Magnetotactic Bacteria: Isolation, Imaging, and Biomineralization

Home , Magnetofossil, Magnetoreception, Magnetosome, Magnetospirillum, Magnetotactic bacteria, Magnetotaxis

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

Zachery Walter John Oestreicher

Graduate Program in Geological Sciences.

The Ohio State University

2012

Committee:

Steven K. Lower, Advisor

Wendy Panero

Olli Tuovinen

Brian H. Lower

Zachery Walter John Oestreicher

2012

Abstract

Magnetotactic bacteria (MTB) are a specialized group of bacteria that produce very small magnets inside their cells. There are a number of reasons that I decided to study these particular microorganisms. MTB are universally found in aquatic environments and they can be isolated with a simple magnet. These bacteria have the distinct ability to synthesize nanometer-scale crystals of magnetite (Fe3O4) or greigite

(Fe3S4) inside their cells. This type of biomineralization serves as a model for mineral formation in more complex organisms such as birds, bees, and fish. The magnetite from

MTB can be used as a biomarker, called magnetofossils, for past life on earth as well as possible extraterrestrial life forms (e.g., putative magnetofossils in Martian meteorites such as the Allan Hills meteorite). Magnetofossils are novel biomarkers because the magnetite from MTB has a specific crystal shape, narrow size range, and flawless chemical composition, which make them easily identified as biological origin. These same crystallographic attributes could also be exploited in biomimicry. For example, in vitro synthesis of magnetic crystals could have applications in medicine, electronic storage devices, and even environmental remediation. The work in this dissertation touches on all of these concepts.

For the environmental isolation of MTB, I collected water samples from two field sites, an arsenic-rich hot spring in Oregon (Mickey Hot Spring) and a freshwater, microbialite-containing lake in British Columbia, Canada (Pavilion Lake). These sites were selected because MTB have never been isolated from these locations, and these two ii sites are often used as proxies for conditions on the early Earth or extraterrestrial bodies.

To isolate MTB from these two samples, I used a relatively simple method that takes advantage of a bar magnet and capillary racetrack created using a cotton-plugged glass pipette. The MTB from Mickey Hot Spring in Oregon were rod to vibrioid-shaped cells that were 2.2 (± 0.6) µm long and 0.62 (± 0.1) µm wide. The magnetosomes were composed of bullet-shaped crystals of magnetite that were 84 (± 17) nm long and 39 (± 8) nm wide. These magnetosomes from the Mickey Hot Spring specimens were usually arranged in a single chain. The 16S rRNA gene sequence analysis identified the Mickey

Hot Spring specimens as part of the Nitrospirae phylum. MTB isolated from Lake

Pavilion in British Columbia were spirillum-shaped cells that were 2.9 (± 0.6) µm long

0.34 (± 0.02) µm wide (n = 7). Their magnetite crystals were 47 (± 5) nm long and 44 (±

5) nm wide. The magnetosomes were arranged in a single chain. The 16S rRNA analysis showed that the Lake Pavilion cells were from the Alphaproteobacteria phylum.

After isolating MTB from two different environments, I turned my attention to the biomineralization of magnetite within MTB. For this portion of my dissertation, I examined a protein called Mms-6, which has recently been shown to play a key role in the nucleation and/or growth of magnetite. I used high-resolution transmission electron microscopy (TEM) to examine gold-conjugated, immunolabeled Mms-6 in thin sections of Magnetospirillum magneticum AMB-1. I found that the Mms-6 proteins are not located on the cell membrane or within the cytoplasm, but are only clustered on the magnetosome membrane. This was confirmed by using confocal laser scanning microscopy on Mms-6 proteins labeled with green fluorescence proteins in cells of M.

iii magneticum AMB-1. These studies constrain the spatial and temporal function of Mms-6 proteins during the mineralization of magnetite by MTB. Mms6 is confined to the magnetosome membrane after invagination from the cell membrane.

The last portion of my dissertation included the high-resolution analysis of two different types of magnetotactic bacteria: M. magneticum AMB-1 and M. gryphiswaldense MSR-1 using atomic force microscopy (AFM) and TEM. In these experiments I examined the ultrastructure and magnetosomes from both species as well as Mms6 proteins and determined the advantages of both techniques to examining MTB.

The main advantage that AFM has over the TEM is that cells or biomolecules can be examined under physiological conditions. This allows direct observation of proteins interacting with magnetite in vitro. However, AFM could not be used to visualize structural details within the cells even when the AFM tip was used as a micro-scalpel to open the outer cell wall of the bacteria. The advantage of TEM is its superior ability to visualize ultrafine, intracellular detail. An obvious disadvantage of TEM is that the bacteria are not living as they are in the AFM. Used together, AFM and TEM offer complementary information for the analysis of biominerals within MTB.

Dedication

This document is dedicated to Chiharu Yaginuma.

Acknowledgments

I have many people to acknowledge and I may have left some people out, but after 75 months of doing my PhD it is hard to remember everyone. Steven Lower and

Brian Lower for advising me, providing me with support and giving me the opportunity to do my PhD, Olli Tuovinen, Wendy Panero, Richard Dick and Linda Dick for using their lab, Chiharu Yaginuma for a lot of things, Uncle Doctor Steve Goldsmith for all the

BBQs, Charles Park for giving me somewhere to write my dissertation, Eric Taylor,

Lumarie Perez-Guzman for help with my dissertation and for the wonderful food, Lijun

Chen, Nadia Casillas-Ituarte, Alyssa Bancroft for being a good friend, Sara Cole and

Richard Montione at CMIF for putting up with all my requests, Todd Matulnik at the fermentation lab, Henk Colijn at CEOF, Nanotech West, Azuma Taoka and Yoshihiro

Fukumori at Kanazawa University, Japan Society for the Promotion of Science and the

National Science Foundation East Asia and Pacific Summer Institutes for providing me with the opportunity to go to Japan, Marit Nilsen-Hamilton and Pierre Palo for sending me the plasmid containing the mms6 gene and for helping me with the protein purification, Dennis Bazylinski and Chris Lefevre for helping me culture AMB-1, Wei

Lin, Jinhua Li and Yongxin Pan at the Chinese Academy of Science the Institute of

Geology and Geophysics for showing me how to isolate MTB, the National Science

Foundation East Asia and Pacific Summer Institutes for giving me another chance to do research in Asia, The Ohio State University for giving me a one year fellowship and for giving me a grant through the Alumni Grants for Graduate Research and Scholarship, the

Geological Society of America for their grant which allowed me to collect all the samples for chapter 3, the Friends of Orton Hall for providing me with money to present at M&M,

Carmen Valverde-Tercedor for her help with my research and dissertation, Brent Curtiss for all the computer assistance, Angeletha Rogers for over six years of help in the

Geology Department, Olivia Amerendes for helping me order countless supplies, Michael

Carter for helping me centrifuge and lyse my bacteria, Aurelie Snyder and Scott Wetzel for all of their confocal knowledge, Jim Pawley who allowed me to participate five amazing times in his live cell microscopy workshop in Vancouver, Peter Styger for giving me the chance to learn confocal microscopy, Takatoshi Karasawa, Amanda Davey for helping out in Senegal, the Microscopy Society of America, Eric Rees at Research and Testing for answering all of questions, PNNL and EMSL for giving me the chance to do research at their facility several times, JoAnn Donohue at the Ohio State Department of Speech and Hearing, Sander Flaum for all of his support and inspiration, my parents

Bill and Candy McHoes, Tonya and Todd Elmer, Juliana Lam, Sherry Cady for inspiring me to do field work and microscopy and getting me started in geomicrobiology, and of course Dr. John Dash for teaching me how to use the TEM way back in September of

1996, for inspiring me to pursue microscopy, and for many other things.

vii

Vita

1997...... B.S. Biology, Portland State University

2004...... M.S. Geology, Portland State University

2006-7…………………………………...... The Ohio State University Graduate School

Fellowship

2007-2011 ...... Graduate Teaching Associate, School of

Earth Sciences, The Ohio State University

2007...... Summer Research Institute Fellow, Pacific

Northwest National Laboratory

2008…...... Summer Research Fellowship in Kanazawa,

Japan, National Science Foundation East

Asia and Pacific Summer Institutes

2010…...... Summer Research Fellowship in Beijing,

China, National Science Foundation East

Asia and Pacific Summer Institutes

Publications

Lower, B. H., Lins, R. D., Oestreicher, Z., Straatsma, T. P., Hochella Jr, M. F., Shi, L., and Lower, S. K., 2008, In vitro evolution of a peptide with a hematite binding motif that may viii

constitute a natural metal-oxide binding archetype. Environmental Science & Technology, 42: 3821-3827. Oestreicher, Z., Lower, S.K., Lin, W., Lower, B.H. 2012, Collection, isolation and enrichment of naturally occurring magnetotactic bacteria from the environment. Journal of Visualized Experiments, 69, e50123, doi:10.3791/50123. Oestreicher, Z., Valverde-Tercedor, C., Chen, L., Jimenez-Lopez, C., Bazylinski, D. A., Casillas- Ituarte, N. N., Lower, S. K., and Lower, B. H., 2012, Magnetosomes and magnetite crystals produced by magnetotactic bacteria as resolved by atomic force microscopy and transmission electron microscopy: Micron, 43: 1331-1335.

Fields of Study

Major Field: Geological Sciences and Microbiology

Table of Contents

Abstract...... ii

Dedication...... v

Acknowledgments...... vi

Vita...... viii

Publications...... viii

Fields of Study ...... ix

List of Tables ...... xiv

List of Figures...... xvi

Chapter 1: Introduction...... 1

Characteristics...... 3

Phylogeny ...... 8

Environment...... 9

Magnetofossils ...... 11

Magnetospirillum magneticum ...... 15

Genetics...... 15

Magnetosomes ...... 19

Magnetosome Formation ...... 20

Proteins ...... 26

Dissertation research...... 29

Chapter 2: Collecting Magnetotactic Bacteria from the Environment ...... 32

Abstract...... 32

Collection...... 33

Magnetotactic Bacteria Isolation ...... 34

Magnetotactic Bacteria Racetrack ...... 36

Magnetotactic Bacteria Enrichment...... 38

Magnetotactic Bacteria Observation by Light Microscopy...... 38

Magnetotactic bacteria Observation by Transmission Electron Microscopy (TEM) ... 39

Discussion...... 42

Chapter 3: Magnetotactic Bacteria from Mickey Hot Springs and Pavilion Lake ...... 46

Abstract...... 46

Introduction...... 47

Methods...... 50

Sample collection...... 50

Magnetotactic Bacteria Enrichment...... 50

Transmission Electron Microscopy ...... 51

Phylogenetic Analysis...... 52

Results...... 53 xi

Sample locations ...... 53

Mickey Hot Springs ...... 56

Pavilion Lake ...... 62

Discussion...... 64

Chapter 4: Localization of Mms6 Proteins ...... 72

Abstract...... 72

Introduction...... 73

Methods...... 76

Protein Purification ...... 76

Mass spectrometry of recombinant protein...... 77

Antibody production ...... 77

Immunoblotting...... 78

Magnetosome isolation and purification...... 78

Cell fractionation ...... 79

Embedding and labeling with Immunogold...... 79

Fluorescence microscopy...... 81

Results...... 82

Recombinant protein purification ...... 82

Immunoblotting...... 82

Microscopy analysis...... 86

Discussion...... 91

xii

Chapter 5: Atomic Force Microscopy and Transmission Electron Microscopy of

Magnetotactic Bacteria ...... 97

Abstract...... 97

Introduction...... 97

Material and methods...... 99

Growth and preparation of bacteria for AFM and TEM...... 99

Atomic force microscopy (AFM) ...... 100

Transmission electron microscopy ...... 100

Results...... 101

Imaging microorganisms, proteins, and antibodies with AFM...... 101

AFM of Mms6 proteins synthesized by M. magneticum AMB-1...... 102

TEM of M. gryphiswaldense MSR-1...... 107

AFM and TEM of magnetosomes produced by M. gryphiswaldense MSR-1 ...... 107

Discussion...... 110

Chapter 6...... 115

Conclusions...... 115

Appendix A: Additional Methods and Data for Chapter 3...... 131

Appendix B: Additional Methods and Data for Chapter 4 ...... 147

xiii

List of Tables

Table 2 Some of the known proteins of magnetotactic bacteria and their function.* ...... 27

Table 3 List of specific reagents and equipment...... 35

Table 4 Geochemistry of Mickey Hot Springs and Pavilion Lake. The units for the ions are mg/L.*...... 55

Table 5 Table showing the measurements of the lengths and width of the cells from

Mickey Hot Springs...... 133

Table 6 Table showing the measurements and frequency calculations for the magnetosomes from the Mickey Hot Springs samples...... 134

Table 7 Table for calculating the frequecy of the magnetosome length to width ratios and the frequency of lengths for the Mickey Hot Springs sample...... 138

Table 8 Table showing the measurements of the lengths and widths of the cells from

Pavilion Lake...... 141

Table 9 Table showing the measurements and frequency calculations for the magnetosomes for the Pavilion Lake sample...... 142 xiv

Table 10 Table for calculating the frequency of the magnetosome length to width ratios and the frequency of lengths for the Pavilion Lake sample...... 145

List of Figures

(Fe3O4). Arrows in both images point to single crystals. Scale bar is 1 µm in A and 0.5

µm in B. Figure in A is from (Lefèvre et al., 2011)...... 5

Figure 2 Diagram showing how earth’s magnetic field influences the swimming direction of magnetotactic bacteria. In the northern hemisphere, the bacteria generally swim towards geographic north (red bacterium), whereas in the southern hemisphere, it is toward geographic south (orange bacterium). This means that they will swim toward the south pole of a magnet in the northern hemisphere and the north pole of a magnetic in the southern hemisphere. (Adapted with permission from (Faivre et al., 2008). Copyright

(2008) American Chemical Society)...... 7

Figure 3 Phylogeny of magnetotactic bacteria demonstrating the 3 different phyla of

Proteobacteria (α-, γ-, and δ-) and the Nitrospirae with different representative magnetotactic bacteria within each one. The most numerous phylum is the α- proteobacteria, which also shows non-magnetic bacteria. (Diagram modified from Jogler and Schüler, 2009)...... 10 xvi

In the presence of higher than optimal oxygen, the cells swim downward to regions of less oxygen. Swimming too far away from the oxygen into more a reduced environment causes the bacteria to swim upward. All along the cells are oriented with the magnetic field that constrains the swimming direction of the bacteria to either up or down.

(Diagram modified from Spring and Bazylinski, 2006)...... 12

Magnetotactic bacterial magnetite (labeled as dotted boxes on the figure) typically lies within the single domain region. (Diagram modified from Chang and Kirschvink, 1989).

...... 14

Figure 6 Transmission electron micrograph of Magnetospirillum magneticum AMB-1 showing magnetosomes inside of cell. Scale bar is 500 nm...... 16

Figure 7 A chain of magnetosomes that were extracted and purified from

Magnetospirillum magnetotacticum AMB-1. The black arrow points to a single crystal of magnetite, which are surrounded by a membrane (black arrowheads). Scale bar is 100 nm...... 21

xvii

Figure 8 Examples of the different crystal habits used by magnetotactic bacteria.

(Reprinted by permission from Macmillan Publishers Ltd: [Nature Review Microbiology]

(Bazylinski and Frankel, 2004), copyright (2004)...... 22

Figure 9 A clear plastic bottle containing a sediment and water sample collected from near the Olentangy River in Columbus, Ohio (USA). The bottle contains approximately one-half sediment and one-half water. The south end of a magnet is placed approximately 1 cm above the sediment for up to several hours (A). After removing some of the fluid from near the magnet on the inside the container, it is placed inside of a capillary racetrack where the MTB swim through a cotton plug (arrow) towards the south end of a bar magnet (B). A close up view of the capillary racetrack showing the sample, cotton, filtered fluid, sealed end of the capillary tube and south end of a bar magnet (C).

...... 37

Figure 10 Once the MTB have been enriched from the racetrack, a small drop can be placed on a coverslip, which is then flipped upside down and placed on an o-ring that is resting on a slide. This slide-o-ring-coverslip sandwich can be placed on a light microscope stage and viewed using a 60X dry objective (oil lenses are inconvenient to use with a hanging drop)...... 40

Figure 11 Bright field microscope image of MTB swimming (thin long arrows) and gathering at the edge of the hanging drop (short arrows) which is next to the south pole of a bar magnet...... 41

Figure 12 Transmission electron microscope images of various MTB isolated from the wetland near the Olentangy River in Columbus, Ohio. There were several different

xviii morphotypes and they all contain crystals (white arrows) and some have inclusions

(white arrowheads). Scale bar for each image is 500 nm...... 43

Figure 13 Map of Western North America. Sample sites are designated with a star.

Pavilion Lake is in British Columbia, Canada, and Mickey Hot Springs is located in southeastern Oregon...... 54

Figure 14 Phylogenetic tree from the 16S rRNA gene sequences of the Mickey Hot

Springs samples, labeled Oestreicher-1:HH5G4AX03*****. Maximum composite likelihood method was used to construct the tree. Each line represents how closely the sequences are related; the scale bar at the bottom shows the percentage divergence. The dominant group of magnetotactic bacteria was from the Nitrospirae phylum. The

Genbank accession numbers from the non-Mickey Hot Springs samples are at the end of each branch and given as a single or double letter followed by 5 or 6 digits. An

Alphaproteobacteria was used to root the tree...... 57

Figure 15 (A) Vibrioid magnetotactic bacterium from Mickey Hot Springs, Oregon USA.

The cells averaged 2.2 µm long and 0.62 µm wide. The magnetosomes were 84 nm long and 39 nm wide. The chain of magnetosomes contains bullet-shaped crystals (white arrow). (B) The magnetite was analyzed using energy-dispersive spectroscopy (EDX) which showed that the crystals are composed of iron and oxygen, which is indicative of magnetite. Copper peaks are from the copper support grid. A single flagellum comes out of one side of the cell (black arrow). Scale bar is 500 nm...... 59

Figure 16 Length versus the width and shape factor (crystal width/crystal length) diagrams for the Mickey Hot Springs sample. The dashed line in A represents width

xix equal to length. The areas labeled in B represents the approximate area of single domain magnetite based on Butler and Banerjee, 1975...... 60

Figure 17 (A) Frequency graphs of the shape factor showing the distribution of different shape factors of the magnetite crystals. (B) Frequency of the width of the magnetosomes showing narrow size distribution of the widths of the magnetite crystals...... 61

The Genbank accession numbers from the non-Pavilion Lake samples are at the end of each branch and given as a single or double letter followed by 5 or 6 digits. An

Alphaproteobacterium, Magnetospirillum gryphiswaldse, was used to root the tree...... 63

Figure 19 (A) Spirillum-shaped magnetotactic bacterium from Pavilion Lake, British

Columbia, Canada. The cells averaged 2.9 µm long and 0.3 µm wide with a single chain of magnetosomes (white arrow) and a single flagellum (black arrow). Scale bar is 500 nm. (B) The EDX showed the crystals were made of iron and oxygen (the copper peak is due to the copper support grid) indicative of magnetite. The crystals averaged 47 nm x

44 nm...... 65

Figure 20 (A) Graph showing the length versus the width of the magnetite crystals from the Pavilion Lake sample illustrating that the length and width are nearly equal. (B) The

xx length versus the shape factor of the magnetosomes showing that the crystals fall within the single domain size range. The figure is based on Butler and Banerjee, 1975...... 66

Figure 21 (A) Graph showing the frequency of the shape factor showing the crystals are nearly equal in width and length. (B) Graph of the frequency of the width of the magnetosomes showing the narrow size distribution of the crystals of magnetite...... 67

Figure 22 The different fractions from the protein purification process resolved using

SDS-PAGE. Lane 1 is the cell lysate, lane 2 is the flow through, lanes 3-6 are the washing steps with increasing concentrations of imidazole (5,mM 5mM, 15mM, and

30mM), lanes 7-9 and 11-13 are the protein elutions using 300mM imidazole, and lane 10 is the protein marker...... 83

Figure 23 SDS-PAGE showing the protein and cell fractions from M. magneticum AMB-

1 used to test the affinity of the Mms6 antibody in the immunoblot. Lane 1) cell membrane fraction (11 µg), lane 2) soluble protein (4 µg), lane 3) magnetosome membrane (11 µg), lane 4) recombinant Mms6 protein (0.05 µg), lane 5) recombinant

Mms6 protein (0.5µg), lane 6) BenchMark protein ladder...... 84

Figure 24 (A) Immunoblot analysis of recombinant Mms6 protein and three different cell fractions from M. magneticum AMB-1 (same as Figure 23). (B) The same immunoblot as A except with 300ug of recombinant Mms6 added to the antibody mixture as a control.

...... 85

Figure 25 Fluorescently labeled Magnetospirillum magneticum AMB-1 cells using anti-

Mms6 (1:400) as the primary antibody and goat anti-rabbit Dylight 488 (1:100) as the secondary. A) The combined Nomarski and fluorescent images showing the outline of

xxi

M. magneticum cells with the fluorescent tag running in the middle of the cells. B) The same field of view as A, but using only Nomarski microscopy. C) Cells excited with the

488 nm laser. Scale bar 5 µm...... 89

(black arrows). Inset shows a detail of the immunolabeling on one of the magnetosomes.

B) Negative control for the immunolabeling showing an ultrathin section treated exactly the same as the ultrathin section in A, but substituting 0.5% BSA Tris-HCl for the primary antibody. C) Another negative control for the immunolabeling treated exactly the same as the section in A, but with 5% pre-immune serum substituted for the primary antibody. Scale bar is 500 nm for all three images...... 90

Figure 27 Purified magnetosomes from M. magneticum AMB-1. A) Magnetosomes

(black arrows) labeled with anti-Mms6 antibody (1:4000) and secondary labeled with goat anti-rabbit antibody conjugated with 10 nm colloidal gold (white arrows). The inset shows a single magnetite crystal labeled with gold conjugated antibody. The space between the white arrows depicts the magnetosome membrane. B) Negative control for the immunolabeling treated exactly the same as A, except substituting 0.5% BSA-Tris-

HCl for the primary antibody. C) Negative control treated exactly the same way as A, but substituting 5% pre-immune rabbit serum in place of the primary antibody. Scale bar is 100 nm in each image...... 92

xxii

Figure 28 (A) Topography and (B) amplitude images of M. gryphiswaldense MSR-1 acquired with tapping mode AFM. (C) Three-dimensional topography/amplitude composite image of M. gryphiswaldense MSR-1. Inclusion bodies (ib) and magnetosome chain (mc) are shown. Scale bars for length are provided inside each AFM image. A monochrome scale bar for height is provided to the right of (C)...... 103

Figure 29 Topography of Mms6 protein molecules deposited on magnetite (Fe3O4) and acquired with tapping mode AFM. Arrows indicate the position of Mms6 proteins. A scale bar for length is provided inside the figure. A color-coded scale bar for height is provided to the right of the topography image...... 104

Figure 30 Topography of anti-GFP molecules deposited on mica and acquired with tapping mode AFM. The anti-GFP molecules were observed to have (A) globular, (B) V- shaped, or (C) Y-shaped morphologies. The drawings to the right show the different conformations of the Fab and Fc fragments of the antibodies (Silverton et al., 1977). A scale bar for length is provided inside (A). A color-coded scale bar for height is provided to the right of the topography images...... 106

Figure 31 TEM image of M. gryphiswaldense MSR-1. Inclusion bodies (ib) and two magnetosome chains (mc) are shown. A scale bar for length is provided...... 108

Figure 32 A) TEM image of magnetosomes and (B) AFM phase image of a magnetosome isolated from M. gryphiswaldense MSR-1. The AFM image was collected using tapping mode. Scale bars for length are provided for both images...... 109

Figure 33 Field site for Mickey Hot Springs. Upper picture is the view of the main pool

(Morning Glory) with a narrow channel coming out of the pool toward the bottom of the

xxiii picture (red arrow) to a smaller pool where the sample was collected (white arrow). The

...... 131

Figure 34 Example of where the EDX spectra were collected from the Mickey Hot

Springs sample (A). The two insets in the lower right of panel A show where the EDX were collected for the crystal (red circle) and for the background of the cell (red rectangle). The EDX spectrum for the crystal showing iron and oxygen as the major components (B). The EDX spectrum for the background of the cell showing large peaks for copper (Cu) (from the grid) and smaller peaks for arsenic (As), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), and iron (Fe)...... 139

Figure 35 Field site for the Pavilion Lake sample. The upper picture is a view of a portion of the lake. The sample was collected from the northern shore shown in the lower image. The lower picture shows where the sediment was disturbed after taking the sample from the lower right of the upper image...... 140

EDX spectrum for the background of the cell showing large peaks for copper (Cu) (from the grid) and smaller peaks for magnesium (Mg), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), and potassium (K)...... 146

Figure 37 The plasmid containing the mms6 gene...... 150

xxiv

Figure 38 Data from the sequencing of the purified Mms6 protein showing that it corresponds to Mms6...... 156

Figure 39 Examples of sections labeled with concentrated primary antibodies. A)

Sections labeled with primary antibody at a concentration of 1:200. B) Sections labeled with primary antibodies at a concentration of 1:400. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images...... 157

Figure 40 Examples of sections labeled with concentrated primary antibodies. A)

Sections labeled with primary antibody at a concentration of 1:600. B) Sections labeled with primary antibodies at a concentration of 1:800. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images...... 158

Figure 41 Examples of sections labeled with concentrated primary antibodies. A)

Sections labeled with primary antibody at a concentration of 1:1000. B) Sections labeled with primary antibodies at a concentration of 1:1200. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images...... 159

Figure 42 Examples of sections labeled with concentrated primary antibodies. A)

Sections labeled with primary antibody at a concentration of 1:1600. B) Sections labeled with primary antibodies at a concentration of 1:2000. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images...... 160

Figure 43 Example of a section labeled with concentrated primary antibodies, in this case the primary antibody was at a concentration of 1:2400 and the secondary antibody concentration was 1:100. Scale bar is 500 nm...... 161

xxv

Figure 44 Examples of sections labeled with concentrated primary antibodies. A)

Sections labeled with primary antibody at a concentration of 1:2000. B) Sections labeled with primary antibodies at a concentration of 1:4000. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images...... 162

xxvi

Chapter 1: Introduction

There are many instances when there is a direct overlap between geology and biology. Perhaps one of the best examples of this is magnetotactic bacteria. These are living organisms that have the ability to synthesize small crystals of magnetite inside their cells. They do this by transporting iron from the environment into their cells where they use proteins to biomineralize the iron into nano-sized magnets of either magnetite

(Fe3O4) or greigite (Fe3S4). These aquatic organisms use the magnets to passively align themselves with Earth’s magnetic field the same way an explorer would use a compass to orient him or herself. Once the bacteria are oriented, they can use their polar flagellum to swim up or down in the water column to a suitable environment. For magnetotactic bacteria a suitable environment is a specific region in aquatic sediments that is very low in oxygen.

The Earth’s magnetic field is at least 3.5 billion years old and we know that life developed around that same time (Schopf, 1993; Tarduno et al., 2010). This adaptation is likely deeply rooted in the biological evolutionary tree. Once the bacteria have completed their lifecycle, the organic components of the cells decompose, but the magnetic crystals can persist and become part of the fossil record. These crystals become part of the geological record in the form of magnetofossils, which record the magnetism of Earth at the time the bacteria mineralized the iron containing crystals. The research in

1 my dissertation investigates some of these alluring geological-biological phenomena within magnetotactic bacteria.

This dissertation is laid out in five chapters. Chapter 1 is a general introduction to magnetotactic bacteria. Chapter 2 explains how magnetotactic bacteria can be collected and isolated from an aquatic environment. Chapter 3 presents data related to the isolation and identification of magnetotactic bacteria from two different unique aquatic environments, one in Oregon (USA) and one in British Columbia (Canada). Chapter 4 discusses the use of transmission electron microscopy (TEM) to examine the location of

Mms6, a key magnetite-mineralizing protein, in Magnetospirillum magneticum AMB-1.

