Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

ProQuest Information and Learning 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 800-521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DETECTION, RECOVERY, ISOLATION AND CHARACTERIZATION OF BACTERIA IN GLACIAL ICE AND LAKE VOSTOK ACCRETION ICE

DISSERTATION

Presented in Partial Fulfillment of the Requirement for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Brent C. Christner, M.S.

The Ohio State University 2002

Dissertation Committee:

Dr. John N. Reeve, Adviser

Dr. Ellen Mosley-Thompson Approved by Dr. Lonnie G. Thompson

Dr. Olli H. Tuovinen

Dr. Charles J. Daniels A d v i s e r

Department of Microbiology

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number 3039458

UMI_ __ (fi)

UMI Microform 3039458 Copyright 2002 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

An extraction system has been constructed that melts

ice from the interior of ice cores and collects the

resulting melt water aseptically. Using this system,

bacteria entrapped in modem and ancient glacial ice from

worldwide locations and in an ice core extending into

accreted Lake Vostok ice have been isolated using

enrichment culture and identified by amplification and

sequencing of DNA-encoding 16S rRNA genes. In general, ice

cores from non-polar locations contained larger numbers and

species of cultivable bacteria than samples from polar

ices, presumably due to the closer proximity of terrestrial

biological ecosystems and exposed landscape. When compared

with other polar locations, higher numbers of isolates were

obtained from ices adjacent to the Dry Valley complex of

Antarctica, consistent with the influx of airborne

biological particles from local environments serving as the

primary factor controlling the numbers of microorganisms

present. The numbers of recoverable bacteria did not

correlate directly with the age of the ice, and isolates ii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were recovered from the oldest samples examined (>500K

years old) . The 16S rDNA sequences from bacterial isolates

and amplicons obtained directly from, samples position

within 6 different bacterial lines of descent (a— , (3—, and

y-proteobacteria, high and low G+C gram positive bacteria,

a n d the Cytophaga/Flavobacterium/Bacteroides) . Some of the

isolated bacteria have close phylogenetic relationships

with species originating from permanently cold

environments, and other ice core sites or different

portions (time periods) of the same core. Macromolecular

synthesis was demonstrated in bacteria frozen under

conditions analogous to those in glacial ice, and the

possibility exists that metabolic activity and repair may

occur during extended periods of glacial entrapment.

Several of the species identified in Lake Vostok accretion

ice are also related to glacial isolates and species from

other cold environments. These ice core studies have

provided a glimpse of the microorganisms likely to

inhabitant this potentially unique subsurface ecosystem.

Investigating microbial survival in ice and exploring

potential habitats for activity within the glacial and

subglacial environment has confirmed that these could have

served as refuge environments for life during periods of

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. global glaciation (Snowball Earth) , and has provided data

for extrapolations to the likelihood of microorganisms

surviving frozen in extraterrestrial habitats or during

interplanetary transport.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS

This study was made possible by financial support from

the National Science Foundation and the scientific

guidance, time and resources supplied by my advisers, John

Reeve, Ellen Mosley-Thompson and Lonnie Thompson. I feel

very fortunate to have worked amongst this fine group.

Their scientific dedication and encouragement have served

as a constant source of motivation. I am also grateful to

my thesis committee members Olli Tuovinen and Charles

Daniels, who provided an interesting subject for my general

exam, and have offered useful criticisms and comments on

project design. Many thanks are also due to Victor

Zagorodnov, who designed and constructed the automated ice

core sampler.

I am indebted to the knowledge and friendship of past

and present members of the Reeve laboratory, including

Kathryn Bailey, Trevor Darcy, Brian Hanzelka, Wen-tyng Li,

Frederic Marc, Rod Morgan, Suzette Pereira, Kathleen

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sandman, Rachel Samson, Divya Soares, Mark Xie, and Li Yu,

a former technician on this project.

I would also like to thank Andrea Wolfe, Laurie

Achenbach, Tom Schmidt, and Joel Klappenbach for assistance

with phylogenetic analysis, Ahmed Yousef for help with the

disinfection procedures, Scott Rogers for recommendations

on PCR amplification from low biomass material, Wade

Jeffrey for advice on macromolecular synthesis, and Dorota

Porazinska and Allison Murray for obtaining ice core

samples from the Canada Glacier. I am also appreciative of

funding supplied by NSF to participate in the Antarctic

Biology Course (2001, McMurdo Station, Antarctica) and by

OSU to participate in the 1999 Woods Hole Microbial

Diversity course.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

July 31, 1970 ...... B o m - Mt.Pleasant, Pennsylvania

1988-1992 ...... B.S. Molecular Biology, Westminster College, New Wilmington, Pennsylvania

1992-1993 ...... Research Technician, Eye and Ear Institute (UPMC) , Pittsburgh, Pennsylvania

1993-1996 ...... M.S. Microbiology, University of Dayton, Ohio

1997-present ...... Graduate Student, Department of Microbiology, The Ohio State University

PUBLICATIONS

Research Publications

1. Christner, B.C., E. Mosley-Thompson, L.G. Thompson. V. Zagorodnov, K. Sandman, and J.N. Reeve. 2000. Recovery and identification of viable bacteria immured in glacial ice. Icarus 144:479-485.

2. Christner, B.C., E. Mosley-Thompson, L.G. Thompson, and J.N. Reeve. 2001. Isolation of bacteria and 16S rDNAs from Lake Vostok accretion ice. Environ. Microbiol. 3:570- 577.

FIELDS OF STUDY

Major Field: Microbiology vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE 07 CONTENTS Page A b s t r a c t ...... ii

Acknowledgments ...... v

V i t a ...... vii

List of Tables ...... x

List of Figures ...... xi

C h a p t e r s :

1. General introduction ...... 1

Glacial ice, paleoclimatology, and the cold biosphere ...... 1

Microorganisms immured in glacial i c e ...... 10

Astrobiology implications of microbiological investigations of terrestrial glacial i c e ...... 24

Objectives of this study ...... 35

2. Procedures used to prevent contamination during sampling and analysis ...... 37

Introduction ...... 37

Materials and methods ...... 38

R e s u l t s ...... 52

Discussion ...... 55

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Isolation and characterization of bacteria and 16S rDNA sequences from glacial ice ...... 62

Introduction ...... 62

Materials and methods ...... 63

Results ...... 73

D i s c u s s i o n ...... 122

4. Macromolecular synthesis under frozen conditions ...... 129

Introduction ...... 129

Materials and methods ...... 130

Results ...... 132

D i s c u s s i o n ...... 141

5. Isolation of bacteria and 16S rDNA sequences from Lake Vostok accretion i c e ...... 148

Introduction ...... 148

Materials and methods ...... 152

R e s u l t s ...... 153

Discussion ...... 168

6. General discussion ...... 171

List of References ...... 179

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

Table Page

3.1 Inventory of glacial ice cores sampled ...... 77

3.2 Bacterial isolates from glacial ice cores ...... 98

3.3 Optimum growth temperature range and antibiotic resistance in isolated bacteria ...... 110

3.4 16S rDNA molecules amplified from >500,000 year old ice from Guliya, China ...... 118

5.1 Media inoculated with melt water from ice core section 3593 ...... 154

5.2 Bacteria isolated from deep Vostok ice core section 3593 ...... 156

5.3 16S rDNA molecules amplified from core 3593 melt water ...... 163

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

Figure Page

2.1 Construction of the sampling head and prototype ice core sampler ...... 42

2.2 Final design of the automated sampler ...... 44

2 . 3 Location of primers used to amplify and sequence 16S rDNA ...... 49

3.1 Global locations of sampling sites and ice cores available for study at the Byrd Polar Research Center ...... 64

3.2 Spread plates of sample from the Guliya ice cap on agar-solidi fed m e d i a ...... 75

3 . 3 Microorganisms and particulates filtered from glacial ice cores visualized by S E M ...... 80

3.4 Distribution of glacial isolates based on phylogenetic assignment to major bacterial divisions ...... 83

3 . 5 Phylogenetic analysis of a- proteobacterial isolates recovered from glacial i c e ...... 84

3 .6 Phylogenetic analysis of 0- and y- proteobacterial glacial isolates, and a member of the C/F/B line of descent ...... 87

3 .7 Phylogenetic analysis of low G+C Gram positive glacial.i s o l a t e s ...... 90

3 .8 Phylogenetic analysis of high G+C Gram positive glacial isolates ...... 93

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 .9 Phylogenetic analysis of y- proteobacterial sequences amplified from >500,000 year-old ice from Guliya, China ...... 115

3.10 Freeze- tolerance in glacial isolates, related species, and E.coli ...... 121

4.1 Incorporation of [3H]-thymidine and [3H]-leucine into TCA-precipitated material by Trans1 and G200-C1 during the freezing process ...... 133

4.2 Incorporation of [3H]-thymidine and [3H]-leucine by strains Transl and G200-C1 during a 206 day incubation at -15°C...135

4.3 Incorporation of [3H]-thymidine and [3H]—leucine by E. coli during a 102 day incubation at -15°C ...... 136

4.4 Incorporation of [3H]-thymidine and the number of cfu m l ' 1 for Transl and G200-C1 during a 100 day incubation at — 1 5 ° C ...... 138

4.5 Incorporation of [3H]-thymidine and [3H]-leucine at -15°C by strains Transl and G200-C1 over 23 days in the presence of DNA and protein synthesis inhibitors ...... 139

4.6 Incorporation of [3H]-thymidine and [3H]-leucine at -15° and -70°C by strains Transl and G200-C1 over 50 d a y s ...... 142

5.1 Origin of deep Vostok ice core section 3593 ...... 149

5 .2 Phylogenetic analysis and scanning electron micrographs of bacterial isolates from core section 3593 ...... 157

5.3 Filaments in anaerobic enrichments ...... 160

xii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.4 Phylogenetic analysis of the 16S rDNAs amplified from core section 3593 ...... 164

5.5 Scanning electron micrographs of cells retained on the surface of a 0.2 nm isopore filter after concentration of core section 3593 melt water ...... 167

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

GENERAL INTRODUCTION

Glacial ice, paleoclimatology# and the cold biosphere

Distribution and formation of glacial ice

Snowfall accumulates into continental ice sheets in

the polar regions and globally at high altitudes.

Depending on the topological nature of the accumulation

environment, high elevation ice fields are termed valley or

alpine glaciers, and ice caps when a flat bedrock surface

or volcanic crater is completely covered in ice. The

expansive ice sheets of Greenland and Antarctica cover -10%

of Earth's terrestrial surface with ice, and contain -70%

of the fresh water on the planet (Patterson 1994). Earth's

climate is currently in an interglacial stage of a 100,000

year cycle, caused largely by episodic changes in the

planets axial tilt and ellipticity of its orbit around the

Sun. During the last glacial maximum 18,000 years ago, sea

levels were «120 m lower than today and the north polar ice

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cap advanced to cover 5 million square kilometers,

blanketing what is now Canada and half of the United States

(Hughes 1998) .

The transformation of firn (granularized and compacted

snowflakes enduring a season without melting) to glacial

ice is a complex process that occurs at rates and depths

dependent on the air temperature, amount of snowfall

accumulation, moisture content of the snow, and whether the

glacial surface experiences annual cycles of melting and

freezing (Patterson 1994). As overlying snowfall applies

pressure, firn crystals glide and bond to other crystal

planes, effectively squeezing intervening air spaces

together into ice-entrapped bubbles. Gases are not able to

diffuse through solid ice; however, air within firn can mix

freely with the atmosphere. Therefore, an air bubble

within glacial ice does not originate from precisely the

same time point as the surrounding ice, and differences

range from hundreds to several thousand years (Patterson

1994) . The firn depth is often only a few meters in

mountain glaciers, but can be a hundred meters or more in

polar ice sheets.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. When a glacier accumulates to greater than 20 meters

in height, the ice flows much like a viscous liquid.

Gravity exerts a vertical force on the surface, causing the

ice to be pushed out laterally (Hughes 1998) . The highest

elevation of a mountain glacier and the interior of an ice

sheet are referred to as the zone of accumulation,

representing the regions where snowfall is added and

compressed into ice. The zone of ablation is the area where

material is lost from the glacier, by melting or the

calving of icebergs from an ice shelf. Based on the

climatie-dependent ratio of gain to loss, or mass balance,

a glacier will expand, contract, or maintain equilibrium

(Patterson 1994; Hughes 1998).

High elevation glaciers, especially those in the

tropics, are limited by regional topography with flow to

lower altitudes and the ablated mass often serving as a

primary water source for many important rivers. Polar ice

sheets, on the other hand, flow outward from a massive

dome-like accumulation zone to a thinner outer margin, and

are typical of the extensive glaciers that covered the

Earth during the climax of the last Ice Age (Hughes 1998) .

Ice shelves are fed by glacial flow and rapidly moving ice

streams capable of transferring large quantities of inland

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ice to costal calving zones. Such remote and seemingly

inconsequential frozen environments have enormous impacts

on the global climate, and are vital to discussions of

apparent global warming trends. An increase in global

temperature sufficient to melt only the West Antarctic ice

sheet would raise global sea level by 5 m (Alley and

Bindschadler 2001) .

Paleoclimatic Inferences from glacial ice cores

Archived chronologically within the Earth's ice caps

and ice sheets are samples of the atmospheric constituents

from different times in the past. Thus, by examining the

physical and chemical properties of ice cores, such proxy

data can be used to elucidate past regional and global

climatic conditions. The ratio of 180/160 and D/H (5180 and

5d, respectively) in precipitation is dependent on

temperature, anion/cation concentrations reflect the

atmospheric chemistry, dust concentrations reveal the

turbidity of the atmosphere, entrapped sulfates record

volcanic eruptions, changes in nitrate concentrations can

indicate the expansion and contraction of local vegetation,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and trace element concentrations can be used to monitor

anthropogenic emissions before and after the industrial

revolution (Patterson 1994).

Natural records such as deep sea sediments, coral

reefs, and tree rings also supply valuable information

about past climate change. However, proxy data from ice

cores contain more detail about more paleoclimatic

parameters, including the precise gaseous composition of

ancient atmospheres. Ice cores retrieved from Greenland

(Dansgaard et al. 1993; Grootes et al. 1993), Antarctica

(Lorius et al. 1985), and the Tibetan Plateau (Thompson et

al. 1997) have been used to reconstruct the last glacial-

interglacial cycle, and the recent completion of the Vostok

(Antarctica) core in 1998 now extends that history back

through the last 4 glacial cycles, covering approximately

420,000 years of climate history (Petit et al. 1999). The

immensity of polar ice sheets has a drastic effect on local

(as well as global) weather patterns, whereas smaller

tropical and subtropical glaciers are incapable of creating

a significant climatic disruption, and are therefore more

sensitive in recording higher frequency climatic

perturbations (Thompson et al. 1997, 1998, 2000). As such,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. data collected worldwide from ice cores are beginning to

provide a more detailed understanding of the mechanisms

which regulate global climate change.

The Neoproterozoic and Paleoproterozoic snowball Earth

The Earth's climate system has always been dynamic,

but perhaps no degree of environmental fluctuation

parallels in severity of the conditions supposedly

responsible for two global periods of glaciation, that

occurred 600-800 million and 2.3-2.5 billion years ago

(Evans et al. 1997; Hoffman et al. 1998; Kirschvink et al.

2000) . In fact, the only evidence for glacial deposition

in 4 billion years of pre-Cambrian history comes from

geological sedimentary materials originating from these

time periods. Budyko (1969) developed a mathematical

climate model that predicted a runaway freeze on the

Earth's surface when ice covers latitudes >30° north or

south of the equator, due to a positive ice-albedo

feedback. Such a drop in global temperature would create a

thick layer of ice that could then seal marine habitats

from the atmosphere, blocking photosynthetic primary

production and subsequently resulting in the world's

ocean's becoming anoxic. The presence of tropical

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Neoproterozoic glacial deposits at sea level locations in

the geological record followed immediately by cap

carbonates and banded iron formations (BIFs) are

interpreted as consistent with a snowball Earth scenario

(Kirshvink 1992; Hoffman et al. 1998).

But what could trigger such an extreme perturbation in

the climate system to allow a runaway freeze, and even more

perplexing, how could the process be reversed? The

clustering of continents in the tropics may have induced

the snowball by remaining ice-free as the Earth cooled and

the poles covered with ice. Under these conditions,

tropical carbon burial coupled with an increased polar

albedo might initiate a global freeze (Hoffman and Schrag

2000) . Decreased levels of atmospheric CO 2 might also

result from the breakup of the Rodinia supercontinent,

which would have increased costal margins and carbon burial

(Hoffman et al. 1998) . Kirshvink (1992) noted that the

absence of a silicate weathering system on a snowball Earth

would not provide a sink for CO 2 . Therefore, CO 2 out-

gassing from volcanic sources would accumulate to very high

levels, eventually reversing the climate abruptly to hot

house conditions, as melting ice decreases albedo and

increases water vapor in the atmosphere, further amplifying

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the warming effect. Hoffman et al. (1998) report rapid

deposition of cap carbonates (~40 cm/year) after escape

from the global freeze, with pre- and post-snowball S13C

data consistent with a significant decrease in biological

activity preceding glaciation, possibly due to the cooling

of the Earth, followed by a gradual rebound to preglacial

v a l u e s .

Since a snowball Earth would take 4-30 million years

to reverse (Hoffman et al. 1998), such a long period of

global freeze and modified geochemistry would have had

drastic consequences on biological ecosystems established

prior to these events. The oceans would have been covered

with 1-2 km of ice, although there was probably no more

than 10 m of ice in the tropics (Hoffman 2001, OSU Geology

Seminar), which may have provided potential ecological

niches for photosynthetic sea ice communities. Sealed

beneath the sea ice, however, the world ocean was

geochemically modified by the circulation of reduced

hydrothermal fluids at mid-oceanic spreading zones, arguing

for the survival of only anaerobic chemoautotrophic or

heterotrophic communities. Hot springs on the sea floor

and terrestrial surface may have served as oases for life

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. during these lengthy crises, providing thermal and chemical

gradients that could support a variety of microbial

lifestyles. Microorganisms might also have survived frozen

within the expansive ice cover, until post-snowball melting

reintroduced environments suitable for activity.

It is noteworthy that the Cambrian explosion, a period

where all 11 animal phyla rapidly evolved, occurred

immediately after the last Neoproterozoic snowball Earth

(Hoffman and Schrag 2000) . Rather than representing some

kind of supposed biological pinnacle, etched from several

billion years of evolutionary fine tuning, the emergence of

metazoan life took place swiftly after a global

catastrophe, and may have been triggered by biological

innovations forged to survive on a frozen planet.

Ironically, the deep-rooted phylogeny of thermophilic

species on the tree of life, generally interpreted to mean

that life originated in a thermophilic setting, may instead

represent the consequence of an evolutionary "bottle-neck"

imposed by the extremes of multiple snowball Earth events

(Kirschvink et al. 2000).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Microorganisms impaired in glacial ice

Cryogenic preservation of airborne biological material deposited in glacial ice

Glacial ice sheets entrap and preserve aerosolized

biological material (i.e. insects, plant fragments, seeds,

pollen grains, fungal spores and microorganisms) deposited

in snowfall. Ice cores extending thousands of meters below

the glacial surface can represent hundreds of thousands of

years of snowfall accumulation, and the assemblages of

microorganisms immured chronologically within a core are

species that were distributed in the atmosphere at

different times in history. Studies indicate that the

topography, local and global environmental conditions, and

proximity of ecosystems contributing biological particles

to a particular air mass influence the concentration and

diversity of airborne microorganisms (Lighthart et al.

1995; Giorgio et al. 1996; Fuzzi et al. 1997; Marshall and

Chalmers 1997). Thus, depending on the geographical

location and climate history of a glacial ice sheet,

microorganisms preserved sequentially within these strata

are mainly derived from nearby ecosystems, but may also

originate from distant ecological sources.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Microorganisms aerosolized from water, exposed soils

and rock surfaces can travel large distances on atmospheric

currents, often in a viable, but dormant state. Amazingly,

some air conditions actually provide a medium for growth,

and microbial metabolism has been detected in fog particles

(Fuzzi et al. 1997) and clouds (Sallter et al. 2001). For

an airborne microorganism to retain viability, the stress

associated with atmospheric transport, namely desiccation

and exposure to ultraviolet (UV) radiation, must not result

in a lethal level of unrepairable cellular damage. During

desiccation, a decrease in the cellular water content

results in structural damage to the cell membrane and

protein denaturation (Marthi and Lighthart 1990; Geiges

1996) . The composition and structure of Gram-negative

bacteria make them more vulnerable to dehydration than

Gram-positive bacteria, which possess a more rigid cell

wall (Marthi 1994) . Many Gram-positive genera also form

spores, which are generally resistant to environmental

abuse. Desiccation is also known to increase survival to

radiation, because much of the substantial damage induced

by exposure to ionizing radiation results from free-radical

formation, which is dependent on the presence of water

(Mattimore and Battista 1996). Microbial strategies for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. high levels of resistance to radiation include possessing

UV-absorbing carotenoid pigments, genetic redundancy, and

efficient enzymatic mechanisms to excise and repair

dimerized thymine bases (Marthi 1994; Atlas and Bartha

1998; Fogg 1998).

In addition to the severe conditions associated with

atmospheric transport, microorganisms deposited in glacial

ice are exposed to the physical and osmotic stress

associated with freezing and thawing, and low temperatures.

Ice crystals initially form in the extracellular phase, and

solute exclusion draws water from the cell, damaging the

integrity of the cell membrane (Fogg 1998) . The stress of

freezing is therefore a problem of water management. Many

plants, animals, and microbes adapted to freezing

conditions therefore produce compatible solutes, like

betaine and glycine, which reduce the shock of such osmotic

imbalance (Marthi and Lighthart 1990). Thermal hysteresis

antifreeze proteins have also been identified in bacteria,

and these cold-induced proteins function to prevent damage

initiated by intra- and extra-cellular ice crystal

formation (Willimsky et al. 1992; Duman et al. 1993).

Another consequence of low temperature is a decrease in

membrane fluidity, so for the cell membrane to avoid the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. "gel-crystalline" state, many organisms increase the

proportion of unsaturated fatty acyl chains in the lipid

bilayer (Finegold 1996) .

These injurious effects, however, are largely

dependent on the internal water content of the cell, and

since most airborne particulates would be desiccated, the

freezing process should cause minimal cellular damage,

considering the anhydrous nature of such microbial

aerosols. On the other hand, this would suggest that

airborne microflora are dormant, and unlikely able to

induce the protection responses discussed above.

Intriguingly, when an organism is stressed under a

particular condition, such as starvation or osmotic stress,

the induced synthesis of stress proteins cross-protects the

cell from various other traumas (Morita 2000) . Similar

strategies are required for survival in both the atmosphere

and under frozen conditions, and indeed, there is a

relationship between freeze- and desiccation-resistance in

many organisms (Fogg 1998) .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Longevity of microorganisms entrapped in ancient materials

Cells and macromolecules remain preserved for millions

of years in permafrost, (Gilichinsky et al. 1993; Shi et

al. 1997;) amber, (DeSalle et al. 1992; Cano et al. 1995;

Greenblatt et al. 1999) and salt crystals (Vreeland et al.

2000) . Inevitably during long-term preservation,

macromolecular damage accumultes, resulting from continuous

exposure to natural radioactive sources. Temperature and

the hydration of nucleic acids and protein strongly

influence the rate of depurination and L-amino acid

racemization, respectively (Lindahl 1993; Bada et al.

1994) . Ancient samples in permanently cold regions

preserve DNA for ~105 years, whereas DNA survival in warmer

climates is restricted to a few thousand years (Poinar et

al. 1996). Racemization rates of amino acid L-enantiomers

were correlated with the preservation of DNA in ancient

amber samples, and Poinar et al. (1996) were only able to

detect DNA templates of amplifiable length in samples with

low D/L ratios. Amino acids in amber have retarded

racemization rates, and the observed stereochemical

preservation is likely a result of the anhydrous nature of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amber, which prevents abstraction of the amino acid a-

hydrogen (bound adjacent to carbonyl group) by water (Bada

et al. 1994).

In addition to preserving biological macromolecules,

viable microorganisms have been reported to survive in

ancient geological materials for millions of years

(Gilichinsky et al. 1992; Shi et al. 1997; Greenblatt et

al. 1999; Vreeland et al. 2000) . Estimations of microbial

longevity based on extrapolations of data collected over

rather short time frames predict survival ranges from

thousands to millions or even billions of years (Kennedy et

al. 1994). Although the degree of thermal decay (i.e.,

increase in entropy) in a microorganism dormant under known

environmental conditions over a given time frame can be

theoretically determined, it requires an estimate of the

minimum macromolecular decomposition needed to be lethal.

Decay rates are known to be drastically reduced when

proteins and nucleic acids are stabilized within materials

with low water activity, such as amber (Bada et al. 1994) ,

and this could also pertain to ice.

It is also possible that entrapped microbes carry out

a slow rate of metabolism, which allows repair of

macromolecular damage, but not growth. Microenvironments

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. within salt crystals (Vreeland et al. 2000), amber

(Greenblatt et al. 1999), and permafrost (Gilichinsky et

al. 1992; Shi et al. 1997) appear to possess the conditions

necessary for active metabolism (i.e., liquid water,

carbon, electron donors and oxidants). Therefore, it is

not possible to determine if biological preservation in the

abovementioned environments results from anabiosis (Morita

2000) or a low level of metabolism that counteracts the

dupurination and racemization processes.

Frozen microbial ecosystems

The biology of permanently cold environments in the

Arctic and Antarctic have received relatively little

investigation. Similar to their high temperature

counterparts, frozen ecosystems are dominated by

microorganisms with trophic assemblages (such as microbial

mats) found only in other extreme environments. In recent

years, a number of investigators have documented microbial

communities in permanently ice-covered lakes (Fritsen and

Priscu 1998; Olson et al. 1998; Paerl and Priscu 1998;

Priscu et al. 1998; Takacs and Priscu 1998; Brambilla et

al. 2001), Antarctic sandstone (Friedmann 1982; McKay and

Friedmann 1985; Hirsch et al. 1998; Sun and Friedmann

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1999), permafrost (Gilichinsky et al., 1993; Shi et al.,

1997; Vorobyova et al. 1997; Rivkina et al. 2000), and sea

ice (Bowman et al. 1997; Gosink et al. 1997; Junge et al.

1998; Staley and Gosink 1999). Many-of these environments

never experience temperatures above the freezing point of

water; life nevertheless persists under arguably the

harshest conditions in the biosphere.

Despite the dry, cold conditions of Antarctica's Dry

Valleys, microbiological surveys of this polar desert have

documented an abundance of cryptoendolithic lichen

(Friedmann 1982; McKay and Friedmann 1985), associations of

fungi and bacteria that together inhabit pores in the

dominating sandstone. The endolithic communities so evades

exposure to abrupt temperature changes and solar

irradiation in these harsh surroundings, and estimates

indicate generation times as long as 104 years (Sun and

Friedmann 1999) . The Dry Valley complex is subjected to

persistent katabatic winds and is a substantial portion of

the Antarctic landscape that has exposed rock surfaces and

soil. The vast majority of the continent (~98%) is covered

in kilometers of glacial ice. The Dry Valleys have 24

hours of daylight during the austral summer, and rock

grains of aeolian origin on the surfaces of the permanently

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. frozen lakes, are warmed by the sun, and melt into the ice.

Pockets of liquid water are created on the surface and

within the ice, and these contain sufficient nutrients to

support the growth of microbial communities (Olson et al.

1998; Paerl and Priscu 1998; Priscu et al. 1998), many of

which are then frozen and trapped completely within the ice

during the austral winter.

The annual freezing of the north and south polar

oceans covers >7% of Earth's surface with thick pack ice

(Staley and Gosink 1999) , which drastically influences the

surface and underlying water column ecology. During sea

ice formation, liquid brine pockets created during ice

crystal consolidation become increasingly concentrated.

