Analysis of Microbes in Greenland Ice Cores from Periods of High and Low Atmospheric Carbon Dioxide Levels

ANALYSIS OF MICROBES IN GREENLAND ICE CORES FROM PERIODS OF HIGH

AND LOW ATMOSPHERIC CARBON DIOXIDE LEVELS

Caitlin Nicole Knowlton

A Thesis Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

May 2013

Committee:

Scott O. Rogers, Advisor

Paul Morris

Vipaporn Phuntumart 1

ABSTRACT

Scott O. Rogers, Advisor

Glacial ice entraps microbial cells and nucleic acids that have been deposited from the atmosphere by precipitation or on dust particles carried by the wind. The ice is deposited in sequential layers representing distinct past periods of time. Metagenomics and metatranscriptomics offer means to perform comprehensive analyses of microbial genes and taxa within the ice. During the past 400,000 years, fluctuations in atmospheric carbon dioxide levels have directly coincided with temperature fluctuations. Three ice core sections from Byrd Station, Antarctica (80° 1’ S, 119° 31’ W) and three ice core sections from GISP 2D (Greenland Ice Sheet Project, core 2D; 72o 36' N, 38o 30'W) were aseptically melted. Two of the ice core sections from GISP 2D (at 1600 m = 10,500 ybp

[years before present] and at 3014 m = 157,000 ybp) were analyzed using metagenomic and metatranscriptomic analyses. The 157,000 year old ice core section was deposited during low atmospheric CO2 levels that would have led to low temperatures, low precipitation rates, and high dust levels, while the 10,500 year old core section was deposited during high CO2 levels that would have led to higher temperatures, higher precipitation rates, and lower dust levels. Nucleic acids from ultracentrifuge- concentrated meltwater were subjected to 454 pyrosequencing. Taxonomic analyses were based on rRNA and mRNA sequences. Taxa in the 10,500 ybp ice primarily included bacterial species of Proteobacteria, Firmicutes, and Cyanobacteria. This sample also contained black particulate matter indicative of volcanic activity. Major volcanic activity was reported in Iceland, Alaska, and from a super volcano in Italy ii

iii during the 10,500 ybp time period. The 157,000 ybp sample also contained a large number of Proteobacteria and Firmicute species, a few Actinobacteria and

Cyanobacteria, but most were common soil bacteria, as well as animal pathogens. The number and frequency of cyanobacterial species was much higher. While some species were in common to both samples, a large number of unique taxa and genes were present in each of the samples. These unique taxa and genes suggest different atmospheric conditions may influence the deposition of microbes in ice.

iii

ACKNOWLEDGEMENTS

It is important to recognize that this research was partially funded by BGSU through FRC-RIG (Faculty Research Committee Research Incentive Grants). The ice in this study was provided by NICL (in Colorado) and NICL-SMO (in New Hampshire).

I acknowledge my advisor, Dr. Scott Rogers, who has always pushed me towards excellence. I am thankful for his gentle guidance and patient teaching in my years at Bowling Green State University. I am grateful for all he has done to make this a successful growing experience.

Many thanks to the educators that have fueled my curiosity to learn: Mr. Mike

Sutter-who introduced me to the mystery of science in high school, Dr. Andrew

Whipple-my undergraduate advisor, Dr. Anita Hopper-for giving me a chance to understand the joy of lab work, Dr. Ray Larsen-for his stimulating classes and enthusiasm to bring me to Bowling Green State University and Dr. Carmen Fioravanti– who provided insight to my research in the beginning stages. I also thank my committee members-Dr. Vipa Phuntumart and Dr. Paul Morris. I appreciate all their friendly advice. I am indebted to Dr. Morris for his generous funding through scholarships during my graduate studies.

I owe a huge thank you to my lab mates, Sammy Juma, Jen Hofacker and especially Yury Shtarkman, who generously conversed and shared ideas with me. It was a pleasure to work alongside such kind individuals. I would also like to thank

Seung-Geuk Shin for his assistance with parts of my lab work. I am grateful for Dr. Ram iv

Veerapaneni and Dr. Tom D’Elia’s comradery and previous work in our lab, that when combine with my work, was invaluable in producing our publication.

I am beyond thankful to family, my father and mother-Rob and Mary Knowlton, and my siblings- Allison and Nate Knowlton, for their unwavering faith in my ability and the support that they have been. I must also thank my friends for the moments of laughter that kept me sane. My friends and family have been a joy and comfort throughout this process. I particularly thank Elizabeth Metz for the rock that she is in my life, and Alexandra Leguire for sharing in my every success throughout my research. Additionally, I am ever grateful to the love of my life, Samuel Shepard, for his constant affirmation. His understanding and encouragement was so important in this journey.

Most importantly, I thank my heavenly Father for the grace He shows to me through life in Jesus Christ. I am honored to study His hand in the intricacy, beauty, and complexity of biology. I would have never made it through this time without continual renewal, peace, and strength from above.

TABLE OF CONTENTS

Page

CHAPTER I. INTRODUCTION TO POLAR MICROBIOLOGY AND

METAGENOMICS/METATRANSCRIPTOMICS ...... 1

1.1 Why study environmental ice deposits? ...... 2

1.1.1 Background...... 2

Application of ice: atmosphere and microbes ...... 2

Atmospheric changes and genome recycling ...... 2

Glacial formation...... 4

Dating ice cores ...... 5

Adaptations of microbes living in ice ...... 6

1.1.2 Evolutionary and practical application of ice study...... 9

1.2 How to study the ice ...... 9

1.2.1 Obtaining ice ...... 9

The ice cores...... 9

The approach ...... 11

1.2.2 Eliminating Contamination ...... 12

1.2.3 Methods ...... 14

Retrieving nucleic acids from ice ...... 14

Sample preparation for sequencing ...... 15

Sequencing ...... 18

Data Analysis...... 20 vi

vii

Assembling Data...... 20

Phylogenetics...... 22

Microscopy of black particulate matter ...... 24

REFERENCES...... 27

CHAPTER II. RESEARCH ...... 34

2.1 Introduction ...... 35

2.2 Materials and Methods ...... 37

2.2.1 Ice core selection and samples ...... 37

2.2.2 Surface sterilization and melting...... 37

2.2.3 Culturing ...... 38

2.2.4 Ultracentrifugation ...... 39

2.2.5 Isolation of RNA ...... 39

2.2.6 Isolation of DNA ...... 40

2.2.7 Preparation of samples for sequencing...... 41

2.2.8 Sequencing ...... 47

2.2.9 Analysis of sequences...... 48

2.2.10 Scanning Electron Microscope (SEM) ...... 49

2.2.11 Sequencing of Cultures ...... 49

2.3 Results……...... 50

2.3.1 Taxa...... 54

2.3.2 Phylogenetics...... 56

2.3.3 Culturing and sequencing ...... 63 vii

viii

2.3.4 Elemental analysis ...... 66

2.4 Discussion ...... 68

2.4.1 Overview...... 68

2.4.2 Expected microbes and found microbes...... 68

Greenland...... 68

Microbes in arid conditions...... 69

Microbes in the atmosphere ...... 69

Microbes previously found in glacial ice...... 71

Temperature influence ...... 72

Actinobacteria...... 73

Microbes found in cold environments...... 74

Cyanobacteria...... 75

Firmicutes...... 75

Extremophiles...... 77

Proteobacteria...... 78

Uncultured bacteria ...... 82

2.5 Conclusions...... 83

REFERENCES...... 84

APPENDIX A. Knowlton, C., R.Veerapaneni, T. D’Elia, and S.O. Rogers. 2013.

Microbial analyses of ancient ice core sections from Greenland and Antarctica. Biology

2:206-232………………………………………………………………………………….. 96

viii

LIST OF TABLES

Table Page

1 Primer construction for 454 sequencing ...... 17

2 Primer names and sequences used in amplification for 454 sequencing...... 45

3 Primer names and sequences used in amplification of cultures...... 50

4 rRNA Sequences from the GISP2D ice core sections...... 52

5 mRNA Sequences from the GISP2D ice core sections...... 53

6 Isolate sequences from the ice core sections ...... 65

LIST OF FIGURES

Figure Page

1 454 sequencing method...... 19

2 Graph of past atmospheric changes ...... 36

3 Gel for quantifying DNA...... 47

4 Phylogram of bacterial small subunit ribosomal RNA sequences ...... 59

5 Phylogram of bacterial large subunit ribosomal RNA sequences ...... 60

6 Bar graph of phyla represented in the study ...... 62

7 SEM micrographs and spectra ...... 67

CHAPTER I.

INTRODUCTION TO POLAR MICROBIOLOGY AND

METAGENOMICS/METATRANSCRIPTOMICS

1.1 Why study environmental ice deposits?

1.1.1 Background

Application of ice: atmosphere and microbes

Ancient ice holds precious information about the past. During the last four decades it has been used to unveil climate and atmospheric changes that have occurred over the past 740,000 years (Augustin, et al. 2004). Gathering information on previous climate change has been helpful in understanding the climate patterns and predicting future climate. Also, during the past two decades, ice research has been used to characterize organisms entrapped in the ice from the past eight million years (Bidle, et al. 2007). Understanding ancient microbial assemblages can provide insights into the past (Miteva, et al. 2009) and potentially be used to predict atmospheric and climatologic conditions and events to come. This has been especially applicable when attempting to deduce the effects of the recent increases in atmospheric carbon dioxide.

Microbes identified in ancient ice may lead to a better understanding of the modes and tempos of evolution (Catranis, et al. 1991), as well as to understand ecological changes in global microbial diversity (Petit, et al. 1999, Rodríguez-Valera 2004). The organisms entrapped in the glacial layers can be compared to each other and to contemporary microbes in order to estimate climatological and evolutionary trends. Some may serve as sentinel species to indicate specific atmospheric or climatic events.

Atmospheric changes and genome recycling

Fluctuations in temperature and CO2 levels are positively correlated (Petit, et al.

1999). Increases in the atmosphere associated with greenhouse gases, particularly CO2 2

3 and methane, are linked to warming events on Earth. The study of polar glacial ice can be used to predict global climate changes. Over time, there have been distinct glacial and interglacial periods (Augustin, et al. 2004). Also, global climate change might play a role in the rate of genome recycling (Rogers, et al. 2004a). Genome recycling occurs when the microbes from the past are reintroduced into current populations through melting of environmental ice. With recent increases in global temperature (Petit, et al.

1999) and records of glacial thawing, reintroduction of entrapped microbes into the environment is increasing. Microbes that have the ability to adapt and survive with minimal resources and harsh conditions can revive upon exposure to conditions above freezing (Vorobyova, et al. 1997). Reintroduction of past microbes into present times can affect evolutionary processes, including the introgression of ancient genotypes into the current populations. For hosts of pathogens, this gene flow poses a concern because immune resistance to ancient pathogens may have been lost. Therefore, organisms may be more susceptible to the ancient pathogenic versions of such microbes upon their reintroduction.

Immunity to infectious agents can be either innate, which is hereditary, or adaptive, which is acquired through external means (Murphy, et al. 2007). Indeed, adaptive immunity can be naturally acquired by exposure to infection or passed from the mother at birth. Artificial adaptive immunity is acquired through vaccination or antibody transfer. While individuals have unique immune histories in response to their exposure to pathogens, vaccines can additionally confer protection against some of the most virulent infectious agents. When a large number of individuals within a 3

4 population are immune to specific contagious microbes, the few that have not received the vaccine still are protected by herd immunity (Murphy, et al. 2007).

Glaciers are stratified. Since different organisms lie in various regions of the ice core, exposure happens at different times. The microbes uncovered from ancient ice are older and may differ genetically and antigenically from the microbes in the environment today. Therefore, they may infect immunologically naïve individuals in host populations.

Glacial formation

Ice accumulates in layers, usually from falling snow. Snow carries dust and biogenic particles from the atmosphere that can become embedded in the glaciers. As the snow compacts due to the weight of the overriding snow layers, firn layers form.

Firn is simply compacted snow from previous years that has not yet formed dense ice layers. The firn layer at the GISP2 (Greenland Ice Sheet Project) site was reported to be approximately 77 m thick (Paterson 1994). While compacted, percolation of water and infusion of gases occurs throughout the firn (Paterson 1994).

Layers of glacial ice occur at greater depth. The ice in these layers has been compacted, which causes fusing of ice crystals and compression of gas bubbles.

Percolation of water and gases usually does not occur below this zone. Reports show the deeper the ice in a core, the higher the density and pressure is in that ice (Paterson

1994). The high-pressure levels and fragile ice structure causes some ice to be unstable and sensitive to lower pressure as the pressure of the expanding gas becomes greater than the bonds holding the ice crystals together. In the lower layers, the deep ice 4

5 becomes more densely compacted due to the weight that pushes the crystals tightly together and the rearrangement of solid water molecules to form stronger crystals.

These compacted, frozen layers preserve a chronological record of past atmospheric gases, particles, and biological components.

Dating ice cores

Research has been undertaken to determine ice core ages based on ice core section depth, ice crystal morphology, chemistry, and isotopic measurements (Hamer, et al. 1994, Meese, et al. 1997, Montagnat, et al. 2000). Many methods have been employed to ensure accurate dating. The Byrd ice core sections from Antarctica have been dated by measuring crystal size and crystal shifting, a procedure known as dislocation density (Montagnat, et al. 2000). Hamer, et al. (1994) used ECM (electrical conductivity method) to date the Byrd cores. Other ice cores are dated by their chemistry. The GISP ice cores were accurately dated using a variety of methods such as

ECM, oxygen isotopic ratios, light scattering from dust, and major ion chemistry

(Meese, et al. 1997). The analysis of glass shards and sulfates in an ice core indicate volcanic activity, some of which can be correlated with dates for known eruptions

(Meese, et al. 1997). Finally, a simple but reliable method of ice dating comes from measurements of visual stratification of the ice (Meese, et al. 1997). In the upper portions of the glaciers and ice sheets, the ice usually has distinct lines to differentiate time periods. However, in deeper ice, the layering becomes less distinct eventually disappearing completely due to increased compaction. After the timeline for the ice has been built, scientists can compare microbes in the ice core sections based on age and the 5

6 atmospheric composition at the time of the ice deposition. Yet, none of this would be possible unless the microbes and/or their biological molecules persist in the ice.

Adaptations of microbes living in ice

A variety of microbes can survive freezing in glacial ice. Microbes enduring

Arctic conditions generally have proven to be robust, resilient, and adaptable (Finegold

1996). The key to successful survival of these organisms can be attributed to the many ways in which these organisms have adapted to thrive in cold environments. Constant low temperature is a factor that may hasten adaptation.

Low concentrations of nutrients and free water limit the acquisition of needed compounds. Slow metabolic activities in frozen organisms have been observed

(Bakermans, et al. 2003), which might allow them to survive in a dormant state.

Dormancy preserves the original microbes from being degraded to the point of being non-viable. It may also slow evolution (Rogers, et al. 2004a) because the organisms undergo no cell division or are doing so at slow rates. Small microchannels in the ice allow small amounts of nutrients to reach the cells providing a means of life (Deming

2002, Price 2000). These nutrients allow some microbes to maintain a low level of metabolic activity (Bakermans, et al. 2003). While microbes are capable of protecting themselves from the cold and they may be metabolically active for extended periods of time, nucleic acids (and other biological molecules) may be detectable for millions of years after the microbes have died (Bidle, et al. 2007).

In addition to their low metabolic activities, microbes in cold environments utilize a variety of proteins and enzymes to overcome the stress of living in ice. Some 6

7 organisms are able to maintain liquid water layers around them by secreting molecules that inhibit water crystallization (Uchida, et al. 2012). One particular mechanism that allows survival in psychrophiles is the induction of cold shock proteins and cold acclimation proteins (Roberts, et al. 1992). These proteins help to induce transcription and translation despite low temperatures (Roberts, et al. 1992). Another general type of protein that can assist in promoting cell survival in freezing conditions are extracellular enzymes, these enzymes play a role in cold-adapted cell metabolism (Deming 2002).

Some of the many enzymes associated with cold-adapted bacteria include xylanases and laminarases, valine dehydrogenase, chitobiase and chitinases, and pectate lyase

(Deming 2002). Interestingly, all of these enzymes have specific amino acid residues located in the domain of the active site that influence the conformational flexibility required of the site for stimulation in the cold (Deming 2002). There are also extracellular proteins and polysaccharides that confer resistance to cold temperatures and freezing. In addition to the biochemical changes in proteins and enzymes, there can be further protective adaptations in membrane structure and pigmentation.

Lipid composition in the membrane changes in response to environmental surroundings (Driessen, et al. 1996). In psychrophilic organisms, the ability to keep the cytoplasm liquid even when supercooled (temperatures below freezing) allows metabolic processes to continue well below 0°C. Unsaturated short (C14-C16) acyl chains increase membrane fluidity (Driessen, et al. 1996). Unsaturated fats, having a lower melting point than saturated fats and greater fluidity, can serve as cryoprotectants. 7

In order for bacteria to remain viable for many years they must have a way of protecting or repairing cell damage from radiation. UV exposure can be higher in polar regions. Both cosmic rays and radioactive decay from elements in the Earth cause damage to the biological molecules in cells. The normal UV repair mechanisms,

“repairases,” are not usually working at such low temperatures (Finegold 1996).

Therefore, there must be another measure to protect the cell from damage. An interesting survival technique found in the cultured organisms from ice, proven to protect the microbes from UV, is pigmentation. The pigment chlorophyll, not only captures light in photosynthesis, but also can be used for shading. Pigments do not repair the damage, but serve to protect (i.e., shade) the cell from damage upon UV exposure. Some pigment molecules can reinforce membranes (Fong, et al. 2001).

Pigments that have been revealed to provide this protection are carotenoids and bacterioruberin (Fong, et al. 2001).

There also are pigments in polar fungi, including melanin, that do block UV, but they have many other functions, including (but not limited to) inhibition of degradative enzymes (Gómez & Nosanchuk 2003). The pigments produced by many bacteria and fungi occur whether or not they are exposed to UV irradiation (or any light), which indicates that their roles in these organisms go beyond protection from UV. This protection facilitated by pigmentation allows microbes from the past to be studied presently.

1.1.2 Evolutionary and practical application of ice study

The study of microbes in ice is important for several reasons. Severe selective pressures are exerted on microbes that are entrapped in the ice. The organisms that have the most beneficial gene set will survive. Thus, evolutionary processes ongoing in icy environments may differ from those occurring at higher temperatures. Adaptive processes in these microbes must be robust, and therefore, it is important to investigate these adaptive mechanisms. Additionally, many scientists involved in polar research are interested in the possibilities of microbial life on icy planets and moons. Between

Jupiter’s ice covered moon, Europa, and Mars, which is assumed to have had environments similar to those on Earth in the past, there are hopes that the search for life can extend far beyond our planet (Willerslev, et al. 2004). Through this research we are investigating the parameters for survival in environmental ice.

1.2 How to study the ice

1.2.1 Obtaining the ice

The ice cores

Ice is an ideal matrix for preserving microbes and nucleic acids. Yet, because of the low concentrations of microbes in the ice, the potential for contamination from external sources is high. Also, the relative fragility of ice requires great care in collecting, transporting, handling and processing of the ice cores. Each core section presents unique challenges and measures must be taken so that the microbes and biomolecules are persevered without the introduction of external contaminants. 9

The ice core sections used in this study were obtained from The University of

Buffalo in New York, in 1991 and from the National Ice Core Laboratory (NICL), in

Lakewood, Colorado in 1995. While some of the cores were obtained from Buffalo, New

York, all of the ice cores are now stored at the US National Ice Core Laboratory (NICL) in Lakewood, Colorado. This -36°C warehouse and -20°C workroom houses ice core sections from various locations around the world that are used for research throughout the nation and the world. The GISP2 (Greenland Ice Sheet Project, core 2) core was drilled from 1989-1993. The decision was made after drilling GISP (the initial ice core at the GISP site in central Greenland) that American scientists would drill GISP2 and

European scientists would drill GRIP (Greenland Ice Core Project), all located in

Greenland (Latitude 72.6 N , Longitude 38.5 W, Altitude 3200 m—at the summit of the

Greenland Ice Sheet). The Byrd ice core was drilled from 1957-1972. The Byrd ice cores, from Antarctica, were also funded by the US. Three ice core sections from Byrd Station,

Antarctica (80° 1’ S, 119° 31’ W) and three ice core sections from GISP 2D (Greenland Ice

Sheet Project, core 2D; 72o 36' N, 38o 30'W) were aseptically melted for this research.

Two of the ice core sections from GISP 2D (at 1600 m = 10,500 years old and at 3014 m =

157,000 years old) were analyzed using metagenomic and metatranscriptomic analyses.

Greenland is primarily a polar region because of the location from 59˚ to 84˚N latitude. The physical characteristics of Greenland, world climate, and wind patterns have each contributed to what has been deposited on the ice sheets and glaciers. Eighty- one percent of Greenland is covered by ice, which comprises over 10% of the ice (or nearly 7% of the freshwater) on Earth. The ice can be up to two miles thick in some locations. The prevailing winds over much of Greenland blow from northeast to west 10

(Great Britain 2012). Because of this, the primary source of dust and microbes is from

Asian deserts (Fischer, et al. 2007). However, winds from North America also reach

Greenland, primarily southern Greenland, where westerly winds often encroach far inland.

Indications of volcanic activity have been described from the GISP (Greenland

Ice Sheet Project) ice cores. Many volcanic periods over the past 110,000 years have been documented from study of the ice cores (Zielinski, et al. 1995). When ice sheets retreated

10,000 years ago, Iceland experienced a surge in volcanic eruptions (Pagli, et al. 2008).

This is close in time to one of the ice core sections being used in this research. There are also reports of volcanic eruptions in the Aleutian Islands near Alaska during that time

(Global Volcanism Program 2012, Detterman, et al. 1989). Though it seems unlikely due to wind patterns and distance that there would be particulate from this eruption in

Greenland ice, Zadanowicz, et al. (1999) have described volcanic matter in GISP ice core sections from eruptions in Oregon.

The approach

A common, recent method of studying biological environmental samples is through a combination of metagenomics and metatranscriptomics, culture-free methods used to sequence all of the nucleic acids in a sample (Cowan, et al. 2005, Delong, et al.

2001, Dupré, et al. 2007, Schloss, et al. 2005, Tringe, et al. 2005). Using these approaches, nucleic acid sequences can be analyzed and compared, taxonomic (often to species) determinations can be made, and metabolic types can be categorized. Microbial communities that represent specific time periods and atmospheric compositions can be 11

12 examined in glacial ice. The advantages to using metagenomics and metatranscriptomics are in the ability to characterize microbes using culture- independent methods. It has been estimated that only about 1% of organisms are currently able to be cultured (Rodríguez-Valera 2004). However, metagenomic methods are capable of yielding sequences from virtually all of the organisms present in a sample, thus increasing the accuracy of the results. Of course, a disadvantage is that physiological (and other) tests are not possible, because only the sequence data is collected.

1.2.2 Eliminating Contamination

An important consideration in the process of studying ancient ice is ensuring that no contamination occurs. Great controversy arises when it comes to verifying that the microbes found in glacial ice studies originate in the ice. Contamination can be introduced into the ice core sample at any point in the process: human handling, drilling (fluids), transportation, melting process, concentration procedures, extraction procedures, or in subsequent molecular biological procedures including sequencing

(Willerslev, et al. 2004). Contamination must be eliminated or somehow prevented

(Rogers, et al. 2004b).

The ice coring process introduces many contaminants onto the surfaces of the ice core sections. For dry drilling (without utilizing drilling fluids), the surfaces of the ice are in contact with the drills and are handled by drill workers. Additionally, the ice may crack, and then contaminants can be introduced into the inner portions of the section. While drilling with a lubricating fluid fewer cracks are created in the ice, but 12

13 fluid may penetrate deep into the core if cracks are formed. There are lists of microbes known to be present in these fluids (Alekhina, et al. 2006) that may be screened out in silico during computer analyses of the sequences. Unfortunately, any species that may be present in the ice, and coincidentally in the drilling fluids, might be unnecessarily removed from the list of microbes that exist in the ice.

Once in the lab, extreme measures must be taken to reduce the introduced contaminants, partly because the concentrations of microbes in the immediate environment are far greater than those in the ice core sections. Rogers, et al. (2004b) reported the results of testing methods to best eliminate contaminating microbes from the outer ice core section surfaces. The only method that assured elimination of all external organisms and nucleic acids, while ensuring the safety of internal organisms and nucleic acids was one that employs a 5.25% sodium hypochlorite treatment applied to the outer ice core section surfaces. To ensure that all contamination from the isolation kits and reactions were detected and subsequently excluded from the lists of microbes found in ice, control samples are required. Air and procedure controls should be used to identify any possible contamination sources.

While human contamination is ever present because nucleic acids and microbes are on all parts of the human body (including shedding skin and hair), as well as in exhaled air, the possibility of contamination events are reduced by keeping work areas, reagents and equipment clean and sterile. Work areas are maintained microbe free by autoclaving supplies, treating surfaces with sterilizing chemicals, and by UV irradiation of workrooms and hoods, and then performing most, or all, of the procedures in a clean 13

14 room and biosafety laminar flow hood.

1.2.3 Methods

Retrieving nucleic acids from ice

The first step in retrieving microbial information from ice requires that the ice core be aseptically melted. Rogers, et al. (2004b) tested to find the process that best facilitates surface sterilization and decontamination. This method requires submersion of the ice core section in Clorox (5.25% sodium hypochlorite with a small amount of NaOH added as a stabilizer). In rigorous tests, this sterilization method eliminated 100% of the viable cells and 100% of the DNA and RNA (as determined by PCR amplification analysis) that had been applied onto the ice core section surface (Rogers, et al. 2004b).

After 10 seconds of immersion in Clorox, the ice core section is rinsed three times in sterile water. Melting of the core is performed in sterile laminar flow hood, to prevent the freshly sterilized core from any possible deposition of lab contaminants. From each

50 ml of meltwater collected, multiple agar cultures (Malt extract agar [MEA] and

Luria-Bertani medium [LB] plates) are inoculated with 200 µl of meltwater and incubated at 15°C to assay for viable microbes from the ice. Research shows that organisms in ice will grow at 4°C but often grow faster at higher temperatures (D'Elia, et al. 2008). Assuming that the microbes from the ice have adapted to cold temperatures, incubation is done at varying temperatures (4°C, 8°C, 15°C and 22°C) to determine the optimal temperature for growth of the organisms. While metagenomics and metatranscriptomics are culture-free methods, cultivation provides a way to confirm the sequence data and allow determinations of growth rates, physiology, and 14

15 other characteristics of the organisms.

In order to increase the likelihood of detecting the nucleic acids and organisms, the meltwater can be ultracentrifuged (100,000 rcf) at 4˚C for 15-17 hours to concentrate cells, viruses, and nucleic acids. Ribosomal RNAs can be used to determine the taxa present in a sample (metatranscriptomics), and mRNAs can be used to determine which genes are being expressed in the sample (metagenomics). Before RNA extraction, the use of sterile, RNase free reagents throughout the procedure is pivotal to preserve intact

RNA. A portion of the concentrated nucleic acids can be used for RNA extraction. Many kits to extract RNAs are available commercially (RNAqueous kit, Micro-to-midi total

RNA purification system, NucleoSpin RNA II, GenElute mammalian total RNA kit, and

RNeasy mini kit, MinElute) (Deng, et al. 2005). Another method employs the use of trizol. The trizol method generally yields high quality RNA (Deng, et al. 2005). Trizol LS reagent consists of guanidine thiocyanate, phenol and chloroform. Following RNA extraction, another portion of the concentrated sample can be used for DNA extraction.

While there are also many kits to extract DNA, our lab has found a CTAB

(cetyltrimethylammonium bromide) method to be most effective (Rogers & Bendich

1985, Rogers, et al. 1989, Rogers & Bendich 1994, Shivji, et al. 1992). CTAB is a detergent that can solubilize cell membranes and complex with DNA, thus allowing extraction of high quality DNA.