In Chapter 5, atomic force microscopy is use to observe biomineralization in M. magneticum AMB-1. Finally, Chapter 6 provides a summary of the dissertation research.

I was attracted to the study of magnetotactic bacteria (MTB) for a number of reasons. This type of bacteria is an important part of the fossil record not only because of their contribution to the paleomagnetic record but also because they represent the first (or one of the first) known examples of biomineralization (Kirschvink and Lowenstam, 1979;

Kopp and Kirschvink, 2008; Petersen et al., 1986). They have been implicated as a proxy for the formation of magnetite in the Martian meteorite ALH84001 (McKay et al., 1996) and as an analogue for biosignatures on Mars (Thomas-Keprta et al., 2002). The process of magnetite formation in MTB has possible applications in the field of medicine for enhanced contrast of magnetic resonance imaging (Benoit et al., 2009). The magnetosomes of MTB can be studied as a model system for how organelles form in prokaryotic cells (Murat et al., 2010). Also, magnetosome formation can be used to help

2 understand the process of biomineralization in more complex systems, such as bones and teeth in animals. MTB use magnetite to assist them in navigation similar to other life forms such as bees, birds, and fish (Blakemore, 1975; Esquivel and de Barros, 1986;

Frankel, 1984; Gould et al., 1978; Mann et al., 1988a; Walcott et al., 1979). Therefore,

MTB could be used to gain understanding of how this process works in larger animals

(Kirschvink et al., 2001; Mann et al., 1988b; Walcott et al., 1979). MTB have an impact on the iron cycle because their magnets contain iron (Simmons and Edwards, 2007b).

MTB have been modified for bioremediation of heavy metals such as cadmium (Tanaka et al., 2010a). Studying MTB can be useful as a model for generating synthetic magnetite crystals (Amemiya et al., 2007; Prozorov et al., 2007a). Finally, MTB also play an important role in larger bacteria communities in nature (Bazylinski et al., 2000).

Characteristics

Salvatore Bellini was the first person to describe MTB in 1963 in his publication

“Su di un particolare comportamento di batteri d’acqua dolce” (Bellini, 1963). His research was published in Italian and therefore he is often not credited with their discovery. The person who is most often recognized as the discoverer of MTB is Richard

Blakemore because he published an article describing MTB in English in the journal

Science in 1975 (Blakemore, 1975). He collected MTB from sediments in a salt marsh next to Eel Pond in Woods Hole, Massachusetts (Blakemore, 1975). The defining characteristic of these organisms is their ability to synthesize magnetic particles within their cells, which are composed of either magnetite (Fe3O4) or greigite (Fe3S4) or both

(Figure 1) (Bazylinski et al., 1995; Bazylinski et al., 1993; Frankel et al., 1979; Heywood

3 et al., 1990). These particles form a chain (some MTB have more than one chain) of varying number of magnetosomes, which are held together by a protein that anchors the chain to the cell wall (Komeili et al., 2006; Komeili et al., 2004). The chain of magnetosomes gives MTB the ability to passively orient themselves relative to the

Earth’s magnetic field. Using their polar flagella, MTB are able to actively swim within their habitat to find an optimal position within the oxygen/sulfide/redox gradient, a process known as magnetoaerotaxis (Frankel et al., 1997). This is a much more efficient method of finding the appropriate redox environment, as opposed to simply relying on tumbling or diffusion because magnetotaxis reduces the search problem from three dimensions, to only one dimension (up or down) (Bazylinski and Frankel, 2004).

Magnetotactic bacteria come in many shapes and sizes; they can be spiral, vibriod, coccoid, or rod shaped. They have a Gram-negative cell wall, and they are motile by means of flagella that can be on one end of the cell, both ends of the cell, or all around the cell (Amann et al., 2007; Balkwill et al., 1980; Spring and Schleifer, 1995).

One unusual type of magnetotactic bacteria is the many celled magnetotactic prokaryote

(MMP). These are bacteria that form clusters of about 10 to 30 coccoid shaped cells,

(Farina et al., 1990; Rodgers et al., 1990). Dozens of magnetotactic bacteria have been discovered, but only a fraction of these have been grown in pure culture in the lab, including for example: Magnetospirillum magneticum AMB-1, Magnetospirillum gryphiswaldense MSR-1, Desulfovibrio magneticus MC-1 and MV-1, and

Magnetospirillum magnetotacticum MS-1 (Blakemore et al., 1979; Lefèvre et al., 2011;

Lefèvre et al., 2011a; Matsunaga et al., 1991; Meldrum et al., 1993; Spring et al., 1993).

Figure 1 Two different types of magnetotactic bacterial cells containing magnetic minerals. (A) Shows MTB, collected from a thermal spring in Nevada, containing several chains of the mineral greigite (Fe3S4). (B) A MTB cell, collected from a marsh near the Olentangy River in Ohio, containing a single chain of the mineral magnetite (Fe3O4). Arrows in both images point to single crystals. Scale bar is 1 µm in A and 0.5 µm in B. Figure in A is from (Lefèvre et al., 2011).

Each of these four species has had their genome sequenced (Matsunaga et al., 2005;

Richter et al., 2007). Because it is difficult to culture magnetotactic bacteria and because they are very diverse, it has been challenging to generate genetic or protein models that explain biomineralization processes within magnetotactic bacteria.

As noted above, all MTB have the ability to synthesize magnetic minerals, which allow the cells to orient in the Earth’s magnetic field. There are two types of magnetotaxis, axial and polar (Frankel et al., 1997). Polar taxis can be seen when a bacterium migrates toward one pole of the magnet, and when the magnet is reversed, the bacterium turns 180°. On the other hand, axial magnetotaxis is seen when a bacterium migrates back and forth along magnetic field lines and not towards one pole or the other.

When the poles are reversed, the axial-bacterium will turn 180° but still continues to swim back and forth. Most bacteria demonstrate polar magnetotaxis, but there is one example of cultured Magnetospirillum that shows axial magnetotaxis (Frankel et al.,

1997).

Most magnetotactic bacteria in the northern hemisphere migrate towards the

North Pole and in the southern hemisphere, they migrate towards the South Pole (Figure

2). However there has been one reported example of magnetotactic bacteria in the northern hemisphere swimming toward the South Pole (Simmons et al., 2006). In the many celled magnetotactic prokaryote, clusters of coccoid MTBs, the response to magnetic fields is not the direct U-turn response to a change in dipole that are seen in unicellular magnetotactic bacteria. Many celled magnetotactic prokaryote are magneto-

receptive, and display four types of responses, ping-pong, rotation, walking, and free motion (Greenberg et al., 2005; Keim et al., 2007b).

Phylogeny

It has been a challenge to create a phylogeny of MTB. Since the first publication of magnetotactic bacteria (Bellini, 1963; Blakemore, 1975), many more MTB have been isolated, but only a few have been cultured. Some have proposed to use the number of magnetosome chains in cells or the morphology of the cells to create phylogenic trees

(Thornhill et al., 1994). But such phylogenic approaches are not very good taxonomic indicators because the number of chains can change within a given type of magnetotactic bacteria (Spring et al., 1994).

The most widely used type of phylogenetic classification for MTB has been the use of 16S rDNA (Burgess et al., 1993; Flies et al., 2005b; Lin et al., 2009; Schüler and

Frankel, 1999). This method of analysis is done by using a bacterial primer (for example,

27F 5’-AGAGTTT GATCCT GGCTCAG-3’ and 1492R 5’-

GGTTACCTTGTTACGACTT-3’) to make copies of the 16S rRNA gene (Lin and Pan,

2009). This method is effective because the 16S rRNA gene is a very conserved region, but has enough variation to be used as a fingerprint to differentiate between different types of bacteria. Figure 3 shows a phylogenetic tree created with sequence data from

16S rDNA.

Another method used in conjunction with 16S rDNA sequencing is fluorescence in situ hybridization (FISH) (Spring et al., 1994). This method uses rRNA probes, which

8 bind to the rRNA in the cell, and fluoresce when exposed to a specific wavelength of light in a light microscope. FISH is often used to confirm and quantify the existence of magnetotactic bacteria and quantify the number of cells in samples collected from the same environment for which the 16S rDNA data have been determined.

Using the 16S approach, most of the magnetotactic bacteria fall into two phylogenetic groups, the Nitrospirae (Lefèvre et al., 2010; Lin et al., 2012) and

Proteobacteria; of which there are three classes: α-proteobacteria (Amann et al., 2007),

γ-proteobacteria (Lefèvre et al., 2011b; Simmons et al., 2004), δ-proteobacteria (Lefèvre et al., 2011b) with the α-proteobacteria being the most common type (Amann et al.,

2007).

Environment

MTB live in the sediments of a wide range of aquatic habitats (both freshwater and saline) as well as wet soils and stratified water columns (Bazylinski et al., 1995;

Bazylinski et al., 2000). Magnetotactic bacteria have been found in estuaries, marine sediments, and freshwater sediments (Flies et al., 2005b). Regardless of the habitat,

MTB prefer to live at the oxic-anoxic transition zone (OATZ) (Bazylinski and Frankel,

2004). This habitat preference is one of the primary reasons that MTB are so difficult to culture in the lab. In short, it is very challenging to reproduce natural oxygen gradients in manufactured culture media in the lab (Bazylinski and Williams, 2007)

When sediments are collected and used to extract magnetotactic bacteria, the bacteria can take several weeks or months to grow. As they continue to grow, different

Figure 3 Phylogeny of magnetotactic bacteria demonstrating the 3 different phyla of Proteobacteria (α-, γ-, and δ-) and the Nitrospirae with different representative magnetotactic bacteria within each one. The most numerous phylum is the α- proteobacteria, which also shows non-magnetic bacteria. (Diagram modified from Jogler and Schüler, 2009).

10 morphotypes successively dominate the community. In laboratory experiments, different species occupy different zones within the oxygen gradient, but the largest number of

MTB cells occurred just below the suboxic zone of the sediment (Flies et al., 2005b).

Magnetotactic bacteria have been studied individually, but not as a community, so their role in the microbial community or their role in iron and sulfur cycles has not been thoroughly examined. In stratified water columns magnetite-producing MTB are more common above the chemocline (the region where there is oxygen), while greigite producers are more common below this (Simmons et al., 2004). Most MTB occupy the oxic-anoxic transition zone (OATZ) in water columns and sediments, which is a gradient consisting of oxygen and reduced sulfur (Bazylinski and Frankel, 2004).

Iron levels also play a role in the occurrence and distribution of MTB in the environment (Stolz, 1993). This is because MTB need to form iron-containing magnets for magnetotaxis. Magnetotaxis allows cells to migrate between the oxic and anoxic transition zones. It is believed that the MTB shuttle themselves back and forth between the micro-oxic zone and the anaerobic zone, accumulating sulfur in the anaerobic zone and then returning to the microxic zone to oxidize the sulfur (Figure 4) (Bazylinski and

Frankel, 2004). Because of this narrow environmental niche, magnetite fossils

(magnetofossils) have been proposed as a paleoenvironmental indicator (Kopp and

Kirschvink, 2008).

Magnetofossils

Magnetite can be found in sediment and is the dominant form of remanence in marine sediments (Petersen et al., 1986). The term magnetofossils was coined to

Figure 4 A schematic representation of how the combination of being aligned with the magnetic field and the presence of oxygen and sulfur provides a redox reaction that manipulates the flagella and causes the bacteria to swim in a specific direction. The red arrows refer to the direction of the dipole moment given by the chain of magnetosomes. In the presence of higher than optimal oxygen, the cells swim downward to regions of less oxygen. Swimming too far away from the oxygen into more a reduced environment causes the bacteria to swim upward. All along the cells are oriented with the magnetic field that constrains the swimming direction of the bacteria to either up or down. (Diagram modified from Spring and Bazylinski, 2006).

12 specifically refer to single domain crystals of magnetite that were formed biogenically

(Kirschvink and Chang, 1984). Single domain crystals are small enough to have only one domain wall, so it is uniformly magnetized in one direction (Butler and Banerjee, 1975), which makes them useful for navigation as opposed to multi-domain magnetite. Figure 5 shows the stability field for magnetite as a function of particle size. As shown on this figure, the magnetite produced by MTB typically fall into the single domain category.

Magnetofossils are not only important for providing information on past environments on Earth, but they also have implications for life on other planets. For example, magnetofossils were purported to be in the Allan Hills meteorite (McKay et al.,

1996; Thomas-Keprta et al., 2000; Thomas-Keprta et al., 2009). Bacterial magnetite has characteristics that distinguish it from nonbiogenic magnetite (Arató et al., 2005; Kopp et al., 2006; Kopp and Kirschvink, 2008). Biologically controlled mineralization of magnetite produces crystals that are chemically pure, species-specific morphology and crystal perfection, narrow size range, unique anisotropic nanometer-sized crystals, and elongation in a specific direction (Benzerara and Menguy, 2009). Other factors that are indicative of biological origin are the presence of chains of magnetite similar to what are found in cells, magnetic properties that are specific to biologic magnetite and size distributions (Arató et al., 2005; Kobayashi et al., 2006; Kopp et al., 2006; Kopp and

Kirschvink, 2008).

Figure 5 Stability field diagram of magnetite as a function of grain size. The left side of the diagram represents elongated parallel-pipe shaped magnetite and the right side represents more cuboctahedral shaped crystals. The solid lines delineate between the different domains of magnetite (multidomain, single domain, and superparamagnetic). Magnetotactic bacterial magnetite (labeled as dotted boxes on the figure) typically lies within the single domain region. (Diagram modified from Chang and Kirschvink, 1989).

Magnetospirillum magneticum

One of the best-characterized species of magnetotactic bacteria is

Magnetospirillum magneticum AMB-1 (Figure 6) (Matsunaga et al., 2005; Matsunaga et al., 1991). Figure 6 shows a transmission electron microscope (TEM) image of a M. magneticum AMB-1 cell. AMB-1 cells are about 0.4 to 0.6 µm in diameter, and are spiral shaped. They are cultured in the lab using magnetospirillum growth media,

Wolfe’s mineral solution, and Wolfe’s vitamin solution (Appendix B). They can grow either in microaerophilic conditions by filling the headspace with 99% nitrogen, or aerobically (Blakemore et al., 1985). But when grown aerobically, they do not produce abundant magnetosomes. Grown microaerobically, the cells produce one chain of magnetosomes averaging over 15 magnetite crystals (Matsunaga et al., 1991). Each crystal is usually approximately 50 nm in diameter (Matsunaga et al., 1991). The bacteria are mobile by means of a single flagellum on each end of the cell.

Genetics

Magnetosome production is under careful genetic control. The genome of magnetotactic bacteria have approximately 4,500 kb with approximately 4,000 open reading frames (ORFs) (Richter et al., 2007). For example M. magneticum AMB-1 has

4,967,148 base pairs and 4,559 ORFs. Approximately 1% of the ORFs in the genome are associated with magnetotaxis (Jogler and Schüler, 2009). Based on the gene sequences of 4 MTB: magnetic coccus strain MC-1, Magnetospirillum magnetotacticum,

Magnetospirillum magneticum, and Magnetospirillum gryphiswaldense, there are 891

Figure 6 Transmission electron micrograph of Magnetospirillum magneticum AMB-1 showing magnetosomes inside of cell. Scale bar is 500 nm.

16 genes shared amongst magnetotactic bacteria and 28 of these are necessary for magnetotaxis (Table 1) (Jogler and Schüler, 2009). Some of the 28 genes are located in a specific region of the genome called the magnetosome gene island (MAI) (Schübbe et al.,

2003).

The MAI is a 130 kb region containing most of the magnetosome membrane genes, hypothetical genes, and many transposable regions (Ullrich et al., 2005). This region is common to all MTB, but it varies amongst different groups of MTB. It is disputable whether or not magnetotactic bacteria acquired the genes necessary for magnetosome biosynthesis independently, or if they were acquired through horizontal gene transfer (HGT) (Jogler et al., 2010; Jogler and Schüler, 2007; Ullrich et al., 2005).

It has been found that the magnetosome formation is very similar between MTB of two different phyla (Jogler et al., 2011). It is most likely that the transposable regions are what allow the MAI to be transferred horizontally among bacteria (Jogler et al., 2009a;

Jogler and Schüler, 2007). Genes involved with magnetosome formation are not just included in the MAI, there are genes outside of the magnetosome island that control magnetosome production, for example, in M. gryphiswaldense 18 genes are located in the

MAI and 10 genes outside (Jogler et al., 2009a).

The diversity of MTB, may be due to the fact that the genes encoded within the

MAI are variable (Jogler and Schüler, 2009). Recently, a comprehensive study was done of many genes within the MAI of AMB-1 cells (Murat et al., 2010). It was found that many of the genes that generate the magnetosome compartment and line up the

Table 1 List of genes necessary for magnetotaxis based on a gene comparison of 4 different magnetotactic bacteria: magnetic coccus strain MC-1, Magnetospirillum magnetotacticum, Magnetospirillum magneticum, and Magnetospirillum gryphiswaldense (From Jogler and Schüler, 2009). Name of Gene Gene Type mamD Magnetosome membrane mamF Magnetosome membrane mamH Magnetosome membrane mamI Magnetosome membrane mamE Magnetosome membrane mamK Magnetosome membrane mamM Magnetosome membrane mamO Magnetosome membrane mamP Magnetosome membrane mamA Magnetosome membrane mamQ Magnetosome membrane mamB Magnetosome membrane mamS Magnetosome membrane mamT Magnetosome membrane mamC Magnetosome membrane mamD Magnetosome membrane mamZ Magnetosome membrane mamX Magnetosome membrane mamY Magnetosome membrane mamW Magnetosome membrane mamL Magnetosome membrane mamN Magnetosome membrane mamR Magnetosome membrane mamG Magnetosome membrane mamH-like Magnetosome membrane mamD-like Magnetosome membrane mms6 Magnetic particle membrane mmsF Magneticspecific particle membrane specific

18 compartments (which are done prior to the crystal formation) are conserved genes. In other words, they are not only common to most MTB, but they have homologues to non- magnetotactic cells including eukaryotic cells. Conversely, the genes involved with magnetite crystal biomineralization are not conserved (i.e. they are more easily lost). It was revealed through several gene deletions that the genes in the MAI have little affect on the phenotype of the cell and it turns out that most of the genes in the MAI affect the mineral (Lohße et al., 2011). However, only about 25% of the genes are dedicated to this, and about 50% of the MAI serves no obvious function (“junk genes”), which is further evidence that the MAI was acquired through horizontal gene transfer.

MTB can spontaneously loose the ability to synthesize magnetosomes, which is also indicative of transposable genes (Frankel et al., 1979; Ullrich et al., 2005). For example, this can happen when the bacteria are grown under oxidative stress or kept in the cold during stationary growth (Schübbe et al., 2003). When mutants occur, the magnetosomes may be smaller and/or misshapen, and the quantity typically diminishes or no magnetosomes are produced (Jogler and Schüler, 2009). Many mutants have been examined in order to elucidate their function. For example the deletion of the mamGFDC is a mutation that caused magnetite crystals to be misshapen, the mamK mutant caused the magnetosome chain to not anchor to the cell, and the mamJ mutant caused no chain to form at all (Komeili et al., 2006; Scheffel et al., 2007).

Magnetosomes

Magnetosomes contain two essential parts, a magnetic mineral surrounded by a membrane (Figure 7) (Gorby et al., 1988). The magnetic mineral found in most MTB is

19 magnetite (Fe3O4), however, some MTB mineralize greigite (Fe3S4) (Bazylinski et al.,

1993). Iron can account for 2-3% of the dry weight of the cell (Bazylinski and Frankel,

2004). There are typically about 20 magnetosomes in a chain per cell, but it can vary from species to species (Balkwill et al., 1980). The three dominant magnetite shapes are cuboidal, anisotropic (arrowhead/bullet/tooth-shaped), and parallelepiped (rectangular)

(Bazylinski and Frankel, 2004). Models for magnetosome formation differ from species to species. In a general the model has four steps: (i) initial invagination of the inner cell,

(ii) followed by protein localization in the magnetosome compartment, (iii) lining up the compartments, and (iv) mineral formation (Murat et al., 2010). Magnetosome membranes are formed prior to the nucleation of magnetite and the magnetosome membrane is the site of crystal nucleation (Faivre et al., 2007). Nucleation probably forms only at one site on the magnetosome membrane because of the single-crystal nature of magnetite, but this has never been resolved (Faivre et al., 2008).

While magnetic crystals in magnetosomes have a species-specific morphology, they share similar size characteristics. The crystals are chemically pure, their size ranges from 30 to 120 nm long, spaced about 10 nm apart, and the crystals have a definitive arrangement inside the cell (Balkwill et al., 1980; Bazylinski and Frankel, 2004). The three main crystal habits of magnetite crystals are cubic [100], dodecahedral [110], and octahedral [111] (Figure 8) (Devouard et al., 1998).

Magnetosome Formation

In the Magnetospirilla bacteria the crystals are oriented with the [111] face

(Figure 6) parallel to the axis of the magnetosome chain. But non-isometric forms have

Figure 7 A chain of magnetosomes that were extracted and purified from Magnetospirillum magnetotacticum AMB-1. The black arrow points to a single crystal of magnetite, which are surrounded by a membrane (black arrowheads). Scale bar is 100 nm.

Figure 8 Examples of the different crystal habits used by magnetotactic bacteria. (Reprinted by permission from Macmillan Publishers Ltd: [Nature Review Microbiology] (Bazylinski and Frankel, 2004), copyright (2004).

22 also been found such as prismatic and bullet shaped crystals (Bazylinski et al., 2007).

Twinning may occur in magnetite crystals as a result of two separate crystals growing together (Devouard et al., 1998). Greigite crystals have similar morphologies and species-specific sizes as magnetite crystals and some bacteria contain both types of minerals in their cells, but in separate chains (Bazylinski et al., 1995; Lefèvre et al.,

2011). Even though the two types of crystals have similarities, it is likely that they are under separate genetic control (Lefèvre et al., 2011).

While there are four general steps involved in the formation of magnetosomes, the formation of the actual crystals of magnetite crystals is less well understood. This is because not all MTB species follow the same mechanism of mineralization. In general, there are three basic steps involved in biomineralization: (i) proteins actively take up iron from the environment, (ii) this iron is transported into the magnetosome membrane, and

(iii) nucleation and growth of magnetite or greigite.

According to Scheffel and Schuler (Scheffel et al., 2007) the first step involves the active uptake of iron from the environment. Depending on the species it can be either in the form of Fe(III) or Fe(II) (Scheffel et al., 2005). The mechanism used by MTB to accumulate iron is poorly understood and varies from species to species. Some acquire

Fe3+ from the environment and bring it into the magnetosome directly (M. gryphiswaldense) while others may use siderophores (M. magnetotacticum &, M. magneticum) and then convert the iron to Fe2+ in the cytoplasm before bringing it into the magnetosome. Then, iron is brought into the magnetosome either from the cytoplasm or from the periplasmic space. Once the iron is inside the magnetosome, it begins to

23 nucleate and the mineral grows (Faivre et al., 2007). It may be that ferrihydrite is an intermediate step in this process and therefore there are proteins in the membrane involved with transforming ferrihydrite into magnetite (Frankel et al., 1983).

The magnetosome membrane is likely the place for mineral nucleation and acts as a boundary that creates a microenvironment inside the magnetosome which allow the reactions to occur that form magnetite (Gorby et al., 1988). This model has been termed a nanoreactor because the vesicles must control the pH, redox, and iron saturation in order to nucleate and grow the crystals (Schüler, 2008). Magnetosomes appear to represent a membrane bound organelle similar to what is found in eukaryotic cells. The current model proposed for magnetosome membrane formation, is that it is nearly pinched off of the plasma membrane, but not completely. This has been determined using electron cryotomography which shows that the magnetosome membrane is still contiguous with the plasma membrane (Komeili et al., 2004). This means that an empty magnetosome vesicle is produced prior to magnetite crystal formation.

Different species of magnetotactic bacteria have different crystal habits, implying that there are differences in the control of the crystal growth. Growth of anisotropic crystals such as elongated crystals would require control over the solution chemistry

(such as pH, iron saturation, and supply direction) inside the magnetosome as the crystal grows. The size of the magnetite crystal is dictated by physics and biology. In terms of the physics, the crystals of magnetite have maximum magnetization when they are between 30-120 nm; this gives the crystals a single domain (Figure 5). Single domain crystals have all the same dipoles lined up parallel so all the magnetization is uniform,

24 which allows the chain to act like a compass needle. Larger crystals have multiple domains (Figure 5), which would give opposite polarity in the same crystal while crystals smaller than 30 nm would not have a stable magnetization due to thermal fluctuations.

In terms of the biology, because of the specificity of the crystals in the magnetosome, it is likely under genetic control. However, the environment can also affect the magnetosome minerals (Bazylinski et al., 1995; Heywood et al., 1990; Lefèvre et al., 2011). It is probably not likely that the crystal size is simply constrained by the size of the magnetosome vesicle, but it is probably due to gene regulation. The growth stage of magnetosomes depends on their shape. Bullet-shaped magnetosomes have a 2- stage growth, beginning with isotropic growth, equal in all directions, then after it reaches

20 nm it grows along one axis (1 0 0) (Li et al., 2010). However, prismatic shaped magnetite begins as isometric cubo-octahedral growth, and then begins to grow in the (1

1 2) direction (Mann et al., 1987). The (1 0 0) direction is not energetically favorable and clearly requires biological control (Mann et al., 1987).

Because the crystals have a species specific shape and size, the biomineralization process is tightly controlled by the bacteria (Komeili et al., 2004). Many of these proteins that control the mineral formation are probably located in the membrane (Gorby et al., 1988; Okuda and Fukumori, 2001; Tanaka et al., 2006). It is possible to create bacterial mutants for specific genes, which have consequences on the characteristics of the magnetosome and minerals. For example, when a mutant for a specific gene (mamA) was examined, it was found that only half of the magnetosomes were produced (Komeili et al., 2004). However, the environment also plays a role in the metal component of the

25 minerals (Bazylinski et al., 1995). For example, when M. magnetotacticum is grown with other ions such as Ti, Cr, Co, Ni, Pb, Cu, and Hg, it was discovered that they replaced some of the iron (Gorby et al., 1988). In one case, researchers managed to use isolated

Mms6 proteins to produce cobalt ferrite (CoFe2O4) with slightly different magnetic properties of magnetite and the crystals were a specific size (Galloway et al., 2011;

Prozorov et al., 2007b). It was also discovered that environmental parameters, such as iron uptake by the cell, could affect the crystal properties in M. gryphiswaldense (Faivre et al., 2008).

Proteins

There are many proteins that are either embedded in the magnetosome membrane, tightly associated with magnetite crystals, or loosely associated with the membrane as determined by the affects of detergents and proteases on the magnetosome (Grunberg et al., 2004). Several of the proteins have been analyzed and their putative function identified (Table 2). Identified proteins include serine proteases (high temperature requirement, HtrA-like), iron transporters (cation diffusion facilitator CDF), actin-like proteins, proteins involved with protein-protein interactions (tetracopeptide repeat TPR), and many other unidentified proteins associated only with magnetotactic bacteria (Okuda and Fukumori, 2001; Scheffel et al., 2007; Tanaka et al., 2006). There are several proteins associated with the magnetosome membrane that have repeating amino acid motifs. These types of repetitive sequences in proteins have been implicated in biomineralization in the shells of mollusks and in cartilage and elastins (Bochicchio et al.,

2001; Sudo et al., 1997).