Solute exclusion during freezing can elevate brine channel

salinities to >1508b that of seawater, allowing supercooled

hypersaline solutions to exist at temperatures as low at

-15°C (Bowman et al. 1997) . Planktonic microorganisms

incorporated into sea ice during crystallization

subsequently develop into spatially defined communities in

and on the ice surface. Biological activity is highest in

the lower 10-20 cm of the ice at the seawater interface,

where the enriched microbial assemblage is termed the sea

ice microbial community (SIMCO) . This consists of diatoms,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. protozoa, bacteria, fungi and invertebrates, contains

higher nutrient levels than seawater, and possesses

psychrophilic species that appear to be specialized for

survival and activity in the sea ice environment (Bowman et

al. 1997; Junge et al. 1998; Staley and Gosink 1999).

Subterranean layers of permafrost in polar locations

contain rocks, sediment, biological detritus, and organisms

(both fossilized and viable) that originated from

previously unfrozen soil environments. Such frozen

deposits underlie -20% of Earth's terrestrial surface.

They contain 107-109 microbial cells per gram, and can be

millions of years old, with the oldest samples studied to

date originating from Siberia (Shi et al. 1997; Vorobyova

et al. 1997) . Gilichinsky et al. (1993) hypothesize that

unfrozen films of water present on mineral surfaces in

permafrost may protect microorganisms from freezing. It is

also suggested that this refuge may serve as a habitat for

activity, given that many permafrost isolates grow at

temperatures as low as -10°C. The significance of liquid

water, which dependent on the permafrost temperature exists

in a film ranging from 5 to 75 A (Anderson 1967), relates

to the physiological need of the cell to exchange ions,

metabolites, and waste products with the external milieu.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rivkina et al. (2000) investigated metabolic activity in

permafrost by measuring [14C]-acetate incorporation into

lipids of samples incubated for 550 days at temperatures

from 5 to -20°C. A sigmoidal curve was observed, and

stationary phase occurred at different levels of

radioisotope incorporation, and this depended on the

incubation temperature. Very little incorporation was

observed in samples at -15 and -20°C.

Habitats for microorganisms in glacial environments

Glaciers are generally perceived as inhospitable

environments, yet microbial habitats are known to exist in

sediment-containing melt water depressions on glacial

surfaces [i.e., cryoconite holes (Wharton et al. 1985)] and

in the water generated through friction and geothermal

heating at the ice-bedrock interface (Skidmore et al.

2000) . Direct microscopic counts of melted snow from the

South Pole ranged from 102-103 cell ml-1, and low levels of

macromolecular synthesis were detected when the snow

samples were incubated at temperatures below -12°C

(Carpenter et al. 2000) . Relatively few studies have

focused on microorganisms entrapped deep within permanent

ice sheets, although, Abyzov et al. (1993, 1998) did

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. examine the microbial species in samples from an ice core

drilled at Vostok Station, Antarctica. They cultured both

prokaryotes and eukaryotes (yeast and fungi) from samples

<10,000 years old, whereas at depths >1500 meters below the

surface, corresponding to ice >100,000 years old, fewer

microorganisms were recovered (less than 5% of cultures

contained growth) , and these were predominantly spore-

forming bacteria (Abyzov 1993). Based on these results, it

was concluded that cells within deep glacier horizons are

dormant (anabiotic) , but that species resuscitated from

these ancient samples were apparently able to commence

metabolism and grow after thousands of years of inactivity.

Much attention is currently focused on the exciting

possibility that the subglacial environments of Antarctica

may harbor microbial ecosystems under thousands of

kilometers of ice, isolated for perhaps as long as the

continent has been glaciated (20-25 million years, Naish et

al. 2001) . The discovery of more than 70 subglacial lakes

in central Antarctica during the early 1970's (Siegert et

al. 1996) went relatively unnoticed by the biological

scientific community, however, curiosity about the nature

of these environments has recently intensified as the

result of international interest in the largest such

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subglacial lake, Lake Vostok. The freshwater in Lake

Vostok originates from the overlying ice sheet which melts

into the lake at the ice-water interface in the north. Ice

accretion occurs at the base of the ice sheet in the

central and southern regions, removing water from the lake

(Kapitsa et al. 1996 ; Jouzel et al. 1999; Siegert et al.

2001) . The Vostok ice core was extended in 1998 to a

record depth of 3623 m, but due to concerns regarding

contamination, drilling stopped at -120 m above the lake-

ice interface. However, 150 m of accretion ice was

recovered (Petit et al. 1999), and the presence of

microorganisms within Lake Vostok accretion ice was

confirmed by Priscu et al. (1999) and Karl et al. (1999),

with intact cells visualized by microscopy, and seven

different 16S rDNA sequences obtained after PCR

amplification. Karl et al. (1999) also found evidence for

respiration by examining 14C 0 2 release from samples

incubated with [14C]-acetate or glucose, however, there was

very little, if any, 14C incorporation into macromolecules.

While it seems inevitable that viable microorganisms from

the overlying glacial ice and in sediment scoured from

bedrock adjacent to the lake are regularly seeded into the

lake, the question remains whether these or pre-existing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. microorganisms have established a flourishing microbial

community within Lake Vostok.

Price (2000) has proposed that aqueous veins exist

between crystal boundaries in glacial ice that could serve

as habitats for microbial activity, possessing all the

necessary conditions, including liquid water, reductants

and oxidants. Salts are insoluble in glacial ice, so only

acids deposited as aerosols become increasingly

concentrated when recrystallization occurs in deep ice.

Liquid water can also be found in lenses on grain

boundaries, around air bubbles, and at salt inclusions

(Patterson 1994) . Electron donors such as formate,

methysulfonic acid (MSA) and acetate, and the electron

acceptors SO4'2 a n d NO3' can be concentrated by several

orders of magnitude in these veins, depending on

temperature and grain size (Price 2000). In contrast to

the highly acidic vein environments in glacial ice, veins

formed in Lake Vostok accretion ice contain concentrated

salts, much like sea ice, because this ice originated as

lake water that contained dissolved ions. Based on the

minimum required input to avoid microbial carbon loss,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Price (2000) calculated that ~10-100 cells cm'3 could

survive for 400,000 years in glacial ice metabolizing the

available MSA alone.

Astrobioloqy implications of microbiological investigations of terrestrial glacial ice

The origin and margins of the biosphere

It would be difficult, if not impossible, to find a

square millimeter of the Earth's surface that is not

inhabited by life. While knowledge of deep subsurface

ecology is scanty, it is clear that the boundaries of the

biosphere extend kilometers deep into the crust (Stevens

and McKinley 1995; Chandler et al. 1998), with

extrapolations predicting that the terrestrial subsurface

contains similar amounts of microbial biomass as the

surface (Whitman et al. 1998). Environments with

severities of temperature, pressure, salinity, or pH are

labeled with the anthropocentric distinction of being

"extreme", yet microorganisms nonetheless occupy, even

thrive in these niches. The resident extremophiles have

evolved for these environmental conditions, with related

species often found only in similar ecological settings

(VanDover et al. 2001). Hence, from a microbial

perspective, the planet's zone of habitation extends from 24

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. high into the atmosphere (Sallter et al. 2001) to the inner

depths of the Earth, where temperatures rise with

increasing depth to exceed those assumed possible for known

carbon-based life.

Life arose early in Earth's history, at least 3.9

billion years ago, and it seems unlikely that life could

have evolved much before this date. The molten surface

took several hundred million years to solidify, and the

early planet was heavily bombarded with asteroids and

comets, which episodically evaporated the atmosphere and

any liquid water on the surface (Lunine 1999) . Sleep et

al. (1989) have suggested that deep marine hydrothermal

ecosystems, sustained by chemoautotrophic primary

production, could have endured these impacts that would

have sterilized the surface, perhaps as early as 4.0-4.4

billion years ago. Stable carbon isotopic ratios (813C) in

>3.8 billion year old kerogen are consistent with

photosynthetic carbonate mineralization (Schidlowski 1987),

and stromatolithic communities fossilized in 3.5 billion

year old chert are strikingly similar to species common in

extant microbial mats (Schopf 1993). Thus, when planetary

conditions were permissive, life arose in a geological

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. instant, with a 100-400 million year window for the

emergence of complex photosynthetic energy-generating

strategies and considerable species diversity.

Theories for an Earth-based origin of life are rooted

on the idea that cellular metabolism, based on

he tero trophy, photoauto trophy, or chemoauto trophy, and a

genetic system evolved spontaneously from an organic soup,

generated by photolysis, prebiotic chemical reactions and

cometary impacts. The evolution of a dissipative chemical

system into the last common ancestor may have involved

intermediary stages, with primitive life possibly

possessing an inorganic crystalline genetic system (Cairns-

Smith 1985), existing as a self-replicating catalytic RNA

(Lazcano and Miller 1996), or perhaps consisting of a

genetically promiscuous ancestral community (Woese 1998).

No current hypothesis, however, is able to provide an

adequate explanation for the unprecedented degree of

biological innovation that occurred between 3.5-3.8 billion

years ago. During life's initial period of existence DNA

replication, transcription, and translation must have

evolved, as evidenced by the conserved genetic nature of

proteins involved in these core processes within the three

domains of life (Fitz-Gibbon and House 1999; Makarova et

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al. 1999). Yet, it would, appear that these major

evolutionary challenges were solved in a geological

instant, with substantially evolved microbial species

living in complex photosynthetic communities by 3.5 billion

years ago (Schopf 1993) .

Panspermia, the concept that life originated from an

extraterrestrial source, was first conceived and termed by

the Greek scientist Anaxagoras, and revisited in the early

twentieth century by the Nobel laureate Svante Arrhenius,

who calculated that bacterial spores ejected into space

could travel on the solar wind to Alpha Centauri in 7,000

years (Margulis and Sagan 1995) . In recent years, Hoyle

and Wickramasinghe (1981) have been the most radical

proponents of panspermia theory. They argue that life is

ubiquitous throughout the universe, and that the Earth is

bombarded at regular intervals with cometary materials

containing extraterrestrial cells and genetic material, and

these shaped the evolution of life on Earth. While

theories for an extraterrestrial genesis of life have also

not supplied a tidy mechanistic explanation for spontaneous

generation, they do provide a cosmological context for such

a discussion, advocating the concept that planetary bodies

in the solar system are not biologically isolated.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Although such notions were once viewed as radical and

generally dismissed by the scientific community, recent

awareness of the tenacity of life and extremophile

diversity now place them in the realm of possibility.

Spores surrounded by radiation absorbing material have been

argued to have the longevity and durability needed to

survive transfer through interstellar space (Weber and

Greenberg 1985), and ejecta from large impacts of the Earth

containing viable microorganisms would provide sufficient

shielding from heat and lethal cosmic radiation to

facilitate transfer to Mars, and vice versa (Melosh 1988;

Weiss et al. 2000). Remarkably, Streptococcus mitis

isolates were revived from a camera on Surveyor 3 that was

recovered by Apollo 12 astronauts after it had spent 2.6

years on the moon (Noever 1998) .

The measures taken by NASA to control the forward

contamination of Mars during the Viking missions seemed

appropriate at the time. However, 25 years of deliberation

and knowledge have made it necessary to reevaluate

appropriate protocols for the protection of planets and

satellites from contamination by Earth microorganisms,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. prompting NASA to create a Planetary Protection Advisory

Committee to assess procedures used in solar system

exploration and sample return missions (Rummel 2001).

Prospects for frozen lifestyles in the solar system

A resurgence of interest in the field of Exobiology,

now often referred to as Astrobiology, is centered on the

notion that geological and physical settings in the

universe similar to those on Earth may also harbor life.

Thus identifying, characterizing, and understanding

terrestrial habitats that could be similar to

extraterrestrial habitats provides an experimentally

tractable system to evaluate the likelihood of

microorganisms surviving in extraterrestrial environments.

These may well offer important clues to lifestyles that

might be encountered on Mars or several of the jovian

moons, and perhaps even within comets or asteroids

containing liquid brine fluid inclusions (Zolensky et al.

1999).

The poles of Mars are covered by ice caps with some

features similar to those of terrestrial ice sheets (Budd

et al. 1986). Although temperatures exceed the frost point

and water ice is unstable at lower latitudes, ice may also

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. still be present in the non-polar regions kilometers below

the surface (Squyres and Carr 1986) . Dust particles from

the surface, elevated into the atmosphere by wind, serve as

condensation nuclei for carbon dioxide and water which is

precipitated, perhaps seasonally, in the martian polar

regions (Clifford et al. 2000). The presence of alternating

clean and dusty layers within these polar ices could

reflect changes in the levels of atmospheric dust, or may

result from the sublimation of frozen volatiles during

periods of high obliquity. Although organic molecules are

likely to be destroyed by the high levels of UV irradiation

and peroxides photochemically-generated in the martian

soil, biochemical traces of life or even viable

microorganisms may well be protected from such destruction

if deposited within polar perennial ice or far below the

planet's surface. During high obliquity, increases in the

temperature and atmospheric pressure at the northern pole

of Mars (McKay and Stoker 1989) could result in the

discharge of liquid water that might create environments

with ecological niches similar to those inhabited by

microorganisms in terrestrial polar and glacial regions.

Periodic effluxes of hydrothermal heat to the surface might

also move microorganisms from the martian subterranean,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. where conditions may be more favorable for extant life

(McKay 2001). The annual partial melting of the ice caps

might then provide conditions compatible with active life

or at least provide water in which these microorganisms may

be preserved by subsequent freezing (McKay and Stoker 1989;

Clifford et al. 2000). The microfossils and chemical

signatures of potential biological origin that were

recently discovered in Alan Hills meteorite ALH84001,

reinvigorated the debate over the possibility of life on

Mars (McKay et al. 1996) . However, such circumstantial

evidence needs confirmation by explorative missions to the

martian surface.

In January 2000, the Galileo spacecraft measured

changes in Europa's magnetic field, and provided the most

convincing evidence to date for the presence of a

subsurface liquid ocean (Kivelson et al. 2000). Geothermal

heating and the tidal forces generated by orbiting Jupiter

are thought to maintain a 50-100 km deep liquid ocean on

Europa with perhaps twice the volume of the Earth's ocean

(Chyba and Phillips 2001), but beneath an ice shell at

least 3-4 km thick (Turtle and Pierazzo 2001). Cold

temperatures (<128 K; Orton et al. 1996) combined with

intense levels of radiation would appear to preclude the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. existence of life on the surface, and the zone of

habitability, where liquid water is stable, may be present

only kilometers below the surface, where sunlight is unable

to penetrate (Chyba and Hand 2001). Europa's surface

appears strikingly similar to terrestrial polar ice floes,

suggesting that the outer shell of ice is periodically

exchanged with the under lying ocean. The ridges in the

crust and the apparent rafting of dislocated pieces implies

that subterranean liquid water flows up through stress-

induced tidal cracks, which may then offer provisional

habitats at shallow depth for photosynthesis or other forms

of metabolism (Gaidos and Nimmo 2000; Greenberg et al.

2000) . However, if one exists, the europan ecology must

differ significantly from any presently known on Earth, all

of which depend on photosynthesis for primary production.

Even communities at marine hydrothermal vents, which have

been widely described as self sufficient ecosystems based

solely on the metabolism of reduced geothermal energy,

ultimately require oxidants such as SO42", CO 2 , a n d O 2 from

the surface to function. Currently, the only possible

example of a photosynthetically-independent ecological

community are the proposed subsurface lithoautotrophic

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. microbial ecosystems [SLiMEs]), with the primary producers

obtaining energy from geothermally produced hydrogen

(Stevens and McKinley 1995; Chapelle et al. 2002).

As discussed, photosynthetic organisms are usually

required to fix CO 2 with reducing power generated by light-

driven reactions, and thus subsequently provide assimilable

carbon for heterotrophic consumption. The oxidant

byproduct, however, is equally crucial for maintaining the

environmental redox gradients exploited by diverse

metabolic lifestyles. Gaidos et al. (1999) argue that

without a source of oxidants, Europa's subsurface ocean

would be destined to reach chemical equilibrium, making

biologically-dependent redox reactions thermodynamically

impossible. However, the surface is continually bombarded

with high-energy particles, producing molecular oxygen and

peroxides, as well as formaldehyde and other organic carbon

sources (Chyba 2000; Chyba and Hand 2001) , and thus it is

conceivable that europan microbial life might subsist

without employing photosynthetic or chemoautotrophic

lifestyles. In this scenario, mixing between the crust and

subsurface need be the only mechanism required to provide

organics and oxygen at levels sufficient to support life

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Chyba 2000) . Tidal heat generation and electrolysis could

also provide sources of energy that could be coupled to

bioenergetic redox reactions (Greenberg et al. 2000).

Development of technology relevant to life detection in extraterrestrial materials

NASA missions are already scheduled to examine Mars

and Europa for liquid water and prospective life, and the

first sample return missions from Mars could launch as

early as 2011, returning to Earth in 2014. A major concern

is the forward contamination of these solar system bodies

with microbes or molecules transported on spacecraft from

Earth, which could then be misinterpreted as evidence for

extraterrestrial life.

Contamination of the Earth with non-terrestrial life

is an equally important consideration, and NASA is now

faced with the challenge of defining appropriate

containment measures (Rummel 2001), and constructing a

facility now designated the Mars Sample Receiving Facility.

To facilitate geochemical and microbiological analysis on

unaltered samples, the curation team has recommended

maintaining samples at below -30°C (Meyer 2001) . The

challenge of identifying appropriate sampling sites and

developing protocols that maintain containment, 34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cleanliness, and cold will benefit from the experience

gained by sampling and studying frozen terrestrial

environments, analogs for Mars and Europa.

Objectives of this study

The Byrd Polar Research Center (BPRC) at Ohio State

collects ice cores from the polar regions, and from

mountain glaciers at low latitude. Stored within this

archive are glacial ice cores ranging in age from <50 to

>750,000 years old. The primary objective of this study

was to sample, isolate, and identify microorganisms

preserved in ice cores of defined ages, and to determine

the longevity of species entrapped in ice for hundreds of

thousands of years. The availability of physical and

chemical data on each core provides the opportunity to

correlate the effect of climatic conditions on

microbiological deposition and survival. As a second

project, the possibility that glacial and subglacial

environments represent microbial habitats was explored.

While establishing the lengths of microbial survival in ice

was of primary biological interest, it seems also possible

that there might be metabolism within these frozen

environments. The data obtained are also relevant to

discussions about the survival of life through periods when

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the Earth was completely covered in ice for millions of

years, and also might well offer valuable information in

considering exobiological lifestyles.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

PROCEDURES USED TO PREVENT CONTAMINATION DURING SAMPLING AND ANALYSIS

Introduction

The major concern in any study employing culture-based

analyses and sensitive molecular amplification approaches

to examine ice cores is obtaining samples without

contamination by extraneous microorganisms and/or nucleic

acids. Care also must be exercised in the subsequent

preparation and handling of media, glassware, and reagents,

with appropriate controls to monitor for contamination at

each experimental step. Although protocols to eliminate

microbial contamination can be developed or implemented,

proving that a positive result is not an artifact is very

difficult. To address this issue, and the inherent concern

it raises for this investigation, techniques and

decontamination procedures developed and employed are

described here in detail, together with the mechanisms used

to verify the authenticity of results obtained.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Materials and Method*

Bacterial strains and control ice cores

Escherichia coli. Bacillus subtilis and Serrratia

marcescens (OSU reference numbers 455, 848, and 234,

respectively) were cultured on tryptic-soy (Difco) broth or

1.5% solidified agar and used as indicator strains and in

the construction of control ice cores. Cells were

suspended in deionized water at concentrations from 102-109

cfu ml'1, and the suspensions were frozen at -30°C in a 100

mm diameter cylinder. Sterile cores of deionized water

were constructed in the same manner. Expansion of the ice

formed within the cylinder made extraction of the cores

difficult. The exterior surface was therefore melted by

running tap water over the cylinder and the extracted core

was then refrozen. Stress fractures caused by freezing

from the outside to the inside limited the length of core

recovered by this method, but workable sample cores of 10-

20 cm were obtained.

Sterilization and removal of nucleic acids from reagents and experimental materials

Liquid reagents and solid materials were autoclaved

for 30 min. at 120°C (20 psi) . Glassware, pipets, test

tubes, and all other solid items were then dried for at 38

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. least 72 h. at 55°C, and then exposed to ethylene oxide in

an Anprolene AN-74 sterilizer (H.W. Andersen Products Inc.,

Haw River, NC) for 12 h. To remove nucleic acids from

water and reagent solutions used in PCR amplifications,

they were filtered through Microcon YM-100 (500 |il

capacity), Biomax-100 Centricon Plus-20 (20 ml capacity) ,

or Biomax-100 Centricon Plus-80 (80 ml capacity)

centrifugation devices following the manufacturers

directions (Millipore Corp. Inc., Cat. no. 42412, UFC2 LGC

02, and UFC5, LGC 02, respectively).

Bleach-rinsed gloves were worn throughout all

procedures. All cultural manipulations and the preparation

of PCR amplification material were carried out in a BioGard

laminar flow hood (Baker Company, Sanford, ME) with a

germicidal UV-C lamp. Before use, all interior surfaces of

the hood were wiped with a solution of 0.24% (w/v) sodium

hypochlorite. Reagents, pipets, and tubes were placed

within the cleaned hood and exposed to UV irradiation for

at least 25 min. before use. Disposable pipet tips with

hydrophobic barriers were used to decrease the likelihood

of aerosol generation and cross-contamination.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rinse method for the decontamination of ice cores

The outermost layer was removed from ice cores using a

dust-free bandsaw, in a -5°C cold room, to expose previously

unhandled ice. The ice was then equilibrated to the

sampling temperature (-5°C) overnight, to reduce the

likelihood of fracturing during washing procedures and

melting. Sections of these ice cores were then held with

sterile forceps or scissor tongs within the laminar flow

hood and washed with -200 ml of -5°C 0.2 nm filtered 95%

ethanol. This disinfected and dissolved away the outermost

ice layer. After a 2 cm annulus has been removed, a fresh

pair of forceps was then used to grasp a cleaned section of

the ice core, and the sample was rinsed with ~100 ml of

chilled, double autoclaved deionized water. To evaluate

the effectiveness of the washing procedures, the wash

materials were collected and assayed for contaminants or

for the removal of bacteria deliberately placed on the ice

exterior in control experiments. The cleaned ice was

placed in a sterile vessel and allowed to melt slowly at

4°C. This took at least 16 h for most samples.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The automated ice core sampler

Heated sampling heads (Fig 2.1A) were constructed with

different diameters (10, 20, and 30 mm) to sample core

sections of varying sizes. The sample head was housed in a

circular holder that guides the ice core position. The

prototype holder was constructed out Teflon (Fig 2. IB), but

the experimental version was made using iodized aluminum

(Fig 2.2A). Prior to sterilization, the device is fitted

to accommodate the size and shape of the ice core to be

sampled. Movable dividers (Fig 2.2B) were fixed to

position the core so that the sampling head remained inside

the core throughout the melting process (Fig 2.2C). The

sample unit accommodates 1/2 to 1/8 core sections >500 cm

in length. The components of the unit were autoclaved for

30 min. at 120°C (20 psi) , dried for at least 72 h. at 55°C,

and then exposed to ethylene oxide treatment [Anprolene AN-

74 sterilizer (H.W. Andersen Products Inc., Haw River, NC)]

for 12 h. The components were assembled inside a laminar

flow hood housed within a -5° C walk-in freezer (Fig 2.2A).

The ice cores are archived at -30°C, but were equilibrated

to -5°C before sampling. Using a dust-free bandsaw, used

only for this purpose, a cross-sectional cut removed a few

millimeters of ice from the end of an ice core exposing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2.1 Construction of the sampling head and prototype ice core sampler.

A. Picture of the sampling head. An external jacket allows the circulation of water that heats the head, which then melts through an ice core under the influence of gravity. The melt water generated is collected through a hole in the center of the melting head and pumped into an external sterile container.

B. The housing for the prototype sampler was initially constructed of Teflon. Here it is shown assembled inside a laminar flow hood housed in a -5°C walk-in freezer. In the device is a control core containing E.coli, used to evaluate the recovery of viable bacteria and the sterility of the melting procedure.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heated water circulates through sampling head

melt water collected externally

Figure 2.1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2.2 Final design of the automated sampler.

A. The complete unit, constructed out of iodized aluminum, with an ice core positioned perpendicularly in the device. All components of the system were sterilized and then assembled inside a laminar flow hood housed in a -5°C walk- in freezer.

B. Moveable separation flanges facilitated melting 1/2 to 1/8 of a core section, permitting duplicate samples to be collected from parallel regions of the same core.

C. The sampling head after movement through a core, illustrating how the melting head removes a cylindrical section from the core's interior.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with Figure 2.2 (Con't) 46

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. previously unhandled ice and provide a uniform, flat

surface for disinfection. This surface was soaked in 0.2 Jim

filtered 95% ethanol for 2 min. to sterilize and dissolve

away a portion of the exterior. In control experiments,

indicator strains were intentionally applied to this

surface to assay the effectiveness of removal of microbial

contaminants by the disinfection procedure. Exposing the

ice core surface to 0.24% sodium hypochlorite (bleach) and

short wavelength ultraviolet (254 nm) light were also

evaluated as sterilization procedures. Prior to sampling,

the sterilized surface of the ice core was swabbed, the

material collected placed in 1 ml of sterile deionized

water, and inoculated into culture media, and included as a

control in molecular amplification procedures. The ice

core was placed in the sampling device such that the

disinfected portion of the core contacted the sampling

head. The melt water generated was collected through a

hole in the center of the melting head and pumped directly

into an external sterile container (Fig 2.1A).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Characterization of contaminating microbes and 16S rDNAs

Nucleic acids amplified and extracted from

microorganisms were analyzed by determining DNA-encoding

16S rRNA gene sequences. A portion of an isolated colony

was resuspended, and the cells lysed directly in PCR

amplification mixtures according to Zeng and Kreitman

(1996) . The primers used for these amplifications are

named by the position of the homologous sequences in the

Escherichia coli 16S rRNA gene (Fig 2.3). Combinations of

the forward primers 2IF (complementary to most archaeal 16S

rDNA) , 27F (bacterial 16S rDNA), 348F (archaeal 16S rDNA)

or 515F (universal 16S rDNA primer) were used with the

reverse primers 1392R (universal 16S rDNA primer) , or 1492R

and 1525R (bacterial and archaeal 16S rDNA) (Lane 1991;

Reysenbach and Pace 1995) .

Tag polymerases from GibcoBRL (catalog # 18038-042),

Sigma (ReadyMix™ Tag cat. no. P4475) and Eppendorf

(MasterTaqF* cat. no. 0032 002.650) were used for

amplification of DNA from cultured cells, and the

chemically-modified, thermally-activated AmpliTaqf’* Gold DNA

polymerase or LD (low DNA) AmpliTaqF* Gold DNA polymerase

(Perkin-Elmer [PE] Biosystems, Foster City, CA [cat. no.

N808-0240 and N808-0107, respectively]) were employed for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2.3 Location of primers used to amplify and sequence 16S rDNA, referenced to the sense strand of E.coli's 16S rDNA sequence.

The positions of 16S rDNA oligonucleotides used as PCR primers are shown schematically, with reference numbers indicting the position of the 3' nucleotide within the 16S rDNA sequence of E.coli. Primer 341F had a 40 bp repeating sequence of GC (GC clamp), and was to amplify DNA that would be subsequently investigated by denaturing gradient (DGGE) or temporal temperature (TTGE) gel electrophoresis (Muyzner et al. 1993). Primers are designated by fill pattern their complementarity to bacterial, eukaryotic, and/or archaeal 16S rDNA sequences.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E. coll 16S rRNA gene + + SOO bp 1000 bp 1500 bp

27F.)

