). This cycle was recently discovered in the hyperthermophilic Archaeum Ignicoccus hospitalis by Huber et al. (Huber et al. 2008), and is presumed to be present only in Archaea. The cycle converts acetyl-CoA and two inorganic carbons (CO2) to succinyl-CoA, using essentially the same enzymes as the rTCA cycle (see Figure 7 B). The carboxylases involved in this cycle, which are responsible for inorganic carbon fixation, are pyruvate synthase and phosphoenolpyruvate (PEP) carboxylase. After the inorganic carbon fixation is complete, the acetyl-CoA must be regenerated. First, the CO2 fixation product succinyl-CoA is reduced to 4- hydroxybutyrate, which is activated to 4-hydroxybutyryl-CoA and then dehydrated to crotonyl- CoA by 4-hydroxybutyryl-CoA dehydratase. 4-hydroxybutyryl-CoA dehydratase is considered a key enzyme in this DC/HB cycle. Crotonyl-CoA is further modified to (S)-3-hydroxybutyryl- CoA, then acetoacetyl-CoA and finally to two acetyl-CoA molecules. The DC/HB cycle generates an additional molecule of acetyl-CoA (see Figure 7 A) and overall uses eight ATP molecules and 10 reducing factors to fix H2CO3 and CO2 into a triose phosphate (glyceraldehyde-6-phosphate). A similar cycle, the hydroxypropionate/hydroxybutyrate (HP/HB) cycle (see Figure 7 B) that was described in the aerobic autotrophic Sulfolobales (Archaea), involves another set of carboxylases for fixing CO2 (the acetyl-CoA/propionyl-CoA carboxylase).
93 Chapter 5
Figure 7. (A) The dicarboxylate/4-hydroxybutyrate cycle described in Desulfurococcales and Thermoproteales; (B) the 3-hydroxypropionate/4-hydroxybutyrate cycle described in Sulfolobales.
Enzymes: 1, pyruvate synthase; 2, pyruvate:water dikinase; 3, PEP carboxylase; 4, malate dehydrogenase; 5, fumarate hydratase; 6, fumarate reductase (natural electron acceptor is not known); 7, succinyl-CoA synthetase; 8, acetyl-CoA/propionyl-CoA carboxylase; 9, malonyl-CoA reductase; 10, malonic semialdehyde reductase; 11, 3- hydroxypropionate-CoA ligase; 12, 3-hydroxypropionyl-CoA dehydratase; 13, acryloyl-CoA reductase; 14, methylmalonyl-CoA epimerase; 15, methylmalonyl-CoA mutase; 16, succinyl-CoA reductase; 17, succinic semialdehyde reductase; 18, 4-hydroxybutyrate-CoA ligase; 19, 4-hydroxybutyryl-CoA dehydratase; 20, crotonyl- CoA hydratase; 21, (S)-3-hydroxybutyryl-CoA dehydrogenase (NAD+); 22, acetoacetyl-CoA-ketothiolase. (Berg 2011).
One of the key enzymes in the DC/HB cycle is the 4-hydroxybutyryl-CoA dehydratase. This enzyme is considered a “radical enzyme”, as it uses radicals as intermediates during the metabolic reactions. Because they are highly reactive towards dioxygen, radicals are often used in catabolic reactions in anaerobic micro-organisms (Buckel and Golding 1998; Matias et al. 2005; Buckel and Golding 2006). Among the 17 proteins that were up-regulated in the dark,
Discussion 94 three enzymes involved in anti-oxidant stress responses such as a superoxide dismutase and two peroxiredoxin were identified (see Table 2 in the research paper 3). Thus, the enhanced activity of the 4-hydroxybutyryl-CoA dehydratase in the dark could result in the production of harmful free radicals that could be inactivated by an increased expression of antioxidant enzymes.
The HB/DC cycle could be hypothetically separated in two independent part, the dicarboxilate and the hydroxybutyrate part. The first is composed of the same enzymes of the initially part of the rTCA cycle (see Figure 8 A), where our up-regulated enzyme (number 2, malate deshydrogenase) reversely transform the oxaloacetate in malate. In others words, this enzyme could be involved in the hypothetical DC/HB cycle but also in the TCA or rTCA cycle, producing ATP and reducing power or fixing CO2 respectively. In the other hand, two others proteins up-regulated in the dark corresponding to the last steps of the DC/HB cycle, which participate also in the PHB degradation pathway (see Figure 8 B). This enzymes reversely catalyzed the final steps of the PHB granules degradation (see Figure 8 B, enzymes PhaB: 3- ketoacyl-ACP reductase and PhaA: acetoacetyl-CoA-ketothiolase). Interestingly, in presence of light the up-regulation of three enzymes involved in the synthesis of PHB globules was shown, which presumes a high concentration of reserve globules in the bacterial cell. How said before, the degradation of the PHB granules appears to produce acetyl-CoA and increase the overall reducing power, such as concentration of available NADPH. In purple non-sulfur bacteria was shown that the form II of the RuBisCO enzyme is commonly involved in the process of CO2 fixation as an electron sink for reducing equivalents derived from the oxidation of reserve compounds (Dubbs and Robert Tabita 2004; Joshi et al. 2009; Laguna et al. 2010). A similar operon structure as that found in purple non-sulfur bacteria is also present in the draft genome of strain Cad16T, with two distinct regions harboring genes coding for RuBisCO form I and form II. So we could suppose that the reducing power resulting from the PHB degradation can be utilized in the process of dark CO2 fixation. Moreover, the excess in acetyl-CoA is most probably used to generate additional reducing power via the TCA or glyoxylate cycle. In both of these cycles, one of our up-regulated enzymes (malate dehydrogenase) reversibly catalyzes the oxidation of malate to oxaloacetate by reducing NAD+ to NADH. 95 Chapter 5
Figure 8. (A) The reductive citric acid (rTCA) cycle. The pathway of acetyl-CoA assimilation to pyruvate, phosphoenolpyruvate (PEP), and oxaloacetate is also shown (Berg 2011). (B) For deviations from this variant of the cycle, see the text. (B) Cyclic metabolism of PHB biosynthesis and degradation in bacteria (Sudesh et al. 2000).
Enzymes rTCA: 1, ATP-citrate lyase; 2, malate dehydrogenase; 3, fumarate hydratase; 4, fumarate reductase (natural electron donor is not known); 5, succinyl-CoA synthetase; 6, ferredoxin (Fd)-dependent b2-oxoglutarate synthase; 7, isocitrate dehydrogenase; 8, aconitate hydratase; 9, Fd-dependent pyruvate synthase; 10, PEP synthase; 11, PEP carboxylase.
Enzymes PHB synthesis/degradation: PhaA, acetoacetyl-CoA-ketothiolase; PhaB, 3-ketoacyl-ACP reductase; PhaC, PHA synthase; PhaZ, PHA depolymerase; 1, dimer hydrolase; 2, (R)-3-hydroxybutyrate dehydrogenase; 3, acetoacetyl-CoA synthetase; 4, NADH-dependent acetoacetyl-CoA reductase.
Discussion 96
5.4. Conclusions and perspectives
The metabolism of phototrophic sulfur bacteria still remain poorly understood. Anoxygenic phototrophic sulfur bacteria occupy regions where light reaches anoxic layers in water columns or in sediments. The mechanism of CO2 fixation in the presence of light by the Calvin-Benson- Bassham cycle (present in PSB) or the reverse tricarboxylic acid cycle (present in GSB) is relatively well characterized, while the assimilation of inorganic carbon in the dark remains poorly understood. However, previous studies have confirmed the pronounced capacity of PSB to fix CO2 in the dark (Wood and Stjernholm 1962; Cohen et al. 1977; Camacho et al. 2001; García-Cantizano et al. 2005; Casamayor et al. 2008; Casamayor 2010; Casamayor et al. 2012).
This study aimed primarily at determining the ecological role of the phototrophic sulfur bacteria in the primary production of the Lake Cadagno. Previous reports showed that despite a volume of only 10% of Lake Cadagno, the chemocline is responsible for up to 40% of the total inorganic carbon photo-assimilation, and that significant rates of CO2 assimilation occurred during the night, indicating that primary production also relies on mechanisms other than photosynthesis (Camacho et al. 2001). In Chapter 2, we confirmed that carbon assimilation occurred at the rate of approximately 297 ng of 14C mL-1 in the light and approximately 231 ng of 14C mL-1 in the dark. Amongst the four major population living in the chemocline tested during the initial in situ experiment, the dominant small-celled species Candidatus “Thiodictyon syntrophicum” strain Cad16T was estimated to be responsible for up to 25% of the total primary production in the chemocline. Following a detailed taxonomic characterization, Cad16T was proposed as the type strain of a new species belonging to the genus Thiodictyon (see Chapter 3). Laboratory experiments using pure cultures of strain Cad16T grown in autotrophic media showed that the maximal CO2 assimilation rate occurred during the first 4 h of light (07:00 to 11:00 AM). Furthermore, mRNA analyses confirmed that the two genes in the draft genome of strain Cad16T coding for RuBisCO forms I (CbbL) and II (CbbM) were differentially expressed. While cbbM was constitutively expressed at a basal level throughout light and dark periods, the expression of cbbL varied during the light-dark cycle and was modulated by the available carbon sources. Yet, peaks in cbbL expression did not correlate with the periods of maximal CO2 assimilation.
Using a proteomic approach (Chapter 4), we were able to show the presence of three proteins that were up-regulated in the dark and may be involved in the anaerobic dicarboxylate/4- hydroxybutyrate (DC/HB) cycle described until now only in Archaea populations (Huber et al. 2008; Berg 2011). However, these three enzymes are also part of two other pathways: the 97 Chapter 5
(r)TCA cycle (malate dehydrogenase) and the PHB degradation pathway (3-ketoacyl-ACP reductase and acetoacetyl-CoA-ketothiolase). Moreover, strain Cad16T could hypothetically fix
CO2 in the dark via the CBB cycle. As shown in other studies, when the two different forms of the enzyme RuBisCO are present, RuBisCO form II is typically involved in maintaining the redox balance when the cells have an excess of reducing power (Dubbs and Robert Tabita 2004; Joshi et al. 2009; Laguna et al. 2010). Both hypotheses need to be tested by additional analyses, such as quantification of key genes by qRT-PCR and other proteomics assays. Nevertheless, the proteomic approach (2D-DIGE) performed in this study using a phototrophic sulfur bacterium has important biological relevance because it provides concrete indications of the mechanism involved in the process of CO2 assimilation in the dark. In the presence of light, PSB accumulate storage compounds such as intracellular sulfur globules, polyhydroxybutyrate (PHB) granules, polyphosphate and glycogen (Del Don et al. 1994; Van Gemerden and Mas 2004). Proteomic data confirmed the presence of enzymes potentially involved in the production of PHB, the most common polyhydroxyalkanoate (PHA). PHB that serves as a carbon and energy reserved, also represents a sink for reducing equivalents. We propose that the energy and reducing power stored in the PHB granules play a key role in the process of dark CO2 fixation. This suggests that
PSB are capable of CO2 fixation in the dark because they accumulated sufficient reserves during the light period. Moreover, PHB is a biotechnologically interesting polymer since it is biodegradable and derived from renewable resources. PHAs in general have attracted much interest and have been used in the development of many technical and medical applications in recent years (Chen 2009; Keshavarz and Roy 2010). Currently, commercial products (e.g., Biopol®, manufactured by Zeneca BioProducts) are synthesized by glucose fermentation using Ralstonia eutropha H16 mutants (Asrar and Gruys 2002). The possibility of producing this commercial product using light instead of glucose could be an interesting future application.