Table 2 Some of the known proteins of magnetotactic bacteria and their function.* Protein Amino acid Isoelectric Characteristics Function length point MamA 221 5.53 TPR-motifs Activation of

(Mam22) magnetosome

MamB 297 5.25 CDF transporter Iron transport

MamC 145 10.36 Control crystal growth

(Mam12) in a cumulative manner

MamD 314 10.6 Leu/Gly-rich motifs

(Mam7)

MamF 111 8.05

MamG 103 8.74 Leu/Gly-rich motifs

MamJ 466 3.80 Asp/Glu-rich repeats Magnetosome chain

formation

MamK 360 MreB-like Chain formation,

cytoskeletal filament

MamM 318 5.82 CDF transporter Inorganic ion transport

MamT 174 10.05 Heme binding

Mms6 157 9.90 Leu/Gly-rich motifs Iron binding, crystal

growth

*Based on (Richter et al., 2007; Scheffel et al., 2007; Scheffel and Schüler, 2007).

In my dissertation, I focused on one critical protein called Mms6, shown in Table

2. Mms6 is a protein specific to magnetotactic bacteria and was first found in M. magneticum AMB-1 (Arakaki et al., 2003). Mms stands for stands for magnetic-particle membrane specific. This protein is generated from a 399 bp gene sequence found in the

MAI, which codes for a 59 amino acid 12.5 kDa (12.7 kDa) protein (Arakaki et al., 2003;

Richter et al., 2007; Wang et al., 2012a). The Mms6 protein has been shown to be closely associated with the surface of magnetite as it forms inside the magnetosome (Amemiya et al., 2007). Mms6 is an acidic protein, which distinguish it from the similar sized Mms5 and Mms7, which are basic. Mms6 has a hydrophobic N-terminal region, which anchors the protein in the magnetosome membrane, and a hydrophilic C-terminal region that is tightly bound to the magnetite crystal. This protein has a leucine-glycine rich region

(LGLGLGLGAWGPXXLGXXGXAGA) similar to Mms5 and Mms7 and self- aggregates like other biomineralizing proteins. Interestingly, deletion of the mms6 gene resulted in a significant decrease of the amount of Mms5, Mms7, and Mms13 on the surfaces of bacterial magnetites (Galloway et al., 2011).

Normally AMB-1 cells produce cubo-octahedral shaped crystals that are approximately 60 nm in diameter and form a single chain with about 20 magnetite crystals. When taken out of the cell, in vitro studies showed that the Mms6 protein produced magnetite crystals of uniform size and shape (Amemiya et al., 2007; Arakaki et al., 2003; Prozorov et al., 2007a). But upon deletion of the mms6 gene (Δmms6), the cells produced smaller and misshapen magnetite particles (Tanaka et al., 2011). The Δmms6

28 cells also showed a decrease in the amount of Mms7 and Mms13 proteins. This suggests that the Mms6 protein is not involved in nucleating magnetite, but is used to produce uniform faces in the crystal (1 0 0) and (1 1 1). Other possible functions of Mms6 proteins in the cell are to stabilize the surface of the growing magnetite crystal, or perhaps it interacts with other proteins that act as a scaffold or enhance/dissolve the mineral surface during mineral growth. Outside the cell, Mms6 can bind other metals

2- such as cobalt to make cobalt ferrite (CoFe2O4), tellurium to make tellurite (TeO3 ), and lanthanide (La3+) (Galloway et al., 2011; Murat et al., 2012; Prozorov et al., 2007b;

Tanaka et al., 2010b).

A surprising result was observed when the mms6 gene was deleted form strain

MSR-1. In this case, the deletion did not have the same effect on the magnetite crystals as in AMB-1 (Lohße et al., 2011). When mms6 and mamGFDC were deleted together, the shape and alignment of the magnetite changed, but the size stayed roughly the same.

The authors believe this is because of the similarity of the proteins, specifically the LG rich region. This suggests that the proteins may have an additive effect rather than each protein having its own distinct function.

Dissertation research

In the first section of my research I discuss how to collect, isolate, and enrich

MTB from aquatic sediments. I do this by listing all the materials that are necessary to do the collection and all the steps necessary to perform the process. I used a wetland as the location because it was convenient to The Ohio State University and which would

29 most likely contain MTB. The samples I isolated were imaged using transmission electron microscopy (TEM).

In the next portion of my research I identified new magnetotactic bacteria from two unique environments. One environment in which I discovered MTB was Mickey

Hot Springs, a group of arsenic-rich thermal springs in southeastern Oregon. I also discovered MTB living in Pavilion Lake, a freshwater lake in south central British

Columbia Canada that produces organo-sedimentary structures of fossilized bacteria.

The magnetotactic bacteria from Mickey Hot Springs were from the phylum Nitrospirae.

The cells were rod to vibrioid shaped and contained bullet-shaped crystals of magnetite usually arranged in a single chain. The magnetotactic bacteria from Pavilion Lake were spirillum-shaped cells with a single chain of cuboctahedral-shaped crystals of magnetite from the α-proteobacteria phylum. These findings demonstrate two novel environments

(high arsenic hot spring and freshwater lake containing microbialites) where magnetotactic bacteria are found. The discovery of MTB in both of these mineralizing environments also opens up the possibility to search for and examine magnetofossils in these environments. This will help us understand magnetofossils in silicified and calcified microfossil deposits.

As noted above, Mms6 is a key protein involved in biomineralization of magnetite in MTB. The third part of my research investigated the sub-cellular localization of Mms6 proteins in single cells of Magnetospirillum magneticum AMB-1.

This was accomplished by labeling Mms6 proteins in the cells with either a gold conjugated antibody for analysis in the TEM or with a fluorophore conjugated antibody

30 for analysis with the confocal microscope. I found that the proteins are not located on the cell membrane or within the cytoplasm, but are only clustered on the magnetosome membrane. These findings constrain the lifecycle and the location of the Mms6 to the magnetosome membrane after the membrane invaginates from the cell membrane.

The last portion of my dissertation included the high-resolution analysis of two different types of magnetotactic bacteria: Magnetospirillum magneticum AMB-1 and

Magnetospirillum gryphiswaldense MSR-1 using atomic force microscopy (AFM) and

TEM. In these experiments I examined the ultrastructure of the cells and the magnetosomes from both species as well as Mms6 proteins and determined the advantages of both techniques to examining MTB. The main advantage that AFM has over the TEM is that you can examine cells under physiological conditions. This allows you to do direct experiments with the proteins and analyze how they interact with magnetite. Whereas in the TEM, the sample is not living, it has to be chemically fixed and dried out before it can be analyzed. Then the sample is subjected to an intense electron beam, which can cause damage to the sample. However, magnetosomes are inside the cell and the TEM allows the analysis of the magnetosomes inside the cells.

The combination of both techniques gives complementary information about the sample and combines separate lines of evidence, which provides for more robust results.

Chapter 2: Collecting Magnetotactic Bacteria from the Environment

Abstract

Magnetotactic bacteria (MTB) are aquatic microorganisms that were first notably described in 1975 from sediment samples collected in salt marshes of Massachusetts

(USA) (Blakemore, 1975). Since then MTB have been discovered in stratified water and sediment columns from all over the world (Blakemore, 1982). One feature common to all

MTB is that they contain magnetosomes, which are intracellular, membrane-bound

magnetic nanocrystals of magnetite (Fe3O4), greigite (FeSO4), or both (Bazylinski et al.,

1995; Lefèvre et al., 2011). In the Northern hemisphere, MTB are typically attracted to the south end of a bar magnet, while in the Southern hemisphere they are usually attracted to the north end of a magnet (Bazylinski et al., 1995; Simmons et al., 2006).

This property can be exploited when trying to isolate MTB from environmental samples.

In this chapter, I describe a relatively inexpensive and effective method for determining whether an environmental sample contains MTB. This protocol relies on the use of simple magnets. A clear plastic container can be used to collect sediment and water from a natural source, such as a freshwater pond. In the Northern hemisphere, the south end of a bar magnet is placed against the outside of the container just above the sediment at the sediment-water interface. After some time, the bacteria can be removed from the inside of the container near the magnet with a pipette and then enriched further by using a capillary racetrack and a magnet. In the racetrack, a sterile cotton plug is used 32 to separate magnetic vs. non-magnetic cells as the MTB swim through the cotton towards a magnet placed at the opposite end of the racetrack. Once enriched, the presence of

MTB can be confirmed by using the hanging drop method and a light microscope to observe MTB swimming in response to the north or south end of a bar magnet. Higher resolution observation can be made by depositing a drop of enriched MTB onto a copper grid and observed with transmission electron microscopy (TEM). Using this method, isolated MTB may be studied microscopically to determine characteristics such as swimming behavior, type and number of flagella, cell morphology, shape of the magnetic crystals, number of magnetosomes, number of magnetosome chains in each cell, composition of the crystals, and presence of intracellular vacuoles.

Collection

When deciding on a freshwater site to collect magnetotactic bacteria (MTB), it is often best to start with a pond or slow-moving stream that has a soft muddy sediment layer. For this isolation protocol, I collected a sample from a wetland near the campus of

The Ohio State University (OSU) in Columbus, Ohio (USA). The protocol described herein is applicable to just about any aquatic location. The materials used in this protocol can be found in Table 3.

Find a location where the depth of the water is between 10 and 100 cm. At such a location, you should collect the upper-most layer of sediment using a clear, screw-top container. Scoop the sediment and water into the container until it is filled with one-third to one-half sediment and the remaining volume with water (Figure 9A). Keep the container submerged until it is filled with water and then tightly cap the container with its

33 screw-top lid. It’s not necessary to mix the sediment. Wipe the outside of the container dry with a towel and then take the sample to your laboratory. It is not necessary to rush the sample back to your laboratory. The MTB should be viable for several weeks to months as long as you store the samples in a cool, dark location with the cap loosened.

Once the sample is in your laboratory, loosen the cap but leave it covering the container to reduce the amount of evaporation. Store the container at room temperature in a dark room, drawer, or completely cover the container with aluminum foil. Allow the sediment and fine particles to completely settle to the bottom of the container by leaving the sample undisturbed for several hours to several days. It is not necessary to mix the sediment, MTB prefer an undisturbed environment. The clear walls of the plastic container will allow you to confirm that the particles have settled to the bottom.

Depending on your sample, MTB can remain alive in the sample for many months.

Magnetotactic Bacteria Isolation

When you are ready to isolate the MTB, place magnets on the sides of the plastic container approximately 1 cm above the sediment-water interface (Figure 9). Be careful not to disturb the sediment in the bottom of the container. Place the south pole of a bar magnet on one side of the container and the north side of another bar magnet on the opposite side (Figure 9). Almost any magnet can be used, such as a magnetic stir bar or large refrigerator magnet. Anything can be used to support the magnets at the correct height above the sediment-water interface. I have found that resting the magnets on the top of a cardboard or plastic box is best, however, the magnets can also be taped to the

Table 3 List of specific reagents and equipment. Item Name Company Catalogue Comments number Glass slides Fisher Scientific S95933 Glass Pasteur pipets Fisher Scientific 13-678-6A O-ring Hardware store Cover slips Fisher Scientific 12-542B Bar magnet Fisher Scientific S95957 Container Any Any plastic or glass container that can hold at least 0.5 L and can be sealed Cotton Any Microscope with Zeiss A 60X dry lens 60X dry lens is not absolutely necessary, but this gives a high NA without using oil Diamond pen Fisher Scientific 08-675 0.22 µm filter Fisher Scientific 09-719C 1 mL syringe Fisher Scientific NC9788564 Microcentrifuge Fisher Scientific 02-681-320 tubes Formvar/Carbon Ted Pella Inc. 01800 200 mesh, copper grids Uranyl acetate Ted Pella Inc. 19481 Tecnai Spirit TEM FEI Tecnai F20 S/TEM FEI

35 outside of the plastic container. Wait 30 minutes to several hours for the bacteria to swim to the magnet.

Use a sterile pipette to carefully collect the water from inside the container

(Figure 9) near the position of the south pole of the bar magnet (for samples collected in the Northern hemisphere). This water should contain the MTB that have been attracted to the south pole of the bar magnet. Next, a capillary racetrack should be used to further enrich the MTB.

Magnetotactic Bacteria Racetrack

In order to enrich your sample with MTB, a capillary racetrack is necessary

(Figure 9B). These need to be made prior to isolating the cells from the clear-plastic container. Use a 5.75 inch (146 mm) glass Pasteur pipette to make a racetrack. Use a diamond pen or file to cut off the top of the pipette, the length of the pipette is not crucial, but it should be able to contain approximately 1-2 mL of water. Next, use a Bunsen burner to melt the tip so that it becomes sealed (Figure 9B). The resulting pipette should have an open end and a sealed end.

Make several of these racetracks and then autoclave. Additionally, you will need to autoclave cotton and several long metal needles. Add filtered sample water to the tip of the racetrack by collecting liquid from the top of the sample shown in Figure 9A, to an autoclaved racetrack using a long metal needle attached to a filtered syringe. The pore size of the filter should be 0.22 µm to eliminate debris and contaminants from the water.

It is important to be absolutely sure that there are no air bubbles in the glass capillary.

Plug the neck of the sealed half of the racetrack with sterile cotton (Figure 9B) about 0.5cm from the sealed tip. Use the metal needle to push the cotton towards the sealed end of the racetrack so it is 0.5 to 1 cm away from the sealed tip (Figure 9C).

Using a sterile pipette, remove MTB-containing fluid (as described in the previous section) from the sample container and add it to the sample reservoir (open end) of a prepared MTB racetrack (Figure 9B).

Magnetotactic Bacteria Enrichment

Once the racetrack is filled with sample fluid, lay it on its side on a horizontal surface (e.g., a benchtop) and place the south pole of a bar magnet (in the Northern hemisphere) next to the sealed tip of the racetrack (Figure 9B & C). Wait 5 to 30 minutes for the MTB to migrate through the cotton. Then you should collect the fluid near the tip of the racetrack. Waiting too long can introduce contaminants, such as other motile bacteria, to the tip of the capillary. Optionally, you could use a light microscope to view the tip of the racetrack and watch the MTB collect at the racetrack’s tip. This will allow you to determine how long it takes the MTB to migrate through the cotton plug.

Then to remove the enriched MTB, gently use the diamond knife to make a little scratch near the cotton plug and snap off the end of the racetrack. Use a 1 mL syringe with a narrow needle (25 or 27 gauge) to remove the fluid from the tip of the racetrack.

This liquid sample should now contain the enriched MTB.

Magnetotactic Bacteria Observation by Light Microscopy

Place a drop (10-20 µL) of the enriched MTB sample onto a coverslip. Quickly flip the coverslip over so the drop is now facing down and hanging from the coverslip.

Place the coverslip onto an o-ring that is resting on a glass slide. The o-ring should have a slightly smaller diameter then the coverslip (about 1 cm; Figure 10). Place this hanging drop onto a light microscope stage and focus on one edge of the drop. A 60X dry objective works very well because most have a high numerical aperture, but do not require oil, which is difficult to use for the hanging drop method (Figure 10). Place the south end of a bar magnet close to the hanging drop and MTB will begin to migrate towards the edge of the drop closest to the magnet (Figure 11). Within a few minutes many MTB should be at the edge of the drop (Figure 11). You are able to prove that the bacteria are magnetic by reversing the pole of the magnet and then observe the bacteria swim in the opposite direction.

Magnetotactic bacteria Observation by Transmission Electron Microscopy (TEM)

Place a drop (∼20 µL) of the enriched MTB onto a flat surface, such as parafilm, and place a copper grid on the drop, and allow the bacteria to adsorb onto the grid for about 10-20 minutes. Wick off excess water with a piece of clean filter paper.

Optionally, the grid can be negatively stained with 2% uranyl acetate, 2% phosphotungstic acid pH 7.2, or 2.5% sodium molybdate (Balkwill et al., 1980; Moench and Konetzka, 1978; Rodgers et al., 1990). This is done by placing the copper grid onto a drop of stain immediately after incubating the grid with the enriched MTB. Incubate the grid with the negative stain; the times will vary depending on the stain used, and then

Figure 11 Bright field microscope image of MTB swimming (thin long arrows) and gathering at the edge of the hanging drop (short arrows) which is next to the south pole of a bar magnet.

wick off the fluid with a piece of clean filter paper. Observe the MTB using transmission electron microscopy (TEM) (Figure 12). For the work described here MTB were adsorbed to Formvar stabilized and carbon coated 200 mesh copper grids (Ted Pella

#01800). The grids were placed with the carbon side down on a drop of cell suspension for up to 10 minutes, then immediately washed one time by placing the grid on a drop of water for 30 seconds. For staining, the grids were placed on a drop of 2% uranyl acetate

(Ted Pella #19481) for 30 seconds to 5 minutes and then dried completely using a piece of filter paper. The grids were analyzed by TEM using either an FEI Tecnai Spirit at

80kV.

Discussion

Magnetotactic bacteria are not necessarily found in every aquatic environment but when they do occur, there can be as many as 100 – 1,000 cells per milliliter (Blakemore,

1982; Moench and Konetzka, 1978). In order to observe the MTB using optical microscopy, you will need approximately 50 bacteria/mL in your sample (Moench and

Konetzka, 1978). If there are none or fewer MTB in your sample, you will need to either select a new environmental site to collect your samples or try an enrichment technique like the one described here. First, a magnet is used to isolate or concentrate magnetotactic bacteria (MTB) contained in environmental samples (Figure 9A). Then a capillary racetrack (Figure 9B) can be used to attract MTB through a cotton plug where they can be separated from non-magnetotactic microorganisms also contained within the environmental sample.

Figure 12 Transmission electron microscope images of various MTB isolated from the wetland near the Olentangy River in Columbus, Ohio. There were several different morphotypes and they all contain crystals (white arrows) and some have inclusions (white arrowheads). Scale bar for each image is 500 nm.

When you bring your environmental sample back to the laboratory, it may be beneficial to wait several days or weeks for the sample to acclimate to laboratory conditions before trying to isolate the MTB using a bar magnet. This acclimation period will allow the bacterial community to mature and repopulate the container leading to higher concentrations of MTB. Another simple technique that often produces more concentrated MTB samples is to leave the bar magnet on the side of the sample container

(Figure 9A) for a longer period of time (e.g., overnight). This should allow the MTB more time to migrate to the magnet. Another technique that may be useful, is to use several racetracks (Figure 9B) at once and then combine the MTB from each racetrack into one sample. Lastly, you should try collecting more sediment from the environment using a large plastic tub (Moench and Konetzka, 1978). This is especially useful if large numbers of unculturable MTB are needed.

If you believe there is a problem with a racetrack or if there are too many contaminating microorganisms (i.e., non-MTB) in your enriched sample, you can place the racetrack under a light microscope to observe the MTB as they swim through the cotton plug and into the tip. This will allow you to determine if contaminating microorganisms are also coming through the cotton plug and when to stop the enrichment process.

I should mention that there are more sophisticated ways to isolate MTB, but these methods require the use of more specialized equipment. One example involves the use of a magnetic coil, instead of a bar magnet, and customized glass vessels to isolate MTB

44 from freshwater sediments (Jogler et al., 2009b; Lins et al., 2003). The protocol described here represents an inexpensive and effective method for determining whether an environmental site contains MTB. This isolation and enrichment protocol is straightforward enough that microbiology students can master and easily “fine-tune” so that higher yields of MTB can be achieved. Once the MTB have been isolated, other analyses such as fluorescence in-situ hybridization, 16S rRNA sequencing for community analysis, energy dispersive spectroscopy (EDS), TEM, optical microscopy, and magnetic measurements can be conducted on the MTB (Li et al., 2010; Lin et al., 2012).

Chapter 3: Magnetotactic Bacteria from Mickey Hot Springs and Pavilion Lake

Abstract

Magnetotactic bacteria (MTB) grow in aquatic sediments all over the world but their geobiological role in their habit as well as their significance in the fossil record is poorly understood. To better understand their complete ecological distribution, magnetotactic bacteria were isolated from two unique environments. One location was

Mickey Hot Springs, an arsenic-rich hot springs in southeastern Oregon, a site that is known to have microfossils in sinter deposits. The other location was Pavilion Lake, a unique freshwater lake in British Columbia, which contains microbialites. Using 16S rDNA analysis, the MTB from Mickey Hot Springs were found to belong to the

Nitrospirae phylum. These cells were rod to vibrioid shape and contained one or more chains of magnetosomes as determined by transmission electron microscopy (TEM). The magnetosomes contained bullet-shaped crystals of magnetite approximately 84 nm long and 39 nm wide. This is the first instance of MTB from an arsenic-rich environment.

The magnetotactic bacteria isolated from Pavilion Lake were Alphaproteobacteria as determined by 16S rDNA analysis. TEM analysis revealed that these cells were spirillum-shaped and contained a single chain of cuboctahedral shaped magnetite crystals that were 40 nm in diameter. The discovery of MTB in these two sites offers the opportunity for others to study these locations in order to better understand the diversity

46 of magnetotactic bacterial habitats, particularly their geobiological function in thermal springs, as well as the search for magnetofossils in the recent rock record.

Introduction

Magnetotactic bacteria (MTB) are found in a variety of aquatic sediments such as marine environments, freshwater lakes and rivers, hot springs, and brackish waters all over the world (Amann et al., 2007; Bazylinski et al., 1995; Bazylinski et al., 2000;

Blakemore, 1975; Lefèvre et al., 2010; Lefèvre et al., 2011; Lin et al., 2009; Moench and

Konetzka, 1978; Spring et al., 1994). The unique feature common to all MTB is their ability to synthesize intracellular membrane-bound crystals of single domain magnetite

(Fe3O4), greigite (Fe3S4), or both (Bazylinski et al., 1995; Bazylinski et al., 1993;

Heywood et al., 1990; Lefèvre et al., 2011; Lins et al., 2007). MTB exist in a chemically stratified water or sediment column at the oxic-anoxic interface. The magnetosomes provide a torque on the cells that passively aligns them with the Earth’s geomagnetic field. This in turn reduces their navigational route from three dimensions to one dimension; shortening the time it takes for cells to navigate to their preferred habitat, the oxic-anoxic interface at the bottom of water bodies (Bazylinski et al., 1995; Frankel et al.,

2007; Frankel, 1984; Mann et al., 1988b).

Magnetite from magnetotactic bacteria has a very specific size and well-defined consistent morphology and is chemically pure (Devouard et al., 1998; Faivre et al., 2008).

Such minerals are preserved in the rock record as “magnetofossils”, which have been found in Mesozoic rocks, and may extend back as far as the pre-Cambrian (Chang and

Kirschvink, 1989; Kirschvink and Chang, 1984; Kopp and Kirschvink, 2008; Kopp et al.,

2007). Differentiating between magnetite made by MTB and abiogenic magnetite in the fossil record is not straightforward, but there are distinctions between the two, such as size, uniformity, chemical purity, and sometimes the magnetosome chain is preserved

(Arató et al., 2005; Benzerara and Menguy, 2009; Benzerara et al., 2011; Chang and

Kirschvink, 1989; Devouard et al., 1998; Kobayashi et al., 2006; Kopp et al., 2006; Kopp and Kirschvink, 2008; Petersen et al., 1986; Thomas-Keprta et al., 2000; Thomas-Keprta et al., 2002). The two sites that I investigated are known to contain fossilized bacteria, which make them potential sites to search for magnetofossils.

As noted above, MTB is not a phylogenetic grouping. Rather, it is a name given to bacteria that share common characteristics, primarily the presence of magnetosomes.

MTB are polyphyletic and have been found to exist in four groups: Alphaproteobacteia,

Deltaproteobacteria, Gammaproteobacteria, and Nitrospirae (Amann et al., 2007;

Burgess et al., 1993; Kawaguchi et al., 1995; Sakaguchi et al., 2002; Spring et al., 1994).

MTB are common in freshwater and marine environments, but recently they have been found in “extreme” environments such as thermal springs and brackish environments

(Lefèvre et al., 2010; Lefèvre et al., 2011). Identifying different MTB is difficult because growing and culturing MTB in the lab remains challenging because of their stringent growth requirements. However, genetic and molecular based phylogenetic approaches have made it much easier to identify MTB directly from environmental samples (Burgess et al., 1993; Flies et al., 2005b; Jogler et al., 2010; Lin et al., 2009; Simmons et al., 2004;

Spring et al., 1994; Spring et al., 1998). This has also permitted the study of the diversity

48 of MTB through time, which has shown that the community of MTB can change within weeks (Lin and Pan, 2010; Simmons et al., 2004).

Because MTB are so widespread, it has been proposed that MTB play a role in iron phosphorus cycles (Keim et al., 2001; Simmons and Edwards, 2007a; Simmons et al., 2007). It has been speculated that different chemicals in the environment such as nitrate, iron, oxygen, and sulfur can affect the types of MTB that exist in a specific location. But, the phylogenetic diversity of MTB within various habitats is largely unexplored (Keim et al., 2007a; Lin et al., 2009; Schüler et al., 1999; Spring and

Bazylinski, 2006).

This study investigates uncultured MTB from two different unique locations: (i) an arsenic rich thermal spring, which possess microbial fossils in sinter, and (ii) a freshwater lake that forms microbialites. The phylogeny of the MTB at both locations was analyzed. The MTB isolated from the arsenic-rich Mickey Hot Springs belonged to the Nitrospirae phylum; whereas the MTB from Pavilion Lake belonged to the

Alphaproteobacteria. Transmission electron microscopy was used to determine the morphology of the cells, size and shape of magnetosomes, while scanning transmission electron microscopy was used to determine the chemical composition of the magnetosomes from both locations. This is the first time MTB have been described from an arsenic-rich environment and from a microbialite-forming environment. This discovery will likely be of great value to others interested in the potential of finding life forms on other planets or the earliest forms of life on Earth, as Mickey Hot Springs is

49 known to contain microbial fossils in sinter and Pavilion Lake contains microbial fossils in freshwater microbialites, which may represent some of the earliest life forms on Earth.

Methods

Sample collection

Sediment samples were collected a two different geographic locations in western

North America (Figure 13). These samples were collected in the summer of 2011 (refer to Appendix A for images of the sample sites). Sediment samples were obtained by scraping the sediment approximately 20-45 cm below the surface of the water with a 1- liter container. The containers contained one-half to three-quarters sediment and the remainder was filled with fluid from the site. The bottles were sealed and brought back to the lab for analysis. At the lab the caps were loosened, but still remained on the container and then kept in a dark at room temperature for several weeks.

At the time of sediment collection, a water sample was also obtained along with measurements of the water temperature and pH. The water was filtered through a

0.22µm filter, acidified in 5% nitric acid (v/v), and sealed. The water samples were analyzed at The Ohio State University Trace Element Research Laboratory using a

Perkin-Elmer Optima 3000DV inductively coupled plasma optical emission spectrometer. Samples were analyzed for boron (B), calcium (Ca), potassium (K), magnesium (Mg), sodium (Na), arsenic (As), aluminum (Al), copper (Cu), phosphorus

(P), iron (Fe), and silicon (Si).

Magnetotactic Bacteria Enrichment

Magnetotactic bacteria were isolated from the sediment following the procedure explained in Chapter 2. Briefly, the south end of a magnet was placed on the outside of the container just above the sediment-water interface, and then the north end of a magnet was placed on the opposite side of the container. After 1 hour the water around the south end of the magnet was extracted with a pipette and placed in a racetrack with a cotton plug at the sealed end (Wolfe et al., 1987). This was repeated 12 times for each sample.

A magnet was placed at the sealed end, and the magnetotactic cells were allowed to swim through the cotton barrier for approximately 30 minutes. The racetrack was taken away from the magnet, the tip snapped off, and the bacteria removed with a syringe. A total of

500-1000 µL were collected from each sample. The hanging drop method was used to confirm the presence of MTB in the sample (see Chapter 2 for description).