68F 3I8F 1392R

515F

Most bacteria llHI Most bacteria and archaea

□ Most bacteria, eukaryotes, and archaea ■A Most archaea

Figure 2.3 50

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amplification of DNA recovered directly from melt water

samples. Depending on the Tag polymerase utilized,

reaction mixtures (10X buffer supplied by the manufacturer,

2-4 mM MgCl 2 , and 5 pmol of each primer) were subjected to

30 or 43 (AmpliTag [PE Biosystems]) cycles of amplification

by denaturation for 1 min at 94°C, annealing for 1 min at

50°C, and extention for 1 min at 60°C (AmpliTaqF*1 [PE

Biosystems]) or 72°C. Before use, AmpliTaqF* requires heat

activation which was achieved by incubation at 95°C for 9

min. Samples of each PCR product was evaluated by agarose

gel (0.8-2.5%) electrophoresis followed by staining with

ethidium bromide.

Amplicons of the expected length from colony isolates

were sequenced by using primers that hybridized within the

amplified DNA fragment. Specifically, primers 515F, 534R,

907R, 1392R and 1492R ([Fig 2.3]; Lane 1991; Medlin et al.

1988) were used to prime ABI Prism BigDye*" terminator cycle

sequencing reactions (catalog# 4303149). The products were

then analyzed on an ABI Prism 3700 DNA analyzer. Some

amplicons were cloned using the pGEM-Teasy (Promega Corp.,

Madison, WI) vector system and sequenced using primers that

annealed the flanking T7 and SP6 promoter sequences. The

sequences obtained were managed using the GeneTool"*

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (catalog# GT10-9808-0333-3452, BioTools Inc., Edmonton, AB)

software package, and were compared and aligned with all

sequences available in the Ribosomal Database Project

(Maidak et al. 2001) and GenBank (Benson et al. 2000) .

R e s u l t s

The effectiveness of removing S. marcescens cells

intentionally contaminated onto ice surfaces in visible

quantities, by rinsing samples with chilled sterile

deionized water, was assessed in preliminary experiments by

collecting the resulting wash-off, and spread plating 200 jil

of the material on agar-solidifed media to assay for the

easily identified indicator organism. A rinse using 300 ml

of water for an oblong ice sample of ~50g was sufficient to

remove the S. marcescens contamination from sterile test

cores, with no colonies of S. marcescens observed in the

final 100 ml used to rinse the ice. Using this method, -20

ml of clean sample was obtained from the original 50g

s a m p l e .

A disadvantage of this procedure was that the

temperature difference between chilled water and ice

frequently resulted in core fractures that often formed so

rapidly, that the entire sample was lost. The ice core

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disintegrated into many fragments that were impossible to

secure with forceps. Such fractures also provide an entry

point for surface contaminants to contaminate deeper into

the core, a n d S. marcescens cells were recovered from

cleaned ice-core samples that cracked during the initial

wash. To overcome this problem, 95% ethanol equilibrated

to the ice temperature (-5°C) was subsequently used to wash

and disinfect cores, followed by a 100 ml rinse with

double-autoclaved distilled water, to dilute the ethanol to

nontoxic concentrations. Ethanol is a potent disinfectant,

but it would not be expected to destroy endospores, but

contamination by endosporulating cultures of B. subtilis

was nevertheless removed by this washing procedure when

such cells were intentionally swabbed onto ice core

surfaces in control experiments.

An alternate method and instrumentation was developed

to melt and collect water only from the interior of an ice

core. Disinfection of the cut surface of an ice core by

soaking in 95% ethanol, sodium hypochlorite, and exposure

to germicidal ultraviolet (UV) light irradiation were

evaluated. Melt water from E. coli control cores treated

with sodium hypochlorite contained 105 cfu ml-1, which was 20

fold less than untreated samples. Although no viable

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cultures were generated from material collected from the

ice core exterior, toxic concentrations of bleach

apparently persisted within collected samples. UV

irradiation of the surface was considered, but given that

ice is transparent to irradiance at wavelengths >190 nm

(Hobbs 1974) , the exposure required to sterilize the core

surface would likely compromise the recovery of sublethally

injured species immured within the ice, and also destroy

preserved DNA. Using 95% ethanol washes had the advantage

that the ethanol remained liquid at -20°C, the ice cores did

not fracture even when soaked for several minutes to

dissolve away potentially contaminated ice, and the ethanol

was easily diluted subsequently to non-toxic

concentrations. When the ethanol disinfection strategy was

assayed to examine the removal of S. marcescens cells

swabbed on the ice and inoculated onto the saw blade used

to cut the ice core, S. marcescens was not then isolated

from the melt water generated and collected by the

automated sampler.

To investigate if there was DNA present on the cut

surface after the ethanol washing, the washed ice surfaces

were swabbed, the material collected was resuspended in 1

ml of sterile water, and used as template DNA in PCR

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amplifications using universal and bacterial 16S rDNA

primers. Amplicons were not generated from material

collected from cleaned ice core surfaces. This was also

the case when sensitive two-stage nested amplification

strategies were utilized (e.g., one round of amplification

with primers 27F and 1525R, followed by a second round of

amplification using primers 515F and 1392R or 1492R) . The

amount of DNA that remained on the cut, ethanol treated ice

core surface must therefore have been below the level of

detection using these primers.

D i s c u s s i o n

Once the fracturing problem had been solved by the use

of ethanol for washing, the rinse method first used in the

initial stages of this project was fully effective at

removing cultivable contaminants from the cut ice surface.

Holding ice samples >100g with forceps proved difficult,

but by using scissor tongs, common food handling utensils,

such ice fragments could be held firmly after a series of

sharp ridges were filed on the grasping surface. These

details are important when one considers the precious

nature and one-time availability of many of the ice cores

examined in this study.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since PCR-based analyses of 16S rDNA was planned, it

was not clear if the ethanol washing approach would be

adequate to remove free nucleic acids, nor was there an

easy way to reliably assay for such contamination. Bleach

destroys DNA, however, rinsing with an aqueous sodium

hypochlorite solution posed the risk of ice fracturing, and

control experiments demonstrated that toxic levels of

bleach could persist within cleaned ice samples.

To develop a more reliable sampling strategy, the

automated sampler (Fig 2.1 and 2.2) was designed,

constructed and tested. The goal was that a heated melting

device with a reservoir to collect the generated melt water

will melt longitudinally through an ice core interior

without contacting the potentially contaminated outer

surface (Fig 2.2C). The central feature is the heated

funnel-shaped sampling head (Fig 2.1A) that collects the

melt water, from which it is pumped to an external

collection system. The advantages of this technology were:

(i) movable dividers (Fig 2.2B) facilitated repeat sampling

through of 1/4 to 1/8 segments of the same core; (ii) by

collecting the melt water as sequential fractions, a high

resolution sampling of a particular deposition period was

possible; (iii) melting only the interior avoided

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. contaminants present on the ice core exterior; (iv) only

the bottom of a freshly cut ice core needed to be

disinfected, reducing the risk of sample loss by

fracturing; and (v) the cut surface was easily swabbed and

assayed for the presence of contaminants. During the

course of this study, only once was a culture obtained from

an ethanol-cleaned ice core surface, and this isolate had a

16S rDNAs sequence consistent with classification as an

endospore-forming Bacillus subtilis species. If the same

species had subsequently been isolated from the interior of

the core it obviously would have been suspected as a

potentially introduced contaminant. But, in the single

case of an apparent breach in sterilization procedures,

this was not the case. The sample in question was

collected from a 200 year old ice core that originated from

the Guliya ice cap. Although there was always a

possibility that microbial and/or nucleic acid

contamination would elude the decontamination used, by

employing quality control measures, and drawing conclusions

from the results obtained from more than one sample from an

ice core, this concern was addressed.

The most difficult aspect of using PCR-based

procedures to amplify 16S rDNAs present in low biomass

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. samples is that the Taq polymerase, a recombinant enzyme

isolated from a genetically engineered E. coli, m ay be

contaminated with small, but significant quantities of

chromosomal DNA. All other components of the PCR (buffer,

M g C l 2 , primers, and water) were passed through Centricon and

Microcon filters (Millipore), which removed double-stranded

DNA molecules larger than 125 base pairs from the filtrate

solution.

The Taq polymerase preparations purchased from Sigma,

Gibco BRL, and Eppendorf possessed no detectable

chromosomal DNA contamination when assayed using universal

and bacterial-specific 16S/18S primer combinations in 30

cycles of amplification in template-free reaction mixtures.

Control experiments revealed that 10-100 pg of

Methanobacterium thermoautotrophicum chromosomal DNA,

equivalent to the amount of DNA present in 104-105 cells,

was required to yield a PCR product under these

amplification conditions. However, with the low cell

numbers anticipated in glacial samples, the use of even

more sensitive PCR strategies, based on two rounds of

nested amplification was necessary, and spurious amplicons

were often generated in controls, namely in template-free

reaction mixtures using these PCR procedures. Several of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. these 16S rDNA amplicons were cloned and sequenced, and the

sequences obtained had >99% identity to E. coli a n d a y-

proteobacterial Pseudomonas species reported as a

contaminant of PCR reagents by Cisar et al. (2000) .

Although the source was not fully established, the results

obtained suggest that the primary sources of contaminating

nucleic acids was the E. coli used to produce Taq

polymerase, and an indigenous microorganism that seems to

colonize laboratory reagents and water supplies in many

laboratories (Cisar et al. 2000; Tanner et al. 1998), based

on efforts to catalog the most frequently encountered

contaminants in negative control libraries. Attempts to

treat the Taq polymerase with deoxyribonuclease I (DNase;

Gibco BRL) either compromised Taq activity or residual

DNase activity remained after heat treatment, and

amplification products were not obtained. Finally, a

genetically modified, stringently purified Taq polymerase

(LD [low DNA] A m p l i Tacf* Gold DNA polymerase, Perkin-Elmer

Biosystems) was used that the manufacturer claimed

contained <10 copies of bacterial 16S rDNA per 2.5 unit of

enzyme. Consistent results were obtained using this enzyme

in nested amplifications protocols that indicated that the

DNA contamination was below the level of detection.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It is over 70 years from the time of the first reports

that claimed the recovery of viable microorganisms from

ancient geological specimens (Lipman 1928; Lieske 1932).

Since then, microbiological investigations have resulted in

similar revivals from ancient rock, salts, soils and

sediments (Kennedy et al. 1994) . Some extraordinary claims

have received much attention, but have also been viewed

with skepticism (Priest and Beckenback 1995; Hazen and

Roedder 2001). Contamination from modern sources could not

always be definitively ruled out, and the ages of presumed

ancient specimens has been a source of debate (Hazen and

Roedder 2001). For example, despite the stringent

precautions and controls used by Cano and Boruki (1995) to

examine the intestinal contents of a bee trapped in

Dominican amber, their recovery of a unique Bacillus

sphaericus species has been questioned (Priest and

Beckenback 1995). Although the major criticisms focused on

the subsequent sequence analyses, this aggressive critique

underlies the issue central to all studies of this nature.

It is difficult to prove the authenticity of a single

result, even if the methods used have a very low

probability of introducing contamination. Therefore, the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. development and assays of the aseptic sampling procedures

used, and the extensive controls used to detect

contamination were of the highest priority for this study.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3

ISOLATION AND CHARACTERIZATION OF BACTERIA AND 16S rDNA SEQUENCES AMPLIFED FROM GLACIAL ICE CORES

Introduction

Studies of ice cores have established past climate

change and geological events, but rarely have these results

been correlated with the insects, plant fragments, seeds,

pollen grains, fungal spores, and bacteria that are also

present, and few attempts have been made to determine the

diversity and longevity of viable species entombed in

glacial ice. Several studies have focused on recovering

viable microorganisms (Abyzov 1993, et al. 1998; Dancer et

al. 1997) and detecting virus particles (Castello et al.

1999) and eukaryotic species (Willerslev et al. 1999)

entrapped in polar ice. However, a thorough evaluation of

the microorganisms preserved in glacial ices from different

geographical locations has not been undertaken, nor have

such isolates been characterized phylogenetically, examined

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for features increasing survival and persistence while

frozen, or related to the local geography and environmental

conditions.

Fortunately, ice cores collected from a global

distribution of polar and non-polar glaciers (Fig 3.1) are

archived in the Byrd Polar Research Center (BPRC) cold

facility at The Ohio State University. These ice cores

have been subjected to extensive physical and chemical

analysis, and provide the opportunity to investigate

microorganisms preserved in glacial ices formed at defined

dates, under known climatic conditions, and at

geographically very different locations. Here, the results

obtained from ice cores originating from China, Bolivia,

Greenland and Antarctica that range in age from 50 to

>500,000 years old are reported.

Materials and Methods

Sample handling

The melt water, which ranged from 50 to >3 00 ml,

maintained at ~5°C during sampling, was collected during a

period of 1-2 h (see Chapter 2). Less than 2 h after

melting, samples were placed in a BioGard laminar flow hood

(Baker Coirpany) maintained at room temperature (~22°C) for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Ice Cap Oasuopu

Puruogangri ' Ice Cap Dunde Ice Cap m Guliya Ice Cap Gregoriev Ice Cap

HP Franz Joaef Land South Pole Ice CoreSites Ice e e Camp Century

Sajama Huascaran 150 W 150 \ \ \ Quelccaya Ice Cap Figure Figure 3.1 Global locations of sampling sites and ice cores available for study at the Byrd Polar Research Center (BPRC). would most For each likely contribute sampling site, the majority the nearest of airborneecosystems particles that are very different. a\

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. processing. The particulates in a large portion of the

melt water were concentrated by filtering through 0.2(Am

Isopore filters (Millipore Corp., cat. no. GTTP04700) , then

resuspended by vortexing in 5 ml of phosphate-buffered

saline. In samples analyzed further by PCR amplification

procedures, the 0.2 Jim filtrate was stored in a final

concentration 1 mM EDTA and 10 mM Tris.

Media and culture conditions

Aliquots of the 0.2 nm filter-concentrated samples were

spread on the surface of agar-solidified growth media that

specifically included tryptose blood, Actinomycetes

isolation, full strength and 1% nutrient broth, full

strength and 1% tryptic soy broth (all supplied by Difco) ,

R2 (Reasoner and Geldreich 1985) , and M9 glucose minimal

salts (Sambrook et al. 1989). Bovine liver catalase (1 KU;

Sigma) was filtered (0.22 Jim) and added to cultures, as a

known enhancer in the recovery of sublethally injured

bacteria (Marthi et al. 1991). Duplicate spread plates

were incubated aerobically at 4° or 10° and 22°C. For

liquid cultures, a 1 ml aliquot of the unfiltered melt

water was inoculated into full strength and 1% nutrient

broth , full strength and 1% tryptic soy broth (Difco) , R2 65

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Reasoner and Geldreich 1985), and M9 glucose minimal salts

(Sambrook et al. 1989) . Liquid enrichments were incubated

without shaking) at 4° and 22°C.

In some instances, medium designed to grow acetogens,

methanogens, and sulfate-reducers were also inoculated

directly with a sample of the melt water. These media are

based on a bicarbonate-buffered salt solution that

contained (per liter) 0.4 g KH2P04, 0.53 g Na2HP04/ 0.3 g

NH4C1, 0.3 g NaCl, 0.1 g MgCl2-6H20, 0.11 g CaC12, 1 ml

trace element solution, 1 ml vitamin solution, 0.5 mg

resazurin, and 4 g NaHCC>3 , 0.25 g Na2S*7H20. The vitamin

and trace element solutions were as previously described

(Stams et al. 1992). In addition, 0.16 mg of 2-mercapto

ethane sulfonic acid (MESA) and 0.27 jig of NiCl2 or 2.8 g of

Na2S04 and 60 jig of FeS04 were included to facilitate the

growth of methanogens and sulphate-reducing bacteria,

respectively. These cultures were done in serum bottles

sealed with rubber stoppers, with a 20 psi gas phase of

N2/C02 or H2/C02 (4:1). Carbon sources (methanol, lactate,

acetate, trimethyl amine syringic acid, and fructose) were

added from 0.5 M stock solutions to final concentrations of

10 mM. Ammonia and nitrate salts media was also used to

enrich for methano- and me thy lo trophic species, and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. prepared according to Patt et al. (1974), with the cultures

incubated in sealed serum bottles with a 1:4 methane-air

gas phase. Duplicate enrichment cultures were incubated at

4° or 10° and 22°C.

16S rDNA amplification from bacterial isolates

A portion of an isolated colony was suspended and

lysed directly in PCR amplification mixtures according to

Zeng and Kreitman (1996). Primer 27F and 1525R (Lane 1991;

Fig 1.3) were used to amplify 16S rRNA gene fragments with

Tag polymerase from GibcoBRL (cat. no. 18038-042), Sigma

(ReadyMix™ Tag cat. no. P4475) or from Eppendorf

(MasterTaqF** cat. no. 0032 002.650) . Reaction mixtures

contained the buffer supplied by the manufacturer, 2 mM

M g C l 2 , 0.05% Nonidet P-40 (Sigma, cat. no. N-6507) and 5

pmol of each primer, and were subjected to 30 cycles of

amplification by denaturing for 1 min at 94°C, annealing for

1 m i n at 50°C, and extending for 1 m i n at 72°C. Samples

from each PCR reaction were evaluated by electrophoresis

through 0.8% agarose gels and subsequently stained with

ethidium bromide.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Direct amplification of 16S rDNA from melt water

Melt water volume (160 m) 1 was filtered through a 0.2nm

Isopore filter (Millipore Corp., cat. no. GTTP04700), and

DNA extracted from the collected particulates by using the

modified protocol of More et al. (1994), with hot detergent

treatment as described by Kuske et al. (1998) . Filters

were placed in a screw cap tubes containing 0.1 mm

zirconium beads (Biospec Products, cat. no. 11079101Z) and

a buffered lysis solution consisting of 4% SDS, 40 mM NaCl,

200 mM Tris (pH 8.0), and 20 mM EDTA. During a 20 min.

incubation at 70°C, samples were vortexed every 5 min for 10

s . Tubes were then placed in a Biospec Mini-Bead Beater

for 5 min. at 5000 rpm, followed by centrifugation for 3

min. at 12,000xg to remove debris. The supernatant was

recovered, and residual DNA still associated with the

pellet was extracted by adding 500 ^il of TE (10 mM Tris and

1 mM EDTA) , vortexing, and centrifuging the mixture as

before. The SDS was precipitated by incubating samples at

4°C for 5 min in the presence of 2 M ammonium acetate and

removed after centrifugation at 12,000xg for 3 min.

Double-stranded DNA molecules >125 nucleotides present in

the extract were then concentrated and exchanged into TE

using Microcon YM-100 centrifugal filters (Millipore Corp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Inc., cat. no. 42412), in accordance with the manufacturers

specifications. Free nucleic acids in the cell-free 0.2nm

filtrate were also concentrated using Biomax-100 Centricon

Plus-80 (80 ml capacity) centrifugation devices (Millipore

Corp. Inc., cat. no. UFC5, LGC 02) . „

The thermally-activated LD (low DNA) AmpliTaqF** Gold

DNA polymerase (Perkin-Elmer [PE] Biosystems, cat. no.

N808-0107) was employed for PCR amplification of DNA

recovered directly from melt water samples. Reaction

mixtures con s i s t e d of 2. 5U of AmpliTaqF**, the b u f f e r

supplied by the manufacturer, 4 mM MgCl 2 , 5 pmol of each

primer (27F or 21F and 1392R, 1492R or 1525R) , and 2 jil of

template. AmpliTaqf** requires heat activation, achieved by

incubating the enzyme at 95°C for 9 min. , which was followed

by 43 cycles of amplification by denaturing for 1 min at

94°C, annea ling for 1 m i n at 50°C, a n d e x t e n d i n g for 1 min

at 60°C. An aliquot (2 til) of the resulting product was

used as the template in a second PCR, with the forward

primer 515F (universal) or 348F (archaeal) combined with

either reverse primer 1392R (archaeal and bacterial) or the

universal reverse primer 1492R (Lane 1991; Reysenbach and

Pace 1995; Fig 2.3) . Samples from each PCR reaction were

evaluated by electrophoresis through 0.8% agarose gels and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. staining with ethidium bromide. Individual DNA molecules

were cloned from the resulting -900 bp populations into

pGEM-Teasy (Promega Corp., Madison, WI) , and the cloned

inserts, amplified using primers for the flanking T7 and

SP6 promoter sequences, were screened for restriction

fragment length polymorphisms (RFLP) using MspI and HinPI

(New England Biolabs).

Sequence and phylogenetic analysis

Nucleotide sequences were determined by using ABI

Prism BigDye1* terminator cycle sequencing (cat. no.

4303149) with an ABI Prism 3700 DNA analyzer. For cultured

isolates, both DNA strands of the -1500 bp amplicon (27F-

1525R) were sequenced by using the internally nested

primers 68F, 515F, 534R, 907R, 1392R and 1492R (Medlin et

al. 1988; Lane 1991; Fig 1.3). Both DNA strands of inserts

cloned into pGEM-Teasy were sequenced using primers that

annealed to the flanking T7 and SP6 promoter sequences.

Multiple reads from sequences were edited and compiled

using the GeneTool™ software package (BioTools Inc., cat.

no. GT10-9808-0333-3452) and compared with those available

in the Ribosomal Database Project (Maidak et al. 2001) and

GenBank (Benson et al. 2000). All sequences were imported

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. into the ARB software environment (Strunk et al. 1998),

aligned based on secondary structures using the ARB

sequence editor, and phylogenetic relationships were

evaluated using neighbor-joining, maximum likelihood (Olsen

et al. 1994), and maximum parsimony methods (Swofford

1999) .

Electron microscopy

For electron microscopy, particulate from ~1 1 of melt

water were concentrated onto an 0.2 nm filter, fixed for 16

h with 3% (v/v) glutaraldehyde in phosphate-buffered saline

and then for 1 h in 1% (w/v) osmium tetroxide. The fixed

cells were dehydrated by sequential passage through

increasing concentrations of ethanol [50% to 100% (v/v)

ethanol in 10% increments], dried in a Pelco CPD-2 critical

point dryer (Ted Pella Inc., Redding, CA) , coated with

gold-palladium for 60 s in a Pelco 3 sputter coater, and

visualized using a Philips XL30 scanning electron

microscope.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nucleic acid quantitation

Cells collected by filtration of 750 to 1000 ml

aliquots of melt water were suspended in 3 ml of phosphate-

buffered saline, and 500 |il of the suspension was lysed with

hot detergent treatment and bead-beating, as described

above. The nucleic acids released, and similarly released

from 100-fold serial dilutions of an E. coli culture, were

denatured in 0.5 N NaOH and 6X SSPE (buffered hybridization

solution) , transferred, and fixed by UV crosslinking onto

nylon membranes (Sambrook et al. 1989). Following

hybridization to the universal 1492R 16S rDNA probe [y-32P]-

ATP end-labeled, the blot was rinsed with a solution of 2X

SSPE and 0.1% SDS at 22°C, used to expose film, which was

developed after 4 days of exposure at -70°C.

Freeze-thaw experiments

Cells from colonies growing on agar-solidified media

were suspended in 1 ml of dH20 by vortexing, and the number

of cfu ml'1 determined by serial dilution plating. Cell

suspensions were frozen at -15°C, thawed, and this cycle was

repeated 18X. The number of cfu remaining was deterimined

after cycle 5, 10, and 18. The known strains investigated

w e r e Aureobacterium suaveolens (ATCC 958) , Arthrobacter

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. globiformis (ATCC 8010) , Brevibacterium linens (ATCC 9172)

Bacillus polymyxa (OSU ref. 443), Micrococcus luteus (OSU

ref. 122), and E. coli (OSU ref. 455).

Antibiotic-resistance assays

Antibiotic disks containing (jig/disk) ampicil lin (10)

gentamicin (10) , k a n amycin (30), er ythromycin (15),

tetracyclin (30), and clindamycin (2) [Dispens-O-Disk***

susceptibility system; Difco] were placed on plates spread

with cell suspensions of the test organisms and then were

incubated at the optimal growth temperature for that

isolate. Zones of inhibition were measured to deterimine

susceptibility, as recommended by the disk manufacturer.

Results

Distribution, geographic, and chronological variation of microorganisms recovered from glacial ice

A range of different enrichment media was used

during the course of this study, however, growth was

observed only under aerobic or microaerobic conditions

using culture media routinely used to enumerate

heterotrophic bacteria and fungi. When these media were

solidified by inclusion of agar and inoculated with equal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amounts of a sample, the largest numbers of colonies were

routinely observed on media containing low levels of

nutrients, such as R2A, Actinomycetes isolation agar, and

nutrient and tryptic soy broth diluted 100-fold below the

manufacturers' recommended concentrations (Fig 3.2). Often

colonies appeared o n l y after 20 days of incubation at 22°C,

or after >70 days of incubation at 10°C. However, most of

these isolates were subsequently able to form colonies

within 2 to 7 days when sub-cultured onto the same growth

medium at 22°C. With melt water samples >500,000 years old

from the Guliya ice cap, colonies were never observed

growing after direct inoculation of agar-solidified media,

even after >100 d, but isolates were recovered from liquid

enrichments (full strength and 1:100 nutrient broth, 1:100

tryptic soy broth) after incubating at 4°C for 30-60 d.

Aliquots of the enrichment cultures were plated on agar-

solidifed media and incubated at 4° and 22°C. Based on the

numbers of cfu ml-1 retrieved from the primary liquid

enrichments, the bacteria did not multiply initially in

these cultures for 3 0 days, apparently requiring a

resuscitation period before any growth became possible.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TBAB NA 1:100 Figure Figure 3.2 Resulting growth after 200 ^1 of a filtered) 40-fold concentrated jim (0.2 aerobically at 22°C. aerobically at 22°C. There was a ~20-fold increase in recovery of cfu on low nutrient sample sample from 200-year old Guliya was ice spread onto rich tryptic (TBAB; blood agar base) and low nutrient 1:100; (NA nutrient agar diluted 100-fold) media and incubated media.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ice cores of different ages from China, Bolivia,

Greenland, and Antarctica (Table 3.1) were sampled to

survey the abundance and range of different bacterial

species that could be revived. No growth was observed in

samples from 150-year old ice from the Antarctic Peninsula,

nor from 1500-, 13K-, 14K-, 20K-, or 22K-year old ice from

the Sajama ice cap (Bolivia) , whereas 180 cfu ml'1 were

recovered from 200-year old ice from the Guliya ice cap

(China) . Low but similar numbers of cfu (<20 cfu ml"1) were

recovered from both modern and 12,000 to 20,000-year old

ice from Sajama, indicating that age had little correlating

effect on the number of recoverable bacteria in ice from

this region. Late Holocene (1800 years old) polar ice from

Taylor Dome (Antarctica) contained only 10 cfu ml'1, but

nevertheless, this was a higher number than from ice of the

same age from the Antarctic Peninsula or from Greenland

(Summit and Dye 2). Similarly low numbers of isolates (1-5

cfu ml'1) were counted by Dancer et al. (1997) from surface

glacial ice from the Canadian high Arctic after enrichment

for coliform bacteria, and even lower numbers (<1 cfu ml'1)

were counted in earlier surveys of polar ice (Abyzov et al.