Another interesting topic for future investigation is the characterization of hypothetical carboxysome-like microcompartments in which form I RuBisCO (CbbL-CbbS) may accumulate and thus enhance CO2 fixation. All of the genes required for the formation of these microcompartments are present in the draft genome of strain Cad16T. As in other microorganisms, the operon containing cbbL (JQ693373) also includes cbbS (JQ693374), as well as six genes putatively coding for components of a carboxysome-like microcompartment (JQ693375-JQ693380) (Cannon et al. 2010).
Discussion 98
In conclusion, our analyses showed that while GSB represent approximately 95% of the total population of phototrophic sulfur bacteria, it is the minor fraction (ca. 5%) of PSB that carry out most of the CO2 fixation in the chemocline of Lake Cadagno. Amongst PSB, it appears that the newly characterized type strain Cad16T which accounts for approximately 30% of the total PSB population, is the most active CO2 fixing microorganism regardless of the presence of absence of light. Surprisingly, transcription patterns of the RuBisCo encoding genes cbbL and cbbM did not T correlate with maximal levels of CO2 fixation indicating that primary production by Cad16 involved more complex mechanisms than just presence of RuBisCO. For example, presence of carboxysomes need to be further explored. In addition, CO2 fixation in absence of light may require reserve polymers, such as PHB, that are accumulated during the day and will provide enough energy and reducing power during the night. The DC/HB and/or the CBB cycles were identified as candidate pathways for the dark CO2 fixation process. This study represents a strong starting point for future analyses to further explore the dark CO2 fixation observed in the model PSB strain Cad16T.
99 Chapter 6
References 100
6. REFERENCES
Abella, C., Montesinos, E. and Guerrero, R. (1980) Shallow Lakes Contributions to their Limnology: Springer Netherlands.
Alban, A., David, S.O., Bjorkesten, L., Andersson, C., Sloge, E., Lewis, S. and Currie, I. (2003) A novel experimental design for comparative two-dimensional gel analysis: Two- dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 3, 36-44.
Alonso-Saez, L., Galand, P.E., Casamayor, E.O., Pedros-Alio, C. and Bertilsson, S. (2010) High bicarbonate assimilation in the dark by Arctic bacteria. The ISME Journal 4, 1581-1590.
Andersen, J.S. and Mann, M. (2000) Functional genomics by mass spectrometry. FEBS Letters 480, 25-31.
Asrar, J. and Gruys, K.J. (2002) Biodegradable Polymer (Biopol). Biopolymers 4, 53-81.
Badger, M.R. and Bek, E.J. (2008) Multiple Rubisco forms in proteobacteria: their functional significance in relation to CO2 acquisition by the CBB cycle. J Exp Botany 59, 1525-1541.
Bak, F. and Cypionka, H. (1987) A novel type of energy metabolism involving fermentation of inorganic sulphur compounds. Nature 326, 891-892.
Barton, L.L. and Fauque, G.D. (2009) Biochemistry, physiology and biotechnology of sulfate- reducing bacteria. Adv Appl Microbiol 68, 41-98.
Berg, I.A. (2011) Ecological aspects of the distribution of different autotrophic CO2 fixation pathways. Appl Environm Microbiol 77, 1925-1936.
Bindoff, N.L., Willebrand, J., Artale, V., Cazenave, A., Gregory, J.M., Gulev, S., Hanawa, K., Le Quere, C., Levitus, S. and Nojiri, Y. (2007) Observations: oceanic climate change and sea level.
Birch, L., Hanselmann, K.W. and Bachofen, R. (1996) Heavy metal conservation in Lake Cadagno sediments: Historical records of anthropogenic emissions in a meromictic alpine lake. Water Research 30, 679-687.
Blankenship, R.E. and Hartman, H. (1998) The origin and evolution of oxygenic photosynthesis. Trends in Biochemical Sciences 23, 94-97. 101 Chapter 6
Boetius, A., Ravenschlag, K., Schubert, C., Rickert, D., Widdel, F., Gieseke, A., Amann, R., Jorgensen, B., Witte, U. and Pfannkuche, O. (2000) A marine microbial consortium apparently mediating anaerobic oxidation of methane. Nature 407, 623-625.
Bosshard, P.P., Santini, Y., Gruter, D., Stettler, R. and Bachofen, R. (2000) Bacterial diversity and community composition in the chemocline of the meromictic alpine Lake Cadagno as revealed by 16S rDNA analysis. FEMS Microbiol Ecol 31, 173-182.
Brocks, J. and Schaeffer, P. (2008) Okenane, a biomarker for purple sulfur bacteria (Chromatiaceae), and other new carotenoid derivatives from the 1640Ma Barney Creek Formation. Geochimica et Cosmochimica Acta 72, 1396-1414.
Brown, K.A. (1982) Sulphur in the environment: a review. Environmental Pollution Series B, Chemical and Physical 3, 47-80.
Brown, S.R. and McIntosh, H.J. (1987) The fossil history of sulfur phototrophs in a meromictic lake ecosystem. Acta Academica Aboensis 47, 83-95.
Brune, D. (1995a) Sulfur compounds as photosynthetic electron donors. Dordrecht/The Netherlands Kluwer Academic Publishers.
Brune, D.C. (1995b) Isolation and characterization of sulfur globule proteins from Chromatium vinosum and Thiocapsa roseopersicina. Arch of Microbiol 163, 391-399.
Buckel, W. and Golding, B.T. (1998) Radical species in the catalytic pathways of enzymes from anaerobes. FEMS Microbiol Rev 22, 523-541.
Buckel, W. and Golding, B.T. (2006) Radical Enzymes in Anaerobes. Annu Rev Microbiol 60, 27-49.
Camacho, A., Erez, J., Chicote, A., Florin, M., Squires, M., Lehmann, C. and Bachofen, R. (2001) Microbial microstratification, inorganic carbon photoassimilation and dark carbon fixation at the chemocline of the meromictic Lake Cadagno (Switzerland) and its relevance to the food web. Aquatic sciences 63, 91-106.
Cannon, G.C., Heinhorst, S. and Kerfeld, C.A. (2010) Carboxysomal carbonic anhydrases: structure and role in microbial CO2 fixation. BBA-Proteins & Proteomics 1804, 382-392.
Casamayor, E.O. (2010) Vertical distribution of planktonic autotrophic Thiobacilli and dark
CO2 fixation rates in lakes with oxygen-sulfide interfaces. Aquatic Microbial Ecol 59, 217-228.
References 102
Casamayor, E.O., Garcia-Cantizano, J. and Pedros-Alio, C. (2008) Carbon dioxide fixation in the dark by photosynthetic bacteria in sulfide-rich stratified lakes with oxic-anoxic interfaces. Limnol Oceanogr 53, 1193-1203.
Casamayor, E.O., Lliros, M., Picazo, A., Barberan, A., Borrego, C.M. and Camacho, A.
(2012) Contribution of deep dark fixation processes to overall CO2 incorporation and large vertical changes of microbial populations in stratified Karstic lakes. Aquatic Sciences-Research Across Boundaries 74, 61-75.
Chen, G.-Q. (2009) A microbial polyhydroxyalkanoates (PHA) based bio-and materials industry. Chemical Society Reviews 38, 2434-2446.
Cloern, J.E., Cole, B.E. and Oremland, R.S. (1983) Autotrophic processes in meromictic Big Soda Lake, Nevada. Limnol Oceanogr 28, 1049-1061.
Cohen, Y., Krumbein, W.E. and Shilo, M. (1977) Solar Lake (Sinai). 2. Distribution of photosynthetic microorganisms and primary production. Limnol Oceanogr 22, 609-620.
Coolen, M.J.L. and Overmann, J. (1998) Analysis of subfossil molecular remains of purple sulfur bacteria in a lake sediment. Appl Environm Microbiol 64, 4513-4521.
Cypionka, H., Smock, A.M. and Böttcher, M.E. (1998) A combined pathway of sulfur compound disproportionation in Desulfovibrio desulfuricans. FEMS Microbiol Lett 166, 181- 186.
Dahl, T.W., Anbar, A.D., Gordon, G.W., Rosing, M.T., Frei, R. and Canfield, D.E. (2010) The behavior of molybdenum and its isotopes across the chemocline and in the sediments of sulfidic Lake Cadagno, Switzerland. Geochimica et Cosmochimica Acta 74, 144-163.
de Wit, R. and van Gemerden, H. (1987) Chemolithotrophic growth of the phototrophic sulfur bacterium Thiocapsa roseopersicina. FEMS Microbiol Letts 45, 117-126.
Decristophoris, P.M.A., Peduzzi, S., Ruggeri-Bernardi, N., Hahn, D. and Tonolla, M. (2009) Fine scale analysis of shifts in bacterial community structure in the chemocline of meromictic Lake Cadagno, Switzerland. J Limnol 68, 16-24.
Del Don, C., Hanselmann, K.W., Peduzzi, R. and Bachofen, R. (1994) Biomass composition and methods for the determination of metabolic reserve polymers in phototrophic sulfur bacteria. Aquatic sciences 56, 1-15. 103 Chapter 6
Del Don, C., Hanselmann, K.W., Peduzzi, R. and Bachofen, R. (2001) The meromictic alpine Lake Cadagno: orographical and biogeochemical description. Aquatic Sciences-Research Across Boundaries 63, 70-90.