Transmission Electron Microscopy

An aliquot of the enriched cells were placed on a 200 mesh copper grid coated with carbon and formvar (Ted Pella) and analyzed in an FEI Tecnai G2 Spirit transmission electron microscope or an FEI Tecnai F20 scanning transmission electron microscope using high angle annular dark field (HAADF). The accelerating voltage of the G2 Spirit was 80 keV, spot size 2, and using the number 2 objective aperture, and images were collected using a Gatan camera and AMT Image Capture software. For the

Tecnai F20, an accelerating voltage of 200keV was used in the HAADF mode. Crystals inside the cells were analyzed using the energy-dispersive X-ray spectrometer on the F20, only a 100 µm condenser aperture was used and the specimen was tilted 5° towards the detector which was an EDAX detector with an ultrathin Moxtek AP3.3 window with an

51 elevation angle of 20°. The size of the cells and the magnetite crystals were analyzed using FIJI software.

Phylogenetic Analysis

Approximately 500µL of sample collected from the racetrack was used to obtain

DNA for phylogenetic analysis. DNA was obtained from the cells by homogenizing the cells and resuspending them in RLT buffer (Qiagen) with β–mercaptoethanol. A Qiagen

DNA kit was used to isolate the DNA. Fragments of the DNA were amplified using PCR using 28F and 519R primers (TTTGATCNTGGCTCAG GWNTTACNGCGGCKGCTG) and a Qiagen hotstart taq mastermix. The DNA was denatured at 95°C for 5 minutes, then 35 cycles at 94°C for 30 seconds, 54°C for 45 seconds, 72°C for 60 seconds, and final extension at 72°C for 10 minutes. The sequence and analysis was done at the

Research and Testing Laboratory, Lubbock, Texas.

The sequences were aligned using the default settings in MUSCLE (Edgar, 2004a;

Edgar, 2004b). The sequences were compared with reference sequences from NCBI. The phylogenetic tree was generated using these sequences with the sum of branch length =

1.92788276 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree (Sneath and Sokal,

1973). The evolutionary distances were computed using the Maximum Composite

Likelihood method and are in the units of the number of base substitutions per site. The analysis involved 24 nucleotide sequences (Tamura et al., 2004). All ambiguous positions were removed for each sequence pair. There were a total of 1793 positions in the final dataset. Evolutionary analyses were conducted in MEGA5 (Tamura et al., 2011).

Results

Sample locations

Samples were collected from two different locations in western North America

(Figure 13). These locations were selected because they represented unique habitats (e.g. high arsenic) where magnetotactic bacteria have never been found. Therefore, I was interested in the diversity of habitats of magnetotactic bacteria.

Mickey Hot Springs is located at the northern end of the Alvord Basin in southeastern Oregon at an elevation of approximately 1200 meters. This area possesses several active hot springs generated from fluids that ascend through deep faults which discharge at the surface (St. John, 1993). Consequently, the fluid contains dissolved ions, one of which is arsenic (Table 4) (Koski and Wood, 2004; St. John, 1993; Wood and

Shannon, 2003). Mickey contains a very high concentration of arsenic, almost 1 mg/L.

For reference, the U.S. Environmental Protection Agency has set the arsenic standard for drinking water at 0.010 µg/L. Water samples from Mickey Hot Springs also contain high amounts of silicon and sodium 76 and 470 mg/L, respectively. The spring where the samples were collected had a temperature of 47°C and the pH was 8.0.

Pavilion Lake is located in south-central British Columbia and is bounded by

Marble Canyon and the lake is fed by groundwater that runs through limestone. This lake contains unique underwater calcareous organo-sedimentary features called microbialites, which host a variety of microbial communities that can become fossilized (Brady et al.,

2010; Ferris et al., 1997; Laval et al., 2000; Lim et al., 2009). The temperature and pH of the lake was not measured at the time of collection, but the results from other work

Figure 13 Map of Western North America. Sample sites are designated with a star. Pavilion Lake is in British Columbia, Canada, and Mickey Hot Springs is located in southeastern Oregon.

Table 4 Geochemistry of Mickey Hot Springs and Pavilion Lake. The units for the ions are mg/L.* Temp. °C pH B Ca K Mg Na Si As Al Cu Fe P Mickey Hot 47 8.0 8.2 0.3 41 <0.2 470 76 0.96 0.13 0.005 0.03 0.06 Springs Pavilion ~20 8.3 - 39.6 3.1 19.8 7.9 4.95 - - - - - Lake *The data for Pavilion Lake is from Lim et al., 2009.

55 reports the lake as having a summer temperature of 20°C and the pH is 8.3 (Lim et al.,

2009).

Mickey Hot Springs

One aliquot of sample from the sediment was analyzed for magnetotactic bacteria by lysing all the cells, extracting the DNA, probing for the 16S ribosomal RNA gene, sequencing it, then comparing it to known sequences. From the sample, there were seventeen different 16S rRNA gene sequences, named Oestreicher-1

HH5G4AX03*****. These sequences were compared with sequences in the NCBI nucleotide database to determine how similar they are to other sequences. Based on the percentage of similarities, a phylogenetic tree was constructed to show how the sequences compared to each and to other magnetotactic bacteria. An

Alphaproteobacterium (Genbank accession number GU271691) was used as a reference point to root the tree. The Mickey Hot Springs samples were all found to affiliate with the Nitrospirae phylum (Figure 14). The horizontal lines refer to the sequence of one organism of interest and the length of the line represents how divergent each sequence is to each other.

For the sample from Mickey Hot Springs, TEM and EDX were used to examine the morphology of the enriched MTB cells and magnetosomes, as well as the composition of the magnetosome crystals. The sample from Mickey Hot Springs contained one type of cell that had a rod to vibrioid morphology. The average length and width of each cell was 2.2 (± 0.6) and 0.62 (± 0.1) µm, respectively (n = 37). Each cell had one flagellum (Figure 15). The magnetosome chain was usually single, but

Figure 14 Phylogenetic tree from the 16S rRNA gene sequences of the Mickey Hot Springs samples, labeled Oestreicher-1:HH5G4AX03*****. Maximum composite likelihood method was used to construct the tree. Each line represents how closely the sequences are related; the scale bar at the bottom shows the percentage divergence. The dominant group of magnetotactic bacteria was from the Nitrospirae phylum. The Genbank accession numbers from the non-Mickey Hot Springs samples are at the end of each branch and given as a single or double letter followed by 5 or 6 digits. An Alphaproteobacteria was used to root the tree.

sometimes two chains would partially overlap. In three cells, I discovered two distinct chains. Each chain averaged 16 (± 5) magnetosomes per chain (n = 50). The EDX analysis established that the magnetosomes were composed of iron and oxygen, which is indicative of magnetite (Figure 15B) (refer to Appendix A for the EDX spectrum of the background analysis of the cells). The crystals were bullet-shaped and they were either aligned with the sharp end of one crystal next to the flat end of another, but some crystals were aligned with the flat ends against each other. The crystals had an average length of

84 (± 17) nm (range 40 to 136 nm) and an average width of 39 (± 8) nm (range 17 to 63 nm) (n = 177) (Figure 16) (refer to Appendix A for all the measurement data). This indicates single domain magnetite (Figure 5).

The shape factor for the magnetite averaged 0.48 (± 0.1) (range 0.22 to 0.81), which falls within the single domain size range (Figure 17). The crystals range from elongated, when the shape factor is low, to rounded as the shape factor approaches zero.

The shape factor is the crystal width divided by the crystal length. This is important because as the shape factor approaches 1, the range of single domain magnets decreases.

Figure 15 (A) Vibrioid magnetotactic bacterium from Mickey Hot Springs, Oregon USA. The cells averaged 2.2 µm long and 0.62 µm wide. The magnetosomes were 84 nm long and 39 nm wide. The chain of magnetosomes contains bullet-shaped crystals (white arrow). (B) The magnetite was analyzed using energy-dispersive spectroscopy (EDX) which showed that the crystals are composed of iron and oxygen, which is indicative of magnetite. Copper peaks are from the copper support grid. A single flagellum comes out of one side of the cell (black arrow). Scale bar is 500 nm.

Figure 16 Length versus the width and shape factor (crystal width/crystal length) diagrams for the Mickey Hot Springs sample. The dashed line in A represents width equal to length. The areas labeled in B represents the approximate area of single domain magnetite based on Butler and Banerjee, 1975.

Pavilion Lake

An aliquot containing enriched magnetotactic bacteria from the Pavilion Lake sample was analyzed by sequencing the 16S rDNA from all the bacteria in the sample.

The sample contained twenty-nine different 16S rDNA sequences, named Oestreicher-1

HH5G4AX03*****. These sequences were compared with sequences in the NCBI nucleotide database in order to construct a phylogenetic tree (Figure 18). The horizontal lines in Figure 18 refer to the sequence of one organism of interest and the length of the line represents how divergent each sequence is to each other. An Alphaproteobacterium

(Genbank accession number DQ482050) was used to root the tree.

From this analysis, it was determined that the MTB from the Pavilion Lake sample were very different than those from Mickey Hot Springs. All species from

Pavilion Lake were found to affiliate with either the Alphaproteobacteria or the

Gammaproteobacteria (Figure 18). Most of the sequences aligned with the

Alphaproteobacteria.

TEM was used to image individual cells from the Pavilion Lake sample. These

MTB were found to be spirillum-shaped, 2.9 (± 0.6) µm long, 0.34 (± 0.02) µm wide (n

= 7), containing a single flagellum (Figure 19). The magnetosomes contained crystals of iron and oxygen indicating magnetite (Figure 19B) (refer to Appendix A for the EDX spectrum of the background analysis of the cells). The crystals were 47 (± 4) nm long

Figure 18 Phylogenetic tree from twenty-nine 16S rDNA sequences of the Pavilion Lake samples, labeled Oestreicher-1:HH5G4AX03*****. Maximum composite likelihood method was used to construct the tree. Each line represents how closely the sequences are related; the scale bar at the bottom shows the percentage divergence. The dominant group of magnetotactic bacteria was from the Alphaproteobacteria phylum, but also from the Gammaproteobacteria; several of the sequences were similar to Magnetospirillum. The Genbank accession numbers from the non-Pavilion Lake samples are at the end of each branch and given as a single or double letter followed by 5 or 6 digits. An Alphaproteobacterium, Magnetospirillum gryphiswaldse, was used to root the tree.

and 44 (± 5) nm wide (range 37-62 nm and 33-56 nm, respectively), (n = 155) with an average of 19 crystals per cell (n = 7) (Figure 20). The magnetite crystals were nearly the same in length and width and most had a shape factor around 0.9 (± 0.05) indicating that the magnetite crystals are single domain magnetite crystals (Figure 21) (refer to

Appendix A for all the measurement data).

Discussion

This study has expanded the range of habitats for MTB. I have discovered that

MTB live in hot springs rich in arsenic as well as freshwater environments that are conducive to the formation of microbialites. 16S rDNA analysis allowed me to determine the phylogeny of the MTB at each site, while TEM and EDX analysis allowed me to characterize the size and shape of individual magnetotactic bacteria and their associated magnetosomes as they (i.e. the mineral crystals) were in place with cells.

The phylogenetic analysis and morphology data of the MTB from one of the thermal springs at Mickey Hot Springs contained rod to vibrio-shaped MTB that were 2.2 µm long by 0.6 µm wide. Most of these cells contained a single chain of magnetosomes.

Each magnetosome had approximately 16 bullet-shaped crystals that were 84 nm by 39 nm. All the operational taxonomic units (OTUs) obtained in the sequence analysis from

Mickey Hot Springs belonged to the Nitrospirae phylum. The association between microbial phylum and shape of the magnetite is consistent with other studies, as all known Nitrospirae MTB have bullet-shaped magnetite crystals just like the ones described herein for Mickey Hot Springs (Lefèvre et al., 2010; Lin et al., 2012).

Figure 19 (A) Spirillum-shaped magnetotactic bacterium from Pavilion Lake, British Columbia, Canada. The cells averaged 2.9 µm long and 0.3 µm wide with a single chain of magnetosomes (white arrow) and a single flagellum (black arrow). Scale bar is 500 nm. (B) The EDX showed the crystals were made of iron and oxygen (the copper peak is due to the copper support grid) indicative of magnetite. The crystals averaged 47 nm x 44 nm.

Figure 20 (A) Graph showing the length versus the width of the magnetite crystals from the Pavilion Lake sample illustrating that the length and width are nearly equal. (B) The length versus the shape factor of the magnetosomes showing that the crystals fall within the single domain size range. The figure is based on Butler and Banerjee, 1975. 66

Nitrospirae have been found at hot springs in Nevada (Great Boiling Springs), which is in the same geographic region as Mickey Hot Springs (Lefèvre et al., 2010;

Nash, 2008). The MTB at the Great Boiling Springs were able to survive up to temperatures of 63°C, which is higher than the temperature measured by me at Mickey

Hot Springs (Table 4). The MTB from Great Boiling Springs contained bullet-shaped magnetite with similar sized cells and magnetosomes as the Mickey Hot Springs specimens described herein.

There was one other major differences between the two hydrothermal sites.

Mickey Hot Springs contains MTB that thrive at high levels of arsenic (~1 mg/L). This is significant because arsenic is nearly universally toxic to organisms (Cervantes et al.,

1994; Hindmarsh et al., 1986; Ratnaike, 2003). The EDS data showed no arsenic in the magnetite (Figure 15B). It is unlikely that the arsenic is being removed before reaching the bacteria because it remains in the fluid (Koski and Wood, 2004). But it is possible that the arsenic is not reaching the bacteria in the sediment. The environment does not seem to be affecting the production of magnetite in the bacteria. The MTB could be living within a community of other organisms that could be taking care of arsenic.

Arsenic metabolizing bacteria have been found from Mickey Hot Springs. These microbes produces realgar (AsS) which has the potential to be used to remediate arsenic waste and as a biomarker (Ledbetter et al., 2007). 16S rRNA is good for diversity assessment, but does not explain ecophysiologic processes or reveal anything about geochemical cycling in the environment (Amann et al., 2007; Simmons and Edwards,

2007a). MTB have been found to have the potential to remediate tellurium contaminated sites (Tanaka et al., 2010b). Magnetite has been shown to be used to remove arsenic (An et al., 2011). Therefore, the ability for MTB to tolerate high arsenic concentrations may be useful because they have the capacity to be used in remediating arsenic contaminated sites.

The MTB from Pavilion Lake contained spirillum-shaped bacteria from the

Alphaproteobacteria that were closely related to Magnetospirillum. The bacteria were

2.9 µm long and 0.3 µm wide and contained a single chain of magnetosomes with an average of 19 cuboidal-shaped magnetite crystals that were 40 nm across. This is similar to other freshwater Magnetosprillum bacteria which have single chains of 15 or more magnetosomes with cuboctahedral shaped magnetite with sizes of ~40-50 nm

(Baumgartner and Faivre, 2011; Faivre and Schüler, 2008; Isambert et al., 2007).

Magnetospirillum are common to many freshwater habitats and are one of the dominant forms of Alphaproteobacteria (Amann et al., 2007).

Environmental parameters have been shown to affect the crystals of magnetosomes (Faivre et al., 2008; Li and Pan, 2012). But the unique environmental parameters do not seem to affect the chemical composition or crystal morphology of the magnetite crystals in the magnetosomes of the magnetotactic bacteria from Pavilion Lake or Mickey Hot Springs. Based on the morphology and phylogenetic analysis, each unique environment did seem to select for one type of MTB. No Nitrospirae were found in Pavilion Lake and no Alphaproteobacteria were found in Mickey Hot Springs. It is possible that there was selection based on using the racetrack method to enrich the MTB

(Lin et al., 2008). But, this same racetrack method was used on samples from both

Mickey Hot Springs and Pavilion Lake. It may also be important to note that the bacteria were not enriched from the samples for several weeks after collection, which could have selected for one dominant type of MTB during the incubation in the lab (Flies et al.,

2005a; Vali et al., 1987)

Locating MTB in mineralizing environments such as Pavilion Lake and Mickey

Hot Springs has important implications for finding microfossils. Nitrospirae have been found throughout the world, they are the deepest branching phylum of magnetotactic bacteria, and they synthesize unique bullet-shaped magnetite (Abreu et al., 2007; Jogler et al., 2011; Lefèvre et al., 2011a; Lin et al., 2012). Magnetite is known to be a robust biomarker in the form of magnetofossils (Chang and Kirschvink, 1989; Kirschvink and

Chang, 1984; Kobayashi et al., 2006; Kopp et al., 2006; Kopp and Kirschvink, 2008).

MTB are known to occupy specific niches in their habitat (Amann et al., 2007; Simmons and Edwards, 2007a; Spring and Bazylinski, 2006; Spring and Schleifer, 1995). If magnetofossils were found in the sinter deposits or microbialites from either of these sites, they would facilitate identifying the paleoenvironment (Oestreicher, 2004).

At Mickey Hot Springs, the fluid is saturated with respect to silica giving the potential for mineralization and the fact that the MTB found here produced elongated anisotropic crystals is beneficial because those are a robust biomarker (Arató et al., 2005;

Lefèvre et al., 2011b). Additionally, there are fossilized bacteria in the older sinter deposits in the Mickey Hot Springs group (Oestreicher, 2004). Furthermore, looking for magnetofossils in silica rich fluid could have implications in astrobiology because

70 magnetite crystals have been found in silica rich inclusions in ALH84001 (Thomas-

Keprta et al., 2009).

Pavilion Lake is a unique environment because it has microbialites that contain microfossils. To date, no magnetofossils have been found in these microbialites (Laval et al., 2000). Here, I show that MTB are known to exist at Pavilion Lake. There is now the potential to find evidence of these MTB in the fossil record. Microbialites in the form of thrombolites have been dated back as far as the Proterozoic and magnetofossils have been dated back to the Cretaceous to late Archean (Kennard and James, 1986; Kopp and

Kirschvink, 2008). This paper describes two unique environments where MTB are found which expand our database of MTB habits and it opens up the possibility of searching for magnetofossils and biomarkers in sinter deposits and calcite-rich microbialites.

Chapter 4: Localization of Mms6 Proteins

Abstract

Magnetotactic bacteria (MTB) are capable of mineralizing intracellular crystals of magnetite (Fe3O4) and or greigite (Fe3S4) inside membrane bound vesicles, called magnetosomes, using a controlled protein-mediated biomineralization process. The biomineralization process is precisely controlled through a series of protein-mediated actions that create a morphologically defined crystal that is species specific. Mimicking this engineering process on a large scale would be beneficial in areas such as medicine, electronics, and pharmaceuticals. The protein Mms6 (the premature peptide was reported to be 12.5 kDa and the mature peptide was 6 kDa, and the His-tagged recombinant protein has been reported to be 10.2 kDa) is very important because it was implicated in the biomineralization process in Magnetospirillum magneticum AMB-1 by controlling the size and shape of magnetite crystals. Furthermore, it has been shown to incorporate other cations into the crystal structure of magnetite. To better understand its role in the biomineralization process, I examined the localization of Mms6 inside cells of M. magneticum AMB-1 using immunogold antibody labeling and fluorescent antibody labeling and then examining the samples using the transmission electron microscope and confocal microscopy. AMB-1 cells typically contain a broken chain of 15 magnetosomes each containing a crystal of cuboctahedral magnetite approximately 50 nm in diameter. I discovered that the protein was only found localized in the magnetosome membrane and 72 nowhere else in the cell. However, the protein was not localized on every magnetosome even though magnetite was present inside the membrane. These two findings suggests that the protein has a limited lifecycle and is not present before the magnetosome membrane is established and also that the protein is no longer present after the magnetite crystal is formed.

Introduction

Magnetotactic bacteria (MTB) are a unique group of bacteria whose commonality is that they all mineralize intracellular, membrane-bound, nanometer-sized crystals of single domain magnetite and or greigite (Balkwill et al., 1980; Bazylinski et al., 1993;

Frankel et al., 1979; Heywood et al., 1990). The morphology of the crystals varies depending on the species, but most are between 35-120 nm in length, which puts them in the single domain size range. The shape can be elongated hexagonal, tooth/bullet-shaped, cuboctahedral, octahedral, (Bazylinski and Frankel, 2004). One of the primary reasons

MTB have been studied is because of their ability to consistently biomineralize magnetite crystals that have a discreet uniform morphology and are strongly magnetic. The ability to make nano-sized crystals is important because of their value to applications in biotechnology for magnetic separations, in medicine for drug and DNA delivery, and in electronics for magnetic data storage, and for environmental clean-up of heavy metals such as cadmium (Matsunaga and Arakaki, 2007; Prozorov et al., 2007b; Tanaka et al.,

2010b; Tang et al., 2011; Tartaj et al., 2005). Magnetite doped with cobalt (CoFe2O4) has unique magnetic properties related to coercivity and magnetic saturation (Galloway et al.,

2011; Prozorov et al., 2007b; Tanaka et al., 2010b). To reproduce the process outside of 73 a MTB cell, it is necessary to first understand the mechanism of magnetite formation in

MTB.

The magnetosome membrane has a particular set of proteins associated with membrane specific proteins (Mam) or magnetic particle membrane specific proteins

(Mms) (Jogler et al., 2009a). Even though the magnetosome membrane is contiguous with the cell membrane, it contains proteins specific to the magnetosome membrane, which do not have homology to nonmagnetic organisms (Gorby et al., 1988; Jogler and

Schüler, 2009; Komeili et al., 2004). It is uncertain how the proteins get emplaced in the magnetosome membrane and very little is known about the timing of the proteins in the formation of magnetite (Komeili et al., 2004; Pradel et al., 2006).

There are 48 magnetosome associated proteins (Matsunaga et al., 2005) but not all bacteria make the same proteins and not all of them mineralize magnetite the same way

(Richter et al., 2007). This is partly due to the fact that many of these proteins are encoded in genes on the highly conserved, but unstable magnetosome island (MAI), which is a cluster of genes unique to MTB (Greenberg et al., 2005; Ullrich et al., 2005).

For example, the mamAB operon was determined to be necessary for magnetite biomineralization in M. gryphiswaldense MSR-1 (Lohße et al., 2011). However, there are several proteins that have been implicated in magnetite biomineralization, one in particular is Mms6 which is found in M. magneticum AMB-1 (Arakaki et al., 2003;

Scheffel et al., 2007).

M. magneticum AMB-1 contains approximately 20 magnetosomes, each 50-100 nm in size, which form a single chain inside the cell. A well-studied protein in AMB-1

74 that has been implicated in controlling the morphology of intracellular magnetite is

Mms6; a 6.3kD protein containing 59 amino acids (Arakaki et al., 2003). The protein has a hydrophilic C-terminal that is tightly bound to the magnetite crystal and a membrane- bound hydrophobic leucine-glycine rich N-terminus (Amemiya et al., 2007; Arakaki et al., 2003; Prozorov et al., 2007a). Mms6 has two phases for binding iron, one that is high affinity and one that is low affinity, but high capacity (Wang, 2011; Wang et al., 2012a).

According to Wang (Wang, 2011), once iron binds to an Mms6 protein, it then undergoes a structural change which allows several Mms6 proteins to pool together on the magnetosome membrane; this allows the magnetite crystal to nucleate.

This protein has been studied in vitro by taking the protein out of the cell and mixing it with iron and examining the type of magnetite that is synthesized (Prozorov et al., 2007a). In another instance, a synthetic peptide based on the Mms6 C-terminal acidic, LG -rich region has been implicated in iron binding and interaction with magnetite crystals (Arakaki et al., 2010; Arakaki et al., 2003). Mms6 binds ferric iron better than

3+ 2- ferrous iron or La , but it can also bind tellurium in the form of tellurite TeO3 biomineralization (Tanaka et al., 2010b; Wang et al., 2012b). Recently, in vivo studies have been performed using an AMB-1 mutant that is missing the mms6 gene and therefore does not produce the protein (Tanaka et al., 2011). The results demonstrated that wild type AMB-1 had magnetite crystals that were only in the [100] and [111] direction, but AMB-1 missing the gene for mms6 (∆mms6) showed that the magnetite was smaller and had a different morphology than the wild-type AMB-1 (Tanaka et al.,

2011).

Most studies have focused on determining the function of Mms6 through proteomics, genetic studies, or in iron binding studies, but no one has actually examined the localization of Mms6 inside of single cells. One possible way to help determine the function of a protein is to use immunolabeling to localize the protein within the cell

(Pradel et al., 2006; Taoka et al., 2007; Taoka et al., 2008). That was the goal of this study. I determined the location of Mms6 proteins inside of individual cells of M. magneticum AMB-1 by using two different immunolabeling techniques, gold immunolabeling and fluorescent labeling. Previous work had showed in vitro that Mms6 was closely attached to the magnetite crystal fraction of lysed cells. The study presented here confirms in vivo that the protein was localized in close proximity to the magnetosomes. However, not all the cells containing magnetosomes were labeled, and those that were labeled, were not labeled along the entire length of the magnetosome chain. This implies that Mms6 has a finite lifetime within a cell.

Methods

Protein Purification

The protein was purified using the technique described by Prosorov et al. (2007a).

Plasmids (pTrcHis-TOPO) containing the gene sequence for the mature Mms6 protein from M. magneticum AMB-1 and a poly His-tag on the N-terminus were used to transform XL-1 Blue Competent Cells (Stratagene). The proteins were overexpressed in the cells using IPTG induction and then purified using Talon metal affinity resin

(Clontech). Most of the proteins were insoluble so 8 M guanidine was used to dissolve the inclusion bodies. The recombinant proteins were dialyzed to refold them into their 76 native state and resolved by SDS-PAGE to confirm that the correct protein was purified

(refer to Appendix B for details of the purification process).

Mass spectrometry of recombinant protein

Mms6 protein was sequenced at The Ohio State University Mass Spectrometry and Proteomics Facility. The protein band was cut from the gel and then digested and sequenced using Capillary-liquid chromatography tandem mass spectrometry (Cap-

LC/MS/MS) using a Thermo Finnigan LTQ mass spectrometer equipped with a

CaptiveSpray source (Bruker Michrom Billerica, MA) operated in positive ion mode.

Sequence information from the MS/MS data was processed by converting the raw data files into a merged file (.mgf) using an in-house program,

RAW2MZXML_n_MGF_batch (merge.pl, a Perl script). The resulting mgf files were searched using Mascot Daemon by Matrix Science version 2.3.2 (Boston, MA) and the database searched against the full SwissProt database version 2012_06 (536,489 sequences; 190,389,898 residues) or NCBI database version 20120515 (18,099,548 sequences; 6,208,559,787 residues) (refer to Appendix B for the sequence comparison).

Protein identifications were checked manually and proteins with a Mascot score of 50 or higher with a minimum of two unique peptides from one protein having a -b or -y ion sequence tag of five residues or better were accepted.

Antibody production

The purified recombinant proteins were sent to ProSci Incorporated (Poway, CA) to produce polyclonal antibodies in rabbits. After 8 weeks, the serum was removed from the rabbit and the Mms6 antibodies were purified from rabbit serum using affinity

77 purification by attaching recombinant Mms6 to a resin and running the serum through a column containing the Mms6 resin. The final concentration of the antibody was 1.42 mg/mL as determined by direct ELISA. Pre-immune serum was removed from the rabbit before injecting the antigen (recombinant Mms6) into the rabbit.

Immunoblotting

Two different amounts of recombinant Mms6 protein (0.5 µg and 0.05 µg) and 3 different cell fractions M. magneticum AMB-1 were resolved using 10 well 16% Tris- glycine gel (Invitrogen) SDS-PAGE. The three cell fractions and the total amount used in each lane were: a) magnetosome membrane (11 µg), b) cellular soluble protein (4 µg), and c) cell membrane (11 µg). Two exact gels were run simultaneously; one gel was used for blotting, the other was stained with Simply Blue Safe Stain (Invitrogen) and imaged. BenchMark prestained protein ladder (Invitrogen) was used on both gels. The proteins were blotted onto a PVDF membrane (Invitrogen), blocked with 5% BSA, labeled with the anti-Mms6 antibody at a concentration of 1:50,000, and labeled with a secondary goat anti-rabbit HRP antibody at a concentration of 1:200 using the Clean-Blot

IP Detection Kit HRP (Thermo-Pierce). The membrane was analyzed on a Gel Logic

1500 using Kodak software. As a control to test whether or not the antibody is binding to

Mms6, the procedure above was repeated, but 300 µg of recombinant Mms6 was added to the antibody solution.