1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vertical Ice Core Max. cfu ml'1 distance/ ' Sample Depth/Age | # of melt /unique ( sample (absVyear) j samples volume isolates designation) I obtained i (cm/ml) I Guliya, ?/50 ! 1 | ?/180 7/11 ■ China (G) ?/200 ! i 34/200 180/13 295/>500K i i i 78/170 0/4b : 296/>500K 1 i 72/170 0/14b Sajama, 5xlOc/ 1 40/100 i 2 <1/7 Bolivia (SB) 24-35c I ! 1 45-51/ 45/150 | 5 <1/6 ! i 140-190 i 7 6 / 1 .5K I 1 36/75 o 1 1 6x10C, i 104/12K i 3 3.5/ 20- 17/12 ! i 28c, 190 26, 17/ t 110/13K 1 2 0 1 180, 114 1 i 111/14K ! 1 28/187 0 1 112/14K I 1 31/191 0 j 36, 7/ 84, 1 ! : 118/19K 3/1 j 2 180 14.5-43/ 1 121/20K 0 < 90-130 ! 14-25/ 90- 124/22K 0 3 155 U y a 2 , Greenland ?/200 1 31/184

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.1 _.Vertical ^ . ! __ Max. cfu _ ml . _i1 Ice Core distance/ . . Sample Depth/Age # of melt /unique (saaple (mba/yaar) samples volume, isolates . . . . designation) (cm/ml), ... obtained Canada 1/? j 1 50/600 8/2 Glacier, Antarctica e (CanClean/ 1 50/200 1.1 x 104/5e CanDirty) 5x10/ n ! 97/150 ! 1 18-30= | 0 41. 47/ j ; isi/

a mbs=meters below surface

b Isolates obtained only from liquid enrichments.

c Melt water from 10 cm sections of each ice core was fractionated in these experiments.

d Only fungal growth observed.

e Sample from a frozen cryoconite hole on the glacial surface.

?=information not available.

Table 3.1 (continued)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In an attempt to measure the total numbers of

microorganisms present in different glacial ices, DNA

released and concentrated from cells in large volumes (750

to 1000 ml) of melt water from 150 year old Dyer Plateau

ice, 200 year old Guliya and Dye 2 ice, 1500 and 12,000

year old Sajama ice were probed by slot blot hybridization.

Following hybridization using the [y-32P]-ATP end-labeled

universal 16S rDNA probe (1510-1492) , signals were observed

only from samples of the 200 year old Guliya ice and from

the E. coli controls (data not shown) . Based on comparison

of signal intensities with the E. coli controls, the Guliya

ice contained a number of rDNA copies equivalent to that

present in -104 bacterial cells ml"1. Consistent with this

result, the highest numbers of colonies recovered (-180 ml"

1) also came from this particular section of the Guliya

core, although this number of cfu apparently represented

only -1% of the cells present in the retentate, based on

the hybridization data.

Scanning electron microscopy (Fig. 3.3) of material

filtered from water from Taylor Dome ice revealed the

presence of microorganisms and some larger particles of

putative biological origin, but very little inorganic

debris. In contrast, EM revealed much larger quantities of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.3 Microorganisms and particulates filtered from glacial ice cores visualized by scanning electron microscopy.

A. Particles trapped in (1) 1800 year-old ice from Taylor Dome, Antarctica; (2) 12,000 year-old ice from Sajama, Bolivia; (3) 200 year-old ice from Guliya, China.

B. (1) Coccal and (2) rod-shaped bacteria from 1800 year- old ice from Taylor Dome. (3) Diatom from 12,000 year-old Sajama ice, and (4) pollen grain from 200 year-old Guliya i c e .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 5 0 0 n m

Figure 3.3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. apparently inorganic granules, in addition to pollen

grains, diatoms, and a variety of bacteria in material

filtered from similar volumes (-1 1) of water from Guliya

and Sajama ice. Macroscopic particles were also visible in

these nonpolar glacial ices, typically forming dust layers

that apparently represented the annual deposition of

particles from nearby environments.

Relatedness of glacial isolates to known bacterial species

Isolates were designated by their geographical origin,

age of the ice in years, and strain number (e.g., G500K-78

is strain no. 78 isolated from Guliya glacial ice that was

500,000 years old), and compared to their closest

phylogenetic relatives by analysis of 16S rDNA sequences

corresponding to the region between nucleotides 27 and 1492

in the Escherichia coli 16S rDNA sequence (Fig. 3.4-3.8;

Table 3.2) . They were further categorized in terms of

growth temperature optima, ability to grow on different

media, and antibiotic resistance patterns (Table 3.3).

Based on having a 16S rDNA sequence >95% identical to that

of documented species residing in the a— , f5—, and

y-proteobacteria, and Cytophaga/ Flavobacterium/Bacteroides,

high, and low G+C Gram positive lines of descent, most

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

GP GP Low G+C G+C Low G+C High Cytophaga/Flavobacterium/ C F B

n p ■ N o n p o l a r □ Antarctic line of descent, and low G+C GP) (Low and high G+C (High GP) G+C Gram a

0 5 10 15 30 20 3 5] 5] 3 25 ice ice are members of the a-, proteobacteria, p~, and y- (CFB) Figure 3.4 Distribution of glacial isolates based on phylogenetic assignment to Bacteroides positive bacteria. major bacterial divisions. Recovered isolates from both polar and non-polar glacial O (0 rtJ d) 10 -P H •H =#= O oo U)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.5 Phylogentic analysis of a-proteobacterial isolates recovered from glacial ice. The 16S rDNA sequences obtained from single-colony isolates, corresponding to nucleotides 27-1492 of the E. coli 16S rDNA, were aligned based on secondary structure using the ARB software package (Struck et al. 1998). Trees were generated using a 1357 nucleotide mask of unambiguously aligned positions. GenBank accession numbers are listed in parentheses. Glacial isolates are in large font, and designated by their geographical origin (G=Guliya, China; SB=Sajama, Bolivia; M3C=Taylor Dome, Antarctica; CanClear=Canada Glacier, Antarctica; SIA=Siple Dome, Antarctica; and V=Lake Vostok accretion ice, Antarctica [see Chapter 5]) age of the ice in years, and strain number (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.

A. Maximum likelihood tree generated using fastDNAml (Olsen et al. 1994). The scale bar indicates 0.1 fixed substitutions per nucleotide position.

B. Maximum parsimony analysis performed with 100 bootstrap replicates, with values shown at the nodes. The scale bar represents 100 evolutionary steps. Of the 381 variable characters in this alignment, 295 were parsimony- informative .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Methylobicteriua radiotoloraas (D32227) 6500K-5 (M395035) ‘ Methylobicteriw fujism eos* (AJ250801)

Hethylobdctoriam a eso p h iliaa (D32225) ' Methylobactorim sp. (Z23156) unidentified bacteriun (AJ223453) 4 V23 (AF324201) Mstbylobsctsrim utorqims (032224) Hethylobictoriaa rhodesitaim (D32228) M3C1. 8K-TD4 (AF395030) HsthylabtctOTiim sp. (U58018) 6500K-15 (AF395034) Bom thiooxidias (AJ250798) A fip it sp. (U87779) alpha proteobacteriue (AF288309) M3C1. 8K-TD7 (AF479379) I Sphiagomoau psuciaobilis (X94100) L G500K-3 (AF395036) ~ Sphiagomoau chloropkenolica (X87161) — Sphiagomoau flava (X87164) M3C1.8K-TD9 (AF479381) Blutomoau aststorius (X73043) Erythromaau ursiacoli (Y10677) Blutomoau aatatoria (AB024288) Blutomoau ursiaicola (AB02489) SIA1K-1A1 (AF395032) isolate "star-like" (AJ001344) B lu to b s c U rsp. (Z23157) Sphiagomoau pzvai (Y09637) Sphiagomoau a u li(Y09638) Sphiagomoau usccborolytic* (Y09639) _j Sphiagomoau sp. Ant. 20 (AF184221) M3C4. 1K-B5 (AF395031) G500K-14 (AF395037) || V21 (AF324199) L Sphiagomoaas sp. (AB033945) ~ Sphiagomoau parapauciaobilis (D13724) Sphiagomoau sp. 1K355 (084523) M3C1.8K-TD1 (AF479378) Sphiagomoau u ag ais (013726) 0.10 Sphingoaona* teh iao id u (AB033944) .— CanClearl (af395038) r Sphiagomoau paucimobilis (016144) Sphiagomoau pucim obilis (013725)

Figure 3.5A 85

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mtthylobictsriaa zadiotolsrus (D32227) G500K-5 (AF395035) HstbylabseUrim fu jiu n m t(U250801) Usthylobactuim msepbiliem (D32225) 83 Mitbfiabietciini ip. (823156) 98 unidutifitd bictuia (U223453) 100 87 V23 (AF324201) — Mrtbjriafaictcriia axtorquwj (032224) 80 Jfettylabictiria iboduitm (032228) 100 86 M3C1.8K-TD4 (AF395030) ■96- MtClqrlobaetcrini sp. (058018) G500K-15 (AF395034) Bom thiooxidm (AJ250798) 59 100 Afipii sp. (087779) alpha protMbactarioi (IF288309) M3C1.8K-TD7 (AF479379) [tt— SphingoMnas pueiaobilis (X94100) 100 10Q 1--- G500K-3 (AF395036) ■ SphingcKBU chloropbeaoha (X87161) 96 ■ Sphingownas f lm (X87164) M3C1.8K-TD9 (AF479381) 100 100 — Blistcaooss oststorios (Z73043) 70 Eiytbrcaom orsincols (110677) TT Blutoaoou u ta to rii(1B024288) Blutoaoou auiaieols (AB02489) 70 SIA1K-1A1 (AF395032) isolata "star-likt" (AJ001344) 95 Blutobsetsr sp. (Z23157) Sphiagcwnas pruni (Y09637) 96 64 83 Spbiogcaooas aali (109638) Sphingansas ssicchirolytia (Y09639) 100 — -— Spbiagoaous sp. Int. 20 (1F184221) M3C4.1K-B5 (AF395031) 54 86 G500K-14 (AF395037) 92 --- V21 (AF324199) Sphingnnnas sp. (AB033945) iphinjfmmp* pfnpinrimfihi 1t« (013724) Sphiogoaoou ip. H355 (084523) M3C1. 8K-TD1 (AF479378) Spbiagoaoou sanguis (013726) Spbiagoaoou ocbiooidu (1B033944) CanClearl (af395038) 81 Spbiagoaoou pueiaobilis (016144) 72 SphingeKUS pueiaobilis (013725)

Figure 3.5B 86

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.6 Phylogenetic analysis of 0- and y- proteobacterial glacial isolates, and a member of the Cytophaga/Flavobacterium/Bacteriodes line of descent (CanDirtyl4) . The 16S rDNA sequences obtained from single colony isolates, corresponding to nucleotides 27-1492 of the E. coli 16S rDNA, were aligned based on secondary structure using the ARB software package (Struck et al. 1998) . Trees were generated using a 1395 nucleotide mask of unambiguously aligned positions. GenBank accession numbers are listed in parentheses. Glacial isolates are in large font, and designated by their geographical origin (G=Guliya, China; SB=Sajama, Bolivia; M3C=Taylor Dome, Antarctica; CanClear/CanDirty=Canada Glacier, Antarctica; and Trans=Ross Ice Shelf, Antarctica) age of the ice in years, and strain number (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.

A. Maximum likelihood tree generated using fastDNAml (Olsen et al. 1994) . The scale bar indicates 0.1 fixed substitutions per nucleotide position.

B. Maximum parsimony analysis performed with 100 bootstrap replicates, with values shown at the nodes. The scale bar represents 10 evolutionary steps. Of the 571 variable characters in this alignment, 536 were parsimony- informative .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pseudoaonis aeruginosa (038445) r Pseudoaonis oleonans (D84018) 1 M3C4.7K-2 (AF479376) Pseudoaonis tlciligenes (Z76653) Pseudominis pseudoilciligenes (Z76666)) Pseudoaonis sp. fro* sea ice (085869) Pseudommis sp. frca sea ice (085868) M3C4.1K-B34 (AF479375) Pseudoaonis uotoforaans (084009) Pseudoaonis synxuthi (Af267911) Psychrobicter ismobilis (085880) C Psychrobicter gltcincoli (085876) Transl2 (AF479327) Psychrabacter soburinos (AJ309940) Psychrobicter airin co li(AJ309941) Acinetobacter jooii (X81664) Acinetobacter johnsonii (Z93439) Acinetobacter hieaolyticus (Z93436) Acinetobacter caicoaceticos (AF159045) M3C1.8K-TD8 (AF479380) Acinetobacter rtdioresistens (X81666) Acinetobacter sp. (Z93445) G50-TB2 (AF479352) CanClear23 (AF479323) SB150-2A1 (AF479373) CanDirtyl2 (AF479324) r Pseudoaonis saccharqpfaiia (AB021407) Matsuebacter chitosmotibidus (AB006851) Leptothrix cholodnii (X97C70) Leptotbrix discophon (L33974) Janthinabacteriua igiricidiano (Y08845) Jmthinobicteziua lividua (Y08846) Pseudoaonis sephitica (AB021388) _ r CanDirty89 (AF479326) Duginella riaig e n (X74914) P ilsto n ii sp. APF11 (AB045276) G500K-6 (AF479337) Ralstonia pickettii (AB004790) Ralstonia solanacearua (028233) j Baloanelh gillinarua (AB035150) JLKL — CanDirtyl4 (AF479331) ~ Chryseobtcteriua bilustinim (M58771) Cbxyseobacteriua pleas (M58772) R ieaeielli coluebini (AF181448) Figure 3.6A 88

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Psaodaooas uagiaou (038445) ■Psaodaooas alcaiigcoas (27(653) ■ Psaoduoaaa psaodoalcaligtoas (27(666)) Psaodaooas oltamzm (D84018) M3C4.7K-2 (AF479376) Psaoduoaas sp. fra su in (085869) Psaodaooas sp. fret su icc M3C4.1K-B34 (AF479375) - Psndotcw uotofoam Pstadowns synmtbi (A£267911) Psjcbzobiette imobilis (085880) Psychrabictir jhciacoh (085876) Transl2 (AF479327) 100 Psycbrobicttr sdm uim (U309940) Psycbnbuttr suincoli (U309941) Jciostobactsr j a i i (281(64) Peinstobictir jobm aii (293439) 65 iciutobictu btmolyticvs (293436) 460 Jcioatobactar cilcuctticas (AF159045) M3C1.8K-TD8 (AF479380) Acinitabicttr ndioruistess (281666) to'natohactar sp. (293445) G50-TB2 (AF479352) CanClear23 (AF479323) SB150-2A1 (AF479373) Can0irtyl2 (AF479324) Psaodaooas sacdiarqphila (AB021407) Mitsoabactar cbitouaotibidos (1B006851) Ltptotbzix ebolodaii (297070) Lqtothrix discopbon (L33974) Jaotbioobacfcaria tguieidmo (108845) Jaotbioahactcria lividm (108846) Psaodaooas asphitica (1B021388) CanDirty89 (AF479326) Dcgaoalli ruigtra (274914) Pilstonii sp. APF11 (AB045276) G500K-6 (AP479337) Pilstonii p iekittii (1B004790) Pilstonii solaoacaara (028233) Ealo io tlli gallioaia (1B035150) CanDirtyl4 (AF479331) Cbysaobactaria balustioa (M58771) Chryuobictuim glaa (M58772) P im m lli colabioi (1F181448) Figure 3.6B 89

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.7 Phylogenetic analysis of Low G+C Gram positive glacial isolates. The 16S rDNA sequences obtained from single-colony isolates, corresponding to nucleotides 27- 1492 of the E. coli 16S rDNA, were aligned based on secondary structure using the ARB software package (Struck et al. 1998). Trees were generated using a 1330 nucleotide mask of unambiguously aligned positions. Glacial isolates are in large font, and designated by their geographical origin (G=Guliya, China; SB=Sajama, Bolivia; and V=Lake Vostok accretion ice, Antarctica [see Chapter 5]) age of the ice in years, and strain number (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.

A. Maximum likelihood tree generated using fastDNAml (Olsen et al. 1994). The scale bar indicates 0.1 fixed substitutions per nucleotide position.

B. Maximum parsimony analysis performed with 100 bootstrap replicates, with values shown at the nodes. The scale bar represents 10 evolutionary steps. Of the 326 variable characters in this alignment, 294 were parsimony- informative .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bacillus avbtilis (026185) G50-TB5 (AF479353) G200-SD1 (AF479349) Bacillus sp. (AB050667) Bacillus sp. (AF411118) H G500K-16 (AF479336) n Bacillus pumilus (AB048252) SB100-9-5-1 (AF479372) Bacillus sp. 82352 (AF227852) Bacillus sp. MK03 (AB062678) G500K-19 (AF479330) G500K-9 (AF479338) Bacillus anthracis (AF176321) Bacillus sp. BS10723 (AF027659) Bacillus caiaus (AF290548) G50-TS3 (AF479356) G500K-2 (AF479333) SB12K-9-4 (AF479367) Mariana Tranch isolata (AB002640) unidantifiad bactariua (AB004761) Bacillus thuzingiansis (016281)) SB100-8-1 (AF479369) Bacillus csrmus (AF290555) Bacillus cobnii (X76437) G200-T16 (AF479350) Bacillus sp. KSM-KP43 (AB055093) Bacillus amgataziua (016273) G500K-18 (AF479335) li~ Bacillus siaplax (078478) Bacillus psychrosaccharolyticus (AB21195) SB100-8-1-1 (AF479370) SB105-2A2 (AF479374) G500K-17 (AF479334) Bacillus aaczoidas (AY030319) Bacillus aazoccanus (X60626) Antarctic isolata froa gaotharaal soil (AJ250318) i Bacillus fizmus (X60616) L_j— — 7^ - _ Bacillus cizculaas (AY043084) G200-T19 (AF479351) — Bacillus badius (X77790) G200-N5 (AF479347) ' Bacillus aminovozans (X62178) SB100-8-1-2 (AF479371) Exiguobactaziua acatylicum (X70313) SB12K-2-2 (AF479360) O Exiguobactaziua undaa (AJ344151) HTiExiguobactaziua acatylicum (AJ297437) Bacillus banzaovozans (D78311) Planococcus okaanokoitas (055729) Planococcus acmaakinii (AF041791) G50-TS4 (AF479357) Planococcus psychzotolazatus (AF324659) Antarctic psychrophila SOS Oranga (AF242541) - f GoOOK-78 (AF395033) — Paanibacillus sp. 7-5 (AB043868) Paanibacillus illinoisansis (085397) ■L G50-TB9 (AF395027) Bacillus longispozus (AJ223991) Paanibacillus aaylolyticus (085396) V22 (AF324200) G200-C15 (AF395028) SB150-2B AF395029 Q.1Q _ r Paanibacillus polyayxa (016276) •— Paanibacillus paoziaa (AJ320494) —r G50-L2 (AF4 79345) Paanibacillus sp. P51-3 (AJ297715) Figure 3.7A 91

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bacillus subtilis (D26185) G50-TB5 (AF479353) G200-SD1 (AF479349) ■Si- Bacillus sp. (AB050667) Bacillus

199 G500K-16 (AF479336) Bacillus pumUus (AB048252) Bacillus banraororans (078311) Antarctic isolate free gaothanal soil (AJ250318) Bacillus firmus (X60616) 65 Bacillus circulans (AX043084) G200-T19 (AF479351) Bacillus badius (X77790) SB100-9-5-1 (AF479372) ----- Bacillus sp. 82352 (AF227852) 100 Bacillus sp. MC03 (AB06267B) 82 [80

100 Bacillus antbracis (AF17&21) Bacillus sp. BSID72323 (AF027659)(M027659) Bacillus caraus (AF290548) 100 G50-TS3 (AF479356) G500K-2 (AF479333) SB12K-9-4 (AF479367) Mariana Tranch isolate (AB002640) unidantifiad bactariua (AB004761) Bacillus tbuzingiansis (D16281)) SB100-8-1 (AF479369) Bacillus caraus (AF290555) Bacillus eohnii (X76437) 73 G200-T16 (AF479350) I 100 99 Bacillus sp. KSM-KP43 (AB055093) Bacillus magatarium (016273) G500K-18 (AF479335) 51 Bacillus siaplax (078478) Bacillus psycbzosacchaiolyticus (AB21195) 99 SB100-8-1-1 (AF479370) 93 SB105-2A2 (AF479374) 99 G500K-17 (AF479334) B acillus macroidas (AI030319) Bacillus aaroccanus (X60626) G200-N5 (AF479347) 99 Bacillus aminovorans (X62178) 12------S B 1 00 -8 -1 -2 (AF479371) Planococcus okaanokoitas (055729) 10 Planococcus acaaalrinfi (AF041791) G50-TS4 (AF479357) 64 Planococcus psycbrotolaratus (AF324659) (95 Antarctic psychrophila SOS Oranga (AF242541) Exiguobactarium acatylicum (X70313) 71 SB12K-2-2 (AF479360) - Exiguobactarium undaa (AJ344151) ■K Ixiguobactarium acatylicum (AJ297437) G500K-78 (AF395033) 1100 69 Paanibacillus sp. 7-5 (AB043868) Paanibacillus illinoisansis (085397) 93 G50-TB9 (AF395027) 100 — Bacillus longisporus (AJ223991) 100 -f”~[55 Paanibac ill us aaylolyticus (085396) -V22 (AF324200) [89 G200-C15 (AF395028) SB150-2B (AF395029) 64 Paanibacillus polympta (016276) 175 Paanibacillus paoriaa (AJ320494) 100 ’ G50-L2 (AF479345) 100 ■ Paanibacillus sp. P51-3 (AJ297715) Figure 3.7B 92

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.8 Phylogenetic analysis of High G+C Gram positive glacial isolates. The 16S rDNA sequences obtained from single-colony isolates, corresponding to nucleotides 27-1492 of the E. coll 16S rDNA, were aligned based on secondary structure using the ARB software package (Struck et al. 1998) . Trees were generated using a 1366 nucleotide mask of unambiguously aligned positions. Glacial isolates are in large font, and designated by their geographical origin (G=Guliya, China; SB=Sajama, Bolivia; CanDirty=Canada Glacier, Antarctica; SIA=Siple Dome, Antarctica; Trans=Ross Ice Shelf, Antarctica; and V=Lake Vostok accretion ice, Antarctica [see Chapter 5] ) age of the ice in years, and strain number (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.

A. Maximum likelihood tree generated using fastDNAml (Olsen et al. 1994). The scale bar indicates 0.1 fixed substitutions per nucleotide position.

B. Maximum parsimony analysis performed with 100 bootstrap replicates, with values shown at the nodes. The scale bar represents 100 evolutionary steps. Of the 4 63 variable characters in this alignment, 375 were parsimony- informative .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mycobacterium alvei (AF023664) ~ Mycobacterium austroafricanum (X93182) Mycobacterium aurum (X55595) SB12K-2-16 (AF479359) j~ Gordona namibiensis (AF380931) Gordona rubropertiactus (X80632) r High G+C isolate (X7318) SIA1K-2A1 (AF479377) Gordona terras (X79286) — Rbodococcus sp. (AB010902) ■ Rbodococcus equi (X80614) Hocardia corynebacteroides (X80615) SB12K-2-5 (AF479363) G200-C39 (AF479344) Microoonospora rbodorangaa (X92612) Micraaonospora ecbinospora (X92607) Microoonospora inyoensis (X92629) — Friedmanniella lacustris (AJ132943) J- SB12K-2-7 (AF479364) Friedmanniella spumicola (AF062535) — Friedmannella antarctica (Z78206) Hocardioides simplex (AF005011) — Hocardioides sp. 0S4 (U61298) Hocardioides plantarum (AF005008) SB12K-2-4 (AF479362) r Bracbybacterium tyroferaentans (X91657) — Bracbybacterium sp. from sea ice (AF041790) — Bracbybacterium faecirn (X91032) r Bracbybacterium conglomerating (X91030) ~ V 1 5 (AF324202) r Kocuria kristinae (X80749) ^ Trans4678 (AF479328) _| Kocuria varians (X87754) Xocuria rbizopbila (Y16264) deep subsurface isolate (X86608) SB19K-1 (AF479368) unidentified sludge eubacterium (Y15857) Micrococcus luteus (AJ409096) Artbrobacter sp. CAB1 (AB39736) G200-A1 (AF479340) CanDirty7 (AF479339) L| Artbrobacter polycbromogenes (X80741) Artbrobacter oxydans (X83408) — Artbrobacter sp. from sea ice (U85895) ' Artbrobacter sp. from sea ice (AF041789) CanDirtyl (AF479325) Artbrobacter agilis (X80748) G50-TB7 (AF479354)

Q.1Q r G200-C1 (AF479341) Ip clone from spacecraft (AF408269) L Artbrobacter sp. from sea ice (085896) Figure 3.8A 94

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. G200-C18 (AF479343) Sanguibacter keddieii (X79450) _r Sanguibacter inulinus (X79451) Sanguibacter suarezii (X79452) J Cellulomonas sp. (X82598) L SB12K-6-2 (AF479365) Promicromonospora enteropbila (X83807) j Cellulomonas turbata (X83806) Oerskovia paurometabola (AJ314851) Agromyces mediolanus (D450S6) Agromyces cerinus (X77448) [~ Agromyces ramosus (X77447) G500K-1 (AF479332) G50-PD1 (AF479348) r Clavibacter michiganensis (AJ310415) Clavibacter michiganensis (U30254) Frigoribacteriurn faeni (Y18807) j Frigoribacterium sp. 227 (AF157478) Frigoribacterium sp. 301 (AF157479) G200-C11 (AF479342) SB12K-2-1 (AF479358) gram positive isolate W5P (AF323267) Leucobacter komagatae (AB007419) gram-positive bacterium str. 1 (AB008511) G50-TB8 (AF479355) Mycetocola sp. 0M-A1 (AB020204) ~ Detolaasinbacter tsukamotoae (ABO12646) j Detolaasinbacter muratae (AB012648) Detolaasinbacter sbiratae (AB012647) Microbacterium arborescens (X77443) SB12K-6-3 (AF479366) Microbacterium sp. VKM Ac-1 (AB042072) Aureobacterium testaceum (X77445) Aureobacterium sp. (Y14699) ~ Aureobacterium liquefaciens (X77444) _j Aureobacterium keratanolyticum (Y14786) G500K-10 (AF479329) rC G50-L3 (AF479346) Microbacterium sp. MAS133 (AJ251194) Microbacterium flavescens (Y17232) SB12K-2-3 (AF479361)

- 0.10

Figure 3.8A (con't) 95

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mycabicttrim alvei (AF023664) Ifycabicttriua mtroafricamu (193182) Hycobicttrina auras (155595) SB12K-2-16 (AF479359) r— Gordon* nuibitnsis (AF380931) — Gordon* rubroptrtinctus (X80632) High G+C isolata (X7318) tM— SIA1K-2A1 (AF479377) Gordona terras (179286) _T7— Bbodococms sp. (AB010902) — Bhodococcns tqni (X80614) i... Soc*rdi* corvnabacteroides (180(15) SB12K-2-5 (AF479363) G200-C39 (AF479344) Microaonotpcr* rbodortngt* (192612) Micremonospor* ecbinoipora (192607) Micrtmonospor* inyotnsit (192629) 7;j | Fritdunnitll* ltevstris (AJ132943) SB12K-2-7 (AF479364) Fritdunnitll* spuaicola (AF062535) FritdunntU* antaretiea (Z78206) Bbeudioidts sitpltx (AF005011) Bocudioidts sp. 0S4 (061298) r~ r. ■ ■ ■ Soc*rdioidts p lin tu a(&F005008) “ -- SB12K-2-4 (AF479362) |'1fln Bnchybicttriim tyroftnmtins (191657) ■ Bnchyb*ct*riaa sp. froa saa ica (AF041790) Bracbybacteriua faacina (191032) M Br*cbyb*ct*rim conglomntm (191030) V15 (AF324202) |!j Kocnri* k ris tin u (180749) 53 Trans4678 (AF479328) KOeurit varitns (187754) Kocuri* rbizophil* (Y16264) deep subsurface isolata (186608) 75 SB19K-1 (AF479368) 100•"iT unidentified sludge eubacteriua (¥15857) Mieroeoeau lutaos (AJ409096) Artbrobacter sp. CIB1 (AB39736) ■55— G200-A1 (AP479340) CanDirty7 (AF479339) ^7— Artbrobacter polycbramogtnas (180741) — Artbrobacter oxydtns (183408) Artbrobacter sp. froe su ice (085895) Artbrobacter sp. froa sea ice (AF041789) CanDirtyl (AF479325) Artbrobacter t g ilit (180748) — G50-TB7 (AF479354) ra— G200-C1 (AF479341) — clone froa spacecraft (AF408269) Artbrobacter sp. froa sea ice (085896) Figure 3.8B 96