Dubbs, J. and Robert Tabita, F. (2004) Regulators of nonsulfur purple phototrophic bacteria and the interactive control of CO2 assimilation, nitrogen fixation, hydrogen metabolism and energy generation. FEMS Microbiol Rev 28, 353-376.
Düggeli, M. (1919) Die Schwefelbakterien. Neujahrsblatt 121 der Naturforschenden Gesellschaft, Zürich.
Düggeli, M. (1924) Hydrobiologische Untersuchungen im Pioragebiet. Aquatic Sciences- Research Across Boundaries 2, 65-205.
Egli, K., Wiggli, M., Klug, J. and Bachofen, R. (1998) Spatial and temporal dynamics of the cell density in a plume of phototrophic microorganisms in their natural environment. Doc Ist Ital Idrobiol 63, 121-126.
Escapa, I.F., Garcia, J.L., Bühler, B., Blank, L.M. and Prieto, M.A. (2012) The polyhydroxyalkanoate metabolism controls carbon and energy spillage in Pseudomonas putida. Environml Microbiol 4, 1049-1063.
Fenchel, T. (2002) Microbial behavior in a heterogeneous world. Science Signaling 296, 1068- 1071.
Fenyö, D., Qin, J. and Chait, B.T. (1998) Protein indentification using mass spectrometric information. Electrophoresis 19, 998-1005.
Fischer, C., Wiggli, M., Schanz, F., Hanselmann, K.W. and Bachofen, R. (1996) Light environment and synthesis of bacteriochlorophyll by populations of Chromatium okenii under natural environmental conditions. FEMS Microbiol Ecol 21, 1-9.
Frank, D.C., Esper, J., Raible, C.C., Büntgen, U., Trouet, V., Stocker, B. and Joos, F. (2010) Ensemble reconstruction constraints on the global carbon cycle sensitivity to climate. Nature 463, 527-530.
Frigaard, N.-U. and Bryant, D. (2006) Chlorosomes: antenna organelles in photosynthetic green bacteria. Complex intracellular structures in prokaryotes 2, 79-114.
Frigaard, N.U. and Dahl, C. (2008) Sulfur metabolism in phototrophic sulfur bacteria. Adv Microb physiol 54, 103-200.
References 104
García-Cantizano, J., Casamayor, E., Gasol, J., Guerrero, R. and Pedrós-Alió, C. (2005)
Partitioning of CO2 incorporation among planktonic microbial guilds and estimation of in situ specific growth rates. Microbial Ecol 50, 230-241.
Garrity, G.M., Lilburn, T.G., Cole, J.R., Harrison, S.H., Euzéby, J. and Tindall, B.J. (2007) Taxonomic outline of the Bacteria and Archaea. Part 6–Phylum “Proteobacteria”: The Deltaproteobacteria and Epsilonproteobacteria. Taxonomic Outline of Bacteria and Archaea Release 7.7, 148-245.
Geider, R.J., Delucia, E.H., Falkowski, P.G., Finzi, A.C., Grime, J.P., Grace, J., Kana, T.M., La Roche, J., Long, S.P. and Osborne, B.A. (2001) Primary productivity of planet earth: biological determinants and physical constraints in terrestrial and aquatic habitats. Global Change Biology 7, 849-882.
George, G.N., Gnida, M., Bazylinski, D.A., Prince, R.C. and Pickering, I.J. (2008) X-ray absorption spectroscopy as a probe of microbial sulfur biochemistry: the nature of bacterial sulfur globules revisited. J Bacteriol 190, 6376-6383.
Gharbi, S., Gaffney, P., Yang, A., Zvelebil, M.J., Cramer, R., Waterfield, M.D. and Timms, J.F. (2002) Evaluation of two-dimensional differential gel electrophoresis for proteomic expression analysis of a model breast cancer cell system. Molecular & Cellular Proteomics 1, 91-98.
Gorlenko, V.M., Dubinina, G.A., Kuznecov, S.I., Ohle, W. and Wareing, H. (1983) The ecology of aquatic micro-organisms. Stuttgart: E. Schweizerbart'sche Verlagsbuchhandlung.
Gregersen, L.H., Habicht, K.S., Peduzzi, S., Tonolla, M., Canfield, D.E., Miller, M., Cox, R.P. and Frigaard, N.U. (2009) Dominance of a clonal green sulfur bacterial population in a stratified lake. FEMS Microbiol Ecol 70, 30-41.
Guerrero, R., Mas, J. and Pedros-Alio, C. (1984) Buoyant density changes due to intracellular content of sulfur in Chromatium warmingii and Chromatium vinosum. Arch Microbiol 137, 350- 356.
Guerrero, R., Montesinos, E., Pedros-Alio, C., Esteve, I., Mas, J., Van Gemerden, H., Hofman, P. and Bakker, J. (1985) Phototrophic sulfur bacteria in two Spanish lakes: vertical distribution and limiting factors. Limnol Oceanogr 30, 919-931. 105 Chapter 6
Habicht, K.S., Miller, M., Cox, R.P., Frigaard, N.U., Tonolla, M., Peduzzi, S., Falkenby, L.G. and Andersen, J.S. (2011) Comparative proteomics and activity of a green sulfur bacterium through the water column of Lake Cadagno, Switzerland. Environme Microbiol 13, 203-215.
Hadas, O., Pinkas, R. and Erez, J. (2001) High chemoautotrophic primary production in Lake Kinneret, Israel: A neglected link in the carbon cycle of the lake. Limnol Oceanogr 46, 1968- 1976.
Halm, H., Musat, N., Lam, P., Langlois, R., Musat, F., Peduzzi, S., Lavik, G., Schubert, C.J., Singha, B. and LaRoche, J. (2009) Co occurrence of denitrification and nitrogen fixation in a meromictic lake, Lake Cadagno (Switzerland). Environm Microbiol 11, 1945-1958.
Hanselmann, K. and Hutter, R. (1998) Geomicrobiological coupling of sulfur and iron cycling in anoxic sediments of a meromictic lake: sulfate reduction and sulfide sources and sinks in Lake Cadagno. Doc Ist Ital Idrobiol 63, 85-98.
Hanson, T.E., Alber, B.E. and Tabita, F.R. (2012) Phototrophic CO2 fixation: Recent insights into ancient metabolisms. Functional Genomics and Evolution of Photosynthetic Systems. 33, 225-251.
Hao, O.J., Chen, J.M., Huang, L. and Buglass, R.L. (1996) Sulfate-reducing bacteria. Critical reviews in environmental science and technology 26, 155-187.
Haywood, G.W., Anderson, A.J., Chu, L. and Dawes, E.A. (1988) The role of NADH-and NADPH-linked acetoacetyl-CoA reductases in the poly-3-hydroxybutyrate synthesizing organism Alcaligenes eutrophus. FEMS Microbiol Lett 52, 259-264.
Hazer, D.B., Kiliçay, E. and Hazer, B. (2012) Poly (3-hydroxyalkanoate) s: Diversification and biomedical applications: A state of the art review. Materials Science and Engineering 32, 637-647.
Heimann, M. and Reichstein, M. (2008) Terrestrial ecosystem carbon dynamics and climate feedbacks. Nature 451, 289-292.
Hell, R., Dahl, C., Knaff, D. and Leustek, T. (2008) Sulfur metabolism. In phototrophic organisms: Springer Netherlands.
Huber, H., Gallenberger, M., Jahn, U., Eylert, E., Berg, I.A., Kockelkorn, D., Eisenreich, W. and Fuchs, G. (2008) A dicarboxylate/4-hydroxybutyrate autotrophic carbon assimilation cycle in the hyperthermophilic Archaeum Ignicoccus hospitalis. Proceedings of the National Academy of Sciences 105, 7851-7856.
References 106
Hutchinson, G.E. (1937) A Contribution to the Limnology of Arid Regions, Primarily Founded on Observations Made in the Lahontan Basin: Connecticut Academy of Arts and Sciences.
Imboden, D.M. and Wüest, A. (1995) Mixing Mechanisms in Lakes. In: Physics and chemistry of lakes, Springer Berlin Heidelberg.
Imhoff, J.F. (2004) Taxonomy and physiology of phototrophic purple bacteria and green sulfur bacteria. In Anoxygenic photosynthetic bacteria, Springer Netherlands.
Ito, A. (2011) A historical meta-analysis of global terrestrial net primary productivity: are estimates converging? Global Change Biology 17, 3161-3175.
Jensen, O.N., Podtelejnikov, A.V. and Mann, M. (1997) Identification of the components of simple protein mixtures by high-accuracy peptide mass mapping and database searching. Analytical Chemistry 69, 4741-4750.
Jones, P.D., Trenberth, K.E., Ambenje, P.G., Bojariu, R., Easterling, D.R., Klein, T.G., Parker, D.E., Renwick, J.A., Rusticucci, M. and Soden, B. (2007) Observations: surface and atmospheric climate change. IPCC, Climate change, 235-336.
Jorgensen, B.B. (2001) Life in the diffusive boundary layer. In The benthic bondary layer, transport processes and biogeochemistry, New York: Oxfort University press.
Jörgensen, B.B. (1990) A thiosulfate shunt in the sulfur cycle of marine sediments. Science 249, 152-154.
Jorgensen, B.B. and Des Marais, D.J. (1986) A simple fiber-optic microprobe for high resolution light measurements: application in marine sediment. Limnol Oceanogr 31, 1376-1383.
Jörgensen, B.B., Fossing, H., Wirsen, C.O. and Jannasch, H.W. (1991) Sulfide oxidation in the anoxic Black Sea chemocline. Deep Sea Research Part A Oceanographic Research Papers 38, S1083-S1103.
Jorgensen, B.B., Revsbech, N.P., Blackburn, T.H. and Cohen, Y. (1979) Diurnal cycle of oxygen and sulfide microgradients and microbial photosynthesis in a cyanobacterial mat sediment. Appl Environm Microbiol 38, 46-58.
Joshi, G.S., Romagnoli, S., VerBerkmoes, N.C., Hettich, R.L., Pelletier, D. and Tabita, F.R. (2009) Differential accumulation of form I RubisCO in Rhodopseudomonas palustris CGA010 under photoheterotrophic growth conditions with reduced carbon sources. J Bacteriol 191, 4243- 4250. 107 Chapter 6
Kämpf, C. and Pfennig, N. (1986) Chemoautotrophic growth of Thiocystis violacea,
Chromatium gracile and C. vinosum in the dark at various O2-concentrations. J Basic Microbiol 26, 517-531.