Sample preparation for sequencing

The extracted RNA can be used to produce complementary DNA (cDNA) using a SuperScript Choice cDNA kit. By using RNA to make cDNA, the rRNA (used in 15

16 taxonomy) and mRNA sequences can be sequenced concurrently. The cDNA is produced with reverse transcriptase using random hexamer primers. Then, the cDNA and the extracted DNA are combined together. EcoRI (NotI) double stranded adapters are added using DNA ligase. After the adapters are added, the samples can be fractionated by column chromatography in order to separate the small and large fragments from the fragments that are of a length that is amenable to amplification and pyrosequencing.

Pyrosequencing, using 454 technology, has the capability to determine sequences averaging 250-500 base pairs (Mardis 2008). The amount of DNA and RNA extracted from the ice is usually low. Therefore, in order to analyze these few sequences, prior to pyrosequencing, they are amplified using usually 2-3 rounds of 35-40 cycles each of

PCR (polymerase chain reaction). PCR uses single stranded DNA primers and thermostable DNA polymerase to make many copies of a targeted region of DNA. The primers are designed to bracket the area to be amplified. After reactions are prepared, they are placed in an automated thermal cycler that performs many cycles of heating and cooling (Saiki, et al. 1988). An advantage to using PCR is that very small quantities of nucleic acids can be used as the starting material. Thus, it is possible to amplify single template molecules. However, this can be a problem, as well. If any contamination is introduced, it also might be amplified. If the concentration of contaminating DNAs is higher than that of the DNAs of interest, then the results can be erroneous. Also, PCR can introduce bias because smaller molecules may be amplified more rapidly than longer molecules, molecules at higher starting concentrations can be amplified more 16

17 rapidly than rare molecules, and some primers may be more effective at initiating polymerization than others. Thus, even after consideration of contaminants, proportional differences among the population of sequences and primers may affect the resulting proportions of sequence types.

The 454 sequencing technology uses a PCR-based method to amplify and sequence the DNA fragments. Two primers are used, denoted A and B. Multiplexing also can be used, incorporating multiplex identifier sequences (called MID) into the synthesized primers. This allows concurrent sequencing of up to 7 samples at a time.

The sequences are then separated in silico by MID sequence after sequencing.

Attachment of the A, B, MID and EcoRI sequences into the sample sequences is accomplished by designing PCR primers that incorporate a combinations of all of these sequences. The EcoRI sections anneal to the templates in the initial stages of amplification. They carry a MID and either A or B sequences on their 5’ ends, which all become part of the amplified products ready for 454 sequencing (Table 1).

Table 1. Primer construction for 454 sequencing Sample Primer Sample 1 A tag sequence-Unique MID sequence 1-EcoRI (NotI) adapter sequence B tag sequence-Unique MID sequence 1-EcoRI (NotI) adapter sequence Sample 2 A tag sequence-Unique MID sequence 2-EcoRI (NotI) adapter sequence B tag sequence-Unique MID sequence 2-EcoRI (NotI) adapter sequence Sample 3 A tag sequence-Unique MID sequence 3-EcoRI (NotI) adapter sequence B tag sequence-Unique MID sequence 3-EcoRI (NotI) adapter sequence Sample 4 A tag sequence-Unique MID sequence 4-EcoRI (NotI) adapter sequence B tag sequence-Unique MID sequence 4-EcoRI (NotI) adapter sequence

Sequencing

Our samples were sent to University of Pennsylvania Medical School for pyrosequencing, many such facilities are available throughout the US and elsewhere.

Quantifying the amount of prepared sample can be accomplished by electrophoresis or spectrophotometry. Based on the concentrations, the samples can be sequenced separately, or for multiplexed samples (when using primers with MID sequences) they can be mixed together in equimolar concentrations.

The sequencing facility uses a Roche/454 GS FLX machine. The 454 pyrosequencing technique uses polymerization and light to determine each nucleotide that has been added (Figure 1). Each nucleotide that is incorporated by DNA polymerase releases pyrophosphate, which begins a series of reactions that generate light from the firefly enzyme luciferase (Mardis 2008).

Figure 1. 454 sequencing method. 454 adapters are attached to the samples. The A adapter hybridizes to the complementary DNA sequence covalently bound to a DNA capture agarose bead. The B adapter primes the pyrosequencing DNA polymerization reaction. Polymerase binds to the B adapter of the DNA attached to an agarose bead to cause a reaction producing light. A camera records an image of the release of the pyrophosphate-producing light indicating nucleotide attachment within the machine until full reads encompass which nucleotide is present at every position in the DNA (http://454.com/img/content/technology-slide-7.jpg).

The 454 adapters attached to the samples have two sequences; the A end hybridizes to the complementary DNA sequence that is covalently bound to a DNA capture agarose bead. The B end is used to prime the pyrosequencing DNA polymerization reaction (Figure 1). Once a sequence has been attached to the bead, it is isolated in an oil/water micelle and emPCR (emulsion PCR) is performed to amplify each individual sequence on the beads to approximately one million copies per bead

(Mardis 2008).

To proceed in the sequencing process, the agarose beads are arrayed into

microwells on picotiter plates (PTP). Each well holds a single bead so that the 19

20 sequencing reactions can be observed on each immobilized bead. Smaller enzyme containing beads are added to the plate and centrifuged, falling into the microwells to surround the agarose bead and expedite the sequence reaction. Once the PTP is on the sequencing machine, pure nucleotide solutions are added one nucleotide at a time. The first four nucleotides of the primer are known and help to calibrate the machine. After a nucleotide has been added, a charge-coupled device (CCD) camera located across from the PTP records an image of the release of the pyrophosphate-producing light indicating nucleotide attachment. The machine then notes which nucleotide has been attached at each position in the plate. The number of the same nucleotide present in a row is indicated by a proportional amount of light. The sequencing machine flows about 100 rounds of each nucleotide over an eight-hour sequencing period. Hundreds of thousands of reads can be produced (Mardis 2008). The entire pyrosequencing process from emPCR to obtaining final sequence reads takes about 16 consecutive hours. After the sequencing service is completed, the fragmented reads are sent back to the researcher and analyzed.

Data Analysis

Assembling Data

Once all the 454 reads are recorded, the data must then be extracted from the data file using database management applications, such as Galaxy and FileMaker

(FileMaker, Inc., Santa Clara, CA). In FileMaker, primers can be identified and isolated from the sequenced fragment. The MID sequences can be located, and the sequences can be categorized by sample. The MID sequences also can be removed from the target 20

21 sequences in silico. On the Ohio Super Computer (OSC, Columbus, OH, USA), or analogous site, the sequences are then assembled, or joined, into “contigs” (contiguous lengths of sequence). This puts fragments together that have identical stretches of nucleotides on their ends. The contigs can be assembled using MIRA 3.0.5 (Whole

Genome Shotgun and EST [Expressed Sequence Tag] Sequence Assembler) or an analogous assembler application.

After assembly, any sequences from the control samples are considered to be contaminants. All sequences present in the control samples are eliminated from the ice core sample files without exclusion. After the isolation of unique sequences in the ice samples, microbes are assumed to have originated from the ancient ice. Unique isolated sequences can be used to identify the microbes. General taxonomic analyses are performed using BLAST (Basic Local Alignment Search Tool) searches. Identities (often to species) are confirmed using phylogenetic methods.

Batch BLASTN (BLAST nucleotide) similarity searches on the OSC (or similar site) can be used to specifically search the GenBank (NCBI) database (and other databases) in order to compare the sample sequences to the database online. The BLAST execution file is set up to retrieve a number of similar sequences (e.g., the top 10 similar sequences) limited by E-value cutoffs (e.g., 1-10). E-value describes how likely your matched score would have been found in the database by chance; it depends on the size of your database and query length. Therefore, identical short alignments have relatively high E values because there is a higher probability of them occurring by chance within the database. The lower the E-value (closer it is to zero) the more 21

"significant" the match is. The information determined is divided by sample and into three categories: rRNA genes, mRNA genes, and other regions. These categories are used in determining the taxonomic and gene identities present in the sample.

Phylogenetics

Phylogenetics helps to confirm the taxonomic identities and can lead to an understanding of the roles each of the species plays in a particular environment. It yields graphical representations of the relationships among the organisms by reconstructing a model of the evolutionary history of the taxa based on the changes in sequences, primarily by using ribosomal RNA gene sequences. Phylogenetics can be performed on a variety of sequences including DNA, RNA, and amino acids.

Nucleotide regions change at different rates. In order to compare the sequences, they must be carefully aligned. Often ribosomal RNA (SSU [small subunit] and LSU [large subunit]) genes are selected because they are found in every organism, vary enough to differentiate at the family to species levels (depending on the region used for alignments), but are conserved enough to align (Rogers 2012) from Bacteria through

Eukarya. It is thought that rRNA is almost always inherited vertically. Therefore, while the sequences used from the dataset are highly similar, they have enough differences to detect phylogenetic signals. Sequences are aligned to compare the nucleotides at each position, indicating similarities, as well as distinct mutations and differences. There is a particular order in which they are aligned: first matching nucleotides are aligned followed by transitions, transversions, and finally gaps (Rogers 2012). A quality alignment is pivotal to obtaining a reliable tree (Rogers 2012). 22

Similarities and differences are analyzed to generate phylogenetic trees. The trees that are produced from phylogenetic analysis can be used to understand the evolution of microbes over time, as well as to evaluate the relationships among the taxa.

Phylogenetic trees are graphical reconstructions of a likely evolutionary history given model assumptions. They consist of bifurcating branches connecting at a node.

Phylograms are scale drawings where the lengths of the horizontal branches (for rectangular trees) represent the differences in the sequences. The shorter branches indicate recent changes and longer branches indicate more distant changes (Rogers

2012). However, the vertical branches on rectangular trees simply connect the horizontal branches and confer no information about evolutionary distance. Often, when a group is distinctly different from the sequences being analyzed, and it is known to be ancestral, it is used as an outgroup (an outgroup is a sister group to the ingroup taxa). Evolutionary direction and timing can be indicated by correctly rooting a tree.

However, when evolutionary direction is unknown, unrooted trees can be presented.

Gaps in alignments represent mutations in sequences such as insertions, deletions, or genetic rearrangements. Gaps can be treated in a variety of ways in some phylogenetic analyses (including counting them as a fifth nucleotide type), but most often they are ignored.

While there are many phylogenetic methods for producing trees, neighbor-joining

(Saitou, et al. 1987) is a method that is relatively computationally simple, but often accurate. In this method, calculations are based on the number of proportional differences between pairs of sequences. The nodes are kept track of through the 23

24 calculations of distances from other taxa. Also, the branch lengths are not averaged, making this method more accurate than others, such as UPGMA (unweighted pair group method using arithmetic means).

Bootstrapping is a statistical analysis that is performed to confirm the support for the branch placement determined in the phylogenetic analysis. Bootstrapping takes data and builds new distance matrices based on the random selection of alignment columns, producing alternative tree topologies. It repeats this process from 100 to 1,000 times (Rogers 2012). The frequency of the presence of the specific branches is calculated as a percentage of occurrences among all of the trees determined in the 100 to 1,000 replications. The higher the percentage of times that a branch is found in the same location on a tree indicates the robustness of the branch at that location.

Microscopy of black particulate matter

Ice cores can also be analyzed by examining the contents of the meltwater microscopically (D'Elia, et al. 2008, 2009). Ice cores are environmental samples that contain not only microbes, but, in some cases, dust particles visible to the eye. There are many methods of microscopy such as light microscopy, fluorescence microscopy, and electron microscopy that can be used to examine ice core meltwater. One fluorescence microscopy technique uses a live-dead stain that differentiates between living and dead cells. Scanning electron microscopy can be used to examine small specimens, but has the disadvantage of only being able to view the surfaces of objects. The inside layers of the objects cannot be seen.

In order to better understand the environmental conditions at the time of snow 24

25 deposition, particulate matter from the ice cores can be chemically characterized using

X-ray elemental analysis in a scanning electron microscope. Before the sample can be put into the SEM microscope for viewing, certain measures must be taken to prepare the sample. The sample must be dehydrated because of the water vaporization that would occur under the low-pressure conditions inside the microscope chamber (Mehta

2012). Water vaporization would mean that part of the sample might be lost, as it becomes a gas, and loss of the vacuum in the microscope would cause problems with viewing of the sample and could contaminate the vacuum pumps with moisture.

Therefore, samples are dehydrated by a critical point dryer that replaces water with

CO2. After dehydration, samples are coated with gold or a mixture of gold/palladium

(60/40) to deflect the electrons in the microscope. To accomplish this, a device called a

“sputter coater” is used. An electric field and argon gas facilitates the formation of a uniform thin gold coating on the sample and stand.

When elemental analyses are performed using a SEM, electrons are aimed at the sample and an X-ray detector records the X-ray spectrum of the sample after the sample is hit with the electron beam. Each peak in X-ray emission is distinct for specific elements. Instead of gold, the sample is coated with carbon for elemental analysis, which stabilizes the sample in the beam of negatively charged electrons.

Some unique features of the scanning electron microscope (SEM) include a larger depth of field and a higher resolution than traditional microscopes (Mehta 2012). These features make the SEM one of the best tools for viewing the surfaces of small objects.

Another unique feature, and the reason that the SEM was chosen in particulate analysis, 25

26 is the chemical analysis capabilities of an energy-dispersive X-ray spectroscopy (EDS) detector in the SEM (Mehta 2012). This device is used to categorize the characteristic x- rays that are emitted by different elements in the electron beam (Mehta 2012). The EDS system allows scientists to determine the elements present in the sample. The black particulate matter in some ice core section samples can be volcanic ash that has been dispersed in the atmosphere and deposited in the ice. By determining the elements present from the X-ray spectrum of the ash/particles, eruptions from specific volcanoes or types of volcanoes can be resolved.

REFERENCES

1. Alekhina, I.A., D. Marie, J.R. Petit, V.V. Lukin, V.M. Zubkov, and S.A. Bulat. 2006.

Molecular analysis of bacterial diversity in kerosene-based drilling fluid from the

deep ice borehole at Vostok, East Antarctica. FEMS Microbiol Ecol 59:289–299.

2. Augustin L., C. Barbante, P.R.F. Barnes, J.M. Barnola, M. Bigler, E. Castellano, O.

Cattani, J. Chappellaz, D. Dahl-Jensen, B. Delmonte, G. Dreyfus, G. Durand, S.

Falourd, H. Fischer, J. Flückiger, M.E. Hansson, P. Huybrechts, G. Jugie, S.J.

Johnsen, J. Jouzel, P. Kaufmann, J. Kipfstuhl, F. Lambert, V.Y. Lipenkov, G.C.

Littot, A. Longinelli, R. Lorrain, V. Maggi, V. Masson-Delmotte, H. Miller, R.

Mulvaney, J. Oerlemans, H. Oerter, G. Orombelli, F. Parrenin, D. A. Peel, J.R.

Petit, D. Raynaud, C. Ritz, U. Ruth, J. Schwander, U. Siegenthaler, R. Souchez, B.

Stauffer, J.P. Steffensen, B. Stenni, T.F. Stocker, I.E. Tabacco, R. Udisti, R.S.W. van

de Wal, M. van den Broeke, J. Weiss, F. Wilhelms, J.G. Winther, E.W. Wolff and

M. Zucchelli (EPICA community members, participants are listed alphabetically).

2004. Eight glacial cycles from an Antarctic ice core. Nature 429:623-628.

3. Bakermans, C., A.I. Tsapin, V. Souza-Egipsy, D.A. Gilichinsky and K.H. Nealson.

2003. Reproduction and metabolism at -10°C of bacteria isolated from Siberian

permafrost. Environ Microbiol 5:321–326.

4. Bidle K.J., S. Lee, D.R. Marchant, and P.G. Falkowski. 2007. Fossil genes and

microbes in the oldest ice on Earth. P Natl Acad Sci-Biol 104:13455–13460.

5. Catranis C.M., W.T. Starmer. 1991. Microorganisms entrapped in glacial ice. Antarc

J US 26:234-236. 27

6. Cowan D., Q. Meyer, W. Stafford, S. Muyanga, R. Cameron, and P. Wittwer. 2005.

Metagenomic gene discovery: past, present and future. TRENDS in Biotechnology

23:321-329.

7. D'Elia, T., R. Veerapaneni, V. Theraisnathan and S.O. Rogers. 2009. Isolation of

fungi from Lake Vostok accretion ice. Mycologia 101:751-763.

8. D'Elia, T., R. Veerapaneni, V. Theraisnathan, and S.O. Rogers. 2008. Isolation of

microbes from Lake Vostok accretion ice. Appl Environ Microbiol 74:4962-4965.

9. DeLong, E.F. and N.R. Pace. 2001. Environmental diversity of bacteria and archaea.

Syst Biol 50:470-478.

10. Deming, J. W. 2002. Psychrophiles and polar regions. Current opinion in

microbiology 5:301-309.

11. Deng, M.Y., H. Wang, G.B. Ward, T.R. Beckham, T.S. McKenna. 2005. Comparison

of six RNA extraction methods for the detection of classical swine fever virus by

real-time and conventional reverse transcription–PCR. J Vet Diagn Invest 17:574–

578.

12. Driessen, A.J.M., J.L.C.M. van de Vossenberg, and W.N. Konings. 1996. Membrane

composition and ion-permeability in extremophiles. FEMS Microbiology Reviews

18:139-148.

13. Dupré J. and M.A. O’Malley. 2007. Metagenomics and biological ontology. Stud

Hist Biol & Biomed Sci 38:834-846.

14. Finegold L. 1996. Molecular and biophysical aspects of adaptation of life to

temperatures below the freezing point. Adv Space Res 18:87-95. 28

15. Fischer, H. M. Siggaard-Andersen, U. Ruth, R. Röthlisberger, and E. Wolff. 2007.

Glacial/interglacial changes in mineral dust and sea-salt records in polar ice cores:

sources, transport, and deposition. Rev Geophys 45:1-26.

16. Fong N.J.C., M.L. Burgess, K.D. Barrow, and D.R. Glenn. 2001. Carotenoid

accumulation in the psychrotrophic bacterium Arthrobacter agilis in response to

thermal and salt stress. Appl Microbiol Biotechnol 56:750-756.

17. Gómez, B.L. and J.D. Nosanchuk. 2003. Melanin and fungi. Curr Opin Infect Dis

16:91–96.

18. Great Britain. Admiralty, Royal Society (Great Britain). Arctic Committee. Manual

of the natural history, geology, and physics of Greenland, and the neighboring

regions: prepared for the use of the Arctic Expedition of 1875, under the direction of

the Arctic Committee of the Royal Society, for the use of the expedition Published by

authority of the Lords Commissioners ...[Internet]. London (UK):H.M. Stationery

Off., printed by G.E. Eyre and W. Spottiswoode; 1875 [cited 2012 January 21].

Available from

http://books.google.com/books?id=SdYBAAAAYAAJ&pg=PA608&lpg=PA608&d

q=greenland+prevailing+winds&source=bl&ots=B_ETIjMynB&sig=E5VNYRcHUdq

V1Ks60UvYTlzIeG8&hl=en&sa=X&ei=JAUbT__jIcr20gHM_aDkCw&ved=0CEgQ6A

EwBg#v=onepage&q=greenland%20prevailing%20winds&f=false

19. Hamer, C.U., H.B. Clausen, C.C. Langway. 1994. Electrical conductivity method

(ECM) stratigraphic dating of the Byrd Station ice core, Antarctica. Annals of

Glaciology 20:115-120. 29

20. Mardis, E.R. 2008. Next-Generation DNA sequencing methods. Annu Rev Genomics

Hum Genet 9:387-402.

21. Meese D.A., A.J. Gow, G.A. Zielinski, P.M. Grootes, M. Ram, K.C. Taylor, P.A.

MayewskiJ, F. Bolzan, and R.B. Alley. 1997. The Greenland Ice Sheet Project 2

depth-age scale: methods and results. J Geophys Res 102:26,411-26,423.

22. Mehta, R. 2012. Interactions, Imaging and Spectra in SEM, Scanning Electron

Microscopy, Viacheslav Kazmiruk (Ed.), InTech, Available

from:http://www.intechopen.com/books/scanning-electron-

microscopy/interactions-imaging-spectra-in-sem.

23. Miteva V, C. Teacher, T. Sowers, and J. Brenchley. 2009. Comparison of the

microbial diversity at different depths of the GISP2 Greenland ice core in

relationship to deposition climates. Environ Microbiol 11:640-656.

24. Montagnat M. and P. Duval. 2000. Rate controlling processes in the creep of polar

ice, influence of grain boundary migration associated with recrystallization. Earth

and Planetary Science Letters 183:179-186.

25. Murphy, K.M., P. Travers, and M. Walport. 2007. Janeway's Immunobiology: The

immune system 7th edition, New York: Garland Science.

26. Pagli, C. and F. Sigmudsson. 2008. Will present day glacier retreat increase volcanic

activity? Stress induced by recent glacier retreat and its effect on magmatism at the

Vatnajökull ice cap, Iceland. Geophys. Res. Lett. 35:L09304.

27. Paterson, W.S.B. 1994. The transformation of snow to ice. In: The physics of glaciers.

Third edition. Burlington, MA. Elsevier. Pp 11-28. 30

http://books.google.com/books?id=Jca2v1u1EKEC&pg=PA11&source=gbs_toc_r&

cad=4#v=onepage&q&f=false.

28. Petit, J.R., J. Jouzel, D. Raynaud, N. I. Barkov, J.M. Barnola, I. Basile, M. Bender, J.

Chappellaz, M. Davis, G. Delaygue, M. Delmotte, V.M. Kotlyakov, M. Legrand,

V.Y. Lipenkov, C. Lorius, L. PÉpin, C. Ritz, E. Saltzman, and M. Stievenard. 1999.

Climate and atmospheric history of the past 420,000 years from the Vostok ice core,

Antarctica. Nature 399:429-436.

29. Price P.B. 2000. A habitat for psychrophiles in deep Antarctic ice. Proc Natl Acad Sci

97:1247-1251.

30. Roberts M.E. and W.E. Innis. 1992. The synthesis of cold shock proteins and cold

acclimation proteins in the psychrophilic bacterium Aquaspirillum arcticum. Curr

Microbiol 25:275- 278.

31. Rodríguez-Valera F. 2004. Environmental genomics, the big picture? FEMS

Microbiol Lett 231:153-158.

32. Rogers S. O., V. Theraisnathan, L.J. Ma, Y. Zhao, G. Zhang, S.G. Shin, J.D.

Castello, and W.T. Starmer. 2004. Comparisons of protocols for decontamination of

environmental ice samples for biological and molecular examinations. Appl Environ

Microbiol 70:2540-2544.

33. Rogers S.O. and A.J. Bendich. 1985. Extraction of DNA from milligram amounts of

fresh, herbarium and mummified plant tissues. Plant Mol Biol 5:69-76.

34. Rogers, S. O. 2012. Phylogenetics. In: Integrated molecular evolution. Boca Raton,

FL: CRC Press. Pp. 209-229. 31

35. Rogers, S.O, W.T. Starmer, and J.D. Castello. 2004. Recycling of pathogenic

microbes through survival in ice. Med Hypoth 63:773-777.

36. Rogers, S.O. and A.J. Bendich. 1994. Extraction of total cellular DNA from plants,

algae and fungi, p. D1: 1-8. In S.B. Gelvin, R.A. Schilperoort (ed.), Plant molecular

biology manual, 3rd Edition. Kluwer Academic Press, Dordrecht, Netherlands.

37. Rogers, S.O., S. Rehner, C. Bledsoe, G.J. Mueller and J.F. Ammirati. 1989.

Extraction of DNA from Basidiomycetes for ribosomal DNA hybridizations. Can J

Bot 67:1235-1243.

38. Saiki, R.K., G.H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G.T. Horn, K.B.

Mullis and H.A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with

thermostable DNA polymerase. Science 239:487-49.

39. Saitou, N. and M. Nei. 1987. The neighbor-joining method: a new method for

reconstructing phylogenetic trees. Mol Biol Evol 4:406-25.

40. Schloss P.D. and J. Handelsman. 2005. Metagenomics for studying unculturable

microorganisms: cutting the gordian knot. Genome Biol 6:229.1-299.4.

41. Shivji, M.S., S.O. Rogers and M.J. Stanhope. 1992. Rapid isolation of high

molecular weight DNA from marine macroalgae. Marine Ecol. Prog. Ser. 84:197-203.

42. Tringe S, and E. Rubin. 2005. Metagenomics: DNA sequencing of environmental

samples. Nature Reviews Genetics 6:805-814.

43. Uchida, T., S. Takeya, M. Nagayama and K. Gohara. 2012. Freezing properties of

disaccharide solutions: inhibition of hexagonal ice crystal growth and formation of

cubic ice. In: Crystallization and materials science of modern artificial and natural 32

crystals. Rijeka, Croatia: InTech. Pp. 203-224.

44. Vorobyova E., V. Soina, M. Gorlenko, N. Minkovskaya, N. Zalinova, A.

Mamukelashvili, D.A. Gilichinsky, E. Rivkina, and T. Vishnivetskaya. 1997. The

deep cold biosphere: facts and hypothesis. FEMS Microbiol Rev. 20:277–290.

45. Willerslev E., A.J. Hansen, and H.N. Poinar. 2004. Isolation of nucleic acids and

cultures from fossil ice and permafrost. Trends Ecol Evol 19:141-147.

46. Zdanowicz, C.M., G.A. Zielinski, and M.S. Germani. 1999. Mount Mazama

eruption: calendrical age verified and atmospheric impact assessed. Geology 27:621-

624.

47. Zielinski, G.A., P.A. Mayewski, L.D. Meeker, S. Whitlow, and M.S. Twickler.

1995. A 110,000-yr record of explosive volcanism from the GISP2 (Greenland) ice

core. Quaternary Res 45:109-111.

CHAPTER II.

RESEARCH

2.1 Introduction

A combined metagenomic/metatranscriptomic approach was used to determine and compare the microbial community structures at a time when carbon dioxide levels and temperatures were at minima (157,000 ybp) (Lüthi, et al. 2008) and at a time when carbon dioxide levels and temperatures were rising nearly to modern levels (10,500 ybp) (Ahn, et al. 2008, Fischer, et al. 1999, Suwa, et al. 2006). During the past half million years, the concentration of atmospheric carbon dioxide has ranged from a low of approximately 200 ppm (parts per million) 150,000 years ago (Lüthi, et al. 2008) to a high of over 390 ppm currently (Glikson 2008), with another high peak of approximately 280 ppm occurring approximately 125,000 years ago (Petit, et al. 1999).

Carbon dioxide levels vary directly with changes in global temperature (Petit, et al.

1999, Figure 2). Low CO2 levels are normally associated with low temperatures, low precipitation rates, and high dust levels, while high CO2 levels generally indicate high temperatures, high precipitation rates, and low dust levels, although dust levels are more variable.

Figure 2. Graph of past atmospheric changes. Atmospheric CO2 concentrations, dust levels, and temperature variation (compared to current mean temperature) during the past 160,000 years, as determined from examination of the Vostok 5G ice core. CO2 concentrations and dust levels are measured in parts per million (ppm). ∆ Temperature is the change in past temperature and current temperature. The two gray bars denote the two periods of time represented in the ice core sections metagenomically/metranscriptomically analyzed. (Petit et al. 1999).

Previous research indicates microbes that are likely to be found in the 157,000 year old ice include Proteobacteria, Firmicutes, and Actinobacteria (Miteva, et al. 2009,

Simon, et al. 2009). Those likely to be found in the 10,500 year old ice include

Proteobacteria and Firmicutes (Miteva, et al. 2009, Simon, et al. 2009). The differences could relate to the global mean temperature being approximately 7-8°C (12-14°F) cooler

157,000 years ago than it was 10,500 years ago (Pagli et al. 2008).

The objective of this study was to compare microbial sequences in two

Greenland glacial ice core sections with differing CO2 levels (and two different 36

37 temperature ranges) using BLAST and/or phylogenetic analyses of the metagenomic and metatranscriptomic sequences.

2.2 Materials and Methods

2.2.1 Ice core selection and samples

Three ice core sections (104 m, 1593 m, and 2131 m; approximately 500, 30,000, and 70,000 ybp, respectively (Hamer, et al. 1994)) from Byrd Station (80° 1’ S, 119° 31’

W), Antarctica and three core sections (99 m, 1601 m, and 3014 m; approximately 500,

10,500, and 157,000 ybp, respectively) from GISP 2D (Greenland Ice Sheet Project, core

2D; 72o 36' N, 38o 30'W) were selected. The ice cores were chosen based on measurements of extremes in CO2 levels and temperatures in the atmosphere. CO2 levels were low 85,000 and 157,000 ybp, and rose 10,500 ybp (Petit, et al. 1999, Figure 2).