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. G200-C18 (AF479343) 68 Sanguibacter keddieii (X79450) Sanguibacter inulinus (X79451) Sanguibacter suarexii (X79452) Cellulomonas sp. (X82598) SB12K-6-2 (AF479365) Promicromonospora enteropbila (X83807) Cellulooonas turbata (X83806) OersJcovia paurometabola (AJ314851) gram-positive bacterium str. 1 (AB008511) G50-TB8 (AF479355) G200-C11 (AF479342) 65 Agromyces mediolanus (D45056) 100 - Agrooyces cerinus (X77448) 72 - '■ Agromyces raaosus (X77447) '*--- G500K-1 (AF479332) G50-PD1 (AF479348) 100 ■ Clavibacter michiganensis (AJ310415) 80 • Clavibacter michiganense (U30254) Frigoribacterium faeni (Y18807) 99 — Frigoribacterium sp. 227 (AF157478) 92 II — Frigoribacterium sp. 301 (AF157479) SB12K-2-1 (AF479358) ■Ti00~r — gram positive isolate NSP (AF323267) 3 2 — Leucobacter komagatae (AB007419) Detolaasinbacter tsukamotoae (AB012646) 99 — Mycetocola sp. OM-A1 (AB020204) 58 ■ Detolaasinbacter muratae (AB012648) £ - Detolaasinbacter shiratae (AB012647) Microbacterium arborescens (X77443) "" SB12K-6-3 (AF479366) ^ Microbacterium sp. VKM Ac-1 (AB042072) 89 — — Aureobacterium testaceua (X77445) Aureobacterium liquefaciens (X77444) 55 “ “ Aureobacterium keratanolyticum (Y14786) G500K-10 (AF479329) 100 59 ibacterium sp. (Y14699) “L3 (AF479346)

{5 (Y17232) (AF479361) Figure 3.8B (con't) 97

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Strain 1i iI designation*/ Nearest phylogenetic % ! Zee Core GenBank I Site j neighbor/GenBank identity0 I accession accession no. i no./# of i sequenced ntb i i i a proteobacteria Sphingomonas 98.4 | G500K-3/ paucimobi 1 is/X94100 AF395036/ Sphingomonas 1395 chlorophenolica/ 96.8 | X87161 1 Me thyl obacteri um 99.3 | G500K-5/ radio tolerans /D32227 AF395035/ Me thyl obacteri um 1402 98.7 | fuj isawaense/AJ250801 i V21* 99.8 G500K-14/ (0 Sphingomonas sp. / i AF395037/ 99.1 i G AB033945 -H 1416 si M3C1.2K-B5d 97.2 ! u M3C1.8K-TD4d 99.5 ' G500K-15/ Methylobacterium sp./ rti 99.5 AF395034/ U58018 i -H rH 1418 Me thyl obacteri um 1 G 99.1 j o extorquens/D32224 i P proteobacteria ! I sp./ i Ralstonia ! G500K-6/ AB045267 ioo AF479337/ Ralstonia pickettii/ 1447 99.8 ! AB004790 i I Y proteobacteria | t Acinetobacter I 99.8 G50-TB2/ radioresistens/X81666 1 I AF479352/ SBl50-2Ald 99.8 1 1 1438 Acinetobacter sp./ 1 99.8 1 Z93445

Table 3.2 Bacterial isolates from glacial ice cores.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 1 Strain i 1 i designation*/ Nearest phylogenetic 1 I c e Core GenBank 1 neighbor/GenBank % Site aceession identity* | accession no. t no./# of ] i i sequenced ntb 1 i Low G+C gram positive bacteria i G50-TB5/ Bacillus subtilis/ 99.5 ! AF4793 53/ D26185 i ! 1464 Bacillus sp./AB050667 99.2 Bacillus cereus/ 99.9 i G50-TS3/ AF290548 AF479356/ G500K-2d 1450 99.9 , SB12K-9-4d 99.9 Paenibacillus 98.2 G50-TB9/ amyl oly t i cus / D8 5 3 9 6 AF395027/ V22* 98.0 : 1479 Paenibacillus fC 97.1 ; C illinoisensis/D85397 1 -H Paenibacillus sp./ a T3 G50-L2/ 99.3 j 0) AJ297715 u 3 AF479345/ C Paenibacillus •H 1470 98.1 td 4J peoriae/AJ320494 ! > 0 •H U Planococcus 98.2 ! rH G50-TS4/ okeanokoites/D55129 5 AF479357/ Planococcus 1465 psychrotoleratus/ 98.1 i i AF324659 G200-SD1/ Bacillus sp./AB050667 99 .7 AF479349/ G50-TB5d 99.1 | 1443 Bacillus sp./AF411118 99.1 | G200-T16/ Bacillus sp./AB055093 99.8 ! AF479350/ / Bacillus cohnii 97 .1 1465 X76437 Bacillus circulans/ G200-T19/ 97.3 AF479351/ AY043084 1447 Bacillus firmus/ 97.0 X60616 Table 3.2 (continued)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Strain designation*/ Nearest phylogenetic 1 Ice Cora GenBank % neighbor/GenBank Sita accession identity* no./# of j accession no. sequenced ntb ! SB100-8-l-2d 97.6 ; G200-N5/ Bacillus aminovorans / 96.3 AF479347/ ; X62178 1442 ! Bacillus badius/ 95.1 j X77790 i I SBl50-2Bd 96.8 G200-C15/ ; Paenibacillus sp./ AF395028/ 94.8 1470 : AJ297715 G50-L2d 94.8 ! G500K-16/ Bacillus pumilus/ 100 i AF479336/ AB048252 i | 1454 | Bacillus sp./AB050667 97.4 i ! Bacillus sp./AB062678 99.6 | i G § G500K-19/ ! -H ; AF479330/ ! G500K-9d 99.4 j 1 JG CD 1450 ! Bacillus sp./AF227852 98.8 i o P G G500K-9/ ; G500K-19d 99.4 j 4J (0 C AF479338/ i Bacillus sp./AB062678 98.6 j >1 o 1 -H o 1458 ! Bacillus sp./AF227852 98.2 ! 1—1 ! Bacillus cereus/ 1 3 100 ! O G500K-2/ j AF290548 ! AF479333/ | Bacillus cereus/ i 100 ! 1452 ' AF290555 SB12K-9-4d 100 1 Bacillus megaterium/ i G500K-18/ 99.5 D16273 i ! AF479335/ Bacillus simplex/ I 1444 98.9 j D78478 Bacillus maroccanus/ 99.9 G500R-17/ X60626 AF479334/ SBl50-2A2d 99.4 1 1436 Bacillus macroides/ 99.3 | AY030319 i Table 3.2 (continued) inn

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Strain

designation*/ Nearest phylogenetic Ice Cor« GenBank % neighbor/GenBank Site accession identity6 accession no. no./# o£ sequenced nt , Paenibacillus sp./ 97 .1 G500K-78/ AB043868 AF395033/ 1446 : Paenibacillus 95.2 : illinoisensis/D85397 High G+C gram positive bacteria ; Artbrobacter agilis/ 99.2 G50-TB7/ ; X80748 AF479354/ ! G200-Cld 98.9 1437 clone from 98.9 spacecraft/AF408269

, Clavibacter (0 99.8 G G50-PD1/ michiganensisAJ30254 1 -I— I T3 AF479348/ i ,£ O Detolaasinbacter •H u G50-TB8/ 96.5 r— I tsukamo toaef ABO 12 646 £ AF479355/ O 1424 Detolaasinbacter 96.4 shira tae/ABO12647 SB12K-2-3* 97.5 G50-L3/ Microbacterium sp./ 97 .4 AF479346/ AJ251194 1404 Microbacteri um 97.3 ; flavescens/Y17 2 3 2 f | Micromonospora G200-C39/ 99.4 | rhodorangea/X9 2 612 i AF479344/ i i 1394 Mi cromonospora 99.2 echinospora / X9 2 6 0 7

Table 3.2 (continued)

ini

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Strain ! f designation*/ Nearest phylogenetic Ice Core GenBank % Site accession neighbor/GenBank accession no. identity6 no./# of i sequenced ntb i Arthrobacter sp./ 99.6 | AB39736 G200-A1/ Arthrobacter i 1 AF479340/ polychromogenes / 99.0 1: 1399 X80741 Arthrobacter oxydans/ 99.0 j X83408 Arthrobacter agilis/ G200-C1/ 99.2 ' X80748 AF479341/ 1403 Arthrobacter s p . from 99.1 sea ice/U85896 Sanguibacter T3 G200-C18/ 97.9 i keddieii / X7 9450

C h i n a i O Frigoribacterium sp./ -H O G200-C11/ 97.3 i r—1 AF157478 1 3 AF479342/ Frigoribacterium sp./ 0 1427 97.3 | AF157479 Agromyces ramosus/ G500R-1/ 98.6 X77447 i AF479332/ 1 Agromyces cerinus/ 1411 97.9 X77448 ; Aureobacterium G500K-10/ kera tanolyti cum/ 98.2 | AF479329/ Y14786 !t 1413 i Aureobacterium sp./ 98.1 | j Y14699 | (0 y proteobacteria <$ •H Acinetobacter e > 99.9 | rfl -H SB150-2A1/ | radioresistens/X81666 *n r—1 AF479373/ I ! (0 o 1435 G50-TB2d 99.8 j U1 PQ lj ______i CanClear23d 99.6 j Table 3.2 (continued) 102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 I ! Strain | I1 I designation*/ ! Nearest phylogenetic | Xc* Cor* | GenBank % neighbor/GenBank ! Sit:* J accession 1 identity6 j accession no. i | no./# o£ I sequenced ntb i | Low G+C gran positive bacteria 1 1 ! SB100-9-5-1/ Bacillus sp./AF227852 i 96.5 ! AF479372/ G500K-19d | 96.3 | 1408 | Bacillus sp./AB062678 | 96.2 i Bacillus cereus/ SB100-8-1/ 100 AF290555 ! AF479369/ 1 Bacillus ! 1460 I 99.9 i | thuringiens is/D162 81 i I i SB150-2A2d ! 98.8 | | SB100-8-1-1/ Bacillus maroccanus/ I j 98.8 I AF479370/ X60626 ; 1462 Bacillus macroides/ i "H I 98.6 J> AY030319 •H Bacillus aminovorans/ ' r H T3 98.4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Strain designation*/ sarest phylogenetic Ic* Cor* GenBank % neighbor/GenBank Sit* accession accession no. identity6 no./# o£ sequenced ntb Exiguobacterium i SB12K-2-2/ 99.4 acetylicum/X70313 | AF479360/ Exiguobacterium i 1470 97.8 ace tyl i cum/AJ2 97437 High G+C gram positive bacteria Mycobacterium aurum/ 99.6 SB12K-2-16/ X55595 AF479359/ Mycobacterium 1427 austroafricanum/ 98.9 X93182 Nocardia fd •H SB12K-2-5/ corynebacteroi des/ 99.6 > AF479363/ X80615 "H ^ 1428 Rhodococcus equif i—I T3 96.8 o ® X 8 0 6 1 4 __ m g Fri edmanni el la SB12K-2-7/ 99.1 spumicola/AF062535 «•? AF479364/ g 8 Fri edmanni el la oj 1428 98.3 *n antarctica/Z7 8206 fd Nocardioides simplex/ CA SB12K-2-4/ 94.8 AF005008 AF479362/ Nocardioides 1416 94.8 sp./U61298 Cellulomonas s p ./ SB12K-6-2/ 99.1 X82598 AF479365/

1426 Cellulomonas turbata/ 97 .7 X83806 gram positive 96.6 SB12K-2-1/ isolate/AF323267 AF479358/ Leucobacter 1448 96.1 komagataef KB007419

Table 3.2 (continued) 104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 | Strain t ' designation*/ i Nearest phylogenetic j Ice Core GenBank neighbor/GenBank 1 % Site accession i accession no. i identity6 j no./# o£ i sequenced ntb I' j Aureobacteri um . SB12K-6-3/ 99.9 1 tes taceuzn/X77445 t j AF479366/ -H ! Microbacterium sp./ I > i 1423 99.9 ! * H _ j AB042072 > I I— 1 'O ! Microbacterium sp./ ! 1 o ® SB12K-2-3/ 98.5 | ffl c j AJ251194 ! AF479361/ i Microbacterium 1433 ! 98.4 I ! flavescens/Y17232 i 1 (conti Micrococcus luteus/ SB19K-1/ 99.6 AJ409096 AF479368/ Sa j a mj aSa , deep subsurface 1385 | 99.6 isolate/X86608 ' i ! 1 a proteobacteria ! G500K-15d 1 99.5 . 8 K - T D 4 / Methylobacterium sp./ 2 H3C1 j 99.2 4-> AF395030/ U58018 i i 1373 Methylobacterium I u 98.7 | D3 2 2 2 4 11 ex torguens / ! G M3C1•8K-TD7/ a proteobacterium/ < 99.9 ! AF479379/ AF288309 ! 1376

> i Sphingomonas sanguis/ M3C1 • 8K-TD1/ 99.4 D13726 f . AF479378/ Sphingomonas s p . / 1405 99.4 D84523

Table 3.2 (continued) 105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 I Strain ■ designation*/ 1 Nearest phylogenetic i 1 Ice Core GenBank % neighbor/GenBank Site accession identity* i accession no. j i no./# o£ sequenced ntb j j Sphingomonas sp. from 99 .8 M3C4.1K-B5/ I Antarctica/AF184221 AF395031/ vai* 99.7 (0 1 U 1408 Sphingomonas s p . / •H 99.1 jj AB033945 i U 1 y proteobacteria (0 1 Acinetobacter j ■1—1 I 99.9 T3 M3C1.8K-TD8/ haemolyticus/Z93436 | c

CanClearl/ Sphingomonas 97.8 paucimobilis/D16144 SI fd AF395038/ -H Sphingomonas U 1419 97 .2 fd a * paucimobilis/D13725 r~i u ! O ! P proteobacteria o i i fd Pseudomonas \ i (d jj c ! saccharophilia/ ! 97.1 j —i < AF479324/ I

Table 3.2 (continued) 106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 | Strain j designation*/ Nearest phylogenetic ; Ice Core GenBank % neighbor/GenBank Site j accession \ accession no. identity6 j no./# of 1 i sequenced ntb Janthinobacterium 97.4 1 : CanOirty892/ lividum/Y08846 1 : AF479326/ i Pseudomonas 97.2 | 1442 | mephitica/AB021388 ' | y proteobacteria j Acinetobacter 99.6 radioresis tens/X81666

1 CanClear23/ '; to ! AF479323/ SB150-2A1* 99.6 u ! 1444 •H -U tJ G50-TB2d Li 99.4 fd i 1! i 4J 1 rj | Cytoplxaga/Flavabacteriuin/Bacteroides 1 ! 1 < T3 • Haloanella

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Strain j designation*/ Nearest phylogenetic Zee Core j GenBank % neighbor/GenBank Site accession identity6 accession no. no./# o£ i sequenced ntb a proteobacteria Sphingomonas zoali/ 95.1 Y09638 i SIA1K-1A1/ Sphingomonas pruni/ 1 £ 95.0 o n AF395032/ Y09637 1 1390 Sphingomonas sp. from O O 95.0 Antarctica/AF184221 X * High G+C gram positive bacteria : Gordona terrae/ w 99.9 SIA1K-2A1/ X79286 AF479377/ High G+C isolate/ 99.4 1415 X7318 j y proteobacteria U_| Psychrobacter Trans12/ 99.6 submarinus/A J 3 09940 a! AF479327/ Psychrobacter 1426 99.4 w £ marincola/AJ309941 High G+C gram positive bacteria ' ) O « Kocuria kristinae/ i M JJ | 99.7 | Trans4678/ X80749 i CO ^ AF479328/ Kocuria rhizophila/ i o 1434 96.0 Ph Y16264 1 * Isolates are designated by their geographic origin, age of the ice in years or thousands (K) of years, and strain number. (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.)

b The number of 16S rDNA nucleotides sequenced for each isolate.

c The percentage identity with the 16S rDNA sequence of the nearest phylogenetic neighbors.

Table 3.2 (continued) 108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2

d Isolate from this study.

* Isolate from Lake Vostok accretion, ice (see Chapter 5).

e Isolate recovered from a frozen cryoconite hole on the glacial surface.

Table 3.2 (continued) 109

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i Optimum 1 Ice Core Strain growth Growth Antibiotic ! Site designation* rangeb at 4°C resistance6 (°C) y proteobacteria G50-TB2 30-37 I NDd Tet, Amp, Cln Low G+C gram positive bacteria G50-TB5 37-45 “ — G50-TS3 45 - Amp G50-TB9 30 C i G50-L2 37 1 G50-TS4 30 — 1

Table 3.3 Optimum growth temperature range and antibiotic resistance in bacterial strains isolated from ice cores.

110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.3

1 Optimum ; i Zee Core Strain j growth Growth ! Antibiotic ! Site designation*> rangeb at 4°C \ resistance6 ! (°C) i 1 SB150-2A2 j 22-37 + Gnt I SB150-2B ! 22-37 + Ery, Tet, Amp j ■ fd SB12K-9-4 ! 22-37 Amp i •H ■ > SB12K-2-2 i 22-37 + I -H _ iH TD High G+C gran positive bacteria | O ^ SB12K-2-16 ! 22-37 + G n t , Amp j m § SB12K-2-5 ! 22-37 +• iI fd o SB12K-2-7 22 + ” i e u (0 SB12K-2-4 22-37 - G n t , Amp j *n SB12K-6-2 + fd 37 Amp i CO SB12K-2-1 22-37 + i SB12K-6-3 22-37 Gn t , Neo SB12K-2-3 22-37 a proteobacteria fu U M3C4.1K-B5 i 22 ! + ! Art®) j •H 1 j O 4-) y proteobacteria C—1 1 e U M 3 C 4 .1K-B34 | 22-37 + j Ery, Amp j fd O (d Eh Q 4_» M3C4.7K-2 22-37 + Ery, Amp, Neo j a High G+C gram positive bacteria i < 1 SIA1K-2A1 j 22-37 j ND !

* Isolates are designated by their geographic origin, age of the ice in years or thousands (K) of years, and strain number. (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.)

b Optimum growth temperature range determined by qualitatively assessing the growth of isolates on agar- solidifed media at 4°, 15°, 22°, 30°, 37°, and 45°C.

Table 3.3 (continued) 111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.3

c Resistance based on agar diffusion test. Antibiotic disks contained (designation; ng/disk) ampicillin (Amp; 10) , gentamicin (Gnt; 10) , kanamycin (Kan; 30) , erythromycin (Ery; 15), tetracyclin (Tet; 30), and clindamycin (Cln; 2).

d Growth at 4°C not determined, but strains did grow at 15°C.

ND=Not determined.

Table 3.3 (continued)

112

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. isolates were assignable to established bacterial genera,

including Acinetobacter, Agromyces, A r throbacter,

Aureobacterium, Bacillus, Blastomonas, Cellulomonas,

Clavibacter, Detolaasinbacter, Exiguobacterium,

Friedmanniella, Frigoribacterium, Gordona, Kocuria,

Methylobacterium, Microbacterium, Micrococcus,

Micromonospora, Mycobacterium, Nocardia, Nocardioides,

Ralstonia, Paenibacillus, Planococcus, Pseudomonas,

Psychrobacter, Sanguibacter, and Sphingomonas. Five

additional isolates (SB12K-2-1, M3C1.8K-TD7, CanDirtyl2,

CanDirtyl4, and CanDirty89) have >95% identity to their

nearest neighbors, but the taxonomic classifications within

these phylogenetic clades are currently in disarray. They

are composed of multiple genera, unidentified isolates, or

misclassified species (Fig. 3.5, 3.6, 3.8). While fungi

were cultured routinely from ice samples from all

geographical locations, their identification was not

undertaken.

16S rDNA amplification and sequencing

Populations of small subunit rDNA molecules were

amplified directly from material concentrated (Biomax-100

Centricon Plus-80 centrifugation devices; Millipore) from

113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 160 ml of Guliya core section 295 filtrate (material

passing through 0.22 nm filter), but not from DNA

extractions conducted on ceils retained on the 0.22 nm

filter. Amplicons were generated using a combination of

Bacteria-specific and universal primers, but not when

Archaea-specific primers and universal primers were used.

Individual DNA molecules were cloned from these populations

and the sequenced determined for the 16S rDNA region

corresponding to nucleotides 515 through 1392 of the E.

coli 16S rDNA sequence. Eight different Y~Prot®obacterial

16S rDNA sequences were obtained (Fig. 3.9; Table 3.4).

Six of the clones (pG500K-80, pG500K-85, pG500K-86,

pG500K-96, pG500K-98, and pG500K-106; Fig. 3.9) cluster in

the genera Pseudomonas, and share between 99.2-100%

identity with sequences from environmental isolates and

clones, with Pseudomonas putida (D86002) being the closest

related type strain (98.9-99.4% identity) . Two of the 6

clones are most similar to each other, and the remainder

are most similar to either a bacterium from a uranium mine

waste pile (AJ295653) or a Pseudomonas isolate studied for

possessing a unique pyoverdin structure (AF321239) [Table

3.4] .

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.9 Phylogenetic analysis of y-proteobacterial sequences amplified from >500,000 year old ice from Guliya, China. The sequences obtained corresponded to nucleotides 515-1392 of the E. coli 16S rDNA, and were aligned based on secondary structure using the ARB software package (Struck et al. 1998). Trees were generated using a 426 nucleotide mask of unambiguously aligned positions. Clones appear in large font (pG500K-), and glacial isolates are marked with asterisk (*), and designated by their geographical origin (G=Guliya, China; SB=Sajama, Bolivia; M3C=Taylor Dome, Antarctica; and CanClear=Canada Glacier, Antarctica) age of the ice in years, and strain number (e.g. G500K-78 is strain no. 78 isolated from Guliya glacial ice that was 500,000 years old.

A. Maximum likelihood tree generated using fastDNAml (Olsen et al. 1994) . The scale bar indicates 0.1 fixed substitutions per nucleotide position.

B. Maximum parsimony analysis performed with 100 bootstrap replicates, with values shown at the nodes. The scale bar represents 10 evolutionary steps. Of the 72 variable characters in this alignment, 58 were parsimony- informative .

115

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. — Pseudoaooas aeruginosa (038445) Pseudcoonas alcaligenes (Z76653) ' Pseudoaooas oleonrans (D84018) M3C4.7K-2* (AF479376) Pseudcoonas pseudoalcaligenes (768(6) Pseudcoonas aandelii (AF058286) Pseudcoonas sp. froa sea ice (085869) Pseudcoonas sp. frca sea ice (085868) M3C4.1K-B34* (AF479375) Pseudooonas synxantba (AF267911) Pseudcoonas azotofoaans (D84009) - pG500K-96 (AF479387) - uncultured sheep aite bacteriua ' pG500K-106 (AF479389) uraniua line bacteriua XF/6S-J636-1 (AJ295653) pG500K~98 (AF479388) Pseudcoonas sp. PM-2001 (AF321239) - uraniua line bacteriua XF/6S-6itt2-41 (AJ295644) ' Pseudcoonas outida (D86002) ~ DG500K-86 (AF479386) r DG500K-80 (AF479384) X pG500K-85 (AF479385) Acinetobacter radioresistens (X81666) CanClear23* (AF479323) Acinetobacter sp. (Z93445) G50-TB2* (AF479352) SB150-2A1* (AF479373) Acinetobacter junii (X81664) Acinetobacter joimsonii (Z93439) M3C1.8K-TD8* (AF479380) Acinetobacter haeoolyticus (Z93436) ~ Acinetobacter calcoaceticus (AF159045) ~ Acinetobacter lvoffii (010875) Mariana Trench isolate (AB002655) Acinetobacter jobnsonii (X95303) Mariana Trench isolate (AB002658) Acinetobacter sp. froa subsurface (X86572) JSL clone froa uraniua aine (AJ296557) clone froa uraniua aine (AJ301569) Lake Baikal isolate (AJ222834) Acinetobacter lmffi (X81665) _| pG500K-l (AF479382) pG500K“41 (AF479383) Figure 3.9A 116

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pseodaonas u a p m (U38445) Pseodaonas aicaiigenes (Z76653) Psedmu oleonaos. M3C4.7K-2* (AF479376) Pseodaonas psntloilaliguu (7(6(0) ■ Pstadmm m k lii ■ Pseodaonas sp. f a sea ice (U85869) ’ Psea&mus sp. f a sea ice (U85868) H3C4.1K-B34* (inrais) Pseodaonas syzmntba (AF267911) 98 Psesimm uotofomis (D84Q09) pG500K-96 (AF479387) uncnltmed shea aite bacteria pG500K*l06 (AF479389) m a n i a nine bacteria IF/GS-JG36-1 (U295653) 51 52 pG500K-98 (AF479388) Pseodaonas sp. n-2001 (AF321239) m a n i a lioe bacteria KF/GS-€itt2-41 (AJ295844) Pseodaonas patidi (08(002) pG500K“86 (AF479386) (AF479384) 64 pG500K-80 pG500K“85 (AF479385) - IriM+nfor+ir r|tff f fffny -CanClear23* (af<79323) -25- - kioetoticter sp. (193445) -G 50-TB2* (AF4T9352) - SB150-2A1* (AF479373) 100 Jcinetobacter junii (I81((4) Jcinetobacter joinsonii (893439) 50 ------M3C1.8K-TD8* (AF479380) 10 59 ^ icuetobacter kiaolytias (893430) Jcinetobacter calcoaceticns (1F159045) 91 Jdnetobacter lvoffii (010875) Kuiana Trench isolate (UD02055) Jdaetobacter jobmiii (195303)

95 Mariana bench isolate (1BD02058) ----- Acinetabicter sp. f a subsurface done f a ma n i a line (AJ296557) done f a ma n i a nine (U301569) 86 Lake Baikal isolate (U222834) Jcinetobacter ln f f i (281(65)

87 pG500K-l (AF479382) pG500K-41 (AF479383) Figure 3.9B 117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ! GenBank t Sequence i accession Nearest phylogenetic designation | % no./# of neighbor/GenBank (no. o£ j identity0 j sequenced accession no. clones)* i 1 ntb i PG500K-41 I 98.8 clone from uranium PG500K-1 | AF479382/ | 98.8 (6) ; 515 mine / AJ3 015 69 Lake Baikal isolate/ 98.4 AJ222834 • i clone from uranium ! 1 99.2 | mine / AJ3 015 69 I ! PG500K-41 i AF479383/ Lake Baikal isolate/ (2) 498 99.2 AJ222834 PG500K-1 1 98.8 1 PG500K-85 j 99.6 PG500K-80 j AF479384/ PG500K-96 99.5 (2) | 818 Pseudomonas s p ./ ! 99.5 AF321239 i I uranium mine ! 99.6 bac teria/AJ2 95653 i PG500K-85 AF479385/ Pseudomonas s p ./ (10) 819 | 99.6 AF321239 PG500K-80 ! 99.6 uranium mine 99.8 bac teria/AJ2 95653 PG500K-86 AF479386/ Pseudomonas s p ./ (8) 823 99.8 AF321239 | PG500K-96 99.8 i PG500K-98 99.8 uranium mine 99.8 bac teri a/AJ2 95653 PG500K-96 AF479387/ (1) 841 Pseudomonas s p ./ 99.8 AF321239

Table 3.4 16S rDNA molecules amplified from >500,000 year old ice from Guliya, China (Core 2, tube 295). All clones classify within the Y~Prote°bacteria• 118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.4 GenBank Sequence . designation- . . . accession , Nearest phylogenetic neighbor/GenBank % (no. o£ “°-/# °* identity6 clone.)* accession no. nt uranium mine 100 bacteria/AJ295653 PG500K-98 AF479388/ Pseudomonas s p ./ (9) 831 ioo ! AF321239 i PG500K-96 99.8 1 PG500K-96 99.6 i uranium mine PG500K-106 AF479389/ 99.6 | bacteria/AJ295653 (1) : 765 1 Pseudomonas s p ./ 99.6 j AF321239

* Number of individual clones with this sequence

b The number of 16S rDNA nucleotides sequenced for each isolate.

c The percentage identity with the 16S rDNA sequence of the nearest phylogenetic neighbors.