Kelly, D.P. (1999) Thermodynamic aspects of energy conservation by chemolithotrophic sulfur bacteria in relation to the sulfur oxidation pathways. Arch Microbiol 171, 219-229.
Keshavarz, T. and Roy, I. (2010) Polyhydroxyalkanoates: bioplastics with a green agenda. Curr Op Microbiol 13, 321-326.
Klose, J. (1975) Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. Humangenetik 26, 231-243.
Koizumi, Y., Kojima, H., Oguri, K., Kitazato, H. and Fukui, M. (2004) Vertical and temporal shifts in microbial communities in the water column and sediment of saline meromictic Lake Kaiike (Japan), as determined by a 16S rDNA-based analysis, and related to physicochemical gradients. Environm Microbiol 6, 622-637.
Krembs, C., Juhl, A.R., Long, R.A. and Azam, F. (1998) Nanoscale patchiness of bacteria in lake water studied with the spatial information preservation method. Limnol Oceanogr 43, 307- 314.
Krige, L.J. (1918) Petrographische Untersuchungen im Val Piora und Umgebung, Dissert. Ecologae geologicae Helveticae 14, 549-654.
Kuuppo-Leinikki, P. and Salonen, K. (1992) Bacterioplankton in a small polyhumic lake with an anoxic hypolimnion. Hydrobiol 229, 159-168.
Laguna, R., Joshi, G.S., Dangel, A.W., Luther, A.K. and Tabita, F.R. (2010) Integrative control of carbon, nitrogen, hydrogen, and sulfur metabolism: the central role of the Calvin- Benson-Bassham cycle. Recent Advances in Phototrophic Prokaryotes 675, 265-271.
Law, C.J., Roszak, A.W., Southall, J., Gardiner, A.T., Isaacs, N.W. and Cogdell, R.J. (2004) The structure and function of bacterial light-harvesting complexes (Review). Molecular membrane biology 21, 183-191.
Lehmann, C., Luehty, L. and Bachofen, R. (1998) Tools for the evaluation of sources and sinks of sulfide in Lake Cadagno. Doc Ist Ital Idrobiol 63, 99-104.
References 108
Leustek, T., Martin, M.N., Bick, J.-A. and Davies, J.P. (2000) Pathways and regulation of sulfur metabolism revealed through molecular and genetic studies. Annual review of plant biology 51, 141-165.
Lin, C.-H., Shen, G.-J., Garcia-Junceda, E. and Wong, C.-H. (1995) Enzymic Synthesis and Regeneration of 3'-Phosphoadenosine 5'-Phosphosulfate (PAPS) for Regioselective Sulfation of Oligosaccharides. Journal of the American Chemical Society 117, 8031-8032.
Lüthy, L., Fritz, M. and Bachofen, R. (2000) In situ determination of sulfide turnover rates in a meromictic alpine lake. Applied and Environm Microbiol 66, 712-717.
Marouga, R., David, S. and Hawkins, E. (2005) The development of the DIGE system: 2D fluorescence difference gel analysis technology. Analytical and bioanalytical chemistry 382, 669-678.
Martinez, J.P., Garay, E., Alcaide, E. and Hernandez, E. (1983) The genus Thiobacillus: Physiology and industrial applications. Acta Biotechnologica 3, 99-124.
Mas, J. and Van Gemerden, H. (2004) Storage products in purple and green sulfur bacteria. In Anoxygenic photosynthetic bacteria, Springer Netherlands.
Matias, P.M., Pereira, I.A.C., Soares, C.M. and Carrondo, M.A. (2005) Sulphate respiration from hydrogen in Desulfovibrio bacteria: a structural biology overview. Progress in biophysics and molecular biology 89, 292-329.
Musat, N., Halm, H., Winterholler, B., Hoppe, P., Peduzzi, S., Hillion, F., Horreard, F., Amann, R., Jørgensen, B. and Kuypers, M. (2008) A single-cell view on the ecophysiology of anaerobic phototrophic bacteria. Proceedings of the National Academy of Sciences 105, 17861- 17866.
Negishi, M., Pedersen, L.G., Petrotchenko, E., Shevtsov, S., Gorokhov, A., Kakuta, Y. and Pedersen, L.C. (2001) Structure and function of sulfotransferases. Arch Biochemistry Biophys 390, 149-157.
O'Farrell, P.H. (1975) High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250, 4007-4021.
Overmann, J. (1997) Mahoney Lake: a case study of the ecological significance of phototrophic sulfur bacteria. Adv Microb Ecol 15, 251-284. 109 Chapter 6
Overmann, J. (2008) Sulfur Metabolism in phototrophic Organisms. In: Ecology of phototrophic sulfur bacteria, Springer Urbana.
Overmann, J., Beatty, J.T., Hall, K.J., Pfennig, N. and Northcote, T.G. (1991) Characterization of a dense, purple sulfur bacterial layer in a meromictic salt lake. Limnol Oceanogr, 36, 846-859.
Overmann, J. and Garcia-Pichel, F. (2006) The phototrophic way of life. In: Prokaryotes, Springer New York.
Overmann, J. and Gemerden, H. (2000) Microbial interactions involving sulfur bacteria: implications for the ecology and evolution of bacterial communities. FEMS Microbiol Rev 24, 591-599.
Pappin, D.J., Hojrup, P. and Bleasby, A.J. (1993) Rapid identification of proteins by peptide- mass fingerprinting. Current biol 3, 327-332.
Parkin, T.B. and Brock, T.D. (1980) The effects of light quality on the growth of phototrophic bacteria in lakes. Arch Microbiol 125, 19-27.
Parkin, T.B. and Brock, T.D. (1981) The role of phototrophic bacteria in the sulfur cycle of a meromictic lake. Limnol Oceanogr 26, 880-890.
Paschinger, H., Paschinger, J. and Gaffron, H. (1974) Photochemical disproportionation of sulfur into sulfide and sulfate by Chlorobium limicola formathiosulfatophilum. Arch Microbiol 96, 341-351.
Pattaragulwanit, K., Brune, D., Trüper, H. and Dahl, C. (1998) Molecular genetic evidence for extracytoplasmic localization of sulfur globules in Chromatium vinosum. Arch Microbiol 169, 434-444.
Pedros-Alio, C. and Guerrero, R. (1991) Abundance and activity of bacterioplankton in warm lakes. IVTLAP 24, 1212-1219.
Peduzzi, R., Bachofen, R. and Tonolla, M. (1998) Lake Cadagno: a meromictic alpine lake. In: Documenta, Istituto Italiano di Idrobiologia.
Peduzzi, R., Demarta, A. and Tonolla, M. (1993) Dynamics of the autochthonous and contaminant bacterial colonization of lakes (Lake of Cadagno and Lake of Lugano as model systems). Studies in Environmental Science 55, 323-335.
Peduzzi, R., Tonolla, M., Demarta, A., DelDon, C., Hanselmann, K. and Bachofen, R. (1991) Role de Chromatium okenii dans le métabolisme d'un lac méromictique (L. Cadagno). In:
References 110
Hommage à F A Forel (J P Vernet, ed), Actes 3ème Conférence internat Limnologues d’expression francaise, Morges.
†Peduzzi, S., †Storelli, N., Welsh, A., Peduzzi, R., Hahn, D., Perret, X. and Tonolla, M. (2012) Candidatus “Thiodictyon syntrophicum”, sp. nov., a new purple sulfur bacterium isolated from the chemocline of Lake Cadagno forming aggregates and specific associations with Desulfocapsa sp. Syst Appl Microbiol 35, 139-144. †: equally contributed.
Peduzzi, S., Tonolla, M. and Hahn, D. (2003a) Isolation and characterization of aggregate- forming sulfate-reducing and purple sulfur bacteria from the chemocline of meromictic Lake Cadagno, Switzerland. FEMS Microbiol Ecol 45, 29-37.
Peduzzi, S., Tonolla, M. and Hahn, D. (2003b) Vertical distribution of sulfate-reducing bacteria in the chemocline of Lake Cadagno, Switzerland, over an annual cycle. Aquatic Microb Ecol 30, 295-302.
Peduzzi, S., Welsh, A., Demarta, A., Decristophoris, P., Peduzzi, R., Hahn, D. and Tonolla, M. (2011) Thiocystis chemoclinalis sp. nov. and Thiocystis cadagnonensis sp. nov., two new motile purple sulfur bacteria isolated from the chemocline of meromictic Lake Cadagno, Switzerland. Int J Syst Evol Microbiol 61, 1682-1687.
Perkins, D.N., Pappin, D.J., Creasy, D.M. and Cottrell, J.S. (1999) Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 3551-3567.
Pfennig, N. (1975) The phototrophic bacteria and their role in the sulfur cycle. Plant and Soil 43, 1-16.
Pfennig, N. (1978) Rhodocyclus purpureus gen. nov. and sp. nov., a ring-shaped, vitamin B12- requiring member of the family Rhodospirillaceae. Int J Syst Bacteriol 28, 283-288.
Pibernat, I.V. and Abella, C.A. (1996) Sulfide pulsing as the controlling factor of spinae production in Chlorobium limicola strain UdG 6038. Arch Microbiol 165, 272-278.
Pourriot, R. and Meybeck, M. (1995) Limnologie générale, Masson Paris.
Prakash, A., Sheldon, R.W. and Sutcliffe Jr, W.H. (1991) Geographic variation of oceanic 14C dark uptake. Limnol Oceanogr 36, 30-39. 111 Chapter 6
Prange, A., Chauvistre, R., Modrow, H., Hormes a, J., Truper, H. and Dahl, C. (2002) Quantitative speciation of sulfur in bacterial sulfur globules: X-ray absorption spectroscopy reveals at least three different species of sulfur. Microbiol 148, 267-276.
Prange, A., Engelhardt, H., Trüper, H.G. and Dahl, C. (2004) The role of the sulfur globule proteins of Allochromatium vinosum: mutagenesis of the sulfur globule protein genes and expression studies by real-time RT-PCR. Arch Microbiol 182, 165-174.
Prange, A. and Modrow, H. (2002) X-ray absorption spectroscopy and its application in biological, agricultural and environmental research. Rev Environm Sci Biotechnol 1, 259-276.
Putschew, A., Scholz-Böttcher, B.M. and Rullkötter, J. (1995) Organic geochemistry of sulfur- rich surface sediments of meromictic Lake Cadagno, Swiss Alps. In: Geochemical Transformations of Sedimentary Sulfur, ACS Publications Washington.