Relatively high CO2 levels were present 500 ybp (Petit, et al. 1999, Figure 2).

2.2.2 Surface sterilization and melting

A 16 cm length of each ice core section was cut with an autoclaved saw and was warmed in a refrigerator at 4˚C for one hour in a sterile beaker. Each section was then moved to a clean room with HEPA-filtered air, in which the stainless steel lab benches and hood surfaces were wiped with Clorox and then 70% ethanol. Additionally, the room and hood had been bathed in UV light for 30 minutes. The ice core sections were melted in a sterile Class II laminar flow biological safety cabinet according to a method previously described (Rogers, et al. 2004b). First, the outer surfaces of the ice were decontaminated with 700 ml of 5.25% sodium hypochlorite (undiluted Clorox, chilled to

4˚C) for 10 to 20 seconds, followed by two rinses of 1200 ml autoclaved water (18.2 MΩ, 37

<1ppb [parts per billion] TOC [total organic carbon], chilled to 4˚C) (Rogers, et al.

2004b). Then, the ice core section was placed into a sterile Büchner funnel and melted in the sterile hood at room temperature. Meltwater was collected into sterile BD Falcon 50 ml conical tubes. Depending on the size of the ice core section, 3-7 tubes were filled with meltwater. Each tube represents a “shell.” The first shell corresponded to the outermost layer of the ice core section, while the last shell corresponded to the innermost layer.

Throughout the entire melting process, control culture plates (2 malt extract agar [Per liter: 12.75 g maltose technical, 2.75 g dextrin, 2.35 g glycerol, 0.78 g peptone, and 15 g agar] and 2 Luria-Bertani medium (LB) [Per liter: 10 g tryptone, 5 g yeast extract, 10 g sodium chloride, and 15 g agar]) were kept open in the hood. At the end of the melting procedure, all of the plates were covered and then incubated at 15˚C. They were monitored every other week for growth for one year to assay for any contaminating organisms that may have been introduced during the process.

2.2.3 Culturing

Immediately after melting, some of the meltwater was used for culturing and the remaining meltwater was stored at -20˚C for later analysis. Ten plates per ice core section (5 malt extract agar and 5 LB agar plates, as above) each were inoculated with

200 µl of meltwater and incubated at 15˚C. Once growth was observed (approximately 1 month after the first melt) individual colonies were subcultured using the same media.

To test growth conditions, the subcultures were placed at three temperatures (4˚, 8˚, and

15˚C) (D’Elia, et al. 2008).

2.2.4 Ultracentrifugation

Two samples were selected for metagenomic/metatranscriptomic analysis, GISP

2D 1601 m and 3014 m. The former sample corresponds to 10,500 ybp (Ahn, et al. 2008), a time of moderate and rising atmospheric CO2 and rising temperatures, while the latter corresponds to 157,000 ybp, a time of low atmospheric CO2, and low temperatures

(Miteva, et al. 2009) (Figure 2). Starting from shell 2 inward, the meltwater was subjected to ultracentrifugation at 32,500 rpm (100,000 rcf) in a Beckman type 60 Ti rotor at 4˚C for 15-17 hours to concentrate cells, viruses, and nucleic acids. The supernatants were placed into sterile BD Falcon 50 ml conical tubes and stored at -20˚C. Each pellet was resuspended in 50 µl of 0.1 X TE (1 mM Tris HCl, pH 8.0; 0.1 mM EDTA, pH 8.0) and then the suspensions from each tube (for a given meltwater sample) were pooled

(~500 µl total) and stored at -20˚C. Two additional samples were used as controls. The first control was autoclaved Nanopure water (18.2 MΩ, <1ppb TOC) that was subjected to ultracentrifugation, as above. Nanopure water (18.2 MΩ, <1ppb TOC) was the second control.

2.2.5 Isolation of RNA

Each concentrated meltwater sample (as well as the two controls) was divided into two 200 µl fractions. One fraction was used for RNA extraction using TRIzol LS reagent (Life Technologies, Grand Island, NY). To carry out homogenization, three volumes of TRIzol LS reagent was added, mixed, and incubated at 55°C to allow complete dissociation of proteins. Next, an equal volume of chloroform was added,

followed by vigorously mixing. After incubation at room temperature, the samples 39

40 were centrifuged in a microfuge at 16,100 rcf, which resulted in three layers. The top aqueous layer, containing the RNA, was removed and the RNA was precipitated by adding 400 µl of isopropanol. This was placed at -20°C for 2 hours, followed by centrifugation in a microfuge to pellet the RNA. This pellet was washed with 800 µl 80% ethanol, air-dried at room temperature in a sterile hood, and stored at -80˚C until needed. Directly before use, the RNA pellet was resuspended in 20µl of sterile RNAse free water.

2.2.6 Isolation of DNA

The second 200 µl fraction of pooled ultracentrifuged meltwater was used for

DNA extraction using a CTAB (cetyltrimethylammonium bromide) method (Rogers &

Bendich 1985, Rogers & Bendich 1994). To begin the extraction process, equal amounts of sample and 2X CTAB solution (2% CTAB w/v, 100 mM Tris [pH 8.0], 20 mM EDTA

[pH 8.0], 1.4 M NaCl) were mixed and placed into a 65˚C water bath for ten minutes.

Then, an equal volume of chloroform/isoamyl alcohol (24:1) was added, followed by vigorous mixing, and centrifugation in a microcentrifuge. The upper layer containing the DNA was transferred to a new microfuge tube and mixed with 0.6 volumes of cold isopropanol. Precipitation was allowed to continue at -20˚C overnight. The precipitated

DNA was pelleted by centrifugation for 15 minutes in a microcentrifuge. Following two, 200 µl 80% ethanol washes, the pellet was dried at room temperature in a vacuum microcentrifuge (Eppendorf, Hamburg, Germany) and then rehydrated in 25 µl of High

Salt TE buffer (10 mM Tris [pH 8.0]; 1mM EDTA; 1 M NaCl). Residual CTAB was removed by again precipitating with 2.5 volumes of cold ethanol. Precipitation was 40

41 allowed to proceed at -20°C overnight. This was followed by centrifugation for 15 minutes in a microcentrifuge to pellet the DNA and washing with 75 µl of 80% ethanol one time. Then, the DNA pellet was dried in a vacuum microcentrifuge for 15 min and rehydrated in 25 µl of 0.1X TE buffer. The DNA solutions were stored at -20˚C until needed.

2.2.7 Preparation of samples for sequencing

Complementary DNA (cDNA) was synthesized from the extracted RNA. The following procedure was performed using SuperScript Choice cDNA kit (Invitrogen,

Grand Island, NY). To begin first strand synthesis, 10 µl of extracted RNA (less than 1 ng/µl) was combined with ~80 pmol of random hexamer primers from the SuperScript

Choice System™, and incubated at 70˚C for 10 minutes, followed by chilling on ice.

Then, the full reaction mixture was added, consisting of: 50 mM Tris-HCl (pH 8.3); 75 mM KCl; 3 mM MgCl2; 10 mM DTT; 500 μM each dATP, dCTP, dGTP, and dTTP; and

200 units SuperScript™ II RT; in 21 μl total volume. This was incubated at 37°C for 1.5 hours, and then placed on ice.

Second strand synthesis was completed in a reaction mixture containing 25 mM

Tris-HCl (pH 7.5); 100 mM KCl; 5 mM MgCl2; 10 mM (NH4)2SO4; 0.15 mM ß-NAD+;250

μM each dATP, dCTP, dGTP, and dTTP; 1.2 mM DTT; 10 units DNA ligase; 40 units

DNA polymerase I; and 2 units RNaseH; in 150 μl total volume. This was incubated at

15˚C for 2.5 hours. A total of 10 units of T4 DNA polymerase were added and the reactions were incubated at 15˚C for and additional 7 minutes. The tube was placed on ice and the reaction was stopped by the addition of EDTA to a concentration of 30 mM. 41

This was followed by the addition of 150 µl of chloroform/isoamyl alcohol (24:1), which was mixed by vortexing to form an emulsion. Then, the phases were separated by centrifugation in a microcentrifuge (16,000 rcf) for 2 minutes. Upon transferring the upper layer into a fresh tube, 100 µl of 5 M NaCl was added and mixed. Then, 500 µl of

100% ethanol (-20°C) was added, mixed, and placed at -20˚C for 10 minutes. Then, they were centrifuged in a microfuge for 20 minutes and the supernatant was decanted. The pellet was washed with 500 µl of cold 80% ethanol, gently mixed, centrifuged for 10 minutes and the ethanol was removed by decanting. The cDNA was dried at 37˚C in a vacuum microcentrifuge for 25 minutes.

The cDNA was then mixed with 10 µl of extracted DNA (less than 1 ng/µl) from the same meltwater sample, and EcoRI (Not I) adapters (AATTCGCGGCCGCGTCGAC, dsDNA) were added using T4 DNA ligase. The final concentration of components in each reaction for EcoRI adapter addition was: 66 mM Tris-HCl (pH 7.6), 10 mM MgCl2 ,

1 mM ATP, 14 mM DTT, 100 pmols EcoRI (Not I) adapters, and 0.5 units of T4 DNA ligase, in 50 µl total volume. The reaction was incubated at 15˚C for 20 hours. Then, the reaction was heated to 70˚C for 10 minutes to inactive the ligase.

The products were size fractionated by column chromatography. Each 2 ml plastic column contained 1 ml of Sephacryl® S-500 HR resin. TEN buffer (10 mM Tris-

HCl [pH 7.5], 0.1 mM EDTA, 25 mM NaCl; autoclaved) was utilized in washing the columns and eluting the samples through the columns. Fractions of approximately 40 µl were collected. After measurement of the volume of each fraction, fractions 6-18 were precipitated to concentrate the DNA. Concentration consisted of adding 0.5 volumes (of 42

43 the fraction size) of 1 M NaCl, and two volumes of -20˚C absolute ethanol. After gentle mixing, each was left to precipitate at -20˚C overnight. The fractions were centrifuged in a microfuge for 20 minutes at room temperature and the liquid was removed by decanting. Then, the pellets were washed with 0.5 ml of -20˚C 80% ethanol and centrifuged for 5 minutes. After removal of the ethanol, each of the DNA pellets was dried under vacuum (as above) and rehydrated in 20 µl of 0.1X TE buffer.

After rehydration, fractions 6-18 were subjected to PCR amplification using EcoRI

(NotI) adapter primers (AATTCGCGGCCGCGCTCGAC). The samples were amplified using a GeneAmp (with AmpliTaq) PCR Reagent Kit (Applied Biosystems, Carlsbad,

California, USA). Each reaction mixture contained: 5 pmol each dATP, dCTP, dGTP, and dTTP; 1X PCR Buffer (10 mM Tris-HCl [pH 8.3], 50 mM KCl, 1.5 mM MgCl2, 0.001%

(w/v) gelatin); 1 unit AmpliTaq DNA polymerase; and 50 pmols EcoRI (NotI) adapter primers, each in 25 µl total volume. In order to carry out the PCR reaction, two BIO-

RAD PTC Peltier Thermal Cyclers were used. The thermal cycling program was: 94˚C for 4 min; then 40 cycles of 94˚C for 1 min, 55˚C for 2 min, 72˚C for 2 min; followed by an incubation for 10 min at 72˚C. Each fraction was subjected to gel electrophoresis on

1% agarose gels in TBE (89 mM Tris base, 89 mM Boric acid, 2 mM EDTA) with 0.5 ug/ml ethidium bromide to confirm amplification.

Primers specific to 454 were used to further amplify the fragments, and to add the specific 454 sequences (required for pyrosequencing) to the ends of the amplified products. All of the primers included the A-tag or the B-tag sequence

(CGTATCGCCTCCCTCGCGCCA, CTATGCGCCTTGCCAGCCCGC, respectively), as 43

44 well as the short 4-base tag sequence (TCAG), unique multiplex identifiers (MID sequences [MID4: AGCACTGTAG, MID5:ATCAGACACG, MID6:ATATCGCGAG,

MID10:TCTCTATGCG]), and EcoRI (NotI) adapter sequences

(AATTCGCGGCCGCGTCGAC)(Table 2).

Table 2. Primer names and sequences used in amplification for 454 sequencinga.

Sample Primer Name Primer Sequence Ice, 1601 MetaMID4_454 CGTATCGCCTCCCTCGCGCCATCAGAGCACTGTAGAATTCGCGGCCGCGTCGAC Ice 1601 MetaMID4B_454 CTATGCGCCTTGCCAGCCCGCTCAGAGCACTGTAGAATTCGCGGCCGCGTCGAC Ice 3014 MetaMID5_454 CGTATCGCCTCCCTCGCGCCATCAGATCAGACACGAATTCGCGGCCGCGTCGAC Ice 3014 MetaMID5B_454 CTATGCGCCTTGCCAGCCCGCTCAGATCAGACACGAATTCGCGGCCGCGTCGAC Control UC H2O MetaMID6_454 CGTATCGCCTCCCTCGCGCCATCAGATATCGCGAGAATTCGCGGCCGCGTCGAC Control UC H2O MetaMID6B_454 CTATGCGCCTTGCCAGCCCGCTCAGATATCGCGAGAATTCGCGGCCGCGTCGAC Control H2O MetaMID10_454 CGTATCGCCTCCCTCGCGCCATCAGTCTCTATGCGAATTCGCGGCCGCGTCGAC Control H2O MetaMID10B_454 CTATGCGCCTTGCCAGCCCGCTCAGTCTCTATGCGAATTCGCGGCCGCGTCGAC aColor code: the A-tag sequence (blue), the B-tag sequence (purple), the short 4-base tag sequence (orange), unique MID sequences (red) and EcoRI (NotI) adapter sequences (green). 45

The samples were amplified using a GeneAmp (with AmpliTaq) PCR Reagent Kit

(Applied Biosystems, Carlsbad, California, USA). The final concentration of each reaction was 5 pmol of dATP, dCTP, dGTP, dTTP, 1X PCR Buffer [10 mM Tris-HCl, pH

8.3, 50 mM KCl, 1.5 mM MgCl2, 0.001% (w/v) gelatin], 1 unit AmpliTaq DNA polymerase and 25 pmol of each A and B MID primer, in 25 µl total volume. The thermal cycling program was: 94˚C for 4 min; then 40 cycles of 94˚C for 1 min, 55˚C for 3 min, 72˚C for 3 min; followed by an incubation for 10 min at 72˚C. All PCR products were cleaned with a PCR purification kit (Qiagen, Valencia, CA). Once the sizes to be used for sequencing were confirmed by gel electrophoresis, the fractions containing distributions from approximately 200 bp to 1.5 kb were pooled. After concentration by ethanol precipitation and rehydration (as above), the samples were quantified on 1% agarose gels in TBE, with ethidium bromide (as above) using serial dilutions of plasmid

(pGEM-3Z, from Promega, Madison, WI) for quantitation (Figure 3). 46

Figure 3. Gel for quantifying DNA. Agarose gel with serial dilutions of plasmid (pGEM®-3Z Vector) used for quantifying DNA concentrations of the samples. The plasmid was diluted to 75 ng, 50 ng, 25 ng, and 10 ng, in lanes 1, 2, 3, 4, respectively. Sample DNAs are in lanes 5-8. One µl of each sample was loaded onto the gel allowing estimates of the concentrations of DNA on the gel. DNA concentrations were estimated to be ~150 ng/µl for GISP 2D 3014 m (sample 1; lane 5); ~200 ng/µl for GISP 2D 1601 m (sample 2; lane 6); ~150 ng/µl for the ultracentrifuge sterile Nanopure water (sample 3, lane 7); and finally the ~250 ng/µl for the unconcentrated Nanopure water (lane 8). Lane 9 contains the DNA marker.

2.2.8 Sequencing

DNA concentrations were estimated to be ~150 ng/µl for GISP 2D 3014 m

(sample 1); ~200 ng/µl for GISP 2D 1601 m (sample 2); ~150 ng/µl for the ultracentrifuge sterile Nanopure water (sample 3); and finally the ~250 ng/µl for the

unconcentrated Nanopure water (Figure 3). These were mixed in equimolar amounts in 47

48 the proportion of 4:3:4:2, respectively, based on their estimated concentrations. Eight µl of sample 1, 6 µl of sample 2, 8 µl of sample 3, and 4 µl of sample 4 were mixed to produce a composite sample for sequencing. To this 26 µl mixture, 28 µl of RNase free water was added. A total of 54 µl (4.6 mg) of the amplified DNA at an approximate concentration of 85 ng/ml was sent to the University of Pennsylvania Medical School

Sequencing Facility for Roche/454 GS FLX service.

2.2.9 Analysis of sequences

Sequences were uploaded to the Ohio Super Computer (OSC) server. They were separated into four files based on their MID sequences (corresponding to the four ice core and control samples). Then, the 454 primers were clipped off (in silico) and sequence assembly was performed. The assembled sequences were used in Megablast searches for determination of sequence, taxon, and gene similarities. The Megablast results (with the E-value cut off parameter 1-10) were converted into a database format using FileMaker Pro11 (FileMaker Inc., Santa Clara, California). Accession numbers for the hit sequences were exported and used in Batch Entrez NCBI to retrieve similar sequences. Database sequences were reorganized based on gene regions. Multiple sequence alignments for rRNA SSU and LSU (ribosomal RNA small subunit and large subunit) genes, separately, were performed with MAFFT (Multiple Sequence

Alignment based on Fast Fourier Transform, Kyoto University, Kyoto, Japan: http://mafft.cbrc.jp/alignment/software/) (Katoh et al. 2002, Katoh et al. 2005) global and local alignment tools. Alignment files were converted into NEXUS format with

SeaView (Université de Lyon, Villeurbanne, France: http://pbil.univ- 48

49 lyon1.fr/software/seaview.html) (Gouy et al. 2010). PAUP (Phylogenetic Analysis

Using Parsimony, Swofford 2001) neighbor-joining phylogenetic reconstructions were performed on the aligned ribosomal (SSU and LSU) gene sequences (Saitou et al. 1987,

Fierer et al. 2007). Files were constructed for each gene region containing sequences similar to uncultured and unidentified organisms that were retrieved from the databases. Positions of uncultured sequences were based on their affinities to known taxa. Results from the phylogenetic analyses of the sequences from uncultured samples allowed us to reevaluate the microbial compositions of our two ice samples.

2.2.10 Scanning Electron Microscope (SEM)

In order to document and characterize the particulate matter in the 10,500 ybp

GISP2D sample, elemental analysis was performed using an SEM, equipped with an energy-dispersive X-ray spectroscopy (EDS) detector, which measures the x-ray spectral profile of components excited by the electron beam. Black particulate matter was taken from the 10,500 ybp concentrated sample. These particles were mounted on stubs and coated with carbon. One carbon stub was used and one silicon (glass) stub was used in order to obtain accurate readings of elements spectra present in the sample.

2.2.11 Sequencing of cultures

After subculturing, DNA was extracted (as above) from the growing cultures of meltwater. The extracted DNA was amplified using a GeneAmp PCR Reagent Kit (as above) and a variety of fungal ITS and SSU bacterial primers (Table 3), in 25 µl total volume. Each reaction contained 25 pmol of the forward and 25 pmol of the reverse

primer. The reaction program was as described above. After amplification the samples 49

50 were quantified on a 1% agarose gel in TBE with ethidium bromide (as above). Samples were cleaned up using Exo-SAP IT (Affymetrix, Inc., USB Products, Cleveland, Ohio).

Exo-SAP IT is composed of Exonuclease I and Shrimp Alkaline Phosphatase. It is a special buffer that removes any remaining primer and dNTPs from the PCR reaction.

Samples were sent to GENEWIZ (South Plainfield, NJ) for Sanger sequencing. The sequences obtained were subjected to BLAST searches and phylogenetic analyses in the same manner as the metagenomic/metatranscriptomic sequences.

Table 3. Primer names and sequences used in amplification of culturesa.

Primer Nameb Primer Sequence Eub16S_1245R CGCTTCTCTTTGTATGCGCCA Eub16S_1336R ATTCTGATCCACGATTACTAGCGAT Eub16S_1054F CATGGCTGTCGTCAGCTCGT Eub16S_611F CCCCGGGCTCAACCTGGGAA Eub16S1 CGGTGGCGAAGGCGGCCC Eub16S2 CACACCGCCCGTCACACCATG Eub16S3 CATGGTGTGACGGGCGGTGTG Eub23S5 ATTACCACGTCCTTCATCGCCTCTG ITS2 GCTGCGTTCTTCATCGATGC ITS3 GCATCGATGAAGAACGCAGC ITS4 TCCTCCGCTTATTGATATGC ITS5 GGAAGTAAAAGTCGTAACAAGG aColor code: the bacterial primers amplifying a region of the 16S ribosomal RNA gene, 16S-23S ribosomal RNA intergenic spacer and/or 23S ribosomal RNA gene (red) and the fungal primers amplifying ITS regions (green). bAll primers beginning with “Eub” were designed in our lab. ITS primers are from White et al. 1990. 2.3 Results

A total of 107,700 high quality 454 sequence reads were obtained from the sequencing facility. After categorization according to MID primers, there were 25,748 high quality reads associated with the MID4 (GISP1601) primer, 29,506 associated with the MID5 (GISP 3014) primer. Primer MID6 (control 1) had 18,015 high quality reads, 50

51 and 34,063 high quality reads were associated with the MID10 (control 2) primer.

During organization by MID primer, 368 reads were immediately eliminated from further consideration because they were tandemly repeated primers. After assembly, a total of 15,891 contigs resulted. BLAST results indicated a large number of sequences from the ice core samples that were in common with the control samples. After removing all sequences that were potentially contaminants, 39 sequences remained that were unique to the ice for subsequent analysis (Table 4 and 5).

Table 4. rRNA sequences from the GISP2D ice core sectionsa. Sequence Accession Percent Name Numbersb E-value Similarity Descriptionc MID4_rep_c4392 KC146571 0 93% Escherichia coli O157:H7 EDL933, 23S ribosomal RNA gene region [AE005174] MID4_rep_c3088 KC146584 8e-90 92% Halomonas neptunia, 23S rRNA gene, strain CECT 5815 [FN257755] MID4_s1867 KC155351 0 98% Lactobacillus crispatus ST1, strain ST1, 23S ribosomal RNA gene region [FN692037] MID4_rep_c2581 KC155352 0 100% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID4_s3382 KC146557 3e-149 99% Microcoleus vaginatus, 16S ribosomal RNA gene [AF355374] MID4_rep_c1832 KC146558 0 99% Uncultured bacterium clone AK4DE1_08B 16S ribosomal RNA gene [GQ397048] MID4_s3934 KC146559 6e-173 93% Uncultured bacterium clone Lou_h61 16S ribosomal RNA gene [EU991250] MID4_rep_c797 KC146560 7e-151 99% Microcoleus vaginatus P0B 16S ribosomal RNA gene [JQ712609] MID4_rep_c2097 KC146554 0 95% Uncultured cyanobacterium clone AN2_1 16S ribosomal RNA gene, 16S-23S ribosomal RNA intergenic spacer and 23S ribosomal RNA gene, [FJ805910] MID5_rep_c3509 KC146585 0 95% Brevundimonas diminuta 23S rRNA gene [X87288] MID5_s3724 KC146586 2e-77 88% Bordetella pertussis gene for 23S ribosomal RNA [X68323] MID5_rep_c4709 KC146572 3e-85 89% Bordetella petrii strain DSM 12804, 23S ribosomal RNA gene region [AM902716] MID5_rep_c833 KC146573 0 98% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_rep_c1018 KC146574 0 98% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_s1137 KC146575 3e-175 97% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_rep_c2575 KC146576 0 100% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_rep_c3738 KC146577 0 98% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_rep_c4301 KC146578 0 100% Lactobacillus helveticus DPC 4571, 23S ribosomal RNA gene region [CP000517] MID5_rep_c1729 KC146569 2e-115 100% Micrococcus luteus NCTC 2665, 16S ribosomal RNA gene region [CP001628] MID5_rep_c2005 KC146570 0 97% Micrococcus luteus NCTC 2665, 16S ribosomal RNA gene region [CP001628] MID5_s3361 KC146561 3e-149 99% Microcoleus vaginatus isolate MOA4-1 clone 148-3B 16S ribosomal RNA gene [AF355374] MID5_rep_c67 KC146562 0 97% Microcoleus vaginatus USPCI-1 16S ribosomal RNA gene [EU586737] MID5_rep_c165 KC146563 0 99% Microcoleus vaginatus USPCI-1 16S ribosomal RNA gene [EU586737] MID5_s1931 (SEQ5) 4e-97 100% Raphidiopsis sp. D9 clone ITS-L 16S ribosomal RNA gene and 16S-23S ribosomal RNA intergenic spacer [EU552069] MID5_rep_c552 KC146564 3e-175 99% Uncultured bacterium clone 199 16S ribosomal RNA gene [GU362407] MID5_rep_c1812 KC146565 0 99% Uncultured bacterium clone AK4DE1_08B 16S ribosomal RNA gene [GQ397048] MID5_rep_c3833 KC146587 1e-89 93% Uncultured bacterium clone F5K2Q4C04IUZMC 23S ribosomal RNA gene [GU926492] MID5_rep_c57 (SEQ6) 1e-16 98% Uncultured bacterium clone FFCH2090 16S ribosomal RNA gene [EU134963] MID5_rep_c1826 KC146566 3e-150 98% Uncultured bacterium clone nbw428c05c1 16S ribosomal RNA gene [GQ094346] MID5_rep_c2330 KC146567 0 99% Uncultured bacterium clone nbw428c05c1 16S ribosomal RNA gene [GQ094346] MID5_rep_c785 KC146568 7e-151 99% Uncultured bacterium clone Ovdat-222 16S ribosomal RNA gene [GQ425453] MID5_rep_c3135 KC146552 0 98% Uncultured cyanobacterium clone AN2_1 16S ribosomal RNA gene, 16S-23S ribosomal RNA intergenic spacer, and 23S ribosomal RNA gene [FJ805910] MID5_rep_c3856 KC146553 0 98% Uncultured cyanobacterium clone AN2_1 16S ribosomal RNA gene, 16S-23S ribosomal RNA intergenic spacer, and 23S ribosomal RNA gene [FJ805910] aColor code: Sequences from GISP2D 1600 m, 10,500 ybp ice core (red) and sequences from GISP2D 3014 m, 157,000 ybp ice core (green) bAll accession numbers starting with KC have been submitted to NCBI, all accession numbers starting with SEQ in parenthesis did not meet the NCBI requirements to be submitted but are still included the in analysis. cName, gene region, and accession number (in square brackets) retrieved from BLAST searches.

Table 5. mRNA sequences from the GISP2D ice core sectionsa. Sequence Accession Percent Name Numbersb E-value Similarity Descriptionc MMID4_rep_c3929 (SEQ1) 3e-57 94% Bradyrhizobium sp. Pter7 Hsp70 class chaperone (dnaK) gene [EF601934] MID4_rep_c1908 (SEQ2) 4e-97 99% Burkholderia vietnamiensis G4 chromosome 1, product="amino acid ABC transporter substrate-binding protein, PAAT family [CP000614] MID4_rep_c1535 (SEQ3) 1e-92 98% Ochrobactrum anthropi ATCC 49188 chromosome 2, product="drug resistance transporter, EmrB/Qac subfamily" [CP000759] MID5_rep_c2094 KC146588 1e-14 84% Geobacillus sp. Y412MC61, product="hypothetical protein" glycerol kinase, diguanylate cyclase with PAS/PAC and GAF sensors [CP002442] MID5_rep_c95 (SEQ4) 1e-77 96% Geobacillus stearothermophilus plasmid pGS18, strain 18, product="putative ISBst12-like transposase” [AM886060] MID5_rep_c2683 KC146579 0 91% Pseudomonas stutzeri A1501, complete genome [CP000304] aColor code: Sequences from GISP2D 1600 m, 10,500 ybp ice core (red) and sequences from GISP2D 3014 m, 157,000 ybp ice core (green) bAll accession numbers starting with KC have been submitted to NCBI, all accession numbers starting with SEQ in parenthesis did not meet the NCBI requirements to be submitted but are still included the in analysis. cName, gene region, and accession number (in square brackets) retrieved from BLAST searches. 53

2.3.1 Taxa

Nucleotide sequences were subjected to BLAST searches to obtain the closest taxa. Even with 100% sequence similarity the BLAST hits cannot reveal the exact organism. However, it is possible to identify an organism to the species level, especially if the BLAST matches are at 99-100% nucleotide identity. Moreover, by additionally assessing phylogenetic clustering, it is possible to more accurately identify the taxonomic affinities of the organisms (often to species) that are trapped in Greenland ice.