Table 3.4 (continued) 119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Clone pG500K-l and pG500K-41 have 96.3 and 99.2%

identity, respectively, to Acinetobacter lwoffi (X81665).

While pG500K-l is equally similar (98.8%) to pG500K-41 and

to a clone from a uranium mine waste pile (AJ301569) ,

PG500K-41 is 99.2% identical to the latter clone and an

isolate from Lake Baikal (AJ222834) [Table 3.4],

Freeze-thaw resistance

The freeze-thaw tolerance of several non-sporulating

ice core isolates and one endospore-forming Bacillus

isolate was investigated by subjecting cell suspensions to

18 cycles of freeze-thaw, and compared with survival

(ability to form colonies on agar-solidified medium) of

type strains of related bacteria and E. coli (Fig. 3.10).

The glacial isolates and related type strain species tested

were very tolerant to repeated cycles of freeze-thaw,

whereas E. coli was very sensitive, with no cfu remaining

after 18 cycles.

Psychro trophy

Fifty-three ice core isolates from Antarctica,

Bolivia, and China were tested for growth at low

temperature (Table 3.3). The majority were psychrotrophic,

120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 o- 0 4 8 12 16 freeze-thaw cycles

SB150-2A2 - -v - - Bacillus polymyxa SB12K-2-2 - -a - - Aureobacterium suaveolens SB12K-9-4 - - o - - Arthrobacter globiformis -SB12K-2-1 - - Micrococcus luteus SB12K-2-3 - -o - - Brevibacterium linens SB12K-2-16 SB12K-6-3 — o* • * Escherichia coli - SB20K-1

- - ¥ - -G200-A1

Fig. 3.10 Freeze-tolerance was examine in glacial isolates, related species, and E. coli. Cell suspensions were serially diluted and plated on agar-solidifed media during 18 cycles of freezing and thawing.

121

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with >55% capable of growth at 4°C, although most were

isolated, and all grew optimally, at mesophilic

temperatures (>20 °C) .

Discussion

Ice samples from non-polar, low-latitude, high-

altitude glaciers in the Andes and Himalayas generally

contained more colony forming units and a greater variety

of recoverable bacterial species than polar ices. This is

consistent with their closer proximities to locations with

substantial vegetation and exposed soils. Ice core

sections from Taylor Dome and from the Canada Glacier,

adjacent to the McMurdo Dry Valley complex of Antarctica,

similarly contained larger numbers (8-10 cfu ml'1) and

species of recoverable bacteria, than other polar ices

(Table 3.1). Rock grains, eroded by the persistent winds

characteristic to this region, are warmed on ice surfaces

during the austral summer, and melt into the ice creating

pockets of liquid water with sufficient nutrients to

support the growth of microbial communities. Such

ecosystems have been documented in lake ice (Olson et al.

1998; Paerl and Priscu 1998; Priscu et al. 1998; Takacs and

Priscu 1998) and on glacial surfaces (Wharton et al. 1985).

Notably two of the isolates from sediment originating from

122

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a cryoconite hole on the Canada Glacier (Fig. 3.8) belong

to the same genera as many isolates obtained in this survey

of polar and nonpolar glacial ices (Arthrobacter) . For

example, the nearest neighbor of one species recovered from

this polar cryoconite environment (CanDirty7; 98.8%) is an

ice core isolate from Guliya, China.

There is no consistent, monotonic decrease in the

number of recoverable bacteria with increasing age within

an ice core. Rather, the numbers of recoverable bacteria

isolated from ice samples from different positions in ice

cores appear to reflect the prevalent climate and

individual events that occurred at the time of deposition.

For example, there were more recoverable bacteria in Sajama

ice deposited -12,000 years ago during cool, wet climate

conditions than in modern ice, deposited at the same

location during a warmer, dryer period (Thompson et al.

1998). Interestingly, 6 of the 7 Sajama isolates from

modem ice belong to genera that form environmentally-

resistant endospores, whereas only one of the 11 species

isolated from older ice deposited at this location during

the Last Glacial Maximum and deglaciation climate reversal

was a Bacillus relative. Wet climate conditions increase

vegetation density and productivity, increasing the

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. concentration of large airborne biological particles, such

as pollen (Liu et al. 1998), and these, in turn, presumably

transport microorganisms. Increased snowfall accumulation

during this period produced thicker annual layers, which

would have reduced exposure to damaging UV radiation at the

glacial surface. The decreased desiccation rates predicted

under these climate conditions may have also contributed to

survival during travel through the atmosphere.

Cells revived from ice core samples have presumably

endured desiccation, solar irradiation, freezing, a period

of frozen dormancy and thawing. Therefore, it is not

surprising that a large number of the isolates recovered

belong to bacterial groups that form spores or have thick

cell walls and polysaccharide capsules. These structures

help overcome the stresses associated with water loss,

namely increased intracellular solute concentrations,

decreased cell size, a weakened cell membrane, and physical

cell rupture caused by freezing and thawing (Fogg 1998).

The high frequency of pigment production (Fig 3.2) is also

consistent with the need to absorb toxic solar irradiation

and so prevent lethal DNA damage. Even though the

surviving cells may have resistant structures and

protective pigments, during extended periods of inactivity,

124

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. they must still incur some radiation and chemical damage.

Long periods (20-70 days) of incubation were often

necessary before visible colonies appeared, and no colonies

were obtained directly from ancient samples (>500,000 years

old) from the Guilya ice cap. Bacteria were resuscitated

from these samples by inoculation of low-nutrient liquid

enrichments and incubation at 4°C for 30 days. This

observation is consistent with aged cells needing time,

before beginning reproductive growth, to repair cellular

damage accumulated during extended periods of dormancy.

The majority of the glacial isolates obtained have

close phylogenetic relationships to either endospore-

forming Bacilli, spore-forming or non-sporulating

Actinomycetes, several of which are known to have life

cycles with radiation and desiccation resistant resting

stages (Morita 1997), or to species from the a— and y-

proteobacterial lines of descent. Bacillus and

Paenibacillus relatives of strains prevalent in soils were

most commonly isolated from nonpolar glacial ices, and

species of Sphingomonas, Methylobacterium, Acinetobacter,

and Arthrobacter were also ubiquitous, and recovered from

both polar and nonpolar locations. The-nearest neighbors

125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of 14 these isolates were isolates from other ice core

locations, or from other portions of the same core, based

on 16S rDNA identity (Table 3.2, in bold) .

While most of the isolates are similar to species

frequently found in environmental surveys from around the

world, some are most closely related to species of

Arthrobacter, Brachybacterium, Exiguobacterium,

Friedmanniella, Frigoribacterium, Janthlnobacterium,

Planococcus, Pseudomonas, Psychrobacter, and Sphingomonas

recovered previously from Antarctic lake mats (Brambilla et

al. 2001), sea ice (Gosink and Staley 1995; Bowman et al.

1997; Junge et al. 1998), and other predominantly cold

environments (Shi et al. 1997; Benson et al. 2000) . It

seems also noteworthy that relatives of the radiation-

resistant type strains Methylobacterium radiotolerans and

Acinetobacter radioresistens were also commonly

encountered. Having very efficient DNA repair mechanisms

is likely to be valuable in terms of extended survival in

all environments including glacial ice. The isolation of

related microbes from many geographically diverse but

predominantly frozen environments argues that these species

probably have features that confer resistance to freezing

and survival under frozen conditions. The enhanced freeze-

126

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. thaw tolerance (Fig 3.10) and psychrotrophy (Table 3.3)

observed for many glacial isolates is consistent with this

notion.

Interestingly, all the 16S rDNA sequences amplified

from a >500,000 year old ice core from Guliya, China are

most similar to clones retrieved from other subsurface

environments. Comparisons of the numbers of organisms and

types isolated by conventional enrichment strategies with

the numbers and types detectable by non-culture based

molecular approaches consistently demonstrate major

discrepancies between how many and who is chere, with who

grows under laboratory conditions (Hugenholtz et al. 1998).

However, 16S rDNA fragments amplified from the >500,000

year old ice core from Guliya did have DNA sequences

phylogenetically-related on the genus level to species of

Pseudomonas and Acinetobacter (highest identity being 99.0

and 97.6%, respectively) recovered from both polar and

nonpolar glacial ices (Fig. 3.9; Table 3.4).

Identifying microorganisms deposited in glacial ice

has provided an indication of the influence of climate and

geography on the composition of entrapped microbial

species. Investigating the isolates has identified common

survival strategies, and provided data confirming microbial

127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. longevity over thousands of years. These isolates, which

originate from different times in the past, have been

deposited in the DOE sponsored Subsurface Microbial Culture

Collection (Balkwill et al. 1997). They are available for

study at request.

128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4

MACROMOLECUIiAR SYNTHESIS UNDER FROZEN CONDITIONS

Xntroduct ion

For a microorganism to remain viable during periods of

dormancy, the damage incurred to the cell must not exceed a

level where effective repair is no longer possible. Amino

acid racemization rates are retarded in amber (Bada et al.

1994) , and decreased rates of macromoleular decay could

also occur in ice. It is also possible, however, that

microorganisms entrapped within ancient specimens are in

fact active, and able to carry out a low level of

metabolism to facilitate the repair of accumulated

macromolecular damage.

Thin veins of liquid water between ice crystals could

potentially provide a microbial habitat within apparently

solid ice (Price 2000) , and studiies of permafrost (Rivkina

et al. 2000) and surface snow (Carpenter et al. 2000) have

demonstrated low levels of metabolic activity at subzero

temperatures. Therefore, to explore the concept that

129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. glacially-entrapped microorganisms could repair incurred

damage in situ, experiments were undertaken to determine if

macromolecular synthesis could be demonstrated in ice.

Materials and Methods

Bacterial strains and culture conditions

Isolate Transl and E. coli (OSU ref. no. 422) were

grown in Luria-Bertani medium (Sambrook et al. 1989), and

G200-C1 was cultured in R2 medium (Reasoner and Geldreich

1985). All cultures (25 ml) were incubated aerobically

(200 rpm) at 22°C in 125 ml Erlenmeyer flasks. Cells from

late exponential growth phase were used to inoculate

cultures at an initial A6oo of 0.2, which were grown to Agoo

0.6-0.8. Cells were then harvested by centrifugation at

17,000xg for 5 min., washed twice with distilled water, and

resuspended at an A6oo of 0.2 in distilled water

(representing 1.3 and 2 x 108 cfu ml'1 of Transl and G200-C1,

respectively). Aliquots (500 nl)of these cell suspensions

were placed in 1.5 ml Eppendorf tubes and chilled to 4°C.

Procedure for macromolecular synthesis assay

Cell suspensions were maintained on ice and used less

than 1 h after harvesting. When added to cell suspensions,

130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chloramphenicol (Sigma, cat. no. 100K9113), nalidixic acid

(Sigma, cat. no. N-314 3), and ciprofloxacin (ICN

Biomedicals, cat. no. 199020 were used at a final

concentration of 15 ng/ml. In each experiment, a control

was included in which the cell and reaction components were

incubated in the presence of 7% trichloroacetic acid (TCA) .

Either [3H]-thymidine (ICN Biomedicals, catalog# 24060;

1 HCi; final concentration 23 nM) or [3H]-leucine (ICN

Biomedicals, catalog# 20036E; 1 jiCi; final concentration 17

nM) was added to the 500 nl samples. The mixture was then

frozen by incubation at -70°C or by immersing the tube in

liquid nitrogen. The samples frozen in liquid nitrogen

were then placed at -70°C. After 1 h, all tubes were

transferred to -15°C for the duration of most experiments.

In one experiment, the tubes were maintained at -70°C for

100 days.

At each experimental time point, 100 (il of 50% TCA was

added to inactivate the mixture, which was then allowed to

melt. After 30 min at 4°C, the acid-insoluble

macromolecules were sedimented by centrifugation at

18,000xg for 15 min, the supernatant removed, and the

resulting pellet washed with 500 Jil of 5% TCA. Following

centrifugation for 10 min. at 18,000xg, the supernatant was 131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. removed and the pellet washed with 500 |Al of 70% ethanol.

Following centrifugation for 5 min. at 18,000xg, the

ethanol was decanted, and the pellet mixed and resuspended

in 1 ml of Ecoscint H scintillation fluid (Life Sciences

Inc., catalog# LS-275). The Eppendorf tube was placed

directly into scintillation vial, and tritium incorporation

was quantitated by 10 min counting in a Beckman model LS-

7500 scintillation counter. Control experiments

demonstrated that 3H was measured at a 61% counting

efficiency.

Results

The possibility of incorporation of [3H]-thymidine and

[3H]-leucine during the freezing process was investigated by

comparing incorporation in samples frozen at -70°C with

samples frozen by submersion in liquid nitrogen (Fig 4.1).

The background level of [3H]-thymidine and [3H] -leucine

associated with TCA-prefixed cell suspensions was lower

than the incorporation obtained from samples frozen

instantly in liquid nitrogen after the addition of label,

and there was no clear difference between the incorporation

by cells frozen by the two methods, even though samples

placed at -70°C remained liquid for 5 to 10 min. before

132

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2450

Transl Transl G200-C1 G200-C1 TdR Leu TdR Leu

Figure 4.1 Incorporation of [3H]-thymidine (TdR) and [3H]-leucine (Leu) into TCA-precipitable material by strains Transl and G200-C1 during the freezing process.

The acid-insoluble macromolecular fraction was precipitated 1 h after freezing. The incorporation in 3 separate reaction mixtures was determined for every data point, including samples frozen in the presence of 7% TCA, frozen by submersion in liquid nitrogen (L-N2) , and frozen at -70°C. The y axis error bars denote standard deviation from the mean. 133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. freezing solid. This suggests that the difference between

the samples and the TCA controls results from acid

conditions effecting chemical binding, rather than the

difference representing incorporation during freezing.

Sampling during the course of a -200 day incubation at

-15°C (Fig 4.2) clearly indicated that both DNA and protein

synthesis occurred during the first 20 days of incubation

in both strains studied, which appeared to continue for the

next 13 0 days, although at reduced rates. A similar

incorporation profile of these precursors into TCA-

precipitated materials was also observed in frozen E.coli

cell suspensions incubated for 100 days under identical -

15°C conditions (Fig 4.3).

Precursor incorporation began to level off after 150

days at -15°C (Fig 4.2), where the maximum amount of

incorporation was observed, ranging between 9-14 x 103 dpm.

After subtracting background counts (isotope count in

samples TCA fixed from the outset) from these values, the

estimated incorporation range for thymidine and leucine was

5850-9690 dpm. Knowing the specific activity of the

precusor (60-90 Ci/mmol of [3H]-thymidine and 40-60 Ci/mmol

of [3H]-leucine) and given that all cells (107-108 cfu ml-1)

in each sample participate in the DNA and protein

134

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200 150 TdR-TCA TdR Leu-TCA d a y s 100 G200-C1 G200-C1 Leu 50 — □ — Transl Leu • Transl ■ Tdr-TCA • ■ • G200-C1 ■ — ■ — Transl TdR — • — G200-C1 • • Transl Leu-TCA G200-C1 • O- - - 6 10 14 - m label. label. The y axis error bars denote standard deviation from the mean. Fig. Fig. 4.2 triplicate, Incorporation and included of identical and [3H]-thymidine (TdR) samples byprefixed [3H]-leucine (Leu) strains in 7% TCA prior to the addition of Transl and G200-C1 during a 206 day incubation at -15°C. Each data point was done in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. coli TdR E. coli Leu coli TdR-TCA • E. coli Leu-TCA

.5 x 10'

5000 -

0 20 40 60 80 100 days

Figure 4.3 Incorporation of [3H]-thymidine (TdR) and [3H]-leucine (Leu) by E.coli during a 102 day incubation at -15°C.

Two samples were taken at each data point. These included samples prefixed in 7% TCA prior to the addition of radioactively labeled precursor. The y axis error bars denote standard deviation from the mean.

136

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. syntheses, then it can be calculated that between 25-500

molecules of thymidine and leucine were incorporated per

cell after 206 days of incubation (for 107-108 cells, 1000

dpm represent 5-50 molecules of thymidine or leucine/cell).

The change in the number of cfu ml-1 was determined

over 100 days concurrently with measurements of [3H] -

thymidine incorporation (Fig 4.4). The results indicate

that the low level of isotope incorporation observed during

this period of incubation was not accompanied by an

increase in the viable number. The number of cfu ml'1

actually decreased >4-fold during the first 20 d of G200-C1

incubation. Unfortunately, due to a technical error, data

on the 2 day time point are the first available for Transl.

For both isolates, the number of cfu ml'1 decreased to a

minimum between 10-25 days post-freezing, and then

increased, appearing to reach a steady-state by 50-100

days, as also observed for thymidine incorporation (Fig

4.4) .

Addition of ciprofloxacin (CFX) and chloramphenicol

(CM) decreased levels of precursor incorporation (Fig 4.5).

The presence of CFX, an inhibitor of DNA gyrase, reduced

thymidine incorporation by 30-40% (Fig 4.5A), and CM, an

inhibitor of protein synthesis, reduced leucine

137

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 C 3 (— O Mi 3 3 x 107 4 4 x 107 2 5 x 5 107 - 1 1 x 107 - - - 2 2 x 107 - 100 80 • • • • G200-C1 • cfu ml • • • • Transl • cfu ml" days 40 20 Transl-TdR G200-Cl-TdR x x x x series, series, due to a technical error. the mean. The y axis error bars denote standard deviation from triplicate. triplicate. For value Transl, represents the 2 day cfu the ml'1 first data point in the and G200-C1 during a 100 day incubation at -15°C. Data at each point was obtained in Fig. Fig. 4.4 Incorporation for Transl of [3H]-thymidine and1 (TdR) the number of cfu ml" U> 00

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.5 Incorporation of [3H]-thymidine (TdR) and [3H]-leucine (Leu) at -15°C by strains Transl and G200-C1 over 23 days in the presence of (A) ciprofloxacin and (B) chloramphenicol.

Data at each point were obtained in duplicate, and the y axis error bars denote standard deviation from the mean. (A) ciprofloxacin [CFX] and (B) chloramphenicol [CM] were added at a concentration of 15 fig/ml, before the addition of radioactively-labeled precursor. Cell suspensions with (.... ) and without (______) antibiotics, and prefixed in TCA (____) were frozen an incubated under identical conditions.

139

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. X > — Transl TdR G200-C1 TdR Transl TdR/CFX •#••• G200-C1 TdR/CFX O - Transl TdR-TCA O - G200-C1 TdR-TCA x

£ a , T3 X

0 4 812 16 20 24 d a y s

10 ‘ Transl Leu ♦ — G200-C1 Leu ■■••• Transl Leu/CFX G200-C1 Leu/ CFX - Transl Leu-TCA O • G200-C1 Leu-TCA

1 X

8 x £ O-. T 3

6 x

4 x

2 x

0 4 8 1216 20 24 d a y s F i g u r e 4.5 140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. incorporation 50-60% (Fig 4.5B) . Curiously for both Transl

and G200-C1, incubations in the presence of a second DNA

synthesis inhibitor, nalidixic acid, resulted in a 2-fold

increase in thymidine incorporation (data not shown) .

Based on growth, G200-C1 was resistant to this antibiotic

at concentrations of 15 Jig/ml, but Transl was inhibited by

nalidixic acid.

In contrast to the incubations at -15°C, neither

thymidine nor leucine incorporation was detected in cells

incubated at -70°C for 50 days (Fig 4.6) . In a repetition

of this experiment, no incorporation was observed at -70°C

after 100 days (data not shown) .

Discussion

Having demonstrated that microorganisms remain viable

during entrapment in glacial ice for hundreds of thousands

of years, the question remained whether metabolic activity

was possible under these frozen conditions. Alternatively,

the species recovered may be particularly successful at

surviving although metabolically dormant over extended time

frames. The presence of liquid water is generally

considered essential for active metabolism, and within ice,

films of water exist on salt inclusions, air bubbles, and

141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.6 Incorporation of [3H]-thymidine (TdR) and [3H]-leucine (Leu) at -15 and -70°C by strains Transl and G200-C1 over 50 days.

Duplicate samples were taken for each data point, and the y axis error bars denote standard deviation from the mean. Aliquots of the same cell suspension were frozen and incubated at either -15° or -70°C, and samples frozen in the presence of 7% TCA (background) were incubated at -15°C. The effects of temperature on both the incorporation of thymidine (A) and leucine (B) are illustrated.

142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■ ■ ' Transl TdR -15°C G200-C1 TdR -15°C Transl TdR -70°C G200-C1 TdR —7 0°C — O - Transl TdR-TCA O - G20 0-C1 TdR-TCA

10 -

6 x 103-

m •JJ

0 10 20 40 5030 B d a y s B Transl Leu -15°C G200-C1 Leu -15°C x ■■••• Transl Leu -70°C ••••• G200-C1 Leu -70°C D- - Transl Leu-TCA O- - G20 0-C1 Leu-TCA ,

10 '

x a. 'CS

0 10 20 30 40 50 d a y s F i g u r e 4 . 6

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between ice crystals (Patterson 1994; Price 2000). Water

has been shown to exist in permafrost at temperatures above

-60°C (Ostroumov and Siegert 1996). Terrestrial glacial ice

is generally at temperatures above this value, and

therefore the possibility exists that microorganisms could

be metabolically active within liquid water in veins

between ice crystals, as proposed by Price (2000).

The two glacial isolates used in this study [Transl

(Fig 3.6) and G200-C1 (Fig 3.8)] were chosen because of

their apparent close phylogenetic relationships to species

native to brine channels in sea ice [>99% 16S rDNA identity

to isolates reported by Junge et al. (1998) and Bowman et

al. (1997)], demonstrated ability to grow at 4°C

(psychrotrophic) , and significantly different cell wall

structure (Transl is a Gram negative Psychrobacter species,

whereas G200-C1 is a Gram positive Arthrobacter species) .

Incorporation of radioactive precursors in TCA-precipitated

material was investigated for cell suspensions held under

frozen conditions (-15°C) for -30 weeks (Fig 4.2). In both

cases after an initial increase, the rate of incorporation

gradually slowed and appeared to reach a steady-state after

200 days (Fig 4.3). The inhibition of precursor

incorporation by CFX and CM was consistent with

144

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. macromolecular synthesis occurring at -15°C (Fig 4.5). By

definition, unfrozen water is predicted not to exist below

-60°C in water ice or within cells (Ostroumov and Siegert

1996) . Consistent with liquid water being present in

samples at -15° but not -70°C, no thymidine or leucine

incorporation into the macromolecular TCA-precipitated

material was observed in cell suspensions at -70°C.

The cells used in these frozen activity studies were

taken from populations in logarithmic growth. Cell

division will have stopped, but some metabolic activity

within the cell will have continued. The difference

between decreased anabolism and continuing catabolism

results in the production of free radicals that damage DNA

and protein (Aldsworth et al. 1999; Stead and Park 2000) .

Aldsworth et al. (1999) have termed this the "suicide

response", and have shown that in contrast to cells

suddenly removed from a growth situation, cells from non

dividing cultures in stationary phase are resistant to this

stress. Their already reduced metabolism and stationary

phase-induced stress proteins apparently protect such cells

from incurring lethal damage.

145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In addition to the physical and osmotic stress imposed

on cells by ice crystal formation, the storage of E. coli

at -20°C has been demonstrated to cause single- and double

stranded chromosomal breakage (Alur and Grecz 1975; Grecz

et al. 1980) . Although substantial damage occurs during

the freeze process, the extent of double-stranded DNA

breakage increased over 4 months, but then began to

decrease by 12 months post-freezing. The authors concluded

that "random reassociation and aggregation of the initial

DNA fragments" was the most likely explanation for the

unusual results (Grecz et al. 1980).

The data obtained are consistent with macromolecular

synthesis being possible when bacteria are within ice at

temperatures in which liquid films exist. Inability to

grow at subzero temperatures does not preclude metabolic

activity at such temperatures. E. coli was unable to grow

below 10°C, but nevertheless, significant precursor

incorporation was observed by suspensions of E. coli held

at -15°C (Fig 4.3). The total incorporation of thymidine

and leucine into macromolecules observed after >200 days at

-15°C (Fig 4.2) corresponded to one cell assimilating

between 25-500 molecules of each labeled precursor. Based

on E.coli's genome and average protein size, this amount of

146

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. activity represents <0.01% of the genome or a total of 10

protein molecules. Interestingly, 50-75% of this

incorporation occurred during the first 15-20 days of

incubation at -15°C; coincidentally, a period of time the

viable count (cfu ml"1) began to increase after initially

decreasing after freezing (Fig 4.4). The amounts of

precursor incorporation detected were insufficient for

growth, and more likely represent incorporation during

cellular repair of damage incurred during the stress of

metabolic arrest and/or freezing. These results obtained

add to the growing body of evidence for metabolism in

environments below freezing (Carpenter et al. 2000; Rivkina

et al. 2000). They support the argument that bacteria

could continue with metabolic activity while entrapped in

water veins within glacial ice.

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5

ISOLATION OF BACTERIA AND 16S rDNA SEQUENCES FROM LAKE VOSTOK ACCRETION ICE

Introduction

More than 70 subglacial lakes have been discovered in

Antarctica (Siegert et al. 1996). The largest, Lake

Vostok, is covered with a layer of -4000 m of glacial ice,

and has been isolated from direct surface input for at

least 420 K years (Petit et al. 1999) . The lake water is

derived from the overlying glacier, with ice melting into

Lake Vostok at the northern ice-water interface, and water

from the lake freezing as accretion ice, below the glacial

ice, over the central and southern regions of the lake

[Fig. 5.1; (Kapitsa et a l . 1996; Jouzel et a l . 1999;

Siegert et al. 2000)] . Glacier movement presumably must

transfer sediment from the adjacent bed rock into the lake,

and both eukaryotic and prokaryotic microorganism have been

detected in samples of glacial ice collected from above

Lake Vostok (Abyzov et al. 1998) . It therefore seems

inevitable that viable microorganisms are seeded into Lake 148

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 5.1 Origin of deep Vostok ice core section 3593.

Schematic illustrating subglacial Lake Vostok, based on Bell (1998) and Siegert (2000) . Chemical and isotopic profiles established that the glacial ice-accretion ice interface occurs at <3540 mbs below the surface (Jouzel et al. 1999), and the deep Vostok ice core section used in this study, section 3593, originated from 3591.965 to 3592.445 mbs. As illustrated, glacial ice melts into Lake Vostok at the ice-water interface in the north, and accretion ice accumulates at the base of the glacial ice over the central and southern regions. Radar measurements have detected the presence of a layer of sediment below the lake water (Kapitsa et al., 1996), with gas hydrates also predicted to be present (Doran et al. 1998) .

149

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vostok

Gore 3591.965- 3592.445 m

220 m of Accreted Ice -500 m of Lake Water

Figure 5.1 150

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vostok, but the nature of the environment and ecosystem

within Lake Vostok remain uncertain. Concerns for

contamination have resulted in a moratorium on direct

sampling of Lake Vostok water and ice core drilling has

been terminated above the ice-water interface. An ice core

has, nevertheless, been retrieved in which the bottom ~150

meters are accretion ice (Fig. 5.1) and this therefore

provides a sample of Lake Vostok water (Petit et al. 1999).