Ravasi, D.F., Peduzzi, S., Guidi, V., Peduzzi, R., Wirth, S.B., Gilli, A. and Tonolla, M. (2012) Development of a real-time PCR method for the detection of fossil 16S rDNA fragments of phototrophic sulfur bacteria in the sediments of Lake Cadagno. Geobiol 10, 196-204.
Reiffenstein, R.J., Hulbert, W.C. and Roth, S.H. (1992) Toxicology of hydrogen sulfide. Annual Rev pharmacology Toxicol 32, 109-134.
Rodrigo, M.A., Miracle, M.R. and Vicente, E. (2001) The meromictic Lake La Cruz (Central Spain). Patterns of stratification. Aquat Sci Res Acr Bound 63, 406-416.
Rojas, J., Giersig, M. and Tributsch, H. (1995) Sulfur colloids as temporary energy reservoirs for Thiobacillus ferrooxidans during pyrite oxidation. Arch Microbiol 163, 352-356.
Romero, L., Camacho, A., Vicente, E. and Miracle, M.R. (2006) Sedimentation patterns of photosynthetic bacteria based on pigment markers in meromictic Lake La Cruz (Spain): paleolimnological implications. J Paleolimnol 35, 167-177.
Rothschild, L.J. (2008) The evolution of photosynthesis... again? Philosophical Transactions of the Royal Society B: Biological Sciences 363, 2787-2801.
Saito, K. (2004) Sulfur assimilatory metabolism. The long and smelling road. Plant Physiol 136, 2443-2450.
Schanz, F., Fischer-Romero, C. and Bachofen, R. (1998) Photosynthetic production and photoadaptation of phototrophic sulfur bacteria in Lake Cadagno (Switzerland). Limnol Oceanogr 43, 1262-1269.
References 112
Schanz, F. and Stalder, S. (1998) Phytoplankton summer dynamics and sedimentation in the thermally stratified Lake Cadagno. In: Documenta, Istituto Italiano di Idrobiologia Pallanza.
Scheele, G.A. (1975) Two-dimensional gel analysis of soluble proteins. Charaterization of guinea pig exocrine pancreatic proteins. J Biol Chem 250, 5375-5385.
Schinck, B. (1999) Habitats of prokaryotes,: Thieme Verlag Stuttgart.
Schramm, A., De Beer, D., Gieseke, A. and Amann, R. (2000) Microenvironments and distribution of nitrifying bacteria in a membrane-bound biofilm. Environm Microbiol 2, 680-686.
Shively, J., Van Keulen, G. and Meijer, W. (1998a) Something from almost nothing: carbon dioxide fixation in chemoautotrophs. Ann Rev Microbiol 52, 191-230.
Shively, J.M., Bradburne, C.E., Aldrich, H.C., Bobik, T.A., Mehlman, J.L., Jin, S. and Baker, S.H. (1998b) Sequence homologs of the carboxysomal polypeptide CsoS1 of the thiobacilli are present in cyanobacteria and enteric bacteria that form carboxysomes-polyhedral bodies. Canadian J Bot 76, 906-916.
Sirevag, R., Buchanan, B.B., Berry, J.A. and Troughton, J.H. (1977) Mechanisms of CO2 fixation in bacterial photosynthesis studied by the carbon isotope fractionation technique. Arch Microbiol 112, 35-38.
Sorokin, J.I. and Donato, N. (1975) On the carbon and sulphur metabolism in the meromictic Lake Faro (Sicily). Hydrobiol 47, 241-252.
Sorokin, Y.I., Sorokin, P.Y., Avdeev, V.A., Sorokin, D.Y. and Ilchenko, S.V. (1995) Biomass, production and activity of bacteria in the Black Sea, with special reference to chemosynthesis and the sulfur cycle. Hydrobiol 308, 61-76.
Sorokin, Y.U.I. (1970) Interrelations between sulphur and carbon turnover in meromictic lakes. Archiv für hydrobiologie 66, 391-446.
Steudel, R. (1996) Mechanism for the formation of elemental sulfur from aqueous sulfide in chemical and microbiological desulfurization processes. Industrial & engineering chemistry research 35, 1417-1423.
Storelli, N., Peduzzi, S., Saad, M.M., Frigaard, N.U., Perret, X. and Tonolla, M. (2013) CO2 assimilation in the chemocline of Lake Cadagno is dominated by a few types of phototrophic purple sulfur bacteria. FEMS Microbiol Ecol 84, 421-432. 113 Chapter 6
Sudesh, K., Abe, H. and Doi, Y. (2000) Synthesis, structure and properties of polyhydroxyalkanoates: biological polyesters. Progress in Polymer Science 25, 1503-1555.
Takahashi, M. and Ichimura, S.-ei (1968) Vertical distribution and organic matter production of photosynthetic sulfur bacteria in Japanese lakes. Limnol Oceanogr, 12, 644-655.
Tang, K., Baskaran, V. and Nemati, M. (2009) Bacteria of the sulphur cycle: an overview of microbiology, biokinetics and their role in petroleum and mining industries. Biochem Eng J 44, 73-94.
Tonge, R., Shaw, J., Middleton, B., Rowlinson, R., Rayner, S., Young, J., Pognan, F., Hawkins, E., Currie, I. and Davison, M. (2001) Validation and development of fluorescence two- dimensional differential gel electrophoresis proteomics technology. Proteomics 1, 377-396.
Tonolla, M., Demarta, A., Hahn, D. and Peduzzi, R. (1998a) Microscopic and molecular in situ characterization of bacterial populations in the meromictic Lake Cadagno. In: Documenta Istituto Italiano Idrobiologia Pallanza.
Tonolla, M., Demarta, A. and Peduzzi, R. (1998b) La chimica del Lago di Cadagno The chemistry of Lake Cadagno. In: Documenta, Istituto Italiano di Idrobiologia Pallanza.
Tonolla, M., Demarta, A., Peduzzi, R. and Hahn, D. (1999) In situ analysis of phototrophic sulfur bacteria in the chemocline of meromictic Lake Cadagno (Switzerland). Appl Environm Microbiol 65, 1325-1330.
Tonolla, M., Demarta, A., Peduzzi, S., Hahn, D. and Peduzzi, R. (2000) In situ analysis of sulfate-reducing bacteria related to Desulfocapsa thiozymogenes in the chemocline of meromictic Lake Cadagno (Switzerland). Appl Environm Microbiol 66, 820-824.
Tonolla, M., Bottinelli, M., Demarta, A., Peduzzi, R. and Hahn, D. (2005a) Molecular identification of an uncultured bacterium (“morphotype R”) in meromictic Lake Cadagno, Switzerland. FEMS Microbiol Ecol 53, 235-244.
Tonolla, M., Peduzzi, R. and Hahn, D. (2005b) Long-term population dynamics of phototrophic sulfur bacteria in the chemocline of Lake Cadagno, Switzerland. Appl Environm Microbiol 71, 3544-3550.
Tonolla, M. and Peduzzi, R. (2006) Lake Cadagno, a model for microbial ecology. In: Documenta Istituto Italiano Idrobiologia Pallanza.
References 114
Tonolla, M., Peduzzi, S., Demarta, A., Peduzzi, R. and Hahn, D. (2004) Phototropic sulfur and sulfate-reducing bacteria in the chemocline of meromictic Lake Cadagno, Switzerland. J Limnol 63, 161-170.
Tonolla, M., Peduzzi, S., Hahn, D. and Peduzzi, R. (2003) Spatio-temporal distribution of phototrophic sulfur bacteria in the chemocline of meromictic Lake Cadagno (Switzerland). FEMS Microbiol Ecol 43, 89-98.
Tonolli, V. (1969) Ecologia e biologia delle acque dolci. In: Introduzione allo studio della Limnologia CNR Istituto Italiano di Idrobiologia Vervania, Pallanza.
Trüper, H.G. (1978) Sulfur metabolism. In The photosynthetic bacteria Plenum Press New York.
Trüper, H.G. (1981) Photolithotrophic sulfur oxidation. In Biology of Inorganic Nitrogen and Sulfur Springer Berlin Heidelberg.
Trüper, H.G. (1984) Phototrophic bacteria and their sulfur metabolism. Stud Inorganic Chem 5, 367-382.
van Gemerden, H. (1967) On the bacterial sulfur cycle of inland waters: JH Pasmans.(Van de Vennestraat 76).
Van Gemerden, H. (1968) On the ATP generation by Chromatium in darkness. Archiv für Mikrobiologie 64, 118-124.
van Gemerden, H. (1986) Production of elemental sulfur by green and purple sulfur bacteria. Arch Microbiol 146, 52-56.
van Gemerden, H. and Beeftink, H. (1981) Coexistence of Chlorobium and Chromatium in a sulfide-limited continuous culture. Arch Microbiol 129, 32-34.
Van Gemerden, H. and Mas, J. (1995) Ecology of phototrophic sulfur bacteria. In Anoxygenic photosynthetic bacteria Springer Netherlands
Van Gemerden, H. and Mas, J. (2004) Ecology of phototrophic sulfur bacteria. In Anoxygenic photosynthetic bacteria Springer Boston.
Van Gemerden, H., Montesinos, E., Mas, J. and Guerrero, R. (1985) Diel cycle of metabolism of phototrophic purple sulfur bacteria in Lake Ciso (Spain). Limnol Oceanogr, 932-943. 115 Chapter 6
Veldhuis, M.J.W. and van Gemerden, H. (1986) Competition between purple and brown phototrophic bacteria in stratified lakes: sulfide, acetate, and light as limiting factors. FEMS Microbiol Lett 38, 31-38.
Vila, X. and Abella, C.A. (2001) Light-harvesting adaptations of planktonic phototrophic micro-organisms to different light quality conditions. Hydrobiol 452, 15-30.
Wang, X., Falcone, D.L. and Tabita, F.R. (1993) Reductive pentose phosphate-independent
CO2 fixation in Rhodobacter sphaeroides and evidence that ribulose bisphosphate carboxylase/oxygenase activity serves to maintain the redox balance of the cell. J Bacteriol 175, 3372-3379.
Wetzel, R.G. (1983) Limnology: Saunders College Publishing, Philadelphia,.
Wetzel, R.G. (2001) Limnology: Lake and River Ecosystems, : 3rd edition, Academic, San Diego, California.