Twelve unique sequences resulted from the 10,500 ybp core section meltwater and 27 unique sequences were present in the 157,000 ybp core section meltwater.

Bacterial species made up 100% of sequences from the metagenomic/metatranscriptomic analysis. Table 4 and 5 shows that deeper, older ice had more total sequences. Table 4 represents the rRNA sequences and Table 5 represents the mRNA sequences found in the two ice cores. There were four species common to both samples: Lactobacillus helveticus (matching CP000517 in a BLAST search were sequences KC155352, KC146573, KC146574, KC146575, KC146576, KC146577,

KC146578), Microcoleus vaginatus (matching AF355374 in a BLAST search were sequences KC146557, KC146561), Uncultured bacterium (matching GQ397048 in a

BLAST search were sequences KC146558, KC146565), Uncultured cyanobacterium

(matching FJ805910 in a BLAST search were sequences KC146554, KC146552,

KC146553). All others were unique to one sample or the other. In the 10,500 ybp sample, all 12 sequences were each unique in the meltwater sample, while in the 157,000 ybp 54

55 sample, 18 species were indicated among the 27 sequences represented in the meltwater.

In general, the taxa in the cores were similar to organisms found in polar ice

(both Arctic and Antarctic), soil, dust, water, hot springs, animals and deep-sea thermal vents. Some of these organisms have been found in multiple environments and are known to be ubiquitous. However, when an organism can be found thriving in multiple environments, it might possess very different growth characteristics.

The majority of species recovered from the ice in this study were environmental species in the NCBI database; that is, they have not been taxonomically identified or characterized. The species indicated in the dataset are specifically from the cold, icy, soil, dust, air, marine, freshwater, hot springs and/or thermal vent environments. The

10,500 ybp ice contained environmental sequences closest to species within

Bradyrhizobium and Microcoleus, as well as some listed as “uncultured bacterium”. Two sequences (KC146557 and SEQ1) were associated with soil, KC146557 specifically with arid land soils. Also, the 10,500 ybp ice sample sequence, KC146584, was closest to a

Halomonas isolate that lives in deep-sea hydrothermal vent environments. The 157,000 ybp ice section sample contained environmental species that were closest to members of

Geobacillus (KC146588 and SEQ4), Micrococcus (KC146569 - with 100% similarity and

KC146570), Microcoleus (KC146561, KC146562, KC146563) and Pseudomonas (KC146579).

Six were associated with soil, three specifically with arid land soils (KC146561,

KC146562, KC146563), one with Pseudomonas (KC146579) and two sequences were closest to Micrococcus, known to be saprotrophic (KC146569 and KC146570). 55

Additionally, two sequences closet to Geobacillus, whose members are thermophilic and associated with hot springs (KC146588 and SEQ4). However, our sequences KC146588, had only an 84% sequence identity to a glycerol kinase gene of Geobacillus (CP002442).

There were bacterial sequences that were associated with animal pathogens present in both ice cores. In the 10,500 ybp ice core section, three species (SEQ2 closest to Burkholderia [CP000614] with 99% similarity, KC146571 closet to Escherichia

[AE005174] with 93% similarity, and SEQ3 closet to Ochrobactrum [CP000759] with 98% similarity) related to human pathogenic bacteria occurred. In the older ice three sequences closest to species of animal pathogens were found. Two different species of

Bordetella (X68323, AM902716), KC146586 with 88% similarity and KC146572 with 89% similarity were matched and KC146579 with 91% similarity to Pseudomonas. While these species are often associated with human pathogenesis, they are also known to be in the soil. Due to the low percent similarity we assume that they could easily be a different species or even a different family unrelated to a human pathogen. Similarly, thermotolerant Lactobacillus, found in both ice cores, while often associated with humans can be found in the soil. Soil microbes could easily be caught up in the dust particles that settle on glacial ice.

2.3.2 Phylogenetics

Of the 39 metagenomic/metatranscriptomic sequences, 32 were taxonomically classified to the phylum level using phylogenetic clustering techniques. Of the 33 rRNA sequences (Table 4), neighbor-joining trees were constructed from 32 of the bacterial rRNA sequences to determine the affinities of all of the sequences, including those that 56

57 could not be identified to the species level using BLAST searches (Figures 4 and 5). In both samples, all the sequences grouped with members of 3 primary phyla:

Cyanobacteria, Firmicutes, and Proteobacteria (Alphaproteobacteria,

Betaproteobacteria, and Gammaproteobacteria), with the exception of KC146569 and

KC146570 which grouped with the Actinobacteria clade and SEQ6 which grouped with

Fusobacteria all from the 157,000 ybp sample. In the rRNA SSU gene phylogram (Figure

4), 11 of the sequences (KC146558, KC146557, KC146561, KC146560, KC146568,

KC146565, KC146562, KC146564, SEQ5, KC146563, KC146559) were within the cyanobacterial clade. While sequences KC146558, KC146557, KC146561, KC146560,

KC146568, and KC146565 formed a separate Cyanobacteria clade, they were within the larger clade that included sequences from Microcoleus vaginatus (AF355374, AF355371,

EU586737, EF654074.1, EF654062) and Phormidium autumnale (FN813344.1) that originated from soils: desert soil, deglaciated soil, arid soil, and Antarctica. The

Actinobacteria clade included two sequences (KC146569, KC146570), closely allied with sequences from Micrococcus luteus (CP1628.1), an environmental sample from soil or dust. Four sequences grouped within the Firmicutes clade (KC146566, KC146567,

KC146556, KC146555). Two (KC146556, KC146555) were from cultures whose sequences grouped within the genus Bacillus. In previous studies of Greenland ice cores, several Bacillus spp. were isolated (Castello 2005). Two metagenomic/metatranscriptomic sequences (KC146566, KC146567) from the 157,000 year old ice core section most closely matched sequences Marinococcus sp. (AY505531.1) and Sinococcus beijingensis (DQ344635.1), both of which had been isolated from saline 57

58 soils. This is consistent with the high atmospheric dust concentration at that time that could easily carry soil particles/organisms (Petit, et al. 1999).

Figure 4. Phylogram of bacterial small subunit ribosomal RNA sequences. Phylogram of bacterial small subunit ribosomal RNA (SSU rRNA) sequences (neighbor- joining using phylogenetic analysis using parsimony (PAUP)) from the GISP2D and Byrd ice cores. Blue and red labels indicate sequences from the GISP2D 1601 and 3014 ice core sections, respectively. Sequences that start with CULT are isolates (green). The closest National Center for Biotechnology Information (NCBI) sequences to the queried sequences were selected from BLAST search results for use in the phylogenetic analysis (black). NCBI accession numbers (in square brackets) and sources of isolates (in parentheses) are provided. Phyla are listed on the right. Bootstrap values (1,000 replications) are given for branches with support greater than 50%.

Figure 5. Phylogram of bacterial large subunit ribosomal RNA sequences. Phylogram of bacterial large subunit ribosomal RNA (LSU rRNA) sequences (neighbor-joining using phylogenetic analysis using parsimony (PAUP)) from GISP2D. Blue and red labels indicate sequences from the GISP2D 1601 and 3014 ice core sections, respectively. The closest National Center for Biotechnology Information (NCBI) sequences to the queried sequences were selected from BLAST search results for use in the phylogenetic analysis (black). NCBI accession numbers (in square brackets) and sources of isolates (in parentheses) are provided. Phyla are listed on the right. Bootstrap values (1,000 replications) are given for branches with support greater than 50%.

Fewer sequences from Cyanobacteria were found in the rRNA LSU gene data set

(Figure 5). KC146554, KC146553, and KC146552 were within a clade that included an 60

uncultured cyanobacterium (FJ805910.1) sequenced from an environmental sample.

Firmicute sequences (KC155351, KC146578, KC155352, KC146577, KC146573, KC146574,

KC146576, KC146575) predominated in the rRNA LSU data set. All were most closely allied with sequences from Lactobacillus crispatus and L. helveticus (FN692037, CP000517, respectively). Both are associated with animals, but similar taxa can be found in soils.

Similarly, one sequence closest to Alphaproteobacterium (KC146585) and three sequences closest to Betaproteobacteria (KC146587, KC146572, KC146586) were most closely related to sequences from bacteria associated with animals, but which can also be found in soil and sludge. Sequence KC146584 most closely matched sequences from

Gammaproteobacteria. It was closest to members of the halophilic and thermotolerant genus Halomonas (FN257755). Species of Halomonas (FN257749.1) and Lactobacillus

(CP000517) are thermotolerant. Coincidentally, dark granular material was found in the 10,500 year old ice core section that resembles volcanic ash.

Figure 6 is a graphical representation of the major phyla found, based on phylogenetic analysis. Bacterial species made up 100% of sequences from the metagenomic/metatranscriptomic analysis. The 10,500 ybp ice sample contained

Proteobacteria (42%), followed by Cyanobacteria (33%), Firmicutes (17%) and uncultured bacteria (8%). The phyla present in the 157,000 ybp ice core section sample were Firmicutes (37%), followed by Cyanobacteria (18% and 15% suspected

Cyanobacteria), Proteobacteria (19%), Actinobacteria (7%) and Fusobacteria (4%).

Fifteen percent of the sequences in the 157,000 ybp sample were suspected to be

Cyanobacteria; sequences KC146562, KC146564, SEQ5, KC146563 appeared to be

Cyanobacteria (Figure 4), yet, no solid conclusion can be made about their phylum with 61

62 out further analysis.

Figure 6. Bar graph of phyla represented in the study. Comparison of major bacterial phyla in GISP 1601 and GISP 3014 based on results presented in Table 4 and 5. Bar graphs are proportional to the total number of taxa for GISP 1601 and GISP 3014. Numbers above bar graphs indicate the total number of unique sequences for each graph. For GISP 1601, the total number of unique sequences was 12. For GISP 3014, the total number of unique sequences was 27. Scales are to the left of each bar graph. Abbreviations: Fi – Firmicutes, C – Cyanobacteria, C? – possible Cyanobacteria, P – Proteobacteria, A – Actinobacteria, Fu – Fusobacteria, and U – uncultured. Proteobacteria and Cyanobacteria dominate GISP 1601 and GISP 3014 appears to be dominated by Firmicutes and Cyanobacteria. 62

2.3.3 Culturing and Sequencing

From approximately 300 plates that were inoculated during ice core melting, 6 cultures grew. While all cultures that grew were subcultured at a variety of temperatures (4˚, 8˚, and 15˚C), the cultures from this study grew best at 15˚C. These isolates do not fit the profile of psychrophiles, whose optimal growth temperature is below 15°C and do not grow at temperatures above 20°C. Rather, as in earlier studies, they are classified as psychrotolerant, surviving and growing at low temperatures

(D’Elia 2008, 2009). In the polar regions, sharp and rapid changes in seasonal temperatures act as the main selective pressure. Significant environmental changes may kill many true psychrophiles, while generalists such as psychrotolerant species have better chances to survive, especially on the surface of the glaciers where temperatures are the most variable.

Culture KC146555 from Byrd, 1593 m and culture KC146556 from Byrd, 104 m grew on LB media (Table 6). Growth on LB media indicates that these organisms were most likely bacterial species. The colonies of isolate KC146556 were white and wrinkled, characteristic of Bacillus colonies. Four of the cultures grew on MEA plates, one each from GISP, 99 m (KC146580), Byrd, 1593 m (KC146581), Byrd, 2131 m (KC146582), and

GISP, 3014 m (KC146583) (Table 6). MEA is used to encourage the growth of a broad range of fungi. The growth of KC146580 from GISP, 99 m was black and fluffy.

KC146582 appeared bright orange in culture. Analysis of the sequences revealed two bacterial hits and four fungal hits on NCBI Blast. Sequences KC146556 and KC146555 were closest to Bacillus (JN999834, JX847129, respectively) species. Three of the fungal 63

64 sequences were closest to Alternaria (JX177676, JX418322, JF440581) and one was closest to Cladosporium (EU272531). 64

Table 6. Isolate sequences from the ice core sectionsa.

Sequence Accession Percent Name Numbersb E-value Similarity Descriptionc CULT_A-1 KC146555 0 99% Bacillus licheniformis strain LZBL-O1 16S ribosomal RNA gene [JX847129] CULT_B-1 KC146556 0 100% Bacillus thioparans strain FJLY3 16S ribosomal RNA gene [JN999834] CULT_C-2 KC146580 2e-146 99% Alternaria sp. cas1 18S ribosomal RNA gene, internal transcribed spacer 1, 5.8S ribosomal RNA gene, and internal transcribed spacer 2, and 28S ribosomal RNA gene [JX177676] CULT_C-8 KC146581 0 99% Alternaria alternata strain A1 18S ribosomal RNA gene, internal spacer 1, 5.8S ribosomal RNA gene, and internal transcribed spacer 2; and 28S ribosomal RNA gene [JX418322] CULT_D-7 KC146582 0 99% Alternaria alternata 18S ribosomal RNA gene, internal transcribed spacer, 1, 5.8S ribosomal RNA gene, internal transcribed spacer 2, and 28S ribosomal RNA gene, region [JF440581] CULT_E-5 KC146583 0 99% Cladosporium tenuissimum isolate 2727 18S ribosomal RNA gene, internal transcribed spacer 1, 5.8S ribosomal RNA gene, and internal transcribed spacer 2, and 28S ribosomal RNA gene [EU272531] aColor code: Sequences from Byrd 1593 m, 30,000 ybp (blue), from GISP2D 104 m, <500 ybp (orange), Byrd 99 m, <500 ybp (black), Byrd 2131 m, 70,000 ybp (purple), and from GISP2D 3014 m, 157,000 ybp ice core (green) bSubmitted to NCBI. cName, gene region, and accession number (in square brackets) retrieved from BLAST search 65

2.3.4 Elemental Analysis

The black particulate matter from GISP, 1601 m (10,500 ybp) was examined in the scanning electron microscope. Four images were taken while viewing the specimen in the scanning electron microscope. Two images and two elemental analyses were taken for each stub. Each image and spectra was from an individual particle on the stub.

Repetition of spectral measurements was helpful in determining the elemental composition of the particulate matter. The images of the black particulate matter from the carbon stub had a coarse texture (Figure 7, a and b). The images of the particulate matter on the glass stub appeared geometric (Figure 7, c and d). The tests on the carbon stub showed traces of carbon, oxygen, sodium, magnesium, aluminum, tungsten, chlorine, potassium, calcium, titanium, and iron (Figure 7, e and f). The carbon could be due to the carbon stub and carbon coating. The elemental analysis results from the particulate matter on the glass stub consisted of carbon, sodium, silicon, chlorine, calcium, oxygen, and zinc (Figure 7, g and h). The peaks in the spectra simply show the presence of an element. The peaks are not reflecting the quantity of the element in the sample.

Figure 7. SEM micrographs and spectra. SEM micrographs and spectra from the black particulate matter in the GISP 1601 sample. a) and b) Micrograph of the particulate matter on a carbon stub. c) and d) Micrograph of the particulate matter on the glass stub. Each micrograph presents different pieces of particulate matter. e) Spectrum from elemental analysis performed on the particulate matter mounted on the carbon stub. The peaks indicate the presence of carbon, oxygen, sodium, aluminum, tungsten, chlorine, potassium, calcium, and iron f) Spectrum from elemental analysis performed on the same particulate matter mounted on a carbon stub. The peaks indicate the presence of carbon, oxygen, sodium, magnesium, aluminum, tungsten, chloride, potassium, calcium, titanium, and iron. g) Spectrum from elemental analysis performed on the particulate matter mounted on the glass stub. The peaks indicate the presence of carbon, sodium, silicon, chlorine, and calcium. h) Spectrum from elemental analysis performed on the same particulate matter mounted on the glass stub. The peaks indicate the presence of carbon, oxygen, sodium, silicon, chlorine, calcium, and zinc. Four spectra were obtained for each different piece of particulate matter (2 on the carbon stub, 2 on the glass stub).

The unique, and possibly defining, element in the analysis was tungsten.

Through looking at which elements are present in the volcanic ash/particulate matter we can likely determine which volcano erupting during this time period had deposited ash in Greenland glaciers.

2.4 Discussion

2.4.1 Overview

The results presented here have indicated that the microbial diversity deposited in glacial ice is lower during periods of high CO2 and warming on Earth. Conversely, during periods of lower temperatures and high dust there were higher numbers of microbial sequences deposited in the ice. This study concentrated on characterizing sequences from microbes present in ancient ice core sections from Greenland, based on short RNA and DNA fragments entrapped in the ice. These originate from viable organisms, deceased organisms, and cellular debris. The primary focus was to compare the microbial community in a 157,000 ybp ice core section with the microbial community in a 10,500 year old ice core section.

2.4.2 Expected microbes and found microbes

Greenland

There have been reports of a variety of organisms from Greenland ice (Ma, et al.

2000, 2005). Greenland’s geographic location and the distance nearest landmasses contribute to the number of sequences retrieved from ice. The GISP2D site lies relatively close to North America, Europe, and Asia, and winds often intersect the GISP2D site from these regions. Also, it has been shown that microbes transported over long distances are often deposited in times of lower temperature because worldwide water is locked up in glaciers—ergo, there is less moisture in the atmosphere and the land is drier leading to more dust in the atmosphere. The large concentrations of dust in the air

allow for increases in long-range transport of these microbes as well as cell remnants. 68

The dust in the air can carry organisms for thousands of miles from desert regions and deposit them in the ice. However, higher proportions of viable organisms correlate with cooler periods with lower dust concentrations (Knowlton, et al. 2013). The reason for this is unclear, but it might relate to the fact that dry periods lead to an increase in the number of dead organisms in the atmosphere, thus decreasing the proportion of viable organisms in the ice. During warmer periods, there is more moisture in the air and more precipitation, which tends to reduce the amount of dust in the atmosphere. The amount of snowfall is significantly higher in Greenland than in other polar regions, such as Eastern Antarctica. The snowfall can be a mechanism for microbes to be deposited on the glacier. However, dust and cell concentrations often are low and therefore robust studies have been difficult to achieve.

Microbes in arid conditions

A microbe found in arid conditions that has a sequence similar to what was found in both the ice cores sections in the metagenomic/metatranscriptomic data was

Microcoleus vaginatus (matching AF355374 in a BLAST search were sequences KC146557,

KC146561 with 99% similarity; Table 4). Microcoleus vaginatus can be found most often in biocrusts of arid land (Starkenburg ,et al. 2011, Belnap & Gardner 1993, Boyer, et al.2002). The phylogenetic analyses based on the rRNA SSU gene sequences (Figure 4) indicated that the majority of microbes that were in the 157,000 year old ice originated from soils, especially those from deserts and saline soils. This is consistent with conditions at the time, which were dry and cold, with high concentrations of dust in the atmosphere (Petit, et al. 1999). Furthermore, the organisms themselves were hardy and 69

70 desiccation resistant. Many Cyanobacteria and Firmicutes were present in the

Greenland (GISP2D) ice core sections (Figures 4 and 5), including many that are similar to species from soils, arid soils, and polar regions. Microorganisms in arid environments must have mechanisms, such as organic compound accumulation, for regulation in the extreme climate (temperature, salt levels, and lack available water; Starkenburg, et al.

2011). Cyanobacteria have been reported to enrich the soil stability due to the protection and adaptability of the polysaccharide sheath that surrounds the filaments (Belnap &

Gardner 1993). This could also enhance their ability to survive in glacial ice. Microcoleus vaginatus is capable of synthesizing and degrading trehalose (Starkenburg, et al. 2011) from maltose and is highly motile (Boyer, et al. 2002). Trehalose is a cryoprotectant that can help this organism survive in the ice (Goodrich, et al. 1998, Snider, et al. 2000).

Microbes in the atmosphere

Types of microbes deposited on glaciers are likely to differ according to climate, characteristics of the microbes, and based on whether they are transported on dust particles or by the various forms of atmospheric water (e.g., clouds, fog, rain, and snow)

(Rogers, et al. 2004a). Reports of Actinobacteria, Firmicutes, Proteobacteria (primarily

Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria) and Bacteroidetes phyla, and genera such as, Pseudomonas, Sphingomonas, Staphylococcus, Streptomyces, and

Arthrobacter have been found in tropospheric clouds (Amato, et al. 2007). These phyla are often found in cold environments, freshwater, marine, and soil, or living on vegetation and animals. When tested for growth at 5°C, half of the organisms found in the Amato, et al. (2007) study grew. This shows that few of the isolated organisms are 70

71 truly psychrophilic but that psychrotolerant organisms dominate the population, similar to the results that have been shown in studies of organisms found in ice. Also, fog water bacteria consisted of Pseudomonas, Bacillus, and Acinetobacter and several fungi

(Amato, et al. 2007). Christner, et al. (2008) reported snow and ice formation by substrates catalyzing ice nucleation in the clouds. Examples of organisms that have ice nucleation ability and are present in the clouds and in our samples are Pseudomonas,

Cladosporium, and Alternaria (Hamilton & Lenton 1998; Knowlton, et al. 2013).

Pseudomonads are Gram-negative bacteria that can oxidize a large variety of organic compounds. Diverse genes control the ability to utilize carbon sources, fix nitrogen, denitrify, degrade aromatic compounds, and synthesize polyhydroxybutyrate, allowing physiological and genetic diversity and the ability of this genus to live in a variety of environments (Yu, et al. 2011, Yan, et al. 2008). One of our sequences

(KC146579) from the metagenomic/metatranscriptomic analysis was 91% similar to the

23S rRNA gene of Pseudomonas stutzeri.

Microbes previously found in glacial ice

The results presented here confirm that an overwhelming majority of sequences in ancient ice are derived from the domain Bacteria (Abyzov, et al. 2005, D’Elia, et al.

2008, D’Elia, et al. 2009, Shtarkman, et al. 2012). There was a not a diverse assemblage of bacterial phyla, but representatives of only a few phyla in the two Greenland ice samples. In general, previous research shows that species in the Proteobacteria and

Actinobacteria are most common in glacial ice (Simon, et al. 2009, Miteva, et al. 2009). In one study by Christner, et al. (2005), molecular analysis of glacial ice from a variety of 71

72 global locations showed approximately 50% of organisms to be Gram-positive, spore forming organisms. This could be due to the thick layer of peptidoglycan that provides strength and protection of the cell during periods of freezing and desiccation.

Studies of Antarctic ice indicate that similar species exist in Greenland ice.

However, they are rarely identical in sequence (Knowlton, et al. 2013), indicating the possibility that species in the northern polar regions differ somewhat from those in the southern polar regions. This suggests that atmospheric transport of cells from one pole to the other might be rare. Ice isolates are psychrotolerant, and some are psychrophiles.

Often the microbes in ice are similar to those that live in aquatic, marine, soil, or sediment environments (D’Elia, et al. 2008). Reports indicate that Actinobacteria,

Proteobacteria, and Firmicutes dominate Antarctic ice (Bidle, et al. 2007). A comprehensive study reported that Sphingomonas, Methylobacteria, Acinetobacter, and

Arthrobacter ubiquitous in both polar and nonpolar ice (Christner, et al. 2005). In

>500,000 year old ice from Tibet, bacteria from Alphaproteobacteria, Betaproteobacteria,

Firmicutes, and Actinobacteria were most frequently found (Christner, et al. 2005).

Similarly, in accretion ice from Lake Vostok (Antarctica), primarily

Alphaproteobacteria, Betaproteobacteria, and Actinobacteria were recovered (Christner, et al. 2005).

Temperature influence

Studies have shown Actinobacteria, Firmicutes, Proteobacteria and Eukarya (in order of prevalence) to be in the highest proportions during cold periods, while in warmer conditions, Firmicutes, Eukarya, Actinobacteria, and Proteobacteria are in the 72

73 highest numbers (Miteva, et al. 2009). In our analyses we have determined that both the

157,000 ybp sample (colder climate) and the 10,500 ybp sample (warmer climate) contained a large number sequences related to Proteobacteria (Alphaproteobacteria,

Betaproteobacteria and Gammaproteobacteria), Firmicutes, and Cyanobacteria.

However, the number and frequency of cyanobacterial species was much higher in the

157,000 ybp sample. They were not recovered as isolates, and therefore the sequences likely are from nonviable sources. During the summer months, pools and lakes can form on the surfaces of glaciers, which provide habitats for cyanobacteria. It is possible that the cyanobacterial sequences originated from cells that inhabited these pools and lakes.

Actinobacteria

Many strains indicated from the sequence data are radiation and desiccation resistant. Pigmentation was recognized in culturing and served to protect from UV damage in airborne travel through the atmosphere and upon deposition on the surface of glacier, also other protective purposes, such as inhibition of degradative enzymes

(Gómez and Nosanchuk 2003). In our results (Table 4), two different sequences

(KC146569, KC146570) from 157,000 ybp were found, one with 100% identity and the other with 97% identity to sequences from Micrococcus luteus in the NCBI database. The

GC-content (74%) of Micrococcus luteus DNA is one of the highest in the bacterial sequences. This high GC-biased mutation pressure plays a large role in the choice of alternative synonymous codon in both the highly and weakly expressed genes (Ohama, et al. 1990). C50 carotenoid sarcinaxanthin found in Micrococcus luteus NCTC2665 seems 73

74 to explain its structure and contribute to strong antioxidative properties of the organism that would be beneficial for endurance in ice (Netzer, et al. 2010). Pigmentation is helpful to protect bacteria from UV radiation that is higher at the North and South Pole, due to lower ozone concentrations in the upper atmosphere in these regions. Castello, et al. (2005) reported finding a sequence related to Rhodococcus erythreus, similar to our two sequences (KC146569, KC146570) that are closely related to Micrococcus luteus

(CP001628), all members of Actinobacteria.

Microbes found in cold environments

Studies of other environments such as sea water, sea ice, and permafrost also indicate the ability of some organisms to survive in ice. The difference in sea ice versus glacial ice is in the formation of the ice sheet. Glacial ice forms in a top down manner from meteoric deposition, while sea ice is primarily formed from the bottom up

(although meteoric deposition also may occur). Nichols (2005) shows high levels of

Alphaproteobacteria in sea water and high levels of Gammaproteobacteria, Gram- positive bacteria and Cytophaga-Flexibacter-Bacteroides in sea ice. This is similar to the reports of Abyzov, et al. 2005 for Vostok ice. Vishnivetskaya, et al. (2005) detected both chemotrophic and phototrophic (green algae and Cyanobacteria) organisms in permafrost. They also found Gram-positive bacteria (including Microbacterium,

Rhodococcus, Microbispora, and Paenibacillus) and Gram-negative bacteria (including

Agrobacterium, Brevundimonas, Afipia, Aminobacter, Pseudomonas, and Xanthomonas).

Interestingly, a correlation was found that when viable algae were present, viable bacteria also were present. This association is a taxonomically complex, metabolically 74

75 interactive, self-sustaining community (Vishnivetskaya, et al. 2005). Pigmentation found in algae from the permafrost is used both for photosynthesis and protection of the microbes. As in our analysis, Cyanobacteria were found in the permafrost studied by

Vishnivetskaya, et al. (2005). Cyanobacteria are nitrogen fixers that thrive in diverse environments and are arguably self-sustaining (although they are probably heterotrophic in dark environments). The presence of Cyanobacteria in permafrost influences mineral nutrient cycling and energy flow.