Magnetosome isolation and purification

M. magneticum AMB-1 cells (ATCC 700264) were cultured in two 19 L carboys containing 15 L of magnetic Spirillum growth medium (ATCC medium 1653) (refer to

Appendix B for the formulation of the media), purged with nitrogen gas for 20 minutes, inoculated with 250 mL of culture and allowed to grow to midlog phase. The cells were harvested and centrifuged using a continuous flow centrifuge (Heraeus 17RS) for 1 hour.

The cell pellet was resuspended in 10 mM Tris-HCl pH 8.0 and the cells were lysed by passing them through a French press 3 times using 18,000lb/in2 of pressure. The magnetosomes were isolated placing the solution in a beaker with two cobalt samarium magnets on the outside of the beaker on opposite sides. After 1 hour the solution was removed and replaced with fresh 10 mM Tris-HCl buffer. After 10 minutes the solution was poured off and replaced with fresh 10 mM Tris-HCl. This was repeated 11 more times after which the magnetosomes were suspended in 10 mM Tris-HCl. An aliquot of this final suspension was used to separate the magnetite crystals from the membranes by incubating the magnetosomes in 10 mM Tris-HCl with 1% SDS for 3 hours. The solution was centrifuged at 12,000 x g for 20 minutes and the supernatant containing only magnetosome membrane was removed. The absence of magnetite in this fraction was checked using Transmission Electron Microscopy (TEM).

Cell fractionation

The solution from the first magnetosome isolation step (above) was used to isolate the cell membrane fraction and the soluble protein fraction by centrifugation at 200,000 x g at 4oC for 3 hours. The supernatant (containing the soluble proteins) was removed and the pellet (containing the cell membrane fraction) was resuspended in 10mM Tris-HCl.

Embedding and labeling with Immunogold

Magnetospirillum magneticum AMB-1 were cultured in a 1L Schott bottle using the formula for magnetic Spirillum growth medium (ATCC medium 1653). The sealed bottles were purged with nitrogen gas for 30 minutes before autoclaving. After autoclaving the bottles were inoculated with AMB-1 and grown at 28°C until they reached midlog phase, after which they were processed for TEM.

The cells were pelleted at 10,000 x g for 10 minutes and then washed with 0.1 M sodium cacodylate, fixed in 3% paraformaldehyde/0.1% glutaraldehyde, dehydrated in a graded ethanol series, and embedded using LR white medium grade resin (EMS) in 00 gelatin capsules and cured at 60°C for 24 hours. Sections 60 nm in thickness were cut from one of the blocks and placed on formvar and carbon coated nickel slot grids (Ted

Pella). The sections were allowed to dry on the grids for several hours to overnight before being immunolabeled. The grids were floated on drops of 0.1 M glycine in Tris-

HCl for 20 minutes, then blocked with 0.1% cold water fish gelatin, 1% BSA, 0.1%

Tween-20, in Tris-HCl for 30 minutes. The grids were incubated overnight at 4°C on a drop of Mms6 antibody diluted 1:2000 in 0.5% BSA Tris-HCl. The grids were then washed with Tris-HCl with 1% BSA and incubated on drops of goat anti-rabbit IgG antibody conjugated with 10 nm gold (Sigma) diluted 1:100 in 0.5% BSA in Tris-HCl for

2 hours. The grids were then rinsed in Tris buffer followed by distilled water, and analyzed using an FEI Spirit at 80keV spot size 2 and imaged using a Gatan camera. No adjustments were made to the image after acquisition other than cropping in Adobe

Photoshop.

Carbon-formvar 200 mesh nickel grids (Ted Pella) were placed on drops of purified magnetosomes that were diluted 1:1250 for 5 minutes and then rinsed in water for 15 minutes. The grids were then blocked with 5% BSA-Tris-HCl, incubated with the anti-Mms6 diluted 1:4000 in 0.5% BSA-Tris HCl for 8 hours at room temperature, blocked again in 1% BSA-Tris-HCl, incubated with goat anti-rabbit IgG antibody conjugated with 10 nm colloidal gold (Sigma) diluted 1:100 in 0.5% BSA in Tris-HCl for

8 hours at room temperature, washed in TBS-HCl, and then washed with water, and analyzed using an FEI Spirit at 80 keV spot size 2 and imaged using a Gatan camera. No adjustments were made to the image after acquisition other than cropping in Adobe

Photoshop.

Fluorescence microscopy

M. magneticum AMB-1 cells were grown using the formula for magnetic

Spirillum growth medium (ATCC medium 1653) in 125 mL serum bottles containing 55 mL media with nitrogen gas in the headspace. Once the cells had reached midlog phase, they were harvested by centrifuging at 10,000 x g for 10 minutes at 4°C and the cell pellet was fixed in 4% paraformaldehyde for 5 minutes. The cells were centrifuged again at 8,000 RPM for 10 minutes and the cell pellet was suspended in water. A drop of cells was placed on a slide and allowed to air dry. The slide was washed with PBS, labeled with primary antibody (1:400) for 1 hour, washed with PBS, labeled with a secondary antibody, goat anti-rabbit IgG anti-body conjugated to Dylight 488 (1:100) (Thermo

Scientific), washed with PBS and water. The cells were analyzed using an Olympus

FluoView 1000; no adjustment was made to the images after acquisition other than cropping in Adobe Photoshop.

Results

Recombinant protein purification

The different fractions from the recombinant protein purification procedure were resolved using SDS-PAGE (Figure 23). The lanes containing the protein elutions are dominated by bands between 10 kDa and 17 kDa corresponds to the recombinant His- tagged Mms6 protein, however the lanes do contain minor amounts of other protein bands above and below the main band. The band in lane 12 was removed and sequenced was found to contain a 12.5 kDa protein that was determined to correspond to a magnetite particle specific iron-binding protein (refer to Appendix B).

Immunoblotting

The SDS-PAGE analysis showed a double band near 15 kDa representing the recombinant Mms6 protein (Figure 23), similar to Wang et al., (Wang, 2011). The less concentrated protein lane showed no band. The lanes with the different cell fractions showed many bands, specifically the magnetosome membrane lane exhibited faint banding between 15 kDa and 6 kDa. There was no obvious banding below 37 kDa in the soluble protein fraction lane.

The immunoblot showed several bands in the more concentrated Mms6 lane, but primarily the labeling was a double band between 15 kDa and 19 kDa ( Figure 24). The lane containing the magnetosome membrane fraction contained only a single band between 6 kDa and 15 kDa ( Figure 24). The other lanes displayed no obvious banding.

Figure 22 The different fractions from the protein purification process resolved using SDS-PAGE. Lane 1 is the cell lysate, lane 2 is the flow through, lanes 3-6 are the washing steps with increasing concentrations of imidazole (5,mM 5mM, 15mM, and 30mM), lanes 7-9 and 11-13 are the protein elutions using 300mM imidazole, and lane 10 is the protein marker.

Figure 23 SDS-PAGE showing the protein and cell fractions from M. magneticum AMB- 1 used to test the affinity of the Mms6 antibody in the immunoblot. Lane 1) cell membrane fraction (11 µg), lane 2) soluble protein (4 µg), lane 3) magnetosome membrane (11 µg), lane 4) recombinant Mms6 protein (0.05 µg), lane 5) recombinant Mms6 protein (0.5µg), lane 6) BenchMark protein ladder.

A B

In the control immunoblot, excess protein was added to the antibody to ensure that the antibody is binding only to Mms6 and not labeling other parts of the cell fractions.

Microscopy analysis

The confocal microscope and the transmission electron microscope (TEM) were used to examine cells that were immunolabeled with anti-Mms6 antibodies. By labeling cells with an antibody specific for Mms6, the specific location of the proteins within the cells of M. magneticum AMB-1was found. Two different methods were used to examine the cells because each technique has different capabilities. Whole, intact cells were examined using confocal microscopy, but the resolution of the confocal microscope does not allow you to see the magnetosomes. On the other hand, the TEM cannot look at intact cells, but it has the capacity to resolve nanometer-sized crystals of magnetite. For the TEM, whole, intact isolated magnetosomes were analyzed as well as ultrathin sections of AMB-1 cells. Because these are less than 0.09 µm thick sections, the whole cell (which is more than 500 µm wide) is not visible and the chain of magnetosomes gets clipped.

The samples were tagged with polyclonal antibodies that were made specifically to target Mms6 epitopes. These were used as a primary antibody, which attaches directly to the Mms6 protein. This was followed by labeling with a secondary antibody, which has affinity for the primary antibody. The secondary antibody has a marker conjugated to it, which makes it visible in the microscope. In the case of the confocal microscope, the secondary antibody contains a fluorophore, which fluoresces when excited by a laser.

For the TEM, the secondary antibody needs to contain an electron opaque marker; in this

86 case it is a 10 nm gold sphere. This appears as an opaque sphere in the micrograph at the site of the protein.

The images in Figure 25 are examples of fluorescently labeled cells analyzed using the confocal microscope. The three images in the figure are taken of the exact same location on the slide, but using two different modes, Nomarski differential interference contrast and confocal laser scanning microscopy. Nomarski imaging is useful for visualizing the whole cell, which give a reference for where the fluorescence is located within the cell. These can then be combined into one image to give an overall picture of the location of fluorescence within the cell. The image in panel A of Figure 25 shows the overlap of the Nomarski image (black and white) and the confocal image

(green areas). This image shows that the cells labeled with the fluorescent tag demonstrate labeling in the center, along the long axis of the cell, corresponding to where a chain of magnetosomes would lie (Figure 25). In some instances the labeling is punctate and does not run in a consistent line along the axis. Panel B and C are the two individual images using the Nomarski method (B) and the confocal image (C). In Figure

25B, a distinct line of features is seen on two of the cells, which denote the chain of magnetosomes. The fluorescence labeling in Figure 25C corresponds to the location of this feature indicating labeling of magnetosomes. A key feature of panel Figure 25A is that no labeling is occurring outside of the middle of the cell, i.e. no labeling on the cell membrane, and no labeling throughout the cell. A negative control was prepared in order to show that this was true labeling and not background labeling by the secondary

87 antibody. The control slide (not shown) did not contain any labeled cells; therefore it was just an opaque image.

Two different types of samples were analyzed using the TEM, 90 nm thick ultrathin sections of AMB-1 cells (Figure 26), and purified chains of magnetosomes from

AMB-1 cells (Figure 27). In Figure 26A the cells displayed labeling near the magnetosomes, which was either directly touching the magnetite crystal or within the matrix surrounding the magnetosomes. However, labeling did not occur on all the magnetosomes, which may be a consequence of the cells being sections, which have a limited number of epitopes at the surface of the section. The other noticeable detail in

Figure 26A is that there was no labeling in other parts of the cell. This demonstrates that labeling was limited to just the magnetosomes and no other area of the cells, such as the cell membrane or the cytoplasm (refer to Appendix B for examples of sections labeled with different concentrations of primary antibodies). Panels B and C of Figure 26 represent two different control experiments, Figure 26B is labeled the same as the thin section in Figure 26A, but with the primary antibody substituted with 0.5% BSA-Tris-

HCl (similar to the control in the confocal microscope sample). Figure 26C was prepared the same as Figure 26A, but with the primary antibody (anti-Mms6) substituted with 5% pre-immune serum from the rabbit that was used to produce the antibodies. The pre- immune serum should not contain any of the antibodies for Mms6. These results clearly demonstrated that the secondary antibody only had affinity for the primary antibody and not for other epitopes within the cell.

Figure 25 Fluorescently labeled Magnetospirillum magneticum AMB-1 cells using anti- Mms6 (1:400) as the primary antibody and goat anti-rabbit Dylight 488 (1:100) as the secondary. A) The combined Nomarski and fluorescent images showing the outline of M. magneticum cells with the fluorescent tag running in the middle of the cells. B) The same field of view as A, but using only Nomarski microscopy. C) Cells excited with the 488 nm laser. Scale bar 5 µm.

Figure 26 Ultrathin sections of Magnetospirillum magneticum AMB-1. A) A single cell showing magnetosomes (white arrow) labeled with anti-Mms6 (1:2000) and then secondary labeled with goat anti-rabbit antibody conjugated with 10nm colloidal gold (black arrows). Inset shows a detail of the immunolabeling on one of the magnetosomes. B) Negative control for the immunolabeling showing an ultrathin section treated exactly the same as the ultrathin section in A, but substituting 0.5% BSA Tris-HCl for the primary antibody. C) Another negative control for the immunolabeling treated exactly the same as the section in A, but with 5% pre-immune serum substituted for the primary antibody. Scale bar is 500 nm for all three images.

TEM was used to analyze chains of magnetosomes isolated from AMB-1 cells. The magnetosomes were labeled with the same gold-conjugated secondary antibody as the thin sections (Figure 27A) (refer to Appendix B for an example of magnetosomes labeled with a more concentrated amount of primary antibody). The significance of this image is that it unmistakably shows labeling completely around the magnetosomes, which correlates nicely with the data from the confocal imaging (Figure 25A). The other two images in Figure 27 are the negative controls exactly like the two controls in the thin sections. Figure 27B is prepared exactly as A, but with 0.5% BSA-Tris-HCl substituted for the primary antibody, and Figure 27C is the primary antibody substituted with 5% pre-immune serum.

Discussion

Magnetotactic bacteria are important because of their unique capacity to biomineralize magnetite with a specific shape, size, and composition. They do this by using proteins that are exclusive to these bacteria. One of the big challenges right now is determining which proteins are specific to magnetosomes in order to understand how

MTB biomineralize crystals. One protein that is directly involved in the mineralization process is Mms6, which is a protein that was first isolated from M. magneticum AMB-1

(Arakaki et al., 2003). This protein has been shown to be associated with the magnetite crystal where it is believed to control the shape of the magnetite crystal (Arakaki et al.,

2010; Arakaki et al., 2003; Galloway et al., 2011; Prozorov et al., 2007a; Tanaka et al.,

2011). Mms6 proteins have been used outside the cells to determine how they interact

Figure 27 Purified magnetosomes from M. magneticum AMB-1. A) Magnetosomes (black arrows) labeled with anti-Mms6 antibody (1:4000) and secondary labeled with goat anti-rabbit antibody conjugated with 10 nm colloidal gold (white arrows). The inset shows a single magnetite crystal labeled with gold conjugated antibody. The space between the white arrows depicts the magnetosome membrane. B) Negative control for the immunolabeling treated exactly the same as A, except substituting 0.5% BSA-Tris- HCl for the primary antibody. C) Negative control treated exactly the same way as A, but substituting 5% pre-immune rabbit serum in place of the primary antibody. Scale bar is 100 nm in each image.

92 with soluble iron and they found that the protein could nucleate nanometer-sized crystals of magnetite and control the shape and size of the crystals (Amemiya et al., 2007;

Prozorov et al., 2007a). But it was also found that this protein could be used to mineralize nonbiological minerals of cobalt ferrite (Galloway et al., 2011; Prozorov et al.,

2007b). It is important to understand how this protein functions because it has been shown to be fundamental to mineralizing magnetite and other bioinspired crystals.

The localization study described here used immunolabeling to answer the question where Mms6 proteins are localized in M. magnetotacticum AMB-1 cells. The analysis with the confocal microscope showed fluorescence running the length of the cell, either intermittently or in a large cluster, corresponding to where the chain of magnetosomes would lie (see Figure 25). Similarly, gold labeling of the isolated magnetosomes clearly showed Mms6 immediately adjacent to the magnetite particle (see

Figure 27). However, there was some labeling where the gold label was not in contact with the magnetite crystal. This could be caused by either the protein epitope existing in the magnetosome matrix or the integrity of magnetosome membrane was destroyed in the preparation and it exposed Mms6 epitopes, causing labeling not immediately adjacent to the magnetite.

An interesting observation was that not all magnetosomes were labeled in the images (see Figure 26 and Figure 27). On the thin sections, this could be because the epitope was not exposed at the surface of the section. But in the confocal image, where the whole cell has the potential of being labeled, and in the isolated magnetosomes where the whole magnetosome is exposed to the antibody, there should be labeling on all of the

93 magnetosomes if Mms6 is present on each and every one of the magnetosomes. But not all of the magnetosomes are labeled which implies that the protein is no longer present after a certain time during the growth of the crystal. Additionally, in the fluorescent images, it was clear that not all the cells were labeled. This could be caused by the lack of magnetosomes in the cells, or that the protein is not always present during the lifecycle of the magnetosomes. If the protein is not always present in the magnetosomes, this could mean that the protein is only expressed during a specific time of the mineral growth cycle. This is consistent with the observation of Tanaka et al. (Tanaka et al., 2011) who said that Mms6 is not involved with the nucleation process and is only involved at certain stages of the mineralization process.

In the immunoblot analysis, the only positive band was found in the lane containing the magnetosome membrane. This indicates that the only significant amount of Mms6 in the cell is contained in the magnetosome membrane. It was shown that the magnetosome membrane is actually an invagination of the cell membrane and not an individual membrane surrounding each magnetosome (Komeili et al., 2006; Komeili et al., 2004). One of the observations made clear in the present study is that Mms6 is not found associated with the cell membrane (no labeling in the microscope images and no bands in the immunoblot), which means that the proteins are emplaced in the magnetosome membrane after the magnetosome membrane becomes invaginated. This further adds credibility to the notion that Mms6 has a specific life cycle and that the

Mms6 proteins are inserted into the membrane and removed from the membrane at different times during the cell cycle.

It is believed that the proteins directly involved with nucleating the magnetosome crystal and controlling biomineralization are found in the magnetosome membrane

(Gorby et al., 1988; Richter et al., 2007; Tanaka et al., 2006). Magnetosome membranes have been found to be contiguous with the cell membrane (Komeili et al., 2004).

Therefore, some proteins are present while the membrane invaginates and before the mineral forms. It would therefore be likely that the proteins that are present before mineralization would be found in the cell membrane. Conversely, you would expect that proteins that are only involved with the mineralization process would no longer be present once the crystals are mature. Therefore, proteins that are only necessary during the biomineralization process would only be present during that process.

There is a significant amount of labeling on the isolated magnetosomes, which ties into the idea that Mms6 controls the morphology of the crystal (Arakaki et al., 2010;

Lohße et al., 2011; Tanaka et al., 2011). However, there does not appear to be uniformity to the gold labeling around each individual magnetosome; some have no labeling, while others have several. This could be due to the randomness of the labeling technique, but it could also be true labeling and be the result of the fact that there were more Mms6 epitopes at some magnetosomes than at others. If this is the case, then labeling is occurring because the protein is present in higher concentrations on some magnetosomes than others; suggesting that the protein is actively performing its function on the individual magnetite crystal. Conversely, the absence of labeling could be caused because the protein is no longer present, perhaps because the crystal has completed its growth cycle. It is unclear in this study because the cells in this study were harvested

95 while they were still actively growing during log phase, making it difficult to see many examples of this.

According to the model proposed by Wang et al. (Wang, 2011) Mms6 proteins cluster together to nucleate and shape the magnetite particle. If this were the case, there should be tighter clustering of the label in the images presented here, however this is not seen. This would suggest that Mms6 is not clustering on the magnetosome membrane.

But it is possible that Mms6 proteins are clustering, but the configurational hindrance of the antibody complex could cause a lack of the label clustering. One way to check for this would be to develop a single label for Mms6 and try the experiment using just primary antibody labeling.

Chapter 5: Atomic Force Microscopy and Transmission Electron Microscopy of Magnetotactic Bacteria

Abstract

Atomic force microscopy (AFM) was used in concert with transmission electron microscopy (TEM) to image magnetotactic bacteria (Magnetospirillum gryphiswaldense

MSR-1 and Magnetospirillum magneticum AMB-1), magnetosomes, and purified Mms6 proteins. Mms6 is a protein that is associated with magnetosomes in M. magneticum

AMB-1 and is believed to control the synthesis of magnetite (Fe3O4) within the magnetosome. I demonstrated how AFM can be used to capture high-resolution images of live bacteria and achieved nanometer resolution when imaging Mms6 protein molecules on magnetite. I used AFM to acquire simultaneous topography and amplitude images of cells that were combined to provide a three-dimensional reconstructed image of M. gryphiswaldense MSR-1. TEM was used in combination with AFM to image M. gryphiswaldense MSR-1 and magnetite-containing magnetosomes that were isolated from the bacteria. AFM provided information, such as size, location and morphology, which was complementary to the TEM images.

Introduction

Environmental microbiologists are often interested in understanding the molecular-scale biogeochemical reactions that occur between a biomolecule, such as a protein, and a mineral surface. For example, microbial proteins can catalyze the

97 synthesis of mineral phases by influencing crystal nucleation and/or growth (Bazylinski and Frankel, 2004; Hernández-Hernández et al., 2008). These reactions in turn control the bioavailability of elements to the surrounding ecosystem, which can have profound effects (both positive and negative) on water quality, mineral distribution, migration of subsurface contaminants, and soil productivity (Barton and Fauque, 2009; Dittrich and

Luttge, 2008; Edwards et al., 2005; Hochella Jr et al., 2008; Madsen, 2011; Weber et al.,

2006).

It is a challenge to probe reactions that occur between a microorganism (or protein) and mineral due to the small size of the reacting space as well as the fact that the interface itself is hidden from direct view. Transmission electron microscopy (TEM) affords the spatial resolution and, with the aid of a diamond microtome, the ability to visualize the interface. But, TEM operates in a vacuum rendering cells dead and biomolecules in a “non-hydrated” state. Atomic force microscopy (AFM) allows one to image living cells in solution so that biomolecules are in their native state. While, AFM is capable of atomic resolution, it can be challenging to achieve high resolution with soft, deformable samples like cells. In this paper, I compare and contrast TEM and AFM analyses of proteins and minerals produced by two magnetotactic bacteria:

Magnetospirillum gryphiswaldense MSR-1 and Magnetospirillum magneticum AMB-1.

These two magnetotactic bacteria use specific proteins (e.g., Mms6) to synthesize magnetite (Fe3O4) nanoparticles within a membrane-bound organelle called a magnetosome (Bazylinski and Frankel, 2004). Each bacterium contains up to 60 magnetosomes and each magnetosome contains one magnetic Fe3O4 nanoparticle

(Bazylinski and Frankel, 2004). The magnetosomes are aligned in a chain-like fashion, which imparts a magnetic dipole to the bacterial cell, allowing the cell to align itself within Earth’s geomagnetic field (Bazylinski and Frankel, 2004). In addition, magnetotactic bacteria have a flagellum, which they use for mobility.

I demonstrate that AFM can be used to obtain high-resolution images of subcellular structures synthesized by a living M. gryphiswaldense MSR-1 bacterium. I show complementary TEM and AFM images of magnetosomes isolated from M. gryphiswaldense MSR-1. I also collected AFM images of individual Mms6 proteins on magnetite immersed in solution. I chose to image Mms6 because it is a magnetosome- associated protein from M. magneticum AMB-1 that has been implicated in mediating the synthesis of magnetite in these bacteria (Bazylinski and Frankel, 2004).

In general, the TEM provides higher resolution of ultrafine features, but the AFM reveals the association of proteins and minerals in their natural environment where protein expression and location are regulated, and where molecular neighbors can influence function. In this way, AFM confirms that TEM records the native structural relationship of proteins and minerals with living cells, for properly prepared specimens.

Ongoing experiments are aimed at using AFM to observe real-time growth of magnetite in the presence of Mms proteins.

Material and methods

Growth and preparation of bacteria for AFM and TEM

Small cultures (e.g., 50 – 100 mL) of the magnetotactic bacterium M. gryphiswaldense MSR-1 were grown in flask standard medium under microaerobic

99 conditions (100 rpm) at room temperature (Heyen and Schüler, 2003). Because these microbes synthesize magnetite (Fe3O4) nanoparticles, simple bar magnets were used to isolate and concentrate the M. gryphiswaldense MSR-1 from the cultures after several days of growth. The concentrated bacterial samples were then spotted onto a carbon coated 200 mesh copper TEM grid (Ted Pella #01800), dried under N2 gas and imaged with AFM or TEM, as described below.

Atomic force microscopy (AFM)

AFM was conducted with either a Digital Instruments Bioscope AFM with

NanoScope IV controller or an Asylum Research MFP3D AFM. Both AFMs rest on inverted optical microscopes (Zeiss Axiovert 200M or Nikon 300TE), which allows accurate positioning of the tip over the sample. AFM began within 10-30 min of preparing the samples. Cantilevers included Olympus AC160TS and AC240TS tips (tip radii < 10 nm) and silicon nitride probes from Bruker (DNP-10; tip radius of 20-60 nm).

AFM imaging was conducted in phosphate buffered saline (PBS), pH 7.4, or air with relative humidity ~40%, at room temperature. For protein experiments, 10 µg/mL of stock solution was spotted onto freshly cleaved mica or cleaned magnetite, incubated for 60 seconds, rinsed with PBS and imaged. For microorganism imaging, bacteria growing in liquid medium were spotted onto OTS-coated slides, incubated for 5 minutes, washed with PBS and then imaged (Yongsunthon and Lower, 2005). Images were recorded at a line frequency of < 1 Hz and 512 x 512 pixels or 1024 x 1024 pixels. Scans of the mica or magnetite were 500 nm x 500 nm or less.

Transmission electron microscopy

100

Magnetotactic bacteria or magnetosomes were adsorbed to formvar stabilized and carbon coated 200 mesh copper grids (Ted Pella #01800). The grids were placed with the carbon side down on a drop of cell suspension or magnetosome suspension for 10 minutes, then immediately washed one time by placing the grid on a drop of water for 30 seconds. For staining, the grids were placed on a drop of 2% uranyl acetate (Ted Pella

#19481) for either 30 seconds or 5 minutes, then dried completely using a piece of filter paper. The grids were analyzed using transmission electron microscopy using either an

FEI Tecnai Spirit at 80kV or with a FEI Tecnai F20 using high angle annular dark field

STEM at 200kV.

Results

Imaging microorganisms, proteins, and antibodies with AFM

Figure 28 shows a tapping-mode AFM image of M. gryphiswaldense MSR-1.

Simultaneous height (Figure 28A) and amplitude (Figure 28B) images were collected on this sample. The height image shows the topography of the sample, that is, the lateral and vertical (z) dimensions of the cell. The amplitude image, on the other hand, is essentially a derivative of the height image, and therefore highlights edges or other features that may be hardly detectable in the height image. While the z information is lost, the appearances of features in an amplitude image (e.g., shape of bacterium) are similar to those seen with an optical or electron microscope.

Subcellular structures are visible in both the height and amplitude images shown in Figure 28. Two intracellular inclusion bodies can be seen as bright white circles with a diameter of approximately 250 nm (Figure 28A). The inclusions can also been seen in

101 the amplitude image (Figure 28B) as gray slightly irregular circles. Figure 28 also shows a chain of approximately 8 magnetosomes running longitudinally down the center of the bacterium. The magnetosomes can be seen much more clearly in the amplitude image

(Figure 28B) and each magnetosome is approximately 75 nm in diameter. A single flagellum is located at the top polar end of the bacterium (Figure 28). The thickness of the flagellum is approximately 100 nm, which is larger than expected but likely an artifact distorted by the edges of the AFM tip. It appears that the bottom polar end of the cell lysed open (Figure 28), which could have been caused by the AFM tip as it was raster scanned across the cell surface during imaging. As a result, some of the bacterium’s intracellular contents were released onto the TEM grid (Figure 28).