Widdel, F., Hansen, T.A., Balows, A., Truper, H.G., Dworkin, M., Harder, W. and Schleifer, K.H. (1992) The dissimilatory sulfate-and sulfur-reducing bacteria. In The prokaryotes: a handbook on the biology of bacteria: ecophysiology, isolation, identification, applications, Springer-Verlag Inc New York
Wirth, S.B., Gilli, A., Dahl, T.W., Niemann, H., Ravasi, D., Lehmann, M.F., Peduzzi, R., Peduzzi, S., Tonolla, M. and Anselmetti, F.S. (2012) 9800 years of meromixis and euxinia in Lake Cadagno (Swiss Alps): Sedimentary and biogeochemical evidence for the formation of a permanently stable chemocline. In EGU General Assembly Conference Abstracts.
Wirth, S.B., Gilli, A., Niemann, H., Dahl, T.W., Ravasi, D., Sax, N., Hamann, Y., Peduzzi, R., Peduzzi, S. and Tonolla, M. (2013) Combining sedimentological, trace metal (Mn, Mo) and molecular evidence for reconstructing past water-column redox conditions: The example of meromictic Lake Cadagno (Swiss Alps). Geochimica et Cosmochimica Acta 120, 220-238.
Wood, H.G. and Stjernholm, R.L. (1962) Assimilation of carbon dioxide by heterotrophic organisms. The bacteria 3, 41-117.
Wüest, A. (1994) Interactions in lakes: Biology as source of dominant physical forces. Limnologica Jena 24, 93-104.
Xiong, J., Fischer, W.M., Inoue, K., Nakahara, M. and Bauer, C.E. (2000) Molecular evidence for the early evolution of photosynthesis. Science 289, 1724-1730.
References 116
Yates, J.R., Speicher, S., Griffin, P.R. and Hunkapiller, T. (1993) Peptide mass maps: a highly informative approach to protein identification. Analytical biochemistry 214, 397-408.
Yeates, T.O., Kerfeld, C.A., Heinhorst, S., Cannon, G.C. and Shively, J.M. (2008) Protein- based organelles in bacteria: carboxysomes and related microcompartments. Nature Rev Microbiol 6, 681-691.
Yoshizawa, Y., Toyoda, K., Arai, H., Ishii, M. and Igarashi, Y. (2004) CO2-responsive expression and gene organization of three ribulose-1,5-bisphosphate carboxylase/oxygenase enzymes and carboxysomes in Hydrogenovibrio marinus strain MH-110. J Bacteriol 186, 5685- 5691.
117 Chapter 7
Unpublished data 118
7. ANNEXES
7.1. Unpublished data
7.1.1. Detection of putative RuBisCO proteins by western blot
The presence of the large subunit of RuBisCO was checked in pure anoxic autotrophic and mixotrophic cultures of Candidatus “T. syntrophicum” strain Cad16T. These cultures were grown with a photoperiod of 12 h of light (7:00 – 19:00) fallowed by 12 h of dark (19:00 – 7:00) until they reached the starting concentration of approximately 107 cell ml-1 (see Material and
Methods of Chapter 2). The total proteins were then extracted every 4 h during 24 h (T1-6), with an extra sample at 9:00 in the period of maximal CO2 fixation activity (see Results of Chapter 2) and checked with polyclonal antibodies against the large subunit of RuBisCO (CbbL-CbbM). Unfortunately, these antibodies are not specific for CbbL or CbbM, but both large subunits of strain Cad16T are recognized (Figure A1).
During both autotrophic (Figure A1, left panel) and mixotrophic (Figure A1, right panel) conditions of growth, the concentration of CbbL-CbbM was very low at the sampling time T1 after 12 h of dark, with values of 0.254 and 0.161 CbbL-CbbM per µl-1, respectively. The maximal quantity of RuBisCO was recorded at T3 after 8 h of light for both autotrophic and mixotrophic growing conditions, with 1.265 CbbL-CbbM per µl-1 and 1.047 CbbL-CbbM per µl- 1, respectively. During the light period (T1.2 to T4), the quantity of CbbL-CbbM in the autotrophic cultures was up to 5-fold higher than in mixotrophic cultures. However, during the dark period (T5, T6 and T1) no relevant differences in the concentration of CbbL-CbbM were measured between the 2 growing conditions. In the autotrophic growing condition, the 14 concentration of the protein appears to correlate with the CO2 assimilation results of the Chapter 3, indicating higher activity during the light phase compared with the dark phase. Moreover, the maximal concentration of the large subunit of the RuBisCO enzyme occurs in T3 exactly during one of the maximal peaks of expression of the cbbL gene (see Results of Chapter 2).
119 Chapter 7
Figure A1. Detection of large subunit of the RuBisCO
Total proteins extracted after autotrophic (left) and mixotrophic growth (right). The purified large subunit of RuBisCO was used as positive control (RbcL = 0.6 ug). The protein samples were taken every 4 h over 24 h (12 h of light followed by 12 h of dark): T1 = 7:00, T1.2 = 9:00, T2 = 11:00, T3 = 15:00, T4 = 19:00, T5 = 23:00 and T6 = 3:00.
With the aim of identifying the large subunit of the form I and the form II of RuBisCO, we decided to use the antibody described above in a 2-dimensional gel (2D-gel). Approximately 10 ng of total proteins extracted from autotrophic cultures at T2 (after 4 h of light 11:00) were loaded on a specific IPG strip (3-10 NL of 7 cm) and divided by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). Figure A2 (right) shows the presence of different protein spots recognized by antibody. Three independent spots were visualized at the level of the hypothetical RuBisCO large subunit, which has a size of approximately 52 KDa in a pI range between 6.0 and 7.5, and these spots were also visible in the silver-stained gel. These protein spots were identified by MALDI-TOF mass spectrometry as the large subunit CbbM of the form II RuBisCO (see Material and Methods of Chapter 4).
Moreover, a line of several spots of approximately 45 KDA were visualized along the same pI range of the 3 independent spots mentioned above. These protein spots were not visible in the silver-stained gel, most likely because of their low quantity. Unfortunately, the low concentration of these proteins also prevented their identification by MALDI-TOF mass spectrometry.
The presence of the enzyme RuBisCO activase (RA) was as also determined (Figure A2 left) using another large-spectrum polyclonal antibody. The presence of different protein spots was demonstrated, and three of these were in the same place as the three spots identified as form II
Unpublished data 120
RuBisCO (CbbM). Unfortunately, we did not have the annotation of the RA gene in our draft genome, so we could not estimate a hypothetical size or pI. The lack of specificity of this second antibody most likely resulted in recognition of protein spots only than the target RA.
Figure A2. Western blot against the large subunit from the RuBisCO in a 2D-GEL of proteins extracted during the light exposure
(center) 2D-PAGE with 10 µg of proteins extracted at the beginning of the light phase (T2 = 11:00) from pure autotrophic cultures of strain Cad16T. The samples were run on a linear isoelectric focusing (IEF) strip (pH 3–10 non-linear) at approximately 85 000 Vh and then separated by 12.5% SDS-PAGE of 7 cm width.
(right) Blot of a same twin gel (12.5% 7 cm 3-10 NL) shown in the center. The presence of the large subunit of RuBisCO was checked with an antibody tagging a key functional region conserved across all known plant, algal and
(cyano)bacterial RbcL protein sequences (form I L8S8 and form II L2) with 100% homology in the amino acid sequence with both Cad16T proteins.
(left) Blot of another 7 cm 2D-PAGE (12.5% 7 cm 3-10 NL) gel run with the same conditions as the gel in the center. The presence of RuBisCO activase was checked with an antibody tagging a key functional region conserved across all known plant, algal and (cyano)bacterial RA protein sequences with 100% homology in the amino acid sequence with this present in strain Cad16T.
121 Chapter 7
Material and Methods
Cell lysis, protein sample preparation, quantification and gel separation. Liquid bacterial cultures (100 ml) of a concentration up to 107 cell ml-1 were washed 3 times with PBS 1x and resuspended in 1 ml of lysis buffer (Tris 10 mM, 60 mg; EDTA 1 mM, 18.6 mg; NADP 0.5 mM, 19.7 mg, deionized water 50 mL, pH adjusted to 6.8 with HCl). The cells were lysed by sonication (SONOPLUS HD 2070, Bandelin electronics, Berlin, Germany) with 5 cycles of 15 sec at approximately 20% of the maximal power with a pause of 2-3 min in ice between each cycle. After cell lysis, the sample was centrifuged for 15 min at 14000 rpm at 4°C, and the protein concentration of the supernatant as measured by Bradford assay (Bio-Rad, cat. no. 500- 0201), according to the manufacturer’s instructions. A quantity of 1 µg of total proteins extracted for each sample was loaded on a SDS-PAGE according to Laemmli (Laemmli 1970).
The total proteins isolated for the 2D-PAGE analysis needed some extra steps and another lysis buffer (see Material and Methods of Chapter 5).
Immunoblot analysis. After SDS-PAGE or 2D-PAGE, the protein samples were transferred to a nitrocellulose membrane as described by Towbin (Towbin et al. 1979). After the transfer, the membrane was blocked for 1 hr at ambient temperature using 10% of DifcoTM Skim Milk (Becton, Dickinson and company, cat. no. 232100) in TBS containing 0.1% Tween 20 (TBS-T). Subsequently, the membrane was incubated for 1 hr at room temperature with TBS-T containing a 1:5000 dilution of the primary rabbit polyclonal antibody anti-Rubisco large subunit (RbcL) or anti- RA/Rubisco activase (Agrisera, cat. no. AS03 037 and AS10 700, respectively). After washing with TBS-T (3 times for 15 min), the membrane was incubated for 30 min at room temperature with a HRP conjugated anti-Rabbit IgG (whole molecule) secondary antibody (Abcam®, cat. no. ab6721) diluted 1:5000 in TBS-T. The antibody detection was conducted by enhanced chemiluminescence substrate detection as recommended by the manufacturer (Western LightningTM Plus-ECL; PerkinElmer, cat. no. NEL103001EA).
Unpublished data 122
References
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.
Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences 76, 4350.