Cyanobacteria

Many of the sequences were grouped into the phylum Cyanobacteria in our phylogenetic analysis (Figure 4 and 5). One of those sequences (SEQ5) was 100% similar to Raphidiopsis (Table 5). Cyanobacteria have been around for approximately 2.8 billion years (Stucken, et al. 2010). They have been found in marine, aquatic and terrestrial environments, including extreme conditions such as hot springs and polar ice (Stucken, et al. 2010). Li, et al. (2008) examined two different Raphidiopsis strains (R. curvata and R. mediterranea) and found the optimal growth temperature to be 10-40˚C. In general, cyanobacteria are known to fix nitrogen, yet Raphidiopsis strains lack the nifD gene that is necessary for nitrogen fixation (Stucken, et al. 2010, Li, et al. 2008). Raphidiopsis are filamentous cyanobacteria. The literature (Vishnivetskaya, et al. 2005) states that cyanobacteria can be found in polar regions, and therefore finding cyanobacterial species in the ice core sections is consistent with this.

Firmicutes

Both culturing and metagenomic/metatranscriptomic analysis revealed 75

76 representatives from the phylum Firmicutes. Castello, et al. 2005 performed an extensive study of Greenland glacial ice (GISP 2 and Dye 3), reporting the recovery of several Bacillus isolates from meltwater. Using similar culturing methods we recovered two isolates (KC146555, KC146556) related to Bacillus. Lactobacillus is a genus closely related to Bacillus (Table 6). In the metagenomic/metatranscriptomic study, eight of our sequences were closely related to Lactobacillus helveticus (CP000517) or Lactobacillus crispatus (FN692037) with 97%-100% similarity (Table 4). Lactobacillus is a genus composed of about 180 known species that are facultatively anaerobic, catalase- negative, non-spore-forming, Gram-positive, rod-shaped organisms, lactic acid producers, with a low G +C content (33-53 mol%) (Falsen, et al. 1999). Lactobacilli can be used in food production and can be found in the human body (Falsen, et al. 1999,

Callanan, et al. 2008, Ojala, et al. 2010). Callanan, et al. reported, in 2008, that 65% to

75% of the genes are conserved between the species of Lactobacillus, despite varying isolation locations. Conservation of the genome and variety of living environments make finding this species in our ice seem highly feasible. Seven of the sequences

(KC155352, KC146573, KC146574, KC146575, KC146576, KC146577, KC14658,) in BLAST searches had highest identities to Lactobacillus helveticus (CP000517), while one was most similar to a sequence from Lactobacillus crispatus (FN692037) (Table 4). All were sequences from the 23S rRNA region of the genome and aligned easily in our phylogenetic analysis (Figure 5).

In our metagenomic/metatranscriptomic analysis, there were two sequences

(KC146588, SEQ4) similar to the genus Geobacillus present in the 157,000 ybp ice. 76

Geobacillus sp. (CP002442) and Geobacillus stearothermophilus (AM886060) were identified in a BLAST search and found to have 84% and 96% similarity, respectively, to our sequences (Table 5). The genus Geobacillus is also closely related to the genus Bacillus.

The lower percent similarities suggest that, while the sequences were most similar to the two NCBI sequences, they could represent a sister taxon, not necessarily Geobacillus.

Bacilli are Gram-positive, thermophilic, neutrophilic, motile, spore-forming rods, and are aerobic or facultatively anaerobic bacteria (Nazina, et al. 2001, Alcaraz, et al. 2010).

Spore formation makes this group of bacteria resilient to environmental stressors.

Bacillus spp. can be found in a wide range of environments, and are capable of broad range of metabolic activity (Alcaraz, et al. 2010). Thermophilic aerobic spore-forming

Bacillus bacteria (such as Geobacillus stearothermophilus) grow best in a temperature range of 45º to >70ºC (Nazina, et al. 2001). The main explanation for finding Bacillus in glacial ice is the spore forming ability of these particular bacteria. Our data not only contained sequences related to species of Bacillus in the metagenomic/metatranscriptomic study, but they also were represented among the isolates recovered from the ice core section meltwater. The length of time that it took for the cultures to grow is characteristic for spore former.

Extremophiles

As mentioned earlier, species of Halomonas and Lactobacillus are thermotolerant and halophilic. Coincidentally, dark granular material was found in the 10,500 year old ice core section that resembles volcanic ash, indicative of possible volcanic activity.

Significant volcanic activity was occurring around the world during the 10,500 ybp time 77

78 period (Pagli, et al. 2008, Zielinski, et al. 1995, Global Volcanism Program 2012,

Detterman, et al. 1989). The results from the SEM elemental analysis helped to verify the elements present in the particulate matter. The spectrum from the carbon stub had a peak for tungsten, which was unique (Figure 7). Volcanoes in the Aleutian Islands, some of which were erupting during that time had a signature trace of tungsten, and therefore might have been the source for the ash in the ice core section (Global

Volcanism Program 2012, Detterman et al. 1989).

Proteobacteria

The sequence (KC146584) that was closest to Halomonas neptunia (FN257755, 92% similarity) grouped with species of Halomonas (Gammaproteobacteria) in the phylogenetic analysis (Figure 5). Gammaproteobacteria have a wide range of morphologies, ways to gather energy, ideal growth temperature, and oxygen requirements (Williams, et al. 2010). Halomonas species belong to the family

Halomonadaceae that is divided into two main groups, halophilic (bacteria that thrive in high salt environments) and non-halophilic bacteria (Mata, et al. 2002). Halomonas neptunia is classified as halophilic bacteria (Kaye, et al. 2004). This genus has been shown to be versatile in physiological function capable of living in high salt environments and temperatures as low as -1° C (Kaye, et al. 2004). They have a G+C content of over 50% and can often be found in hypersaline environments such as lagoons, salty lakes, cured and salty foods, briny petroleum reservoirs, hypersaline desert soils, sea ice, and deep-sea, hydrothermal-vents (Kaye, et al. 2004). The versatility in physiological function contributes to the ability of these particular bacteria to 78

79 possibly be found in ice in our study.

One of the metagenomic/metatranscriptomic sequences (SEQ1 from GISP2

1601m) was 94% identical to a heat shock protein from Bradyrhizobium sp (EF601934-

Alphaproteobacteria) (Table 5). If this were indeed a sequence of a heat shock protein

(HSP) it would help in explaining the survival of this organism in ice. HSPs are conserved throughout evolution and they assist in survival mechanisms in stresses, infection, injury, oxidative damage, hypoxia, freezing, and thermal stress (McConnell, et al. 2011).

SEQ2 was 99% similar to Burkholderia vietnamiensis (Table 5); these

Betaproteobacteria are found as human and plant pathogens or in the soil, specifically the rhizosphere of plant roots (Los Santos, et al. 2001). This species is an N2-fixing species (Los Santos, et al. 2001). The genus Burkholderia has been recently been under study because of their emergence as an opportunistic pathogens. This genus has been infecting the lungs of immunocompromised patients specifically with cystic fibrosis

(Conway & Greenway 2002). In addition, this particular isolate degrades environmental pollutants. Conway & Greenway (2002) examined the production of several acyl-HSLs in B. vietnamiensis to gain a better understanding the role of quorum sensing in the degradation of environmental pollutants, such as trichloroethylene. Both B. vietnamiensis and B. cepacia promote rice plant growth and maize growth, respectively

(Los Santos, et al. 2001). This organism’s ability to utilize nitrogen may provide a means of survival in ice. Found in the soil, this species may have been transported on dust and deposited on the ice, forming a plausible explanation of it being present in our sample. 79

Escherichia coli (AE005174) was 93% identical to our sequence (KC146571) (Table

4). However, this sequence did not fit into the alignment with the other portions of 23 rRNA, hence it is not included in the phylogenetic analysis (Figure 5). When aligning sequences there can be such great variation in the portion of the gene region segment being aligned that one particular sequence, such as KC146571, may not align with the other sequences and must be excluded. Escherichia coli O157 is well known to cause food borne illness (Conner & Kotrola 1995, Hayashi, et al. 2001). It is important to note general characteristics of E. coli because of the low percentage of similarity of our sequence. The genus Escherichia belongs to the family of Enterobacteriaceae from

Gammaproteobacteria, that are Gram-negative, facultative anaerobes (Lukjancenko, et al. 2010). Some strains of E. coli can be pathogenic to humans, however, there are many strains of E. coli that can be found in the human body that cause no harm, but rather are essential as part of the normal flora. Escherichia coli can also be readily found in the environment. The K-12 strain is a model organism that is studied in many laboratories.

The ubiquitous nature of E. coli reassures us that it is highly likely to find E. coli or something very similar in our sample.

One sequence (SEQ3) was 98% similar to Ochrobactrum anthropi (CP000759) (Table

5). This organism is highly drug resistant (Menuet, et al. 2008). Ochrobactrum anthropi has two chromosomes that have an average G+C content of 56 mol% (Chain, et al.

2011). Organisms in the order Rhizobiales are Gram-negative, non-fermenting, flagellated, nitrogen fixing bacteria. The species from the genus Ochrobactrum have been isolated from all over the world. This genus can also be found in soil, especially on 80

81 wheat roots and in internal root tissues from a variety of plant on many different continents (Lebuhn, et al. 2000). This common soil bacterium has been reported in a clinical setting, especially in endocarditis of immunocompromised patients (Lebuhn, et al. 2000, Chain, et al. 2011). As mentioned before, the ability of this organism to fix nitrogen may be advantageous in survivability in ice.

Two of the sequences (KC146586, KC146572) from the metagenomic/metatranscriptomic study were closest to species in the genus Bordetella

(Betaproteobacteria). Bordetella pertussis was 88% similar to our sequence and Bordetella petrii was 89% similar to our sequence (Table 4). The family Alcaligenaceae includes

Bordetella associated mostly with human pathogens and the genus Achromobacter includes Bordetella associated with environmental or facultative pathogenic organisms

(Von Wintzingerode, et al. 2001). While BLAST searches indicated that our sequences were associated with human pathogens, they had low percent similarity. Bordetella pertussis is not able to survive outside of a human host (Parkhill, et al. 2003) and while our sequence to be related to B. pertussis, it is possible that it could actually be a different genus in Alcaligenaceae or even a different family in the order

Burkholderiales. If in fact the organisms fell into a different but related genus or family that was more compatible with the environment it would increase their viability and survival in ice.

Sequence KC146585 shares 95% identity with B. diminuta (X87288) and falls in the phyla Alphaproteobacteria in our phylogenetic analysis (Table 4, Figure 5). In 1994,

Segers, et al. helped to define the genus Brevundimonas, as separate from Pseudomonas. 81

The defining features that separate Brevundimonas from Pseudomonas are that the former has short-wavelength polar flagella, restricted biochemical activity, different polyamine and ubiquinone patterns, and different fatty acid compositions (Segers, et al.1994).

Brevundimonas spp. are Gram-negative, rod shaped, and non-lactose-fermenting. It can be associated with the environmental water and human body in a clinical setting

(Menuet, et al. 2008, Segers et al.1994). Association with environmental water and soil would provide means of explanation to finding a sequence highly related to

Brevundimonas in our sample.

Uncultured bacteria

Many of the uncultured sequences were within the Cyanobacteria in our phylogenetic analysis. Rodríguez-Valera (2004) stated that less than one percent of the known bacteria are culturable. Uncultured organisms are organisms that have never been able to be cultured, but many have been partially sequenced. Many of the uncultured bacteria are sequences directly from environmental samples and therefore termed "environmental sequences." Despite not being able to culture the bacteria, phylogenetic analyses on metatranscriptnomically-derived sequences can be used in the determination of affinities to known taxa. Our uncultured sequences (KC146558,

KC146559, KC146554) in the 10,500 ybp core grouped with Actinobacteria and

Cyanobacteria in the phylogenetic analysis (Figure 4). Our uncultured sequences

(KC146564, KC146565, KC146587, SEQ6, KC146566, KC146567, KC146568, KC146552,

KC146553) in the 157,000 ybp core grouped with Cyanobacteria, Betaproteobacteria,

Fusobateria and Firmicutes in the phylogenetic analysis (Figure 4 and 5). 82

2.5 Conclusions

As the glacial ice in Greenland continues to melt, viable microbes will be released back into the environment, and may contribute to contemporary microbial genotypes through the process of temporal gene flow, also known as genome recycling (Rogers, et al. 2004a, Smith, et al. 2004). Melting ice in the environment releases an estimated 1017 to

1021 viable microbes every year (Smith, et al. 2004). This research allowed for determination of the community of microbes that have been entrapped in the ice through aeolian transport. While some species appeared in both samples, unique taxa and genes present in each sample may potentially indicate different atmospheric conditions during the time of ice deposition. Phylogenetic clustering indicated that the distributions of phyla were quantitatively similar but the number of taxa was higher in the 157,000 year-old sample (Table 4 and 5). This is likely due to increased dust deposition, because of lower temperatures and decreased precipitation in the atmosphere. Additionally, while the total number of sequences was higher in the

157,000 year old ice, previous research shows that the number of viable microbes was higher in the 10,500 year old sample (Knowlton, et al. 2013). This is likely because the dust was carrying a larger number of non-viable cells while less dust reached the glacier during the warmer periods. Although the distribution of phyla was similar, the species differed. Thus, while a small number of phyla are readily transported to, and deposited on, the glaciers, the species deposited vary with time and atmospheric conditions.

Additional correlations of microbial community composition and atmospheric conditions await further and more extensive investigations. 83

REFERENCES

1. Abyzov, S.S., M.N. Poglazova, J.N. Mitskevich, and M.V. Ivanov. 2005. Common

features of microorganisms in ancient layer of the Antarctic ice sheet, Pp. 240-250. In

J.D. Castello and S. O. Rogers (ed.), Life in ancient ice. Princeton University Press,

Princeton, NJ.

2. Ahn, J. and E. Brook. 2008. Atmospheric CO2 and climate on millennial time scales

during the last glacial period. Science 322:83-85.

3. Alcaraz, L.D., G. Moreno-Hagelsieb, L.E. Eguiarte, V. Souza, L. Herrera-Estrella,

G. Olmedo. 2010 . Understanding the evolutionary relationships and major traits of

Bacillus through comparative genomics. BMC Genomics 11:332-349.

4. Amato, P., M. Parazols, M. Sancelme, P. Laj, G. Mailhot and A. Delort. 2007.

Microorganisms isolated from the water phase of tropospheric clouds at the Puy de

Dôme: major groups and growth abilities at low temperatures. FEMS Microbiol Ecol

59:242–254.

5. Belnap, J. and J.S. Gardner. 1993. Soil microstructure in soils of the Colorado

plateau: the role of the cyanobacterium Microcoleus Vaginatus. Great Basin Naturalist

53:40-47.

6. Bidle K.J., S. Lee, D.R. Marchant, and P.G. Falkowski. 2007. Fossil genes and

microbes in the oldest ice on Earth. P Natl Acad Sci-Biol 104:13455–13460.

7. Boyer S.L., J.R. Johansen, V.R. Flechtner, G.L. Howard. 2002. Phylogeny and

genetic variance in terrestrial Microcoleus species based on the sequence analysis of

the 16S rRNA gene and associated 16S-23S ITS region. J Phycol 38:1222–1235. 84

8. Callanan M., P. Kaleta, J. O’Callaghan, O. O’Sullivan, K. Jordan, O. McAuliffe, A.

Sangrador-Vegas, L. Slattery, G.F. Fitzgerald, T. Beresford, and R.P. Ross. 2008.

Genome sequence of Lactobacillus helveticus, an organism distinguished by selective

gene loss and insertion sequence element expansion. J Bacteriol 190:727-735.

9. Castello J.D., S.O. Rogers, J.E. Smith, W.T. Starmer, and Y. Zhao. 2005. Plant and

bacterial viruses in the Greenland Ice Sheet , Pp. 196-207. In J.D. Castello and S. O.

Rogers (ed.), Life in ancient ice. Princeton University Press, Princeton, NJ.

10. Chain, P.S.G., D.M. Lang, D.J. Comerci, S.A. Malfatti, L.M. Vergez, M. Shin, R.A.

Ugalde, E. Garcia, and M.E. Tolmasky. 2011. Genome of Ochrobactrum anthropi

ATCC 49188T, a versatile opportunistic pathogen and symbiont of several

eukaryotic hosts. J Bacteriol 193:4274–4275.

11. Christner B.C., E. Mosley-Thompson, L.G. Thompson, and J.N. Reeve. 2005.

Classification of bacteria from polar and nonpolar glacial ice. Pp. 227-239. In J.D.

Castello and S. O. Rogers (ed.), Life in ancient ice. Princeton University Press,

Princeton, NJ.

12. Christner, B.C., C.E. Morris, C.M. Foreman, R. Cai, D.C. Sands. 2008. Ubiquity of

biological ice nucleators in snowfall. Science 319:1214.

13. Conner, D.E. and J.S. Kotrola. 1995. Growth and survival of Escherichia coli O157:H7

under acidic conditions. Appl Environ Microbiol 61:382–385.

14. Conway B.A. and E.P. Greenberg. 2002. Quorum-sensing signals and quorum-

sensing genes in Burkholderia vietnamiensis. J Bacteriol 184:1187-1191.

15. D'Elia, T., R. Veerapaneni, V. Theraisnathan and S.O. Rogers. 2009. Isolation of 85

fungi from Lake Vostok accretion ice. Mycologia 101:751-763.

16. D'Elia, T., R. Veerapaneni, V. Theraisnathan, and S.O. Rogers. 2008. Isolation of

microbes from Lake Vostok accretion ice. Appl Environ Microbiol 74:4962-4965.

17. Detterman, R.L., J.E. Case, S.E. Church, J.G. Frisken, F.H. Wilson, and M.E. Yount.

1989. The Alaska Mineral Resource Assessment Program: Background Information

to Accompany Folio of Geologic and Resource Maps of the Ugashik, Bristol Bay, and

Western Part of Karluk Quadrangles, Alaska. From Library of Congress.

http://www.dggs.alaska.gov/webpubs/usgs/c/text/c-1046.PDF (accessed

November 14, 2012) .

18. Falsen, E, C. Pascual, B. Sjödén, Maria Ohlén, and M.D. Collins. 1999. Phenotypic

and phylogenetic characterization of a novel Lactobacillus species from human

sources: description of Lactobacillus iners sp. nov. Int J Syst Bacteriol 49:217-221.

19. Fierer N., M. Breitbart, J. Nulton, P. Salamon, C. Lozupone, R. Jones, M. Robeson,

R. A. Edwards, B. Felts, S. Rayhawk, R. Knight, F. Rohwer, and R.B. Jackson. 2007.

Metagenomic and small-subunit rRNA analyses reveal the genetic diversity of

bacteria, archaea, fungi, and viruses in soil. Appl Environ Microbiol 73:7059–7066.

20. Fischer, H., M. Wahlen, J. Smith, D. Mastroianni, and B. Deck. 1999. Ice core

records of atmospheric CO2 around the last three glacial terminations. Science

283:1712-1714.

21. Glikson A. 2008. Milestones in the evolution of the atmosphere with reference to

climate change. Aust J Earth Sci 55:125-139. 86

22. Global Volcanism Program. 2012. Aleutian Islands. Retrieved from

http://www.volcano.si.edu/world/region.cfm?rnum=1101.

23. Gómez, B.L. and J.D. Nosanchuk. 2003. Melanin and fungi. Curr Opin Infect Dis

16:91–96.

24. Goodrich, R.P., T.M. Handel, and J.D. Baldeschwieler. 1988. Modification of lipid

phase behaviour with membrane-bound cryoprotectants. Biochimica Biophysica

Acta 938:143–154.

25. Gouy M., S. Guindon, and O. Gascuel. 2010. SeaView version 4: a multiplatform

graphical user interface for sequence alignment and phylogenetic tree building.

Molecular Biology and Evolution 27:221-224.

26. Hamer, C.U., H.B. Clausen, C.C. Langway. 1994. Electrical conductivity method

(ECM) stratigraphic dating of the Byrd Station ice core, Antarctica. Annals of

Glaciology 20:115-120.

27. Hamilton W.D. and T.M. Lenton. 1998. Spora and Gaia: how microbes fly with their

clouds. Ethol Ecol Evol 10:1-16.

28. Hayashi T., K. Makino, M. Ohnishi, K. Kurokawa, K. Ishii, K. Yokoyama, C.G.

Han, E. Ohtsubo, K. Nakayama, T. Murata, M. Tanaka, T. Tobe, T. Iida, H.Takami,

T. Honda, C. Sasakawa, N.Ogasawara, T. Yasunaga, S. Kuhara, T. Shiba, M.

Hattori, and H. Shinagawa. 2001. Complete genome sequence of enterohemorrhagic

Eschelichia coli O157:H7 and genomic comparison with a laboratory strain K-12.

DNA Res 8:11-22.

29. Katoh, K., K. Kuma, H. Toh, and T. Miyata. 2005. MAFFT version 5: improvement 87

in accuracy of multiple sequence alignment. Nucleic Acids Res 33:511-518.

30. Katoh, K., K. Misawa, K. Kuma, and T. Miyata. 2002. MAFFT: a novel method for

rapid multiple sequence alignment based on fast fourier transform. Nucleic Acids

Res 30:3059-3066.

31. Kaye, J.Z., M.C. Márquez, A. Ventosa, and J.A. Baross. 2004. Halomonas neptunia sp.

nov., Halomonas sulfidaeris sp. nov., Halomonas axialensis sp. nov. and Halomonas

hydrothermalis sp. nov.: halophilic bacteria isolated from deep-sea hydrothermal-vent

environments. Int J Syst Evol Micr 54:499–511.

32. Knowlton, C., R.Veerapaneni, T. D’Elia, and S.O. Rogers. 2013. Microbial analyses

of ancient ice core sections from Greenland and Antarctica. Biology 2:206-232.

33. Lebuhn M., W. Achouak, M. Schloter, O. Berge, H. Meier, M. Barakat, A.

Hartmann, and T. Heulin. 2000. Taxonomic characterization of Ochrobactrum sp.

isolates from soil samples and wheat roots, and description of Ochrobactrum tritici

sp. nov. and Ochrobactrum grignonense sp. nov. Int J Syst Evol Micr 50:2207–2223.

34. Li, R., S.W. Wilhelm, W.W. Carmichael, M.M. Watanabe. 2008. Polyphasic

characterization of water bloom forming Raphidiopsis species (Cyanobacteria) from

central China. Harmful Algae 7:146–153.

35. Los Santos, P.E., R. Bustillos-Cristales, and J. Caballero-Mellado. 2001.

Burkholderia, a genus rich in plant-associated nitrogen fixers with wide

environmental and geographic distribution. Appl Environ Microbiol 67:2790-2798.

36. Lukjancenko O., T.M. Wassenaar, and D.W. Ussery. 2010. Comparison of 61

sequenced Escherichia coli genomes. Microb Ecol 60:708–720. 88

37. Lüthi, D., M.L. Floch, B. Bereiter, T. Blunier, J.M. Barnola, U. Siegenthaler, D.

Raynaud, J. Jouzel, H. Fischer, K. Kawamura, and T.F. Stocker. 2008. High-

resolution carbon dioxide concentration record 650,000–800,000 years before present.

Nature 453:379-382.

38. Ma, L.J., C.M. Catranis, W.T. Starmer, and S.O. Rogers. 2005. The significance and

implications of the discovery of filamentous fungi in glacial ice. Pp. 159-180. In J.D.

Castello and S. O. Rogers (ed.), Life in ancient ice. Princeton University Press,

Princeton, NJ.

39. Ma, L.J., S.O. Rogers, C. Catranis, and W.T. Starmer. 2000. Detection and

characterization of ancient fungi entrapped in glacial ice. Mycologia 92:286-295.

40. Mata J.A., J. Martínez-Cánovas, E. Quessada, and V. Béjar. 2002. A Detailed

phenotypic characterisation of the type strains of Halomonas Species. System Appl

Microbiol 25:360–375.

41. McConnell K.W., A.C. Fox, A.T. Clark, N.N. Chang, J.A. Dominguez, A.B. Farris,

T.G. Buchman, C.R. Hunt, and C.M. Coopersmith. 2011. The role of HSP70 in

mediating age-dependent mortality in sepsis. J Immunol 186:3718–3725.

42. Menuet, M., F. Bittar, N. Stremler, J. Dubus, J. Sarles, D. Raoult, and J. Rolain.

2008. First isolation of two colistin-resistant emerging pathogens, Brevundimonas

diminuta and Ochrobactrum anthropi, in a woman with cystic fibrosis: a case report. J

Med Case Reports 2:373-376.

43. Miteva V, C. Teacher, T. Sowers, and J. Brenchley. 2009. Comparison of the

microbial diversity at different depths of the GISP2 Greenland ice core in 89

relationship to deposition climates. Environ Microbiol 11:640-656.

44. Nazina, T.N., T.P. Tourova, A.B. Poltaraus, E.V. Novikova, A.A. Grigoryan, A.E.

Ivanova, A.M. Lysenko, V.V. Petrunyaka, G.A. Osipov, S.S. Belyaev and M.V.

Ivanov. 2001. Taxonomic study of aerobic thermophilic bacilli: descriptions of

Geobacillus subterraneus gen. nov., sp. nov. and Geobacillus uzenensis sp. nov. from

petroleum reservoirs and transfer of Bacillus stearothermophilus, Bacillus

thermocatenulatus, Bacillus thermoleovorans, Bacillus kaustophilus, Bacillus

thermoglucosidasius and Bacillus thermodenitrificans to Geobacillus as the new

combinations G. stearothermophilus, G. thermocatenulatus, G. thermoleovorans, G.

kaustophilus, G. thermoglucosidasius and G. thermodenitrificans. Int J Syst Evol Micr

51:433–446.

45. Netzer, R., M.H. Stafsnes, T. Andreassen, A. Goksøyr, P. Bruheim, and T.

Brautaset. 2010. Biosynthetic pathway for λ-cyclic sarcinaxanthin in Micrococcus

luteus: heterologous expression and evidence for diverse and multiple catalytic

functions of C50 carotenoid cyclases. J Bacteriol 192:5688–5699.

46. Nichols, D.S. 2005. The growth of prokaryotes in Antarctic sea ice: implications for

ancient ice communities, Pp. 50-68. In J.D. Castello and S. O. Rogers (ed.), Life in

ancient ice. Princeton University Press, Princeton, NJ.

47. Ohama, T., A. Muto and S. Osawa. 1990. Role of GC-biased mutation pressure on

synonymous codon choice in Micrococcus luteus, a bacterium with a high genomic

GC-content. Nucleic Acids Res 18:1565-1569.

48. Ojala T., V. Kuparinen, J.P. Koskinen, E. Alatalo, L. Holm, P. Auvinen, S. 90

Edelman, B. Westerlund-Wikström, T.K. Korhonen, L. Paulin, and M. Kankainen.

2010. Genome sequence of Lactobacillus crispatus ST1. J Bacteriol 192:3547–3548.

49. Pagli, C. and F. Sigmudsson. 2008. Will present day glacier retreat increase volcanic

activity? Stress induced by recent glacier retreat and its effect on magmatism at the

Vatnajökull ice cap, Iceland. Geophys Res. Lett 35:L09304.

50. Parkhill J., M. Sebaihia, A. Preston, L.D. Murphy, N. Thomson, D.E. Harris,

M.T.G. Holden, C.M. Churcher, S.D. Bentley, K.L. Mungall, A.M. Cerdeño-

Tárraga, L. Temple, K. James, B. Harris, M.A. Quail, M. Achtman, R. Atkin, S.

Baker, D. Basham, N. Bason, I. Cherevach, T. Chillingworth, M. Collins, A.

Cronin, P. Davis, J. Doggett, T. Feltwell, A. Goble, N. Hamlin, H. Hauser, S.

Holroyd, K. Jagels, S. Leather, S. Moule, H. Norberczak, S. O’Neil, D. Ormond, C.

Price, E. Rabbinowitsch, S. Rutter, M. Sanders, D. Saunders, K. Seeger, S. Sharp,

M. Simmonds, J. Skelton, R. Squares, S. Squares, K. Stevens, L. Unwin, S.

Whitehead, B. G. Barrell, and D. J. Maskell. 2003. Comparative analysis of the

genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella

bronchiseptica. Nat Genet 35:32-40.