Microbial cells in melt water from sections of this

accretion ice core that originated 3590 and 3603 meters

below the surface (mbs) have been detected by

epifluorescence and scanning electron microscopy (Priscu et

al. 1999; Karl et al. 1999), and seven small-subunit

bacterial ribosomal RNA-encoding DNAs (16S rDNA) were

amplified from the 3590 melt water that originated from a-

and P-proteobacteria and from an Actinomyces (Priscu et al.

1999) . Evidence for respiration was also obtained by

measuring 14C-C02 release during incubations at 3°C and 23°C

after the addition of 14C-acetate or 14C-glucose to melt

water from the 3603 section (Karl et al. 1999) . However,

there was very little, if any, 14C-incorporation into

macromolecules.

151

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This chapter documents the results of experiments

undertaken to determine if viable bacteria could be

recovered directly from Lake Vostok accretion ice. Four

different isolates have been obtained, and additional 16S

rDNAs have been amplified from a section of the accretion

ice core that originated at 3593 mbs. The results are

consistent with the concept that Lake Vostok is seeded

regularly with bacteria initially immured in the overlying

glacial ice, and is likely to contain bacteria similar to

species found in other cold environments.

Materials and Methods

Ice core origin and sampling

A section of the deep Vostok ice core extending from

3591.965 to 3592.445 mbs, here designated core section 3593

(Fig. 5.1), was obtained from the National Ice Core

Laboratory (Denver, CO). Core 3593 was broken in transit,

and the longer resulting section (~33 cm) was subjected to

automated melt water sampling, and the smaller fragment

washed with 95% ethanol and water, as described in Chapter

2 .

152

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Measurements made using a Finnigan Mat mass

spectrometer, revealed that the mean stable isotope ratios

in the core were -56.238b for 5180 and -44 6 . 248b for 5d

(Henderson et al. 1999), similar to the values reported for

samples from 3540 to 3750 mbs by Jouzel et al. (1999), and

consistent with a water-freezing-to-ice origin.

Microbiological and molecular procedures

Enrichment cultures, colony isolations, 16S rDNA

amplification and sequencing, and electron microscopy were

conducted as described in Chapter 3.

Results

Enrichment isolates

Melt water that was obtained from inside Vostok deep

core section 3593 using the automated ice core sampling

system was used to inoculate a wide range of different

growth media (Table 5.1) . After 7 days, growth was

observed in M9 glucose-minimal salts (Sambrook and Maniatis

1989) and in R2 medium, a low nutrient medium designed to

recover stressed bacteria (Reasoner and Geldreich 1985), in

cultures incubated aerobically at 25°C. Single colony

isolates were obtained from these enrichment cultures by

153

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5.1 Media inoculated with melt water from ice core section 3593. When used in petri plates, media were solidified by the addition of 1.5% (w/v) agar.

Used under aerobic conditions* Designed to Medium Source isolate/enumerate Tryptose blood Fastidious Difco, Inc. agar base mi croorgani sms Nutrient agar Oligotrophic Difco, Inc. diluted 1:100 microorganisms R2A Hetero trophi c Reasoner and microorganisms in Geldreich, potable water 1985 Actinomycetes Actinomycetes from Difco, Inc. isolation agar soil or water Low tryptone Cytophaga/ E. yeast extract Myxobacteria Leadbettera M9 minimal Auxotrophic mutants of Sambrook et E. coli al., 1989 Ammonia and Methylotrophs Patt et al. nitrate minimal 1974 salts medium Used under anaerobic conditions! Medium Designed to Source isolate/enumerate Nutrient agar Heterotrophic Difco, Inc. diluted 1:100, nitrate-reducing/ R2A, M9 minimal denitrifying bacteria with 50 mM KN 0 3 - Basal salt medium Methanogenic archaea C.M. Plugge3 with H2/C02, methanol, acetate, and fructose Basal salt medium Acetogens and C.M. Plugge3 with H 2 /CO2 , sulfate-reducers methanol, acetate, fructose, BESb, and 20 mM Na 2S0 4 C a 1999 Microbial Diversity Course, Woods Hole, MA. b BES (bromoethanesulfonic acid) inhibits the growth of methanogenic archaea at a concentration of 5 mM. c N a 2S0 4 was added to cultures to facilitate the growth of sulfate-reducing bacteria. 154

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. plating on agar-solidified M9-glucose and R2 media. Three

isolates from the M9-enrichment culture, designated V15,

V18 and V19, that formed colonies with reproducibly

different morphologies were further investigated. They

were all resistant to fl-lactam antibiotics, and sensitive to

gentamycin, tetracyclin and neomycin but V19 alone was also

resistant to erythromycin. Surprisingly, all three had the

same 16S rDNA sequence, indicating a close phylogenetic

relationship to Brachybacterixxm conglomeration (Table 5.2;

Fig. 5.2A), and all of the single colony isolates

investigated from the R2 enrichment culture similarly also

had this Brachybacterium-related 16S rDNA sequence.

Media inoculated with core section 3593 melt water and

incubated for at 4° C under conditions used to enrich for

acetogenic bacteria and methanogenic archaea, appeared to

contain growth after 9 months. These cultures contained

very long filaments, observed initially by light microscopy

and then confirmed by SEM (Fig 5.3), ranging from 100 to

300 /zm in length, that were not present in cultures

incubated at 25°C or in negative control cultures. To date

analyses have not detected methane or organic acid

155

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5.2 Bacteria isolated from deep Vostok ice core section 3593.

Sequence ali.gnaent Nearest GenBank pbylogenet ic Isolate accession neighbor (GenBank

number # of accession no; base % identity1* origin) pairs* Brachbac teri um V I 5 AF324202 1454 99.4 congl omera turn (X91030; cheese) Guliya 500K-14 99.8 (AF395037; Guliya ice core) V 2 1 AF324199 1409 Sphingomonas sp. 99.0 (AB033945; not available) Paenibacillus V 2 2 AF324200 1485 99.2 amylolytlcus (D85396; soil) Unidentified bacterium 99.3 (AU223453; not available) V 2 3 AF324201 1406 Afe thyl obac teri um sp. (Z23156; 98.2 biofilm on cooling fan)

a The number of 16S rDNA nucleotides used for the alignment. b The percentage identity with the 16S rDNA sequence of the nearest phylogenetic neighbour.

156

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 5.2 Phylogenetic analysis and scanning electron micrographs of bacterial isolates from core section 3593.

A. The 16S rDNA sequences obtained from single-colony isolates, corresponding to nucleotides 2711492 of the E. coli 16S rDNA, were aligned based on secondary structure using the ARB software package (Strunk et al., 1998), and a phylogenetic tree was created with maximum likelihood using a 1321 nucleotide mask of unambiguously aligned positions and using FASTDNAML (Olsen et al., 1994). Bootstrap values generated from 100 replicates using the maximum parsimony method are shown at the nodes. Evolutionary distance is defined as the number of fixed nucleotide changes per position. The scale bar indicates 0.1 fixed substitutions per nucleotide position. Isolates V21 and V23 position within the a-subdivision (a) of the proteobacteria, V15 in the high-G+C-containing Gram- positive (GP) group, and V22 in the low-G+C-containing GP group. GenBank accession numbers and the percentage identity of the corresponding 16S sequence with that of the most similar Lake Vostok isolate are listed in parentheses. Ice core isolates from Guliya (China) and Sajama (Bolivia), and from Taylor Dome (TD), Canada glacier (CanClear) and Siple Dome (SIA) in Antarctica are listed. These are designated by their geographic origin, age of the ice in years or thousands (k) of years and strain number, e.g. Guliya500k-78 is strain no. 78 isolated from Guliya glacial ice that was <500 000 years old.

B. Scanning electron micrographs of cells from cultures of V15, V21, V22 and V23.

157

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. High G+C GP G+C High LowG+C GP

(X%) 9 1657,97.1 98.2%)

(X91030,99.4%)

(X91032,

95.7%) sp. in sea ice brine (AF041790,97.0%) brine ice in sea sp.

(D85396,99.2%)

(D85397,96.5%)

(AJ223991,97.6%)

99.3%)

(AJ250801,

(AF324202) 98.2%) 95.6%) 94.9%) Brachybacterium M Brachybacterium tyrofermentans MBrachybacterium Brachybacterium faecium Brachybacterium

(AF324200) V'-Brachybacterium conglomeration V'-Brachybacterium 94.5%) 6 9 too *— Guliya200-C (AF395028,94.2%) 15 *— ,ool)V15 ,ool)V15 97.1%) sp. (Z23156,sp. Bacillus longisporus Bacillus

ifloj— Sajama 100-2B (AF395029,93.6%) Sajama ifloj— ------rPaenibacillus amylolyticus rPaenibacillus -Guliya50-TB9 (AF395027,98.1%) — LPaenibacillus illinoisensis LPaenibacillus jji— Guliya500k-78 (AF395033,93.8%) |V22 90 lool 100 (AF324201)

unidentified bacterium (AJ223453, bacterium unidentified rV23 rV23 ^Methylobacterium fujisawaense ^Methylobacterium *Guliya500k-15 (AF395034, ioorGuliva500k-5 (AF395035, M-Methylobacterium 99.0%) 99.0%) too JM-TD1.8IC-4 (AF395030, 87 99.8%) too sp. Antarcticfrom (A soil sp. F184221,

Guliya500k-3 (AF395036,92.8%) too sp. (AB033945,sp.

CanClearl (AF395038,94.5%) -SIAlk-lAl (AF395032,94. 76 (AF324199) Sphingomonas 57 68 yTD4.2kB-5 (AF395031,97.4%) Sphingomonas Guliya500k-14 (AF395037, V21

9 0 60 cn h H* CD CD C iQ

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V21

V23

Figure 5.2B 159

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.3 Filaments in anaerobic enrichments. After 9 months of incubation at 4°C, filaments ranging in length from 100 to 300 |im were observed in several anaerobic enrichment cultures, but were not present in identical cultures incubated at 25°C, nor in negative control cultures. No methane or organic acid production was observed, and attempts to isolate these filaments failed 160

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. production, indicative of methanogenic and acetogenic

activity, respectively, and attempts to isolate and

characterize these filaments were not successful.

Colony isolations

Particulates, collected by filtration from 100-150 ml

of melt water, were resuspended at ~30-fold the original

concentration, and aliquots (200 Ail) of the resuspended

materials were spread on the surface of agar-solidified

media. A total of three colonies were obtained, all on

agar-solidified R2 medium on plates incubated aerobically

at 25°C. The growth of colonies was never observed on any

other agar-solidified medium (see Experimental procedures) ,

even though the plates were incubated for >3 months at both

4°C and 25°C. On subculture, all three isolates, designated

V21, V22 and V23, grew most rapidly on R2 medium at 25°C,

although V22 and V23 also grew at temperatures as low as

4°C, and V21 grew as low as 10°C. Based on their 16S rDNA

sequences, V21, V22 and V23 are most closely related

phylogenetically to Sphingomonas, Paenibacillus and

Methylobacterium species, respectively (Table 5.2; Fig.

5.2A) .

161

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16S rDNA amplification and sequencing

Populations of small subunit rDNA molecules were

amplified directly from core 3593 melt water by using

universal and Bacteria-specific primers, but not when

Archaea-specific primers were used. Individual DNA

molecules were cloned from these populations and sequenced,

revealing the presence of bacterial 16S rDNAs from five

different phylogenetic lines of descent (Table 5.3; Fig.

5.4). Sequence pA419 originated from an a-proteobacterium

whose nearest cultured neighbor was isolated from Lake

Baikal in Russia, and pA419 is also -86% similar to a 16S

rDNA sequence retrieved by Benson et al. (2000) from an

Antarctic lake. Although perhaps not so striking, in terms

of such very cold freshwater environments, the other 16S

rDNAs amplified from the 3593 melt water also have

freshwater-isolate relatives. Specifically, sequences

pA3178 and pA42B412 are from P-proteobacteria and are most

similar to the 16S rDNA sequences of an Aquabacterium and a

facultative hydrogen-autotroph, formerly designated

Pseudomonas saccharophila, respectively. Sequence pA47 is

93.6% identical to that of the 16S rDNA of Sphingobacterium

heparinum, a member of the Cytophaga/Flavobacterium/

Bacteroides lineage, and sequences pD12 and pD4 are 93.5%

162

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5.3 16S rDNA molecules amplified from core 3593 melt w a t e r .

Sequence Nearest alignment Clone GenBank phylogenet ic (no. accession neighbor (GenBank % obtained)a number # of accession no; base identity0 origin) pairsb Aquabacterium sp. PA3178 (AF089858; AF324205 842 98.7 (4) drinking water biofilm) Lake Baikal pA419 isolate (AJ001426; AF324207 896 96.4 (3) 400 mbsd in Lake Baikal, Russia) Pseudomonas PA42B412 sa ccharophi lia AF324206 839 99.0 (12) (AB021407; not available) Sphingobacteri um pA47 (1) AF324208 827 93 .6 heparinum (Ml 165 7; not available) Rubrobacter xylanophilus AF324203 845 98.9 pD4 (2) (AJ243871; hot spring) Alkalibacteri um olivoapovlitic AF324204 838 93 .5 pD12 (1) (AF143513; olive wash water)

a The number of individual clones sequenced that had this sequence.

b The number of 16S rDNA nucleotides used for the alignment.

c The percentage identity with the 16S rDNA sequence of the nearest phylogenetic neighbour.

d mbs=meters below surface

163

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 5.4 Phylogenetic analysis of the 16S rDNAs amplified, from core section 3593.

Sequences, PCR-amplified from melt water, that correspond to nucleotides 515-1392 of the E. coli 16S rRNA-encoding gene were designated pA419, pA3178, pA42B412, pA47, pD4 and pD12. The phylogenetic tree was created with maximum likelihood using an 813 nucleotide mask of unambiguously aligned positions (Olsen et al., 1994) . Bootstrap values generated from 100 replicates using the maximum parsimony method are shown at the nodes. Evolutionary distance is defined as the number of fixed nucleotide changes per position. The scale bar indicates 0.1 fixed substitutions per nucleotide position. GenBank accession numbers are provided in parentheses.

164

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. eu p. O o o V -t- o u £ ■§J CO. ►3 £

5 -g «*. ^^jofrccfm irw AfpcrMir« (M 11657) aSs S's-i’S ' s J S S c->S S— ae J Si 46 "I 5 S£- « m , 1 s t =t l i t 3 *3* ®> si- *ss § V

W-"

?. S S' _ s* E S 53 -S ® ^ f S 8 1 a * is i as t a I ? ? U 2

F i g u r e 5.4 165

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a n d 98.9% identical to the 16S rDNA sequences of

Alkalibacterium olivoapovlitic and Rubrobacter

xylanophilus, respectively, positioning them within the

low- and high-G+C Gram positive groups.

Extensive precautions were taken, but the

possibilities of contamination and of DNA molecules being

generated artifactually during the PCR amplifications were

always serious concerns. Infrequently, an amplicon was

generated in a negative control reaction, and when these

were cloned and sequenced they had sequences almost

identical to a 16S rDNA sequence that has been shown

previously to arise in PCR controls (Cisar et al., 2000).

This sequence (GenBank Accession number AF195876) is

related to 16S rDNA sequences from y-proteobacterial

pseudomonads and is not closely related to any of the

experimental sequences used in constructing Figs 5.2 and

5.4.

SEM analysis of Lake Vostok Accretion ice

Scanning electron microscopic analysis of 0.2 um

filtered accretion ice section 3593 revealed the presence

of several morphotypes (Fig 5.5), and little, if any

166

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. l l m m Figure 5.5 Scanning electron micrographs of apparently prokaryotic cells, retained on the surface of a a 0.2 pm isopore (Millipore) filter after concentration of core section 3593 melt water.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inorganic or organic detritus. Too few cells were present

on the filter sections analyzed to obtain a reliable cell

c o u n t .

Discussion

Based on the estimates of Siegert et al. (2000), the

48 cm ice core (section 3593) sampled in this study

accreted over a period of 10-25 years. It did not contain

any macroscopically-visible solid inclusions, and therefore

most likely formed over a relatively deep portion of the

lake (Jouzel et al. 1999). Scanning electron microscopy of

materials filtered from core 3593 melt water revealed

little inorganic debris, and although sparsely distributed,

a number of particles were identified with sizes and

morphologies consistent with bacterial cells (Fig 5.5).

Epifluorescence microscopy of DNA-stained samples also

revealed the presence of low cell numbers (2.3 x 103 and 2-3

x 102 cells m l ' 1) in melt water from flanking Vostok ice

core sections 3590 and 3603, respectively (Priscu et al.

1999; Karl et al. 1999). Presumably, therefore, only a

very small percentage of the cells present in core 3593

were recovered, and it is perhaps noteworthy that V21, V22

and V23 are related, although not identical, to species

168

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. recovered previously from both polar and non-polar glacial

ices (Chapter 3). These and related isolates survive

repeated cycles of freezing and thawing, even though V22 is

the only member of a bacterial group (Paenibacillus) t hat

is known to differentiate into cells (endospore) that

specifically facilitate airborne transport, resist

desiccation and provide long-term survival under non-growth

conditions (Cano and Borucki 1995; Vreeland et al. 2000) .

The results obtained predict that representatives of

at least five bacterial lineages are likely to be present

in Lake Vostok, some of which are related, in terms of 16S

rDNA sequences, to isolates from other cold, potentially

very similar environments. For example, sequence pD12 is

-92% identical to 16S rDNA sequences from two

Carnobacterium species that were isolated from ice-covered

Antarctic lakes (Franzmann et al. 1991; Bratina et al.

1998), and sequence pA419 clusters both with the 16S rDNA

sequence of an isolate recovered from 400 mbs of Lake

Baikal in Siberia and the sequence of an amplicon from a

frozen Antarctic lake (Fig. 5.4). Extrapolations from

rDNA sequence similarities to similarities in life style

and physiology are clearly very tenuous, but these results

169

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. do argue that Lake Vostok probably contains bacteria

similar to species found in other permanently cold

environments.

170

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 6

GENERAL DISCUSSION

The microorganisms recovered in this study from

glacial ice have endured desiccation, solar irradiation,

freezing, an extended period of no growth, and subsequent

thawing. It is therefore not surprising that m a n y of these

isolates belong to bacterial groups that form spores (Fig

3.4, 3.7, and 3.8), structures known to confer resistance

to environmental abuses. Many also have thick cell walls

or polysaccharide capsules, and resist repeated cycles of

freezing and thawing (Fig 3.10). The frequent isolation of

related genera from geographically different ice core sites

and from other frozen or permanently cold environments (Fig

3.5-3.8) suggests these are species adapted to surviving

freezing, and that persist under cold and non-growth

conditions. Members of the genera Sphingomonas,

Acinetobacter, and Arthrobacter were often isolated from

glacial samples, and these are also the most frequently

isolated genera in enrichment surveys of terrestrial

171

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subsurface environments (Balkwill et al. 1997). This is

consistent with these genera having many species that can

survive for extended times under low nutrient, non-growth

conditions, and that similar survival strategies are in

effect in ice and in deep subsurface situations.

Ice cores from non-polar, low-latitude, high-altitude

glaciers generally contained more colony forming units

(Table 3.1) and a greater variety of bacterial species (Fig

3.4) than polar ices. Similarly, the highest recovery of

viable bacteria from polar ice cores was obtained from

Antarctic regions adjacent to exposed soils and rock

surfaces of the McMurdo Dry Valley complex. These results

are consistent with increased microbial deposition in

glaciers contiguous to environments that supply airborne

rock grains, soils, and biological particles. Not

surprisingly, the highest numbers of bacteria were isolated

from sections of ice cores that were "dusty", visibly

contaminated with macroscopic debris that presumably would

have transported and protected attached bacteria. Abyzov

et al. (1998) also reported a correlation between the

number of total cells and the concentration of dust in the

Vostok core.

172

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Analyses of DNA isolated directly from glacial ices

revealed that only non-polar ices containing

macroscopically visible particles possessed sufficient

biomass for DNA detection by slot blot hybridization

assays, and therefore nucleic acid-based quantitation

techniques and direct cell counting were not feasible. The

information gathered on the microbiological content of

glacial ices was determined by more sensitive means,

specifically enrichment culturing (which can detect 1

viable cell) and the PCR. Even with the amplification

potential of the PCR, isolates were often obtained from ice

samples for which no amplicons were subsequently generated

when extracted DNA was amplified from cells concentrated on

0.2 um filters. The very low biomass and possible presence

of PCR-inhibiting substances (Wilson 1997; Wintzingerode et

al. 1997) presumably limited the sensitivity of PCR

amplification in many of the ice cores samples examined.

Concentrating DNA released from cells in the melt water

during thawing and filtering provided more routinely

successful templates for PCR amplification than extractions

from cells concentrated on 0.2 um filters. It seems

possible that inhibitors may have been sufficiently diluted

in these filtrates, or removed during concentration in

173

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. these fractions, or that the quantity of released DNA in

these filtrates simply exceeded the DNA present in unbroken

c e l l s .

It is remarkable that microorganisms can maintain

viability over hundreds of thousands of years trapped in

glacial ice. The isolates obtained certainly appear to

possess features that might enhance their survival while

dormant, but the thermodynamic reality is that in the

absence of metabolic activity, cells must incur a

significant amount of macromolecular damage over such long

periods of time. This point has often been raised in

discussions of reported microbial revivals from ancient

salt, amber, and permafrost (Gilichinsky et al. 1993; Cano

et al. 1995; Shi et al. 1997; Greenblatt et al. 1999;

Vreeland et al. 2000) In contrast, however, to the

isolates from these once active environments that then

became impermeable geological materials, the microorganisms

deposited in glacial ice are most likely dormant and

sublethally injured, and then presumably still endure

thousands of years of additional damage while remaining

v i a b l e .

174

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It was from this perspective that the experiments were

undertaken to determine if metabolic activity was possible

under conditions comparable to those in glacial ice. DNA

and protein synthesis at -15°C were investigated to

determine if such macromolecular syntheses or possibly

repair were possible during apparently frozen storage.

According to the calculations of Price (2000), the supply

of organic carbon in the liquid veins within ice is

sufficient to support a small population of cells (-10-102

cells/cm3) for hundreds of thousands of years. The evidence

presented here is consistent with macromolecular synthesis

occurring at ice temperatures in which water exists between

ice crystals, and on the surfaces of entrapped cells and

air bubbles (Fig 4.2-4.6). Indirect evidence for microbial

activity in glacial ice was also obtained when analysis of

the air bubbles in cores from Vostok and Sajama revealed

isotopic fractionation profiles consistent with in situ

microbiological production of nitrous oxide and methane,

respectively (Sowers 2001; T. Sowers, personal

communication) . Geochemical anomalies attributed to

microbial activity in Greenland ice have also been reported

(Souchez et al. 1995; Souchez et al. 1998), and this issue

175

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. must now be experimentally addressed. Perhaps 14C-dating of

the microbial fraction in an ice core sequence might serve

as a practical first step.

Regardless of metabolic status, cells entombed in

glacial ice remain viable for >500,000 years, and the

possibility exists that microbes transported through ice

sheets have established unique ecosystems in the subglacial

environment. Examination of an ice core recovered from

within the accretion zone of Lake Vostok has provided a

glimpse of the microbial inhabitants in a environment

isolated from the surface for at least 0.5 million years,

and perhaps as long as the continent has been glaciated.

Several of the isolates recovered from the lake water are

close phylogenetic relatives of bacteria commonly recovered

in this study of glacial ice. The water in Lake Vostok

originates from the overlying glacier (Jouzel et al. 1999),

and our results are consistent with models of circulation

within the lake predicting the accreted ice is formed from

low-salinity water at the ice ceiling, which is composed

chiefly of the most recent glacial melt to enter the lake

(Siegert et al. 2001).

176

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. If a flourishing microbial ecosystem were found to

exist within the water or sediment of this subsurface

environment, it would represent one of the most extreme in

the biosphere, and ice core studies at OSU and by others

(Karl et al. 1999; Priscu et al. 1999) have made a

preliminary microbiological analysis possible. Our results

predict that Lake Vostok contains viable bacteria, and it

seems noteworthy that several are most closely related to

species identified in glacial ice (Fig 5.2A), and

additional 16S rDNA sequences obtained are most similar to

species common to freshwater ecosystems (Fig 5.4).

Identifying bacteria recovered from glacial ice cores

originating from worldwide locations has provided a way to

examine the influence of geography and climate change on

the composition of entrapped species, observe common

survival strategies, investigate longevity while frozen,

and explore potential habitats for activity in the glacial

and subglacial environment. Studying microbial survival

and preservation under some of the most extreme conditions

on the planet is intrinsically interesting, but has also

yielded clues about potential refuge environments for life

during periods of global ice cover, such as those predicted

by the snowball Earth hypothesis (Hoffman et al. 1998;

177

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kirshvink 1992) . The results obtained are clearly relevant

to discussions of the likelihood of microbial survival in

frozen extraterrestrial environments and during

interplanetary transport.

178

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF REFERENCES

Abyzov, S.S. 1993. Microorganisms in the Antarctic ice. In Antarctic Microbiology (E.I. Friedmann, Ed.), pp. 265- 295, Wiley-Liss, Inc., New York.

Abyzov, S.S., I.N. Mitskevich, and M.N. Poglazova. 1998 Microflora of the deep glacier horizons of central Antarctica. Microbiology (Moscow) €7:66-73.

Aldsworth, T.G. , R.L. Sharman, and C.E.R. Dodd. 1999. Bacterial suicide through stress. Cell. Mol. Life Sci. 56:378-383.

Alley, R.B. and R.A. Bindschadler. 2001. The west Antarctic ice sheet and sea-level change. In The West Antarctic Ice Sheet: Behavior and Environment (B. Alley and R.A. Bindschadler, Eds.), pp. 1-11, Antarctic research series; v. 77, American Geophysical Union, Washington, D.C.

Alur, M.D., and N. Grecz. 1975. Mechanism of injury of Escherichia coli by freezing and thawing. Biochem. Biophys. Res. Comm. 62:3 08-312.

Anderson, D.A. 1967. Ice nucleation and the substrate-ice interface. Nature 216:563-566.

Atlas, R.M. and R. Bartha. 1998. Microbial ecology: fundamentals and applications. 4th edition, Benjamin- Cummings Science Publishing, Menlo Park, California.'

Bada, J.L., X.S. Wang, H.N. Poinar, S. Paabo, and G.O. Poinar. 1994. Amino acid racemization in amber-entombed insects: implications for DNA preservation. Geochim. Cosmochim. Acta. 58:3131-3135.

Balkwill, D.L., R.H. Reeves, G.R. Drake, J.Y. Reeves, F.H. Crocker, M.B. King, and D.R. Boone. 1997. Phylogentic characterizationof bacteria in the subsurface microbial culture collection. FEMS Microbiol. Rev. 20:201-216. 179

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bell, R.E. 1998. Review of Lake Vostok Studies. In Lake Vostok Workshop Final Report. Bell, R.E and Karl, D.M. (eds.) Washington, DC: NSF Press, pp 9-12.

Benson, D.A., Karsch-Mizrachi, I., Lipman, D. J. , Ostell, «J., Rapp, B.A., and Wheeler, D.L. 2000. GenBank. Nucl. Acids Res. 28:15-18.

Bowman, J.P., S.A. McCammon, M.V. Brown, D.S. Nichols, and T.A. McMeekin 1997. Diversity and association of psychrophilic bacteria in Antarctic sea ice. Appl. Environ. Microbiol. 63:3068-3078.

Brambilla, E., H. Hippe, A. Hagelstein, B.J. Tindall, and E. Stackebrandt. 2001. 16S rDNA diversity of cultured and uncultured prokaryotes of a mat sample from Lake Fryxell, McMurdo Dry Valleys, Antarctica. Extremophiles 5:23-33.

Bratina, B. J. , B.S. Stevenson, W.J. Green, and T.M. Schmidt. 1998. Manganese reduction by microbes from the oxic regions of the Lake Vanda (Antarctica) water column. Appl. Environ. Microbiol. 64:3791-3797.