123 Chapter 7
7.1.2. Expression of cbbL and cbbM directly in Lake Cadagno (in situ)
The expression of both RuBisCO genes (cbbL and cbbM) of Candidatus “T. syntrophicum” strain Cad16T were also fallowed by qRT-PCR (see Material and Methods of Chapter 2) in dialysis bags pure cultures incubated in the chemocline of the Lake Cadagno (see Material and Methods of Chapter 2). The aim of this analysis was to assess whether transcription profiles of Cad16T cells grown in laboratory conditions (i.e., in vitro) were similar to those of bacteria incubated in the Lake Cadagno chemocline (i.e., in situ). For mRNA expression analysis, 100 ml of pure cultures of PSB Candidatus T. syntrophicum strain Cad16T were incubated for five weeks (August 20 to September 24, 2009) in eight 30 cm long dialysis bags (same characteristics as above) at a depth of 12 m. Cells were then sampled every 4 h for a 24 h period, starting at 3:00 PM on September 23. In this period of the year, the sun shone over the lake from 7:48 to 18:34 affording a period of light-exposure 74 min shorter than the 12 h photoperiod selected for the in vitro experiments. Interestingly, the transcription profiles of cbbM in Cad16T cells grown in laboratory conditions or incubated in situ differed considerably, with a single peak of cbbL synthesis measured in T4 = 3.39x10-6 pg of mRNA per cell during the night at 23:00.
On the other hand, the expression of cbbM displayed similar levels of expression compared with those recorded in laboratory conditions (between 10-7 and 10-8 pg of mRNA per cell), without any major peak of expression.
Unpublished data 124
Figure A3. Expression of cbbL and cbbM in situ
Expression of the cbbL gene (full square) and cbbM (empty square) incubated in dialysis bags in the chemocline of the Lake Cadagno of pure Cad16 cultures were monitored every 4 hours for 24 h by RT-qPCR. The sampling points were as follows: T0 = 7:00, T1 = 11:00, T2 = 15:00, T3 = 19:00, T4 = 23:00 and T5 = 3:00. The quantity of mRNA was normalized to the cell number to obtain a direct measurement of the relative cell contribution. All the experiments were performed in triplicate. 125 Chapter 7
FEMS research fellowship 126
7.2. DENMARK: FEMS Research Fellowship rapport
7.2.1. Transcriptomic analyses on the purple sulfur bacterium Thiodictyon sp. Cad16
Nicola Storelli1, Niels-Ulrik Frigaard3, Donald A. Bryant4 and Mauro Tonolla1,2
1 Cantonal Institute of Microbiology, Microbial Ecology Group, Microbiology Unit, Plant biology Dept. UNIGE, Via Mirasole 22a, CH-6500 Bellinzona
2 Centro di Biologia Alpina (CBA), Piora, CH-6777 Quinto
3 Copenhagen Biocenter, Department of Biology University of Copenhagen, Ole Maaløes Vej 5, 2200 Copenhagen N (Denmark)
4 Department of Biochemistry an Molecular Biology, The Pennsylvania State University, University Park, PA 16802 USA
ABSTRACT
Purple sulfur bacteria (PSB) play an essential role in the turnover of carbon and sulfur in many aquatic ecosystems. Their physiological diversity as photoautotrophs, photoheterotrophs, and sometimes as chemotrophs has been recognized for decades, but the significance of this diversity in natural environments is not known. The genome of Thiodictyon sp. Cad16 has been sequenced using pyrosequencing. Preliminary analyses indicate that the genome is unexpectedly large, at least 8 Mbp, and contains approximately 8000 genes with encoding functions related to anoxic, oxic, phototrophic, and chemotrophic lifestyles. This study aimed to investigate the physiology of the PSB Thiodictyon sp. Cad16, which is abundant in our model ecosystem (Lake Cadagno, Switzerland) and grows readily in pure culture. We produced transcriptomic sequence information to correlate changes and adaptations in physiology with changes in environmental conditions. To this end, cultures of Thiodictyon sp. Cad16 were grown under different, defined laboratory conditions (anoxic photoautotrophy, anoxic photoheterotrophy, anoxic chemotrophy, oxic chemotrophy, etc.), mRNA was isolated and cDNA prepared. The cDNA was then sequenced using Illumina’s Solexa approach and compared to the whole genome sequence. 127 Chapter 7
INTRODUCTION
Lake Cadagno is a crenogenic, meromictic lake located in the Piora valley in the southern Alps of Switzerland (46°33' N, 8°43'E). It is characterized by a compact chemocline with high concentrations of sulfate; steep gradients of oxygen, sulfide, and light; and a turbidity maximum that correlates to large numbers of bacteria (up to 107 cells ml-1) (Tonolla, Demarta et al. 1999; Luthy, Fritz et al. 2000; Camacho, Erez et al. 2001). The chemocline of the meromictic Lake Cadagno contains a dense community of phototrophic purple sulfur bacteria (PSB) belonging to the genera Lamprocystis, Thiocystis, Thiodictyon and Chromatium, as well as the green sulfur bacteria (GSB) of the genus Chlorobium (Tonolla, Demarta et al. 1998; Tonolla, Demarta et al. 1999).
Currently, no genome sequence is available for any PSB from freshwater environments. Therefore, we decided to sequence the genome of the Thiodictyon sp. strain Cad16 isolated from Lake Cadagno and usually growing in pure cultures (Peduzzi and Tonolla, data not published). This PSB is currently being sequenced at the laboratory of D.A. Bryant, Pennsylvania, USA, using a pyrosequencing technology approach (Margulies, Egholm et al. 2005). Preliminary analyses indicate that the genome is unexpectedly large, at least 8 Mbp, and contains approximately 8000 genes encoding functions related to anoxic, oxic, phototrophic, and chemotrophic lifestyles.
PSB, similar to GSB, play an essential role in the turnover of carbon and sulfur in many aquatic ecosystems (Gemerden 1986; Camacho and Vicente 1998; Camacho, EREZ et al. 2001). Their physiological diversity as photoautotrophs, photoheterotrophs, and sometimes as chemotrophs has been recognized for decades (Brune 1995), but the significance of this functional diversity in natural environments is not known.
We expect the outcome of this project to be of importance in the understanding of the physiological capabilities and adaptation strategies of PSB in natural environments and should provide the basis for the annotation of the genome sequence.
FEMS research fellowship 128
MATERIALS AND METHODS
Bacterial strain and growth condition
Thiodictyon sp. Cad16 was grown in liquid medium (Pfenning's Medium I, by DSMZ; -1 - hereafter also referred to as “PSB medium”) containing 0.5 g KH2PO4 liter , 0.34 g NH4Cl liter 1 -1 -1 -1 - , 0.5 g MgSO4 · 7H2O liter , 0.25 g CaCl2 · 2H2O liter , 0.34 g KCl liter , 1.5 g NaHCO3 liter 1 , 0.5 ml of trace element solution SL10, and 0.02 mg of vitamin B12 (Eichler and Pfennig 1988). -1 The medium was reduced with 0.3 g liter Na2S · 9H2O (1.10 mM final concentration) and adjusted to approx. pH 7.2. Media were prepared in a 2-liter bottle with an N2/CO2 (80%/20%) gas phase according to Widdel and Bak (Widdel and Bak 1992). All cultures were incubated at room temperature (20 to 25°C) with a 12 h light/12 h dark photoperiod at light intensities of 4 to 8 µE m-2 s-1, generated with an incandescent 60-W bulb (Eichler and Pfennig 1988). The growth of the cultures was monitored at a wavelength of 650 nm in a UV/VIS Spectrometer Lambda 2S (Perkin-Elmer) with a 1-cm cuvette.
DNA and RNA analysis
Total mRNA extraction. Four tubes containing 2 ml of Cad16 liquid culture (OD650 0.5< x <1.0) were harvested after 10 min centrifugation at 7,000 g. Then, 0.25 ml of TRIZOL was added to the pellets in the tubes. The material was gently shaken for few minutes to resuspend and lyse the cells, and finally, all samples were collected in one 1.5 ml Eppendorf tube. We extracted the RNA and resuspended it in a 1 ml Eppendorf tube containing approximately 1 ml of TRIZOL-cell TRIZOL (Invitrogen™l). In addition, 0.2 ml of chloroform was added and after a rapid resuspension and incubation for 2-3 min at room temperature, the suspension was centrifuged for 15 min at 12,000 g at 4°C. The aqueous phase was then transferred into a 1.5 ml RNase-free tube, purified on ice with isopropanol and ethanol 70%, and centrifuged twice for 10 min at 12,000 g and 4°C. The RNA was then air-dried for 5-10 min and resuspended in 50 µl of RNA-free water. The RNA was stored at -80°C or used for the next step of purification.
DNase treatment. The RNA extracted was purified using the “TURBO DNA-freeTM kit” (Ambion®) following the manufacturer’s instructions.
RNA concentration. The purified RNA was concentrated using the “RNeasy MinElute Cleanup kit” (Qiagen®) following the manufacturer’s instructions. 129 Chapter 7
During the purification procedures with this kit, we lost a large amount of RNA. Therefore, we decided to skip this step and concentrate the 50 µl recovered after DNase treatment with a SpeedVAC to a volume of 10-15 µl (approximately 20-30 min).
mRNA enrichment. The concentrated RNA was treated with a “MICROBExpressTM kit” (Ambion®), following the manufacturer’s instructions to remove the 16S and 23S ribosomal RNAs (rRNA and tRNA).
Amplification of mRNA. The mRNA was transformed in aRNA (mRNA with a PAP tail in the 3’ extremity) and amplified using a “MessageAmp II-Bacteria kit” (Ambion®), following the manufacturer’s instructions.
Double-stranded cDNA synthesis from aRNA. The synthesis of cDNA was carried out using the “SuperScriptTM Choice System for cDNA Synthesis kit” (InvitrogenTM), following the manufacturer’s instructions.
Ligation of all cDNA fragments. The fragments of cDNA were ligated with T4 DNA ligase (Biolabs, M02025). This step stabilizes the cDNA for the next sequencing step. Then, 1 µl of T4 DNA ligase was added to a final reaction volume of 20 µl, which contained 4 µl of 10x buffer provided with the T4 DNA ligase, and incubated overnight at 16°C. After the reaction, the sample was incubated for an additional 20 min at 65°C to inactivate the enzyme and then stored at -80°C.
cDNA sequencing
The cDNA will be sent to another laboratory to be sequenced using Illumina’s Solexa approach. At present, we are evaluating the best offers.
Protein collection
The pellets harvested by centrifugation for 20 min at 9,000 g in two 50 ml Falcon tubes from approximately 80 ml of liquid culture with an OD650 between 0.5 and 1.0 were resuspended in 4 ml of KH2PO4 0.1 M and aliquoted in six 1.5 ml Eppendorf tubes. The 6 tubes were centrifuged for 10 min at 17,500 g, and the supernatant was discarded and the pellets stored at –80°C. All centrifugation steps were performed at 4°C.