51. Petit, J.R., J. Jouzel, D. Raynaud, N.I. Barkov, J.M. Barnola, I. Basile, M. Bender, J.

Chappellaz, M. Davis, G. Delaygue, M. Delmotte, V.M. Kotlyakov, M. Legrand,

V.Y. Lipenkov, C. Lorius, L. PÉpin, C. Ritz, E. Saltzman, and M. Stievenard. 1999.

Climate and atmospheric history of the past 420,000 years from the Vostok ice core,

Antarctica. Nature 399:429-436.

52. Rodríguez-Valera F. 2004. Environmental genomics, the big picture? FEMS 91

Microbiol Lett 231:153-158.

53. Rogers S.O. and A.J. Bendich. 1985. Extraction of DNA from milligram amounts of

fresh, herbarium, and mummified plant tissues. Plant Mol Biol 5:69-76.

54. Rogers S.O., V. Theraisnathan, L.J. Ma, Y. Zhao, G. Zhang, S.G. Shin, J.D.

Castello, and W.T. Starmer. 2004. Comparisons of protocols for decontamination of

environmental ice samples for biological and molecular examinations. Appl Environ

Microbiol 70:2540-2544.

55. Rogers, S.O, W.T. Starmer, and J.D. Castello. 2004. Recycling of pathogenic

microbes through survival in ice. Med Hypoth 63:773-777.

56. Rogers, S.O. and A.J. Bendich. 1994. Extraction of total cellular DNA from plants,

algae and fungi, Pp. D1:1-8. In S.B. Gelvin, R.A. Schilperoort (ed.), Plant molecular

biology manual, 3rd Edition. Kluwer Academic Press, Dordrecht, Netherlands.

57. Saitou, N. and M. Nei. 1987. The neighbor-joining method: a new method for

reconstructing phylogenetic trees. Mol Biol Evol 4:406-25.

58. Segers, P., M. Vancanneyt, B. Pot, U. Torck, B. Hoste, D. Dewettinck, E. Falsen, K.

Kersters, and P. De Vos. 1994. Classification of Pseudomonas diminuta Leifson and

Hugh 1954 and Pseudomonas vesicularis Büsing, Döll, and Freytag 1953 in

Brevundimonas gen. nov. as Brevundimonas diminuta comb. nov. and Brevundimonas

vesicularis comb. nov., Respectively. Int J Syst Bacteriol 44:499-510.

59. Shtarkman, Y.M., Z.A. Koçer, R. Edgar, R. Veerapaneni, T. D’Elia, P.F. Morris, and

S.O. Rogers. 2012. Subglacial Lake Vostok (Antarctica) accretion ice contains a

diverse set of sequences from aquatic and marine Bacteria and Eukarya. PLOS ONE, 92

in revision.

60. Simon C., A. Wiezer, A.W. Strittmatter, and R. Daniel. 2009. Phylogenetic diversity

and metabolic potential revealed in a glacier ice metagenome. Appl Environ

Microbiol 75:7519–7526.

61. Smith, A.W., D.E. Skilling, J.D. Castello, and S.O. Rogers. 2004. Ice as a reservoir

for pathogenic human viruses: specifically, caliciviruses, influenza viruses, and

enteroviruses. Med Hypotheses 63:560-566.

62. Snider, C.S., T. Hsiang, G. Zhao, M. Griffith. 2000. Role of ice nucleation and

antifreeze activities in pathogenesis and growth of snow molds. Phytopathology

90:354-361.

63. Starkenburg, S.R., K.G. Reitenga, T. Freitas, S. Johnson, P.S.G. Chain, F. Garcia-

Pichel, and C.R. Kuske. 2011. Genome of the Cyanobacterium Microcoleus vaginatus

FGP-2, a photosynthetic ecosystem engineer of arid land soil biocrusts worldwide. J

Bacteriol 193:4569–4570.

64. Stucken, K., U. John, A. Cembella, A.A. Murillo, K. Soto-Liebe, J.J. Fuentes-

Valdés, M. Friedel, A.M. Plominsky, M. Vásquez, and G. Glöckner. 2010. The

smallest known genomes of multicellular and toxic Cyanobacteria: comparison,

minimal gene sets for linked traits and the evolutionary implications. PLOS ONE

5:1-15.

65. Suwa, M., J.C. Fischer, M.L. Bender, A. Landais, and E.J. Brook. 2006. Chronology

reconstruction for the disturbed bottom section of the GISP2 and the GRIP ice cores:

implications for termination II in Greenland. Journal of Geophysical Research 93

111:D02101.

66. Vishnivetskaya T.A., L.G. Erokhina, E.V. Spirina, A.V. Shatilovich, E.A.

Vorobyova, A.I. Tsapin, and D.A. Gilichinsky. 2005. Viable phototrophs:

Cyanobacteria and green algae from the permafrost darkness, Pp. 140-158. In J.D.

Castello and S. O. Rogers (ed.), Life in ancient ice. Princeton University Press,

Princeton, NJ.

67. Von Wintzingerode, F., A. Schattke, R.A. Siddiqui, U. Rösick, U.B. Göbel, and R.

Gross. 2001. Bordetella petrii sp. nov., isolated from an anaerobic bioreactor, and

emended description of the genus Bordetella. Int J Syst Evol Micr 51:1257-1265.

68. White T.J., T. Bruns, S. Lee, J. Taylor. 1990. Amplification and direct sequencing of

fungal ribosomal RNA genes for phylogenetics, Pp. 315-322. In M.A. Innis, D.H.

Gelfand, J.J. Sninsky, T.J. White, eds. PCR Protocols: a guide to methods and

applications. Academic Press, New York.

69. Williams, K.P., J.J. Gillespie, B.W.S. Sobral, E.K. Nordberg, E.E. Snyder, J.M.

Shallom, and A.W. Dickerman. 2010. Phylogeny of Gammaproteobacteria. J

Bacteriol 192:2305–2314.

70. Yan, Y., J. Yang, Y. Dou, M. Chen, S. Ping, J. Peng, W. Lu, W. Zhang, Z. Yao, H. Li,

W. Li, S. He, L. Geng, X. Zhang, F. Yang, H. Yu, Y. Zhan, D. Li, Z. Lin, Y. Wang, C.

Elmerich, M. Lin, and Q. Jin. 2008. Nitrogen fixation island and rhizosphere

competence traits in the genome of root-associated Pseudomonas stutzeri A150. Proc

Natl Acad Sci USA 105:7564–7569.

71. Yu, H., M. Yuan, W. Lu, J. Yang, S. Dai, Q. Li, Z. Yang, J. Dong, L. Sun, Z. Deng, 94

W. Zhang, M. Chen, S. Ping, Y. Han, Y. Zhan, Y. Yan, Q. Jin, and M. Lin. 2011.

Complete genome sequence of the nitrogen-fixing and rhizosphere-associated

bacterium Pseudomonas stutzeri Strain DSM4166. J Bacteriol 193:3422–3423.

72. Zielinski, G.A., P.A. Mayewski, L.D. Meeker, S. Whitlow, and M.S. Twickler.

1995. A 110,000-yr record of explosive volcanism from the GISP2 (Greenland) ice

core.

APPENDIX A

Biology 2013, 2, 206-232; doi:10.3390/biology2010206 OPEN ACCESS biology ISSN 2079-7737 www.mdpi.com/journal/biology Article Microbial Analyses of Ancient Ice Core Sections from Greenland and Antarctica

Caitlin Knowlton 1, Ram Veerapaneni 2, Tom D’Elia 3 and Scott O. Rogers 1,*

1 Department of Biological Sciences, Bowling Green State University, Bowling Green, OH 43403, USA; E-Mail: [email protected] 2 Department of Biological Sciences, Bowling Green State University, Firelands Campus, Huron, OH 44839, USA; E-Mail: [email protected] 3 Biological Sciences, Indian River State College, 32021 Virginia Avenue, Fort Pierce, FL 34981, USA; E-Mail: [email protected]

* Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: +1-419-372-9157; Fax: +1-419-372-2024.

Received: 10 December 2012; in revised form: 8 January 2013 / Accepted: 11 January 2013 / Published: 25 January 2013

Abstract: Ice deposited in Greenland and Antarctica entraps viable and nonviable microbes, as well as biomolecules, that become temporal atmospheric records. Five sections (estimated to be 500, 10,500, 57,000, 105,000 and 157,000 years before present, ybp) from the GISP2D (Greenland) ice core, three sections (500, 30,000 and 70,000 ybp) from the Byrd ice core, and four sections from the Vostok 5G (Antarctica) ice core (10,500, 57,000, 105,000 and 105,000 ybp) were studied by scanning electron microscopy, cultivation and rRNA gene sequencing. Bacterial and fungal isolates were recovered from 10 of the 12 sections. The highest numbers of isolates were found in ice core sections that were deposited during times of low 97 atmospheric CO2, low global temperatures and low levels of atmospheric dust. Two

of the sections (GISP2D at 10,500 and 157,000 ybp) also were examined using metagenomic/metatranscriptomic methods. These results indicated that sequences from microbes common to arid and saline soils were deposited in the ice during a

time of low temperature, low atmospheric CO2 and high dust levels. Members of Firmicutes and Cyanobacteria were the most prevalent bacteria, while Rhodotorula species were the most common eukaryotic representatives. Isolates of Bacillus, Rhodotorula, Alternaria and members of the Davidiellaceae were isolated from both Greenland and Antarctica sections of the same age, although the sequences differed between the two polar regions. Keywords: glacial ice; microbes; bacteria; fungi; Greenland; Antarctica; metagenomic; metatranscriptomic; cultures

1. Introduction

Microorganisms have been identified and isolated from glacial ice samples from many different regions of the world. The global distribution of these microorganisms in snow and glacial ice is the result of wind and atmospheric circulation. Bacteria and fungi have been detected and isolated from many different frozen environments such as permafrost, cryopegs, sea ice, glacial ice, and accretion ice from subglacial lakes [1–17]. Bacteria and fungi have been isolated from both Arctic and Antarctic regions [4,5,7–9]. Most of the bacteria and fungi isolated from these permanently cold environments were psychrotolerant as opposed to psychrophilic, having optimal growth temperatures well above freezing [4,5]. Glacial ice has constant temperatures, making it ideal for long-term preservation of microorganisms and biomolecules [16–19]. Environmental ice acts as a protective matrix for microorganisms over extended periods of time and provides a record of microbial evolution and ancient biodiversity [8]. While the numbers of viable organisms and intact nucleic acid sequences decline with depth, they have been recovered from ice that is more than a million years old [4,5,14,16–18,20]. Study of the organisms isolated from environmental ice has yielded insights into microbial longevity. However, studies have not been performed to compare microorganisms belonging to similar timescales entrapped in glacial ice from geographically distant sites. The Arctic and the Antarctic are geographically distant regions, but also are distinct for other reasons. Much of Antarctica is a desert with little precipitation, and the continent is far from other large landmasses. In general, Greenland receives more precipitation, and is geographically less isolated, receiving winds from Europe, Asia and North America [21]. Study of microbes isolated from these two distinct locations, dating back to similar time periods may provide insights into microbial community 98

composition through time and transportation of microbes in the atmosphere, as well as deposition and preservation of microbes and biomolecules in ice. Sections from the GISP2D (Greenland Ice Sheet Project 2), Vostok 5G (Antarctica) and Byrd (Antarctica) ice cores were selected for comparison (Figure 1). Sections were chosen that were representative of times of high and low atmospheric carbon dioxide, dust and mean global temperature [22–26]. Scanning electron microscopy, cultivation, sequencing and phylogenetic analyses were used to assess the influences of atmospheric characteristics on microbe deposition and survival in ice. Metagenomic/metatranscriptomic analyses were used to determine and compare the microbial communities at times when carbon dioxide levels and temperatures were at minima (157,000 ybp) [22,23] and at a time when carbon dioxide levels and temperatures were rising nearly to modern levels (10,500 ybp) [22–26]. During the past 420,000 years, the concentration of atmospheric carbon dioxide has ranged from a low of approximately 200 ppm 150,000 years ago [22,23] to a high of >390 ppm currently [27], with another high peak of approximately 280 ppm occurring approximately 125,000 years ago [23]. Carbon dioxide levels vary directly with changes in global temperature. Low CO2 levels are normally associated with low temperatures, low precipitation and high dust levels, while high CO2 levels generally indicate high temperatures, high precipitation rates, and low dust levels, factors that may influence transportation of microbes and overall community composition. Here we report research that was designed to test whether the same species are concurrently deposited in ice at both poles, and whether changes in atmospheric carbon dioxide, dust and global temperature affect deposition of microbes in glacial ice.

Figure 1. Locations of ice core sections used in this study. Sections from the Greenland Ice Sheet Project 2 (GISP2) (Greenland), Byrd (Antarctica) and Vostok 5G (Antarctica) ice cores were examined. See Experimental Section for more details about the core sections.

Greenland! Antarctica! GISP2!.!

Vostok! Byrd.! ! .! 99

2. Results and Discussion

2.1. Results

2.1.1. Scanning Electron Microscopy (SEM)

Most of the cells in the GISP2D and Vostok 5G core sections that were examined were rod- shaped bacteria (Figures 2 and 3). A few were coccoid, including diplococcoid forms (Figure 2, panels 16 and 22; Figure 3, panel 23). Two were spiral shaped (Figure 2, panels 18 and 19). Only three appeared to be fungi (Figure 3, panels 27–29), more complex shapes and larger cells, some of which may be conidia.

2.1.2. Revival and Molecular Identification of Glacial Isolates

There are geographical and temporal differences in the microbial composition of ice core samples (Table 1). Of the 27 cultures recovered from the 10 ice core sections, 19 were from Greenland ice and only eight were from Antarctica ice. Also, in general, the number of viable microbes, as indicated by number of cultures, was inversely proportional to the age of the ice core (Figure 4). The only exceptions to this were for the youngest ice sections that were analyzed, which were less than 500 years old (from Byrd and GISP2D). In both cases, few cultures were obtained. The ice core sections that were 10,500 years old from the GISP2D (1,601 m) and Vostok 5G (316 m) cores yielded higher numbers of cultures (seven fungi and two bacteria; Table 1). The number of cultures decreased with increasing age of the ice core sections (Table 1, Figure 4). There was a correlation between the number of cultures obtained and the atmospheric conditions at the time of ice deposition (Figure 5). Higher numbers of fungi and bacteria were isolated when atmospheric CO2 levels were below 240 ppmv (parts per million per volume), temperatures were at least 3 °C lower than the present mean global temperature, and dust levels were <0.4 ppm (atmospheric measurements are from [23]).

Figure 2. Scanning electron micrographs (SEM) of bacteria from the GISP2D ice core sections (panels 1–6, 1,601 m; 7–11, 2,501 m; 12–18, 2,777 m; 19–22, 3,014 m). Microbes in micrographs 1, 3, 5 and 6 are rod shaped bacteria. The microbes in micrographs 1 and 6 have a sheath-like cover over the exterior, similar to the microbes previously observed in Vostok accretion ice (red arrow; [4]). The organisms in micrograph 3 are similar to the rod shaped bacteria observed in previous GISP2D sediment ice studies (white arrow; [28]). Microbes in micrographs

7 and 11 are similar to long rod shaped bacteria observed in the Vostok ice cores 100

(green arrow; [4]). Organisms in micrograph 14 and 15 are rod shaped bacteria, similar to those previously observed from Vostok ice cores (blue arrow; [4]). The organism in micrograph 16 has a diplococcoid form (yellow arrow) and has been observed in other studies of the Vostok and GISP2D ice cores and from several cold environments including the Siberian permafrost [4,14,17]. This microbe is similar to Psychrobacter, a diplococcoid bacterium commonly found in the cold environments. The microbe in micrograph 18 has an unusual spiral structure (orange arrow) similar to organisms isolated from the Vostok core [4]. These may be related to the order Spirochaetales, which have long helically coiled cells. The microbes in micrographs 8 and 13 resemble cells of Caulobacter spp. while the curving form in micrograph 21 cannot be identified.

Figure 3. Scanning electron micrographs of bacteria from the Vostok 5G ice core section (micrographs 23–25, 316 m; 26–27, 900 m; 28–30, 1,529 m). No organisms were observed in the Vostok 5G 2,149 m ice core. Also, no cultures were obtained from this core section (Table 1). Diplococcoid organisms similar to the ones observed in the GISP2D cores were evident in micrograph 23 (yellow arrow). The microbe in micrograph 24 has a spiral structure and may be related to Spirochaetales. 101 Spore-like structures were observed commonly (red arrow). A fungal hyphae-like

structure was observed in the Vostok 5G 900 m ice core (panel 27). The spherical form in micrograph 29 appears to be a fungal coniduim. The microbes in Panels 25 and 30 are very small, similar to ultramicrobacteria found previously in the GISP2D ice core [28,29].

Table 1. Summary of BLAST search results. Included are sequences from cultures and metagenomic/metatranscriptomic analyses.

Depth Age Sequence Accession Identity Core Closest BLASTn match b Phylum c (m) (ybp) number a number (%) Byrd 99 <500 CULT C-2 KC146580 Alternaria sp. As 99 d 1,593 30,000 CULT C-8 KC146581 Alternaria alternata As 99 d CULT A-1 KC146555 Bacillus lichniformis Fi 99 d 2,131 70,000 CULT D-7 KC146582 Alternaria alternata As 99 d GISP 2D 104 <500 CULT B-1 KC146556 Bacillus thioparans Fi 100 e 1,601 10,500 GI860 KC206482 Alternaria alternata As 100 d c2581 KC155352 Lactobacillus helveticus Fi 100 d s3382 KC146557 Microcoleus vaginatus Cy 99 e c797 KC146560 Microcoleus vaginatus Cy 99 e c1832 KC146558 Uncultured bacterium Cy 99 e GI859 KC206491 Fusarium culmorum As 98 d s1867 KC155351 Lactobacillus crispatus Fi 98 e GI862 KC206492 Penicillium corylophilum As 98 d GI867 KC206485 Rhodotorula mucilacinosa Ba 98 e GI861 KC206488 Cladosporium tennuissimum As 97 d GI866 KC206494 Rhodotorula mucilacinosa Ba 97 d c2097 KC146554 Uncultured cyanobacterium Cy 95 e GI865 KC206477 Caulobacter crescentus Ap 94 e c4392 KC146571 Escherichia coli Gp 93 e s3934 KC146559 Uncultured bacterium Ac 93 e c3088 KC146584 Halomonas neptunia Gp 92 e 2,501 57,000 GI868 KC206495 Aspergillus restrictus As 99 d

102

Table 1. Cont.

Depth Age Sequence Accession Identity Core Closest BLASTn match b Phylum c (m) (ybp) number a number (%) GI869 KC206484 Aureobasidium pullulans Ba 99 d GI871 KC206489 Aspergillus conicus As 98 d GI873 KC206481 Rhodotorula mucilacinosa Ba 98 d GI872 KC206478 Caulobacter crescentus Ap 86 e 2,777 105,000 GI875 KC206483 Cryptococcus magnus Ba 99 d GI876 KC206487 Rhodotorula mucilacinosa Ba 99 d GI874 KC209502 Bacillus subtilis Fi 98 e 3,014 157,000 c2575 KC146576 Lactobacillus helveticus Fi 100 f c4301 KC146578 Lactobacillus helveticus Fi 100 f c1729 KC146569 Micrococcus luteus Ac 100 f CULT E-5 KC146583 Cladosporium tenuissimum As 99 d s3361 KC146561 Microcoleus vaginatus Cy 99 e c1654 KC146563 Microcoleus vaginatus Cy 99 e c552 KC146564 Uncultured bacterium Cy 99 e c1812 KC146565 Uncultured bacterium Cy 99 e c2330 KC146567 Uncultured bacterium Fi 99 e c785 KC146568 Uncultured bacterium Cy 99 e c3738 KC146577 Lactobacillus helveticus Fi 98 f c833 KC146573 Lactobacillus helveticus Fi 98 e c1018 KC146574 Lactobacillus helveticus Fi 98 f GI855 KC206493 Penicillium chrysognum As 98 d GI858 KC206480 Rhodotorula mucilacinosa Ba 98 d c1826 KC146566 Uncultured bacterium Fi 98 e c3135 KC146552 Uncultured cyanobacterium Cy 98 e c3856 KC146553 Uncultured cyanobacterium Cy 98 e s1137 KC146575 Lactobacillus helveticus Fi 97 e c2005 KC146570 Micrococcus luteus Ac 97 e c67 KC146562 Microcoleus vaginatus Cy 97 e c3509 KC146585 Brevudimonas diminuta Ap 95 f c3833 KC146587 Uncultured bacterium Bp 93 e c2683 KC146579 Pseudomonas stutzeri Gp 91 f c4709 KC146572 Bordetella petrii Bp 89 f s3724 KC146586 Bordetella pertussis Bp 88 f c2094 KC146588 Geobacillus sp. Fi 84 g Vostok 5G 316 10,500 GI878 Bacillus amyloliquifaciens Fi 100 e GI877 Davidiella tassiana As 99 d 900 57,000 None ------1,529 105,000 GI879 Alternaria tenuissimum As 100 d GI880 Rhodotorula mucilacinosa Ba 99 d 2,149 157,000 None ------a Cultures are indicated in bold font. Metagenomic/metatranscriptomic sequences are in standard 103 font. b Fungi are in red font. c Phyla: Ac—Actinobacteria; Ap—Alphaproteobacteria; As—

Ascomycota; Ba—Basidiomycota; Bp—Betaproteobacteria; Cy—Cyanobacteria; Fi—Firmicutes; Gp—Gammaproteobacteria. d Ribosomal RNA internal transcribed spacer (ITS1 and ITS2). e Ribosomal RNA small subunit gene. f Ribosomal RNA large subunit gene. g Glycerol kinase gene.

Figure 4. Number of cultures from GISP2D, Byrd and Vostok 5G ice core sections, based on age of ice. Black bars indicate total number of cultures, red bars indicate fungal cultures and blue bars indicate bacterial cultures.

! GISP 2D!

Number of Cultures Number T! F!B! 0! 0! <500! 10,500! 57,000! 105,000! 157,000!

! Name of Ice Core Section (ybp)! 5!

Byrd + Vostok 5G!

T!F!B! 0! 0!0!0! 0! 0! 0!0!0! Number of Cultures Number <500! 10,500! 30,000! 57,000! 70,000! 105,000! 157,000! Name of Ice Core Section (ybp)!

Figure 5. Cultures obtained from ice core sections based on atmospheric conditions at the time of deposition. Black bars indicate total number of cultures, red bars indicate fungal cultures and blue bars indicate bacterial cultures. Values for atmospheric measurements are from reference [23]. 104

20! !

T! 10! F! Number of Cultures Number B! 0! <210! 210-240! >240! <-3! -3 to 0! >0! <0.4! 0.4-1.0! >1.0! CO2 (ppmv)! Temperature (°C)! Dust (ppm)! ! ! Atmospheric Conditions!

2.1.3. Metagenomic/metatranscriptomic Analyses

A total of 107,700 high quality 454 sequence reads were obtained from the sequencing facility. After categorization by the MID primers, there were 25,748 high quality reads associated with the MID4 (GISP2D, 1,601 m) primer, 29,506 associated with the MID5 (GISP2D, 3,014 m) primer. 18,015 high quality reads associated with the MID6 (control 1) primer, and 34,063 high quality reads associated with the MID10 (control 2) primer. After assembly, a total of 15,891 contigs resulted. Sequences that were in common with the control samples were removed, leaving 33 sequences that were unique to the ice for analysis (Table 1). Nine unique Basic Local Alignment Search Tool (BLAST) hits were present in the 10,500 ybp GISP2D core section and 24 unique hits were present in the 157,000 ybp core section. All of the metagenomic/metatranscriptomic sequences were from bacteria (Table 1). There were five species common to both samples. All others were unique to one sample or the other. In the 10,500 ybp sample, all nine sequences were unique, while in the 157,000 ybp sample, 15 species were indicated among the 24 sequences. In general, the taxa in the cores were similar to organisms found in polar ice (both Arctic and Antarctic), soil, dust, water, hot springs, marine environments, deep-sea thermal vents and associated with animals. A number of isolates and sequences recovered from the ice in this study were those that have yet to be scientifically described. In the National Center for Biotechnology Information (NCBI) database, they are termed “environmental taxa” or “unidentified” or “uncultured.” In the 10,500 ybp ice, five of the sequences were closest to unidentified environmental taxa that were similar to Microcoleus and uncultured bacteria. Also, the 10,500 ybp GISP2D ice sample contained one sequence closest to a sequence of a Halomonas isolate that lives in marine environments, including deep-sea thermal vents. In the 157,000 ybp ice

section sample, fifteen sequences were closest to those from unidentified environmental species. 105

They were closest to members of Geobacillus, Micrococcus, Microcoleus and Pseudomonas. Five were associated with soil, three specifically with arid land soils, and two sequences were closest to Micrococcus, known to be saprotrophic.

2.1.4. Phylogenetics

For bacteria, phylogenetic trees were produced using SSU rRNA (small subunit ribosomal RNA; Figure 6) and LSU rRNA (large subunit rRNA; Figure 7) sequences. For fungi, the rRNA internal transcribed spacers (ITS1 and ITS2) were used, due to their increased precision for species determination for Eukarya. Phylogenetics was used to confirm the BLAST search results, as well as to refine species identities. The bacteria grouped into four clades, corresponding to phyla: Actinobacteria, Cyanobacteria, Firmicutes, and Proteobacteria (including Alphaproteobacteria, Betaproteobacteria and Gammaproteobacteria), the majority being in either the Cyanobacteria or the Firmicutes. Three sequences from the 157,000 year old ice core were closest to sequences from Microcoleus vaginatus and Phormidium autumnale that were isolated from soils, desert soils or deglaciated soil (including one from Antarctica). The other six sequences formed a separate Cyanobacteria clade, which was within the larger clade that included sequences from other Antarctic and arid soil taxa. The Actinobacteria clade included three sequences, all closely allied with sequences from environmental samples from soil or dust. Two were closest to sequences from Micrococcus luteus. Four sequences grouped within the Firmicutes clade. Two were from cultures whose sequences were within the genus Bacillus. In previous studies of Greenland ice cores, several Bacillus spp. were isolated [3]. Two metagenomic/metatranscriptomic sequences from the 157,000 year old ice core section most closely matched sequences from Marinococcus sp. and Sinococcus beijingensis, both of which had been isolated from saline soils. This is consistent with the low temperature and high atmospheric dust concentration at that time [23].