AFM of Mms6 proteins synthesized by M. magneticum AMB-1

I also used AFM to collect images of purified Mms6 proteins on Fe3O4 (Figure

29). Mms6 is a magnetosome-associated protein that is synthesized by M. magneticum

AMB-1 and is believed to be involved in the biomineralization of Fe3O4 (Amemiya et al.,

2007; Bazylinski and Frankel, 2004). A 100 µL drop of purified Mms6 (10 µg/mL) was spotted onto a cleaned Fe3O4 crystal and incubated for 10 minutes at room temperature.

Next, the sample was gently washed with 1 mL of PBS and imaged by tapping mode

AFM (Figure 29). Figure 29 shows the basic morphology that was observed for Mms6.

All of the Mms6 proteins that were observed with AFM had the same oblong morphology and an approximate size of 25 nm x 50 nm (Figure 29). The height of each Mms6 protein was approximately 7 nm (Figure 29).

Since Mms6 had never before been imaged with AFM, I wanted to compare

102

103

104

the morphology of Mms6 (Figure 29) to protein structures, such as antibodies, that have been previously imaged by AFM. Therefore, I used AFM to collect images of purified green fluorescent protein (GFP) antibody molecules (Figure 30). A 100 µL drop of

1:1000 anti-GFP (RDI Fitzgerald Inc.) diluted in PBS was spotted onto freshly cleaved mica. Mica was used in place of magnetite because the antibodies readily bind to mica.

The antibodies were incubated for 10 minutes at room temperature, gently washed with 1 mL of PBS and imaged by tapping mode AFM.

Figure 30 shows the three basic morphologies that were observed for the anti-GFP molecules. Some molecules appeared to have one domain (Figure 30A) with a size of approximately 8 nm x 12 nm. Other molecules appeared to have a V-shaped morphology consisting of two domains with roughly equal dimensions (6 nm x 8 nm), which were connected to one another by a smaller 3 nm x 3 nm domain. Finally, a third Y-shaped morphology was observed on the mica. These structures consisted of three approximately equal oval-shaped domains. Each domain was roughly 6 nm x 8 nm. The height of all the antibodies observed in Figure 30 appeared to be 4 nm. For comparison, the three-dimensional morphologies and sizes of antibodies, including their Fab and Fc fragments, are provided in Figure 30 beside each topographic image (Silverton et al.,

1977). The sizes and shapes of the antibodies observed in our topographic images

(Figure 30) are consistent with Silverton et al., 1977 and previously published images

(Kienberger et al., 2004; San Paulo and Garcia, 2000).

105

106

TEM of M. gryphiswaldense MSR-1

Figure 31 shows a single M. gryphiswaldense MSR-1 bacterium that was imaged by TEM. A single flagellum, approximately 8 µm long, can be seen attached to the bacterium (Figure 31). Pieces of flagella < 2 µm in length can also be seen in the TEM image of M. gryphiswaldense MSR-1 (Figure 31). Despite great care in sample preparation, these pieces of flagella were likely sheared off during TEM sample preparation.

The TEM image (Figure 31) also shows two intracellular inclusion bodies, which can be seen as light gray irregularly shaped circles. TEM also revealed two chains of approximately 20 magnetosomes (black particles) each running along the longitudinal axis of the M. gryphiswaldense MSR-1 bacterium (Figure 31). These subcellular structures that were observed in the TEM image of Figure 31 are similar to those seen in the AFM image of M. gryphiswaldense MSR-1 (Figure 28).

AFM and TEM of magnetosomes produced by M. gryphiswaldense MSR-1

Magnetosomes were purified from M. gryphiswaldense MSR-1 using a previously described protocol (Grünberg et al., 2004) and imaged by both TEM and AFM (Figure

32). The TEM image (Figure 32A) shows nine cubo-octohedral magnetosomes arranged in a chain. The magnetosome membrane can been seen as a light-gray structure encasing dark-black/gray Fe3O4 crystals (Figure 32A). The membrane can be clearly seen connecting several of the magnetosomes (e.g., the first and second magnetosome in

Figure 32A). The magnetosomes range in size from approximately 30 – 75 nm.

107

Figure 31 TEM image of M. gryphiswaldense MSR-1. Inclusion bodies (ib) and two magnetosome chains (mc) are shown. A scale bar for length is provided.

108

109

Figure 32B shows a phase contrast image of a single magnetosome particle collected using AFM. The phase contrast image is shown here because it highlights nanometer-scale variations in the magnetosome particle that are not apparent in topography images. The magnetosome is approximately 120 nm in diameter (Figure

32B), which is approximately twice a large as the magnetosomes imaged by TEM (Figure

32A). The sides of the particle appear to adhere to the AFM tip more readily (bright white color along the perimeter of the particle) than the top of the particle (Figure 32B).

The top of the magnetosome (gray/black color) is similar to the substrate that the magnetosome is sitting upon (Figure 32B). As expected, this suggests that there are variations in the material composition of the magnetosome (i.e., organic membrane surrounding an inorganic Fe3O4 crystal).

Discussion

Imaging live microorganisms with AFM can be a powerful tool to study biogeochemical processes in-situ. Sample preparation is fairly simple because the only requirement for imaging is that the biological sample be adsorbed to a solid substrate

(e.g., mica, magnetite), which can then be placed into the microscope. In Figure 28 I demonstrated how tapping-mode AFM could be used to image live M. gryphiswaldense

MSR-1 cells and subcellular structures such as inclusion bodies and magnetosomes.

Inclusion bodies and magnetosomes are both intracellular structures, but I was still able to image them with AFM (Figure 28) even though it is only capable of imaging surfaces.

The reason I could image them here was because the inclusion bodies and magnetosomes

110 pushed up against the inside of the bacterial membrane causing their structures to “show” through the cell membrane, which allowed us to image them with AFM.

TEM, on the other hand, is able to image intracellular structures. When TEM was used to image M. gryphiswaldense MSR-1 I observed inclusion bodies and two magnetosome chains (Figure 28). The resolution of the TEM image (Figure 31) is definitely better than that of the AFM image collected on M. gryphiswaldense MSR-1

(Figure 28). However, it must be remembered that the TEM image of M. gryphiswaldense MSR-1 (Figure 31) is of a bacterium that has been fixed on a TEM grid and so it is no longer alive when it’s imaged. The M. gryphiswaldense MSR-1 bacterium that was imaged by AFM (Figure 28) was not fixed prior to imaging and was presumably still alive when I imaged it with AFM. The microorganisms imaged by AFM (Figure 28) and TEM (Figure 31) provide complementary data about the subcellular structures (i.e., size, shape, location) that are synthesized by the microbe. While it would be possible, although difficult, to image the exact same microorganism I did not do this here because the bacteria were obtained from the same culture and so they should be essentially identical microorganisms.

Many smaller geobiological samples, such as the magnetosomes shown in Figure

32, require scan sizes on the order of 500 nm x 500 nm. The resolution achieved by

AFM (Figure 32B) was comparable to the resolution achieved by TEM (Figure 32A).

The magnetosome membrane surrounding the Fe3O4 nanocrystals was visible by TEM

(Figure 32A). In the TEM image (Figure 32A), the membrane appeared as a gray

“sticky” substance between the black Fe3O4 nanocrystals. Very little membrane was

111 observed surrounding the black Fe3O4 nanoparticles. However, in the AFM phase image of the magnetosome (Figure 32B), it is clear that a “sticky” biological membrane (shown in white in Figure 32B) surrounds the entire Fe3O4 nanocrystal. The size of the individual magnetosome imaged by AFM (Figure 32B) is larger than the individual magnetosomes imaged by TEM (Figure 32A). This difference is due to the fact that the

AFM image (Figure 32B) was collected under “physiological” conditions (i.e., PBS, pH

7.4), whereas the sample prepared for TEM was fixed, dried and imaged in vacuum

(Figure 32A). The AFM image (Figure 32B) likely shows how the magnetosome, with its lipid bilayer coating, exists inside a living bacterium. It should be noted that the size and shape of the magnetosome observed by AFM (Figure 32B) was similar to images published in a previous study that also used AFM to image magnetosomes isolated from a magnetotactic bacterium (Yamamoto et al., 2010).

Inside living magnetotactic bacteria, the magnetosomes exist as a chain

(Bazylinski and Frankel, 2004). While I was able to successfully image an isolated magnetosome chain by TEM (Figure 32A), despite multiple attempts, I was never able to locate and image a chain of magnetosomes by AFM. Therefore, the AFM image provided in Figure 32B shows a single magnetosome. Nonetheless, Figure 32 demonstrates that by combining AFM with TEM a more thorough characterization of the geobiological material is achievable. One method permits the visualization of the magnetosome chain, while the other method allows for the examination of the magnetosome’s lipid bilayer.

112

In Figure 29 I used AFM to image Mms6 protein molecules from M. magneticum

AMB-1. The proteins were deposited onto a magnetite crystal and imaged with tapping- mode AFM. Magnetite was selected as the substrate because the proteins would readily bind to this mineral and because magnetite is the natural substrate for Mms6 in M. magneticum AMB-1. Mms6 proteins had an oblong-spherical shape with an approximate size of 25 nm x 50 nm and height of 7 nm (Figure 29). Since this was the first time that a biomineralizing protein from magnetotactic bacteria had been imaged by AFM, I wanted to compare its size and shape to a known protein. Anti-GFP was selected because it is a well-characterized protein that has a known structure, morphology and size (Kienberger et al., 2004; San Paulo and Garcia, 2000; Silverton et al., 1977).

Mms6 displayed one shape when imaged by AFM (Figure 29), but Anti-GFP molecules displayed three distinct morphologies (Figure 30), each with an approximate size of 10-25 nm x 10-25 nm and height of 5 nm. While the heights of Anti-GFP and

Mms6 were comparable (i.e., 5 nm for Anti-GFP; 7 nm for Mms6), the size of Mms6 was approximately twice as large as the Anti-GFP molecules. This suggests that the Mms6 molecules aggregated together on the magnetite substrate such that the oblong-spherical shapes observed in Figure 29 were actually >1 Mms6 molecule. These results were consistent with previous in-vitro AFM studies of purified MamA proteins that formed oligomeric complexes on mica (Yamamoto et al., 2010). MamA is localized to the surface of the magnetosome membrane in M. magneticum AMB-1 and is believed to function in magnetosome chain assembly, activation and perhaps stabilization (Komeili et al., 2004; Yamamoto et al., 2010). Our results suggest that Mms6 behaves like MamA

113 in-vitro and exists as oligomer. I am currently purifying other Mms proteins from M. magneticum AMB-1 so that I can image them by AFM to determine if they also form protein-protein complexes on magnetite. Such information would provide beneficial insight into possible protein-protein complexes that may be important in Fe3O4 biomineralization in-vivo.

In addition to imaging, AFM can be used to perform nanoscale manipulations of biological samples. In Figure 28 the AFM tips appears to have dissected the M. gryphiswaldense MSR-1 cell causing its internal cellular material to be released onto the substrate (Figure 28). While the cellular lysis was accidental here, this example illustrates the utility of using AFM-assisted micro-dissection to extract and manipulate intracellular structures, such as magnetosomes or inclusion bodies, which may not be accessible to the AFM tip as it is raster scanned across the surface of a live microorganism. As such, the AFM cantilever could be used to perform microdissection on a live microorganism and subsequently image these intracellular structures by AFM.

Yamamoto et al., (Yamamoto et al., 2010) actually demonstrated the utility of this technique when they used AFM-assisted microdissection to obtain images of magnetosomes adsorbed onto mica substrates (Yamamoto et al., 2010).

114

Chapter 6

Conclusions

Magnetotactic bacteria can be isolated from aquatic habitats using a relatively simple method. Using a magnet, the cells are first isolated from the sample container, which contains the sediment sample and water from the environment. This is then transferred to a racetrack that further enriches the sample. The isolated magnetotactic bacteria can be transferred to a microscope slide or a copper grid to be analyzed in the light microscope or transmission electron microscope, respectively. This allows the cells to be analyzed for magnetotaxis, cell morphology, and analysis of the magnetosomes.

Two atypical aquatic environments were investigated for the presence of magnetotactic bacteria. One site was Mickey Hot Springs, an arsenic-rich thermal spring from Oregon and the second was from Pavilion Lake, a freshwater lake in British

Columbia, Canada that contains microbialites. Using the racetrack method, magnetotactic bacteria were isolated from both of these locations. The transmission electron microscope was used to analyze the morphology of the cells and the magnetosomes. Mickey Hot Springs was dominated by rod/vibrioid-shaped bacteria that were 2.2 (± 0.6) µm long and 0.62 (± 0.1) µm wide. The cells contained had bullet- shaped crystals of magnetite that were 84 (± 17) nm long and 39 (± 8) nm wide and usually in a single chain. According to the 16S rRNA gene sequence analysis, the bacteria were from the Nitrospirae phylum. The Pavilion Lake sample was dominated by 11 5 spirillum-shaped bacteria that were 2.9 (± 0.6) µm long 0.34 (± 0.02) µm wide (n = 7) and contained magnetite crystals that were similar in length and width, 47 (± 4) nm long and 44 (± 5) nm wide. The magnetosomes were arranged in a single chain. The 16S rRNA analysis showed that the cells were from likely from the Alphaproteobacteria phylum.

Transmission electron microscopy and confocal microscopy were used to analyze the location of Mms-6 proteins of Magnetospirillum magneticum AMB-1. The TEM was used to analyze ultrathin sections of AMB-1 cells and also purified magnetosomes and the confocal microscope was used to analyze intact cells. The cells and magnetosomes were labeled with an antibody that had affinity for Mms-6 and then this antibody was labeled with a secondary antibody specific for the primary antibody. The secondary had a tag on it that allowed it to be visualized within the sample. It was determined that the

Mms-6 proteins occur all around the magnetosome, but not on every one. It is believed that the unlabeled magnetosomes do not contain any protein because the magnetite crystal is mature and is no longer being acted upon by the protein. Additionally, the protein was not localized on the cell membrane of cells, implying that the protein is not present before the magnetosome membrane forms. This investigation constrains the timing and location of the Mms-6 protein to the magnetosome membrane during the development of the magnetite crystal.

Atomic force microscopy (AFM) and transmission electron microscopy (TEM) were used to analyze two different types of magnetotactic bacteria, Magnetospirillum gryphiswaldense MSR-1 and Magnetospirillum magneticum AMB-1. In addition to this,

116 the purified magnetosomes and purified Mms6 proteins were also imaged. Both techniques offer nanoscale resolution, but have their own advantages and disadvantages.

The TEM can analyze intracellular structures, but the cells need to be fixed and sectioned. The sample preparation for the AFM is much less complicated than the TEM, but the imaging process is more time consuming and elaborate. However, even though I was able to visualize some internal structures using the AFM, they were not as resolved as the TEM images. The AFM has the ability to image small molecules at high resolution without the use of staining and the proteins can be imaged in fluid so they are closer to their native state. For example, Mms6 proteins were found to form oligomers similar to MamA proteins, which were 25 nm x 50 nm and 7 nm in height. Both techniques have their benefits and when used in conjunction can reveal more information about the magnetotactic bacteria.

117

References

Abreu, F., Martins, J. L., Silveira, T. S., Keim, C. N., de Barros, H. G. P. L., Filho, F. J. G., and Lins, U., 2007, 'Candidatus Magnetoglobus multicellularis&apos' a multicellular, magnetotactic prokaryote from a hypersaline environment: International Journal of Systematic and Evolutionary Microbiology, v. 57, no. 6, p. 1318-1322. Amann, R., Peplies, J., and Schüler, D., 2007, Diversity and taxonomy of magnetotactic bacteria: Magnetoreception and magnetosomes in bacteria, p. 25-36. Amemiya, Y., Arakaki, A., Staniland, S. S., Tanaka, T., and Matsunaga, T., 2007, Controlled formation of magnetite crystal by partial oxidation of ferrous hydroxide in the presence of recombinant magnetotactic bacterial protein Mms6: Biomaterials, v. 28, no. 35, p. 5381-5389. An, B., Liang, Q., and Zhao, D., 2011, Removal of arsenic(V) from spent ion exchange brine using a new class of starch-bridged magnetite nanoparticles: Water Research, v. 45, no. 5, p. 1961-1972. Arakaki, A., Masuda, F., Amemiya, Y., Tanaka, T., and Matsunaga, T., 2010, Control of the morphology and size of magnetite particles with peptides mimicking the Mms6 protein from magnetotactic bacteria: Journal of Colloid and Interface Science, v. 343, no. 1, p. 65-70. Arakaki, A., Webb, J., and Matsunaga, T., 2003, A Novel Protein Tightly Bound to Bacterial Magnetic Particles in Magnetospirillum magneticum Strain AMB-1: The Journal of Biological Chemistry, v. 278, no. 10, p. 8745-8750. Arató, B., Szányi, Z., Flies, C., Schüler, D., Frankel, R. B., Buseck, P. R., and Posfai, M., 2005, Crystal-size and shape distributions of magnetite from uncultured magnetotactic bacteria as a potential biomarker: American Mineralogist, v. 90, no. 8-9, p. 1233-1240. Balkwill, D. L., Maratea, D., and Blakemore, R. P., 1980, Ultrastructure of a magnetotactic spirillum.: Journal of Bacteriology, v. 141, no. 3, p. 1399-1408. Barton, L. L., and Fauque, G. D., 2009, Biochemistry, physiology and biotechnology of sulfate-reducing bacteria: Adv Appl Microbiol, v. 68, p. 41-98. Baumgartner, J., and Faivre, D., 2011, Magnetite biomineralization in bacteria: Molecular Biomineralization, p. 3-27. Bazylinski, D., and Williams, T., 2007, Ecophysiology of magnetotactic bacteria: Magnetoreception and magnetosomes in bacteria, p. 37-75.

118

Bazylinski, D. A., and Frankel, R. B., 2004, Magnetosome formation in prokaryotes: Nature Review Microbiology, v. 2, no. 3, p. 217-230. Bazylinski, D. A., Frankel, R. B., Heywood, B. R., Mann, S., King, J. W., Donaghay, P. L., and Hanson, A. K., 1995, Controlled Biomineralization of Magnetite (Fe3O4) and Greigite (Fe3S4) in a Magnetotactic Bacterium.: Applied and Environmental Microbiology, v. 61, no. 9, p. 3232-3239. Bazylinski, D. A., Frankel, R. B., and Konhauser, K. O., 2007, Modes of Biomineralization of Magnetite by Microbes: Geomicrobiology Journal, v. 24, no. 6, p. 465-475. Bazylinski, D. A., Heywood, B. R., Mann, S., and Frankel, R. B., 1993, Fe3O4 and Fe3S4 in a bacterium: Nature v. 366, no. 6452, p. 218-218. Bazylinski, D. A., Schlezinger, D. R., Howes, B. H., Frankel, R. B., and Epstein, S. S., 2000, Occurrence and distribution of diverse populations of magnetic protists in a chemically stratified coastal salt pond: Chemical Geology, v. 169, no. 3, p. 319- 328. Bellini, S., 1963, Ulteriori studi sui’batteri magnetosensibili’: Instituto di Mocrobiologia Dell’Università di Pavia. Benoit, M. R., Mayer, D., Barak, Y., Chen, I. Y., Hu, W., Cheng, Z., Wang, S. X., Spielman, D. M., Gambhir, S. S., and Matin, A., 2009, Visualizing Implanted Tumors in Mice with Magnetic Resonance Imaging Using Magnetotactic Bacteria: Clinical Cancer Research, v. 15, no. 16, p. 5170-5177. Benzerara, K., and Menguy, N., 2009, Looking for traces of life in minerals: Comptes Rendus Palevol, v. 8, no. 7, p. 617-628. Benzerara, K., Miot, J., Morin, G., Ona-Nguema, G., Skouri-Panet, F., and Férard, C., 2011, Significance, mechanisms and environmental implications of microbial biomineralization: Comptes Rendus Geoscience, v. 343, no. 2-3, p. 160-167. Blakemore, R., 1975, Magnetotactic bacteria: Science, v. 190, no. 4212, p. 377-379. Blakemore, R. P., 1982, Magnetotactic bacteria: Annual Review of Microbiology, v. 36, no. 1, p. 217-238. Blakemore, R. P., Maratea, D., and Wolfe, R. S., 1979, Isolation and pure culture of a freshwater magnetic spirillum in chemically defined medium.: Journal of Bacteriology, v. 140, no. 2, p. 720-729. Blakemore, R. P., Short, K. A., Bazylinski, D. A., Rosenblatt, C., and Frankel, R. B., 1985, Microaerobic conditions are required for magnetite formation within aquaspirillum magnetotacticum: Geomicrobiology Journal, v. 4, no. 1, p. 53-71. Bochicchio, B., Pepe, A., and Tamburro, A., 2001, On (GGLGY) synthetic repeating sequences of lamprin and analogous sequences: Matrix Biology, v. 20, no. 4, p. 243-250. Brady, A. L., Slater, G. F., Omelon, C. R., Southam, G., Druschel, G., Andersen, D. T., Hawes, I., Laval, B., and Lim, D. S. S., 2010, Photosynthetic isotope biosignatures in laminated micro-stromatolitic and non-laminated nodules associated with modern, freshwater microbialites in Pavilion Lake, B.C.: Chemical Geology, v. 274, no. 1-2, p. 56-67.

119

Burgess, J. G., Kawaguchi, R., Sakaguchi, T., Thornhill, R. H., and Matsunaga, T., 1993, Evolutionary relationships among Magnetospirillum strains inferred from phylogenetic analysis of 16S rDNA sequences.: Journal of Bacteriology, v. 175, no. 20, p. 6689-6694. Butler, R. F., and Banerjee, S. K., 1975, Theoretical single-domain grain size range in magnetite and titanomagnetite: Journal of Geophysical Research, v. 80, no. 29, p. 4049-4058. Cervantes, C., ji, G., Ramirez, J., and Silver, S., 1994, Resistance to arsenic compounds in microorganisms: FEMS Microbiology Reviews, v. 15, no. 4, p. 355-367. Chang, S. B. R., and Kirschvink, J. L., 1989, Magnetofossils, the magnetization of sediments, and the evolution of magnetite biomineralization: Annual Review of Earth and Planetary Science, v. 17, p. 169. Devouard, B., Posfai, M., Hua, X., Bazylinski, D. A., Frankel, R. B., and Buseck, P. R., 1998, Magnetite from magnetotactic bacteria; size distributions and twinning: American Mineralogist, v. 83, no. 11-12 Part 2, p. 1387-1398. Dittrich, M., and Luttge, A., 2008, Microorganisms, mineral surfaces, and aquatic environments: Learning from the past for future progress: Geobiology, v. 6, no. 3, p. 201-213. Edgar, R., 2004a, MUSCLE: a multiple sequence alignment method with reduced time and space complexity: BMC bioinformatics, v. 5, no. 1, p. 113. Edgar, R. C., 2004b, MUSCLE: multiple sequence alignment with high accuracy and high throughput: Nucleic Acids Research, v. 32, no. 5, p. 1792-1797. Edwards, K. J., Bach, W., and McCollom, T. M., 2005, Geomicrobiology in oceanography: microbe–mineral interactions at and below the seafloor: TRENDS in Microbiology, v. 13, no. 9, p. 449-456. Esquivel, D. M. S., and de Barros, H. G. P. L., 1986, Motion of magnetotactic microorganisms: Journal of Experimental Biology, v. 121, no. 1, p. 153-163. Faivre, D., Böttger, L. H., Matzanke, B. F., and Schüler, D., 2007, Intracellular Magnetite Biomineralization in Bacteria Proceeds by a Distinct Pathway Involving Membrane-Bound Ferritin and an Iron(II) Species: Angewandte Chemie International Edition, v. 46, no. 44, p. 8495-8499. Faivre, D., Menguy, N., Posfai, M., and Schüler, D., 2008, Environmental parameters affect the physical properties of fast-growing magnetosomes: American Mineralogist, v. 93, no. 2-3, p. 463-469. Faivre, D., and Schüler, D., 2008, Magnetotactic bacteria and magnetosomes: Chemical reviews, v. 108, no. 11, p. 4875. Farina, M., Esquivel, D. M. S., and de Barros, H. G. P. L., 1990, Magnetic iron-sulphur crystals from a magnetotactic microorganism: Nature, v. 343, no. 6255, p. 256- 258. Ferris, F. G., Thompson, J. B., and Beveridge, T. J., 1997, Modern freshwater microbialites from Kelly Lake, British Columbia, Canada: Palaios, p. 213-219. Flies, C. B., Jonkers, H. M., Beer, D., Bosselmann, K., Bottcher, M. E., and Schüler, D., 2005a, Diversity and vertical distribution of magnetotactic bacteria along

120

chemical gradients in freshwater microcosms: FEMS Microbiology Ecology, v. 52, no. 2, p. 185-195. Flies, C. B., Peplies, J., and Schüler, D., 2005b, Combined Approach for Characterization of Uncultivated Magnetotactic Bacteria from Various Aquatic Environments: Applied and Environmental Microbiology, v. 71, no. 5, p. 2723-2731. Frankel, R., Williams, T., and Bazylinski, D., 2007, Magneto-aerotaxis: Magnetoreception and magnetosomes in bacteria, p. 1-24. Frankel, R. B., 1984, Magnetic guidance of organisms: Annual Review of Biophysics and Bioengineering, v. 13, no. 1, p. 85-103. Frankel, R. B., and Bazylinski, D. A., 2004, Magnetosome mysteries: ASM News, v. 70, p. 176-183. Frankel, R. B., Bazylinski, D. A., Johnson, M. S., and Taylor, B. L., 1997, Magneto- aerotaxis in marine coccoid bacteria: Biophysical Journal, v. 73, no. 2, p. 994- 1000. Frankel, R. B., Blakemore, R. P., and Wolfe, R. S., 1979, Magnetite in freshwater magnetotactic bacteria: Science, v. 203, no. 4387, p. 1355. Frankel, R. B., Papaefthymiou, G. C., Blakemore, R. P., and Oapos;Brien, W., 1983, Fe3O4 Precipitation in Magnetotactic Bacteria: Biochimica et biophysica acta, v. 763, no. 2, p. 147-159. Galloway, J. M., Arakaki, A., Masuda, F., Tanaka, T., Matsunaga, T., and Staniland, S. S., 2011, Magnetic bacterial protein Mms6 controls morphology, crystallinity and magnetism of cobalt-doped magnetite nanoparticles in vitro: Journal of Materials Chemistry, v. 21, no. 39, p. 15244. Gorby, Y. A., Beveridge, T. J., and Blakemore, R. P., 1988, Characterization of the bacterial magnetosome membrane.: Journal of Bacteriology, v. 170, no. 2, p. 834- 841. Gould, J. L., Kirschvink, J., and Deffeyes, K., 1978, Bees have magnetic remanence: Science (New York, NY), v. 201, no. 4360, p. 1026. Greenberg, M., Canter, K., Mahler, I., and Tornheim, A., 2005, Observation of Magnetoreceptive Behavior in a Multicellular Magnetotactic Prokaryote in Higher than Geomagnetic Fields: Biophysical Journal, v. 88, no. 2, p. 1496-1499. Grünberg, K., Müller, E. C., Otto, A., Reszka, R., Linder, D., Kube, M., Reinhardt, R., and Schüler, D., 2004, Biochemical and proteomic analysis of the magnetosome membrane in Magnetospirillum gryphiswaldense: Applied and environmental microbiology, v. 70, no. 2, p. 1040-1050. Hernández-Hernández, A., Rodríguez-Navarro, A., Gómez-Morales, J., Jiménez-López, C., Nys, Y., and García-Ruiz, J. M., 2008, Influence of model globular proteins with different isoelectric points on the precipitation of calcium carbonate: Crystal Growth and Design, v. 8, no. 5, p. 1495-1502. Heyen, U., and Schüler, D., 2003, Growth and magnetosome formation by microaerophilic Magnetospirillum strains in an oxygen-controlled fermentor: Applied microbiology and biotechnology, v. 61, no. 5, p. 536-544.