Budd, W.F., D. Jenssen, J.H.I. Leach, I.N. Smith, and U. Radok. 1986. The north polar ice cap of Mars as a steady- state system. Polarforsch. 56:43-63.

Budyko, M. I. 1969. The effect of solar radiation variations on the climate of the earth. Tellus 21:611-619.

Cairns-Smith, A.G. 1985. Seven clues to the origin of life. Cambridge University Press, Cambridge, UK.

Cano, R.J., and M.K. Boruki 1995. Revival and identification of bacterial spores in 25- to 40-million- year-old Dominican amber. Science 268:1060-1064.

Carpenter, E.J., S. Lin and D.G. Capone. 2000. Bacterial activity in South Pole snow. Appl. Environ. Microbiol. 66:4514-4517.

Castello, J.D., S.O. Rogers, W.T. Starmer, C.M. Catranis, L. Ma, G.D. Bachand, Y. Zhao, andJ.E. Smith. 1999. Detection of tomato mosaic tobamovirus RNA in ancient glacial ice. Polar Biol. 22:207-212.

180

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chandler, D.P., F.J. Brockman, T.J. Bailey, and J.K. Fredrickson. 1998. Phylogenetic diversity of archaea and bacteria in deep subsurface paleosol. Microb. Ecol. 36:37- 50.

Chapelle, F.H., K. O'Neill, P.M. Bradley, B.A. Methe, S.A. Ciufo, L.L. Knobel, and D.R. Lovley. 2002. A hydrogen-based subsurface microbial community dominated by methanogens. Nature 415:312-315.

Chyba, C.F., and K.P. Hand. 2001. Life without photosynthesis. Science 292:2026-2027.

Chyba, C.F., and C.B. Phillips. 2001. Possible ecosystems and the search for life on Europa. Proc. Natl. Acad. Sci. 98:801-804.

Chyba, C.F. 2000. Energy for microbial life on Europa. Nature 403:381-382.

Cisar, J.O., D.-Q. Xu, J. Thompson, W. Swaim, L. Hu, and D.J. Kopecko. 2000. An alternative interpretation of nanobacterial-induced biomineralization. Proc. Natl. Acad. Sci. USA 97:11511-11515.

Clifford, S.M., and 27 colleagues. 2000. The state and future of Mars polar science and exploration. Icarus 1441210-242.

Dansgaard, W., S.J. Johnsen, H.B. Clausen, D. Dahl-Jensen, N.S. Gundestrup, C.U. Hammer, C.S. Hvidberg, J.P. Steffensen, A.E. Sveinbjornsdottir, J. Jouzel, and G. Bond. 1993. Evidence for general instability of past climate from a 250-kyr ice-core record. Nature 364:218-220.

DeSalle, R., J. Gatesy, W. Wheeler, and D. Grimaldi. 1992. DNA sequences from a fossil termite in Oligo-Miocene amber and their phylogenetic implications. Science 257:1933-1936.

Doran, P.T., M. Voytek, D. Karl, L. Becker, J. Tiedje, and K. Moran. 1998. In Lake Vostok Worshop Final Report. Bell, R.E., and D.M. Karl (eds) . Washington, D.C. : National Science Foundation, pp. 31-32.

181

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Duman, J.G. and T.M. Olsen. 1993. Thermal hysteresis protein activity in bacteria, fungi, and phylogenetically diverse plants. Cryobiology 30:322-328.

Evans, D.A., N.J. Beukes, and J.L. Kirschvink. 1997. Low- latitude glaciation in the palaeoproterozoic era. Nature 386:262-266.

Finegold, L. 1996. Molecular and biophysical aspects of adaptation of life to temperatures below the freezing point. Adv. Space Res. 18:87-95.

Fitz-Gibbon, S.T., and C.H. House. 1999. Whole genome-based phylogenetic analysis of free-living microorganisms. Nucl. Acids Res. 27:4218-4222.

Fogg, G.E. 1998. The Biology of polar habitats. Oxford Univ. Press, New York.

Franzmann, P.D., P. Hopfl, N. Weiss and B.J. Tindall. 1991. Psychrotrophic, lactic acid-producing bacteria from anoxic waters in Ace Lake, Antarctica: Carnobacterium fundi turn sp. nov. and Carnobacterium alter fundi turn sp. nov. Arch. Microbiol 156:255-262.

Friedmann, E.I. 1982. Endolithic microorganisms in the Antarctic cold desert. Science 193:1045-1053.

Fritsen, C.H. and J.C. Priscu. 1998. Cyanobacterial assemblages in permanently ice covers on Antarctic lakes: distribution, growth rate, and temperature response of photosynthesis. J. Phycol. 34:587-597.

Fuzzi, G., P. Mandrioli, and A. Perfetto. 1997. Fog droplets - an atmospheric source of secondary biological aerosol particles. Atmosph. Environ. 31:287-290.

Gaidos, E.J., and F. Nimmo. 2000. Tectonics and water on Europa. Nature 405:637.

Gaidos, E.J., K.H. Nealson, and J.L. Kirschvink. 1999. Life in ice-covered oceans. Science 284:1631-1633.

Geiges, 0. 1996. Microbial processes in frozen food. Adv. Space Res. 18:109-118.

182

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gilichinsky, D.A., V.S. Soina, and M.A. Petrova 1993. Cryoprotective properties of water in the earth cryolithosphere and its role in exobiology. Origin Life Evol. Biosphere 23:65-75.

Giorgio, C.D., A. Krempff, H. Guiraud, P. Binder, C. Tiret, and G. Dumenil. 1996. Atmospheric pollution by airborne microorganisms in the city of Marseilles. Atmosph. Environ. 30:155-160.

Gosink, J. J. , R.P. Herwig and J.T. Staley. 1997. Octadecabacter arcticus gen. nov., sp. nov., a n d O. antarcticus, sp. nov., nonpigmented, psychrophilic gas vacuolated bacteria from polar sea ice and water. System. Appl. Microbiol. 20:356-365.

Gosink, J.J., and J.T. Staley. 1995. Biodiversity of gas vacuolate bacteria from Antarctic sea ice and water. Appl. Environ. Microbiol. 61:3486-3489.

Grecz, N., T.L. Hammer, C.J. Robnett, and M.D. Long. 1980. Freeze-thaw injury: evidence for double strand breaks in Escherichia coli DNA. Biochem. Biophys. Res. Comm. 93:1110- 1113.

Greenberg, R., P. Geissler, B.R. Tufts, and G.V. Hoppa. 2000. Habitability of Europa's crust: the role of tidal- tectonic processes. J . Geophys. Res. 105:17551-17562.

Greenblatt, C.L., A. Davis, B.G. Clement, C.L. Kitts, T. Cox, and R.J. Cano. 1999. Diversity of microorganisms from amber. Microb. Ecol. 38:58-68.

Grootes, P.M., M. Stuiver, J.W.C. White, S. Johnsen, and J. Jouzel. 1993. Comparison of oxygen isotope records from the GISP2 and GRIP Greenland ice cores. Nature 366:552-554.

Hazen, R.M. and E. Roedder. 2001. How old are bacteria from the Permian age? Nature 411:155.

Henderson, K.A., L.G. Thompson and P.-N. Lin. 1999. Recording of El NifLo in ice core lsO records from Nevado Huascaran, Peru. J. Geophys. Res 104:31053-31065.

183

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hirsch, P., W. Ludwig, C. Hethke, M. Sittig, B. Hoffmann and C.A. Gallidowski. 1998. Hymenobacter roseosalivarius gen. nov., sp. nov. from continental Antarctic soils and sandstone: bacteria of the Cytophaga/Flavobacterium/ Bacteroides line of phylogenetic descent. System. Appl. Microbiol. 21:374-383.

Hobbs, P.V. 1974. Ice Physics. Clarendon, Oxford.

Hoffman, P.F. and D.P. Schrag. 2000. Snowball Earth. Scientific American 282:68-75.

Hoffman, P.F., A.J. Kaufman, G.P. Halverson, and D.P. Schrag. 1998. A neoproterozoic snowball Earth. Science 281:1342-1346.

Hoyle, F. and C. Wickramasinghe. 1981. Evolution from space: a theory of cosmic creationism. Simon and Schuster, N e w York.

Hugenholtz, P., B.M. Goebel, and N.R. Pace. 1998. Impact of culture-independent studies on the emerging phylogenetic view of bacterial diversity. J. Bacteriol. 180:4765-4774.

Hughes, T.J. 1998. Ice sheets. Oxford University Press Inc., New York.

Jouzel, J., J.R. Petit, R. Souchez, N.I. Barkov, V.Y. Lifenkov, D. Raymond, M. Stievenard, Nil. Vassiliev, V. Verbeke, and F. Vimeux. 1999. More than 200 meters of lake ice above subglacial Lake Vostok, Antarctica. Science 286:2138-2141.

Junge, K. , J.J. Gosink, H.G. Hoppe, and J.T. Staley 1998. Arthrobacter, Brachybacterium a n d Planococcus isolates identified from Antarctic sea ice brine. Description of Plannococcus mcceekinii, sp. nov. System. Appl. Microbiol. 21:306-314.

Kapitsa, A.P., J.K. Ridley, G. deQ. Robin, M.J. Siegert, and I.A. Zotikov. 1996. A large deep freshwater lake beneath the ice of central East Antarctica. Nature 381:684-686.

184

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Karl, D.M., D.F. Bird, K. Bjorkman, T. Houlihan, R. Shackelford, and L. Tupas. 1999. Microorganisms in the accreted ice of Lake Vostok, Antarctica. Science 286:2144- 2147.

Kennedy, M.J., S.L. Reader, and L.M. Swierczynski. 1994. Preservation records of micro-organisms: evidence of the tenacity of life. Microbiol. 140:2513-2529.

Kirschvink, J .L ., E .J . Gaidos, L .E . Bertani, N.J. Beukes, J. Gutzmer, L.N. Maepa, and R.E. Steinberger. 2000. Paleoproterozoic snowball Earth: extreme climatic and geochemical global change and its biological consequences. Proc. Natl. Acad. Sci. 97:1400-1405.

Kirschvink, J.L. 1992. in The Proterozoic biosphere: a multidisciplinary study, pp. 51-52, ed. Schopt, J.W., C. Klein, and D. Des Maris. Cambridge University Press, Cambridge, U.K.

Kivelson, M.G. , K.K. Khurana, C.T. Russell, M. Volwerk, R.J. Walker, and C. Zimmer. 2000. Galileo magnetometer measurements: a stronger case for a subsurface ocean at Europa. Science 289:1340-1343.

Kuske, C.R., K.L. Banton, D.L. Adorada, P.C. Stark, K.K. Hill, and P.J. Jackson. 1998. Small-scale DNA sample preparation method for field PCR detection of microbial cells and spores in soil. Appl. Environ. Microbiol. 64:2463-2472.

Lane, D.J. 1991. 16S/23S rRNA sequencing. In Nucleic Acid Techniques in Bacterial Systematics. pp. 133, Stakebrandt, E, . and Goodfellow, M. (eds) . Chichester, N.Y. : Wiley P r e s s .

Lazcano, A. and S.L. Miller. 1996. The origin and early evolution of life: prebiotic chemistry, the pre-RNA world, a n d time. Cell 85:793-798.

Lieske, R. 1932. Uber das vorkommen von bakterien in kohlen-flozen (bacteria in coal). Biochem. Z. 250:339-351.

Lighthart, B., and B.T. Shaffer. 1995. Airborne bacteria in the atmospheric surface layer: temporal distribution above a grass seed field. Appl. Environ. Microbiol. 61:1492-1496. 185

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lindahl, T. 1993. Instability and decay of the primary structure of DNA. Nature 362:709-715.

Lipman, C.B. 1928. The discovery of living microorganisms in ancient rocks. Science 68:272-273.

Liu, K., Z. Yao, and L.G. Thompson 1998. A pollen record of Holocene climate changes from the Dunde ice cap, Qinghai- Tibetan Plateau. Geology 26:135-138.

Lorius, C., J. Jouzel, C. Ritz, L. Merlivat, N.E. Barkov and Y.S. Korotkevich. 1985. 150,000-year climatic record from Antarctic ice. Nature 316:591-595.

Lunine, J.I. 1999. Earth: evolution of a habitable world. Cambridge University Press, Cambridge, UK.

Maidak, B.L., J.R. Cole, T.G. Lilburn, C.T. Parker Jr, P.R. Saxman, R.J. Farris, G.M. Garrity, G.J. Olsen, T.M. Schmidt, and J.M. Tiedje. 2001. The RDP-II (Ribosomal Database Project). Nucl. Acids Res. 29:173-174.

Makarova, K.S., L. Aravind, M.Y. Galperin, N.V. Grishin, R.L. Tatusov, Y . I. Wolf, and E.V. Koonin. 1999. Comparative genomics of the archaea (euryarchaeota) : evolution of conserved protein families, the stable core, and the variable shell. Genome Res. 9:608-628.

Margulis, L. and D. Sagan. 1995. What is life? Simon and Schuster, New York.

Marshall, W.A. and M.O. Chalmers. 1997. Airborne dispersal of Antarctic algae and cyanobacteria. Ecography 20:585-594.

Marthi, B., and B. Lighthart. 1990. Effects of betaine on enumeration of airborne bacteria. Appl. Environ. Microbiol. 56:1286-1289.

Marthi, B., B.T. Shaffer, B. Lighthart, and L. Ganio. 1991. Resuscitation effects of catalase on airborne bacteria. Appl. Environ. Microbiol. 57:2775-2776.

Marthi, B. 1994. Resuscitation of microbial bioaerosols. In: Atmospheric Microbial Aerosols: Theory and Applications, pp. 210-225, eds. B. Lighthart and A.J. Mohr. Chapman and Hall, New York. 186

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mattimore, V. and J.R. Battista. 1996. Radioresistance of Deinococcus radiodurans: functions necessary to survive ionizing radiation are also necessary to survive prolonged desiccation. J. Bacteriol. 178:633-637.

Medlin, L., Elwood, H.J., Stickel, S., and Sogin, M.L. 1988. The characterization of enzymatically amplified eukaryotic 16S-like rRNA coding regions. Gene 71:491-499.

Meyer, C. 2001. Definition of the apparent requirements for cold curation of Mars samples. NASA Technical Memo

McKay, C.P. 2001. The deep biosphere: lessons for planetary exploration. In: Subsurface Microbiology and Biogeochemistry, pp. 315-327. Eds. J.K. Fredrickson and M. Fletcher. Wiley-Liss Inc.

McKay, C.P., and C.R. Stoker. 1989. The early environment and its evolution on Mars: implications for life. Rev. Geophys. 27:189-214.

McKay, C.P. and E.I. Friedmann. 1985. The cryptoendolithic environment in the Antarctic cold desert: temperature variations in nature. Polar Biol. 4:19-25.

McKay, D.S., E.K. Gibson, K.L. Thomas-Keptra, H. Vali, S. Romanek, S.J. Clemett, X.D.F. Chillier, C.R. Maechling, and N. Zare. 1996. Search for past life on Mars: possible relic biogenic activity in martian meteorite ALH84001. Science 273:924-930.

Melosh, H.J. 1988. The rocky road to panspermia. Nature 333:687-688.

Moree, M.I., J.B. Herrick, M.C. Silva, W.C. Ghiorse, and E.L. Madsen. 1994. Quantitative cell lysis of indigenous microorganisms and rapid extraction of microbial DNA from sediment. Appl. Environ. Microbiol. 60:1572-1580.

Morita, R.Y. 1997. Bacteria in oligotrophic environments: starvation-survival lifestyle. Chapman and Hall, New York.

Morita, R.Y. 2000. Is H 2 the universal energy source for long-term survival? Microb. Ecol. 38:307-320.

187

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Muyzer, G., E.C. DeWaal, and A.G. Uitterlinden. 1993. Profiling complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59:695-700.

Naish, T.R. and 32 colleagues. 2001. Orbitally induced oscillations in the East Antarctic ice sheet at the Oligocene/Miocene boundary. Nature 413:719-723.

Noever, D. 1998. Earth microbes on the moon. NASA Space Science News, September 1.

Olsen, G.J., J.H. Natusda, R. Hagstrom and R. Overbeek. 1994. fastDNAml: a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput. Appl. Biosci. 108:41-48.

Olson, J.B., T.F. Steppe, R.W. Litaker, and H.W. Paerl 1998. N 2 -fixing microbial consortia associated with the ice cover of Lake Bonney, Antarctica. Microb. Ecol. 36:231- 238 .

Orton, G. S., J. R. Spencer, L. D. Travis, T. Z. Martin, and L. K. Tamppari. 1996. Galileo Photopolarimeter- Radiometer Observations of Jupiter and the Galilean Satellites. Science 274x389-391.

Ostroumov, V.E. and C. Siegert. 1996. Exobiologial aspects of mass transfer in microzones of permafrost deposits. Adv. Space Res. 18:79-86.

Patt, T.E., G.C. Cole, J. Bland and R.S. Hanson. 1974. Isolation and characterization of bacteria that grow on methane and organic compounds as sole sources of carbon and energy. J. Bacteriol. 120:955-964.

Patterson, W.S.B. 1994. The physics of glaciers, 3rd edition. Elsevier Science Inc., Tarrytown, New York.

Paerl, H.W., and J.C. Priscu 1998. Microbial phototrophic, heterotrophic, and diazotrophic activities associated with aggregates in the permanent ice cover of Lake Bonney, Antarctica. Microb. Ecol. 36:221-230.

188

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Petit, J.R., J. Jouzel, D. Raynaud, N.I. Barkov, J.M. Bamola, I. Basile, M. Benders, J. Chappellaz, M. Davis, G. Delaygue, M. Delmotte, V.M. Dotlyakov, M. Legrand, V.Y. Lipendoc, C. Lorius, L. Pepin, 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.

Price, B.P. 2001. Life in solid ice? In: Life in Ancient Ice, Oregon Workshop Proceedings, Summer 2001.

Price, B.P. 2000. A habitat for psychrophiles in deep Antarctic ice. Proc. Natl. Acad. Sci. 97:1247-1251.

Priest, F.G. and A.T. Beckenbach. 1995. Age of bacteria from amber. Science 270:2015-2017.

Poinar, H.N., M. Hoss, J.L. Bada, and S. Paabo. 1996. Amino acid racemization and the preservation of ancient DNA. Science 272:864-866.

Priscu, J.C., E.E. Adams, W.B. Lyons, M.A. Voytek, D.W. Mogk, R.L. Brown, C.P. McKay, C.D. Takacs, K.A. Welch, C.F. Wolf, J.D. Kirschtein, and R. Avci. 1999. Geomicrobiology of subglacial ice above Lake Vostok, Antarctica. Science 286:2141-2144.

Priscu, J.C., C.H. Fritsen, E.E. Adams, S.J. Giovannoni, H.W. Paerl, C.P. McKay, P.T. Doran, D.A. Gordon, B.D. Lanoil, and J.L. Pinckney. 1998. Perennial Antarctic lake ice: an oasis for life in a polar desert. Science 280:2095-2098.

Reasoner, D.J. and E.E. Geldreich. 1985. A new medium for the enumeration and subculture of bacteria from potable water. Appl. Environ. Microbiol. 46:1-7.

Reysenbach, A.-L. and Pace, N.R. 1995. Reliable amplification of hyper thermophilic archaeal 16S rRNA genes by the polymerase chain reaction. In Archaea: a Laboratory Manual. Robb, F.T., and Place, A.R. (eds) Cold Spring Harbor Laboratory, N.Y.: Cold Spring Harbor Press.

189

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rivkina, E.M., E.X. Friedmann, C.P. McKay, and D.A. Gilichinsky. 2000. Metabolic activity of permafrost bacteria below the freezing point. Appl. Environ. Microbiol. 66:3230-3233.

Rummel, J.D. 2001. Planetary exploration in the time of astrobiology: protecting against biological contamination. Proc. Natl. Acad. Sci. 98:2128-2131.

Sambrook, J., E.F. Fritsch and T. Maniatis. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, N.Y.: Cold Spring Harbor Press.

Sattler, B., H. Puxbaum, and R. Psenner. 2001. Bacterial growth in supercooled cloud droplets. Geophys. Res. Lett. 28:239-242.

Schidlowski, M. 1987. Application of stable carbon isotopes to early biochemical evolution on Earth. Annu. Rev. Earth Planet Sci. 15:47-72.

Schopf, J.W. 1993. Microfossils of the early archaean apex chert: new evidence of the antiquity of life. Science 260:640-646.

Shi, T. , R.H. Reeves, D.A. Gilichinsky, and E.I. Friedmann 1997. Characterization of viable bacteria from Siberian permafrost by 16S rDNA sequencing. Microb. Ecol. 33:169- 179.

Siegert, M.J., J.C. Ellis-Evans, C. Mayer. J.R. Petit, A. Salamatin, and J.C. Priscu. 2001. Physical, chemical and biological processes in Lake Vostok and other Antarctic subglacial lakes. Nature 414:603-609.

Siegert, M.J., R. Kwok, C. Mayer and B. Hubbard. 2000. Water exchange between subglacial Lake Vostok and the overlying ice sheet. Nature 403:643-646.

Siegert, M.J., J.A. Dowdeswell, M.R. Gorman and N.F. McIntyre. 1996. An inventory of Antarctic subglacial lakes. Antarctic Science 8:281-286.

Skidmore, M.L., J.M. Foght and M.J. Sharp. 2000. Microbial life beneath a high Arctic glacier. Appl. Environ. Microbiol. 66:3214-3220. 190

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Souchez,R., A. Bouzette, H.B. Clausen, S.J. Johnsen, and J. Jouzel. 1998. A stacked mixing sequence at the base of the Dye 3 core. Geophys. Res. Lett. 25:1943-1946.

Souchez, R., M. Janssens, M. Lemmens, and B. Stauffer. 1995. Very low oxygen concentration in basal ice from Summit, Central Greenland. Geophys. Res. Lett. 22:2001- 2004.

Sowers, T. 2001. The N 2O record spanning the penultimate deglaciation from the Vostok ice core. J. Geograph. Res. 106:31903-31914.

Squyres, S.W., and M.H. Carr. 1986. Geomorphic evidence for the distribution of ground ice on Mars. Science 231:249-252.

Staley, J.T. and J.J. Gosink. 1999. Poles apart: biodiversity and biogeography of sea ice bacteria. Annu. Rev. Microbiol. 53:189-215.

Stams, A.J.M., K.C.F. Frijters, C.T.M.J. and J.B. Van Lier. 1992. Enrichment of thermophilic propionate-oxidizing bacteria in syntrophy with Methanobacterium thermoautotrophicum or Methanobacterium thermoformicicum. Appl. Environ. Microbiol. 58:346-352.

Stead, D., and S.F. Park. 2000. Roles of Fe superoxide dismutase and catalase in resistance of Campylobacter coli to freeze-thaw stress. Appl. Environ. Microbiol. 66:3110- 3112.

Stevens, T.O., and J.P. McKinley. 1995. Lithoautotrophic microbial ecosystems in deep basalt aquifers. Science 270: 450-454.

Strunk, 0., O. Gross, B. Reichel, M. May, S. Hermann, N. Struckmann, B. Nonhoff, M. Lenke, A. Vilbig, T. Ludwig, A. Bode, K.H. Schleifer and W. Ludwig. 1998. ARB: a software environment for sequence data. Department of Microbiology, Technical University of Munich, Munich, Germany.

Sun, H.J. and E.I. Friedmann. 1999. Growth on geological time scales in the Antarctic cryptoendolithic microbial community. Geomicrobiol. J. 16:193-202. 191

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Swofford, D. L . 1999. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version beta 4a. Sinauer Associates, Sunderland, MA.

Takacs, C.D., and J.C. Priscu 1998. Bacterioplankton dynamics in the McMurdo Dry Valley Lakes, Antarctica: production and biomass loss over four seasons. Microb. Ecol. 36:239-250.

Tanner, M.A. , B.M. Goebel, M.A. Dojka, and N.R. Pace. 1998. Specific ribosomal DNA sequences from diverse environmental settings correlate with experimental contaminants. Appl. Environ. Microbiol. 64:3110-3113.

Thompson, L.G., T. Yao, E. Mosley-Thompson, M.E. Davis, K.A. Henderson and P.-N. Lin. 2000. A high-resolution millennial record of the South Asian Monsoon from Himalayan ice cores. Science 289:1916-1919.

Thompson, L.G., M.E.Davis, E.M.Thompson, T.A.Sowers, K.A.Henderson, V.S.Zagorodnov, P.N.Lin, V.N. Mikhalenko, R.K.Campen, J.F.Bolzan, J.Cole-Dai and B.Francou. 1998. A 25,000 year tropical climate history from Bolivian ice cores. Science 282:1858-1864.

Thompson, L.G., T. Yao, M.E. Davis, K.A. Henderson, E. Mosley-Thompson, P.N. Lin, J. Beer, H.-A. Synal, J. Cole- Dai, and J.F. Bolzan. 1997. Tropical climate instability: The last glacial cycle from a Qinghai-Tibetan ice core. Science 276:1821-1825.

Touzel, J.P. and G. Albagnac. 1983. Isolation and characterization of Methanococcus mazei stra i n MC 3 . FEMS Microbiol. Lett. 16:241-245.

Turtle, E.P. and E. Pierazzo. 2001. Thickness of a Europan ice shell from impact crater simulations. Science 294:1326- 1328.

Van Dover, C.L., and 26 colleagues. 2001. Biogeography and ecological setting of Indian Ocean hyrdrothermal vents. Science 294:818-823.

192

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vorobyova, E., V. Soina, M. Gorlenko, N. Minkovskaya, N. Zalinova, A. Mamukelashvili, D. Gilichinsky, E. Rivkina and T. Vishnivetskaya. 1997. The deep cold biosphere: facts and hypothesis. FEMS Microbiol. Rev. 20:277-290.

Vreeland, R.H., W.D. Rosenzweig, and D.W. Powers. 2000. Isolation of a 250 million-year old halotolerant bacterium from a primary salt crystal. Nature 407:897-900.

Weber, P., and J.M. Greenberg. 1985. Can spores survive in interstellar space? Nature 316:403-407.

Weiss, B.P. J.L. Kirschvink, F.J. Baudenbacher, H. Vali, N.T. Peters, F .A. MacDonald, and J.P. Wikswo. 2000. A low temperature transfer of ALH84001 from Mars to Earth. Science 290:791-795.

Wharton, R.A., C.P. McKay, G.M. Simmons and B.C. Parker. 1985. Cryoconite holes on glaciers. BioScience 35:499-503.

Willerslev, E., A.J. Hansen, B. Christensen, J.P. Steffensen, and P. Arctander. 1999. Diversity of Holocene life forms in fossil glacier ice. Proc. Natl. Acad. Sci. 96:8017-8021.

Willimsky, G., H. Bang, G. Fischer, and M.A. Marahiel. 1992. Characterization of cspB, a Bacillus subtilis inducible cold shock gene affecting cell viability at low temperatures. J. Bacteriol. 174:6326-6335.

Wilson, I.G. 1997. Inhibition and facilitation of nucleic acid amplification. Appl. Environ. Microbiol. 63:3741-3751.

Wintzingerode, F.V., U.B. Goebel, and E. Stackebrandt. 1997. Determination of microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis. FEMS Microbiol. Rev. 21:213-229.

Woese, C. 1999. The universal ancestor. Proc. Natl. Acad. Sci. 95:6854-6859.

Zeng, L.W. and M. Kreitman. 1996. Simple strategy for sequencing cDNA clones. Biotechniques 21:446-450.

193

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Zolensky, M.E., R.J. Robnar, E.K. Gibson Jr., L.E. Nyquist, Y. Reese, C. Shih, and H. Wiesmann. 1999. Asteroidal water within fluid inclusion-bearing halite in an H5 chrondrite, Monahans (1998). Science 285:1377-1379.

194

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.