Figure 6. Phylogram of bacterial small subunit ribosomal RNA (SSU rRNA) sequences (maximum parsimony using phylogenetic analysis using parsimony (PAUP) [30]) from the GISP2D and Byrd ice cores. Sequences determined in this study are indicated in bold font. Sequences that start with CULT are isolates. All other sequences in bold were from the metagenomic/metatranscriptomic analysis. Ice core location and depth (meters below the glacial surface) are indicated. Phyla are on the right. The closest National Center for Biotechnology Information (NCBI) sequences to the queried sequences were selected from BLAST search results for use in the phylogenetic analysis. NCBI accession numbers (in parentheses) and sources of isolates (square brackets) are provided. Bootstrap

values (1,000 replications) are given for branches with support greater than 50%. 106

107

Figure 7. Phylogram of bacterial large subunit rRNA (LSU rRNA) sequences (maximum parsimony using PAUP [30]) from the GISP2D ice core. Sequences determined in this study are indicated in bold font. Ice core location and depth (meters below the glacial surface) are indicated for each. Phyla are on the right. The closest NCBI sequences to the queried sequences were selected from BLAST search results for use in the phylogenetic analysis. NCBI accession numbers (in parentheses) and sources of isolates (square brackets) are provided. Bootstrap values (1,000 replications) are given for branches with support greater than 50%. 108

s1867 GISP2D 1601 m (KC155351)! c2581 GISP2D 1601 m (KC155352)! c3738 GISP2D 3014 m (KC146577) 55 c833 GISP2D 3014 m (KC146573) c1018 GISP2D 3014 m (KC146574) c2575 GISP2D 3014 m (KC146576) 80 s1137 GISP2D 3014 m (KC146575) c4301 GISP2D 3014 m (KC146578) Firmicutes Lactobacillus crispatus (FN692037) [animal associated] Lactobacillus helveticus (CP000517.1) [animal associated, thermotolerant] 61 Streptococcus hyointestinalis (AB168127.1) [animal associated]

100 Bacillus mycoides (AF267883.1) [soil, animal associated] 96 Bacillus cereus (S43429.1) [soil, animal associated] Bacillus stearothermophilus (S43424.1) [soil, animal associated] 99 c3509 GISP2D 3014 m (KC146585) Alphaproteobacteria Brevundimonas diminuta (X87288.1) [soil, plant/animal associated] 97 Bordetella pertussis (X68323.1) [animal associated] Bordetella petrii (AM902716) [soil, animal associated] 80 Uncultured bacterium (GU926492.1) [sludge] 99 Betaproteobacteria 69 c3833 GISP2D 3014 m (KC146587) c4709 GISP2D 3014 m (KC146572) 80 s3724 GISP2D 3014 m (KC146586) Neisseria macacae (JQ042812.1) [animal associated]

97 Halomonas neptunia (FN257755.1) [hydrothermal vents] Gammaproteobacteria 64 Halomonas alimentaria (FN257749) [halophilic, soil, oil] c3088 GISP2D 1601 m (KC146584)! Ilyobacter polytropus (AJ307975.1) [mud] Fusobacteria 56 Borrelia burgdorferi (M88330.1) [animal associated] Spirochaetes Fibrobacter succinogenes (AB003380) [animal associated] Fibrobacteres Verrucomicrobium spinosum (AF245377.1) [aquatic, soil] Verrucomicrobia Thermotoga thermarum (AJ877022.1) [aquatic, thremophilic] Thermotogae Acidobacterium capsulatum strain DSM 11244 (AM180893.1) [microbial mat] Acidobacteria

100 Planctomyces limnophilus (AF245365.1) [aquatic] Planctomycetes Chlamydia trachomatis (U68443.2) [animal associated] Chlamydiae c2097 GISP2D 1601 m (KC146554)! 79 c3856 GISP2D 3014 m (KC146553)! Cyanobacteria Uncultured cyanobacterium (FJ805910.1) [environmental] c3135 GISP2D 3014 m (KC146552)! Mycoplasma pneumoniae (JQ390376.1) [animal associated] Tenericutes Chlorobium tepidum (NC_002932.3) [mud] Chlorobi Chlorobium limicola (M62805.1) [mud] 50 changes Fewer sequences from Cyanobacteria were found in the LSU rRNA gene data set. However, they all were within a clade that included an uncultured cyanobacterial sequence from an unidentified environmental sample. Firmicute sequences predominated in the LSU rRNA data set. All were most closely allied with sequences from Lactobacillus crispatus and L. helveticus. 109 Both are associated with animals, but similar taxa can be found in soils. Similarly, one

Alphaproteobacterium and three Betaproteobacteria were most closely related to sequences from bacteria associated with animals, but which can also be found in soil and sludge. One sequence most closely matched sequences from Gammaproteobacteria. It was closest to members of the halophilic and thermotolerant genus Halomonas. Species of Halomonas and Lactobacillus are thermotolerant. Coincidentally, dark granular material was found in the 10,500-year-old ice core section that resembles volcanic ash (data not shown). All of the fungi that were isolated and sequenced (internal transcribed spacer (ITS) regions) were either in the Basidiomycota or the Ascomycota. Rhodotorula was the most common genus (Figure 8). Several isolates (GI858, GI867, GI873 and GI876) were recovered from sections of the GISP2D ice core that were 10,500, 57,000 and 157,000 years old, as well as from a 105,000- year-old Vosok 5G ice core section (GI880). Most grouped closest to R. mucilaginosa, R. glutius and related isolates that had previously been recovered from marine, deep-sea or Lake Vostok ice samples. However, isolate GI866 was distant from all other isolates, as well as from sequences on NCBI. While it appears that it could be within the Sporidiobolales, it might be outside of the genus Rhodotorula. Isolates of Penicillium and Aspergillus also were common. The ITS sequence from isolate GI855 (from GISP2D, 3,014 m) was closest to P. chrysogenum isolates from marine and deep- sea sediments. Another isolate (GI862, from GISP2D, 1,601 m) also was close to a deep-sea isolate. Two isolates (GI868 and GI871, both from Vostok 5G, 2,501 m) were closest to an isolate of Aspergillus from Lake Vostok accretion ice. All of these isolates, as well as the isolates described in the NCBI database originated from cold high-pressure environments. This is likely a common trait for organisms that are able to survive in deep ice. The sequence of isolate GI869 (from Vostok 5G, 2,501 m) is closest to marine isolate within the Dothioraceae, which includes the genus Aureobasidium. However, the sequence differs from all other such sequences in this family on the NCBI database, as indicted by the branch length (Figure 8). Sequences from two isolates (GI861, from GISP2D, and GI877, from Vosok 5G; both at 10,500 ybp) are within the Cladosporium/Davidiella complex. They were closest to marine and Greenland ice taxa [8]. One isolate of Alternaria was recovered from the 105,000-year-old section from Vostok 5G. The sequence was closest to A. tenuissima, as well as to isolates from Greenland and from soil. One isolate (GI859) was closest to sequences from Fusarium isolates recovered from Greenland and northern Spain. Isolate GI860 (from GISP2D, 1,601 m) could not be placed within a specific taxon. It appears to be approximately midway between Alternaria and Fusarium (Figure 8). One additional basidiomycete was isolate GI875 (from GIDP2D, 2,777 m), whose sequence is closest to Cryptococcus magnus, a species isolated from soil that is also an opportunistic pathogen of humans and other mammals. 110

Figure 8. Phylogram of fungal rRNA internal transcribed spacer (ITS) sequences (with maximum parsimony on PAUP, [30]) from the GISP2D and Vostok 5G ice cores. Sequences that start with GI (in bold font) are the isolates from the ice sections. Ice core location and depth (meters below the glacial surface) are also given. Genera and phyla designations are on the right. Organisms belonging to the genera Rhodotorula, Alternaria and Cladosporium have previously been isolated from the GISP2D and the Vostok 5G ice cores. The closest NCBI sequences to the queried sequences were selected from BLAST search results for use in the phylogenetic analysis. NCBI accession numbers (in parentheses) and sources of isolates (square brackets) are provided. Bootstrap values (1,000 replications) are given for branches with support greater than 50%. 111

P. sp. (AF177735.1) [glacier]! 79 P. sp. (DQ681340.1)! GI 855 GISP2D 3014 m (KC206493)! P. sp. (DQ767595.1) [yellow sea]! 68 68 P. chrysogenum (AY373903.1)! P. chrysogenum (AM158209.1) [deep sea sediment]! P. sp. (AF177739.1) [glacier]! Penicillium P. sp. (DQ191308.1) [Glacial ice]! P. expansum (AF218786.1) [Antarctica]! 100 Penicillium sp. (AY373906.1)! 77 GI 862 GISP 2D 1601 m (KC206492)! Penicillium sp. (EU497960.1) [deep sea]! Penicillium sp. (AY391833.1) [ancient]! A. sp. (AY373864.1)! 100 A. sp. (AM901671.1)! GI 868 GISP2D 2501 m (KC206495)! Aspergillus A. sp. (EU108742) [Lake Vostok accretion ice]! GI 871 GISP 2D 2501 m (KC206489)! Dothioraceae sp. (EF060638.1) [marine]! 100 A. sp. (EF690466.1)! Aureobasidium A. sp. (AF455533.1)! GI 869 GISP 2D 2501 m (KC206484)! Uncultured soil fungus (DQ421000.1)! C. sp. (AF177734.1) [Greenland]! GI 877 Vostok 316 m (KC2064867))! Mycosphaerella sp. (AF362049.1)! 83 Apiosporina sp. (AY166451.1)! Cladosporium Davidiella sp. (EF679368.1)! 100 C. cladosporioides (AY251074.2)! Ascomycota C. sp. (DQ317332.1) [Ross sea]! GI 861 GISP 2D 1601 m (KC206488)! A. sp. (AF261661.1) [Greenland]! 95 A. sp. (EU326185.1) [soil]! 100 A. sp. (AY154712.1)! Alternaria 100 Uncultured soil fungus (DQ420890.1)! GI 879 Vostok 1529 m (KC206496)! A. tenuissima (EU315002.1)! GI 860 GISP 2D 1601 m (KC206482)! Fusarium sp. (AF261662.1) [Greenland]! 100 Fusarium sp. (DQ655726.1) [North Spain]! Fusarium Fusarium sp. (DQ459870.1)! GI 859 GISP 2D 1601 m (KC206484)! Sporidiobolales sp. (EF060603.1)! 100 R. glutinis (EF194846.1) [Marine]! Rhodosporidium sp. (AF444499.1)! GI 867 GISP 2D 1600m (KC206485)! 56 GI 873 GISP 2D 2501 m (KC206481)! GI 858 GISP 2D 3014 m (KC206480)! GI 880 Vostok 1529 m (KC206490)! 87 R. sp. (EU108795) [Vostok glacial ice]! GI 876 GISP 2D 2777 m (KC206487)! Rhodotorula R. sp. (EU108787) [Lake Vostok accretion ice]! R. sp.(EU108786) [Lake Vostok accretion ice]! R. mucilaginosa (EU108793) [Lake Vostok accretion ice]! R. sp. (DQ643075.1) [Deep sea]!

100 R. mucilaginosa (AM117824.1) [Deep sea]! Sporidiobolales sp. (EF060587.1) [marine]! 83 R. mucilaginosa (EF190221.1) [marine]! R. mucilaginosa (EF192223.1) [marine]! R. sp. (DQ386306.1)! 64 R. sp. (DQ015695.1) [Antarctica]! GI 866 GISP 2D 1601 m (KC206494)! C. sp. (AF444400.1)! 96 C. capitatum (AY052492.1) [cold adapted]! Uncultured soil fungus (AF294943.1)! Cystofilobasidium 77 C. sp. (AY029346.1) [Antarctica]! 99 Uncultured soil fungus (DQ421217.1)! C. magnus (AF190009.1)! Basidiomycota 100 GI 875 GISP 2D 2777 m (KC206483)! Cryptococcus Filobasidium globisporum (AF444336.1)! C. victoriae (AY040656.1) [Antarctic]! Ustilago maydis (AY345004.1)! 100 Pseudozyma proliﬁca (AM160639.1)! Pseudozyma Pseudozyma antarctica (AY641557.1)!

10 changes! 2.2. Discussion

2.2.1. Atmospheric Conditions and Microbial Deposition

Microbial cells are transported by wind, clouds and precipitation [12,31]. Some become entrapped in glaciers, becoming records of the atmosphere at specific points in time. In this study, two major questions are addressed: (1) Are the same species of microbes concurrently 112

deposited in ice in the Arctic and Antarctic?, and (2) Do atmospheric conditions affect the deposition of microbes in polar ice? The answer to Question 1 is, no. The sequences found in the Arctic differed from those found during the same time periods in the Antarctic, although very similar sequences were sometimes found in ice from both polar regions (Table 1, Figures 6–8). The answer to Question 2 is, yes. The highest numbers of isolates were from ice core sections that contained ice deposited during times of moderate to low atmospheric CO2 (<240 ppmv), low temperature (<−3 °C below current global mean) and low dust (<0.4 ppm). These correlations to the cultivation results may indicate increased preservation of the microbes during times of colder global mean temperatures. The metagenomic/metatranscriptomic data indicated that sequences from microbes common to arid and saline soils were deposited in the ice during a time of low temperature, low atmospheric CO2 and high dust levels. The presence of nucleic acids does not indicate whether viable cells are present, and therefore, some of the organisms represented in these samples may have been nonviable. The metagenomic/metatranscriptomic data might be more indicative of the nucleic acids that are in the highest concentrations in the ice, rather than an indication of intact microbes. The lower number of microbes in the Vostok sections could partly be due to the high elevation of the site and the fact that the Vostok site lies within a cold desert that receives much less snowfall compared to Greenland. Conversely, the higher number of microbes isolated from the Greenland ice cores may be related to a much warmer climate in the recent history compared to the Antarctic cores. During the Eemian interglacial (130,000 to 114,000 ybp) the northern hemisphere had warm temperatures comparable or higher to the temperatures of the Holocene period [32,33]. During this period, part of Greenland (southernmost part of Greenland) was covered with forests including a diverse array of conifers [34]. The warmer temperatures along with the higher precipitation and the geographically closer location to forests may be responsible for the higher number of organisms isolated from the Greenland ice cores. Distances to the nearest land masses with temperate climates likely contributed to the differences observed. The GISP2D site lies relatively close to North America, Europe and Asia, and winds often intersect the GISP2D site from these regions. On the other hand, most of the winds that reach the Vostok and Byrd sites originate in Antarctica, and the closest land mass is the narrowest part of South America. Therefore, the deposition of organisms from land masses onto Antarctica glaciers would be expected to be much less than onto Greenland glaciers. The numbers of cosmopolitan species isolated in Arctic and Antarctic regions along with spore trap data add credence to the point [35].

2.2.2. Isolated Fungi and Bacteria

All of the bacterial and fungal isolates grew at 22 °C. These isolates do not fit the profile of psychrophiles, whose optimal growth temperature is below 15 °C and do not grow at 113

temperatures above 20 °C. The isolates can be classified as psychrotolerant, as they can survive and grow at low temperatures while having the optimal and maximum growth temperatures above 20 °C. Earlier studies indicated that most of the polar ice habitats are dominated by psychrotolerant organisms [4,5,31,36]. Psychrophiles thrive and flourish in continuously cold environments like the sea ice where long periods of sustained cold temperatures act as the major selective pressure. However, in the polar regions, sharp and rapid changes in seasonal temperatures act as the main selective pressure. Significant environmental changes may kill many true psychrophiles, while generalists such as psychrotolerant species have better chances to survive. The organisms in these environments should be able to survive freeze thaw cycles that occur during the seasonal changes. Sharp changes in the surface temperatures have been observed in the Antarctic soils and have been recorded to go as high as 30 °C [37]. If fungi and bacteria from these surrounding environments are deposited in the glacial ice, we would expect to find psychrotolerant species among our isolates. It is important to note that the highest number of cultures were from ice core sections that were deposited during times of cold global temperature (e.g., 30,000, 70,000 and 157,000 ybp; Table 1, Figure 5). This might be caused by increased preservation of the microbes during transport and deposition due to the cold temperatures in the atmosphere. Species from three genera were isolated from both Greenland and Antarctic ice: Alternaria, Bacillus and Rhodotorula. However, in all cases, the species, as indicated by the rRNA sequences, differed from one pole to the other (Table 1, Figures 6–8). The taxon that was most often isolated was the genus Rhodotorula (Table 1, Figure 8), as was the case in our previous studies of Greenland and Antarctic ice core sections [5,12]. Members of this genus are hardy and adaptable, and have been isolated frequently from polar regions [12]. All of the Rhodotorula isolates were pigmented (shades of yellow, orange, pink and red), which might be important to survival in cold environments, and might explain their abundance in the cores analyzed [38]. Rhodotorula spp. remain viable after many freeze thaw cycles, which is another explanation for the abundance in the ice cores analyzed [12]. Isolates from the Greenland core sections included sequences closely related to Aspergillus and Penicillium. Isolate KC206495 (99% similar to Aspergillus restrictus) and KC206489 (99% similarity to Aspergillus conicus) were isolated from the 57,000-year-old ice core. Isolate KC206492 (98% similarity with Penicillium corylophilum) was isolated from the 10,500-year-old ice core and KC206493 (98% sequence similarity with Penicillium chrysogenum) was isolated from a 157,000-year-old ice core. Several species of Penicillium are tolerant to cold conditions and are known to survive for long periods of time [39]. Four of the isolates from Greenland ice cores were closely related to the species of Cladosporium, Cryptococcus and Aureobasidium. All have been frequently isolated from cold environments, such as Antarctic moss, which is one of the 114 richest microhabitats for fungi [40]. Cladosporium spores have been found in high numbers in

the air, especially in the temperate and tropical environments; the total fungal spore load in the Arctic region air is only about 5% [38]. Spore counts in the Antarctic atmosphere varied with the seasons, and highest spore counts were observed during the summer months, attributed to winds from northern regions [38]. One of the isolates from the Vostok ice core, one from GISP2D and three from the Byrd ice core were closely related to Alternaria sp. isolated from cryopegs in Siberia [15]. All of the cultured fungal sequences from this study had a 97% similarity or higher to NCBI sequences.

Several bacterial species isolated from the ice cores had high sequence similarities to the genera Bacillus and Caulobacter. Organisms belonging to these genera are tolerant to low temperatures and have been isolated from glacial and accretion ice samples from Lake Vostok [4]. Castello et al. [3] performed an extensive study of Greenland glacial ice (GISP2D and Dye 3) to report isolation of Bacillus through culturing meltwater. In a very similar method to theirs, we obtained our sequences related to Bacillus. All of the organisms isolated from the Arctic and Antarctic ice cores were closely related to species that are known to survive and thrive in cold environments and have been isolated from similar environments, indicating that the isolates probably are from the glacial ice and not contaminants.

2.2.3. Metagenomic/Metatranscriptomic Analysis

While we saw a decrease in the number of viable microbes cultured from older ice (Figure 4), the metagenomic/metatranscriptomic results suggest that more nucleic acid (probably carried within living and dead microbes) deposition in glaciers occurred during periods of low CO2, lowered temperatures and high dust (e.g., 157,000 ybp). The cultivation results indicated that more viable microbes were present in ice that had been deposited in the glacier during periods of moderate to low CO2, moderate to low temperatures and low dust (e.g., 500, 10,500 and 105,000 ybp). Together, the results indicate that cold temperatures are needed to retain cell viability, and to lessen degradation of nucleic acids. While decreases in temperature and CO2 were positively correlated with the recovery of viable microbes, increases in the amount of dust were negatively correlated. There have been reports of a variety of organisms present in Greenland ice (e.g., Ma et al. [8,9]). It has been shown that more long-range transport and deposition of microbes occurs during times of lower temperature, but it has been assumed that dust particles are responsible for the increases in transport. The results presented here argue against this assumption. Rather, temperature might be more important than the presence of dust particles. Microorganisms can be deposited through precipitation. Reports of Actinobacteria, Firmicutes, Proteobacteria (primarily Alphaproteobacteria, Betaproteobacteria and

Gammaproteobacteria) and Bacteroidetes, and genera such as, Pseudomonas, Sphingomonas, 115 Staphylococcus, Streptomyces and Arthrobacter were found in tropospheric clouds [31]. These

phyla are often found in cold environments, freshwater, marine water, soil or vegetation. Also, fog water bacteria consisted of Pseudomonas, Bacillus, Actinetobacter and several fungi [31]. It may be that most of the organisms isolated in this study were preserved and deposited in snow, not dust. Another important component to long-term survival is the ability to survive cold and desiccation. An example of a microbe found in arid conditions that was found in the ice cores was Microcoleus vaginatus (99% identity). This bacterium is found most often in biocrusts of arid land [41–43]. The phylogenetic analyses based on the SSU rRNA gene sequences (Figure 6) indicated that the majority of microbes that were in the 157,000 year old ice originated from soils, especially those from deserts and saline soils, and that they were hardy and desiccation resistant. This is consistent with conditions at the time, which were dry and cold [23]. Cyanobacteria and Firmicutes were present in the Greenland (GISP2D) ice core sections (Figures 5 and 6), including many that are similar to species from soil, arid soil and polar regions. Microorganisms in both hot and cold arid environments have mechanisms for adaptation to extremes in climate (temperature, salt levels and lack available water), such as accumulation of organic compounds that aid in resistance to desiccation and freezing [41]. Cyanobacteria have been reported to enrich the soil stability due to the protection and adaptability of the polysaccharide sheath that surrounds the filaments [42]. Microcoleus vaginatus is capable of synthesizing and degrading trehalose [41] from maltose and is highly motile [43]. Trehalose is a cryoprotectant that can help this organism survive in the ice [44,45]. Firmicutes surround themselves with sheaths that protect against desiccation, osmotic shock and temperature changes. The results presented here confirm that an overwhelming majority of sequences in ancient ice are derived from bacteria [1,4,5,46], while the greatest number of isolates were fungi. This might be due to the relative numbers of total microbes to the total viable microbes in each core section, although rigorous testing of this supposition will have to be performed in subsequent research. There was a diverse assemblage of bacterial phyla in the two ice samples analyzed by metagenomics/metatranscriptomics. In general, it is consistent with previous research, which indicates that Proteobacteria and Actinobacteria are commonly present in glacial ice [47,48]. Castello et al. [3] reported finding a sequence related to Rhodococcus erythreus, similar to our two sequences closely related to Micrococcus luteus, which are also in the Actinobacteria. In one study by Christner et al. [49], molecular analysis of glacial ice from a variety of global locations indicated approximately 50% of organisms to be Gram-positive, spore forming organisms. This could be due to the thick layer of peptidoglycan that provides strength and protection of the cell. Many of the microbes found in this study also form spores. Studies have shown Actinobacteria, Firmicutes, Proteobacteria and Fungi to be in the highest proportions in ice cores [48]. In our analyses, the same taxa were also found, with the addition of Cyanobacteria. The number of members of the Cyanobacteria was lower in the 10,500 ybp core section than in the 157,000 ybp 116 section. The reason for this remains unclear.

2.2.4. Global Mixing, Gene Flow and Genome Recycling

Organisms were isolated from 10 of the 12 ice cores used. However, there were differences among the ice core sections. Ice cores from the Arctic region consistently contained more viable organisms than those from Antarctica. Furthermore, species differed in the Arctic and Antarctic, suggesting that atmospheric transport, gene flow and genome recycling [50] between the poles occur infrequently. It appears that the two polar regions entrap microbes only from geographically local regions. Therefore, it is likely that latitudinal global mixing, leading to the global distribution of single genotypes, is inefficient over the long distance between the polar regions. Additionally, there appears to be a temporal component to preservation and deposition in polar ice. Far more isolates and sequences were found in ice core sections that had been deposited during periods of global cooling, when dust levels also were low. Thus, local and global conditions appear to be influential in the patterns of preservation of microbes and nucleic acids in Arctic and Antarctic ice.

3. Experimental Section

3.1. Ice Core Selection and Samples

Three ice core sections (104 m, 1,593 m, and 2,131 m; approximately 500, 30,000, and 70,000 ybp, respectively; [51]) from Byrd Station (80°1' S, 119°31' W, elevation 1,553 m), Antarctica; five core sections (99 m, 1,601 m, 2,501 m, 2,777 m and 3,014 m; approximately 500, 10,500, 57,000, 105,000 and 157,000 ybp, respectively [23]) from GISP 2D (Greenland Ice Sheet Project, core 2D; 72°36' N, 38°30' W, elevation 3,203 m); and four sections (316 m, 900 m, 1,529 m and 2,149 m; approximately 10,500, 57,000, 105,000 and 157,000 ybp, respectively [22,24,26]) from the Vostok 5G ice core (Vostok Station, Antarctica, 78°52' S, 106°50' W, elevation 3,488 m) were selected (Figures 1 and 9). The ice cores sections were chosen based on atmospheric CO2, dust, and global temperature at the time of ice deposition [23].

Figure 9. Depths and estimated ages of ice core sections used in this study.

Atmospheric CO2 levels (all in ppmv): 500 ybp = 250; 10,500 and 105,000 ybp = 230; 70,000 ybp = 210; 30,000 = 200; 57,000 and 157,000 = 180. Global temperature: (relative to present temperature, in °C): 500 ybp = +2; 10,500 and 105,000 = −3; 57,000 = −4; 30,000, 70,000 and 157,000 = −6. Atmospheric dust (ppm): 500 = <0.1; 105,000 = 0.1; 10,500 and 57,000 = 0.2; 70,000 = 0.5; 30,000 = 0.7; 157,000 = 1.2. Values for atmospheric measurements are from reference [23]. All values are from reference [23]. 117

Byrd! estimated ! GISP 2D! estimated ! Vostok 5G! (Antarctica)! age of ice! (Greenland)! age of ice! (Antarctica)! surface! (ybp)! (ybp)! 104 m! 500! 99 m! 316 m!

10,500! 900 m! 1000! ! !

1529 m! 1593 m! 30,000! 1601 m! 57,000! Ice Core Depth Core Ice (m below surface) below (m 2000! 2131 m! 70,000! 105,000! 2149 m!

2501 m! 157,000! 2777 m!

3000! 3014 m!

3.2. Surface Sterilization and Melting

The outer surfaces of the ice core sections were decontaminated before melting. Clorox (a 5.25% sodium hypochlorite solution) was used as the decontaminating agent. The decontamination protocol developed by Rogers et al. [52] was used for this procedure.

3.3. Scanning Electron Microscopy

An aliquot (meltwater from ice cores) of 5 mL was filtered through a 0.2 µm polycarbonate filter using a sterile syringe. The filter was transferred to a Petri dish with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.2). The filter was fixed in this solution for 1 h, followed by three 10 min rinses in 0.1 M phosphate buffer (pH 7.2). The filter was dehydrated with 40%, 60%, 118

80% and 95% ethanol solutions, sequentially for 10 min each and finally with 100% ethanol three times (10 min each). After dehydration, the filter was dried using a Samdri 780A critical point dryer. Then, the filter was cut into four pieces to be mounted into a coating unit. A Polaron E500 SEM coating unit was used to sputter coat the filters with a 5 nm thick gold-palladium coat. The filters then were observed in a scanning electron microscope (Hitachi S-2700, SEM). Control filters were prepared for SEM using sterilized water, taken through the fixation and dehydration procedure and observed.

3.4. Culturing

The meltwater (200 µL) from each shell of the ice cores was spread onto several types of solid media in duplicate, and incubated at 8 °C to test for the presence of viable microorganisms. The following media were used: malt extract agar [1.28% maltose, 0.27% dextrin, 0.24% glycerol, 0.08% peptone, 1.5% agar (pH 4.7)], potato dextrose agar [0.4% potato starch, 2% dextrose, 1.5% agar (pH 5.6)], rose bengal agar [0.5% soytone, 1% dextrose, 0.1% monopotassium phosphate, 0.005% rose bengal, 1.5% agar (pH 7.2)], nutrient agar [0.3% beef extract, 0.5% peptone, 1.5% agar (pH 6.8)], oatmeal agar [6% oatmeal, 1.25% agar (pH 6.0)], Sabouraud dextrose agar [1% enzymatic digest of casein, 2% dextrose, 2% agar (pH 7.0)], yeast extract agar [3% yeast extract, 3% malt extract, 0.5% peptone, 1% dextrose, 2% agar (pH 6.2)], acidic yeast extract agar [3% yeast extract, 3% malt extract, 0.5% peptone, 1% dextrose, 2% agar (pH 4.5)], meat-liver agar [2% meat liver base, 0.075% D(+)-glucose, 0.075% starch, 0.12% sodium sulfite, 0.05% ammonium ferric citrate, 1.1% agar (pH 7.6)], blood agar [1.5% pancreatic digest of casein, 0.5% papaic digest of soybean meal, 0.5% sodium chloride, 5% sheep’s blood, 1.5% agar (pH 7.3)], R2A [0.05% yeast extract, 0.05% proteose peptone No. 3, 0.05% casamino acids, 0.05% dextrose, 0.05% soluble starch, 0.03% sodium pyruvate, 0.03% dipotassium phosphate, 0.005% magnesium sulfate, 1.5% agar (pH 7.2)], Luria-Bertani agar [1% tryptone, 0.5% yeast extract, 0.5% sodium chloride, 1.5% agar (pH 7)], water agar [2% agar]. They were incubated for 2 weeks, monitored for any growth and transferred to 15 °C for 2 weeks. The plates then were incubated at 22 °C and monitored periodically for microbial growth. The cultures obtained were recorded and subcultured for future use. Throughout the decontamination and culturing procedures, control plates (two malt extract agar (MEA) and two lysogeny broth agar (LBA)) were employed in the laminar flow hood. Each plate was placed at one corner and exposed to the environment in the hood. They were incubated along with the culture plates and observed regularly for any growth.