121

Heywood, B. R., Bazylinski, D. A., Garratt-Reed, A., Mann, S., and Frankel, R. B., 1990, Controlled biosynthesis of greigite (Fe3S4) in magnetotactic bacteria: Naturwissenschaften, v. 77, no. 11, p. 536-538. Hindmarsh, J. T., McCurdy, R. F., and Savory, J., 1986, Clinical and Environmental Aspects of Arsenic Toxicity: Critical Reviews in Clinical Laboratory Sciences, v. 23, no. 4, p. 315-347. Hochella Jr, M. F., Lower, S. K., Maurice, P. A., Penn, R. L., Sahai, N., Sparks, D. L., and Twining, B. S., 2008, Nanominerals, mineral nanoparticles, and Earth systems: science, v. 319, no. 5870, p. 1631-1635. Isambert, A., Menguy, N., Larquet, E., Guyot, F., and Valet, J. P., 2007, Transmission electron microscopy study of magnetites in a freshwater population of magnetotactic bacteria: American Mineralogist, v. 92, no. 4, p. 621-630. Jogler, C., Kube, M., Schübbe, S., Ullrich, S., Teeling, H., Bazylinski, D. A., Reinhardt, R., and Schüler, D., 2009a, Comparative analysis of magnetosome gene clusters in magnetotactic bacteria provides further evidence for horizontal gene transfer: Environmental Microbiology, v. 11, no. 5, p. 1267-1277. Jogler, C., Lin, W., Meyerdierks, A., Kube, M., Katzmann, E., Flies, C., Pan, Y., Amann, R., Reinhardt, R., and Schüler, D., 2009b, Toward cloning of the magnetotactic metagenome: identification of magnetosome island gene clusters in uncultivated magnetotactic bacteria from different aquatic sediments: Applied and environmental microbiology, v. 75, no. 12, p. 3972-3979. Jogler, C., Niebler, M., Lin, W., Kube, M., Wanner, G., Kolinko, S., Stief, P., Beck, A. J., de Beer, D., Petersen, N., Pan, Y., Amann, R., Reinhardt, R., and Schüler, D., 2010, Cultivation-independent characterization of ‘Candidatus Magnetobacterium bavaricum’ via ultrastructural, geochemical, ecological and metagenomic methods: Environmental Microbiology, v. 12, no. 9, p. 2466-2478. Jogler, C., and Schüler, D., 2007, Genetic analysis of magnetosome biomineralization, p. 133-161. Jogler, C., and Schüler, D., 2009, Genomics, Genetics, and Cell Biology of Magnetosome Formation: Annual Reviews of Microbiology, v. 63, no. 1, p. 501-521. Jogler, C., Wanner, G., Kolinko, S., Niebler, M., Amann, R., Petersen, N., Kube, M., Reinhardt, R., and Schüler, D., 2011, Conservation of proteobacterial magnetosome genes and structures in an uncultivated member of the deep- branching Nitrospira phylum: Proceedings of the National Academy of Science, v. 108, no. 3, p. 1134-1139. Kawaguchi, R., Burgess, J. G., Sakaguchi, T., Takeyama, H., Thornhill, R. H., and Matsunaga, T., 1995, Phylogenetic analysis of a novel sulfate-reducing magnetic bacterium, RS-1, demonstrates its membership of the δ-Proteobacteria: FEMS Microbiology Letters, v. 126, no. 3, p. 277-282. Keim, C., Lopes Martins, J., Lins de Barros, H., Lins, U., and Farina, M., 2007a, Structure, behavior, ecology and diversity of multicellular magnetotactic prokaryotes: Magnetoreception and magnetosomes in bacteria, p. 103-132.

122

Keim, C., Lopes Martins, J., Lins de Barros, H., Lins, U., and Farina, M., 2007b, Structure, behavior, ecology and diversity of multicellular magnetotactic prokaryotes, p. 103-132. Keim, C. N., Lins, U., and Farina, M., 2001, Elemental analysis of uncultured magnetotactic bacteria exposed to heavy metals: Canadian Journal of Microbiology, v. 47, no. 12, p. 1132-1136. Kennard, J. M., and James, N. P., 1986, Thrombolites and stromatolites: two distinct types of microbial structures: Palaios, p. 492-503. Kienberger, F., Mueller, H., Pastushenko, V., and Hinterdorfer, P., 2004, Following single antibody binding to purple membranes in real time: EMBO reports, v. 5, no. 6, p. 579-583. Kirschvink, J. L., and Chang, S. B. R., 1984, Ultrafine-grained magnetite in deep-sea sediments: Possible bacterial magnetofossils: Geology, v. 12, no. 9, p. 559-562. Kirschvink, J. L., and Lowenstam, H. A., 1979, Mineralization and magnetization of chiton teeth: Paleomagnetic, sedimentologic, and biologic implications of organic magnetite: Earth and Planetary Science Letters, v. 44, no. 2, p. 193-204. Kirschvink, J. L., Walker, M. M., and Diebel, C. E., 2001, Magnetite-based magnetoreception: Current Opinion in Neurobiology, v. 11, no. 4, p. 462-467. Kobayashi, A., Kirschvink, J. L., Nash, C. Z., Kopp, R. E., Sauer, D. A., Bertani, L. E., Voorhout, W. F., and Taguchi, T., 2006, Experimental observation of magnetosome chain collapse in magnetotactic bacteria: Sedimentological, paleomagnetic, and evolutionary implications: Earth and Planetary Science Letters, v. 245, no. 3-4, p. 538-550. Komeili, A., Li, Z., Newman, D. K., and Jensen, G. J., 2006, Magnetosomes Are Cell Membrane Invaginations Organized by the Actin-Like Protein MamK: Science, v. 311, no. 5758, p. 242-245. Komeili, A., Vali, H., Beveridge, T. J., and Newman, D. K., 2004, Magnetosome vesicles are present before magnetite formation, and MamA is required for their activation: Proceedings of the National Academy of Science, v. 101, no. 11, p. 3839-3844. Kopp, R., Weiss, B., Maloof, A., Vali, H., Nash, C., and Kirschvink, J., 2006, Chains, clumps, and strings: Magnetofossil taphonomy with ferromagnetic resonance spectroscopy: Earth and Planetary Science Letters, v. 247, no. 1-2, p. 10-25. Kopp, R. E., and Kirschvink, J. L., 2008, The identification and biogeochemical interpretation of fossil magnetotactic bacteria: Earth Science Reviews, v. 86, no. 1-4, p. 42-61. Kopp, R. E., Raub, T. D., Schumann, D., Vali, H., Smirnov, A. V., and Kirschvink, J. L., 2007, Magnetofossil spike during the Paleocene-Eocene thermal maximum: Ferromagnetic resonance, rock magnetic, and electron microscopy evidence from Ancora, New Jersey, United States: Proceedings of the National Academy of Science, v. 22, no. 4, p. PA4103. Koski, A. K., and Wood, S. A., 2004, The geochemistry of geothermal waters in the Alvord Basin, southeastern Oregon: Proceedings of the 11th International Symposium on Water–Rock Interaction, Energy and Environment, Saratoga 123

Springs, New York, Lawrence Livermore National Laboratory, Livermore, California. Laval, B., Cady, S. L., Pollack, J. C., McKay, C. P., Bird, J. S., Grotzinger, J. P., Ford, D. C., and Bohm, H. R., 2000, Modern freshwater microbialite analogues for ancient dendritic reef structures: Nature, v. 407, no. 6804, p. 626-629. Ledbetter, R. N., Connon, S. A., Neal, A. L., Dohnalkova, A., and Magnuson, T. S., 2007, Biogenic Mineral Production by a Novel Arsenic-Metabolizing Thermophilic Bacterium from the Alvord Basin, Oregon: Applied and Environmental Microbiology, v. 73, no. 18, p. 5928-5936. Lefèvre, C., Abreu, F., Schmidt, M. L., Lins, U., Frankel, R. B., Hedlund, B. P., and Bazylinski, D. A., 2010, Moderately Thermophilic Magnetotactic Bacteria from Hot Springs in Nevada: Applied and Environmental Microbiology, v. 76, no. 11, p. 3740-3743. Lefèvre, C., Menguy, N., Abreu, F., Lins, U., Posfai, M., Prozorov, T., Pignol, D., Frankel, R. B., and Bazylinski, D. A., 2011, A Cultured Greigite-Producing Magnetotactic Bacterium in a Novel Group of Sulfate-Reducing Bacteria: Science, v. 334, no. 6063, p. 1720-1723. Lefèvre, C. T., Pósfai, M., Abreu, F., Lins, U., Frankel, R. B., and Bazylinski, D. A., 2011a, Morphological features of elongated-anisotropic magnetosome crystals in magnetotactic bacteria of the Nitrospirae phylum and the Deltaproteobacteria class: Earth and Planetary Science Letters, v. 312, no. 1-2, p. 194-200. Lefèvre, C. T., Viloria, N., Schmidt, M. L., sfai, M. a. l. P. o., Frankel, R. B., and Bazylinski, D. A., 2011b, Novel magnetite-producing magnetotactic bacteria belonging to the Gammaproteobacteria: International Society for Microbial Ecology, v. 6, no. 2, p. 440-450. Li, J., and Pan, Y., 2012, Environmental Factors Affect Magnetite Magnetosome Synthesis in Magnetospirillum magneticumAMB-1: Implications for Biologically Controlled Mineralization: Geomicrobiology Journal, v. 29, no. 4, p. 362-373. Li, J., Pan, Y., Liu, Q., Yu-Zhang, K., Menguy, N., Che, R., Qin, H., Lin, W., Wu, W., Petersen, N., and Yang, X., 2010, Biomineralization, crystallography and magnetic properties of bullet-shaped magnetite magnetosomes in giant rod magnetotactic bacteria: Earth and Planetary Science Letters, v. 293, no. 3-4, p. 368-376. Lim, D. S. S., Laval, B. E., Slater, G., Antoniades, D., Forrest, A. L., Pike, W., Pieters, R., Saffari, M., Reid, D., and Schulze-Makuch, D., 2009, Limnology of Pavilion Lake, BC, Canada Characterization of a microbialite forming environment: Fundamental and Applied Limnology, v. 173, no. 4, p. 329-351. Lin, W., Li, J., and Pan, Y., 2012, Newly Isolated but Uncultivated Magnetotactic Bacterium of the Phylum Nitrospirae from Beijing, China: Applied and Environmental Microbiology, v. 78, no. 3, p. 668-675. Lin, W., Li, J., Schüler, D., Jogler, C., and Pan, Y., 2009, Diversity analysis of magnetotactic bacteria in Lake Miyun, northern China, by restriction fragment length polymorphism: Systematic and Applied Microbiology, v. 32, no. 5, p. 342- 350. 124

Lin, W., and Pan, Y., 2009, Specific primers for the detection of freshwater alphaproteobacterial magnetotactic cocci.: International Microbiology, v. 12, no. 4, p. 237-242. Lin, W., and Pan, Y., 2010, Temporal variation of magnetotactic bacterial communities in two freshwater sediment microcosms: FEMS Microbiology Letters, v. 302, no. 1, p. 85-92. Lin, W., Tian, L., Li, J., and Pan, Y., 2008, Does capillary racetrack-based enrichment reflect the diversity of uncultivated magnetotactic cocci in environmental samples?: FEMS Microbiology Letters, v. 279, no. 2, p. 202-206. Lins, U., Freitas, F., Keim, C. N., Barros, H. L., Esquivel, D. M. S., and Farina, M., 2003, Simple homemade apparatus for harvesting uncultured magnetotactic microorganisms: Brazilian Journal of Microbiology, v. 34, no. 2, p. 111-116. Lins, U., Keim, C. N., Evans, F. F., Farina, M., and Buseck, P. R., 2007, Magnetite (Fe3O4) and Greigite (Fe3S4) Crystals in Multicellular Magnetotactic Prokaryotes: Geomicrobiology Journal, v. 24, no. 1, p. 43-50. Lohße, A., Ullrich, S., Katzmann, E., Borg, S., Wanner, G., Richter, M., Voigt, B., Schweder, T., and Schüler, D., 2011, Functional Analysis of the Magnetosome Island in Magnetospirillum gryphiswaldense: The mamAB Operon Is Sufficient for Magnetite Biomineralization: Plos One, v. 6, no. 10, p. e25561. Madsen, E. L., 2011, Microorganisms and their roles in fundamental biogeochemical cycles: Current opinion in biotechnology, v. 22, no. 3, p. 456-464. Mann, S., Sparks, N., Walker, M. M., and Kirschvink, J. L., 1988a, Ultrastructure, morphology and organization of biogenic magnetite from sockeye salmon, Oncorhynchus nerka: implications for magnetoreception: Journal of Experimental Biology, v. 140, no. 1, p. 35-49. Mann, S., Sparks, N. H., Walker, M. M., and Kirschvink, J. L., 1988b, Ultrastructure, morphology and organization of biogenic magnetite from sockeye salmon, Oncorhynchus nerka: implications for magnetoreception: Journal of Experimental Biology, v. 140, no. 1, p. 35-49. Mann, S., Sparks, N. H. C., and Blakemore, R. P., 1987, Ultrastructure and Characterization of Anisotropic Magnetic Inclusions in Magnetotactic Bacteria, v. 231, no. 1265, p. 469-476. Matsunaga, T., and Arakaki, A., 2007, Molecular bioengineering of bacterial magnetic particles for biotechnological applications: Magnetoreception and Magnetosomes in Bacteria, p. 227-254. Matsunaga, T., Okamura, Y., Fukuda, Y., Wahyudi, A. T., Murase, Y., and Takeyama, H., 2005, Complete Genome Sequence of the Facultative Anaerobic Magnetotactic Bacterium Magnetospirillum sp. strain AMB-1: DNA Research, v. 12, no. 3, p. 157-166. Matsunaga, T., Sakaguchi, T., and Tadakoro, F., 1991, Magnetite formation by a magnetic bacterium capable of growing aerobically: Applied Microbiology and Biotechnology, v. 35, no. 5, p. 651-655. McKay, D. S., Gibson Jr, E. K., Thomas-Keprta, K. L., Vali, H., Romanek, C. S., Clemett, S. J., Chillier, X. D. F., Maechling, C. R., and Zare, R. N., 1996, Search 125

for Past Life on Mars: Possible Relic Biogenic Activity in Martian Meteorite ALH84001: Science, v. 273, p. 16. Meldrum, F. C., Mann, S., Heywood, B. R., Frankel, R. B., and Bazylinski, D. A., 1993, Electron Microscopy Study of Magnetosomes in Two Cultured Vibrioid Magnetotactic Bacteria: Proceedings of the Royal Society of London, v. 251, no. 1332, p. 237-242. Moench, T. T., and Konetzka, W., 1978, A novel method for the isolation and study of a magnetotactic bacterium: Archives of Microbiology, v. 119, no. 2, p. 203-212. Murat, D., Falahati, V., Bertinetti, L., Csencsits, R., Körnig, A., Downing, K., Faivre, D., and Komeili, A., 2012, The magnetosome membrane protein, MmsF, is a major regulator of magnetite biomineralization in Magnetospirillum magneticum AMB- 1: Molecular Microbiology, v. 85, no. 4, p. 684-699. Murat, D., Quinlan, A., Vali, H., and Komeili, A., 2010, Comprehensive genetic dissection of the magnetosome gene island reveals the step-wise assembly of a prokaryotic organelle: Proceedings of the National Academy of Science, v. 107, no. 12, p. 5593-5598. Nash, C., 2008, Mechanisms and Evolution of Magnetotactic Bacteria [PhD Dissertation]: California Institute of Technology, 75 p. Oestreicher, Z. W. J., 2004, Geobiological Investigation of Mickey Hot Springs, Southeastern Oregon [Master's Thesis]: Portland State University, 171 p. Okuda, Y., and Fukumori, Y., 2001, Expression and characterization of a magnetosome- associated protein, TPR-containing MAM22, in Escherichia coli: FEBS Letters, v. 491, no. 3, p. 169-173. Petersen, N., von Dobeneck, T., and Vali, H., 1986, Fossil bacterial magnetite in deep-sea sediments from the South Atlantic Ocean: Nature, vol. 320, no. 17, p. 611-615. Pradel, N., Santini, C. L., Bernadac, A., Fukumori, Y., and Wu, L. F., 2006, Biogenesis of actin-like bacterial cytoskeletal filaments destined for positioning prokaryotic magnetic organelles: Proceedings of the National Academy of Science, v. 103, no. 46, p. 17485-17489. Prozorov, T., Mallapragada, S. K., Narasimhan, B., Wang, L., Palo, P., Nilsen-Hamilton, M., Williams, T. J., Bazylinski, D. A., Prozorov, R., and Canfield, P. C., 2007a, Protein-Mediated Synthesis of Uniform Superparamagnetic Magnetite Nanocrystals: Advanced Functional Materials, v. 17, no. 6, p. 951-957. Prozorov, T., Palo, P., Wang, L., Nilsen-Hamilton, M., Jones, D., Orr, D., Mallapragada, S. K., Narasimhan, B., Canfield, P. C., and Prozorov, R., 2007b, Cobalt Ferrite Nanocrystals: Out-Performing Magnetotactic Bacteria: ACS Nano, v. 1, no. 3, p. 228-233. Ratnaike, R. N., 2003, Acute and chronic arsenic toxicity: Postgraduate Medical Journal, v. 79, no. 933, p. 391-396. Richter, M., Kube, M., Bazylinski, D. A., Lombardot, T., Glockner, F. O., Reinhardt, R., and Schüler, D., 2007, Comparative Genome Analysis of Four Magnetotactic Bacteria Reveals a Complex Set of Group-Specific Genes Implicated in Magnetosome Biomineralization and Function: Journal of Bacteriology, v. 189, no. 13, p. 4899-4910. 126

Rodgers, F. G., Blakemore, R. P., Blakemore, N. A., Frankel, R. B., Bazylinski, D. A., Maratea, D., and Rodgers, C., 1990, Intercellular structure in a many-celled magnetotactic prokaryote: Archives of Microbiology, v. 154, no. 1, p. 18-22. Sakaguchi, T., Arakaki, A., and Matsunaga, T., 2002, Desulfovibrio magneticus sp. nov., a novel sulfate-reducing bacterium that produces intracellular single-domain-sized magnetite particles.: International Journal of Systematic and Evolutionary Microbiology, v. 52, no. 1, p. 215-221. San Paulo, A., and Garcia, R., 2000, High-resolution imaging of antibodies by tapping- mode atomic force microscopy: attractive and repulsive tip-sample interaction regimes: Biophysical Journal, v. 78, no. 3, p. 1599-1605. Scheffel, A., Gardes, A., Grunberg, K., Wanner, G., and Schüler, D., 2007, The Major Magnetosome Proteins MamGFDC Are Not Essential for Magnetite Biomineralization in Magnetospirillum gryphiswaldense but Regulate the Size of Magnetosome Crystals: Journal of Bacteriology, v. 190, no. 1, p. 377-386. Scheffel, A., Gruska, M., Faivre, D., Linaroudis, A., Plitzko, J. M., and Schüler, D., 2005, An acidic protein aligns magnetosomes along a filamentous structure in magnetotactic bacteria: Nature, v. 440, no. 7080, p. 110-114. Scheffel, A., and Schüler, D., 2007, The Acidic Repetitive Domain of the Magnetospirillum gryphiswaldense MamJ Protein Displays Hypervariability but Is Not Required for Magnetosome Chain Assembly: Journal of Bacteriology, v. 189, no. 17, p. 6437-6446. Schopf, J. W., 1993, Microfossils of the Early Archean Apex chert: new evidence of the antiquity of life: Science, v. 260, no. 5108, p. 640-646. Schübbe, S., Kube, M., Scheffel, A., Wawer, C., Heyen, U., Meyerdierks, A., Madkour, M. H., Mayer, F., Reinhardt, R., and Schüler, D., 2003, Characterization of a spontaneous nonmagnetic mutant of Magnetospirillum gryphiswaldense reveals a large deletion comprising a putative magnetosome island: Journal of Bacteriology, v. 185, no. 19, p. 5779-5790. Schüler, D., 2008, Genetics and cell biology of magnetosome formation in magnetotactic bacteria: FEMS Microbiology Reviews, v. 32, no. 4, p. 654-672. Schüler, D., and Frankel, R. B., 1999, Bacterial magnetosomes: microbiology, biomineralization and biotechnological applications: Applied Microbiology and Biotechnology, v. 52, no. 4, p. 464-473. Schüler, D., Spring, S., and Bazylinski, D. A., 1999, Improved Technique for the Isolation of Magnetotactic Spirilla from a Freshwater Sediment and their Phylogenetic Characterization: Systematic and Applied Microbiology, v. 22, no. 3, p. 466-471. Silverton, E., Navia, M. A., and Davies, D. R., 1977, Three-dimensional structure of an intact human immunoglobulin: Proceedings of the National Academy of Sciences, v. 74, no. 11, p. 5140-5144. Simmons, S., and Edwards, K., 2007a, Geobiology of magnetotactic bacteria: Magnetoreception and magnetosomes in bacteria, p. 77-102. Simmons, S., and Edwards, K., 2007b, Geobiology of magnetotactic bacteria, p. 77-102.

127

Simmons, S. L., Bazylinski, D. A., and Edwards, K. J., 2006, South-seeking magnetotactic bacteria in the Northern Hemisphere: Science, v. 311, no. 5759, p. 371-374. Simmons, S. L., Bazylinski, D. A., and Edwards, K. J., 2007, Population dynamics of marine magnetotactic bacteria in a meromictic salt pond described with qPCR: Environmental Microbiology, v. 9, no. 9, p. 2162-2174. Simmons, S. L., Sievert, S. M., Frankel, R. B., Bazylinski, D. A., and Edwards, K. J., 2004, Spatiotemporal distribution of marine magnetotactic bacteria in a seasonally stratified coastal salt pond: Applied and Environmental Microbiology, v. 70, no. 10, p. 6230-6239. Sneath, P. H. A., and Sokal, R. R., 1973, Numerical taxonomy. The principles and practice of numerical classification. Spring, S., Amann, R., Ludwig, W., Schleifer, K.-H., Schüler, D., Poralla, K., and Petersen, N., 1994, Phylogenetic Analysis of Uncultured Magnetotactic Bacteria from the Alpha-Subclass of Proteobacteria: Systematic and Applied Microbiology, v. 17, no. 4, p. 501-508. Spring, S., Amann, R., Ludwig, W., Schleifer, K. H., Van Gemerden, H., and Petersen, N., 1993, Dominating role of an unusual magnetotactic bacterium in the microaerobic zone of a freshwater sediment: Applied and Environmental Microbiology, v. 59, no. 8, p. 2397-2403. Spring, S., and Bazylinski, D. A., 2006, Magnetotactic bacteria: Prokaryotes, v. 2, p. 842- 862. Spring, S., Lins, U., Amann, R., Schleifer, K. H., Ferreira, L. C. S., Esquivel, D. M. S., and Farina, M., 1998, Phylogenetic affiliation and ultrastructure of uncultured magnetic bacteria with unusually large magnetosomes: Archives of Microbiology, v. 169, no. 2, p. 136-147. Spring, S., and Schleifer, K.-H., 1995, Diversity of Magnetotactic Bacteria: Systematic and Applied Microbiology, v. 18, no. 2, p. 147-153. St. John, A., 1993, Hydrogeochemical Characterization of the Alvord Valley Known Geothermal Resources Area, Harney County, Oregon [Master's: Portland State University, 177 p. Stolz, J. F., 1993, Magnetosomes: Journal of General Microbiology, v. 139, no. 8, p. 1663-1670. Sudo, S., Fujikawa, T., Nagakura, T., Ohkubo, T., Sakaguchi, K., Tanaka, M., Nakashima, K., and Takahashi, T., 1997, Structures of mollusc shell framework proteins: Nature, v. 387, p. 563-564. Tamura, K., Nei, M., and Kumar, S., 2004, Prospects for inferring very large phylogenies by using the neighbor-joining method: Proceedings of the National Academy of Sciences of the United States of America, v. 101, no. 30, p. 11030-11035. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., and Kumar, S., 2011, MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods: Molecular biology and evolution, v. 28, no. 10, p. 2731-2739.

128

Tanaka, M., Arakaki, A., and Matsunaga, T., 2010a, Identification and functional characterization of liposome tubulation protein from magnetotactic bacteria: Molecular Microbiology, v. 76, no. 2, p. 480-488. Tanaka, M., Arakaki, A., Staniland, S. S., and Matsunaga, T., 2010b, Simultaneously Discrete Biomineralization of Magnetite and Tellurium Nanocrystals in Magnetotactic Bacteria: Applied and Environmental Microbiology, v. 76, no. 16, p. 5526-5532. Tanaka, M., Mazuyama, E., Arakaki, A., and Matsunaga, T., 2011, MMS6 Protein Regulates Crystal Morphology during Nano-sized Magnetite Biomineralization in Vivo: Journal of Biological Chemistry, v. 286, no. 8, p. 6386-6392. Tanaka, M., Okamura, Y., Arakaki, A., Tanaka, T., Takeyama, H., and Matsunaga, T., 2006, Origin of magnetosome membrane: Proteomic analysis of magnetosome membrane and comparison with cytoplasmic membrane: Proteomics, v. 6, no. 19, p. 5234-5247. Tang, Y.-S., Wang, D., Zhou, C., Ma, W., Zhang, Y.-Q., Liu, B., and Zhang, S., 2011, Bacterial magnetic particles as a novel and efficient gene vaccine delivery system: Gene Therapy, p. 1-9. Taoka, A., Asada, R., Wu, L. F., and Fukumori, Y., 2007, Polymerization of the Actin- Like Protein MamK, Which Is Associated with Magnetosomes: Journal of Bacteriology, v. 189, no. 23, p. 8737-8740. Taoka, A., Umeyama, C., and Fukumori, Y., 2008, Identification of Iron Transporters Expressed in the Magnetotactic Bacterium Magnetospirillum magnetotacticum: Current Microbiolgy, v. 58, no. 2, p. 177-181. Tarduno, J. A., Cottrell, R. D., Watkeys, M. K., Hofmann, A., Doubrovine, P. V., Mamajek, E. E., Liu, D., Sibeck, D. G., Neukirch, L. P., and Usui, Y., 2010, Geodynamo, solar wind, and magnetopause 3.4 to 3.45 billion years ago: science, v. 327, no. 5970, p. 1238-1240. Tartaj, P., Morales, M. P., González-Carreño, T., Veintemillas-Verdaguer, S., and Serna, C. J., 2005, Advances in magnetic nanoparticles for biotechnology applications: Journal of Magnetism and Magnetic Materials, v. 290-291, p. 28-34. Thomas-Keprta, K. L., Bazylinski, D. A., Kirschvink, J. L., Clemett, S. J., McKay, D. S., Wentworth, S. J., Vali, H., Gibson Jr, E. K., and Romanek, C. S., 2000, Elongated prismatic magnetite crystals in ALH84001 carbonate globules:: Potential Martian magnetofossils: Geochimica et Cosmochimica Acta, v. 64, no. 23, p. 4049-4081. Thomas-Keprta, K. L., Clemett, S. J., Bazylinski, D. A., Kirschvink, J. L., McKay, D. S., Wentworth, S. J., Vali, H., Gibson Jr, E. K., and Romanek, C. S., 2002, Magnetofossils from ancient Mars: a robust biosignature in the Martian meteorite ALH84001: Applied and Environmental Microbiology, v. 68, no. 8, p. 3663- 3672. Thomas-Keprta, K. L., Clemett, S. J., McKay, D. S., Gibson, E. K., and Wentworth, S. J., 2009, Origins of magnetite nanocrystals in Martian meteorite ALH84001: Geochimica et Cosmochimica Acta, v. 73, no. 21, p. 6631-6677.