FEMS research fellowship 130
RESULTS
Choice of the different growing conditions
The techniques used for the growth in vitro of the PSB is completely different from those currently employed for common bacteria such as E. coli or Pseudomonas aeruginosa. PSB is anaerobic, therefore we have to utilize special, hermetically sealed containers with anaerobic conditions (1:5 - CO2:N2); the cultivation in plates is most problematic. Liquid cultures of Thiodictyon sp. Cad16 need between 6 and 7 days to reach a reasonable optic density at standard heterotrophic growth condition (OD650 of 0,5-1,0). Despite the growth problems with our bacteria, we were able to grown it at 7 different conditions at which the cultures reached a reasonable OD650, but for some of them, detailed below, a pre-incubation during approximately 5 days at standard heterotrophic growth conditions were necessary to minimize the incubation time:
heterotrophy heterotrophy in continuous light heterotrophy with oxygen (microaerophily) * heterotrophy with oxygen in the dark (chemoheterotrophy) * heterotrophy in the dark * autotrophy autotrophy with oxygen in the dark (chemoautotrophic) *
* For growth in the presence of oxygen under continuous darkness, the culture was incubated first under normal conditions (atmosphere 1:5 CO2:N2 and with light cycles of 12 hours light/dark) and then for 2-3 more days at special conditions.
More accurate explanations of each growth condition are described in the appendix A at the end of the rapport. From these conditions mRNA was isolated and cDNA prepared. This cDNA will be sequenced using Illumina’s Solexa approach.
mRNA isolation and cDNA preparation
The preparation of the ligated cDNA from the liquid cultures was carried out using more specific kits, as previously explained. After each purification and translation step a quantification 131 Chapter 7 of the mRNA or cDNA was performed using a NanoDrop spectrophotometer. The nucleic acid concentration is presented in Table F1.
Table F1. Nucleic acid concentration. Growth RNA mRNA aRNA cDNA cDNA tot (10µl) condition** ng/µl ng/µl ng/µl ng/µl µg *** *** *** *** * 2x more 5-10µl 100 µl 10 µl TRIZOL 1. 12h L/D 160 60 595 220 2.2 2. 12h L/D ØAC 135 40 1350 480 4.8 3. 24h DO2 100 45 700 300 3.0 4. 24h DO2 ØAC 75 6 950 360 3.6 5. 24h L 85 45 40 220 2.2 6. 12h L/D O2 125 50 125 216 2.16 7. 8h D 85 95 475 690 6.9 8. 8h D ØAC 30 30 95 225 2.25 9. 8h L 200 110 550 800 8.0 10. 8h L ØAC 100 30 75 175 1.75 11. 24h D 260 190 1440 675 6.75 *2.0 µg is the minimum amount of cDNA needed for a sequencing analysis; ** growth conditions detailed in the appendix A; *** for each purification/translation step some aliquots were stored at -80°C as reserve.
The different yield between the growth conditions was imputable to essentially 2 reasons, the first was that the starting OD650 were not always the same, the second that I carried out the experiment of the aRNA synthesis in 2 times and with 2 different MessageAmp II-Bacteria kits. The first one yielded less aRNA (5, 8 and 10 in Table 1.) than the new one. The aRNA of samples 1 and 2 were synthesized with both kits (old and new) because they were grown at the 2 most important growth conditions (heterotrophy and autotrophy).
Collection of protein
Parallel to the mRNA manipulation, we stored a pellet derived from the centrifugation of about 80 ml of liquid culture aliquoted in six 1.5 ml Eppendorf tubes at –80°C. This pellet will be used for proteomic analysis to be carried out using 2D-gel electrophoresis (DIGE) and the results will be correlated to genomic and transcriptomic analysis.
FEMS research fellowship 132
DISCUSSION
The host laboratory headed by Dr. Niels-Ulrik Frigaard has extensive experience with bioinformatics and the functional genomics of GSB (Eisen, Nelson et al. 2002) and has recently moved into the field of PSB. This move is the result of a collaboration with Dr. Tonolla. Bioinformatics analyses of the PBS genome sequence data will allow us to learn more about the physiology of these organisms, which are an important part of the ecosystem of Lake Cadagno (Tonolla, Peduzzi et al. 2005).
My work in Denmark was essentially to grow the PSB Thiodyction sp. Cad16 in all possible conditions and to collect the largest variety of mRNA. This total mRNA was later converted into cDNA, and all of the fragments were ligated together to form a large cDNA molecule. This cDNA is more stable for storage and can be sequenced more easily. We now shall send our cDNA to the best laboratory (quality-price) for sequencing The sequencing results will then be compared with sequences present in Internet databases (NCBI, etc.), thus enabling us to identify a large part of genes presents in the genome. After the genomic and transcriptomic analyses, we shall conduct some proteomic experiments with 2D-gel electrophoresis (DIGE) to better understand the life of this PSB.
133 Chapter 7
REFERENCES
Brune, D., Ed. (1995) Sulfur compounds as photosynthetic electron donors. In: Anoxygenic Photosynthetic Bacteria Kluwer Academic Publishers Dordrecht/The Netherlands
Camacho, A., J. EREZ, et al. (2001) Microbial microstratification, inorganic carbon photoassimilation and dark carbon fixation at the chemocline of the meromictic Lake Cadagno(Switzerland) and its relevance to the food web. Aquatic sciences 63, 91-106.
Camacho, A., J. Erez, et al. (2001) Microbial microstratification, inorganic carbon photoassimilation and dark carbon fixation at the chemocline of the meromictic Lake Cadagno (Switzerland) and its relevance to the food web.Aquatic Sciences-Research Across Boundaries 63, 91-106.
Camacho, A. and E. Vicente (1998) Carbon photoassimilation by sharply stratified phototrophic communities at the chemocline of Lake Arcas(Spain). FEMS Microbiol Ecol 25, 11-22.
Eichler, B. and N. Pfennig (1988) A new purple sulfur bacterium from stratified freshwater lakes, Amoebobacter purpureus sp. nov. Arch Microbiol 149, 395-400.
Eisen, J., K. Nelson, et al. (2002) The complete genome sequence of Chlorobium tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacterium. Proceedings of the National Academy of Sciences 99, 9509-9514.
Gemerden, H. (1986) Production of elemental sulfur by green and purple sulfur bacteria. Arch Microbiol 146, 52-56.
Luthy, L., M. Fritz, et al. (2000) In situ determination of sulfide turnover rates in a meromictic alpine lake. Appl Environm Microbiol 66, 712-717.
Margulies, M., M. Egholm, et al. (2005) Genome sequencing in microfabricated high-density picolitre reactors. Nature 437, 376-380.
Tonolla, M., A. Demarta, et al. (1998) Microscopic and molecular in situ characterization of bacterial populations in the meromictic Lake Cadagno. In: Documenta Istituto italiano di. Idrobiologia Pallanza.
Tonolla, M., A. Demarta, et al. (1999) In situ analysis of phototrophic sulfur bacteria in the chemocline of meromictic Lake Cadagno (Switzerland). Appl Environm Microbiol 65, 1325- 1330.
FEMS research fellowship 134
Tonolla, M., R. Peduzzi, et al. (2005) Long-Term population dynamics of phototrophic sulfur bacteria in the chemocline of Lake Cadagno, Switzerland. Appl Environm Microbiol 71, 3544- 3550.
Widdel, F. and F. Bak (1992) Gram-negative mesophilic sulfate-reducing bacteria. The Prokaryotes 4, 3352-3378.
135 Chapter 7
7.2.2. APPENDIX A: Growth conditions
Explanation of the codes used on tubes (growth condition):
12h L/D: heterotrophic growth conditions with 2 mM of acetate added to a basic PSB medium; normal condition of light exposure with a photoperiod of 12 hours of light followed by
12 hours of darkness during 7 days; starting OD650=0.589 (in a 1ml cuvette) at 10:00 am.
12h L/D ØAC: autotrophic growth conditions in basic PSB medium; normal condition of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 8 days; starting OD650=0.483 (in a 1ml cuvette) at 10:00 am.
24h DO2: heterotrophic growth conditions with 2 mM of acetate added to a basic PSB media; first at normal light condition with a photoperiod of 12 hours of light followed by 12 hours of darkness during 6 days to an OD650=0.579 (in a 1ml cuvette), and then changing the atmosphere by adding 5% oxygen and storing the sample in complete darkness during 1 day
(chemoheterotrophic growth conditions); starting OD650=0.704 (in a 1ml cuvette) at 2:00 p.m..
24h DO2 ØAC: : autotrophic growth conditions in basic PSB medium; first at normal condition of light with a photoperiod of 12 hours of light followed by 12 hours of darkness during 6 days to an OD650=0.364 (in a 1ml cuvette), and then changing the atmosphere by adding 5% oxygen and storing the sample in complete darkness during 1 day (chemoautotrophic growth conditions); starting OD650=0.473 (in a 1ml cuvette) at 2:00 p.m..
24h L: heterotrophic growth condition with 2 mM of acetate added to basic PSB medium; special conditions of light with a photoperiod of 24h continuous light during 8 days; starting
OD650=0.317 (in a 1ml cuvette) at 3:30 p.m..
12h L/D O2: heterotrophic growth condition with 2 mM of acetate added to basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by
12 hours of darkness during 7 days (OD650=0.573), and then changing the atmosphere by adding
5% oxygen and incubating for one more day; starting OD650=1.088 (in a 1ml cuvette) at 3:30 p.m..
8h D: : heterotrophic growth conditions with 2 mM of acetate added to basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 7 days; starting OD650=0.584 (in a 1ml cuvette); sampling after 8 hours of darkness (03:00 am).
FEMS research fellowship 136
8h D ØAC: autotrophic growth conditions in basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 7 days; starting OD650=0.467 (in a 1ml cuvette); sampling after 8 hours of darkness (03:00 am).
8h L: heterotrophic growth conditions with 2 mM of acetate added to basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 7 days; starting OD650=0.795 (in a 1ml cuvette); sampling after 8 hours of light phase (15h003:00 p.m.; 12 hours after the 8h D, same culture).
8h L ØAC: autotrophic growth conditions in basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 7 days; starting OD650=0.437 (in a 1ml cuvette); sampling after 8 hours of light phase (3:00 p.m.; 12 hours after the 8h D, same culture).
24h D: heterotrophic growth conditions with 2 mM of acetate added to basic PSB medium; normal conditions of light exposure with a photoperiod of 12 hours of light followed by 12 hours of darkness during 7 days (OD650=1.040), and then storing the sample in complete darkness during 3 day; starting OD650=1.105 (in a 1ml cuvette) at 4:30 p.m..
Figure F1. Schematized growth conditions.