3.5. PCR Amplification 119

DNA from the fungal and bacterial isolates obtained by culturing was subjected to polymerase chain reaction (PCR) amplification. DNA was extracted using a CTAB (cetyltrimethylammonium bromide) extraction method [53–55]. ITS4 and ITS5 primers were used to amplify the ribosomal DNA (rDNA) internal transcribed spacers (ITS1 and ITS2) and the 5.8S gene in fungi [56]. For bacterial isolates, primers 16S-2 and 23S-7 were used to amplify the rDNA intergenic spacer region (ITS1), along with portions of the small subunit and large subunit genes [57]. A GeneAmp PCR Reagent kit (Applied Biosystems, Carlsbad, CA, USA) was used for amplification. The composition of each 50 µL reaction was as follows: five microliters of cell suspension, 50 pmol of each primer, 10 pmol of each dNTP, 2U Taq DNA polymerase, 50 mM

KCl and 1.5 mM MgCl2. The program cycle used for amplification was: 95 °C for 8 min, 40 cycles of 94 °C for 1 min, 54 °C for 1 min 30 s and 72 °C for 2 min, followed by 72 °C for 8 min, and finally cooled to 4 °C. PCR reactions were subjected to electrophoresis on 1% agarose gels with TBE (90 mM Tris-Borate, 2 mM EDTA, pH 8.0) and 0.5 µg/mL ethidium bromide was used to view the amplification using UV light. For PCR amplification, sterilized water was used as a negative control. Rhodotorula mucilaginosa was used as the positive control for PCR with fungal primers and Bacillus subtilis as positive control for bacterial PCR.

3.6. Cloning and Sequencing

A TOPO TA cloning kit (Invitrogen, Grand Island, NY, USA) was used for cloning the amplified products. The amplified DNA from the fungal and bacterial isolates was ligated into the PCR 4-TOPO vector. The ligation reaction included: two and a half microliters of PCR product, 1.0 µL of salt solution (200 mM NaCl, 10 mM MgCl2) and 1.5 µL of vector (10 ng µL−1). The ligation reaction was carried out following the manufacturer’s directions. One Shot® TOP10 Competent E. coli cells were transformed with the ligation reaction. The plasmid DNA was isolated from transformed cells using the Cyclo-Prep Plasmid DNA isolation kit (Amresco, Solon, OH, USA). The plasmids then were tested for the presence of inserts using EcoRI digestion. The products were subjected to electrophoresis on 1% agarose gels with TBE and 0.5 µg/mL ethidium bromide to analyze the digestion products. Plasmids containing the inserts then were sent to Gene Gateway LLC (Hayward, CA, USA) for sequencing.

3.7. Direct Sequencing of Cultures

After subculturing, DNA was extracted from the growing cultures of meltwater. The extracted DNA was amplified using a GeneAmp (with AmpliTaq) PCR Reagent Kit (Applied Biosystems, Carlsbad, CA, USA). A final concentration of each reaction was 5 pmol of dATP, dCTP, dGTP, dTTP, 1× PCR Buffer [10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 0.001% (w/v) gelatin], 1 unit AmpliTaq DNA polymerase and fungal rRNA ITS and bacterial rRNA SSU 120 primers, in 25 µL total volume. Each reaction had 25 pmol of a forward and 25 pmol of a reverse

primer. The reaction was carried in a thermocycler and the program was: 94 °C for 4 min; then 40 cycles of 94 °C for 1 min, 55 °C for 3 min, 72 °C for 3 min; followed by an incubation for 10 min at 72 °C. After amplification the samples were quantified on 1% agarose gels in TBE (as above). Samples were cleaned using a Exo-SAP IT kit (Affymetrix, Inc., USB Products, Cleveland, OH, USA). Exo- SAP IT is composed of Exonuclease I and Shrimp Alkaline Phosphatase. It is a special buffer that removes any remaining primer and dNTPs from the PCR reaction. Samples were sent to GENEWIZ (South Plainfield, NJ, USA) for Sanger sequencing. They were examined and aligned using Clustal X or MAFFT [58,59], and then were subjected to phylogenetic analysis using maximum parsimony with PAUP [30].

3.8. Metagenomic/Metatranscriptomic Analyses

3.8.1. Ultracentrifugation

Two samples were selected for metagenomic/metatranscriptomic analysis, GISP2D 1,601 m and 3,014 m. The former sample corresponds to 10,500 ybp [22,24,26], a time of moderate and rising atmospheric CO2 and rising temperatures, while the latter corresponds to 157,000 ybp, a time of low atmospheric CO2 and low temperatures [50]. Starting from Shell 2 inward, the meltwater was subjected to ultracentrifugation at 32,500 rpm in a Beckman type 60 Ti rotor (100,000 rcf) at 4 °C for 15–17 h to concentrate cells, viruses and nucleic acids. The supernatants were placed into sterile BD Falcon 50 mL conical tubes and stored at −20 °C. Each pellet was resuspended in 50 µL of 0.1× TE (1 mM Tris HCl, pH 8.0; 0.1 mM EDTA, pH 8.0) and then the suspensions from each tube (for a given meltwater sample) were pooled (~500 µL total) and stored at −20 °C. Two additional samples were used as controls. The first control was autoclaved Nanopure water (18.2 MΩ, <1 ppb TOC) that was subjected to ultracentrifugation, as above. Nanopure water (18.2 MΩ, <1 ppb TOC) was the second control.

3.8.2. Isolation of Nucleic Acids

Each concentrated meltwater sample (as well as the two controls) was divided into two 200 µL fractions. One fraction was used for RNA extraction using TRIzol LS reagent (Life Technologies, Grand Island, NY, USA). To carry out homogenization, three volumes of TRIzol LS reagent were added, mixed and incubated to allow complete dissociation of proteins. Next, an equal volume of chloroform was added, followed by vigorous mixing. After incubation at room temperature, the samples were centrifuged in a microfuge at 16,100 rcf, which resulted in three layers. The top aqueous layer, containing the RNA, was removed and the RNA was precipitated 121

by adding 400 µL of isopropanol. This was placed at −20 °C for 2 h, followed by centrifugation in a microfuge to pellet the RNA. This pellet was washed with 800 µL 80% ethanol, air-dried and stored at −80 °C until needed. Directly before use, the RNA pellet was resuspended in 20 µL of sterile RNAse free water. The second 200 µL fraction of pooled ultracentrifuged meltwater was used for DNA extraction using a CTAB method, as described above.

3.8.3. Preparation of Samples for Sequencing

Complementary DNAs (cDNAs) were synthesized from the extracted RNAs. The procedure was performed using SuperScript Choice cDNA kit (Invitrogen, Grand Island, NY, USA), according to the manufacturer’s instructions, using 10 µL of the extracted RNA and 80 pmol of random hexamer primers. The cDNA was then mixed with 10 µL of extracted DNA (less than 1 ng/µL) from the same meltwater sample, and EcoRI (NotI) adapters (AATTCGCGGCCGCGTCGAC, dsDNA) were added using T4 DNA ligase. The final concentration of components in each reaction for EcoRI adapter addition was: 66 mM Tris-HCl

(pH 7.6), 10 mM MgCl2, 1 mM ATP, 14 mM DTT, 100 pmols EcoRI (NotI) adapters and 0.5 units of T4 DNA ligase, in 50 µL total volume. The reaction was incubated at 15 °C for 20 h. Then, the reaction was heated to 70 °C for 10 min to inactive the ligase. [Note: The cDNA were mixed in order to maximize the biomass of nucleic acids, necessary for successful pyrosequencing. Thus, the cDNA comprises a metatranscriptomic fraction and the DNA comprises the metagenomic fraction of each sample]. The products were size fractionated by column chromatography. Each 2 mL plastic column contained 1 mL of Sephacryl® S-500 HR resin. TEN buffer (10 mM Tris-HCl [pH 7.5], 0.1 mM EDTA, 25 mM NaCl; autoclaved) was utilized in washing the columns and eluting the samples through the columns. Fractions of approximately 40 µL were collected by chromatography. After measurement of the volume of each fraction, fractions 6–18 were precipitated to concentrate the DNA. Concentration consisted of adding 0.5 volumes (of the fraction size) of 1 M NaCl, and two volumes of −20 °C absolute ethanol. After gentle mixing, each was left to precipitate at −20 °C overnight. The fractions were centrifuged in a microfuge for 20 min at room temperature and decanted. Then, the pellets were washed with 0.5 mL of −20 °C 80% ethanol and centrifuged for 5 min. Finally, each of the DNA pellets was dried under vacuum and rehydrated in 20 µL of 0.1× TE buffer. After resuspension, fractions 6–18 were subjected to PCR amplification using EcoRI (NotI) adapter primers (AATTCGCGGCCGCGCTCGAC). The samples were amplified using a GeneAmp PCR Reagent Kit (as described above). The thermal cycling program was 94 °C for 4 min; then 40 cycles of 94 °C for 1 min, 55 °C for 2 min, 72 °C for 2 min; followed by an incubation for 10 min at 72 °C. Each fraction was subjected to gel electrophoresis on 1% agarose gels in TBE (as above) to confirm amplification, and to determine the size distributions. 122

454-specific primers were used to amplify the fragments, which also added the specific 454 sequences to the ends of the amplified products. All of the primers included the A-tag or the B- tag sequence (CGTATCGCCTCCCTCGCGCCA, CTATGCGCCTTGCCAGCCCGC, respectively), as well as the short four-base tag sequence (TCAG), unique multiplex identifiers (MID sequences (MID4: AGCACTGTAG, MID5: ATCAGACACG, MID6: ATATCGCGAG, MID10: TCTCTATGCG)), and EcoRI (NotI) adapter sequences (AATTCGCGGCCGCGTCGAC). The samples were amplified using a GeneAmp PCR Reagent Kit (as described above). All PCR products were cleaned with a PCR purification kit (Qiagen, Valencia, CA, USA). Once the sizes to be used for sequencing were confirmed by gel electrophoresis, the fractions containing distributions from approximately 200 bp to 1.5 kb were pooled. After concentration by precipitation and rehydration, the samples were quantified on 1% agarose gels in TBE (as above), using serial dilutions of plasmid (pGEM-3Z, from Promega, Madison, WI, USA) for quantitation. 3.8.4. Sequencing

DNA concentrations were estimated to be ~150 ng/µL for GISP 2D 3,014 m (sample 1); ~200 ng/µL for GISP 2D 1,601 m (sample 2); ~150 ng/µL for the ultracentrifuge sterile Nanopure water (sample 3); and finally the ~250 ng/µL for the unconcentrated Nanopure water. These were mixed in equimolar amounts in the proportion of 4:3:4:2, respectively, based on their estimated concentrations. Eight µL of Sample 1 (GISP 2D 3,014 m), 6 µL of Sample 2 (GISP 2D 1,601 m), 8 µL of Sample 3 (control ultracentrifuged water), and 4 µL of Sample 4 (control water) were mixed to produce a composite sample for sequencing. To this 26-µL mixture, 28 µL of RNase free water was added. A total of 54 µL (4.6 mg) of the amplified DNA at an approximate concentration of 85 ng/mL was sent to the University of Pennsylvania Medical School Sequencing Facility for Roche/454 GS FLX service.

3.8.5. Analysis of the Metagenomic/Metatranscriptomic Sequences

The sequences were assembled using MIRA 3.0.5 [60] and then were separated into four files, based on their multiplex identifier (MID) sequences (corresponding to the four ice core and control samples). Then, the 454 primers were clipped off (in silico) and sequence assembly was performed. The assembled sequences were used in Megablast searches for determination of sequence, taxon and gene similarities. The Megablast results (with the cut off parameter 1e-10) were retrieved for further analysis. Database sequences were reorganized based on gene regions. Multiple sequence alignments for ribosomal RNA small subunit (rRNA SSU) and large subunit (rRNA LSU) genes, separately, were performed with MAFFT (Multiple Sequence Alignment based on Fast Fourier Transform, Kyoto University, Kyoto, Japan) [58,59,61] global and local

alignment tools. Alignment files were converted into NEXUS format with SeaView (Université 123 de Lyon, Villeurbanne, France) [62,63]. Sequences from bacterial isolates CULT A-1 and CULT

B-1 were included in the phylogenetic analyses. PAUP (Phylogenetic Analysis Using Parsimony, [30]) maximum parsimony phylogenetic reconstructions were performed on the aligned SSU and LSU rRNA gene sequences. Bootstrap analyses (1,000 replications) were used to examine the support for the branches. Results from the phylogenetic analyses of the sequences from uncultured samples allowed a reevaluation of the microbial compositions of the two ice samples.

3.9. Phylogenetic Analysis for Fungal and Bacterial Sequences

The sequences obtained from cultures were subjected to BLAST searches and phylogenetic analysis in the same manner as the metagenomic sequences. The rRNA gene sequences obtained from both the bacterial and fungal isolates were used in BLAST searches of the GenBank NCBI- database to identify sequences of related taxa. Sequence alignments were created using ClustalX 2.0 for bacteria and fungi using the isolate sequences and related NCBI sequences. Alignment files were used to generate maximum parsimony phylogenetic trees using the program PAUP [30]. The phylogenetic trees were created using the heuristic search option, and gaps were treated as a fifth base. Bootstrap support using 1,000 replications also was determined using the same criteria.

4. Conclusions

A variety of microbes and their nucleic acids were recovered from Greenland and Antarctica ice core sections. In general, the viability of microbes decreased with increasing ice core age (Figure 4). While similar species were found in Arctic and Antarctic ice of the same age, the species differed, and in all cases the species composition within each ice core section was unique (Table 1, Figures 6–8). Among the bacteria, species of Bacillus were common, while among the fungi, species of Rhodotorula were the most common. Cultivation and metagenomic/metatranscriptomic studies indicated that viability was dependent on cold temperatures at the time of deposition (Figure 5). Low concentrations of atmospheric dust and

CO2 also were related to increased viability.

Acknowledgments

We thank NICL-SMO and the Ice Core Working group for allowing us access to the ice core sections used in this research. This research was partially funded by Bowling Green State University. We thank Marilyn Cayer for her valuable help with scanning electron microscopy.

References

1. Abyzov, S.S.; Poglazova, M.N.; Mitskevich, J.N.; Ivanov, M.V. Common features of microorganisms in ancient layers of the Antarctic ice sheet. In Life in Ancient Ice; Castello, 124

J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 240– 250. 2. Bell, R.; Studinger, M.; Tikku, A.; Castello, J.D. Comparative biological analyses of accretion ice from subglacial Lake Vostok. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 251–267. 3. Castello, J.D.; Rogers, S.O.; Smith, J.E.; Starmer, W.T.; Zhao, Y. Plant and bacterial viruses in the Greenland ice sheet. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 196–207. 4. D’Elia, T.; Veerappaneni, R.; Rogers, S.O. Isolation of microbes from Lake Vostok accretion ice. Appl. Environ. Microb. 2008, 74, 4962–4965. 5. D’Elia, T.; Veerappaneni, R.; Theraisnathan, V.; Rogers, S.O. Isolation of fungi from Lake Vostok accretion ice. Mycologia 2009, 101, 751–763. 6. Kellogg, D.E.; Kellogg, T.B. Frozen in time: The diatom record in ice cores from remote drilling sites on the Antarctic ice sheets. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 69–93. 7. Ma, L.J.; Catranis, C.; Starmer, W.T.; Rogers, S.O. Revival and characterization of fungi from ancient polar ice. Mycologist 1999, 13, 70–73. 8. Ma, L.J.; Rogers, S.O.; Catranis, C.; Starmer, W.T. Detection and characterization of ancient fungi entrapped in glacial ice. Mycologia 2000, 92, 286–295. 9. Ma, L.J.; Catranis, C.M.; Starmer, W.T.; Rogers, S.O. The significance and implications of the discovery of filamentous fungi in glacial ice. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 159–180. 10. Nichols, D.S. The growth of prokaryotes in Antarctic sea ice: Implications for ancient ice communities. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 50–68. 11. Rivkina, E.; Laurinavichyus, K.; Gilichinsky, D.A. Microbial life below the freezing point within permafrost. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 106–117. 12. Starmer, W.T.; Fell, J.W.; Catranis, C.M.; Aberdeen, V.; Ma, L.J.; Zhou, S.; Rogers, S.O. Yeasts in the genus Rhodotorula recovered from the Greenland ice sheet. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 181–195. 13. Vishnivetskaya, T.A.; Erokhina, L.G.; Spirina, E.V.; Shatilovich, A.V.; Vorobyova, E.A.; Tsapin, A.I.; Gilichinsky, D.A. Viable phototrophs: Cyanobacteria and green algae from the permafrost darkness. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 140–158. 125

14. Gilichinsky, D.A.; Khlebnikova, G.M.; Zvyagintsev, D.G.; Fedorov-Davydov, D.G.; Kudryavtseva, N.N. Microbiology of sedimentary materials in the permafrost zone. Int. Geol. Rev. 1989, 31, 847–858. 15. Gilichinsky, D.; Rivkina, E.; Bakermans, C.; Shcherbakova, V.; Petrovskaya, L.; Ozerskaya, S.; Ivanushkina, N.; Kochkina, G.; Laurinavichuis, K.; Pecheritsina, S.; et al. Biodiversity of cryopegs in permafrost. FEMS Microbiol. Ecol. 2005, 53, 117–128. 16. Abyzov, S.S. Microorganisms in Antarctic ice. In Antarctic Microbiology; Friedmann, E.I., Ed.; Princeton University Press: Princeton, NJ, USA, 1993; pp. 265–295. 17. Gilichinsky, D.A.; Vorobyova, E.; Erokhina, L.G.; Fyordorov-Dayvdov, D.G.; Chaikovskaya, N.R. Long-term preservation of microbial ecosystems in permafrost. Adv. Space Res. 1992, 12, 255–263. 18. Rogers, S.O.; Ma, L.J.; Zhao, Y.; Theraisnathan, V.; Shin, S.G.; Zhang, G.; Catranis, C.M.; Starmer, W.T.; Castello, J.D. Recommendations for elimination of contaminants and authentication of isolates in ancient ice cores. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 5–21. 19. Willerslev, E.; Hansen, E.J.; Poinar, H.N. Isolation of nucleic acids and cultures from fossil ice and permafrost. Trends Ecol. Evol. 2004, 19, 141–147. 20. Bidle K.J.; Lee, S.; Marchant, D.R.; Falkowski, P.G. Fossil genes and microbes in the oldest ice on Earth. Proc. Natl. Acad. Sci. USA 2007, 104, 13455–13460. 21. Lamb, H.H.; Woodroffe, A. Atmospheric circulation during the last ice age. Quat. Res. 1970, 1, 29–58. 22. Lüthi, D.; Floch, M.L.; Bereiter, B.; Blunier, T.; Barnola, J.M.; Siegenthaler, U.; Raynaud, D.; Jouzel, J.; Fischer, H.; Kawamura, K.; et al. High-resolution carbon dioxide concentration record 650,000–800,000 years before present. Nature 2008, 453, 379–382. 23. Petit, J.R.; Jouzel, J.; Raynaud, D.; Barkov, N.I.; Barnola, J.M.; Basile, I.; Bender, M.; Chappellaz, J.; Davis, M.; Delaygue, G.; et al. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 1999, 399, 429–436.

24. Ahn, J.; Brook, E. Atmospheric CO2 and climate on millennial time scales during the last glacial period. Science 2008, 322, 83–85. 25. Fischer, H.; Siggaard-Andersen, M.; Ruth, U.; Röthlisberger, R.; Wolff, E. Glacial/interglacial changes in mineral dust and sea-salt records in polar ice cores: Sources, transport, and deposition. Rev. Geophys. 2007, 45, 1–26. 26. Suwa, M.; Fischer, J.C.; Bender, M.L.; Landais, A.; Brook, E.J. Chronology reconstruction for the disturbed bottom section of the GISP2 and the GRIP ice cores: Implications for termination II in Greenland. J. Geophys. Res. 2006, 111, D02101. 27. Glikson, A. Milestones in the evolution of the atmosphere with reference to climate change. Aust. J. Earth Sci. 2008, 55, 125–139. 126

28. Miteva, V.I.; Sheridan, P.P.; Brenchley, J.E. Phylogenetic and physiological diversity of microorganisms isolated from a deep Greenland ice core. Appl. Environ. Microb. 2004, 70, 202–213. 29. Loveland-Curtze, J.; Miteva, V.I.; Brenchley, J.E. Herminiimonas glaciei sp. nov., a novel ultramicrobacterium from 3,042 m deep Greenland glacial ice. Int. J. Syst. Evol. Microbiol. 2009, 59, 1272–1277. 30. Swofford, D. PAUP: Phylogenetic Analysis Using Parsimony, Version 4.0b2a (PPC); Sinauer and Associates: Sunderland, MA, USA, 1999. 31. Amato, P.; Parazols, M.; Sancelme, M.; Laj, P.; Mailhot, G.; Delort, A. Microorganisms isolated from the water phase of tropospheric clouds at the Puy de Dôme: Major groups and growth abilities at low temperatures. FEMS Microbiol. Ecol. 2007, 59, 242–254. 32. Zachos, J.; Pagani, M.; Sloan, L.; Thomas E.; Billups, K. Trends, rhythms, and aberrations in global climate change 65 Ma to present. Science 2001, 292, 686–693. 33. Utescher, T.; Bruch, A.A.; Micheels, A.; Mosbrugger, V.; Popova, S. Cenozoic climate gradients in Eurasia—A paleo-perspective on future climate change? Paleogeog. Paleoclim. Paleoecol. 2011, 304, 351–358. 34. Willerslev, E.; Cappellini, E.; Boomsma, W.; Nielsen, R.; Hebsgaard, M.B.; Brand, T.B.; Hofreiter, M.; Bunce, M.; Poinar, H.N.; Dahl-Jensen, D.; et al. Ancient biomolecules from deep ice cores reveal a forested southern Greenland. Science 2007, 317, 111–114. 35. Vincent, W.F. Evolutionary origins of Antarctic microbiota: Invasion, selection and endemism. Antarct. Sci. 2000, 12, 374–385. 36. Franzmann, P.D. Examination of Antarctic prokaryotic diversity through molecular comparisons. Biodivers. Conserv. 1996, 5, 1295–1305. 37. Marshall, W.A. Seasonality in Antarctic airborne fungal spores. Appl. Environ. Microb. 1997, 63, 2240–2245. 38. Sonjak, S.; Frisvad, J.C.; Gunde-Cimerman, N. Penicillium mycobiota in arctic subglacial ice. Microb. Ecol. 2006, 52, 207–216. 39. Tosi, S.; Casado, B.; Gerdol, R.; Caretta, G. Fungi isolated from Antarctic mosses. Polar Biol. 2002, 25, 262–268. 40. Starkenburg, S.R.; Reitenga, K.G.; Freitas, T.; Johnson, S.; Chain, P.S.G.; Garcia-Pichel, F.; Kuske, C.R. Genome of the Cyanobacterium Microcoleus vaginatus FGP-2, a photosynthetic ecosystem engineer of arid land soil biocrusts worldwide. J. Bacteriol. 2011, 193, 4569–4570. 41. Belnap, J.; Gardner, J.S. Soil microstructure in soils of the Colorado plateau: The role of the cyanobacterium Microcoleus vaginatus. Great Basin Nat. 1993, 53, 40–47. 42. Boyer, S.L.; Johansen, J.R.; Flechtner, V.R.; Howard, G.L. Phylogeny and genetic variance in terrestrial Microcoleus species based on the sequence analysis of the 16S rRNA gene and 127 associated 16S-23S ITS region. J. Phycol. 2002, 38, 1222–1235.

43. Goodrich, R.P.; Handel, T.M.; Baldeschwieler, J.D. Modification of lipid phase behaviour with membrane-bound cryoprotectants. Biochim. Biophys. Acta 1998, 938, 143–154. 44. Snider, C.S.; Hsiang, T.; Zhao, G.; Griffith, M. Role of ice nucleation and antifreeze activities in pathogenesis and growth of snow molds. Phytopathology 2000, 90, 354–361. 45. Shtarkman, Y.M.; Koçer, Z.A.; Edgar, R.; Veerapaneni, R.; D’Elia, T.; Morris, P.F.; Rogers, S.O. Subglacial Lake Vostok (Antarctica) accretion ice contains a diverse set of sequences from aquatic and marine Bacteria and Eukarya. PLoS One 2012, in press. 46. Simon, C.; Wiezer, A.; Strittmatter, A.W.; Daniel, R. Phylogenetic diversity and metabolic potential revealed in a glacier ice metagenome. Appl. Environ. Microb. 2009, 75, 7519– 7526. 47. Miteva, V.; Teacher, C.; Sowers, T.; Brenchley, J. Comparison of the microbial diversity at different depths of the GISP2 Greenland ice core in relationship to deposition climates. Environ. Microbiol. 2009, 11, 640–656. 48. Christner, B.C.; Mosley-Thompson, E.; Thompson, L.G.; Reeve, J.N. Classification of bacteria from polar and nonpolar glacial ice. In Life in Ancient Ice; Castello, J.D., Rogers, S.O., Eds.; Princeton University Press: Princeton, NJ, USA, 2005; pp. 227–239. 49. Rogers, S.O.; Starmer, W.T.; Castello, J.D. Recycling of pathogenic microbes through survival in ice. Med. Hypotheses 2004, 63, 773–777. 50. Smith, A.W.; Skilling, D.E.; Castello, J.D.; Rogers, S.O. Ice as a reservoir for pathogenic human viruses: Specifically, caliciviruses, influenza viruses, and enteroviruses. Med. Hypotheses 2004, 63, 560–566. 51. Hamer, C.U.; Clausen, H.B.; Langway, C.C. Electrical conductivity method (ECM) stratigraphic dating of the Byrd Station ice core, Antarctica. Ann. Glaciol. 1994, 20, 115– 120. 52. Rogers, S.O.; Theraisnathan, V.; Ma, L.J.; Zhao, Y.; Zhang, G.; Shin, S.G.; Castello, J.D.; Starmer, W.T. Comparisons of protocols for decontamination of environmental ice samples for biological and molecular examinations. Appl. Environ. Microb. 2004, 70, 2540–2544. 53. Rogers, S.O.; Bendich, A.J. Extraction of DNA from milligram amounts of fresh, herbarium and mummified plant tissues. Plant Mol. Biol. 1985, 5, 69–76. 54. Rogers, S.O.; Bendich, A.J. Extraction of total cellular DNA from plants, algae and fungi. In Plant Molecular Biology Manual, 2nd ed.; Gelvin, S.B., Schilperoort, R.A., Eds.; Kluwer Academic Press: Dordrecht, The Netherlands, 1994; pp. D1:1–D1:8. 55. Rogers, S.O.; Rehner, S.; Bledsoe, C.; Mueller, G.J.; Ammirati, J.F. Extraction of DNA from Basidiomycetes for ribosomal DNA hybridizations. Can. J. Bot. 1989, 67, 1235–1243. 56. White T.J.; Bruns, T.; Lee, S.; Taylor, J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenteics. In PCR Protocols, a Guide to Methods and Applications; Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J., Eds.; Academic Press, 128 Inc. Harcourt Brace Janovich Publishers: New York, NY, USA, 1990; pp. 315–322.

57. Kabadjova, P.; Dousset, X.; Le Cam, V.; Prevost, H. Differentiation of closely related Carnobacterium food isolates based on 16S-23S ribosomal DNA intergenic spacer region polymorphism. Appl. Environ. Microb. 2002, 68, 358-366. 58. Katoh, K.; Misawa, K.; Kuma, K.; Miyata, T. MAFFT: A novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002, 30, 3059– 3066. 59. Katoh, K.; Kuma, K.; Toh, H.; Miyata, T. MAFFT version 5: Improvement in accuracy of multiple sequence alignment. Nucleic Acids Res. 2005, 33, 511–518. 60. Chevreux, B.; Wetter, T.; Suhai, S. Genome Sequence Assembly Using Trace Signals and Additional Sequence Information. Comput. Sci. Biol. Proc. German Conf. Bioinform. 1999, 99, 45–56. 61. MAFFT Version 7. Multiple Alignment Program for Amino Acid or Nucleotide Sequences. http://mafft.cbrc.jp/alignment/software/ (accessed on 25 August 2011). 62. Gouy, M.; Guindon, S.; Gascuel, O. SeaView version 4: A multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Mol. Biol. Evol. 2010, 27, 221–224. 63. SeaView, at the Université Lyon, Villeurbanne, France. http://pbil.univ-lyon1.fr/software/ seaview.html/ (accessed 8 October 2011).

© 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

129