129

Thornhill, R. H., Grant Burgess, J., Sakaguchi, T., and Matsunaga, T., 1994, A morphological classification of bacteria containing bullet-shaped magnetic particles: FEMS Microbiology Letters, v. 115, no. 2, p. 169-176. Ullrich, S., Kube, M., Schübbe, S., Reinhardt, R., and Schüler, D., 2005, A hypervariable 130-kilobase genomic region of Magnetospirillum gryphiswaldense comprises a magnetosome island which undergoes frequent rearrangements during stationary growth: Journal of Bacteriology, v. 187, no. 21, p. 7176-7184. Vali, H., Förster, O., Amarantidis, G., and Petersen, N., 1987, Magnetotactic bacteria and their magnetofossils in sediments: Earth and Planetary Science Letters, v. 86, no. 2, p. 389-400. Walcott, C., Gould, J. L., and Kirschvink, J. L., 1979, Pigeons have magnets.: Science. Wang, L., 2011, Studies of the structure and function of Mms6, a bacterial protein that promotes the formation of magnetic nanoparticles [PhD Dissertation]: Iowa State University, 126 p. Wang, L., Prozorov, T., Palo, P. E., Liu, X., Vaknin, D., Prozorov, R., Mallapragada, S., and Nilsen-Hamilton, M., 2012a, Self-Assembly and Biphasic Iron-Binding Characteristics of Mms6, A Bacterial Protein That Promotes the Formation of Superparamagnetic Magnetite Nanoparticles of Uniform Size and Shape: Biomacromolecules, v. 13, no. 1, p. 98-105. Wang, W., Bu, W., Wang, L., Palo, P. E., Mallapragada, S., Nilsen-Hamilton, M., and Vaknin, D., 2012b, Interfacial Properties and Iron Binding to Bacterial Proteins That Promote the Growth of Magnetite Nanocrystals: X-ray Reflectivity and Surface Spectroscopy Studies: Langmuir, v. 28, no. 9, p. 4274-4282. Weber, K. A., Achenbach, L. A., and Coates, J. D., 2006, Microorganisms pumping iron: anaerobic microbial iron oxidation and reduction: Nature Reviews Microbiology, v. 4, no. 10, p. 752-764. Wolfe, R., Thauer, R., and Pfennig, N., 1987, A 'capillary racetrack' method for isolation of magnetotactic bacteria: FEMS Microbiology Ecology, v. 45, no. 1, p. 31-35. Wood, S. A., and Shannon, W. M., 2003, Rare-earth elements in geothermal waters from Oregon, Nevada, and California: Journal of Solid State Chemistry, v. 171, no. 1-2, p. 246-253. Yamamoto, D., Taoka, A., Uchihashi, T., Sasaki, H., Watanabe, H., Ando, T., and Fukumori, Y., 2010, Visualization and structural analysis of the bacterial magnetic organelle magnetosome using atomic force microscopy: Proceedings of the National Academy of Science, v. 107, no. 20, p. 9382-9387. Yongsunthon, R., and Lower, S. K., 2005, Force Measurements Between a Bacterium and Another Surface In Situ: Advances in applied microbiology, v. 58, p. 97-124.

130

Appendix A: Additional Methods and Data for Chapter 3

Figure 33 Field site for Mickey Hot Springs. Upper picture is the view of the main pool (Morning Glory) with a narrow channel coming out of the pool toward the bottom of the picture (red arrow) to a smaller pool where the sample was collected (white arrow). The

131 picture is a detail of where the sample was collected. The blue arrow points to the 1 L container used to collect the sample from the small pool.

132

Table 5 Table showing the measurements of the lengths and width of the cells from Mickey Hot Springs. Length (µm) Width (µm) 2553.4 572.2 3008.7 646.0 1971.7 611.8 2236.1 569.5 2188.3 583.6 1426.9 367.8 2529.8 582.9 2557.8 673.7 1799.9 623.4 2234.1 600.0 2342.7 676.9 1879.7 555.1 1971.1 597.7 2308.5 551.6 3010.4 944.1 2463.4 607.7 2236.5 466.7 2254.8 468.9 1719.7 487.1 2050.1 514.3 2006.0 520.8 2350.0 757.8 2257.1 679.5 2295.9 799.6 1719.7 712.5 2393.8 831.9 2617.0 871.3 2909.6 738.0 1915.1 613.6 1743.5 648.0 1818.8 543.8 2198.2 619.9 2091.8 502.1 2829.5 736.2 2128.2 587.2 2028.4 624.0 2417.1 599.9 2228.8 624.0 Average 369.4 117.7 Std dev

133

Table 6 Table showing the measurements and frequency calculations for the magnetosomes from the Mickey Hot Springs samples. Length Width (µm) (µm) Width/Length 72.0 44.1 0.6 87.9 43.2 0.5 57.8 39.3 0.7 76.6 41.5 0.5 91.4 44.9 0.5 89.4 40.9 0.5 73.5 47.1 0.6 83.0 43.1 0.5 71.1 46.9 0.7 65.4 50.1 0.8 88.3 46.4 0.5 84.9 54.0 0.6 88.1 46.9 0.5 116.2 55.2 0.5 92.9 49.3 0.5 75.0 47.9 0.6 128.7 47.4 0.4 91.3 53.7 0.6 91.9 50.6 0.6 94.1 50.6 0.5 91.3 48.4 0.5 92.7 44.0 0.5 88.8 54.9 0.6 80.7 51.3 0.6 90.9 51.7 0.6 112.6 49.2 0.4 112.2 37.1 0.3 104.6 34.0 0.3 115.8 37.4 0.3 84.7 36.2 0.4 111.0 36.2 0.3 77.6 36.2 0.5 86.5 47.3 0.5 72.2 41.8 0.6 118.2 25.7 0.2 103.2 34.8 0.3 105.2 28.7 0.3 108.3 31.1 0.3 78.1 32.0 0.4 97.2 34.4 0.4 100.5 31.1 0.3 113.3 37.4 0.3 82.1 37.8 0.5 81.2 43.2 0.5 89.1 36.2 0.4 79.6 39.2 0.5 134

78.8 40.2 0.5 90.8 43.2 0.5 98.7 36.6 0.4 93.5 43.2 0.5 104.0 28.4 0.3 95.6 43.6 0.5 80.2 37.6 0.5 67.4 44.3 0.7 88.1 42.4 0.5 71.9 43.1 0.6 90.8 39.5 0.4 64.1 42.2 0.7 62.8 41.8 0.7 80.3 39.5 0.5 81.8 38.5 0.5 92.6 42.3 0.5 80.9 44.1 0.5 91.6 37.7 0.4 76.5 35.7 0.5 77.0 46.8 0.6 45.3 36.6 0.8 70.3 41.0 0.6 80.9 43.3 0.5 109.0 36.8 0.3 84.8 42.0 0.5 104.1 41.2 0.4 86.1 41.1 0.5 60.2 37.6 0.6 100.3 47.9 0.5 71.0 39.7 0.6 72.9 43.5 0.6 78.0 35.3 0.5 76.5 35.8 0.5 71.7 31.7 0.4 79.0 39.7 0.5 107.3 45.8 0.4 91.8 53.6 0.6 90.0 46.6 0.5 84.1 50.0 0.6 77.8 53.3 0.7 100.9 57.8 0.6 101.5 37.6 0.4 86.5 57.5 0.7 63.6 44.5 0.7 95.8 55.0 0.6 82.4 62.9 0.8 78.6 59.3 0.8 86.0 49.4 0.6 90.0 41.4 0.5 77.1 54.7 0.7 66.8 46.4 0.7 135

64.5 51.4 0.8 84.8 53.3 0.6 91.8 37.9 0.4 52.1 37.1 0.7 119.5 37.1 0.3 100.9 39.8 0.4 64.9 38.5 0.6 87.7 37.6 0.4 63.1 32.0 0.5 87.8 31.3 0.4 102.0 30.7 0.3 75.0 33.2 0.4 117.6 30.6 0.3 112.5 34.5 0.3 85.6 35.5 0.4 77.7 35.0 0.5 108.6 32.0 0.3 88.2 29.3 0.3 72.2 35.4 0.5 73.5 32.8 0.4 77.0 33.6 0.4 67.8 38.6 0.6 68.4 31.7 0.5 77.3 39.3 0.5 103.5 45.3 0.4 76.1 42.5 0.6 76.1 36.8 0.5 72.9 31.1 0.4 78.8 29.9 0.4 61.7 40.9 0.7 83.1 35.6 0.4 51.5 29.0 0.6 113.8 38.6 0.3 96.6 39.9 0.4 127.6 34.6 0.3 88.3 39.6 0.4 95.2 35.3 0.4 85.3 25.7 0.3 136.3 40.3 0.3 79.2 26.3 0.3 97.6 33.0 0.3 53.7 27.1 0.5 76.7 36.4 0.5 66.8 34.4 0.5 75.6 31.6 0.4 70.7 33.6 0.5 71.6 30.4 0.4 105.2 33.7 0.3 98.7 32.7 0.3 101.6 37.3 0.4 74.8 35.8 0.5 136

75.9 36.7 0.5 78.2 32.6 0.4 79.4 34.1 0.4 81.0 39.1 0.5 54.9 33.5 0.6 77.8 30.1 0.4 92.4 37.6 0.4 61.7 29.8 0.5 50.0 24.1 0.5 39.9 16.6 0.4 39.9 28.8 0.7 84.6 27.9 0.3 83.0 22.3 0.3 119.9 31.6 0.3 74.3 30.7 0.4 66.6 28.8 0.4 81.9 28.9 0.4 76.8 32.8 0.4 95.4 39.3 0.4 57.4 25.7 0.4 70.3 27.3 0.4 48.7 30.2 0.6 74.6 29.0 0.4 60.2 28.3 0.5 98.5 33.7 0.3 82.3 28.6 0.3 54.4 25.7 0.5 78.2 33.2 0.4 91.5 26.5 0.3 84.2 38.8 0.5 Average 17.4 8.2 0.1 Std dev 136.3 62.9 0.8 Max 39.9 16.6 0.2 Min

137

Table 7 Table for calculating the frequecy of the magnetosome length to width ratios and the frequency of lengths for the Mickey Hot Springs sample. Cumul. Cumul. Bins Frequency Freq % Freq Bins Frequency Freq % Freq 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.0 0.0 0.0 1.0 0.1 0.0 0.0 0.0 10.0 0.0 0.0 1.0 0.1 0.0 0.0 0.0 15.0 1.0 0.0 3.0 0.2 0.0 0.0 0.0 20.0 0.0 0.0 24.0 0.2 0.0 0.0 0.0 25.0 2.0 0.0 61.0 0.3 1.0 0.0 1.0 30.0 21.0 0.1 109.0 0.3 10.0 0.1 11.0 35.0 37.0 0.2 140.0 0.4 21.0 0.1 32.0 40.0 48.0 0.3 157.0 0.4 14.0 0.1 46.0 45.0 31.0 0.2 172.0 0.5 31.0 0.2 77.0 50.0 17.0 0.1 176.0 0.5 36.0 0.2 113.0 55.0 15.0 0.1 177.0 0.6 18.0 0.1 131.0 60.0 4.0 0.0 177.0 0.6 16.0 0.1 147.0 65.0 1.0 0.0 177.0 0.7 12.0 0.1 159.0 70.0 0.0 0.0 0.7 9.0 0.1 168.0 75.0 0.0 0.0 0.8 4.0 0.0 172.0 Total 80.0 0.0 1.0 % 0.8 4.0 0.0 176.0 85.0 0.0 0.9 1.0 0.0 177.0 90.0 0.0 0.9 0.0 0.0 177.0 95.0 0.0 1.0 0.0 0.0 177.0 100.0 0.0 1.0 0.0 0.0 177.0 105.0 0.0 Total 177.0 1.0 110.0 0.0 115.0 0.0 120.0 0.0 125.0 0.0 130.0 0.0 135.0 0.0 140.0 0.0 145.0 0.0 150.0 0.0 155.0 0.0 160.0 0.0 165.0 0.0 170.0 0.0 175.0 0.0 180.0 0.0 185.0 0.0 190.0 0.0 195.0 0.0 200.0 0.0 Sum 177.0

138

Figure 34 Example of where the EDX spectra were collected from the Mickey Hot Springs sample (A). The two insets in the lower right of panel A show where the EDX were collected for the crystal (red circle) and for the background of the cell (red rectangle). The EDX spectrum for the crystal showing iron and oxygen as the major components (B). The EDX spectrum for the background of the cell showing large peaks for copper (Cu) (from the grid) and smaller peaks for arsenic (As), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), and iron (Fe).

139

140

Table 8 Table showing the measurements of the lengths and widths of the cells from Pavilion Lake. Length (µm) Width (µm) 3450.1 328.8 3374.8 331.6 2978.2 366.2 2081.4 322.2 3509.2 310.8 2405.4 323.9 2535.8 358.2 2905.0 334.5 Average 570.4 20.2 Std dev

141

Table 9 Table showing the measurements and frequency calculations for the magnetosomes for the Pavilion Lake sample. Length Width (µm) (µm) Width/Length 52.2 51.6 1.0 55.1 52.9 1.0 54.3 50.8 0.9 61.5 56.0 0.9 53.1 46.1 0.9 46.0 46.0 1.0 50.9 40.0 0.8 50.2 44.5 0.9 44.1 43.2 1.0 48.9 46.1 0.9 47.5 46.9 1.0 49.1 46.1 0.9 48.9 46.1 0.9 52.6 49.7 0.9 50.2 48.5 1.0 50.5 48.7 1.0 42.3 38.0 0.9 41.8 41.4 1.0 43.2 40.6 0.9 45.5 38.6 0.8 50.8 50.7 1.0 51.2 48.6 0.9 48.1 44.7 0.9 44.7 36.9 0.8 52.8 48.5 0.9 43.2 39.0 0.9 50.1 46.1 0.9 43.9 43.7 1.0 44.6 37.0 0.8 36.8 36.1 1.0 38.6 38.1 1.0 39.4 39.2 1.0 45.4 40.6 0.9 44.4 43.8 1.0 42.8 39.2 0.9 56.7 51.7 0.9 47.5 42.9 0.9 47.7 45.5 1.0 53.3 48.0 0.9 44.2 43.6 1.0 43.8 41.8 1.0 50.6 49.1 1.0 48.9 48.7 1.0 43.4 41.3 1.0 46.5 41.5 0.9 44.8 40.0 0.9 142

54.2 50.9 0.9 51.8 51.0 1.0 56.6 46.3 0.8 49.0 48.6 1.0 61.3 50.8 0.8 54.5 52.5 1.0 39.4 35.2 0.9 43.8 40.7 0.9 45.2 38.2 0.8 37.7 36.1 1.0 43.3 42.5 1.0 47.0 39.1 0.8 41.4 35.6 0.9 45.7 41.1 0.9 46.5 42.4 0.9 45.7 41.5 0.9 43.2 41.3 1.0 50.2 49.5 1.0 43.2 40.4 0.9 48.4 47.1 1.0 42.5 42.5 1.0 49.7 46.7 0.9 42.2 41.9 1.0 46.3 46.2 1.0 45.7 39.4 0.9 49.3 48.4 1.0 43.4 40.8 0.9 37.5 36.9 1.0 40.7 35.0 0.9 37.1 33.9 0.9 37.9 32.9 0.9 38.3 37.9 1.0 41.2 38.8 0.9 43.7 40.0 0.9 49.3 48.1 1.0 55.0 50.3 0.9 51.6 45.8 0.9 50.8 48.4 1.0 43.6 39.5 0.9 44.4 42.1 0.9 46.9 46.2 1.0 54.8 54.1 1.0 43.0 40.3 0.9 43.4 42.1 1.0 49.2 42.3 0.9 45.6 45.4 1.0 41.5 34.7 0.8 51.9 46.4 0.9 48.0 48.0 1.0 43.7 43.4 1.0 46.2 45.8 1.0 143

50.8 45.2 0.9 48.1 47.6 1.0 49.6 47.8 1.0 46.9 43.8 0.9 Average 5.1 5.1 0.1 Std dev 61.5 56.0 1.0 Max 36.8 32.9 0.8 Min

144

Table 10 Table for calculating the frequency of the magnetosome length to width ratios and the frequency of lengths for the Pavilion Lake sample Cumul. Cumul. Bins Frequency Freq % Freq Bins Frequency Freq % Freq 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 5.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 10.0 0.0 0.0 0.0 0.7 0.0 0.0 0.0 15.0 0.0 0.0 0.0 0.8 1.0 0.0 1.0 20.0 0.0 0.0 0.0 0.9 24.0 0.2 25.0 25.0 0.0 0.0 4.0 1.0 75.0 0.8 100.0 30.0 0.0 0.0 24.0 1.1 0.0 0.0 100.0 35.0 4.0 0.0 54.0 1.2 0.0 0.0 100.0 40.0 20.0 0.2 88.0 1.3 0.0 0.0 100.0 45.0 30.0 0.3 99.0 1.4 0.0 0.0 100.0 50.0 34.0 0.3 100.0 1.5 0.0 0.0 100.0 55.0 11.0 0.1 100.0 Total 100.0 1.0 60.0 1.0 0.0 100.0 65.0 0.0 0.0 100.0 70.0 0.0 0.0 75.0 0.0 0.0 80.0 0.0 0.0 Total % 85.0 0.0 1 freq 90.0 0.0 95.0 0.0 100.0 0.0 105.0 0.0 110.0 0.0 115.0 0.0 120.0 0.0 125.0 0.0 130.0 0.0 135.0 0.0 140.0 0.0 145.0 0.0 150.0 0.0 155.0 0.0 160.0 0.0 165.0 0.0 170.0 0.0 175.0 0.0 180.0 0.0 185.0 0.0 190.0 0.0 195.0 0.0 200.0 0.0 Total % 100

145

Figure 36 Example of where the EDX spectra were collected from the Pavilion Lake sample (A). The red circle labeled 1 and rectangle show where the EDX were collected for the crystal (red circle) and for the background of the cell (red rectangle). The EDX spectrum for the crystal showing iron and oxygen as the major components (B). The EDX spectrum for the background of the cell showing large peaks for copper (Cu) (from the grid) and smaller peaks for magnesium (Mg), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), and potassium (K). 146

Appendix B: Additional Methods and Data for Chapter 4

Magnetic Spirillum Growth Medium based on ATCC media 1653

Distilled water...... 1.0 L

Wolfe's Mineral Solution (see below)...... 5.0 ml

0.1% Resazurin...... 0.45 ml

KH2 PO4 ...... 0.68 g

NaNO3 ...... 0.12 g

Ascorbic acid...... 0.035 g

Tartaric acid...... 0.37 g

Succinic acid...... 0.37 g

Sodium acetate...... 0.05 g

Add components in the order given with stirring. Adjust to pH 6.75 with NaOH.

Autoclave the medium at 121°C for 15 minutes. Aseptically fill screw-capped containers to full capacity with sterile medium. Inoculate heavily leaving no headspace of air, and screw down closures tightly.

After autoclaving add:

Aseptically add Wolfe's Vitamin Solution (see below)...... 10.0 ml

0.01 M Ferric Quinate (see below)...... 2.0 ml

147

To make 0.01 M Ferric Quinate:

FeCl3 ...... 0.27 g

Quinic Acid (Sigma Q-0500)...... 0.19 g

Distilled water...... 100.0 ml

Dissolve and autoclave at 121°C for 15 minutes.

To make Wolfe’s vitamin solution:

Biotin...... 2.0 mg

Folic acid...... 2.0 mg

Pyridoxine hydrochloride...... 10.0 mg

Thiamine HCl...... 5.0 mg

Riboflavin...... 5.0 mg

Nicotinic acid...... 5.0 mg

Calcium D-(+)-pantothenate...... 5.0 mg

Vitamin B12...... 0.1 mg p-Aminobenzoic acid...... 5.0 mg

Thioctic acid...... 5.0 mg

Distilled water...... 1.0 L

To make Wolfe’s mineral solution:

Nitrilotriacetic acid...... 1.5 g

MgSO4 . 7H2O ...... 3.0 g

148

MnSO4 . H2O ...... 0.5 g

NaCl...... 1.0 g

FeSO4 . 7H2O ...... 0.1 g

CoCl2 . 6H2O ...... 0.1 g

CaCl2 ...... 0.1 g

ZnSO4 . 7H2O ...... 0.1 g

CuSO4 . 5H2O ...... 0.01 g

AlK(SO4)2 . 12H2O...... 0.01 g

H3BO3 ...... 0.01 g

Na2MoO4 . 2H2O...... 0.01 g

Distilled water...... 1.0 L

Add nitrilotriacetic acid to approximately 500 ml of water and adjust to pH 6.5 with KOH to dissolve the compound. Bring the volume to 1.0 L with remaining water and add remaining compounds one at a time.

149

Figure 37 The plasmid containing the mms6 gene.

150

The amino acid sequence and the DNA sequence for the Mms6 protein and the mms6 gene, respectively.

Mms6 amino acid sequence*, 157 amino acids, 14691 Daltons, 474 base pairs mpaqiangvicppgapagtkaaaamgemeregaaakagaaktgaaktgtvaktgiaaktgvatavaapaapanvaaaq gagtkvalgagkaaagakvvggtiwtgkglglglglglgawgpiilgvvgagavyaymksrdiesaqsdeevelrdala

DNA sequence**

GTGCCAGCTCAGATCGCCAACGGAGTTATTTGCCCCCCAGGGGCCCCGGCCG

GAACCAAGGCCGCCGCCGCCATGGGCGAGATGGAGCGCGAGGGCGCCGCCG

CCAAGGCCGGGGCTGCCAAGACGGGCGCCGCCAAGACCGGAACCGTCGCCA

AGACCGGCATCGCCGCCAAGACGGGTGTTGCCACCGCCGTTGCCGCTCCGGC

GGCTCCTGCCAATGTTGCCGCCGCCCAGGGCGCCGGGACCAAGGTCGCCCTT

GGCGCGGGCAAGGCCGCCGCCGGTGCCAAGGTCGTCGGTGGAACCATCTGG

ACCGGTAAGGGGCTGGGCCTCGGTCTGGGTCTCGGTCTGGGCGCGTGGGGGC

CGATCATTCTCGGCGTTGTTGGCGCCGGGGCGGTTTACGCGTATATGAAGAG

CCGTGATATCGAATCGGCGCAGAGCGACGAGGAAGTCGAACTGCGCGACGC

GCTGGCCTGA

*From the NCBI website: http://www.ncbi.nlm.nih.gov/protein/83310055?report=fasta

**From the NCBI website: http://www.ncbi.nlm.nih.gov/protein/83310055?report=graph

151

Mms6 Protein Purification Protocol

Start with frozen cell pellet from Superbroth/IPTG added cells.

Lysis buffer (100mL) good for 1 hour:

0.477 g HEPES (20 mM)

2.922 g NaCl (500 mM)

10 mL glycerol (10%v/v)

0.100 mL NP-40 (0.1%v/v)

69.9 µL 14.3 M 2-mercaptoehtanol (10 mM)

500 µL 0.2 M PMSF (1 mM) pH = 7.9

Add 4 mL of lysis buffer to each 1 g of cell pellet

Pipette up and down until the pellet is dissolved

Pool about 20 mL together

Centrifuge for 15 minutes at 4ºC for 12,500RPM

Pour off supernatant, add 10 mL of E-buffer to each tube, pipette up and down to homogenize.

E-buffer (100mL):

1.192 g HEPES (50mM)

5 mL glycerol (5% v/v)

3.49 µL 2-mercaptoethanol (0.5mM)

0.05 g Na-deoxy-cholate (0.05% w/v)

152

1mL NP-40 (1% v/v) pH = 7.9

Combine into 1 tube and centrifuge at 4ºC at 14,500 rpm for 20 minutes

Pour off supernatant

Add 20-25 mL of S-buffer and pipette up and down and let the solution incubate on a shaker at 250 rpm at RT (temperature doesn’t really matter) for 16 hours (or overnight).

The solution gets really foamy.

S-buffer (100mL):

0.238 g HEPES (10 mM)

57.38 g Guanidine–HCl (6 M)

34.9 µL 14.3 M 2-Mercaptoethanol (5 mM) pH = 7.9

Centrifuge the cell lysate for 10-15 minutes at 12,500 rpm

Get the resin ready

Talon Clontech resin #635503 100 mL

BioRad columns 732-1010 20 mL chromatography column

Add 5 mL of pure resin into two tubes (total of 10 mL)

Centrifuge resin to get rid of alcohol (4 minutes at 7500 rpm)

Pour off alcohol; this gives 2.5 mL of resin in each tube (5mL total)

153

Wash resin 3X with S-buffer supplemented with 5 mM imidazole, add 5 µL to 10 mL of buffer (really it is 1X with about 5-10 mL of buffer)

Centrifuge for 4 minutes at 7500 rpm at 4oC

Aliquot 1 M imidazole and store at -20oC

Mix 16 hour lysate with resin for 2 hours at 4oC

Pour into column

Use 50 mL Falcon tube to collect flow thru.

BC 500 (100mL);

0.243 g Tris-HCl (20 mM)

20 ml glycerol (20 % v/v)

3.728 g KCl (500 mM)

50 µl NP-40 (0.05 % v/v)

69.8 µl 14.3 M 2-mercaptoethanol (10 mM)

100 µl 0.2M PMSF (0.2 mM) pH=7.9

W1 = S-buffer + 5 mM imidazole (125 µL imidazole in 25 mL of S-buffer), 25mL is 5X resin bed

W2 = BC 500 + 5 mM imidazole

W3 = BC 500 + 15 mM imidazole

W4 = BC 100 + 30 mM imidazole

BC 100 (100 mL)

154

Same as BC 500 but 0.7456 g of KCl (100mM)

Elute protein with BC 100 + 300 mM imidazole

Dialyze protein using:

20mM Tris, 500 mM KCl, 0.2 mM PMSF, 2 M urea, pH 7.5, 1L for 2 hours

20mM Tris, 500 mM KCl, 0.2 mM PMSF, 1 M urea, 1mM, EDTA, pH 7.5, 1L for 2 hours

20mM Tris, 500 mM KCl, 0.5 M urea, 1 mM EDTA pH 7.5, 1L for 2 hours

20mM Tris, 500mM KCl, pH 7.5 4 Dialysis against 20mM Tris, 100mM KCl, pH7.5, 1L for 2 hours, 2 times

155

Figure 38 Data from the sequencing of the purified Mms6 protein showing that it corresponds to Mms6.

156

Figure 39 Examples of sections labeled with concentrated primary antibodies. A) Sections labeled with primary antibody at a concentration of 1:200. B) Sections labeled with primary antibodies at a concentration of 1:400. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images.

157

Figure 40 Examples of sections labeled with concentrated primary antibodies. A) Sections labeled with primary antibody at a concentration of 1:600. B) Sections labeled with primary antibodies at a concentration of 1:800. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images. 158

Figure 41 Examples of sections labeled with concentrated primary antibodies. A) Sections labeled with primary antibody at a concentration of 1:1000. B) Sections labeled with primary antibodies at a concentration of 1:1200. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images.

159

Figure 42 Examples of sections labeled with concentrated primary antibodies. A) Sections labeled with primary antibody at a concentration of 1:1600. B) Sections labeled with primary antibodies at a concentration of 1:2000. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images.

160

161

Figure 44 Examples of sections labeled with concentrated primary antibodies. A) Sections labeled with primary antibody at a concentration of 1:2000. B) Sections labeled with primary antibodies at a concentration of 1:4000. In both images, the secondary antibody concentration was 1:100. Scale bar is 500 nm in both images.

162